Triptolide derivatives as potential multifunctional anti-Alzheimer agents: Synthesis and structure–activity relationship studies

Article abstract of DOI:10.1016/j.bmcl.2018.01.019

Owning to the promising neuroprotective profile and the ability to cross the blood–brain barrier, triptolide has attracted extensive attention. Although its limited solubility and toxicity have greatly hindered clinical translation, triptolide has nonetheless emerged as a promising candidate for structure–activity relationship studies for Alzheimer's disease. In the present study, a series of triptolide analogs were designed and synthesized, and their neuroprotective and anti-neuroinflammatory effects were then tested using a cell culture model. Among the triptolide derivatives tested, a memantine conjugate, compound 8, showed a remarkable neuroprotective effect against Aβ_1–42 toxicity in primary cortical neuron cultures as well as an inhibitory effect against LPS-induced TNF-α production in BV2 cells at a subnanomolar concentration. Our findings provide insight into the different pharmacophores that are responsible for the multifunctional effects of triptolide in the central nervous system. Our study should help in the development of triptolide-based multifunctional anti-Alzheimer drugs.

Full text of DOI:10.1016/j.bmcl.2018.01.019

Accepted Manuscript
Triptolide Derivatives as Potential Multifunctional Anti-Alzheimer Agents:
Synthesis and Structure–Activity Relationship Studies
Chengqing Ning, Liumei Mo, Xuwei Chen, Wentong Tu, Jun Wu, Shengtao
Hou, Jing Xu
PII:
S0960-894X(18)30029-5
https://doi.org/10.1016/j.bmcl.2018.01.019
BMCL 25545
DOI:
Reference:
To appear in:
Bioorganic & Medicinal Chemistry Letters
Received Date:
Revised Date:
Accepted Date:
11 September 2017
5 January 2018
12 January 2018
Please cite this article as: Ning, C., Mo, L., Chen, X., Tu, W., Wu, J., Hou, S., Xu, J., Triptolide Derivatives as
Potential Multifunctional Anti-Alzheimer Agents: Synthesis and Structure–Activity Relationship Studies,
Bioorganic & Medicinal Chemistry Letters (2018), doi: https://doi.org/10.1016/j.bmcl.2018.01.019
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. 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.

Triptolide Derivatives as Potential Multifunctional Anti-Alzheimer Agents: Synthesis and
Structure–Activity Relationship Studies
a
Chengqing Ning, Liumei Mo, Xuwei Chen, Wentong Tu, Jun Wu,* Shengtao Hou,* and
b, c
d
e
, b
, d
,
a
Jing Xu*
a
SUSTech Academy for Advanced Interdisciplinary Studies and Department of Chemistry, Southern
University of Science and Technology, Shenzhen 518055, China
b
Department of Neurology, Peking University Shenzhen Hospital, Shenzhen 518036, China
c
Shantou University Medical College, Shantou 515041, China
d
Brain Research Centre and Department of Biology, Southern University of Science and Technology,
Shenzhen 518055, China
e
Department of Chemistry, Southern University of Science and Technology, Shenzhen 518055, China
ABSTRACT
Owning to the promising neuroprotective profile and the ability to cross the blood–brain barrier,
triptolide has attracted extensive attention. Although its limited solubility and toxicity have greatly
hindered clinical translation, triptolide has nonetheless emerged as a promising candidate for
structure–activity relationship studies for Alzheimer’s disease. In the present study, a series of
triptolide analogs were designed and synthesized, and their neuroprotective and
anti-neuroinflammatory effects were then tested using a cell culture model. Among the triptolide
derivatives tested, a memantine conjugate, compound 8, showed a remarkable neuroprotective
effect against Aβ1–42 toxicity in primary cortical neuron cultures as well as an inhibitory effect
against LPS-induced TNF-α production in BV2 cells at a subnanomolar concentration. Our
findings provide insight into the different pharmacophores that are responsible for the
multifunctional effects of triptolide in the central nervous system. Our study should help in the
development of triptolide-based multifunctional anti-Alzheimer drugs.
KEYWORDS: triptolide; memantine; neuroprotection; neuroinflammation; Aβ
Alzheimer’s disease (AD), the most common form of dementia, is a major public health problem,
with an ever-increasing number of affected individuals as the world’s population ages. It is
estimated that 47 million people worldwide were living with dementia in 2015, and this number
1
will increase to more than 131 million by 2050. At present, there is no cure for AD, and there is
no way to halt the neurological damage. The current US FDA-approved drugs (Fig. 1) only help
2
lessen or stabilize cognitive symptoms. Thus, there is an urgent need for new effective therapies
for AD.
Fig. 1. Chemical structure of FDA-approved anti-AD drugs.

Although the neuropathology of AD is still not fully understood, deposition of β-amyloid peptides
and neurofibrillary tangles consisting of hyperphosphorylated tau are the two primary pathological
3
-5
hallmarks of the disease.
these amyloid plaques and neurofibrillary tangles plays an important role in the pathogenesis of
AD.
Accumulating evidence indicates that the inflammatory response to
6
-9
The inflammatory response, caused by the activation of microglial cells and astrocytes,
leads to the overexpression of proinflammatory factors such as TNF-α and IL-1β, contributing to
1
0-12
neuronal dysfunction and death.
failures of several clinical candidates (such as solanezumab and verubecestat) targeting the
In light of the highly complex pathology of AD and recent
1
amyloid or tau cascades, multifunctional compounds with multiple modes of action targeting
3
1
4-15
several specific pathogenetic AD processes might have promise as disease-modifying agents
and have better clinical efficacy.
Triptolide, a diterpene triepoxide, is one of the major active components of the Chinese herb
Tripterygium wilfordii Hook F (Thunder God Vine), which has been used in Traditional Chinese
Medicine to treat autoimmune and inflammatory diseases, such as rheumatoid arthritis, for
decades. Since it was isolated, triptolide has been reported to possess a broad spectrum of
biological actions, including anticancer, anti-inflammatory, immunosuppressive and anti-fertility
1
6-19
activities.
mouse model of AD.
More recently, triptolide was shown to reduce AD-like pathology in a transgenic
2
0-22
Although the mechanisms underlying the therapeutic effects of triptolide
in the AD model remain to be fully elucidated, it was found that the activation of neurotrophic
pathways, inflammatory cascades, modulation of oxidative stress and inhibition of BACE1
expression were involved, suggesting that triptolide might be a potent multifunctional natural lead
compound for the treatment of AD.
Despite the promising neuroprotective activities of triptolide, poor aqueous solubility and toxicity
have impeded preclinical development and clinical translation of the compound. Previous
1
6
structure–activity relationship (SAR) studies of triptolide identified the key pharmacophores that
may account for its antitumor effect. In the present study, we explore whether these key
pharmacophores are important for its neuroprotective activities. To this end, a series of triptolide
derivatives were designed and synthesized, and their neuroprotective effects were evaluated with
SAR studies, leading to the identification of conjugate 8, a potential multifunctional anti-AD
agent.
RESULTS AND DISCUSSION
Fig. 2. Modification of the 12,13-epoxy moiety.
Epoxy groups and the C-14-hydroxyl group have been reported to be key functional groups,
1
6
accounting for the antitumor effect of triptolide. We therefore examined the role of these groups

in its neuroprotective action. We first focused on the epoxy groups. Among the three epoxy groups,
C-12,13-epoxy is the most sensitive to nucleophilic attack and is the only group involved in the
2
covalent binding with the cysteine of xeroderma pigmentosum B (XPB), a receptor reported to
3
2
4
25
be involved in its antitumor effect. It has been demonstrated that derivatives with nucleophilic
ring-opening at the C-12 position, such as tripchlorolide (Fig. 2), completely lack
immunosuppressive activity while retaining the anti-inflammatory activity. However,
tripchlorolide can be converted into triptolide in vivo. We therefore synthesized C-12,13-epoxy
opening analogs that cannot be efficiently converted to triptolide in vivo (Fig. 2). In addition, the
C-7,8-β-epoxide-modified analog, compound 4 (Scheme 1), was synthesized to assess the roles of
the different epoxy groups on neuroprotective activity.
The synthetic route is shown in Scheme 1. Compound 1 was synthesized via boron trifluoride
6
2
etherate-mediated hydrofluorination,
C-12,13-epoxy in PBS buffer (pH = 4.0) at 100 ºC. SmI
while Compound 2 was prepared by hydrolysis of
-induced reduction of α,β-epoxy ketone
2
7
2
2
8
23
provided compound 3. When reacted with LiBH
selectively opened at room temperature, giving rise to product 4.
4
•THF and BF
3
•Et
2
O,
C-7,8-β-epoxy is
We next evaluated the neuroprotective activities of these synthesized compounds. Because soluble
2
9-30
Aβ oligomers play a key role in the pathogenesis of AD,
we used oligomeric Aβ1–42 as a
31
neurotoxin to establish a cell-based screen for neuroprotection. Primary cortical neurons were
incubated with Aβ oligomers in the absence or presence of varied concentrations of the different
compounds for 24 h, and the neuroprotective effect was evaluated by measuring cell viability
using the CCK8 assay. Doses showing no neurotoxicity were selected for the assay (data not
3
2
shown). Aβ oligomers were prepared according to the reported method. Triptolide was selected
as positive control. As shown in Fig. 3, exposure to Aβ1–42 at a concentration of 4 µM for 24 h
resulted in a 40% reduction in primary cortical neuronal viability, compared with the control
group. Pretreatment with triptolide completely rescued cells from Aβ1–42 toxicity. An increase in
cell viability, although slight, was also observed when primary cortical neurons were pretreated
with compounds 1–4, indicating that both C12,13-epoxy and C-7,8-β-epoxy are involved in
neuroprotection. Hydrofluorination, hydrolysis or reduction resulted in the loss of the
neuroprotective activity against Aβ toxicity.
Scheme 1. Synthesis of compounds 1–4.

Fig. 3. Neuroprotective profiles of compounds 1–4 against Aβ1–42 neurotoxicity. Primary cortical
neurons were treated with Aβ1–42 (4 µM) alone or in the presence of the indicated dose of triptolide
derivatives for 24 h. Cell viability was determined by CCK8 assay. All results are presented as
mean ± SD of three independent experiments.
We then turned our attention to the modification of the 14-β-OH group. As shown in Scheme 2,
oxidation of the 14-OH group with Dess–Martin periodinane (DMP) at room temperature
produced the ketone analog, 5. Triptolide treated with DAST under ice bath conditions produced
the 14-α-fluorinated derivative, 6. The neuroprotective assay showed that compounds 5 and 6
completely lost their protective activities against Aβ cytotoxicity towards primary cortical
neuronal cells (Fig. 4), suggesting that the 14-β-OH group is also a key pharmacophore for its
neuroprotective action.
Given the importance of the 14-β-OH group for the bioactivity of triptolide, we next examined the
effect of modifying this group by the addition of various moieties. We conjugated triptolide to
memantine via a carbamate linker to generate a triptolide-memantine conjugate, compound 8. The
rational for selecting memantine as the moiety included the following: 1) the introduction of
memantine might increase its neuroprotective activity, and the safety is known, as it is an NMDA
receptor antagonist approved by the FDA for the treatment of AD; 2) memantine is a privileged
scaffold used in CNS drug discovery, and a triptolide-memantine conjugate has not been reported
in the literature. To explore the influence of moieties on neuroprotective effect, we also conjugated
triptolide to a bulky moiety lacking neuroprotective activity via a carbamate linker to generate
compound 9 as control. Conjugate 8 and 9 were prepared from triptolide via 2 steps as outlined in
Scheme 2. CCK8 assay revealed that compound 8 dose-dependently reduced cell viability (Fig.
4C). No toxicity was observed at concentrations ≤ 10 nM. At 25, 50 and 100 nM, it induced a 7.7
±
1.8%, 32.8 ± 0.7% and a 73.5 ± 2.6% reduction in viable cells, respectively. In comparison, cells
exposed to triptolide at these concentrations showed a 27.2 ± 3.4%, 62.8 ± 5.9 and a 76.5 ± 2.9%
reduction in cell viability, respectively. This demonstrates that the toxicity of conjugate 8 is
slightly lower than that of triptolide in cortical neuronal cultures. Conjugate 8 completely
protected against Aβ toxicity at a concentration of 0.01 nM, with a neuroprotective profile similar
to triptolide, indicating that the conjugated large rigid moiety memantine does not reduce
neuroprotective activity. Interestingly, when the triptolide conjugated to tert-butylamine
(
compound 9), no neuroprotective effect against Aβ toxicity or even toxic effect on cell viability
was observed at a concentration of 1000 nM (Fig. 4E). These results also confirm that 14-β-OH is
an important pharmacophore for neuroprotective activity. Chemical modification of 14-β-OH may
result in loss of activity. The different neuroprotective profile between compound 8 and 9 suggests
that the conjugated moiety has a significant influence on the neuroprotective activity. We assumed

that a bulky moiety linked to 14-β-OH group would reduce the activity of triptolide, while a
moiety having neuroprotective effect, such as memantine, might recover the activity. However, it’s
worth noting that only at concentrations ≥ 10 µM can memantine demonstrate neuroprotective
3
3, 34
effect against Aβ toxicity in cortical neurons.
neuroprotective activity of conjugate 8 could be due to the synergistic effects of triptolide and
Thus, a possible explanation for the remarkable
memantine in cortical neuronal cells.
To investigate the stability of carbamate group under non-enzymatic hydrolysis conditions,
conjugate 8 was incubated with neurobasal cell culture medium (supplemented with B-27 and N2)
o
at 37 C for 24h. The sample was then analyzed by LC-HRMS to determine the concentration of
conjugate 8. The results showed that 13% of conjugate 8 as well as positive control triptolide
underwent decomposition or hydrolysis, which suggested that the conjugate 8 is stable under
conditions used for neuroprotective assay.
Scheme 2. Synthesis of triptolide derivatives 5-6 and 8-9.

Fig. 4. Neuroprotective effects of triptolide derivatives against Aβ1–42 cytotoxicity in primary
cortical neuronal cultures. Primary cortical neurons were treated with Aβ1–42 (4 µM) alone or in the
presence of the indicated dose of triptolide analogs for 24 h. All results are presented as the mean
±
SD of three independent experiments.
To further investigate whether the neuroprotective effects of these derivatives are related to their
anti-inflammatory activities, we assessed the ability of these compounds to repress TNF-α
production in LPS-activated BV-2 microglial cells, given that high TNF-α levels in the brain are
1
0, 35
reported to disrupt microglial clearance of Aβ and cause synaptic dysfunction.
We measured
the cytotoxicity of these compounds using the MTS assay for cell viability. BV-2 cells were
incubated with various doses of the compounds for 24 h, and then cell viability was determined to
evaluate the dose–effect relationship (data not shown). Doses that did not affect cell viability were
selected for the anti-inflammation assay. BV-2 cells were pretreated with various doses of
compounds for 1 h and stimulated with LPS (1 µg/mL) for another 23 h. ELISA kits from
eBioscience (BMS607-3) were used to measure TNF-α levels at 24 h. As shown in Fig. 5,
treatment with LPS produced a robust increase in TNF-α levels, compared with the vehicle control
group. This significant increase in TNF-α expression was inhibited by the various compounds in a
dose-dependent manner to different degrees, except for compounds 3 and 4, which displayed only
a weak anti-inflammatory activity in BV-2 cells. Compound 8 had the strongest anti-inflammatory
activity, similar to that of triptolide, which completely suppressed the expression of TNF-α
induced by LPS at a concentration of 10 nM. The anti-inflammatory effects of compounds 5 and 6
were slightly weaker than that of conjugate 8. They completely inhibited TNF-α production at 20
nM.

Fig. 5. Effects of triptolide derivatives on LPS-induced TNF-α expression in BV-2 cells. BV-2
cells were treated with LPS (1 µg/mL) alone or in the presence of the indicated concentrations
(
nM) of the various triptolide derivatives for 24 h. TNF-α levels in the cell supernatants were
determined by ELISA assay. All results are presented as mean ± SD of three independent
experiments.
Although the various compounds demonstrated moderate to potent anti-inflammatory activities,
the activities of some of the derivatives are inconsistent with their neuroprotective activities
towards Aβ cytotoxicity, such as compounds 5 and 6. As described above, no neuroprotective
activities were observed for compounds 5 and 6 in Aβ-treated cortical neurons, although they
completely suppressed TNF-α expression. This suggests that different pharmacophores are
responsible for the neuroprotective and anti-neuroinflammatory effects. It seems that both epoxy
and 14-β-OH groups are required for the neuroprotective activities, whereas epoxy groups, but not
the 14-β-OH group, are involved in the anti-neuroinflammatory activities. Indeed, triptolide has
been reported to interact with several cellular proteins, including XPB, dCTP pyrophosphatase 1
3
6
(DCTPP1), polycystin-2 (PC-2), disintegrin and metalloprotease 10 (ADAM10). Therefore, the
various moieties in triptolide might participate in binding to these proteins.
In summary, we designed, synthesized and assessed the neuroprotective activities of triptolide
derivatives in a cell culture model of AD. Preliminary SAR investigation revealed that both the
epoxy and 14-β-OH groups were required for the neuroprotective activity against Aβ1–42 toxicity
towards primary cortical neurons. Because it was reported that the neuroprotective effect of
triptolide was related to its anti-inflammatory activity, we examined the anti-neuroinflammatory
activities of the various compounds. We found that the neuroprotective activities did not correlate
well with the anti-inflammatory activities, indicating that different pharmacophores are
responsible for the multifunctional effects of triptolide in the CNS. Together, our results
demonstrate that the epoxy groups, but not the 14-β-OH group, are necessary for the
anti-neuroinflammatory action of triptolide. SAR investigation showed that conjugate 8 exhibits
promising neuroprotective and anti-neuroinflammatory profiles at subnanomolar concentration. In
addition, a slightly lower cytotoxicity against cortical neurons was observed compared with
triptolide. Given the encouraging cell-based neuroprotective and anti-neuroinflammatory activities,
conjugate 8 merits further investigation, as it might open up new avenues for generating leads and
drug candidates endowed with multifunctional neuroprotective profiles against AD. Thus, our
study provides a foundation for the discovery of triptolide derivatives as potential anti-AD agents.
Furthermore, our study provides a good lead (triptolide-memantine conjugate 8) for further
anti-AD drug discovery and development.
ACKNOWLEDGEMENTS

We gratefully acknowledge the National Thousand Young Talents Program, Shenzhen
Tech-Innovation Program from Shenzhen Science, Technology and Innovation Commission (No.
Pu20150267) and Shenzhen “Peacock Team” Program for financial support of this work. We also
thank Jiana Li for technical assistance. Financial supports to STH were from the National Natural
Science Foundation of China Grant (81571287), Shenzhen Science and Technology Innovation
Committee Basic Science Research Grant (JCYJ20160301112230218); SUSTech Peacock
Program Start-up Fund to STH (22/Y01226109) and SUSTech Brain Research Centre Fund.
SUPPLEMENTARY DATA
Detail about experimental information and spectra of target compounds.
REFERENCES AND NOTES
1.
Prince M, Comas-Herrera A, Knapp M, Guerchet M, Karagiannidou M. World Alzheimer
Report 2016. Alzheimer’s Disease International (ADI), London.
2.
Citron M. Nat Rev Drug Discov. 2010; 9:387.
3.
Selkoe DJ. Physiol Rev. 2010; 81:741.
4.
Blennow K, de Leon MJ, Zetterberg H. Lancet. 2006; 368:387.
5
.
Alberto SP, Matthew PF, Eliezer M, Bradley TH. Perspect Med. 2011; 1:a006189.
Heneka MT, Carson MJ, El Khoury J, Landreth GE, Brosseron F, Feinstein DL, Jacobs AH,
Wyss-Coray T, Vitorica J, Ransohoff RM, Herrup K, Frautschy SA, Finsen B, Brown GC,
Verkhratsky A, Yamanaka K, Koistinaho J, Latz E, Halle A, Petzold GC, Town T, Morgan D,
Shinohara ML, Perry VH, Holmes C, Bazan NG, Brooks DJ, Hunot S, Joseph B,
Deigendesch N, Garaschuk O, Boddeke E, Dinarello CA, Breitner JC, Cole GM, Golenbock
DT, Kummer MP. Lancet Neurol. 2015; 14:388.
6.
7
.
Shadfar S, Hwang CJ, Lim MS, Choi DY, Hong JT. Arch Pharm Res. 2015; 38:2106.
McGeer PL, McGeer EG. Acta Neuropathol. 2013; 126:479.
8.
9
1
1
1
.
Wyss-Coray T. Nat Med. 2006; 12:1005.
0. Koenigsknecht-Talboo J, Landreth GE. J Neurosci. 2005; 25, 8240.
1. Griffin WS, Sheng JG, Roberts GW, Mrak RE. J Neuropathol Exp Neurol. 1995; 54:276.
2. Griffin WS, Sheng JG, Royston MC, Gentleman SM, McKenzie JE, Graham DI, Roberts G
W, Mrak RE. Brain Pathol. 1998; 8:65.
1
1
1
1
1
1
1
2
3. Hardy J, De Strooper B. Brain. 2017; 140:853.
4. Zheng H, Fridkin M, Youdim M. Pharmaceuticals (Basel). 2014; 7:113.
5. Rosini M, Simoni E, Caporaso R, Minarini A. Future Med Chem. 2016; 8:697.
6. Zhou ZL, Yang YX, Ding J, Li YC, Miao ZH. Nat Prod Rep. 2012; 29:457.
7. Zheng Y, Zhang WJ, Wang XM. CNS Neurosci Ther. 2013; 19:76.
8. Liu Q. Int Immunopharmacol. 2011; 11:377.
9. Qiu D, Kao PN. Drugs R D. 2003; 4:1.
0. Cui YQ, Wang Q, Zhang DM, Wang JY, Xiao B, Zheng Y, Wang XM. Curr Alzheimer Res.
2
016; 13:288.
2
1. Wang Q, Xiao B, Cui S, Song H, Qian Y, Dong L, An H, Cui Y, Zhang W, He Y, Zhang J,
Yang J, Zhang F, Hu G, Gong X, Yan Z, Zheng Y, Wang X. Dis Model Mech. 2014; 7: 1385.
2. Cheng S, LeBlanc KJ, Li L. PLoS One. 2014; 9:e108845.
2
2
3. He QL, Titov DV, Li J, Tan M, Ye Z, Zhao Y, Romo D, Liu JO. Angew Chem Int Ed Engl.
2
015; 54:1859.
24. Titov DV, Gilman B, He QL, Bhat S, Low WK, Dang Y, Smeaton M, Demain AL, Miller PS,

Kugel JF, Goodrich JA, Liu JO. Nat Chem Biol. 2011; 7:182.
2
2
2
2
2
3
3
5. Yu DQ, Zhang DM, Wang HB, Liang XT. YaoXueXueBao. 1992; 27:830.
6. Zderic JA, Limon DC, Ringold HJ, Djerassi C. J Am Chem Soc. 1959; 81:3120.
7. Dai D, Yuan H, Musser JH. WIPO Pat Appl. 2002070472.
8. Molander GA, Hahn, G. J Org Chem. 1986; 51:2596.
9. Viola KL, Klein WL. Acta Neuropathol. 2015; 129:183.
0. Tu S, Okamoto S, Lipton SA, Xu H. Mol Neurodegener. 2014; 9:48.
1. Primary cortical neurons were prepared according to previously described: Hou ST, Jiang SX,
Zaharia LI, Han X, Benson CL, Slinn J, Abrams SR. J Biol Chem. 2016; 291:27007.
2. Dahlgren KN, Manelli AM, Stine WB, Baker LK, Krafft GA, LaDu MJ. J Biol Chem. 2002;
3
2
77:32046.
3
3. Song MS, Rauw G, Baker GB, Kar S. Eur J Neurosci. 2008; 28:1989.
4. Tremblay R, Chakravarthy B, Hewitt K, Tauskela J, Morley P, Atkinson T, Durkin JP. J
Neurosci. 2000; 20:7183.
3
3
5. Ralay Ranaivo H, Craft JM, Hu W, Guo L, Wing LK, Van Eldik LJ, Watterson DM. J
Neurosci. 2006; 26:662.
3
6. Manzo SG. Zhou ZL, Wang YQ, Marinello J, He JX, Li YC, Ding J, Capranico G, Miao ZH.
Cancer Res. 2012; 72:5363.

Graphical Abstract