Design, synthesis and biological evaluation of novel O-carbamoyl ferulamide derivatives as multi-target-directed ligands for the treatment of Alzheimer's disease

Article abstract of DOI:10.1016/j.ejmech.2020.112265

A novel series of O-carbamoyl ferulamide derivatives were designed by multitarget-directed ligands (MTDLs) strategy, the derivatives were synthesized and evaluated to treat Alzheimer's disease (AD). In vitro biological evaluation demonstrated that compound 4f was the best pseudo-irreversible hBChE (human butyrylcholinesterase) inhibitor with an IC₅₀ value of 0.97 μM 4f was a potent selective MAO-B (monoamine oxidase-B) inhibitor (IC₅₀ = 5.3 μM), and could inhibit (58.2%) and disaggregate (43.3%) self-mediated Aβ aggregation. 4f also could reduce the levels of pathological tau and APP clearance, and displayed a wide safe range hepatotoxicity on LO2 cells. The in vivo studies revealed that 4f exhibited fascinating dyskinesia recovery rate and response efficiency on AlCl₃-mediated zebrafish, and demonstrated significant protective effect on vascular injury caused by Aβ_1-40. PET-CT imaging demonstrated that [¹¹C]4f exhibited high BBB penetration (especially could reach to hippocampus and striatum of brain) and had a fast brain uptake after intravenous bolus injection. Furthermore, compound 4f could improve scopolamine-induced cognitive impairment. Further, the metabolism in vitro of 4f was also investigated, and presented 3 metabolites in rat liver microsome metabolism, 4 metabolites in human liver microsome, and 4 metabolites in rat intestinal flora, providing previous data for the preclinical study. Therefore, these results implied that compound 4f was an advanced multi-function agent and deserved further preclinical study against mild-to-serve Alzheimer's disease.

Full text of DOI:10.1016/j.ejmech.2020.112265

Journal Pre-proof
Design, synthesis and biological evaluation of novel O-carbamoyl ferulamide
derivatives as multi-target-directed ligands for the treatment of Alzheimer's disease
Zhipei Sang, Keren Wang, Ping Bai, Anguo Wu, Jian Shi, Wenmin Liu, Gaofeng Zhu,
Yiling Wang, Yu Lan, Zude Chen, Yiyang Zhao, Zhanpin Qiao, Changning Wang,
Zhenghuai Tan
PII:
S0223-5234(20)30232-4
DOI:
https://doi.org/10.1016/j.ejmech.2020.112265
EJMECH 112265
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 18 February 2020
Revised Date: 19 March 2020
Accepted Date: 20 March 2020
Please cite this article as: Z. Sang, K. Wang, P. Bai, A. Wu, J. Shi, W. Liu, G. Zhu, Y. Wang, Y.
Lan, Z. Chen, Y. Zhao, Z. Qiao, C. Wang, Z. Tan, Design, synthesis and biological evaluation
of novel O-carbamoyl ferulamide derivatives as multi-target-directed ligands for the treatment of
Alzheimer's disease, European Journal of Medicinal Chemistry (2020), doi: https://doi.org/10.1016/
j.ejmech.2020.112265.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
©
2020 Published by Elsevier Masson SAS.

Design, synthesis and biological evaluation of novel
O-carbamoyl ferulamide derivatives as multi-target-directed
ligands for the treatment of Alzheimer’s disease
*
,#,1
#,1
#,2,3
,4
1
1
Zhipei Sang , Keren Wang , Ping Bai , Anguo Wu , Jian Shi , Wenmin Liu ,
5
4
2
2
1
1
Gaofeng Zhu , Yiling Wang , Yu Lan , Zude Chen , Yiyang Zhao , Zhanpin Qiao ,
*
,2
*,6
Changning Wang , Zhenghuai Tan
1
College of Chemistry and Pharmaceutical Engineering, Nanyang Normal University,
Nanyang, 473061, China
2
Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology,
Massachusetts General Hospital, Harvard Medical School, Charlestown, MA 02129,
USA.
3
Chengdu Institute of Biology, Chinese Academy of Sciences, Chengdu 610041, China,
4
73061, China
4
Sichuan Key Medical Laboratory of New Drug Discovery and Drugability Evaluation,
Southwest Medical University, Luzhou, 646000, China
5
State Key Laboratory of Functions and Applications of Medicinal Plants, Guizhou
Medical University, Guiyang, 550014, China
6
Institute of Traditional Chinese Medicine Pharmacology and Toxicology, Sichuan
Academy of Chinese Medicine Sciences, Chengdu, 610041, China
*
Corresponding Authors.
E-mail: sangzhipei@126.com (Zhipei Sang)

E-mail: tanzhh616@163.com (Zhenghuai Tan)
E-mail: cwang15@mgh.harvard.edu (Changning Wang)
#
These authors contributed equally.

ABSTRACT: A novel series of O-carbamoyl ferulamide derivatives were designed by
multitarget-directed ligands (MTDLs) strategy, the derivatives were synthesized and
evaluated to treat Alzheimer’s disease (AD). In vitro biological evaluation demonstrated
that compound 4f was the best pseudo-irreversible hBChE (human butyrylcholinesterase)
inhibitor with an IC value of 0.97 μM. 4f was a potent selective MAO-B (monoamine
5
0
oxidase-B) inhibitor (IC = 5.3 μM), and could inhibit (58.2%) and disaggregate (43.3%)
5
0
self-mediated Aβ aggregation. 4f also could reduce the levels of pathological tau and APP
clearance, and displayed a wide safe range hepatotoxicity on LO2 cells. The in vivo
studies revealed that 4f exhibited fascinating dyskinesia recovery rate and response
efficiency on AlCl -mediated zebrafish, and demonstrated significant protective effect on
3
11
vascular injury caused by Aβ1-40. PET-CT imaging demonstrated that [ C]4f exhibited
high BBB penetration (especially could reach to hippocampus and striatum of brain) and
had a fast brain uptake after intravenous bolus injection. Furthermore, compound 4f could
improve scopolamine-induced cognitive impairment. Further, the metabolism in vitro of
4
f was also investigated, and presented 3 metabolites in rat liver microsome metabolism,
4
metabolites in human liver microsome, and 4 metabolites in rat intestinal flora,
providing previous data for the preclinical study. Therefore, these results implied that
compound 4f was an advanced multi-function agent and deserved further preclinical
study against mild-to-serve Alzheimer’s disease.
KEYWORDS. Alzheimer’s disease; O-carbamoyl ferulamide derivatives; Multi-function
agents; Zebrafish AD; PET-CT imaging; Scopolamine-induced cognitive impairment;

Metabolism in vitro.
1
. INTRODUCTION
Alzheimer's disease (AD) is first described by Dr. Alois Alzheimer in 1906. It is
characterized by a decline in memory, problem-solving and other cognitive skills that
1
affects a person’s ability to do everyday activities. Up to now, 50 million people are
suffering from dementia globally. The report estimates that by 2050 over 152 million
people worldwide will have dementia, tripling the number of people who have it now.
The total estimated worldwide cost of dementia in 2018 in US$ 1 trillion, and the figure
2
will rise to US$ 2 trillion by 2030. It has been a global problem.
The cause of AD is poorly understood, pathologic brain changes mainly include low
levels of acetylcholine (ACh), amyloid-β (Aβ) deposits, neurofibrillary tangles, and
1
, 3
vascular injury.
The current clinical drugs are mainly focused on cholinesterase
inhibitors, including donepezil, rivastigmine and galantamine, which could improve ACh
levels by inhibiting AChE. Long-term clinical uses indicate that these drugs just
temporarily improve cognition or slow the rate of cognitive decline, and these drugs can
not prevent or reverse the progression of AD. In addition, during the long-term clinical
uses of ChEs inhibitors, the patients have been accompanied by serious side effects, such
4
, 5
as nausea, vomiting and diarrhea, even hepatotoxicity, like tacrine.
Therefore, the
development of more powerful drugs is urgent.
Given that the complex pathogenesis of AD, the development of
multi-target-directed ligands (MTDLs) has been used to address multiple factors involved

6
-8
in the progression of Alzheimer’s disease. These drugs, which possess two or more
complementary biological activities, may represent a promising clinical advance in the
future.
The classical cholinergic hypothesis shows that the AChE levels significantly increase
in the early stage of AD, and the AChE inhibitors approved by FDA have been
successfully applied and improve the symptom on mild AD patients, but do not produce
positive effects on the late stage of AD patients. Actually, recent studies display that the
AChE levels in the brain only increase during the mild to moderate AD patients, and the
level declines progressively as the progress of AD goes, at the late stage of AD patients,
the AChE levels slightly decrease and the BChE levels soar to 165% of the normal level,
9
this implying BChE takes over the hydrolysis of ACh in the late stage of AD. Due to the
altered BChE/AChE ratio and the demonstration of improving cognitive effects of BChE
inhibition in vivo using selective BChE inhibitors, targeting selective BChE inhibition has
1
0-12
become a viable strategy to treat serve AD patients.
In addition, selective BChE
inhibition could circumvent the classical cholinergic side effect caused by AChE
1
3
inhibition. Therefore, selective BChE inhibition would be a promising strategy for the
late stages of AD patients.
Aβ plaques and neurofibrillary tangles are the classical pathological feature of the
brain of AD patients, Aβ plaques are formed by the extracellular aggregates of Aβ, and
neurofibrillary tangles are formed by intraneuronal accumulations of
2
, 14
hyperphosphorylated tau.
It is clear that the ubiquitin-proteasome system (UPS) is the

main route of protein and organelle clearance in eukaryotic cells and is also important of
the degradation of misfolded proteins (such as Aβ plaques and hyperphosphorylated Tau
1
5, 16
in AD) in the endoplasmic reticulum.
Therefore, modulation of UPS may lead to the
clearance of Aβ aggregates and hyperphosphorylated Tau and provide a new promising
strategy to treat AD.
In addition, recent studies demonstrate that MAO-B level, one of the monoamine
oxidase (MAO) form serving the metabolism of monoamine neurotransmitters, is
increased in the hippocampus and cerebral cortex of AD patients. The in vivo studies
display that the selective MAO-B inhibitor selegiline produces a beneficial effect on
1
7
cognitive symptoms in patients with moderate to severe AD. Thus, the multi-function
agents could simultaneously regulate these pathological features (such as BChE, MAO-B,
Aβ, and tau) might be an important step for the searching of powerful drugs against AD.
Ferulic acid (4-hydroxy-3-methoxy cinnamic acid, FA) was first isolated from
Ferula foetida. It is considered to be as a promising agent for AD therapy due to its free
radical scavenging effects, Aβ aggregation inhibition properties, anti-inflammatory
1
8, 19
property and neuroprotective effects.
ferulic acid-based derivatives as multifunction agents.
that introduction of 1-benzylpiperidine (LD-1) and 1,2,3,4-tetrahydroisoquinoline (LD-2)
Many groups have successfully developed
2
0-27
In our lab, it had been verified
2
8
fragments into ferulic acid could enhance the BChE inhibitory activity , and then
introducing the O-alkylamine fragments into the LD-1 to get ferulic acid-O-alkylamine
29
derivatives, compound EJ presenting good selective BChE inhibitors , in order to

explore the structure-activity-relationship, novel ferulic acid derivatives were
3
0
synthesized , and compound AC showed the best selective BChE inhibitory activity,
while it did not significantly improve cognitive disorder, the possible reason may be the
neurotoxicity due to the high BChE inhibitory potency. So, in order to develop novel
selective BChE inhibitor, we plan to replace the alkylamine fragment with carbamate
fragment which is the pharmacophore of rivastigmine to create novel O-carbamoyl
ferulamide derivatives (Figure 1), hoping these derivatives could be promising agents for
the treatment of AD.
In this work, novel O-carbamoyl ferulamide derivatives were designed and synthesized,
the in vitro biological activities are evaluated by AChE/BuChE inhibition,
MAO-A/MAO-B inhibition, and self-induced Aβ aggregation inhibition and
disaggregation. The docking studies and reversible inhibition assay are performed to
provide a reasonable explanation for the binding action of the optimal O-carbamoyl
ferulamide compound. The representative compound is further studied by modulation of
UPS to clear the misfolded proteins. Furthermore, the optimal active molecule is studied
in depth by the zebrafish AD model, PET-CT imaging and scopolamine-induced
cognitive impairment. The preliminary preclinical study was assessed by the metabolism
of liver microsome and intestinal flora.
2
. RESULTS AND DISCUSSION
2
.1 Chemistry. The synthetic of O-carbamoyl ferulamide derivatives 4a~4u were
described in Scheme 1, FA (1a) or cinnamic acid (1b) was reacted with secondary amine

1
2
28
NR R (A~F) in the presence of EDCI) and HOBt, to obtain the key intermediate 2a~2i,
yield from 30%~60%, and then the target compounds 4a~4u were afforded by reacting
with N,N-disubstituted carbamoyl chlorides (3a~3e), yield from 65%~83%.
2
.2 Inhibition Effects on AChE and BChE. The inhibitory potency of novel
O-carbamoyl ferulamide derivatives 4a~4u were performed using AChE (eeAChE and
3
0
hAChE) and BChE (eqBChE and hBChE) through the reported Ellman’s method. The
positive compounds FA (1a), cinnamic acid (1b) and rivastigmine were also investigated.
As displayed in Table 1, the parent compounds FA and cinnamic acid did not show
any eeAChE and eqBChE inhibitory activity, while the derivatives 4a~4u indicated
significant inhibitory activities against eeAChE and eqBChE. More interestingly, most of
the derivatives displayed potent selectivity for eqBChE over eeAChE, with the exception
of compounds 4h, 4i, 4m, 4n, 4o, 4q and 4r. In addition, the target compounds 4a~4r
displayed moderate AChE inhibitory activities with dozens of micromoles. Among these
derivatives, 4f (IC = 0.92μM) was the best eqBChE inhibitor with the selectivity index
5
0
1
2
(
SI) value was 12.3. The results indicated that the introduction of the NR R moiety of the
3
4
amide group and N, N-disubstituted carbamoyl NR R could increase the ChEs inhibitory
3
4
activity. From the screening data, when the side chain NR R was
N-ethyl-N-methylcarbamate, compound 4b with 4-benzylpiperidine side chain indicated
potent eqBChE inhibitory activity (IC50 = 3.9 μM), removal of benzyl group of 4b to
obtain compound 4m (IC = 19.6 μM), the eqBChE inhibitory activity sharply decreased.
5
0
Replacing 4-benzylpiperidine of 4b with 1,2,3,4-tetrahydroisoquinoline, compound 4f

(
IC50 = 0.92μM) was obtained and demonstrated the best eqBChE inhibitory potency, and
then ring opening of 1,2,3,4-tetrahydroisoquinoline, we got the compound 4p (IC50 = 1.5
μM) demonstrating similar eqBChE inhibitory activity compared to 4f. The phenomenon
1
2
displayed that the benzene ring of NR R side chain was considered to be as an important
role in eqBChE inhibition. When changing the N-ethyl-N-methylcarbamate moiety of 4b
with N,N-diethylcarbamate to gain compound 4a (IC = 2.8 μM), the eqBChE inhibitory
5
0
activity showed slightly increase. Similarly, replacing 4-benzylpiperidine of 4b with
,2,3,4-tetrahydroisoquinoline and benzylpiperazine, to get compounds 4e (IC = 2.0 μM)
1
5
0
and 4j (IC = 8.7 μM), respectively. It implied that 1,2,3,4-tetrahydroisoquinoline side
5
0
chain could increase inhibitory potency, while benzylpiperazine containing two basic N
atom remarkably decreased inhibitory activity. Sequentially, exchanging
N-ethyl-N-methylcarbamate moiety of 4b with N,N-dimethylcarbamate to obtain
compound 4c (IC = 1.3 μM), which showed the best eqBChE inhibitory activity in the
5
0
4
1
-benzylpiperidine series derivatives. Changing the 4-benzylpiperidine of 4c with
,2,3,4-tetrahydroisoquinoline to get compound 4g, which indicated that the eqBChE
inhibitory potency slightly decreased to 3.7 μM, while replacing 4-benzylpiperidine with
benzylpiperazine getting compound 4k (IC50 = 12.1 μM), the eqBChE inhibitory potency
sharply decreased. Changing 4-benzylpiperidine of 4c with piperidine to acquire
compound 4n (IC50 = 26.3 μM) showed sharply declined eqBChE inhibitory activity, and
then ring opening of piperidine of 4n to obtain compound 4r (IC = 19.2 μM), the
5
0
eqBChE inhibitory activity indicated a weak change. In addition, replacing

N-ethyl-N-methylcarbamate moiety of 4b with N,N-diphenylcarbamate to obtain
compound 4d, which showed the weakest eqBChE inhibitory potency with 6.6%
inhibition rate at 25 μM, and then we changed the 4-benzylpiperidine of 4d with
1
,2,3,4-tetrahydroisoquinoline, benzylpiperazine, piperidine, N-ethylbenzylamine and
diethylamine to obtain 4h (IC = 23.9 μM), 4l (9.7%), 4o (20%) and 4q (IC = 24.6 μM),
5
0
50
respectively, all of them exhibited weak eeAChE and eqBChE inhibitory activities. For
the best BChE inhibitor 4f (IC50 = 0.92 μM, SI = 12.3), we replaced the
N-ethyl-N-methylcarbamate moiety of 4f with N-methoxy-N-methylcarbamate to get
compound 4i, which was a modest selective AChE inhibitor (IC = 11.8 μM). In order to
5
0
investigate the effect of methoxy group in FA on ChEs inhibition, we synthesized
derivatives 4s~4u. The data in Table 1 displayed that derivatives 4s and 4t were selective
eqBChE inhibitors, while 4u was dual weak AChE/BChE inhibitory activity. Removing
the methoxy group of derivatives 4b (IC50 = 3.9 μM) and 4f (IC50 = 0.92 μM) to get
compounds 4s (IC = 5.3 μM) and 4t (IC = 3.8 μM), respectively, showing decreased
5
0
50
eqBChE inhibitory activity. So, the data indicated that the methoxy group of FA was
important for the eqBChE inhibitory activity. In addition, we further explored the
structure-activity-relationship by comparing our previous compounds EJ (IC50 = 0.07 μM)
2
9, 30
and AC (IC50 = 8.9 nM),
inhibiting that introducing the O-alkylamine fragment
showed higher BChE inhibitory activity than introducing the O-carbamate fragment.
In order to further study the ChEs inhibition, the synthesized derivatives were
re-evaluated using hAChE and hBChE. As presented in Table 1, derivatives 4a~4u

showed moderate to good ChEs inhibitory activity. Generally speaking, for the
derivatives 4a-4r, derivatives 4a, 4b, 4c, 4e, 4f and 4p exhibited good BChE inhibitory
activities, particularly, compound 4f (IC50 = 0.97 μM) displayed the best hBChE
inhibitory potency, which was better than rivastigmine (IC = 3.6 μM) approved by FDA.
5
0
1
2
The inhibitory activities and selectivity were remarkably affected by the NR R moiety of
3
4
the amide group and N, N-disubstituted carbamoyl NR R , briefly, compounds (4a~4i)
1
2
with the NR R moiety of the amide group (such as 4-benzylpiperidine and
1
,2,3,4-tetrahydroisoquinoline) showed better inhibitory activity than other compounds.
Compounds (4b, 4f, 4m and 4p) with N-ethyl-N-methylcarbamate fragment showed the
remarkable inhibitory potency, while, the compounds (4d, 4h, 4l, 4o and 4q) with
N,N-diphenylcarbamate fragment showed the lowest inhibitory activities. On the other
hands, we removed the methoxy group of FA to obtain derivatives 4s~4u, the data
showed that derivatives 4s and 4t were selective hBChE inhibitors, while 4u was
selective AChE inhibitor (IC = 14.8 μM). In addition, 4s (IC = 5.6 μM) showed better
5
0
50
hBChE inhibitory activity than 4b (IC50 = 6.5 μM), and 4t (IC50 = 3.7 μM) displayed
lowed hBChE inhibitory activity than 4f (IC50 = 0.97 μM), displaying that the methoxy
group played key role on hBChE inhibitory activity. Therefore, based on the results of
eeAChE/hAChE and eqBChE/hBChE, compound 4f was selected to further evaluation.
2
.3 Molecular modeling studies of 4f. According to the crystal structure of hBChE (PDB
code: 4tpk), docking was used to investigate the action mechanism of compound 4f with
30
hBChE. In the hBChE−4f (Figure 2A~C), 1,2,3,4-tetrahydroisoquinoline via sigma-π

interaction with crucial residue Trp82 and two π-π interactions were observed between
the benzene ring of 1,2,3,4-tetrahydroisoquinoline and important residue Trp430, which
3
0
had been presented in the high hBChE inhibitor AC in our previous work , showing the
1
,2,3,4-tetrahydroisoquinoline group was important for BChE inhibitory activity. In
addition, from the docking results, the N-ethyl-N-methylcarbamate fragment showed
strong hydrophobic interaction with residues Pro285, Gln119 and Ser287, while the
O-alkylamine fragment in our previous compound AC showed strong hydrophobic
interaction with residues Phe329, Val288 and Ala328, which had been identified as key
1
0, 30
active site key residues for hBChE inhibitory activity.
Therefore, the above observed
phenomenon could provide a reasonable binding mechanism for good BChE inhibitory
activity of 4f, but lower than compound AC.
2
.4 hBChE reversibility of inhibition by 4f. The reversibility of 4f toward hBChE
inhibition was investigated after dilution (Figure 3A) and with time monitoring (Figure
3
1
3
B). For the hBChE reversibility of inhibition, as shown in Figure 3A, compared to
controls, the 0.1 × IC50 of donepezil increased the hBChE activity to 15.3%, while the 0.1
IC50 of standard rivastigmine and compound 4f increased the hBChE activity to 1.1%
×
and -1.1%, respectively, implying that hBChE level slowly recovered after treating
donepezil, while rivastigmien and 4f did not indicate any recovery of hBChE under the
same conditions. The further recovery of hBChE inhibitors inhibition after dilution with
time monitoring was displayed in Figure 3B. The hBChE activity restored to 98.1% with
0
.1× IC donepezil at 120 min, but the 0.1 × IC of rivastigmine increased the hBChE
50 50

activity from 48.2% (0 min) to 73.3% (240 min), and the 0.1 × IC50 of 4f increased the
hBChE activity from 44.6% (0 min) to 66.2% (240 min), 4f presented the similar results
with rivastigmine. Thus, 4f was a pseudo-irreversible hBChE inhibitor, which was
consistent with rivastigmine.
2
.5 Inhibition of MAOs in Vitro. The recent evidences displayed that the abnormal
MAO-B expression in astroglia was observed around Aβ plaques in the brains of AD
3
2
patient. Thus, the inhibitory potencies towards the synthesized compounds against
3
3
hMAO-A and hMAO-B (recombinant human enzyme) were performed. As displayed in
Table 2, all the novel O-carbamoyl ferulamide derivatives showed moderate
MAO-A/MAO-B inhibitory activities. Seen from the data, the MAO-B inhibitory
1
2
potencies were remarkably affected by the NR R moiety of the amide group and N,
3
4
1
2
N-disubstituted carbamate moiety NR R . When the NR R moiety of the amide group
was 4-benzylpiperidine, compound 4a possessing N,N-diethylcarbamate showed good
MAO-B inhibitory activity (IC = 5.13 μM), changing N,N-diethylcarbamate of 4a with
5
0
N,N-diphenylcarbamate to get compound 4d (IC50 =13.3 μM), which displayed the lowest
MAO-B inhibitory potency, this phenomenon was also observed in compounds 4h, 4l, 4o
and 4q, it implied that compounds with N,N-diphenylcarbamate moiety remarkably
decreased MAO-B inhibitory activity. Replacing N,N-diethylcarbamate of 4a with
N-ethyl-N-methylcarbamate to get compound 4b, which was a dual MAO-A/MAO-B
1
2
inhibitor. When the NR R moiety of the amide group was 1,2,3,4-tetrahydroisoquinoline,
compounds 4e and 4h were dual MAO-A/MAO-B inhibitor, while, compounds 4f, 4g,

and 4i acted as selective MAO-B inhibitors with SI was 3, 3.6 and 3.3, respectively.
Among of which, compound 4f served as the best selective MAO-B inhibitor (IC50 = 5.3
μM),
and then ring opening of side chain 1,2,3,4-tetrahydroisoquinoline of 4f to obtain
compound 4p, a dual MAO-A/MAO-B inhibitor, demonstrating slightly lower MAO-B
1
2
inhibitory potency than that of 4f. However, when the NR R moiety of the amide group
was benzylpiperazine, piperidine, and diethylamine, we got compounds 4j, 4k, 4l, 4m, 4n,
4
o and 4r, which displayed weak inhibitory activities. On the other hand, we removed the
methoxy group of FA to get derivatives 4s~4u, the data displayed that derivatives 4s and
t were selective MAO-B inhibitors, while 4u was selective MAO-A inhibitor (IC = 9.8
4
5
0
μM). In addition, replacing the methoxy group of FA in derivatives 4b (IC = 7.52 μM
)
50
and 4f (IC50 = 5.3 μM) to get derivatives 4s (IC50 = 7.8 μM) and 4t (IC50 = 8.3 μM),
respectively, the MAO-B inhibitory activity showed a slight decrease, implying that the
methoxy group was important for the MAO-B inhibition.
2
.6 Molecular docking studies of 4f with MAO-B. The possible binding action of
33
compound 4f with MAO-B was carried out using human MAO-B (PDB code: 2v60). In
hMAO-B-4f complex (Figure 4), the carbamoyl group moiety was close to the FAD
(flavin adenine dinucleotide) cofactor. The compound 4f interacted with Tyr60 via an
intermolecular hydrogen bond and sigma-π interaction, respectively. In addition, the
carbonyl group of ferulic acid via two intermolecular hydrogen bonds interacted with
Thr201 and one intermolecular hydrogen bond with Gln206, respectively, showing that
the ferulic acid skeleton played key role in inhibiting MAO-B. Moreover, some

hydrophobic interactions were also presented between compound 4f and Pro102, Ser200,
Ile199, Thr201, Ile316, Thr314, Glu84, Leu88, Phe99, Leu328, Thr327, Tyr326, Met341
and Phe343.
2.7 Effect on Self-induced Aβ1-42 Aggregation. Many shreds of evidence had
demonstrated that high levels of BChE could improve the extracellular Aβ1-42 deposits,
1
2
BChE inhibitors might affect Aβ aggregation. Thus, the abilities of representative
compounds 4a, 4b, 4c, 4f, 4p, 4s and 4t to inhibit self-induced Aβ1-42 aggregation were
2
9
tested at 25μM by Thioflavin T (ThT) fluorescence. And curcumin was also tested for
compared purpose. As provided in Table 2, the target derivatives 4a, 4b, 4c, 4f, 4p, 4s
and 4t at 25μM showed good inhibitory potency with 49.7%, 53.7%, 56.6%, 58.2%,
5
0.7%, 48.1%, and 52.9% respectively. Particularly, 4f exhibited slightly better inhibitory
activity than that of curcumin (55.1%). Moreover, the data did not present obvious
structure-activity-relative. Compound 4f was selected to further study.
2
.8 Effect on Abundance of Aβ Fibrils by 4f. In order to supplement the ThT assay,
transmission electron microscopy (TEM) was used to probe Aβ1-42 aggregation (Figure
3
0, 31
5
A).
The results demonstrated that fresh Aβ1-42 formed fibrils after 24 h incubation,
however only small bulk aggregates were observed after adding 4f into Aβ1-42. The TEM
phenomenon was well consistent with the ThT assay, implying that 4f could inhibit Aβ1-42
aggregation.
2
.9 Molecular Docking of 4f with Aβ1-42. The docking was implemented to explore the
3
3
mechanism of 4f with Aβ1-42 (PDB: 1iyt). As demonstrated in Figure 6A~6C, two

intermolecular hydrogen bonds were presented between 4f and important residue Lys16.
29, 30
Compared with compounds EJ and AC,
the substituent side chain groups did not
show obviou effect on inhibiting Aβ1-42 aggregates. In addition, hydrophobic effects were
demonstrated between compound 4f and His6, Ser8, Glu11, Val12, Gln15, Lys16, Phe19,
Phe20, and Asp23, which might contribute to the binding of Aβ1-42 and compound 4f.
2
.10 Disaggregation of Self-mediated Aβ1-42 by 4f. The ThT assay was also used to
3
0, 33
assess the ability of 4f to disassemble self-mediated Aβ1-42 aggregation.
As observed
in Table 2, 4f could disassemble Aβ1-42 (43.3% disaggregation rate), meanwhile,
curcumin exhibited 58.3% disaggregation rate at 25μM. As shown in Figure 5B, the
TEM images in depth approved the ThT assay data. Therefore, 4f presented the inhibition
and disaggregation activity for self-induced Aβ1-42 aggregation.
2
.11 Neuroprotective Effect of 4f. The neuroprotective effect of 4f to protect SH-SY5Y
3
4
neurotoxicity caused by Aβ1-42 was performed by MTT assay. And the result was shown
in Figure 7, 4f did not show any cytotoxicity until 50 μM in Figure 7A, it meant that 4f
has a wide therapeutic safety range. As shown in Figure 7B, 25 μM Aβ1-42 was exposed
to SH-SY5Y cell for 48h, the cell viability remarkably declined to 50.3 % (p < 0.01 vs
untreated group), 4f showed remarkable protective properties at the given concentrations
1
and 10 μM with 68.1% (p < 0.01) and 83.8% (p < 0.05) cell viability, respectively. This
data clarified that 4f exhibited potent neuroprotective potency against SH-SY5Y
neurotoxicity caused by Aβ1-42
.12 4f Promoted Pathological Tau and APP Clearance by UPS. The Aβ plaques and
.
2

neurofibrillary tangles are the main pathological features in AD patient brains, the
clearance of Aβ aggregates and hyperphosphorylated would be beneficial for the
treatment of AD. The degradation pathways acting on such abnormal aggregate-prone
toxic proteins include the UPS and autophagy. In addition, enhancing autophagy could
improve cognitive disorder in the early-stage of AD, but accelerated progress during the
late-stage of AD. The UPS predominantly degraded short-lived nuclear and cytosolic
proteins. Thus, the modulation of UPS would be a one of the possible promising strategy
1
5, 35
anti-AD.
Firstly, we determined the safety range of 4f using MTT method on HT22 mouse
hippocampal neuronal cells. As shown in Figure 8, the cell viability did not show
obvious change at 10 μM of 4f, while when treatment with 20 μM and 80 μM 4f, the cell
viability reduced to 79.8% and 8.4%, respectively. This revealed that compound 4f had a
wide safety therapy range.
Secondly, HT22 cells were transfected with pRK5-GFP-Tau plasmid as tau protein
cell model with green fluorescence, and the tau protein levels were evaluated by the
fluorescence number and western blot after adding compound 4f. As shown in Figure 9A,
when transfecting HT22 cells with pRK5-GFP-Tau plasmid (as model group), the
fluorescence intensity significantly enhanced compared with the untreated group (control
group), implying the tau protein levels significantly increased. While treatment with 4f
(2.5, 5, 10 and 20 μM), the fluorescence intensity potentially decreased (Figure 9B and
9
C), the fluorescence intensity of model group added up to 40.9 (p < 0.001) compared

with the control group (0.0100), when adding compound 4f (2.5, 5, 10 and 20 μM), the
fluorescence intensity decreased to 32.6, 32.6, 33.3 and 29.2, respectively. Moreover,
according to the screening data in Figure 9D and 9E, the western blot analysis revealed
that compound 4f mildly decreased tau protein levels.
Moreover, when HT22 cells were transfected with pRK5-GFP-Tau P301L plasmid
as another tau protein cell model with green fluorescence, the fluorescence intensity did
not show obvious change between model group and compound 4f in Figure 10A. While
in Figure 10B and 10C, the fluorescence intensity of model group increased to 33.6 (p <
0
.001) compared with the control group (0.070), treatment with 4f (2.5, 5, 10 and 20 μM),
the fluorescence intensity reduced to 30.3, 23.9, 18.7 and 17.9, respectively, in a
dose-response manner. In addition, the western blot analysis in Figure 10D and 10E
displayed that compound 4f slightly decreased tau protein levels.
Further, HT22 cells were transfected with pEGFP-N1-APP plasmid as APP protein
cell model with green fluorescence, and the APP protein levels were tested when
treatment with compound 4f. As shown in Figure 11A, the HT22 cells transfected with
pEGFP-N1-APP plasmid (model group) showed significant fluorescence compared with
the control group, and when treatment with compound 4f (2.5, 5, 10 and 20 μM), the
fluorescence gradually disappeared. And the evidences in Figure 11B and 11C provided
strong support, the model group displayed remarkable fluorescence intensity with 33.2 (p
<
0.001) compared with the control group (0.13), when treatment with 4f (2.5, 5, 10 and
2
0 μM) reduced to 29.8, 13.6, 13.3 and 14.9, respectively, in a dose-dependent manner.

Further, the western blot analysis in Figure 11D and 11E indicated that 10 μM and 20
μM of 4f could significantly eliminate APP protein levels.
According to the above results, compound 4f could decrease tau and App protein
levels, and could be one potential candidate for the treatment of AD by degrading
abnormal hyperphosphorylated tau and Aβ aggregates protein.
2
.13 In Vitro Blood−Brain Barrier Permeation Assay. The parallel artificial membrane
permeation assay of the blood−brain barrier (PAMPA-BBB) was carried out to test the
3
6
29, 30
possible BBB permeability of 4f. Based on our previous work
, the derivatives with
−
6
permeability above 3.44 × 10 cm/s was considered to cross BBB, and we tested the
−
6
permeability of 4f was 21.42 × 10 cm/s. So, derivative 4f presented good BBB
permeability in vitro. Take our previous compounds EJ and AC into account, we could
get a conclusion that introducing O-alkylamine or O-carbamoyl fragments into ferulic
acid significantly enhanced BBB permeability.
2
.14 Hepatotoxicity of 4f. As we all known, tacrine, an excellent ChE inhibitor, is
approved by the FDA in 1993 against AD, but it has been forced to withdraw because of
3
7
its serve hepatotoxicity. Therefore, hepatotoxicity test is crucial for 4f to develop as
anti-AD candidate compound, the hepatotoxicity of compound 4f was tested on human
3
8
embryonic liver cell line LO2 using MTT assay. As shown in Figure 12, 4f did not
show hepatotoxicity until the concentration added up to 25 μM, while, when treatment
with 50μM 4f, the cell viability decreased to 46.9%, indicating that 4f had a wide safety
range of hepatotoxicity under 25μM.

2
.15 Theoretical Prediction of the ADME Properties. To correlate better the druglike
properties of 4f, the theoretical prediction of the absorption, distribution, metabolism, and
excretion (ADME) properties were calculated applying the Molinspiration property
program, and the results were displayed in Table 3.
From the data obtained for the theoretical evaluation of the ADME properties, 4f do
not break any point of the Lipinski’s rule of five. Therefore, 4f was chosen for further AD
model study.
2
.16 Efficiencies of 4f on AlCl -medicated zebrafish AD. AlCl -mediated zebrafish AD
3 3
3
9
model had drawn wide attention and were used to rapidly screen the anti-AD drugs.
(
1) Maximum non-lethal concentration (MNLC) and lethal concentration (LC ) of
10
4
f. To evaluate the safety profile of compound 4f, zebrafish were treated at given
concentrations of 3.125, 6.25, 12.5, 25, 50 and 100 µg/mL of 4f. The results observed
from Table 4 displayed that the MNLC and LC of 4f were 3.125 µg/ mL and 6.25 µg/
1
0
mL, respectively.
2) The acute toxicity phenotypes. As presented in Table 5, the zebrafish did not present
dead at the given concentrations of 0.35, 1.04 and 3.125 µg/mL for 4f, but leading to
3.3% mortality at 6.25 µg/mL of 4f. In addition, the zebrafish did not present acute
(
1
toxicity after adding 0.35 µg/ mL4f, which was consistent with the vehicle group (1%
DMSO) and the untreated group. While, treating with 4f at the dose of 1.04 and 3.125 µg/
mL induced 13.3% and 40% liver yolk sac absorption delay, respectively. As the
concentration increased to 6.25 µg/mL, presenting more toxicity including 100% liver

yolk sac absorption delay, 46.7% heart pericardial edema, 16.7% circulation absent blood
flow, 16.7% circulation slow blood flow, 26.7% brain small or misshapen, 33.3% jaw
short or misshapen jaw, 26.7% small or misshapen eye(s), 23.3% liver absent, 33.3%
kidney edema, 36.7% degenerating muscle within the somites and 53.3% body shape.
Generally speaking, the main toxicity phenotypes of 4f were heart pericardial edema,
circulation absent/slow blood flow and yolk sac absorption delay. Moreover, presenting
yolk sac absorption delay at a dose of 1.04 µg/ mL, presenting heart pericardial edema
and circulation absent/slow blood flow at a dose of 6.25 µg/mL and presenting 13.3%
death at a dose of 6.25 µg/mL. The acute toxicity phenotypes of zebrafish were shown in
Figure S1.
(
3) Assessing the effect of 4f on AlCl -medicated zebrafish. In order to evaluate
3
whether 4f could improve dyskinesia and reaction capacity on zebrafish model, the
2
5
distance and speed change experiments on AlCl -induced zebrafish AD were performed.
3
According to the data in Table 6, the MTC (maximum tolerated concentration) of
compound 4f was 1 μg/mL. And then three various concentrations for compound 4f group
(
0.11 μg/mL, 0.33 μg/mL and 1 μg/mL) were designed. Meanwhile, the experiments
contained donepezil group (DPZ) at a dose of 8.0 μM, normal zebrafish (untreated group)
and AlCl -mediated zebrafish AD (model group). As shown in Figure 13A~13B, the
3
distance of the model group obviously decreased to 5346 mm compared with the
untreated group (7299 mm, p < 0.01). The distance sharply climbed to 6477 mm (DPZ
group, p < 0.05) after treating with 8.0 μM donepezil compared with the model group,

and the efficacy was 57.9%. Moreover, treatment with compound 4f with three
concentrations 0.11 μg/mL, 0.33 μg/mL and 1 μg/mL, the distance gradually added up to
6
418 mm (p < 0.05), 6763 mm (p < 0.01) and 7024 mm (p < 0.05), respectively, and the
efficacy were 54.9%, 72.6% and 85.9%, respectively. In particular, the 0.33 μg/mL (6763,
p < 0.01, 72.6%) and 1.0 μg/mL (7024, p < 0.05, 85.9%) 4f groups demonstrated longer
distance than that of donepezil (6477, p < 0.05, 57.9%).
According to the results in Figure 13C~13D, speed change after light stimulus
alteration of model group sharply decreased to 1.08 mm/s compared with the untreated
group (2.32 mm/s, p < 0.001). While, the speed change added up to 1.88 mm/s after
adding 8.0 μM donepezil (DZP group, p < 0.01, 64.5% efficacy). Meanwhile, treatment
with compound 4f (0.11 μg/mL, 0.33 μg/mL and 1 μg/mL), the number gradually rose to
1
.79 (p < 0.05), 2.06 (p < 0.01) and 2.15 mm/s (p < 0.05), respectively, and the efficacy
were 57.3%, 79.0% and 86.3%, respectively. In particular, 4f at dose of 0.33 μg/ mL and
1
.0 μg/ mL displayed more efficacy than that of DPZ group.
Therefore, based on the results observed, 1.0 μg/mL compound 4f presented the best
dyskinesia recovery rate and response efficiency.
.17 Effect on Aβ1-40-mediated zebrafish vascular injury by 4f. Cerebral blood flow in
2
the grey matter can be reduced by over 40%, and indeed reduced cerebral blood flow is
the earliest biomarker of the development of AD. Recent studies displayed that
3
exogenous Aβ reduced cerebral blood flow, and increased Aβ aggregation. Thus,
protecting Aβ-induced vascular injury could be beneficial for the development of anti-AD

drugs.
Vascular injury model in zebrafish was built by treating with Aβ1-40, treatment with
compound 4f at the dose of 0.11 μg/ mL (1/9 MTC), 0.33 μg/ mL (1/3 MTC) and 1
3
0, 31
μg/mL (MTC).
The results were listed in Figure 13E, treatment with Aβ1-40, the
vascular injury rate sharply increased to 73.3% (model group) compared with the
untreated zebrafish (untreated group, 0). While treatment with compound 4f at given
concentration of 0.11 μg/mL, 0.33 μg/mL and 1 μg/mL, the injury rate gradually
decreased in a dose-independent manner. In particular, 1 μg/mL 4f showed the lowest
vascular injury rate (53.3%, p < 0.05) and the most efficacy (27.3%, p < 0.05), while,
0
.11 μg/mL 4f group (vascular injury rate 80%, p > 0.05) and 0.33 μg/mL 4f (vascular
injury rate 66.7%, p > 0.05) group indicated no significant difference compared with
model group, both of them showed -9.1% and 9.0% efficacy respectively. The phenotype
of 4f on Aβ1-40-induced vascular injury was presented in Figure S2. Thus, the screening
phenomenon presented that 4f (1.0 μg/mL) produced a significant protective effect on
vascular injury caused by Aβ1-40
.18 In vivo PET-CT imaging with [ C]4f in mice
To investigate the blood-brain-barrier (BBB) permeability ability of 4f, we further
.
11
2
11
40, 41
performed rodent PET-CT imaging in mice with [ C]4f.
1) Preparation of the precursor and radiosynthesis of [ C]4f. The Precursor 5 was
prepared from 4f via direct demethylation in the presence of BBr at ‒78 ℃ (Scheme 2).
11
(
3
11
11
For the [ C]4f synthesis, 4f was reacted at 110 °C with [ C]CH I in DMF (0.3 mL) in
3

2
the presence of NaOH. The reaction was consequently quenched with H O and the
radiolabelled crude product was purified via semi-preparative reversed-phase HPLC. It
was prepared in 30 to 35 minutes after the end of bombardment with 20–28%
radiochemical yields and >97% radiochemical purity with a specific activity ranged from
1
00 ± 20 mCi/μmol.
11
(
2) In vivo PET-CT imaging with [ C]4f in mice. In our PET-CT studies, as presented
11
in Figure 14, [ C]4f exhibited good BBB penetrability and had a fast uptake after
11
intravenous injection. Based on the analysis focused on the brain, the uptake of [ C]4f in
the brain reached to 8.76 % ID/cc at around 2 min and sustained binding during the 60
min scanning time.
11
The specificity of [ C]4f was investigated by PET imaging in mice with a 5.0 min
11
pretreatment of 4f (1 mg/kg) and Rivastigmine (1 mg/kg). We observed that [ C]4f
binding in brain was blocked after administration of 4f and Rivastigmine pretreated mice,
11
which demonstrated the binding selectivity and specificity of [ C]4f. In the 4f and
Rivastigmine pretreated mice, we found a similar binding with an approximate 45%
reduction compared with baseline, which suggested that 4f and rivastigmine have the
similar binding affinity to relative targets.
11
Next, the detailed distribution of [ C]4f in different regions of the brain was
4
2, 43
investigated with the mouse (Ma-Benveniste-Mirrione) VOI atlas
. Nine significant
brain functional regions (cerebellum, striatum, hippocampus, amygdala, thalamus, cortex,
brain stem, hypothalamus, and central gray) were selected to analyze. Form the data we

11
acquired, the distribution of [ C]4f in the 9 brain functional regions displayed similar
44
11
distribution patterns (Figure 15). Meanwhile, the uptake of [ C]4f in different brain
functional regions indicated significant reduce in the blocking study. The selected brain
regions of the mice using a 5.0 min pretreatment of 4f (1 mg/kg) and Rivastigmine (1
mg/kg) have a significant reduction of radioactivity absorption, which showed the
binding selectivity of 4f in those regions. It is also worth note that the regions closely
related to the regulation of memory and cognitive functions, such as the hippocampus and
striatum, have high radioactive absorption (~ 9 %ID/cc), which suggested 4f had good
targets binding affinity related to AD.
The distribution of major organs in mice was analyzed through the PBAS module in
PMOD (PMOD 4.003, PMOD Technologies Ltd., Zurich, Switzerland). From the data in
11
Figure 16, we could conclude that [ C]4f had the highest uptake in the heart (11.80 ±
0
0
.28 %ID/cc) at 5 min. In the brain (6.21 ± 0.61 %ID/cc), heart, and lungs (4.68 ±
.12 %ID/cc), maximum uptake was achieved at 5 min. Then the radioligand washout
11
from the brain regions, the highest uptake time point of [ C]4f in liver, kidney and spleen,
11
was later than other organs. The highest absorption time point of [ C]4f in liver and
kidney was at 30 min was 14.46 ± 2.55 %ID/cc and slightly declined in 60 min; while the
radioactivity uptake peaked in the spleen at 30 min was 11.49 ± 1.83 %ID/cc and then
washout slowly.
2
.19 In Vivo Assay
(1) Acute Toxicity: In order to further evaluate the safety profile of compound 4f, the

acute toxicity experiment at doses of 50, 100, 250, 500 and 1000 mg/kg was performed
using Kunming mice. There was no mortality changes of mice were monitored after
treating with the five groups of compound 4f. However, some abnormal behavior
presented at 250, 500 and 1000 mg/kg groups, such as dyspnea, decreased autonomic
activity, hypokinesia, lethargy but could get right due to vigilance, these symptoms could
be worsen as the time increased. The phenomenon of piloerection was observed after the
first 4h, but the mice could restore to the normal level after 24h. The above symptoms
were serious treating with 500 and 1000 mg/kg groups, presenting mild symptoms at 250
mg/kg group, and almost no symptoms were observed at 100 mg/kg or below. The
emerged symptoms could recover completely after 24h. After 7 days of compound
administration, all the experiment mice were sacrificed, and the heart, liver, spleen, lung
and kidneys were detected without damage in compound-treatment groups compared to
the vehicle control group. Therefore, compound 4f did not present any acute toxicity or
mortality immediately at dose up to 1000 mg/kg.
Scopolamine-induced amnesia had been widely used to assess lead compound against
3
1, 45
AD.
To test the ability of 4f to improve cognitive impairment, the Y-maze test was
carried out as previously described. As demonstrated in Figure 17A, the total arm entry
demonstrated that there was no significant difference between the untreated group, model
group (3 mg/kg scopolamine), 4f (5, 10 and 20 mg/kg) group and rivastigmine (Riva., 6
mg/kg) group. And the data exhibited that the agents had no potent effect on spontaneous
alternation behavior. While treated with 3 mg/kg scopolamine before evaluating, the

spontaneous alternation significantly decreased to 40.8% compared with the untreated
group (65.3%, p < 0.01). While, the spontaneous alternation gradually increased after
adding compound 4f (5 mg/kg, 10 mg/kg and 20 mg/kg; p > 0.05, p < 0.01, and p< 0.01,
respectively), in particular, the high dose of 4f (20 mg/kg) showed the highest
spontaneous alternation (61.6%) and demonstrated slightly higher alternation than that of
rivastigmine (6 mg/kg, 60.2%) (Figure 17B). The data displayed that 4f could improve
memory. Moreover, we compared 4f with the previous compound AC, compound AC did
not show significant improved cognitive performance at doses of 10 and 100 mg/kg,
while compound 4f could improve cognitive disorder at a dose of 20 mg/kg, exhibiting
the high hBChE inhibitor may lead to peripheric side effect and could not improve
cognitive performance.
In order to explore the mechanisms of 4f, the mice were executed after completing
above behavioral experiment, and then the AChE and choline acetyltransferase (ChAT)
vitality in hippocampi were assessed. The data were listed in Figure 17C~17D, treatment
with scopolamine alone (model group), the AChE vitality remarkably increased to 0.211
U/mg compared with the vehicle-treated mice (untreated group, 0.105 U/mg, p< 0.01).
Meanwhile, the AChE vitality obviously declined after adding compound 4f (5, 10 and 20
mg/kg). The high dose group (20 mg/kg) indicated the lowest AChE vitality (0.10 U/mg,
p < 0.01) and exhibited similar vitality compared with 6 mg/kg rivastigmine (0.098 U/mg,
p < 0.01) (Figure 17C). In addition, from the results in Figure 17D, treatment with
scopolamine alone (model group), the ChAT vitality remarkably declined to 28.8 U/g

compared with untreated group (59.1 U/g, p< 0.01). Meanwhile, treatment with 4f (5, 10
and 20 mg/kg), the ChAT vitality improved in a dose-dependent manner. Particularly, the
high dose group (20 mg/kg) demonstrated the highest ChAT vitality (54.1 U/g, p< 0.01)
and closed to the ChAT vitality of rivastigmine (6 mg/kg, 55.3 U/g). The phenomenon
exhibited that 4f presented precognitive effects by increasing brain cholinergic activity.
2
.20 Metabolism in vitro. Drug metabolism, a discipline in drug discovery and
4
6
development, served as an important role. So, the metabolism in vitro of compound 4f
was investigated using rat/human liver microsomes and rat intestinal flora.
(1) Liver microsomal metabolism. The rat and human liver microsomes were applied to
3
1
study the metabolites of 4f through UPLC G2-XS Qtof. The HPLC analysis for rat liver
microsomal metabolism of 4f was displayed in Figure S1 and [4f+H] peak with
3
95.19646 m/z (real time = 6.28 min, response values = 28145) was detected in Figure
S2A using low-energy collision-induced dissociation (CID). The metabolic pathway of 4f
and 3 metabolites were determined in Figure 18, the correspondent [4f-C H NO
4
7
(
dissociation) + O
3
+ Na] peak with m/z = 380.11215, [4f+O+H] peak with 411.19094
+O+H] peak were detected in Figure S2B~S2D,
m/z and [4f-CH (dissociation)+H O
2
2 2
which were further determined by the dissociation fragments by high-energy CID.
The HPLC analysis for human liver microsomal metabolism of 4f was indicated in
Figure S3, 4 metabolites were confirmed Figure 19, and the correspondent [4f+H] peak
with 395.19552 m/z, [4f-CH (dissociation) + H O +O+Na] peak with m/z = 453.16625,
2
2
2
[
4f-C H NO (dissociation) + O + Na] peak with m/z = 380.11223, [4f-C H N
4 7 3 9 9