Discovery of thienopyrrolotriazine derivatives to protect mitochondrial function against Aβ-induced neurotoxicity

Article abstract of DOI:10.1016/j.ejmech.2017.09.033

Recovery of mitochondrial dysfunction has gained increasing attention as an alternative therapeutic strategy for Alzheimer's disease (AD). Recent studies suggested that the 18 kDa mitochondrial translocator protein (TSPO) has the potential to serve as a drug target for the treatment of AD. In this study, we generated a structure-based pharmacophore model and virtually screened a commercial library, identifying SVH07 as a virtual hit, which contained a tricyclic core structure, thieno[2′,3’:4,5]pyrrolo[1,2-d][1,2,4]triazine group. A series of SVH07 analogues were synthesized and their effects on the mitochondrial membrane potential and ATP production were determined by using neuronal cells under Aβ-induced toxicity. Among these analogues, compound 26 significantly recovered mitochondrial membrane depolarization and ATP production. In vitro binding assays indicated that SVH07 and 26 showed high affinities to TSPO with the IC₅₀ values in a nanomolar range. We believe that compound 26 is a promising lead compound for the development of TSPO-targeted mitochondrial functional modulators with therapeutic potential in AD.

Full text of DOI:10.1016/j.ejmech.2017.09.033

Accepted Manuscript
Discovery of thienopyrrolotriazine derivatives to protect mitochondrial function against
Aß-induced neurotoxicity
TaeHun Kim, Woo Seung Son, Mohammad Neaz Morshed, Ashwini M. Londhe, Seo
Yun Jung, Jong-Hyun Park, Woo-Kyu Park, Sang Min Lim, Ki Duk Park, Sung Jin
Cho, Kyu-Sung Jeong, Jiyoun Lee, Ae Nim Pae
PII:
S0223-5234(17)30747-X
10.1016/j.ejmech.2017.09.033
EJMECH 9748
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 26 April 2017
Revised Date: 18 September 2017
Accepted Date: 18 September 2017
Please cite this article as: T. Kim, W.S. Son, M.N. Morshed, A.M. Londhe, S.Y. Jung, J.-H. Park, W.-K.
Park, S.M. Lim, K.D. Park, S.J. Cho, K.-S. Jeong, J. Lee, A.N. Pae, Discovery of thienopyrrolotriazine
derivatives to protect mitochondrial function against Aß-induced neurotoxicity, European Journal of
Medicinal Chemistry (2017), doi: 10.1016/j.ejmech.2017.09.033.
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.

ACCEPTED MANUSCRIPT
Discovery of Thienopyrrolotriazine Derivatives to Protect Mitochondrial
Function against Aß-induced Neurotoxicity
a,b, ǂ
a,c, ǂ
a,d
a,b
TaeHun Kim
, Woo Seung Son , Mohammad Neaz Morshed , Ashwini M. Londhe ,
a
a
e
b,f
a,b
Seo Yun Jung , Jong-Hyun Park , Woo-Kyu Park , Sang Min Lim , Ki Duk Park , Kyu-Sung
c
g,*
a,b,*
Jeong , Jiyoun Lee , and Ae Nim Pae
Graphical abstract

ACCEPTED MANUSCRIPT
Discovery of Thienopyrrolotriazine Derivatives to Protect Mitochondrial
Function against Aß-induced Neurotoxicity
a,b, ǂ
a,c, ǂ
a,d
a,b
TaeHun Kim
, Woo Seung Son , Mohammad Neaz Morshed , Ashwini M. Londhe ,
a
a
e
b,f
a,b
Seo Yun Jung , Jong-Hyun Park , Woo-Kyu Park , Sang Min Lim , Ki Duk Park , Sung Jin
g
c
h,*
a,b,*
Cho , Kyu-Sung Jeong , Jiyoun Lee , and Ae Nim Pae
a
Convergence Research Center for Diagnosis, Treatment and Care System of Dementia, Korea
Institute of Science and Technology (KIST), Hwarangno 14-gil 5, Seongbuk-gu, Seoul 02792,
Republic of Korea
b
Biological Chemistry, Korea University of Science and Technology, Gajeong-ro 217, Yuseong-
gu, Daejon 34113, Republic of Korea
c
Department of Chemistry, Yonsei University, Seodaemun-gu, Seoul 120-749, Korea
d
Center for Advanced Research in Sciences (CARS), University of Dhaka, Dhaka-1000,
Bangladesh
e
Center for Drug Discovery Technology, Korea Research Institute of Chemical Technology,
Gajeong-ro 141, Yuseong-gu, Daejon 34114, Republic of Korea
f
Center for Neuro-Medicine, Korea Institute of Science and Technology (KIST), Hwarangno 14-
gil 5, Seongbuk-gu, Seoul 02792, Republic of Korea
g
New Drug Development Center, Daegu-Gyeongbuk Medical Innovation Foundation,
Daegu, Republic of Korea
h
Department of Global Medical Science, Sungshin University, Seoul 01133, Republic of Korea
ǂ
These authors contributed equally to this work.
Corresponding authors:
Ae Nim Pae, Ph.D.
E-mail: anpae@kist.re.kr, Tel: +82- 2-958-5185, Fax: +82-2-958-5189
Jiyoun Lee, Ph.D.
Email: jlee@sungshin.ac.kr, Tel: +82- 2-920-7281, Fax: +82-2-920-2027
Keywords. Thieno[2’,3’:4,5]pyrrolo[1,2-d][1,2,4]triazine; Translocator protein; Mitochondrial
function; Amyloid beta; Alzheimer’s disease; Ro5-4864.
1

ACCEPTED MANUSCRIPT
Abstract
Recovery of mitochondrial dysfunction has gained increasing attention as an alternative
therapeutic strategy for Alzheimer’s disease (AD). Recent studies suggested that the 18 kDa
mitochondrial translocator protein (TSPO) has the potential to serve as a drug target for the
treatment of AD. In this study, we generated a structure-based pharmacophore model and
virtually screened a commercial library, identifying SVH07 as a virtual hit, which contained the
tricyclic core structure, a thieno[2’,3’:4,5]pyrrolo[1,2-d][1,2,4]triazine group. A series of SVH07
analogues were synthesized and their effects on mitochondrial membrane potential and ATP
production were determined by using neuronal cells under Aβ-induced toxicity. Among these
analogues, compound 26 significantly recovered mitochondrial membrane depolarization and
ATP production. In vitro binding assays indicated that SVH07 and 26 showed high affinities to
TSPO with the IC values in a nanomolar range. We believe that compound 26 is a promising
5
0
lead compound for the development of TSPO-targeted mitochondrial functional modulators with
therapeutic potential in AD.
2

ACCEPTED MANUSCRIPT
1. Introduction
Alzheimer’s disease (AD) is an incurable neurodegenerative disorder characterized as the
progressive loss of cognitive function and memory [1]. The pathogenesis of AD has not been
fully understood, but age-related protein misfolding and aggregation is considered to be the
major culprit [2, 3]. In particular, misfolded protein deposits such as amyloid plaques and
neurofibrillary tangles are the hallmarks of AD, and accumulation of these misfolded proteins
disrupts metabolic processes and communication of neurons, eventually leading to neuronal cell
death [4]. Research efforts have focused on the removal of the misfolded proteins, specifically β-
amyloid (Aβ), by applying antibodies and enzyme inhibitors [5, 6]. Unfortunately, most of these
anti-amyloid agents have failed to show therapeutic benefits in clinical trials while some of these
candidates are still in long-term clinical trials for preventive effects [7-9]. Given that currently
available drugs, such as acetylcholinesterase (AchE) inhibitors and N-methyl-D-aspartate
(NMDA) receptor antagonists, can only provide symptomatic relief without significantly
affecting the disease course [1, 10], identifying a novel therapeutic target may offer an
alternative approach to treat AD.
Mitochondria are thought to hold promise as a potential therapeutic target in
neurodegenerative diseases, because the most prominent roles of mitochondria are cellular
metabolism and energy production, both of which are crucial to maintain healthy neuronal
structures and normal brain functions [11-14]. Several mitochondrial proteins have been studied
as possible drug targets for the treatment of AD [15-21]. Specifically, recent studies found that
intracellular Aβ appears to promote the formation of the mitochondrial permeability transition
pore (mPTP) by interacting with one of the putative components cyclophilin D (CypD) [22, 23].
The knock-down of CypD recovered cognitive functions in AD model mice [23], and the
3

ACCEPTED MANUSCRIPT
administration of a CypD inhibitor, cyclosporine A (CsA), restored Aβ-induced decline of ATP
production and cell viability [24]. These evidences suggest that prolonged opening of the mPTP
disrupts mitochondrial membrane potential, decreases ATP production, and induces extensive
swelling of mitochondrial membrane, leading to mitochondrial dysfunction and cell death [22].
Therefore, blocking the formation of the mPTP not only prevent Aβ from facilitating
mitochondrial dysfunction, but may also afford an alternative treatment option that can change
the disease course [24-26].
The 18 kDa translocator protein (TSPO) is another mitochondrial protein that may
provide a possible drug target for the treatment of AD [27-30]. TSPO appears to participate in
the formation of the mPTP, although the exact role of TSPO is under debate [31-33]. As
illustrated in Figure 1, several synthetic TSPO ligands have been developed as therapeutic
agents for neurodegenerative diseases [34, 35]. Compounds 1 - 6 are highly potent TSPO ligands
that have been reported as neuroprotective agents and their efficacies have been evaluated in
preclinical and clinical trials [31, 36-39]. It has been demonstrated that Ro5-4864 (5) reduced
neuronal cell death by regulating neurosteroid biosynthesis, opening of the mPTP, and cellular
oxidative stress [40, 41]. Moreover, Ro5-4864 reduced anxiety and improved learning and
memory in AD mice by reducing gliosis and hippocampal Aβ accumulation [17]. Inspired by
these approaches, we have previously developed TSPO-targeted mitochondrial functional
modulators for the treatment of AD [42]. In our previous study, we generated a common feature
pharmacophore based on the structures of compounds 1-4, and further optimized and designed a
new series of compounds containing a benzimidazole group as the core scaffold [42]. Initially,
we also included Ro5-4864 to generate a common pharmacophore, however, possibly due to its
distinct binding mode, we failed to generate a single, reliable pharmacophoric model. Therefore,
4

ACCEPTED MANUSCRIPT
we decided to generate an independent pharmacophore model based on the structure of Ro5-
4
864.
Figure 1. Representative TSPO ligands.
In this study, we designed a series of TSPO-targeted mitochondrial functional modulators
via a ligand-based method. A new pharmacophore model was generated with the structure of
Ro5-4864 and commercially available chemical libraries were virtually screened with the
pharmacophore model for optimization. A total of 26 compounds were selected from the virtual
screening and their biological activities associated with mitochondrial functions were evaluated
in vitro, resulting in a hit compound that contained a thieno[2’,3’:4,5]pyrrolo [1,2-
d][1,2,4]triazine group as the core scaffold. On the basis of these findings, we designed and
synthesized a library of thienopyrrolotriazine analogues, and assessed their biological activity in
Aβ-induced mitochondrial dysfunction.
5

ACCEPTED MANUSCRIPT
2
. Results and discussion
.1. Hit identification from the shape-added pharmacophore model of Ro5-4864 (5)
2
To generate a common feature pharmacophore model for TSPO ligands, we initially started
with compounds 1-5 (Figure 1) as the training set compounds, however, failed to generate a
reliable pharmacophore model. The common feature pharmacophore mainly consisted of
hydrophobic and hydrogen bond acceptor features (Supplementary data, Figure S1). Based on
the fit values generated for representative compounds, Ro5-4864 demonstrated smaller fit values
for each pharmacophore compared to other compounds (Supplementary data, Table S1). We
speculated that Ro5-4864 belongs to a benzodiazepine class, which may have a distinct binding
mode due to its ring size as well as ring strain, thus cannot be accommodated in one common
pharmacophore model. Considering that Ro5-4864 is a potent TSPO ligand with neuroprotective
effects in neurons and AD model mice [17], we decided to build a pharmacophore model solely
based on the structure of Ro5-4864. As described in Figure 2A, five pharmacophoric features
were generated from Ro5-4864: one hydrophobic feature, two hydrophobic aromatic features,
and two hydrogen bond acceptors. One hydrophobic feature was generated from the N-methyl
group, while two hydrophobic aromatic features were generated from each aromatic ring in Ro5-
4
864. Hydrogen bond acceptor features were generated from N and O atoms from the
benzodiazepine core. For structure-based ligand search, we additionally allowed the presence of
S atom in place of hydrogen bond acceptor features to diversify the core scaffold; on the other
hand, we combined the pharmacophore-based search and shape-based similarity search to reduce
the number of virtual hits. This combined structure search using the pharmacophore model
6

ACCEPTED MANUSCRIPT
retrieved 393,076 hits out of 1,047,671 compounds. Among these initial hit compounds, a total
of 97 compounds met the filtering criteria by fit values (4.00 out of 5.00). Finally, 26 out of 97
compounds (SVH01 – SVH26, Supplementary data, Table S2) were selected manually by
considering structural diversity and the presence of essential functionality for activity.
A
B
O
N
Cl
N
Cl
(Ro5-4864)
5
Figure 2. (A) The shape-added pharmacophore model of 5 (Ro5-4864). (B) The structure of the
hit compound (SVH07) and its mapping to the generated pharmacophore model. The 2D
structures of compounds are illustrated on the left-hand side (C atom; black, N atom; blue, O
atom; red, S atom; yellow, and Cl atom; green). On the right-hand side, the pharmacophore
features are represented as color-coded spheres: H-bond acceptor, green; and hydrophobic
feature, light blue.
Biological activities of the selected virtual hit compounds, SVH01 – SVH26, were evaluated
by mitochondrial functional assays in HT22 mouse hippocampal cells (Supplementary data,
Table S2). Effect of each compound on the mitochondrial membrane potential (∆Ψ ) was first
m
determined by the JC-1 (5,59,6,69-tetrachloro-1,19,3,39-tetraethylbenz-imidazolocarbocyanine
m
iodide) assay [43]. Healthy mitochondria maintain ∆Ψ that is essential for mitochondrial energy
production and cellular respiration. Aβ-induced opening of the mPTP causes depolarization of
the membrane potential, which significantly decreases the red/green fluorescence intensity ratio
of the JC-1 dye [44, 45]. To determine percent recovery of ∆Ψ by the selected compounds, we
m
7

ACCEPTED MANUSCRIPT
treated HT-22 cells with each compound at 5 µM, and then incubated with oligomeric Aβ (5 µM)
to induce mitochondrial toxicity. Based on changes of the red/green fluorescence intensity ratio
between normal cells and Aβ-treated cells, the percent recovery value of ∆Ψ was calculated for
m
each compound. 8 out of 26 virtual hits showed remarkable percent recovery values of ∆Ψm
(SVH07, SVH12, SVH13, SVH14, SVH17, SVH20, SVH21, and SVH24), ranging from 41%
to 75%. Two additional mitochondrial functional assays were carried out in HT22 cells to assess
biological activity of the selected compounds; a luciferase-based ATP (adenosine-5’-
triphosphate) production assay and the MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide) assay. When treated at 5 µM in the presence of Aβ (5 µM), only
two compounds, SVH07 and SVH14, recovered ATP production by 8 and 14% of the control
respectively; while both compounds did not affect cellular viability in the MTT assays (> 95%).
Among these two compounds, SVH07 exhibited a higher fit value than SVH14 for the given
pharmacophore model (SVH07, fit value = 4.55 and SVH14, fit value = 4.17, Table S2). On the
basis of in vitro assay results as well as fit values, we finally chose SVH07 as our lead
compound.
2
.2. Molecular Docking Studies
To investigate the binding modes of Ro5-4864 and SVH07, we performed docking
studies by using a human TSPO homology model from our previous study [42]. Docking
analysis of Ro5-4864 (Figure 3A, Glide docking score= -10.427) showed that the oxygen atom
in the benzodiazepine ring formed one hydrogen bond with Ser23, and both benzene rings
interacted with Trp53 and Trp95 via strong pi-pi stacked interactions. In addition, the chlorine
atoms formed hydrophobic interactions with Leu49 (alkyl), and Trp95 (pi-alkyl), respectively.
SVH07 (Figure 3B, Glide docking score= -9.984) tightly bound inside the hydrophobic pocket
8

ACCEPTED MANUSCRIPT
that was formed by various aromatic amino acid residues. Thieno[2',3':4,5]pyrrolo[1,2-
d][1,2,4]triazine core exhibited the pi-pi stacked interactions with Trp53 and Trp95, and the N-
methylaniline ring interacted through pi-pi stacking interactions with His43, Tyr34, and Phe99.
A pi-sulfur interaction was also observed between sulfur atom and Trp95. From the docking
analysis, we observed that Ro5-4864 and lead compound SVH07 share the same binding pocket
but do not overlap with each other as demonstrated in Figure 3C.
A
C
B
Figure 3. The predicted binding modes of Ro5-4864 and the hit compound SVH07. (A) Ro5-
864 (yellow stick); (B) SVH07 (orange stick); (C) the binding poses of two compounds
4
overlapped inside the human TSPO. Binding cavity is shown by hydrophilic (blue) and
hydrophobic (gray) surface.
To compare the binding modes of other TSPO ligands, we also performed docking
studies with compounds 1-4. We observed that the binding poses of FGIN-1-27 (1), DAA1097
(2), and Ro5-4864 (5) were overlapped inside the central cavity of the human TSPO binding
pocket. Compound 1 (Figure 4A, Glide docking score= -9.628) formed pi-pi stacked interactions
with Trp53 and Trp95. The oxygen atom of the acetamide group formed a hydrogen bond with
Trp53. The two hexyl side chains also participated in hydrophobic interactions with the side
chains of His43, Pro44, Leu150, and Val154. Docked pose analysis of 2 (Glide docking score= -
9
.917) shows that the oxygen atom of the acetamide group formed two hydrogen bonds with
9

ACCEPTED MANUSCRIPT
Tyr57 and Asn92 as depicted in Figure 4B. The three benzene rings of 2 firmly held in the
binding site by pi-pi stacked interactions with Trp53 and Trp95. Compounds 2 and 3 (DAA1106,
Glide docking score= -10.236) appeared to share the same binding cavity, but in a different
orientation (Figure 4C). The oxygen atom of the acetamide in 3 interacted with Trp53 and
Trp143 by two hydrogen bonds. The methoxy oxygen appeared to form one hydrogen bond with
Asn151. The chlorine atom formed two halogen bonds with the main chain and the side chain of
Asn92. Similar to compound 2, the three benzene rings of compound 3 also tightly packed inside
the binding cavity via pi-pi stacked interactions with Trp53 and Trp95. Among compounds 1-5,
compound 4 (PK11195, Glide docking score= -10.809) demonstrated the highest docking score.
The oxygen atom of the amide group in 4 forms two hydrogen bonds with Trp53 and Trp143
(Figure 4D). The isoquinoline ring of 4 fitted inside the binding cavity by firm pi-pi stacked
interactions with Trp95. Overall, we observed that all five compounds bind at the central site of
the human TSPO. While compounds 1, 2, and 5 demonstrated a superimposable binding
orientation inside central pocket, compounds 3 and 4 appeared to occupy a different subsite,
sharing several common binding site residues at the interface of each subsite.
(A)
(B)
(C)
(D)
Figure 4. The predicted binding modes of compounds 1-4. Binding site interactions of 1 (red), 2
magenta), 3 (orange), and 4 (cyan) is shown in A, B, C, and D respectively.
(
10

ACCEPTED MANUSCRIPT
2
.3. Chemistry
1
2
As illustrated in Figure 5, we decided to diversify three different positions (R , R , and
3
1
R ) of SVH07 to design a library of analogues. Our pharmacophore model indicated that R and
2
R positions represent the hydrophobic interaction of the chlorobenzene ring; therefore, we
limited the size of substituents in these two positions to small alkyl chains (methyl and ethyl) and
a chlorine atom. On the other hand, both the S atom and the carbonyl group in SVH07 act as
hydrogen bond acceptors that mimic the carbonyl group and the N atom in the benzodiazepine
ring; therefore, we decided to modify the position and the functional group in the 6-position of
the triazine ring, resulting in two different groups of compounds, the S-alkylated and the N-
3
alkylated derivatives. For the substituents in the R position, we included various amine
containing aromatic rings as well as a dimethylamine group to further explore possible hydrogen
bond interactions at this specific position.
Figure 5. Newly designed scaffolds of thieno[2’,3’:4,5]pyrrolo[1,2-d][1,2,4]triazine derivatives
(
HY; Hydrophobic group, H ; H-bond donor group).
1
11

ACCEPTED MANUSCRIPT
The S-alkylated thieno[2’,3’:4,5]pyrrolo[1,2-d][1,2,4]triazines were synthesized by
following the pathway described in Scheme 1. First, thiophenecarbaldehydes 7a-c were
condensed with methyl 2-azidoacetate to produce compounds 8a-c. Compounds 8a-c were
thermolyzed to build a bicyclic 4H-thieno[3,2-b]pyrrole, and generated compounds 9a-c [46, 47].
Compounds 9a-c were treated with hydrazine monohydrate to produce hydrazone compounds
1
1
0a-c, which were further cyclized to tricyclic compounds 11-15 [48]. Thionation of compounds
1-15 yielded compounds 16-20 [49]. Finally, S-alkylation of compounds 16-20 with
corresponding 2-bromo-N-substituted acetamides generated the desired products 21-46.
The N-Alkylated thieno[2’,3’:4,5]pyrrolo[1,2-d][1,2,4]triazine derivatives were
synthesized via S 2 reactions of thienopyrrolotriazine fragments 11 or 14 (Scheme 2). Refluxing
N
compounds 11 or 14 with the corresponding 2-bromo-N-substituted acetamides under the basic
condition afforded the final compounds 47-55 [49].
12

ACCEPTED MANUSCRIPT
Scheme 1. Synthesis of the S-alkylated thieno[2’,3’:4,5]pyrrolo[1,2-d][1,2,4]triazine derivatives
a
(SVH07 and 21 - 46).
a
o
Reagents and conditions: (a) methyl 2-azidoacetate, NaOMe (25 wt.% in MeOH), MeOH, 0 C to rt, 4 h, 77 - 98%,
b) Toluene, reflux, 12 h, 50 - 94%, (c) NH NH ·H O, EtOH, reflux, 18 h, 87 - 95%, (d) R -C(OMe) , DMF, reflux,
(
2
2
2
2
3
1
3
2 h, 83 - 97%, (e) Lawesson’s reagent, benzene, reflux, 12 h, 88 - 99%, (f) 2-bromo-N-substituted acetamide, Et N,
o
MeCN, 60 C, 2 h, 40 – 85%.
13

ACCEPTED MANUSCRIPT
Scheme 2. Synthesis of the N-alkylated thieno[2’,3’:4,5]pyrrolo[1,2-d][1,2,4]triazine derivatives
a
(47 - 55).
a
2 3
Reagents and conditions: (a) 2-bromo-N-substituted acetamide, K CO , KI, MeCN, reflux, 8 h, 56 – 99%.
2.4. Recovery of Aβ-induced mitochondrial dysfunction
As summarized in Table 1, the S-alkylated compounds 21 – 46 and the N-alkylated
compounds 47 – 55 were evaluated by cell-based mitochondrial functional assays (JC-1 assay,
luciferase-based ATP production assay, and MTT reduction assay) in HT22 cells at a single dose
of 5 µM where our reference compound CsA consistently showed 50-55 % of ∆Ψm recovery
without affecting cellular viability. The percent recovery values of ∆Ψm showed that newly
synthesized compounds mostly recovered mitochondrial membrane potential to a similar or
greater extent than SVH07 (46%), ranging from 41% to 70%, except compounds 40, 51 and 55.
Interestingly, the percent recovery value of CsA, a known mPTP blocker, was comparable (55%)
to that of SVH07 and other analogues, whereas the percent recovery values of Ro5-4864
14

ACCEPTED MANUSCRIPT
appeared to be lower (12%) than the most of SVH07 analogues, but demonstrated a similar
activity to compounds 51 and 55. Between the two groups of compounds, the S-alkylated
derivatives 21-46 and the N-alkylated derivatives 47-55, the S-alkylated compounds appeared to
have similar or slightly better percent recovery values (48% to 70%) than the N-alkylated
compounds (41% to 57%), however, these differences in ∆Ψm are likely to depend on the
1
3
1
substituents in the R -R positions. Among the S-alkylated series, compounds with R = H
1
(
compounds 21-25, 32-37) demonstrated slightly lower percent recovery compared to their R =
Me and Cl counterparts (compounds 26-31, 38-46). Compounds in the N-alkylated series showed
1
a reversed trend in their activities, showing that compounds with R = H (47-51) exhibited
1
slightly higher percent recovery values than their R = Me counterparts (52-55). However, the
overall differences in percent recovery values in these compounds did not appear to be
2
significant. Similarly, two different R substituents, methyl and ethyl group, did not appear to
1
2
produce distinct effects, possibly because these R and R substituents are relatively small-sized
hydrophobic groups mimicking the chlorophenyl substituent of Ro5-4864. Rather, more
3
pronounced differences were observed with respect to the various R substituents. Among the S-
3
alkylated series, compounds with the N-benzylamine substituent (R = D, compounds 23, 29, 35,
and 40) demonstrated relatively low percent recovery values (35% to 55%); among the N-
alkylated series, compounds with the N-dimethyl substituent (51, 55) exhibited significantly
reduced percent recovery of ∆Ψm (12% and 17%).
15

ACCEPTED MANUSCRIPT
Table 1. Results of the JC-1 assay, a luciferase-based ATP assay, and the MTT assay of the S-
a
alkylated compounds 21-46 and the N-alkylated compounds 47-55 at a single dose of 5 µM.
Recovery of
ATP
Recovery of
ATP
∆
Ψ
m
%
Cell
∆Ψ
m
%
Cell
Cpd
Cpd
b
d
b
d
recovery
viability
recovery
viability
c
c
production
production
SVH07
46%
48%
56%
50%
56%
53%
61%
61%
62%
55%
67%
70%
56%
59%
55%
48%
63%
59%
8%
-3%
14%
-15%
1%
97%
91%
82%
54%
99%
68%
91%
92%
91%
29%
82%
99%
99%
96%
99%
73%
99%
98%
39
40
58%
35%
66%
69%
61%
66%
69%
50%
50%
56%
57%
56%
12%
41%
52%
45%
17%
55%
12%
19%
-50%
26%
-9%
22%
4%
88%
61%
97%
99%
78%
95%
94%
99%
93%
40%
100%
91%
98%
100%
97%
90%
96%
90%
100%
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
41
42
43
-4%
44%
22%
18%
-3%
25%
41%
25%
5%
44
45
7%
46
- 2%
- 9%
- 17%
- 11%
9%
47
48
49
50
51
- 3%
17%
- 8%
9%
52
0%
53
6%
54
8%
55
0%
0%
CsA
- 46%
9%
Ro5-
3
8
56%
33%
87%
4864
a
b
All data are reported as the average of duplicates. The increase of fluorescence ratio (green/red) after the
treatment of each compound (5 µM) and Aβ (5 µM) in HT22 cell was normalized by dividing the difference of the
c
ratio between 0% (Aβ-induced damaged condition) and 100% (normal condition in the absence of Aβ). Recovery
d
of ATP production at 5 µM of each test compound against Aβ-induced mitochondrial ATP reduction. HT22 cell
viability measured from the MTT assay after the treatment with 5 µM of each compound only.
Cellular ATP production is another important measure to assess mitochondrial function;
therefore, we also determined percent recovery values of ATP production in HT-22 cells treated
16

ACCEPTED MANUSCRIPT
with each compound in the presence of A
β. Both SVH07 and Ro5-4864 appeared to recover
ATP production slightly (8% and 9% respectively) whereas CsA (-46%) did not have any
beneficial effect on the ATP production suggesting that these compounds may act via distinct
mechanisms. We observed that the S-alkylated analogues generally demonstrated superior
percent recovery to the N-alkylated analogues. More importantly, we noticed a clear trend
1
showing that most of compounds with a methyl group in the R position demonstrated excellent
1
percent recovery values (18% to 44%, compounds 26-28, 30-31, 38, 39, 41) compared to their R
2
=
H counterparts (-15% to 25%, compounds 21-25, 32-37). When we included the R substituent
1
2
for further analysis, we noticed that compounds with R = Me and R = Et groups showed
1
2
generally high percent recovery values, and compounds with R = R = Me groups followed
1
2
thereafter. Considering that SVH07 has R = H and R = Et as the substituents, addition of a
1
small hydrophobic group in the R position while replacing the ethyl group with a methyl group
2
in the R position greatly enhanced the recovery of mitochondrial function. The substituents in
3
the R position also appeared to affect biological activity. Specifically, compounds having the N-
3
3
methyl-N-phenyl group in the R position (R = A; compounds 26, 32, 38, 43, 52, except 47)
1
2
demonstrated the best activity within analogues containing the same R and R group, whereas
3
compounds with the N-benzylamine group in the same position (R = D; compounds 23, 29, 35,
and 40) showed minimal activity or even further suppressed ATP production.
When we analyzed the percent recovery values of ∆Ψm and ATP production in parallel
3
with cell viability data, it was clear that all compounds having the R = D (23, 29, 35, and 40)
demonstrated low cell viability (29% to 73%), suggesting that this specific substituent might be
responsible for low activity and toxicity. Given the assay results from all three reference
compounds (SVH07, CsA, and Ro5-4864), we considered that compounds showing the percent
17

ACCEPTED MANUSCRIPT
recovery values greater than 50% of ∆Ψm and 10% of ATP production with higher than 90% of
cell viability could be suitable candidates for further development. Overall, we were able to
identify a total of six compounds (26-28, 31, 32, and 41) that can restore mitochondrial function
under Aβ-induced cellular toxicity.
2.5. Binding affinity to TSPO
Our virtual hit compound SVH07 was generated based on the pharmacophore model of
Ro5-4864, a known TSPO ligand; therefore, we decided to measure the binding affinity of the
identified analogues by in vitro competition assays. As summarized in Table 2, the binding
affinity to TSPO was measured with the selected compounds (26, 27, 28, 31, 32, and 41)
together with SVH07 and two representative TSPO ligands PK11195 (4) and Ro5-4864. SVH07
and its three derivatives (26, 32, and 41) showed high binding affinity to TSPO with the IC50
values in a nanomolar range (SVH07, IC = 24.8 ± 8.0 nM; 26, IC = 39.9 ± 15.0 nM; 32, IC
50
5
0
50
=
55.1 ± 9.8 nM; and 41, IC50 = 122.5 ± 12.4 nM). These high affinity analogues generally
showed excellent percent recovery values of ∆Ψm and ATP production, supporting our
hypothesis that TSPO ligands may participate in the regulation of mitochondrial function.
3
Interestingly, the three most potent compounds, SVH07, 26, and 32 have the same R substituent,
3
the N-methyl-N-phenyl group (R = A), which demonstrated beneficial effects on ATP
3
production, suggesting that the specific binding interaction between the R group and TSPO
might be particularly important to maintain mitochondrial function.
It should be noted that compounds 26-28, and 31 share a similar structure, differing only
3
in the R group; however, compound 26 showed a remarkably high binding affinity compared to
compounds 27, 28, and 31. Given that these compounds also demonstrated comparable activity
18

ACCEPTED MANUSCRIPT
for the recovery of mitochondrial function, we believe that they may interact with different
binding regions or act on an additional cellular target.
a
Table 2. Competitive binding affinity to TSPO.
b
Compound
SVH07
IC50 (nM)
24.8 ± 8.0
2
2
2
3
3
4
6
7
8
1
2
1
39.9 ± 15.0
1,667.0 ± 1,061.2
> 10,000
2,225.0 ± 1,084.5
55.1 ± 9.8
122.5 ± 12.4
7.5 ± 2.1
4
(PK11195)
(Ro5-4864)
5
11.1 ± 2.3
a
50
3
b
IC values were measured based on the competition binding of [ H]PK11195 by using the rat cerebral cortex;
Mean ± SD
Although none of the N-alkylated derivatives demonstrated a satisfactory activity in the
1
2
3
cell-based assays, we additionally tested compound 47 which has the same R , R , and R groups
as SVH07 for a comparison. The IC value of compound 47 was 157.3 ± 68.6 nM, 6.3 times
5
0
higher value than SVH07, suggesting that the S-alkylated derivatives may have stronger binding
interactions than the N-alkylated analogues such as 47.
We also tested SVH07 and compound 26 for the surface plasmon resonance (SPR)
studies to confirm their potent binding affinity to TSPO. As shown in Figure 6, the binding
affinity (K
assays (SVH07, K
.45 nM).
D
) of each compound corresponded to the IC50 values from the competitive binding
D
= 39.6 ± 26.6 nM; 26, K = 61.2 ± 48.8 nM; and PK11195 (4), K = 11.1 ±
D
D
0
19

ACCEPTED MANUSCRIPT
SVH07, K = 39.6 ± 26.6 nM
D
A
1
1
1
1
6
4
2
0
8
6
4
2
0
2
5
1
0
0
µM
µM
.5 µM
.01 µM
0.001 µM
-
-
50
0
50
100 150 200 250 300 350 400
Time (sec)
Compound 26, K = 61.2 ± 48.8 nM
D
B
1
1
1
1
6
4
2
0
8
6
4
2
0
2
5
1
0
0
0
µM
µM
.5 µM
.01 µM
.001 µM
-
-
50
0
50
100 150 200 250 300 350 400
Time (sec)
Figure 6. Surface plasmon resonance (SPR) sensorgrams of SVH07 (A) and compound 26 (B).
The K values are expressed as mean ± SD from three independent measurements.
D
20

ACCEPTED MANUSCRIPT
2
.6. In vitro metabolic stability
In vitro metabolic stability of the selected compounds was assessed by measuring
cytochrome P450 (CYP) activity and microsomal and plasma stability. First, we determined the
remaining activity of a panel of CYP isozymes in the presence of each compound at 10
Table 3). SVH07 appeared to inhibit four out of the five isozymes (1A2, 2D6, 3A4, and 2C19)
µM
(
with a remaining activity less than 50 %. Similarly, compounds 27, 31, and 32 inhibited four
different isozymes, leaving only either 2C9 or 2D6 with a remaining activity greater than 50%.
Compounds 28 and 41 inhibited greater than 50% of activity of multiple isozymes (3A4, 2C19
for 28, 2D6, 2C9, and 2C19 for 41). Although, compound 26 inhibited all of the tested isozymes
to some extent, all of these isozymes still maintained greater than 50% of enzyme activity;
therefore, we further tested compound 26 for liver microsomal stability and plasma stability.
Table 3. CYP inhibition.
Remaining activity at 10
CYP2D6 CYP2C9
49.5
µ
M (% of control)
Compound
SVH07
CYP1A2
33.7
CYP3A4
CYP2C19
16.8
70.5
64.8
95.9
50.0
38.8
124.4
21.1
32.2
64.1
34.1
44.7
20.6
41.3
51.4
2
2
2
3
3
4
6
7
8
1
2
1
50.7
49.2
65.5
40.1
46.8
70.9
60.4
17.0
54.2
81.7
48.8
38.4
52.2
44.1
30.9
20.3
28.8
10.2
21

ACCEPTED MANUSCRIPT
As described in Table 4, SVH07 and compound 26 showed excellent plasma stability in
human and rat. However, both compounds demonstrated poor liver microsomal stability in all
the tested species. Compound 26 appeared to be slightly more stable than SVH07, although less
than 15% of the compound remained intact after 30 min. On the basis of these findings, we
decided not to pursue further in vivo studies with compound 26 despite its excellent in vitro
activity. Our future study will focus on the optimization of compound 26 to improve its drug-like
properties, especially liver microsomal stability.
Table 4. Liver microsomal stability and plasma stability of compounds SVH07 and 26.
Liver microsomal stability
Plasma Stability
(% Remaining)
(
% Remaining during 30 min)
Cpd
Human
Rat
Human Dog
Rat Mouse
(%)
(%)
(%)
(%)
3
0 min 120 min 30 min 120 min
SVH07
5.0
8.8
5.1
9.8
6.4
9.0
8.2
> 100
99.3
96.8
93.6
94.0
96.6
72.4
80.6
2
6
12.8
3
. Conclusion
In this work, we developed a series of TSPO-targeted mitochondrial functional
modulators that contain a thieno[2’,3’:4,5]pyrrolo[1,2-d][1,2,4]triazine group as the core scaffold.
The thienopyrrolotriazine scaffold was identified via a ligand-based pharmacophore model and
virtual screening efforts using Ro5-4864, a neuroprotective TSPO ligand. Virtual screening and
cell-based assays of a commercial chemical library afforded SVH07 as the best hit. Based on the
chemical structure of SVH07, we designed and synthesized a series of analogues that can be
divided into the S-alkylated and the N-alkylated derivatives. Among these analogues, compound
2
6 demonstrated excellent in vitro activity, reversing mitochondrial membrane depolarization
22

ACCEPTED MANUSCRIPT
and impaired ATP production in cells under A
β
-induced mitochondrial dysfunction. Competitive
binding assays and SPR studies confirmed the virtual hit SVH07 and compound 26 directly
interact with TSPO. CYP inhibition assays indicated that compound 26 moderately inhibited
catalytic activity of the tested isozymes, however all the tested isozymes still maintained over 50%
of activity. Compound 26 also exhibited excellent plasma stability, however, demonstrated poor
liver microsomal stability, less than 15% of remaining compound after 30 min. Although we
decided not to pursue further in vivo studies with compound 26 due to its poor metabolic
stability, we believe that compound 26 is a promising lead compound for the development of
TSPO-targeted mitochondrial functional modulators. Further optimization of drug-like properties
of compounds containing the thienopyrrolotriazine scaffold is currently underway, potentially
offering a novel therapeutic option for the treatment of AD.
4
4
4
. Experimental Section
.1. Molecular modeling
.1.1. Generation of a shape-added pharmacophore model and virtual screening
Shape-added pharmacophores were generated with Ro5-4864. First of all, a 3D
pharmacophore model was generated following the standard procedure for a single molecule of a
commercial program, Discovery studio 4.0 (Accelrys). And then shape query was added. A
shape query is defined by a set of 3D coordinates, each with a corresponding radius. Typically,
these points are the centers of atoms in a template molecule, and the radius associated with each
point depends on the atom type. Shapes can be constructed automatically from a single
conformation of a molecule or from a set of aligned molecules. Here shape matching is done
using shape similarity. That is, a shape query defines a volume in space that is the union of a set
23

ACCEPTED MANUSCRIPT
of spheres. During a search, each candidate molecule is converted to a similar volume. The two
volumes are aligned, and their intersection and union are approximated. The similarity of the
shape query to the molecule is computed as the volume of the intersection divided by the volume
of the union of the aligned shapes. Depending on the similarity tolerance that is set, a hit may
have more or less volume that protrudes from the shape. A shape in a mixed pharmacophore that
contains pharmacophore features and a shape component is processed somewhat differently from
a shape-only query. In a shape-only query, the shape is aligned to each candidate compound, the
similarity is computed, and, if the similarity tolerance is satisfied, the candidate is considered a
hit. The shape in a mixed query is also aligned to the candidate compound, and the tolerance is
tested. In addition, the transformation used to align the location constraints to the molecule is
applied to the shape. Then, the similarity of the shape to the molecule is computed without any
further alignment. If this similarity value satisfies the tolerance, then the candidate is considered
a hit.
Virtual screening was carried out by using the generated shape-added pharmacophore model
to obtain new compounds. The commercial library Asinex (AsinexGold, 229,398 compounds,
AsinexPlatinum, 125,231 compounds, Asinex, Moscow, Russia, www.asinex.com) and
ChemDiv (693,042 compounds, ChemDiv, Inc. California, USA, www.chemdiv.com) have been
utilized for virtual screening. The best pharmacophore model was used for the virtual screening
experiment, by selecting the fast flexible database search option. The first screening selected new
compounds with similar functional and spatial properties defined in 3D pharmacophore query.
Initially compounds were selected by the highest fit value, which was calculated from the
Discovery studio 4.0 program by aligning to the generated shape-added pharmacophore model.
24

ACCEPTED MANUSCRIPT
Compounds obtained by fit value were further filtered by visual inspection. Selected compounds
were purchased for biological assay.
4
.1.2 Docking Study
Docking study was performed by Schrodinger software (Maestro, v9.2, Schroꢀdinger LLC,
New York, NY). Ligands were prepared at pH 7.4 using the LigPrep wizard with OPLS force
field 2005. The Human TSPO homology model from our previous study[42] was utilized for this
docking study. Compounds were docked into the hydrophobic central cavity using extra
precision (XP) – flexible docking methodology. Images with detail ligand-protein interactions
were prepared using Discovery studio software.
4
.2. Chemistry
All reactions were performed with commercially available reagents without further
purification processes in an oven-dried glassware under a nitrogen atmosphere. All reactions
were carefully monitored by analytical thin-layer chromatography (TLC) plates (Merck, catalog
no. 1.05715) with UV light (
λ
= 254 nm). In general, the reaction products were purified by flash
1
column chromatography on silica gel 60 (Merck, catalog nos. 1.07734). H (300 or 400 MHz)
1
3
and C (75 or 100 MHz) NMR spectra were recorded using an internal standard,
tetramethylsilane (TMS), in CDCl or in DMSO-d . Chemical shifts ( ) are recorded in parts per
δ
3
6
million (ppm) values relative to TMS, and the coupling constants (J) are recorded in hertz (Hz)
values. Melting points were measured by using an OptiMelt melting point apparatus (Stanford
Research System, Inc.) without correction. The purity of synthesized products were verified to be
higher than 95% by using an analytical HPLC (Waters E2695 system) with SunFire C18 column
25

ACCEPTED MANUSCRIPT
(4.6 mm × 150 mm; 5
µ
2
m). HPLC data were recorded using parameters as follows: H O/MeCN,
7
0/30
→ 0/100 in 20 min, +3 min isocratic, flow rate of 1.0 mL/min, λ = 254 and 280 nm.
4
.2.1. Representative procedure for the preparation of methyl-2-azido-3-(thiophen-2-
yl)acrylate (8a).
To a stirred solution of NaOMe 25 wt. % in MeOH (2.97 mL, 13.7 mmol) and methyl
azidoacetate (1.58 mL, 13.7 mmol) in dehydrated MeOH (8 mL) was added 2-
o
thiophenecarbaldehyde 7a (361 mg, 3.93 mmol) dropwise at 0 C. The reaction mixture was
o
stirred for 30 min at 0 C and for 4 h at room temperature. And then, an aqueous solution of
saturated NH
4
Cl was added, and the resulting mixture was stirred for 30 min at room temperature.
~ 30 mL) and washed with H O (~ 30 mL).
The resulting solution was extracted with EtOAc (3
×
2
The organic layer was dried over anhydrous MgSO , and concentrated in vacuo. The residue was
4
purified by column chromatography (SiO2, n-hexane/EtOAc 8/1) to yield the title product 8a
1
(
632 mg, 77%) as yellow oil; R
f
= 0.50 (n-hexane/EtOAc 8/1); H NMR (400 MHz, CDCl
3
)
δ
7
.53-7.55 (m, 1H), 7.35 (d, J = 5.2 Hz, 1 H), 7.21 (s, 1H), 7.11 (dd, J = 5.2, 3.6 Hz, 1H), 3.93 (s,
1
3
1
H); C NMR (101 MHz, CDCl )
δ
163.66, 136.57, 132.08, 130.52, 127.08, 122.50, 119.56,
3
5
2.76.
4
.2.1.1. Preparation of methyl-2-azido-3-(5-methylthiophen-2-yl)acrylate (8b).
Following the procedure of 8a, NaOMe 25 wt. % in MeOH (684
µL, 3.17 mmol), methyl
azidoacetate (308 µL, 3.17 mmol), and 5-methylthiophene-2-carbaldehyde 7b (100 mg, 0.79
f
mmol) in MeOH (1.2 mL) gave the title product 8b (136 mg, 77%) as yellow oil; R = 0.70 (n-
1
hexane/EtOAc 8/1); H NMR (300 MHz, CDCl
3
)
δ
7.16 (d, J = 3.6 Hz, 1 H), 7.13 (s, 1H), 6.77
(m, 1H), 3.92 (s, 3H), 2.57 (s, 3H).
4
.2.1.2. Preparation of methyl-2-azido-3-(5-chlorothiophen-2-yl)acrylate (8c).
26

ACCEPTED MANUSCRIPT
Following the procedure of 8a, NaOMe 25 wt. % in MeOH (11.8 mL, 54.4 mmol), methyl
azidoacetate (3.91 mL, 34.0 mmol), and 5-chlorothiophene-2-carbaldehyde 7c (2.0 g, 13.6 mmol)
in MeOH (34 mL) gave the title product 8c (3.26 g, 98%) as yellow oil; R = 0.60 (n-
f
1
hexane/EtOAc 4/1); H NMR (300 MHz, CDCl )
δ 7.08 (d, J = 3.9 Hz, 1 H), 7.06 (s, 1H), 6.92
3
(d, J = 3.9 Hz, 1H), 3.93 (s, 3H).
4
.2.2. Representative procedure for the preparation of methyl 4H-thieno[3,2-b]pyrrole-5-
carboxylate (9a).
A solution of methyl-2-azido-3-(thiophen-2-yl)acrylate 8a (500 mg, 2.39 mmol) in toluene (12
mL) was heated under the reflux for 12 h and then cooled to room temperature. The reaction
mixture was extracted with EtOAc (3
×
~ 30 mL) and washed with H O (~ 30 mL). The organic
2
layer was dried over anhydrous MgSO
4
, and concentrated in vacuo. The residue was purified by
column chromatography (SiO2, n-hexane/EtOAc 8/1) to yield the title product 9a (216 mg, 50%)
1
as a yellow solid; R = 0.30 (n-hexane/EtOAc 8/1); H NMR (300 MHz, CDCl )
δ 9.06 (br s, 1H),
f
3
1
3
7
.37 (d, J = 5.3 Hz, 1 H), 7.18 (s, 1H), 7.00 (dd, J = 4.8, 0.9 Hz, 1H), 3.94 (s, 3H); C NMR
101MHz, CDCl3) 162.17, 141.48, 129.61, 126.67, 124.81, 111.12, 107.72, 51.77.
(
δ
4
.2.2.1. Preparation of methyl 2-methyl-4H-thieno[3,2-b]pyrrole-5-carboxylate (9b).
Following the procedure for 9a, methyl-2-azido-3-(5-methylthiophen-2-yl)acrylate 8b (100 mg,
0
0
1
1
4
f
.45 mmol) in toluene (2.2 mL) gave the title product 9b (60 mg, 69%) as a yellow solid; R =
1
.15 (n-hexane/EtOAc 8/1); H NMR (300 MHz, CDCl )
δ
8.97 (br s, 1H), 7.32 (s, 1 H), 7.08 (s,
3
13
H), 6.68 (s, 1H), 3.92 (s, 3H); C NMR (101 MHz, CDCl )
δ 162.18, 145.05, 141.28, 124.98,
3
23.26, 109.38, 107.88, 51.61, 17.01.
.2.2.2. Preparation of methyl 2-chloro-4H-thieno[3,2-b]pyrrole-5-carboxylate (9c).
27

ACCEPTED MANUSCRIPT
Following the procedure for 9a, methyl-2-azido-3-(5-chlorothiophen-2-yl)acrylate 8c (790 mg,
3
0
1
1
4
.24 mmol) in toluene (16 mL) gave the title product 9c (656 mg, 94%) as a yellow solid; R =
f
1
.25 (n-hexane/EtOAc 4/1); H NMR (300 MHz, CDCl )
δ
9.07 (br s, 1H), 7.07 (s, 1 H), 6.93 (s,
3
1
3
H), 3.98 (s, 3H); C NMR (75 MHz, CDCl )
δ 161.78, 138.04, 133.57, 125.79, 122.26, 111.14,
3
07.67, 51.83.
.2.3. Representative procedure for the preparation of 4H-thieno[3,2-b]pyrrole-5-
carbohydrazide (10a)
To a stirred solution of methyl 4H-thieno[3,2-b]pyrrole-5-carboxylate 9a (200 mg, 1.10 mmol)
in anhydrous EtOH (5 mL) was added hydrazine monohydrate (268
µ
L, 5.52 mmol). The
o
reaction mixture was heated under the reflux for 18 h and then cooled to 0 C. The precipitate
was filtered off, washed on the filter with H O (5 mL) and cold EtOH (5 mL), then dried in
2
vacuo to give the title product 10a as a white solid (174 mg, 87%); R = 0.10 (n-hexane/EtOAc
f
1
3
7
4
/1); H NMR (300 MHz, DMSO-d )
δ
11.70 (br s, 1H), 9.48 (br s, 1H), 7.35 (d, J = 5.4 Hz, 1H),
6
.04 (s, 1H), 6.95 (d, J = 5.4 Hz, 1H), 4.39 (br s, 2H).
.2.3.1. Preparation of 2-methyl-4 -thieno[3,2- ]pyrrole-5-carbohydrazide (10b)
H
b
Following the procedure of 10a, methyl 2-methyl-4H-thieno[3,2-b]pyrrole-5-carboxylate 9b
(210 mg, 1.07 mmol) and hydrazine monohydrate (209
µ
L, 4.30 mmol) in EtOH(5.5 mL) gave
1
the title product 10b (190 mg, 91%) as a white solid; R = 0.20 (n-hexane/EtOAc 3/1); H NMR
f
(300 MHz, DMSO-d6) δ 11.51 (br s, 1H), 9.37 (br s, 1H), 6.95 (s, 1H), 6.69 (s, 1H), 4.35 (br s,
2
H), 2.47 (s, 3H).
.2.3.2. Preparation of 2-chloro-4H-thieno[3,2-b]pyrrole-5-carbohydrazide (10c)
4
Following the procedure of 10a, methyl 2-chloro-4H-thieno[3,2-b]pyrrole-5-carboxylate 9c (500
mg, 2.31 mmol) and hydrazine monohydrate (450 L, 9.27 mmol) in EtOH (23 mL) gave the
µ
28