Usnic acid derivatives as tau-aggregation and neuroinflammation inhibitors

Article abstract of DOI:10.1016/j.ejmech.2019.111961

Accumulation of tau protein aggregation plays a crucial role in neurodegenerative diseases, such as Alzheimer's disease (AD). Uncontrollable neuroinflammation and tau pathology form a vicious circle that further aggravates AD progression. Herein, we repor

Full text of DOI:10.1016/j.ejmech.2019.111961

Journal Pre-proof
Usnic acid derivatives as tau-aggregation and neuroinflammation inhibitors
Cun-Jian Shi, Wan Peng, Jin-Hua Zhao, Hua-Li Yang, Lai-Liang Qu, Cheng Wang,
Ling-Yi Kong, Xiao-Bing Wang
PII:
S0223-5234(19)31113-4
DOI:
https://doi.org/10.1016/j.ejmech.2019.111961
EJMECH 111961
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 21 November 2019
Revised Date: 9 December 2019
Accepted Date: 9 December 2019
Please cite this article as: C.-J. Shi, W. Peng, J.-H. Zhao, H.-L. Yang, L.-L. Qu, C. Wang, L.-Y. Kong, X.-
B. Wang, Usnic acid derivatives as tau-aggregation and neuroinflammation inhibitors, European Journal
of Medicinal Chemistry (2020), doi: https://doi.org/10.1016/j.ejmech.2019.111961.
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.
©
2019 Published by Elsevier Masson SAS.

Graphical abstract:

Usnic Acid Derivatives as Tau-Aggregation and Neuroinflammation Inhibitors
Cun-Jian Shi, Wan Peng, Jin-Hua Zhao, Hua-Li Yang, Lai-Liang Qu, Cheng Wang,
Ling-Yi Kong*, Xiao-Bing Wang *
Jiangsu Key Laboratory of Bioactive Natural Product Research and State Key
Laboratory of Natural Medicines, Department of Natural Medicinal Chemistry,
School of Traditional Chinese Pharmacy, China Pharmaceutical University, Nanjing
2
10009, China
*
Corresponding Authors.
Tel/Fax: +86-25-83271405;
E-mail: xbwang@cpu.edu.cn (Xiao-Bing Wang); cpu_lykong@126.com (Ling-Yi
Kong)
1

Abstract
Accumulation of tau protein aggregation plays a crucial role in neurodegenerative
diseases, such as Alzheimer’s disease (AD). Uncontrollable neuroinflammation and
tau pathology form a vicious circle that further aggravates AD progression. Herein,
we reported the synthesis of usnic acid derivatives and evaluation of their inhibitory
activities against tau-aggregation and neuroinflammation. The inhibitory activity of
the derivatives against the self-fibrillation of the hexapeptide AcPHF6 was initially
screened by ThT fluorescence assay. Using circular dichroism and transmission
electron microscopy, compound 30 showed the most potent inhibitory activity against
AcPHF6 self-fibrillation. Compound 30 was further confirmed to inhibit the
aggregation of full-length 2N4R tau protein by a heparin-induced mechanism. In
addition, we investigated the anti-inflammatory activity of compound 30, and showed
that compared with sodium usnate, it reduced NO release in LPS-stimulated mouse
microglia BV2 cells. More importantly, 30 showed significant protective effects
against okadaic acid-induced memory impairment in rats. Thus, 30 was a novel
tau-aggregation and neuroinflammation inhibitor that represented a potential
therapeutic candidate for AD.
Keywords: Alzheimer’s disease, Usnic acid derivatives, Tau anti-aggregation,
Anti-inflammatory
2

1
. Introduction
Alzheimer’s disease (AD), a neurodegenerative disease and the most common
form of dementia, is characterized by senile plaques formed by amyloid-β (Aβ)
outside neurons in the brain and neurofibrillary tangles (NFTs) consisting of
hyperphosphorylated-tau inside neurons [1, 2]. This abnormal accumulation and
misfolding of proteins exacerbate the damage and death of neurons [3, 4]. In addition,
oxidative stress [5], neuroinflammation [6], metal ion disorders [7], and low level of
abnormal deposition acetylcholine (ACh) [8] can also lead to the pathology of AD.
Owing to the complexity of the pathological mechanism of AD, researchers propose
that the cure of a single target is not realistic [9]. Although multitarget-directed
ligands (MTDLs) have made great progress [10], there are still no drugs available for
treating AD. Obviously, selection of appropriate targets is of utmost importance.
Aβ hypothesis still occupies a predominant place in pharmaceutical product
pipelines (either single or multitarget). According to a scientific data list, in 2018,
drugs addressing Aβ targets accounted for 54% of the AD drugs in phase III clinical
trials [11]. Nevertheless, all drugs targeting Aβ have failed, and this hypothesis has
been questioned [12]. One explanation is that Aβ may not be particularly harmful in
itself, but induces tau hyperphosphorylation and enhances tau-seeded pathology, and
finally exerts further negative effects [13]. Some findings have suggested that the
pathological development of tau is more related to the degree of cognitive impairment
than to Aβ [14, 15]. In addition, tau imaging is also superior to Aβ imaging in
predicting the progression and status of AD [16, 17]. All these findings indicate that
3

tau is a more meaningful target of AD than Aβ.
Tau, a microtubule-associated protein predominantly expressed in nerve cells, has
an important role in microtubule assembly and stabilization [18, 19]. In fact, natively
unfolded tau is highly soluble and shows little tendency for aggregation. In
pathological conditions, both kinase activation (e.g. glycogen synthase kinase-3β,
GSK-3β) and phosphatase inhibition (e.g. protein phosphatase 2A, PP2A) are able to
hyperphosphorylate tau, which detach from microtubules easily and tends to
aggregate into neurotoxic NTFs [4, 20-22]. Correspondingly, microtubules are
depolymerized and subsequently lose their function owing to the shedding of tau [23].
Once the tau aggregation cascade is initiated, the process begins to self-replicate.
Unlike Aβ, proteopathic tau seeds are transferred from a “donor cell” to a “recipient
cell” in a prion-like manner, where the seeds transform normal tau into aggregates
(new seeds), driving the progression of neuronal degeneration and death [24-26]. The
overall level of NFTs is closely related to degree of dementia [27, 28].
Inhibition of tau aggregation is a suitable therapeutic strategy, and considerable
progress has been made. There are currently two major approaches for addressing tau
aggregation. One is to search for protein kinase inhibitors that inhibit tau
hyperphosphorylation to prevent tau aggregation [29]. The other is to search for direct
inhibitors of the tau aggregation process [30]. However, no matter which one is
targeted alone, the tau cascade will not be easily broken. Therefore, the relationship
between neuroinflammation and tau pathology has received increasing attention [31,
3
2]. Not only that, systemic inflammation also provides a non-negligible contribution
4

to AD [33]. Although inflammation may not be the initiating factor of AD, the vicious
circle between it and neuronal lesions lead to increased neuronal degeneration [34, 35].
Aggregated
tau
activates
NLRP3-ASC
inflammasome
to
aggravate
neuroinflammation, and the occurrence of neuroinflammation not only increases
abnormal tau hyperphosphorylation but also aggravates tau-seeded tau pathology
[
36-38]. Therefore, we aimed to develop anti-neuroinflammatory drug that can inhibit
the tau aggregation process.
Natural products play a leading role in the discovery of drugs for the treatment of
human diseases. Usnic acid (Figure 1), a dibenzofuran derivative, is a high-level
secondary metabolite in lichen [39]. Usnic acid has various biological characteristics
owing to its special skeleton. Previous studies have shown that usnic acid has
anticancer, antioxidant, anti-inflammatory, antiviral, and anti-pyretic activities [40].
Especially, the anti-inflammatory ability of usnic acid is potent. Different biological
activities of usnic acid may be determined by its various fragments. The "triketone"
moiety of usnic acid is an important fragment contributing to its bactericidal,
anticancer, and cytotoxic activities [41], but it has little to do with the activities we
need for the treatment of AD. Thus, we modified this (+) - moiety of usnic acid with
the hope that the derivatives retain its anti-inflammatory activity and confer its ability
to inhibit tau protein aggregation. Considering the poor water solubility of usnic acid,
we used (+) - sodium usnate (Figure 1) to improve its water solubility and
bioavailability [42], and as a reference.
5

Figure 1. Chemical structures of (+) - usnic acid and (+) - sodium usnate.
Herein, two series of compounds, compounds 1-25 as series A and 26-50 of series
B, were synthesized and screened for AcPHF6 self-aggregation inhibitory activity by
ThT assay. An optimal compound, 30, was identified through CD spectrum and TEM
image analyses. Noticeably, 30 also displayed an inhibitory activity against full-length
tau aggregation. Further, 30 was initially evaluated for its anti-neuroinflammatory
activity in BV-2 cells and for behavioral performance in a rat model of AD.
2
. Results and Discussion
2
.1 Chemistry
Usnic acid derivatives were prepared according to previously published methods
[
43, 44]. Enamine compounds 1, 4-25 were prepared through condensation of usnic
acid with primary amines or amino acids (Scheme 1). Notably, the basicity of the
amino group also affects the formation of an enamine bond with usnic acid. For
example, 2-aminoquinoline and 2-aminothiazole cannot react with usnic acid under
ordinary catalytic conditions, whereas 8-aminoquinoline can. In addition, pH (9.5-10)
control is extremely important in the reaction of usnic acid with amino acids for
preparing of 2 and 3. The target compounds 26-50 were synthesized from the reaction
of usnic acid with the corresponding hydrazines and hydrazides (Scheme 1).
6

Interestingly, under the same conditions, aliphatic hydrazines and usnic acid did not
react to form pyrazole rings. The structures of all target compounds were
1
13
characterized by H NMR, C NMR, and HRMS.
Scheme 1. Synthesis of compounds 1-50. Reagents and conditions: (a) EtOH, 80 °C,
o
2
-4 h. (b) EtOH/H O, KOH (pH~9.5), 80 C, 4 h.
2
2
.2 Biological Assays
2
.2.1 ThT fluorescence assay
AcPHF6 (Ac- VQIVYK -NH ) is a short peptide in the third repeat of the
3
06
311
2
microtubule-binding region of tau protein. Studies have shown that AcPHF6 plays a
dominant role in the initiation of tau aggregation [45]. As AcPHF6 forms a cross-β
structure fiber spontaneously in vitro, it is a suitable model to study small-molecule
inhibitors of tau aggregation [46]. Polyphenolic flavone myricetin and dye orange G
(Figure 2) are well-known inhibitors of AcPHF6 [47, 48]. Aurones (Figure 2), a
subclass of the naturally occurring flavonoids, exhibited comparable inhibitory
7

activity to myricetin [47]. And 5-nitro-a-cyanocarboxamide derivatives of caffeic acid
Figure 2) showed a higher inhibitory activity (90% inhibition) at 50 µM [49]. In
(
addition, peptides and macrocyclic peptides (Figure 2) based on the structure of
AcPHF6 also effectively reduced the aggregation process of AcPHF6 [50, 51].
Figure 2. Chemical structures of AcPHF6 aggregation inhibitors: myricetin; orange G;
aurones; caffeic acid derivatives; pepetid D-TLKIVW and macrocyclic peptides.
In the screening phase, we evaluated the ability of the target compounds to inhibit
amyloid formation using ThT fluorescence assay. When ThT interacts with
β-sheet-rich amyloid fibrils, its excitation and emission maxima are red-shifted to 440
and 490 nm, respectively [52]. In general, emission intensity at 490 nm reflects the
amount of amyloid fiber. The kinetics of AcPHF6 aggregation (100 µM) in the
absence of inhibitors was assessed in 50 mM phosphate buffer in the presence of ThT
(20 µM). As reported previously [48], AcPHF6 aggregates rapidly at 25 °C and
reaches a fluorescence maximum after an average time of 40 min (Figure 3). On the
8

basis of the above conditions, we tested the inhibitory ability of usnic acid derivatives
at a concentration of 10 µM with myricetin as control. Excitingly, 38 of the 50
compounds exhibited inhibitory activity (Table 1, 2). Among them, compounds 18
and 30 were the most active compounds, with inhibition rates of 58% and 52%,
respectively.
In series A, compounds 1, 2, and 3 were reported and used in antiviral and
bactericidal applications. Same as sodium usnate, 1 did not exhibit inhibitory effect on
tau aggregation. Interestingly, the introduction of different amino acids to usnic acid
resulted in significant differences in activity, as evident for 2 and 3. The tryptophan
and carbonyl groups on both sides of compound 2 may constitute an important active
group, which was similar to the NQTrp reported by M. Frenkel-Pinter et al. [53, 54].
Addition of long chains (4) did not improve the inhibitory activity of sodium usnate.
After the addition of poly-substituted phenols, hydroxyl substituents in the ortho
position of enamine (5-8) show a certain inhibitory activity. However, when the
hydroxyl group is in its para position (9-11), there is no activity regardless of the
steric hindrance around the hydroxyl group. Compounds 12-16 showed that strong
electron-donating and strong electron-withdrawing groups brought significant
contribution to the activity observed, and that the position of the N, N-dimethyl group
in meta (16) was less important than that in para (15), but the increase in steric
hindrance (14) did not further increase the inhibitory activity. The addition of
heterocycles indicated that small moieties, such as pyridine or pyrazine, as in 17-20,
were significantly better than the bulky moieties 21-25 in testing. In contrast to phenyl,
9

pyridine as a substituent lead to the hydroxyl group in the para position of the
enamine bond (18) exhibiting the best tau-inhibitory activity. Correspondingly, the
replacement of the amino group at the same position (19) reduced the activity,
showing the importance of the hydroxyl group.
In series B, the starting compound, 26, had a moderate tau-inhibitory capacity,
which may be related to its high planarity. Addition of fluorophenyl, tert-butylphenyl,
methoxyphenyl, or benzonitrile, as in 27-28, 29, 34-35, and 44, seemed not to affect
their activity, compared with that of 26. However, different substituted benzoic acid
groups (30-32) and 4-benzenesulfonamide (37) resulted in a good inhibitory activity.
Compared with that of 31 and 32, the tau-inhibitory activity of 30 was successively
increased, suggesting that the position of the carboxyl group was more beneficial in
the para position than that in the meta and ortho positions. However, replacing a
carboxyl group with an ethyl ester group (33) greatly reduced tau-inhibitory activity,
which was consistent with the result of replacing a sulfonamide group (37) and a
methylsulfone group (38). The introduction of heterocycles was only moderately
improved by pyridazine (42) and pyrazine (43), and most of the other heterocycles (38,
4
0-41, and 45-47) even showed decreased inhibitory activity compared to that of 26.
Notably, when a sulfo group was added between the pyrazole ring and the benzene
ring, as in 48-50, the compounds did not exhibit inhibitory activity regardless of the
substituent on the phenyl ring, which may be attributed to the sulfo group blocking
the conjugation and the planar structure.
1
0

Table 1 Effect of compounds 1-25 on AcPHF6 self-aggregation
a
Compd.
1
R
Inhibition%(10µM) Compd.
R
Inhibition%(10µM)
b
n.a.
47.6±16.4
n.a.
14
15
16
17
18
19
20
21
22
23
48.2±1.7
2
3
4
5
6
7
8
9
48.1±3.6
11.3±8.8
48.0±7.6
58.3±4.6
46.4±3.4
44.2±4.7
n.a.
n.a.
22.0±12.0
48.4±12.2
12.7±9.7
25.3±20.1
n.a.
n.a.
1
0
n.a.
39.2±18.4
1
1
n.a.
24
25
27.6±13.5
29.8±9.5
1
1
2
3
5.4±2.0
27.6±13.4
85.9±1.3
Myricetin
Sodium Usnate
n.a.
a
The inhibitory rates are shown as the mean ± SD from three separate experiments,
1
1

each run in triplicate.
b
n.a. means no activity.
Table 2 Effect of compounds 26-50 on AcPHF6 self-aggregation
Compd.
R
Inhibition%(10µM) Compd.
R
Inhibition%(10µM)
2
2
2
2
3
3
3
6
7
8
9
0
1
2
22.5±9.4
30.3±5.1
26.9±9.0
22.3±10.4
51.7±3.6
40.4±6.1
32.8±4.2
39
40
41
42
43
44
45
29.8±5.2
17.1±8.5
7.7±2.8
F
40.9±5.0
38.0±3.9
20.1±9.4
12.2±9.2
3
3
14.5±4.8
46
n.a.
3
3
3
3
3
4
5
6
7
8
27.2±3.2
24.0±1.5
28.7±9.8
46.8±1.1
8.5±7.7
47
48
49
50
9.3±1.3
n.a.
n.a.
n.a.
Myricetin
85.9±1.3
Sodium Usnate
n.a.
1
2

Figure 3 shows the aggregation kinetics of AcPHF6 in the presence of the
optimal compounds 18 and 30. Indeed, the spectral properties of ThT combined with
fiber aggregates are influenced by a number of factors [52], such as the nature of the
compound, composition of the medium, and protein-to-dye ratios. Therefore, to
exclude false-positives that may be generated in ThT fluorescence assay, we further
verified the activity of 18 and 30 using circular dichroism spectroscopy.
Figure 3. Kinetics of 100 µM AcPHF6 aggregation in the absence (blue point) and
presence of 10 µM 18 (purple point) and 30 (red point), as monitored by ThT
fluorescence.
2
.2.2 Circular dichroism (CD) spectroscopy
Based on the experimental conditions of the ThT assay, circular dichroism was
used to further elucidate the degree of AcPHF6 aggregation inhibition by usnic acid
derivatives 18 and 30 (Figure 4). Owing to the morphological diversity of
1
3

β-sheet-rich structures, we did not quantify the content of the secondary structure by
CD spectrum [55]. The CD spectrum of AcPHF6 in the absence of an inhibitor (black
line) showed a negative band at 221 nm and a positive band at 200 nm, which is a
typical feature of the β-sheet structure [56]. A decrease in intensity was observed for
the CD spectrum of AcPHF6 in the presence of 18 (blue line) and 30 (red line),
suggesting that the abundance of β-sheet structures was lower. Unexpectedly, the
activity of 30 was significantly higher than that of 18, and this result was inconsistent
with the ThT fluorescence result that the inhibitory activity of 18 was slightly higher
than that of 30.
Figure 4. CD spectra obtained after incubation of 100 µM AcPHF6 for 2 h at 25 °C in
the absence (black) or presence of 10 µM of 18 (blue) and 30 (red).
In addition, the combination of compounds with amyloid fibers in the form of
probes rather than inhibitors can also reduce the intensity of the CD spectrum [47].
1
4

Thus, we used transmission electron microscopy (TEM) to verify our findings.
2
.2.3 Transmission electron microscopy
TEM allows visual observation of the aggregation pattern and progress of
amyloid fibers [57]. We first observed the fiber morphology of AcPHF6
spontaneously accumulated in phosphate buffer for 2 h, as shown in Figure 5A. The
hexapeptide aggregated into a network of filament fibers. Although the addition of 18
(Figure 5B) led to the fiberation process of AcPHF6, the long, twisted fiber structure
obviously still existed. However, we found an encouraging result that, as shown in
Figure 5C, 30 exerted strong inhibitory activity against AcPHF6 aggregation, with
almost no fibrillar material was observed in the field of view, and finally a dot-like
structure (<5 nm) was formed.
Figure 5. TEM images of inhibition of AcPHF6 self-aggregation: (A) AcPHF6 (200
µM) alone; (B) AcPHF6 (200 µM) + 18 (20 µM) and (C) AcPHF6 (200 µM) + 30 (20
µM).
Taken together, CD data and TEM results confirmed the inhibitory capacity of
1
5

usnic acid derivatives 18 and 30, with 30 as the most effective among the two
derivatives. Therefore, we considered 30 the best aggregation inhibitor of AcPHF6
and used it for further studies.
2
.2.4 Full-length tau aggregation and inhibition studies
The adult human brain contains six major tau isoforms [19]. We used the most
representative full-length 2N4R tau to verify the inhibitory ability of 30. Indeed,
owing to the lack of post-translational modifications, recombinant tau protein does not
substantially undergo self-aggregation under physiological conditions. Addition of
anionic cofactors, such as heparin and arachidonic acid, is a method that promotes the
use of recombinant tau aggregation [58]. Here, we evaluated the effect of 30 (10 µM)
on the tau aggregation process under heparin induction, which was monitored by ThT
assay. Figure 6 shows that treatment of full-length tau protein with 30 significantly
slowed down the aggregation process and reduced the amount of formed aggregates
(50% inhibition). Importantly, this result corresponded to the inhibition of AcPHF6
self-aggregation by 30, confirming the potential of 30 as a tau anti-aggregation
compound.
1
6

Figure 6. Kinetics of 30 activity on heparin-induced full-length 2N4R tau aggregation,
as monitored by ThT fluorescence.
2
.2.5 Neuro- and hepatotoxicity assessments
Considering the toxicity of usnic acid [59], we tested the cytotoxicity of its
derivative, 30, in vitro with the hope that its safe range would be acceptable. Human
neuroblastoma SK-N-SH (SH-SY5Y) cells and human hepatocyte (LO2) cells are
suitable models for detecting neurotoxicity and hepatotoxicity of compounds. First,
SH-SY5Y cells were treated with 30 at concentrations of 10 to 40 µM for 24 h, and
cell viability was assessed by the MTT test. Figure 7A shows that 30 at 30 µM
reduced the survival of SH-SY5Y cells. Similarly, LO2 cells were incubated with 30,
and as show in Figure 7B, 30 exerted no significant hepatotoxicity even at high
concentrations (up to 40 µM).
1
7

Figure 7. (A) Viability of SH-SY5Y cells was determined by the MTT assay after 24
h of incubation with various concentrations of 30. (B) Viability of LO2 cells was
determined by the MTT assay after 24 h of incubation with various concentrations of
3
0. The results are expressed as a percentage of control cells. Values are reported as
the mean and SD of three independent experiments, each run in triplicate.
2
.2.6 Inhibition of NO release studies
Excessive activation of microglia plays a crucial role in neuronal damage and
death caused by neuroinflammation in AD [60]. Some studies revealed that tau
pathology is prone to trigger activation of microglia [61]. Correspondingly, the
inflammatory factors released by activated microglia can also aggravate tauopathies,
either by inducing tau aggregation or hyperphosphorylation [32, 36]. Therefore,
inhibition of neuroinflammatory cytokine production is considered an effective
strategy for the treatment of AD.
NO release is an important feature of microglial activation, and it may lead to an
uncontrolled increase in neuroinflammation [62, 63]. In a culture medium, NO is
rapidly oxidized to nitrite, and nitrite (NO ) concentration is an indicator of NO
2
1
8

production, which can be detected by the Griess assay. Before this, we determined the
cytotoxicity of 30 and sodium usnate against the viability of BV2 cells by using MTT
assay, and the results showed that 30 at a concentration of 20 µM exerted no effect on
BV2 cell viability (Figure 8A), whereas sodium usnate at the same concentration
exhibited acute cytotoxic effect (Figure 8B). Thus, we investigated whether 30 exerts
anti-inflammatory activity without affecting cell viability, with 10 µM sodium usnate
as a control. As shown in Figure 8C, 30 at a concentration of 10 µM retained the
anti-inflammatory activity of sodium usnate and inhibited NO release rate by 41%.
Figure 8. Viability of BV2 cells was determined by the MTT assay after 24 h of
incubation with various concentrations of 30 (A) and sodium usnate (B). The results
are expressed as a percentage of control cells. (C) Effect of 30 on LPS-stimulated
production of the inflammatory mediator NO in BV-2 cells, compared with that of
sodium usnate. Griess assay was used to detect NO production. Results are expressed
as percent of cells treated with LPS alone. All values are reported as the mean ± SD of
three independent experiments, each run in triplicate.
1
9

2
.2.7 Compound 30 ameliorates learning and memory impairments in AD model rats
Okadaic acid (OA)-induced memory impairment in rats is a suitable model for
assessing the therapeutic efficiency of anti-AD agents in vivo. OA is a selective
inhibitor of the serine/threonine phosphatases 2 (PP2A), and thereby induces tau
hyperphosphorylation [64]. We injected OA to the ipsilateral hippocampus of rats to
establish AD models after 7 days of intraperitoneal injection of 30 (at 5 and 10 mg/kg)
or saline. The rats develop dementia after 3 days of OA injection into the brain. After
OA injection was completed, the rats continued to be intraperitoneally administered
3
0 for 7 days, and the Morris water maze test was carried out to evaluate the
cognition-improving effect of 30.
During the experimental period, 30 exerted no acute toxicity symptoms, such as
mortality or abnormal body weight change. Figure 9A shows changes in body weight
of rats injected intraperitoneally with 30 for 14 days. During the 5-day training, the
average escape latency and search distance of each group of mice gradually decreased,
but model rats usually spent more time and walked further distances to find the hidden
platform. Figure 9B shows the mean values of the escape latencies to the hidden
platform at day 1 to 5. Compared with that on day 1 of training, the average escape
time of model-operated rats on day 5 decreased from 50.0 to 41.8 s, whereas the
average escape time of control-operated rats decreased from 37.2 to 12.5 s over the
course of the same training days. Compared with the control group, the high-dose (10
mg/kg) 30-treated group had a shorter escape latency than the low-dose (5 mg/kg)
3
0-treated group, showing an average escape time from 45.9 to 18.5 s. At the same
2
0

time, we also recorded the average swimming speed of each group of rats in the maze.
There was no difference in the average swimming speed (Figure 9C) between the
groups, indicating that intake of 30 during the experiment did not affect the rats’
normal physiology activity. Furthermore, in the probe trial on day 6, after the platform
was removed, the time of rats staying in the target quadrant and the number of
crossing the virtual platform (the original platform location) were significantly higher
in the high-dose and control groups than in the model group (Figure 9D,E).
Representative trajectories of the rats in the Morris water maze during the spatial
probe trial period (Figure 9F) showed purposeful and regular search for hidden
platforms in the treatment and control group rats, but showed random motion paths in
model rats. These results further revealed that 30 administration led to a substantial
improvement of the conventional reference spatial memory and cognitive abilities of
rats.
2
1

Figure 9. Compound 30 attenuates OA-induced spatial learning and memory deficits
in the training session of Morris water maze task. Data are presented as the mean ±
ns
SD (n = 8); statistical significance was evaluated by two-way ANOVA: p > 0.05, *p
<
0.05, **p < 0.01 and ***p < 0.001 compared with the model group. Error bars
represent standard error of the mean. (A) Mean daily body weight profile of the
control, 5 mg/kg 30, and 10 mg/kg 30 groups during the 14-day drug administration
period. (B) Escape latency time of each group was counted on day 1 to 5 during the
training period. (C) Average swimming speed of rats in each group for 5 days. (D)
Time spent in the virtual platform quadrant. (E) Number of virtual platform (the
original platform location) crossings. (F) Representative tracks of rats in the Morris
2
2

water maze during the spatial probe trial period. The blue solid coil represents the
range of the maze, and the green dotted circle represents the position of the virtual
platform. The red square point and the blue square point are the starting and ending
positions of the rat, respectively.
3
. Conclusion
In summary, two series of usnic acid derivatives were synthesized and tested for
their tau-aggregation inhibitory activities. We used the tau-derived hexapeptide
AcPHF6 model, a fragment that mainly promotes tau aggregation, to evaluate the
activity of the derivatives. Among them, compounds 18 and 30 were the most
prominent, with inhibitory activities of 58% and 52%, respectively. As the ThT
fluorescence assay was affected by many factors, there might have been a certain
degree of false positives in the results. Therefore, we used CD spectrum and TEM
assays to determine whether 30 inhibits AcPHF6 self-polymerization, and TEM
results showed that 30, compared with the control, stopped the formation of AcPHF6
filaments and promoted the formation AcPHF6 dot structures. Encouragingly, 30 also
exerted excellent inhibitory activity against the aggregation of full-length 2N4R tau
protein. In addition, 30 at concentrations below 20 µM showed no apparent
cytotoxicity to SH-SY5Y cells and mouse microglia BV2 cells, and it also showed
low hepatotoxicity even at a high concentration of 40 µM. Moreover, 30 at 10 µM
significantly reduced NO release by LPS-induced BV2 cells, indicating that it had
excellent anti-neuroinflammatory activity. Most importantly, the results of the Morris
2
3

water maze test showed that 30 enhanced the cognitive ability of OA-induced AD
model rats. All these results highlighted 30 as a potential candidate for the
development of anti-AD drugs.
4
. Experimental section
4
.1 Chemistry
All conventional reagents and solvents were purchased directly from commercial
companies and without further purification. Reaction progress was monitored using
analytical thin layer chromatography on precoated silica gel GF254 plates (Qingdao
Haiyang Chemical Plant, Qingdao, China) and the spots were detected under UV light
(254 nm). After post-treatment of the reaction, the product was separated by rapid
1
purification preparative liquid chromatography (Biotage, Isolera One, Sweden). H
1
3
NMR and C NMR spectra were measured with a Bruker ACF-500/600 spectrometer
at 25 °C and referenced to TMS. The residual solvent line was designated as an
internal standard and the chemical shift was reported in ppm (δ). High-resolution
electrospray ionization (HRESI) mass spectra were carried out using an Agilent
6
4
4
520B Q-TOF mass spectrometer (Agilent Technologies, Santa Clara, CA, USA).
.2 Synthesis
.2.1 Preparation of compound 1
(+)-Usnic acid (1 equiv.) and concentrated ammonium hydroxide (10 equiv.) were
added to absolute EtOH under nitrogen, and the reaction mixture was stirred at 80 °C
for 2 h. After the reaction was completed, the solution was cooled to 25 °C and then
2
4

acidified with 1 N hydrochloric acid. The mixed liquid was extracted twice with ethyl
acetate. The combined extracts were then washed with H O and brine, and dried over
2
anhydrous Na SO . After concentration under reduced pressure, the residue was
2
4
quickly purified by using a silica gel column (PE/EtOAc).
R,E)-6-acetyl-2-(1-aminoethylidene)-7,9-dihydroxy-8,9b-dimethyldibenzo[b,d]furan-
,3(2H, 9bH)-dione (1).
Yield 72%, pale yellow powder. m.p. 255-256 °C. H NMR (500 MHz,
DMSO-d ) δ 13.39 (s, 1H), 12.29 (s, 1H), 11.54 (bs, 1H), 9.80 (bs, 1H), 5.85 (s, 1H),
(
1
1
6
1
3
2
.63 (s, 3H), 2.53 (s, 3H), 1.97 (s, 3H), 1.63 (s, 3H). C NMR (126 MHz, DMSO-d )
6
δ 200.8, 197.5, 188.4, 175.7, 172.9, 162.4, 157.6, 155.6, 106.2, 105.0, 102.5, 101.2,
-
1
3
4
00.8, 56.0, 31.5, 30.9, 24.4, 7.4. HRMS (ESI) m/z 342.0983 [M - H] (calcd for
42.0986, C H NO ).
18
16
6
.2.2 General procedures for the preparation of 2-3
The reaction of (+) - usnic acid with amino acid was carried out as previously
described [65], but with some modifications. Amino acid (3 equiv.) was suspended in
aqueous ethanol (1:1, v/v) and KOH was added in portions while stirring until the
o
amino acid was completely dissolved at 80 C. (+) - Usnic acid was added to the
clarified mixture and the pH was adjusted to 9.5 with KOH. The mixture was heated
until the reaction was complete. Solution was cooled to room temperature and was
acidified with 1 N hydrochloric acid. The mixed liquid was extracted twice with ethyl
acetate, and the combined extracts were washed with H O and brine, dried over
2
anhydrous Na SO . After concentration under reduced pressure, the residues were
2
4
2
5

quickly purified by silica gel column (PE/EtOAc).
S)-2-(((E)-1-((R)-6-Acetyl-7,9-dihydroxy-8,9b-dimethyl-1,3-dioxo-1,9b-dihydrodibe
(
nzo[b,d]furan-2(3H)-ylidene)ethyl)amino)-3-(1H-indol-3-yl)propanoic acid (2).
1
Yield 80%, yellow powder. m.p. 150-152 °C. H NMR (600 MHz, DMSO-d ) δ
6
1
3.64 (bs, 1H), 13.41 (s, 1H), 13.37 (bs, 1H), 12.13 (s, 1H), 10.99 (s, 1H), 7.46 (dt, J
8.2, 0.9 Hz, 1H), 7.33 (dt, J = 8.2, 0.9 Hz, 1H), 7.13 (d, J = 2.4 Hz, 1H), 7.05 (ddd,
J = 8.0, 6.9, 1.1 Hz, 1H), 6.94 (ddd, J = 8.0, 6.9, 1.1 Hz, 1H), 5.84 (td, J = 7.8, 4.8 Hz,
=
1
3
1
1
5
H), 5.05 (m, 1H), 3.30-3.40 (m, 2H), 2.64 (s, 3H), 2.37 (s, 3H), 1.97 (s, 3H), 1.62 (s,
1
3
H). C NMR (126 MHz, DMSO-d ) δ 200.8, 197.4, 188.6, 174.3, 172.8, 171.1,
6
62.4, 157.5, 155.7, 136.0, 127.0, 124.4, 121.0, 118.5, 118.0, 111.3, 107.5, 106.2,
05.0, 102.3, 101.7, 100.8, 56.9, 56.2, 31.5, 30.9, 28.3, 18.5, 7.4. HRMS (ESI) m/z
-
29.1616 [M - H] (calcd for 529.1612, C H N O ).
2
9
26
2
8
(S)-2-(((E)-1-((R)-6-Acetyl-7,9-dihydroxy-8,9b-dimethyl-1,3-dioxo-1,9b-dihydrodibe
nzo[b,d]furan-2(3H)-ylidene)ethyl)amino)-3-phenylpropanoic acid (3).
1
Yield 86%, pale yellow powder. m.p. 239-240 °C. H NMR (600 MHz,
DMSO-d ) δ 13.72 (bs, 1H), 13.40 (s, 1H), 13.37 (bs, 1H), 12.10 (s, 1H), 7.27-7.19
6
(m, 5H), 5.88 (s, 1H), 5.09 (td, J = 7.6, 4.9 Hz, 1H), 3.28-3.14 (m, 2H), 2.64 (s, 3H),
1
3
2
1
1
.38 (s, 3H), 1.97 (s, 3H), 1.64 (s, 3H). C NMR (126 MHz, DMSO-d ) δ 201.4,
6
98.2, 189.2, 175.1, 173.4, 171.2, 163.0, 158.1, 156.2, 135.9, 130.0, 128.9, 127.6,
06.8, 105.6, 102.9, 102.3, 101.4, 57.8, 56.9, 38.5, 32.1, 31.5, 19.0, 8.0. HRMS (ESI)
-
m/z 490.1507 [M - H] (calcd for 490.1511, C H NO ).
2
7
24
8
4
.2.3 General procedures for the preparation of 4-25
2
6

(+)-Usnic acid (1 equiv.) and primary ammonia compounds (1.1 equiv.) were
o
added in absolute EtOH under nitrogen and the reaction mixture was stirred at 80 C
for 4 h. For the hydrochloride salt materials, they (1.1 equiv) were added to absolute
ethanol, and an equivalent amount of pyridine (1.1 equiv) was added dropwise thereto
and stirred at 80 °C under nitrogen. After 5 minutes, (+) - usnic acid was added and
the reaction mixture was stirred for another 4 h. After reaction was complete, the
reaction solution was cooled and concentrated under reduced pressure. The residues
were quickly purified by silica gel column (PE/EtOAc).
(R,E)-6-acetyl-7,9-dihydroxy-2-(1-((5-hydroxypentyl)amino)ethylidene)-8,9b-dimeth
yldibenzo[b,d]furan-1,3(2H,9bH)-dione. (4).
1
Yield 79%, pale yellow powder. m.p. 127-129 °C. H NMR (500 MHz,
DMSO-d ) δ 13.39 (s, 1H), 13.05 (t, J = 5.3 Hz, 1H), 12.32 (s, 1H), 5.86 (s, 1H), 4.38
6
(t, J = 5.1 Hz, 1H), 3.54 (t, J = 6.7 Hz, 2H), 3.42 (q, J = 6.2 Hz, 2H), 2.64 (s, 3H),
1
3
2
.59 (s, 3H), 1.97 (s, 3H), 1.65(m, 5H), 1.53-1.44 (m, 2H), 1.43-1.37 (m, 2H). C
NMR (126 MHz, DMSO-d ) δ 201.4, 197.7, 189.5, 175.3, 173.5, 163.0, 158.2, 156.2,
6
1
8
06.8, 105.6, 102.8, 102.0, 101.4, 60.9, 56.8, 44.0, 32.4, 32.2, 31.5, 28.6, 23.4, 18.6,
-
.0. HRMS (ESI) m/z 428.1715 [M - H] (calcd for 428.1730, C H NO ).
2
3
27
7
(R,E)-6-acetyl-2-(1-((5-chloro-2-hydroxyphenyl)amino)ethylidene)-7,9-dihydroxy-8,9
b-dimethyldibenzo[b,d]furan-1,3(2H,9bH)-dione. (5).
1
Yield 89%, yellow powder. m.p. 154-155 °C. H NMR (500 MHz, DMSO-d ) δ
6
1
4.53 (bs, 1H), 13.40 (s, 1H), 12.02 (s, 1H), 10.62 (s, 1H), 7.47 (d, J = 3.5 Hz, 1H),
.30 (dd, J = 8.5, 3.5 Hz, 1H), 7.03 (d, J = 8.5 Hz, 1H), 6.00 (s, 1H), 2.66 (s, 3H),
7
2
7

1
3
2
1
.52 (s, 3H), 1.99 (s, 3H), 1.71 (s, 3H). C NMR (126 MHz, DMSO-d ) δ 201.4,
6
98.6, 190.0, 175.0, 174.2, 163.1, 158.0, 156.2, 151.2, 129.6, 127.1, 124.7, 122.9,
118.2, 107.0, 105.5, 103.0, 102.7, 101.5, 57.2, 32.1, 31.5, 20.8, 8.0. HRMS (ESI) m/z
-
4
68.0856 [M - H] (calcd for 468.0852, C H ClNO ).
27 19 7
(R,E)-6-acetyl-2-(1-((4-chloro-2-hydroxyphenyl)amino)ethylidene)-7,9-dihydroxy-8,9
b-dimethyldibenzo[b,d]furan-1,3(2H,9bH)-dione. (6).
1
Yield 84%, yellow powder. m.p. 147-149 °C. H NMR (500 MHz, DMSO-d ) δ
6
1
7
2
1
4.52 (bs, 1H), 13.40 (s, 1H), 12.03 (s, 1H), 10.87 (s, 1H), 7.36 (d, J = 8.5 Hz, 1H),
.05 (d, J = 2.3 Hz, 1H), 6.98 (dd, J = 8.5, 2.3 Hz, 1H), 5.98 (s, 1H), 2.66 (s, 3H),
1
3
.51 (s, 3H), 1.98 (s, 3H), 1.71 (s, 3H). C NMR (126 MHz, DMSO-d ) δ 201.4,
6
98.5, 190.0, 174.8, 174.1, 163.0, 158.0, 156.2, 153.0, 133.4, 128.7, 122.8, 119.7,
116.7, 107.0, 105.5, 103.0, 102.7, 101.4, 57.2, 32.1, 31.5, 20.8, 8.0. HRMS (ESI) m/z
-
4
68.0856 [M - H] (calcd for 468.0857, C H ClNO ).
27 19 7
(R,E)-6-acetyl-7,9-dihydroxy-2-(1-((2-hydroxy-5-methylphenyl)amino)ethylidene)-8,
9
b-dimethyldibenzo[b,d]furan-1,3(2H,9bH)-dione. (7).
1
Yield 82%, yellow powder. m.p. 134-136 °C. H NMR (500 MHz, DMSO-d ) δ
6
1
7
2
4.50 (bs, 1H), 13.41 (s, 1H), 12.12 (s, 1H), 10.05 (s, 1H), 7.11 (d, J = 2.0 Hz, 1H),
.06 (dd, J = 8.2, 2.0 Hz, 1H), 6.92 (d, J = 8.2 Hz, 1H), 5.98 (s, 1H), 2.66 (s, 3H),
1
3
.51 (s, 3H), 2.24 (s, 3H), 1.99 (s, 3H), 1.71 (s, 3H). C NMR (126 MHz, DMSO-d )
6
δ 201.4, 198.4, 190.0, 174.7, 174.0, 163.0, 158.0, 156.2, 149.6, 130.4, 128.8, 127.4,
1
23.1, 116.8, 106.9, 105.6, 102.8, 102.8, 101.4, 57.1, 32.2, 31.5, 20.8, 20.4, 8.0.
-
HRMS (ESI) m/z 448.1402 [M - H] (calcd for 448.1399, C H NO ).
25
22
7
2
8