Bioorganic & Medicinal Chemistry Letters
High-throughput screening of bioactive compounds via new catalytic
reaction in the pooled mixture
a,⁎
b,c,⁎
Ayano Satoh , Yuta Nishina
b
c
a
Graduate School of Interdisciplinary Science and Engineering in Health Systems, Okayama University, Okayama, Japan
Graduate School of Natural Science and Technology, Okayama University, Okayama, Japan
Research Core for Interdisciplinary Sciences, Okayama University, Okayama, Japan
A R T I C L E I N F O
A B S T R A C T
Keywords:
To increase the chances of finding new candidate molecules with medicinal properties, while expending less
resource and effort, the present study used pooled substrates as starting materials. A bisindole compound that
showed inhibitory activity was then isolated from the mixture, and the activity was improved by optimizing the
substituents on the indole skeleton.
Golgi
Secretion
Brefeldin A
Bisindole
High-throughput screening
High-throughput screening of compound libraries has long been a
mainstay of drug discovery. However, this type of drug discovery
generally requires a large amount of resource and effort since enormous
numbers of chemicals need to be synthesized, isolated, and screened.
Therefore, more efficient ways to identify drug candidates are required.
Hartwig and MacMillan independently attempted to use pools of
to improve the performance. Using this method, we found a series of
bisindole compounds inhibiting secretion.
Results and discussion
We selected starting substrates so as to produce heterocycle- and
fluorine-containing molecules, since the addition of heterocycle func-
tionality to a compound generally confers diverse physical, chemical,
1
,2
substrates to identify unexpected catalytic reactions. Each substrate
pool, containing common functional groups, was systematically reacted
with another substrate pool and a catalyst in a small reaction volume.
The resultant reaction mixtures were analyzed by gas chromato-
graphy–mass spectrometry (GC–MS) to identify newly formed products.
Using this approach, several expected and unexpected reaction pro-
ducts were found and screened simultaneously. This reduced the effort
required to find new reactions, as compared to reacting individual
substrates. In this context, we have developed a method that uses a pool
of substrates to find new chemicals with particular pharmacological
functions or activities. In our method shown in Fig. 1, (1) a pool of
substrates and a catalyst are reacted to generate a pooled catalytic re-
action mixture, which is composed of many expected and unexpected
products obtained via catalytic processes, in addition to the remaining
substrates. Using the mixture, (2) modulation of protein secretion are
evaluated, as this could indicate an effect on intracellular trafficking. In
case there is no protein secretion response, move back to (1). When a
positive response is observed, (3) each compound is isolated by column
chromatography, distillation, and/or recrystallization. After isolation,
3
–5
and biological properties.
In particular, functionalized heterocyclic
compounds with fluorine-containing substituents often exhibit phar-
macokinetic advantages, such as improved drug stability and absorp-
6
tion in the body.
The present study utilized pooled compounds as starting materials,
as shown in Fig. 2A. The pool of compounds, consisting of 11 chemicals
and 1 catalyst, was heated at 120 °C for 24 h to generate the pool. Then,
the pool was screened for cytotoxicity and inhibition of secretion. The
diluted pool was incubated with the same number of HeLa cells stably
expressing soluble secretory alkaline phosphatase. Three hours later,
the number of live cells and the enzymatic activity of alkaline phos-
phatase in the culture medium were measured in order to evaluate
cytotoxicity and secretion inhibition, respectively. As shown in Fig. 2B,
the pool (labeled as “Mixture”) inhibited secretion (right panel),
without producing cytotoxicity (left panel). Brefeldin A (BFA; 18 µM)
was included as a positive control. This fungal metabolite is known to
inhibit secretion and to disrupt the Golgi.
(
4) protein secretion is again performed to identify an active com-
This pool was further analyzed by GC–MS to identify individual
compounds (Fig. 2C). After isolating those compounds (8 identified and
pound. Finally, (5) derivatives of the active compound are synthesized
⁎
Received 16 March 2019; Received in revised form 19 June 2019; Accepted 28 June 2019
Available online 29 June 2019
0960-894X/ © 2019 Elsevier Ltd. All rights reserved.