Discovery of novel phosphatidylcholine-specific phospholipase C drug-like inhibitors as potential anticancer agents

Article abstract of DOI:10.1016/j.ejmech.2019.111919

Phosphatidylcholine–specific phospholipase C (PC-PLC) is a promising target for new anticancer treatment. Herein, we report our work in the discovery of novel drug-like PC-PLC inhibitors. Virtual screening led to the identification of promising hits from

Full text of DOI:10.1016/j.ejmech.2019.111919

Journal Pre-proof
Discovery of novel phosphatidylcholine-specific phospholipase C drug-like inhibitors
as potential anticancer agents
Chatchakorn Eurtivong, Lisa I. Pilkington, Michelle van Rensburg, Reuben M. White,
Harpreet Kaur Brar, Shaun Rees, Emily K. Paulin, Chris Sun Xu, Nabangshu Sharma,
IvanhoeK.H. Leung, Euphemia Leung, David Barker, Jóhannes Reynisson
PII:
S0223-5234(19)31071-2
DOI:
https://doi.org/10.1016/j.ejmech.2019.111919
EJMECH 111919
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 23 September 2019
Revised Date: 11 November 2019
Accepted Date: 26 November 2019
Please cite this article as: C. Eurtivong, L.I. Pilkington, M. van Rensburg, R.M. White, H.K. Brar, S.
Rees, E.K. Paulin, C.S. Xu, N. Sharma, I.H. Leung, E. Leung, D. Barker, Jó. Reynisson, Discovery of
novel phosphatidylcholine-specific phospholipase C drug-like inhibitors as potential anticancer agents,
European Journal of Medicinal Chemistry (2019), doi: https://doi.org/10.1016/j.ejmech.2019.111919.
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.

Discovery of Novel Phosphatidylcholine-Specific Phospholipase C
Drug-like Inhibitors as Potential Anticancer Agents
Chatchakorn Eurtivong,1, 2 Lisa I. Pilkington,1 Michelle van Rensburg,1
Reuben M. White,1 Harpreet Kaur Brar,1 Shaun Rees, 1Emily K. Paulin, 1 Chris
Sun Xu,3 Nabangshu Sharma,1 Ivanhoe K. H. Leung,1,4 Euphemia Leung,3 David
Barker1 and Jóhannes Reynisson*1,5
1 School of Chemical Sciences, The University of Auckland, Private Bag 92019, Victoria
Street West, Auckland 1142, New Zealand
2 Chemical Science Program, Chulabhorn Graduate Institute, Chulabhorn Royal Academy
of Science, Bangkok 10210, Thailand
3 Auckland Cancer Society Research Centre, The University of Auckland, Private Bag
92019, Victoria Street West, Auckland 1142, New Zealand
4 Maurice Wilkins Centre for Molecular Biodiscovery, The University of Auckland,
Private Bag 92019, Victoria Street West, Auckland 1142, New Zealand
5 School of Pharmacy and Bioengineering, Keele University, Hornbeam building,
Staffordshire ST5 5BG, United Kingdom
*To whom correspondence should be addressed: School of Pharmacy and Bioengineering,
Keele University, Hornbeam building, Staffordshire ST5 5BG, United Kingdom
E-mail: j.reynisson@keele.ac.uk, Tel. +44 (0)1782 733985
Abstract
Phosphatidylcholine–specific phospholipase C (PC-PLC) is a promising target for new
anticancer treatment. Herein, we report our work in the discovery of novel drug-like PC-PLC
inhibitors. Virtual screening led to the identification of promising hits from four different
structural series that contain the molecular scaffold of benzenesulphonamides (10),
pyrido[3,4-b]indoles (22), morpholinobenzoic acid (84) and benzamidobenzoic acid (80). 164
structural analogues were tested to investigate the chemical space around the hit series and to
generate preliminary structurally activity relationships (SAR). Two of the pyrido[3,4-
b]indoles (22_10 and 22_15) had comparable or better potency as D609, an established but
non-drug-like PC-PLC inhibitor. Furthermore, three morpholinobenzoic acids (84, 84_4 and
1

84_5) had superior potency than D609. Therefore, this study paves the way towards the
development of drug-like PL-PLC inhibitors as potential anticancer agents.
Keywords: virtual screening; similarity searching; molecular modelling; synthesis;
pyrido[3,4-b]indoles; morpholinobenzoic acid; chemical space
1. Introduction
Phospholipase C (PLC) is a family of enzymes that mediate diverse cellular functions [1].
PLCs cleave phospholipids to generate by-products that play central roles in regulation of
cellular processes: proliferation, differentiation, motility, apoptosis and gene expression [1-
2]. A common by-product is diacylglycerol (DAG) that activates protein kinase C (PKC) to
further phosphorylate downstream proteins and propagate an array of signalling events [1-2].
Unregulation of these signalling events can lead to the development of cancer. Thus, the
homeostasis of the signalling pathways is crucially important to cellular health and to prevent
the development of diseases such as cancer [1-2].
Phosphatidylcholine – specific phospholipase C (PC-PLC) is a PLC subtype that functions by
preferentially cleaving phosphatidylcholine to generate DAG and phosphocholine as by-
products. A wealth of evidence suggests that PC-PLC is involved in cancer development,
including; increased expression in ovarian tumour cells [3] and highly metastatic triple-
negative MDA-MB-231 breast cancer cell line [4], downregulation of HER2 oncogene upon
PC-PLC inhibition [5], and linkages to hepatocarcinoma [6] and leukaemia progression [7].
Furthermore, it has been demonstrated that PC-PLC plays a crucial role in the development
of atherosclerosis making PC-PLC and interesting target for the development of
cardiovascular drugs [8].
A handful of PC-PLC inhibitors are known but are not considered to be drug-like. These are
2-aminohydroxamic acids [9], univalent anions [10], N,N’-dihydroxyureas [11], phospholipid
2

analogues [12] and the established xanthate, D609 (see molecular structures in Fig. 1) [13-
14]. D609 is also known to modulate other biochemical targets such as the the K+ channel,
KCNQ1/KCNE1 [15] and sphingomyelin synthase [16]. Hence, it is desirable to identify
novel inhibitors of PC-PLC with drug-like properties and with tractable synthetic routes for
further development.
Fig. 1. Examples of known PC-PLC inhibitors and their reported inhibition activities. IC50 – the 50% inhibition
concentration, Ki – dissociation constants.
The structure of eukaryotic PC-PLC is unknown and only prokaryotic PC-PLC enzymes have
been structurally characterised. The most studied PC-PLC is isolated from Bacillus cereus
[17]. In this paper, we describe a search for hit compounds against human PC-PLC for
potential anticancer drug development using the virtual high throughput screening (vHTS)
methodology, an effective method for identifying novel inhibitors for biomolecular targets
[18-21]. The process relies on the structural similarity between PC-PLCBc and mammalian
PC-PLC, i.e., the bacterial crystal structure was used as a model for its human counterpart.
3

This approach can be justified, as it was reported that PC-PLCBc has antigenic similarity to
mammalian PC-PLC [22] and can mimic similar responses to mammalian PLC, e.g.,
enhancement of prostaglandin biosynthesis [23].
2. Results and Discussion
2.1. Pilot Virtual Screen
1 × 104 molecular entities were downloaded from the ChemBridge diversity collection [24].
These compounds were filtered based on Lipinski’s rules [25], resulting in 8373 drug-like
molecules. The docking scaffold used was the wild-type PC-PLCBc crystal structure (structure
shown in Fig. 2) [17]. It can be seen that the catalytic site has three Zn2+ ions ligated to amino
acids and water molecules. Zn1 is bridged to Zn3 by a water molecule and ligates to residues
Asp55, His69, His118 and Asp122. Zn2 is ligated to two water molecules and amino acid
residues His128, Glu146 and His142. Zn3 is ligated to amino acids Trp1, His14 and Asp122.
Zn3 was selected as the centre of binding as it was reported that the Zn2+ ions play an
important role for the catalytic cleavage by binding to charged phosphate groups on the
phospholipid substrates [11,17,26].
4

His142
Zn2
Glu146
His128
Zn3
Zn1
Trp1
Asp55
His69
Asp122
His14
His118
Fig. 2. The structure of wild-type PC-PLCBc enzyme (PDB: 1AH7) [17]. The phospholipase monomer is
displayed as red ribbons embedded with three catalytic Zn2+ ions and water molecules depicted as grey and red
spheres respectively (left). Coordinating amino acids to the Zn2+ ions are represented as sticks and coloured by
atom types (right).
Four scoring functions, GoldScore (GS) [27], ChemScore (CS) [28,29], Piecewise Linear
Potential (ChemPLP) [30] and Astex Statistical Potential (ASP) [31] were implemented at
30% docking efficiency. Molecules without hydrogen bonding (HB = 0) and those that had
scores GS < 55, CS < 26, ChemPLP < 65 and ASP < 25 were eliminated. The known D609
inhibitor (33) was docked to the catalytic site, and the output scores were used as criteria for
the elimination process (endo-isomer: GS = 65.8, CS = 28.7, ChemPLP = 55.1, ASP = 55.1
and exo-isomer: GS = 76.9, CS = 30.0, ChemPLP = 58.8, ASP = 19.3). D609 (33) was used
as a positive control throughout this study [13,14].
The initial elimination process resulted in 1721 molecules. These were re-docked at a higher
efficiency (100%), i.e., higher quality pose of the ligand was predicted. Molecules with HB =
0, GS < 60, CS < 34, ChemPLP < 65 and ASP < 28 were eliminated. Additionally, molecules
with undesirable reactive, cytotoxic and chemically unstable moieties [32]. The remaining
molecules were checked for the best predicted poses between the scoring functions and
5

whether a plausible binding mode was predicted, e.g., unstrained poses, chemical groups
directed towards Zn2+ ions to form electrostatic interactions and no lipophilic moieties facing
the aqueous environment. In total, thirty-two compounds (see molecular structures 1-33 Fig.
S1 in the Supplementary Information) were identified to be the most promising and were
acquired from ChemBridge [24]. The compounds were tested with the PC-PLC Amplex Red
assay (Molecular Probe, Inc.) that measures the activity of PC-PLCBc and the results are
shown in Fig. 3.
90
80
70
60
50
40
30
20
10
0
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
Compounds
Fig. 3. Inhibition of PC-PLCBc activity by the 32 virtual hits from the pilot screen at 10 µM single concentration
using the Amplex Red biochemical assay. The standard deviations are shown. The positive control used is D609
(33) in green. Active hits were identified as having >15% inhibitory activity resulting in compounds 10 and 22
as bracketed in red.
The ligands were considered to be hits if they achieved levels of enzyme inhibition >15%,
based on the fluorescence intensity emitted relative to the blank. Two hits were identified: 3-
[(4-cyclohexylphenyl)sulfonamido]benzoic acid (10) and 1-[4-(trifluoromethyl)phenyl]-
2,3,4,9-tetrahydro-1H-pyrido[3,4-b]indole-3-carboxylic acid (22) having 25% and 37%
inhibition at 10 µM, the molecular structures are shown in Fig. 4. D609 (33), the accepted
standard, gave inhibition close to 75%. Interestingly, both 10 and 22 have carboxylic acid
6

moieties that can potentially coordinate to the catalytic Zn2+ ions in the binding site
suggesting that they ‘activate’ the phosphate group on the phosphatidylcholine substrate
instigating nucleophilic cleavage steps [26,33].
Fig. 4. The two hits 10 and 22 identified with the Amplex Red biochemical assay for activity against PC-PLC.
The known inhibitor D609 (33) was used as a positive control [13,14]. The chiral centres (*) are shown in 22.
In order to explore the structure activity relationship (SAR) of 10 and 22, similar analogues
were identified using eMolecules.com [34], a website that hosts commercially-available
compounds. Similar ligands were found from ChemBridge [24], Chemical Diversity Inc
(ChemDiv) [35] and InterBioScreen Ltd (IBS) [36]. The chemical series resulting from hits
10 and 22 were classified as the N-phenylbenzenesulphonamides and pyrido[3,4-b]indoles,
respectively, based on their core scaffolds.
2.1.1. N-Phenylbenzenesulphonamides
Hit 10 demonstrated 25% inhibition in the biochemical assay and has a N-
phenylbenzenesulphonamide core. A similarity coefficient of 0.7 was used to search for
structurally
similar
compounds
in
eMolecules.com
[34].
Twelve
N-
7

phenylbenzenesulphonamides were purchased and tested; the molecular structures and results
are given in Table S1 in the Supplementary Information. Only modest inhibitory activities
against PC-PLCBc was achieved by the members of this series and none of the derivatives was
more potent than the original hit 10. A detailed discussion on the SAR and molecular
modelling are given in the Supplementary Information (Pages S7 –S9).
2.1.2. Pyrido[3,4-b]indoles
Hit compound 22 had a measured inhibition of 37%. It had a 2,3,4,9-tetrahydro-1H-
pyrido[3,4-b]indole heterocyclic core with the pyridine moiety linked to a phenyl ring and a
carboxylic acid with unknown stereochemistry at the two chiral centres, shown in Figure 4. A
similarity coefficient value of 0.7 was used to search for structurally similar compounds using
eMolecules.com [34]. Thirty-five indoles were purchased and their structures and inhibition
data are given in Table S2 in the SI.
There is a clear indication of PC-PLCBc inhibition for this series, with several derivatives
showing improved potency over to the original hit (>37%) and even surpassing the
established reference compound 33. Generally, the best compounds have alkoxy and
hydroxyl electron-donating groups at meta and para positions on the phenyl ring. 3,4-
Dihydroxysubstitution on the phenyl rind (compound 22_10) had the strongest inhibition
effect (~75%), followed by 3,5-dimethoxy-4-hydroxy substitution (22_14) at ~60% and
following closely by 22_12, 22_15 (p-N,N-dimethyl), 22_11 and 22_13 ranging between 43 –
53% inhibition. Alkoxy substitutions alone showed modest inhibitory effects (22_3, 22_5,
22_6, and 22_8). Single halogen substitutions did not improved efficacy with the exception
of m-fluoro (22_16), whereas 2,4-dichloro (22_20) demonstrated improved inhibition
(~44%). The remaining substituents on the phenyl ring include nitro (22_26 and 22_25) and
8

m-CF3 (22_23), hydrogens (22_ 1) and alkyls (22_28, 22_27, 22_29, 22_30 and 22_32),
which did not improve potency. Replacing phenyl with pyridine (22_34) did not improve
activity, whereas improved activity was seen with a naphthyl ring (22_35). Methyl (22_33)
substantially reduced inhibition suggesting the importance of bulky aromatic systems.
Introducing acetyl on the piperidine ring at R5 (22_31) had detrimental effect on the potency.
Molecular modelling was conducted to understand the inhibition mechanisms of these
ligands. Since they were tested as potential racemic mixtures, the diastereoisomer
corresponding to the best activity was unclear. This problem was investigated by docking all
four diastereoisomers (Table 1) using the most active compound 22_10. The results revealed
that the (1R,3S)-diastereoisomer scored best for the CS, ChemPLP and ASP functions. It also
scored comparably well to the highest GS scoring (1S,3S)-diastereoisomer. It can therefore
be argued that the (1R,3S)-diastereoisomer is predicted to be the most active. Interestingly,
visual inspection showed plausible binding modes (see. Fig. 5) for all four diastereoisomers,
e.g., phenyl and indole moieties occupy lipophilic areas and the carboxylic acid moiety is
facing the Zn2+ ions.
Table 1 Docking scores of the four possible diastereoisomers of the most active ligand 22_10.
Compound
PC_079 (SS)
PC_079 (RS)
PC_079 (RR)
PC_079 (SR)
Compound
GS
CS
ChemPLP
84.3
ASP
43.8
41.5
42.4
45.5
75.6
69.5
64.1
73.8
45.0
44.1
43.0
46.4
22_10 (1S,3S)
22_10 (1S,3R)
22_10 (1R,3R)
22_10 (1R,3S)
83.5
77.6
95.3
9

(1S,3R)
(1S,3S)
(1R,3R)
(1R,3S))
Fig. 5. The docked configurations of the four 1-(3,4-dihydroxyphenyl)-2,3,4,9-tetrahydro-1H-pyrido[3,4-
b]indole-3-carboxylic acid (22_10) diastereomers in the PC-PLCBc binding site. The protein surface is rendered.
Red and blue surfaces depicts negative and positive partial charge respectively whereas grey shows neutral
lipophilic areas. The hydrogens are removed for the ligands for clarity.
All 1-phenyl-(1R,3S)-2,3,4,9-tetrahydro-1H-pyrido[3,4-b]indoles 22_1 to 22_32, despite
different substitution patterns on the phenyl ring, show high similarity in binding modes and
scores. The general binding mode and interactions are represented by 22_10 in Fig. 6. The
phenyl group is situated close to lipophilic areas partly consisting of Phe66 and Phe70
residues where they are expected to form π−π stacking, whereas the tricyclic 2,3,4,9-
tetrahydro-1H-pyrido[3,4-b]indole heterocycle occupies a lipophilic region formed by
residues Ala3, Tyr56 and Phe66. Hydrogen bonding was predicted with His69 via the
carboxylic acid. In some cases, Glu146 formed hydrogen bond with pyridine, e.g., ligand
22_15. The carboxylic acid functional group facing the deeply buried positively charged
regions are expected to form coordination bonds with the Zn2+ ions.
10

Ala3
His69
Phe66
Tyr56
Phe70
Figure 6. Binding interactions of (1R,3S)-22_10 in the PC-PLCBc binding site has hydrogen bonding with His69
residue depicted as a green dotted line. Zn2+ ions are depicted as unconnected grey spheres and are coordinated
by the carboxylic acid functionality shown as red dotted lines. Hydrophobic amino acid residues Ala3, Tyr56,
Phe66 and Phe70 are shown. The hydrogens are removed for the ligand and amino acid residues for clarity.
In order to extend the SAR of the indoles further, thirty-six derivatives were synthesised. A
detailed description of the synthesis, synthetic procedures, characterisation of compounds is
given in the Supplementary Information (Pages S12 – S34). The stereochemistry of these
ligands was controlled for the carbon 3 site to better understand its effect. The derivatives
were therefore tested as isomeric mixtures of cis/trans isomers. Three main types of ligands
were made; carboxylic acids, their esters and planar aromatic derivatives with both chiral
centres removed. The phenyl ring was decorated with methoxyl and hydroxyl substituents.
The molecular structures and their activity against PC-PLC are given in Tables S9 – S11 in
the SI.
Unfortunately, only three of the synthesised derivatives had good potency, two carboxyl acid
derivatives and one ester. The aromatisation of the piperidine moiety to pyridine rendered the
derivatives marginally active or simply inactive. The acids (3S)-22_S5 and (3R)-22_S8 do
not have the same stereochemistry but both are active, supporting the modelling of 22_10,
i.e., all four stereoisomers can fit into the binding pocket. The ester derivative (3S)-22_S15
11

has the same substitution pattern on the phenyl ring as (3S)-22_S5, m-OCH3 and p-OH. A
detailed discussion of the SAR is given in the SI pages S37 and S38.
Interestingly, tadalafil (see Fig. S3 in the SI), a marketed drug, consists of the 2,3,4,9-
tetrahydro-1H-pyrido[3,4-b]indole core as its molecular scaffold. Tadalafil is a
phosphodiesterase type 5 (PDE5) inhibitor and is orally available as a treatment for erectile
dysfunction [37]. This suggests the 2,3,4,9-tetrahydro-1H-pyrido[3,4-b]indole core is likely
to be compatible for use in the clinic with acceptable pharmacokinetic profile.
2.2 Main Screen
The success of the pilot screen encouraged further search of more inhibitors using the same
methodology. The rest of the ChemBridge´s Diversity collection, 4 × 104 molecular entities,
were filtered using Lipinski’s Rule of Five, leaving ~3.3 × 104 drug-like molecules. All four
scoring functions available in the GOLD software suite were used: GS, CS, ChemPLP and
ASP. The consensus concept used in the pilot screen was applied, i.e., only ligands that
scored well with all scoring functions and have predicted hydrogen bonding were taken
forward [38, 39]. The first filter criteria was: GS > 55, CS > 25, ChemPLP > 65, ASP > 25
and HB ≥ 1. This resulted in 5249 candidates, which were re-docked at 100% search
efficiency. The candidates were filtered through a second score criteria: GS > 55, CS > 25,
ChemPLP > 70 and ASP > 25. Furthermore, compounds with good hydrogen bonding
activities were selected (HB ≥ 1) resulting in 871 candidates. Subsequently, ligands with
carboxylic acid functionality and moieties that ionise to form negative charges were taken
forward. This approach assumed the best inhibitors had the ability to coordinate to the Zn2+
ions. One hundred and fifty-five candidates were selected that had the carboxylic and
sulphonic acid moieties. Candidates with undesirable reactive, cytotoxic and chemically
12

unstable moieties [32]. All the ligands were visually inspected for plausible binding modes
and similar binding pose between the four scoring functions, i.e., good overlays between the
predicted binding poses. In total, fifty-four compounds 34-87 were purchased for testing and
they are shown in Fig. S4 in the SI.
Fig. 8. The molecular structures of the hits 63 (65%), 69 (60%), 71 (54%), 80 (47%) and 84 (80%) tested with
the Amplex Red biochemical assay for PC-PLC inhibition at 10 µM.
Four compounds emerged as strong hits with inhibition >50% using the Amplex Red
biochemical assay (see Fig. 8). Hit 84 showed the best potency (80%) followed by 63 (65%),
69 (60%) and 71 (54%). Hits with medium inhibitory activities (25 – 50%) were also seen 37
(25%), 48 (43%), 49 (28%), 73 (40%), 74 (29%) and 80 (47%). Interestingly, multiple
compounds of the same family appeared as actives from the main screen: indole-3-acetic
13

acids (63 and 69) and benzamidobenzoic acids (48, 74 and 80). Four more benzamidobenzoic
acid derivatives were also found to exhibit weak activities (<25%): 55, 62, 72 and 78.
Similarity searches were conducted based on the molecular scaffolds of 71 and 84, as well as
the more modest benzamidobenzoic acid ligand 80. The search results revealed no similar
compounds to 71 at the 0.7 threshold limit. Forty-three derivatives of 84 and twenty
derivatives of 80 were obtained using the search coefficient value of 0.7. Although the
indole-3-acetic acids 63 and 69 displayed strong potency, they were omitted from the search
as potential false positives. The compounds resemble the plant growth hormone, auxin (see
molecular structure in Fig. S5 in the SI). Naturally, auxin is biosynthesised from its amino
acid precursor, tryptophan [40], and is reported to possess anticancer activities in the
presence of horseradish peroxidase [41-43]. The mechanism proposed was generation of free
radicals that resulted in apoptosis [41-43]. Auxin alone did not induce cytotoxicity in human
melanoma cells [44], which confirmed their inactivity in vitro.
2.2.1. 2-Morpholinobenzoic acids
Hit 84 emerged as the strongest inhibitor from the main screen with the impressive 80%
inhibition (at 10 µM). Compound 84 has a 2-morpholinobenzoic acid scaffold linked to a
benzyl group by an amine, with the amine positioned para to the morpholine ring. Forty-one
analogues of 84, classified as the 2-morpholinobenzoic acids, were purchased for testing; the
molecular structures are shown in Table S12 in the SI as well as the assay results.
Compounds 84_1 to 84_33 have various substitution groups around the phenyl ring.
Compounds 84_34 to 39 have various heterocyclic and aromatic groups linked to the 5-amino
functionality. The 5-amino group is replaced with a hydrogen for 84_40. Compound 84_41
contains a piperazine ring instead of a morpholine.
14

Generally, various substitution patterns around the phenyl ring, did not impact on the
inhibitory effects of the compound with levels mainly ranging between 65 – 75% for
compounds 84_1 to 84_33. Unsubstituted phenyl (84_1) had 67% potency against the PC-
PLC enzyme. Most single halogen substitutions (84 and 84_2 to 7) had inhibition >70% with
o- (84_4) and m-chloro (84_5) excellent inhibition at 89%, whereas dual o- and p-chloro
substitution (84_8) had reduced activity at 65%. Introducing o-CF3 (84_9), p-SCH3 (84_10),
single alkoxys (84_11 to 16), double alkoxys (84_17 to 24), halogen and alkoxy
combinations (84_25 to 29), single alkyls (84_30 to 32) and 1-pyrrolidine (84_33) resulted in
inactivity.
Compounds 84_34 to 39 showed activity in the absence of benzyl groups at the R1 site and
with inhibition > 68%. A notable exception was the 2-phenylacetamide functional group,
which displayed significant loss in activity suggesting detrimental effects of amides to the
SAR. The absence of 5-amino group (84_40) and replacement of morpholine for 4-
methylpiperazine-1-yl (84_41) also resulted in significant loss in potency.
(A)
(B)
Ala3
His69
Asp55
Phe66
Fig. 9. The docked configuration of 2-morpholinobenzoic acid 84_5 (A) in the PC-PLCBc binding site. The
protein surface is rendered. Red and blue surfaces depicts negative and positive partial charge whereas grey
shows neutral lipophilic areas. (B) Binding interactions with 84_5 show Zn2+ ions as grey unconnected spheres
form coordinate bonds depicted as red dotted lines, whereas hydrogen bonding are depicted as green dotted lines
with residues Ala3 and Asp55.
15

Modelling of the 2-morpholinobenzoic acid 84_5 (see Figure 9 A and B) showed the phenyl
group occupying a lipophilic space surrounded side chains of Phe70, Tyr79 and Thr133. The
benzoic acid core is seen directly above a hydrophobic basin formed by Phe66 that could
form π−π stacking stabilising the enzyme-inhibitor complex. Furthermore, the morpholine to
Ala3 hydrogen bond interaction can be rationalised as a selective interaction. The reason that
piperazine analogue (84_42) demonstrated significant reduction in activity is most likely due
to it having no hydrogen bond with Ala3.
2.2.2. Benzamidobenzoic acids
Benzamidobenzoic acids 48, 74 and 80 were identified from the main screen with relative
inhibition 25 – 50%, whereas 55, 62, 72 and 78 showed were weaker inhibitors (<25%).
From this series, hit 80 showed the strongest inhibition (47%) featuring a 3-methoxyphenyl
moiety linked to benzoic acid by an amide functionality. Hits 48 and 74 had non-aromatic
ethylcyclohexanes and benzyl moieties as replacements of the phenyl ring. Hit 55 replaces
phenyl with tolylsuphane, and hits 62, 72 and 78 replaced phenyl with benzyl moieties.
Twenty structurally similar derivatives were identified from a similarity search based on the
molecular scaffold of 80 and were acquired for testing, the results and structures are given in
Table S13 in the SI. Unfortunately, this series had weak activity and no improved ligands
were found. Detailed discussion of the SAR and modelling of 80 is given in the SI (pages S48
– S49).
2.3. Inhibition and binding studies
16

The pilot and main screens resulted in the 1H-pyrido[3,4-b]indoles 22 and 2-
morpholinobenzoic acids 84 as the most promising series for inhibition of PC-PLC activity.
The best ligands were chosen for dose response measurements: two compounds, 22_10 and
15, from the 1H-pyrido[3,4-b]indoles and three compounds, 84 and derivatives 84_4 and,
84_5, from the 2-morpholinobenzoic acid series. The derived IC50 values (50% inhibition
concentration) are shown in Table 2. In addition, the binding constant values (KD) for 84 and
derivative 84_4 against recombinant PC-PLCBc were also measured (Supplementary Figures
S7–S9). Interestingly, only 22_15 displayed weaker activity than D609, whereas the most
active 2,3,4,9-tetrahydro-1H-pyrido[3,4-b]indole derivative, 22_10, showed comparable
potencies to the 2-morpholinobenzoic acids 84. The binding constants of 84 and 84_4 are
similar to 33, in agreement with the measured IC50 values.
Table 2. Inhibition concentration at 50% (IC50) and binding constant (KD) values of the 2,3,4,9-tetrahydro-1H-
pyrido[3,4-b]indole and 2-morpholinobenzoic acid derivatives against PC-PLC. The IC50 values are the
averages of five replicates using the Amplex Red biochemical assay. Six different concentrations were used to
derive the IC50 values: 0, 0.1, 0.4, 1.1, 3.3 and 10.0 µM. The KD values are the averages of three replicates using
the intrinsic fluorescence assay. * The KD value for 84_5 was not determined due to spectral overlap in
fluorescence emission. Compound 33 (D609) is the positive control. KDI2a/2b (Known Drug Index) values
reflecting the drug-likens of the ligands. KDI2a has theoretical maximum of 6.0 and KDI2b of 1.0, the higher
values the more drug-like.
Compound
KD(µM)
KDI2a/2b
IC50 (µM)
8.08 ± 0.65
3.10 ± 0.03
10.86 ± 0.56
3.69 ± 0.23
3.60 ± 0.19
4.90 ± 0.21
216 ± 8.0
4.66 / 0.19
5.06 / 0.29
5.71 / 0.74
5.94 / 0.94
5.94 / 0.94
5.94 / 0.94
33 (D609)
22_10
22_15
84
-
-
342 ± 3.9
325 ± 3.0
n.d.*
84_4
84_5
17

It can be stated that two hit series have been established for further development against PC-
PLC, both displaying enhanced potency as compared to the standard D609 (33) ligand.
2.5. Chemical Space
The calculated molecular descriptors; molecular weight (MW), rotatable bonds (RB),
hydrogen bond donors and acceptors (HD/HA), octanol / water partition coefficient (log P)
and polar surface area (PSA) for all the compounds are within the boundaries of drug-like
chemical space, i.e., compounds satisfied Lipinski’s filters [25], and Veber’s rules [45]. The
molecular descriptors and the results of the scoring functions are given in Tables S14 – S16
in the SI.
Furthermore, the drug-like qualities of the compounds were derived for the hit compounds by
calculating the Known Drug Index (KDI) and the results are shown in (see Table 2) [46]. The
KDI reflect the overall balance of the molecular properties calculated based on statistical
distribution of Known Drug Space for the six molecular descriptors used. KDI2a is additive
with a theoretical maximum of 6.0 and for KDI2b the indexes for each descriptor are
multiplied giving 1.0 as its maximum. The average for KDI2a for known drugs is 4.08 (±1.27)
and all of the new hit compounds have values > 5 except for D609 (33). The same trend is
seen of the KDI2b, the hit compounds have much higher values than D609 with the 84 series
all having 0.94, and with the average of know drugs being 0.18 (±0.20), very well balanced
physicochemical properties for biological systems.
3. Conclusion
A virtual screen was conducted against PC-PLC, a promising anticancer target. 5×105
molecular entities of the ChemBridge Diversity library were screened and 86 of those were
18

tested using the Amplex Red biochemical assay at a single concentration (10 µM). Four
potent hits with the molecular scaffolds of benzenesulphonamides, pyrido[3,4-b]indoles,
morpholinobenzoic acids and benzamidobenzoic acids were identified. In order to establish a
SAR 164 structural analogues were tested. The inhibition at 50% values (IC50) were derived
for the five most active compounds. The morpholinobenzoic acids 84, 84_4 and 84_5 had
IC50 values of 3 - 5 µM with the pyrido[3, 4-b]indole 22_10 at similar potency, 3.10 µM. For
comparison, the established inhibitor 33 (D609) was measured to have an IC50 value of 8.08
µM. The Known Drug Indexes (KDI), which gauge the drug-likeness of ligands, gave very
favourable results, far surpassing the established D609 ligand. This means that two novel
drug-like inhibitor classes were identified for PC-PLC enabling a drug discovery programme
to be developed for this bio-molecular target.
4. Materials and Methods
4.1. Modelling
The crystal structure was obtained from Protein Data Bank (PDB) [47] ID: 1AH7[17] with
resolution 1.50 Å. The Scigress v2.6 program [48] was used to prepare the crystal structure
for docking, i.e., hydrogen atoms were added and crystallographic water molecules removed.
The centre of PC-PLCBc binding pocket was defined as the position of Zn2+ ion (x = 42.4820,
y = 22.996, z = 8.556) with 10 Å radius. The basic amino acids lysine and arginine were
defined as protonated. Furthermore, aspartic and glutamic acids were assumed to be
deprotonated. The GoldScore (GS)[27], ChemScore (CS)[28][29], Piecewise Linear Potential
(ChemPLP)[30] and Astex Statistical Potential (ASP)[31] scoring functions were
implemented to validate the predicted binding modes and relative energies of the ligands
using the Genetic Optimisation for Ligand Docking software package (GOLD) version 5.4.
This screening protocol used has been successfully applied to identify inhibitors for different
19

molecular targets such as phospholypase C- γ [49], autophagy complexation Atg5-Atg16
[50], tyrosyl-DNA phosphodiesterase 1 [51] and heat shock protein 90 [52]. The virtual high
throughput screen was conducted with the ChemBridge diversity collection of 5 × 104
entities. Substructure and Tanimoto similarity methods [53] were used to identify structurally
similar compounds through commercially available databases (see results and discussion).
The Dragon version 7.0 [54] software was used to calculate for six mainstream molecular
descriptors; molecular weight (MW), octanol-water partition coefficient P (Log P), rotatable
bonds (RB), hydrogen bond donors (HD) and acceptors (HA) and polar surface area (PSA).
4.2. In vitro PC-PLC activity assays
Relative changes of PC-PLC activities were determined using the Amplex Red assay kit
(Molecular Probe, Inc.), as described by the manufacturer. Briefly, PC-PLC activity was
determined by adding the 0.01U PC-PLC protein to a reaction mixture containing 0.4 mM
Amplex Red, 1 unit/mL horseradish peroxidase, 4 unit/mL alkaline phosphatase, 0.1 unit/mL
choline oxidase, and 0.5 mM phosphotidylcholine (PtdCho) in 1X Reaction Buffer (50 mM
Tris-HCl, pH 7.4, 0.14 M NaCl, 10 mM dimethylglutarate, 2 mM CaCl2). Phosphocholine
released from PtdCho by PC-PLC was converted to choline by alkaline phosphatase, which
choline was oxidized to form H2O2. In the presence of horseradish peroxidase, the H2O2
reacts with Amplex Red to generate the fluorescent product, resorufin, which was detected
using λex = 560 nm and λem = 590 nm on the EnSpire multimode plate reader.
4.3. Recombinant PC-PLCBc production
Plasmid encoding the full length PC-PLCBc was obtained from Dr Ries Langley of Faculty of
Medical and Health Sciences, The University of Auckland. The vector was a modified pET-
21a(+) that enables the production of recombinant proteins with an N-terminal His6 tag and a
20

3C Protease cleavage site. The plasmid was transformed into Escherichia coli BL21(DE3)
competent cells in the presence of 50 µg/mL ampicillin. Expression of PC-PLCBc was
induced using 0.5mM isopropyl 1-thio-β-D-galactopyranoside (IPTG) at 18°C for 16 hours.
The harvested cells were resuspended in lysis buffer (50 mM Tris pH 8.0, 150 mM NaCl, 5
mM imidazole, 0.1 mM ZnSO4, 5% glycerol) and then lysed by sonication. The protein was
purified using a HisTrap HP column (GE Healthcare) charged with Ni2+ ions, connected to an
ÄKTA START protein purifier system (GE Healthcare). The polyhistidine-tagged PC-PLCBc
was eluted using elution buffer (250 mM imidazole, 50 mM Tris-HCl pH 8.0, 150 mM NaCl,
5% glycerol, 0.1 mM ZnSO4). The polyhistidine tag was then removed by incubation with
GST-3C protease (PreScission Protease) in the presence of 2 mM dithiothreitol (DTT) at 4°C
for 12 hours. This was followed by reverse purification using a His-Trap HP column to
remove the cleaved polyhistidine tag. The recombinant PC-PLCBc produced had a 14-amino
acid pro-peptide extension at its N-terminal, which was cleaved by incubation with trypsin
(sequencing grade modified Trypsin, Promega) using a 1:100 protease to protein ratio, at
37°C for 20 min at 90 rpm to obtain the fully functional protein [55, 56]. The trypsin was
deactivated by heating at 50 °C and the protein was further purified by size exclusion
chromatography using a HiPrep 16/60 Sephacryl S-100HR gel filtration column.
4.4. Binding constant measurement by intrinsic tryptophan fluorescence
Assay contained 5 µM of PC-PLCBc and varying concentrations of compounds in 1 mM 3,3-
dimethylglutaratic acid (pH 7.5) and 0.1 mM ZnSO4 (90 µL volume). All experiments were
conducted by using 96-blackwell microplates (Eppendorf). Fluorescence were measured by
using a multimode plate reader (Perkin Elmer). Excitation wavelength was 280 nm and the
fluorescence emission spectra were measured between 300-450 nm. KD was obtained by
21

using the following equation for non-linear curve fitting using SigmaPlot 14.0 (Systat
Software, USA).
∆
max ×[LT ]
∆
=
obs
KD +[LT ]
In which [LT] denotes the total ligand concentration, ∆obs denotes changes in fluorescence
intensity and ∆max as the maximum change in fluorescence intensity.
Acknowledgements
We are very grateful for the financial support from the Cancer Society of New Zealand (Ref:
17.01) and the Breast Cancer Partnership, Health Research Council, New Zealand (Ref:
17/671).
References
[1] C.M. Vines, Phospholipase C, Adv. Exp. Med. Biol. 740 (2012) 235-254.
[2] J.B. Park, C.S. Lee, J.H. Jang, J. Ghim, Y.J. Kim, S. You, D. Hwang, P.G. Suh, S.H. Ryu,
Phospholipase signalling networks in cancer, Nat. Rev. Cancer 12 (2012) 782-792.
[3] F. Spadaro, C. Ramoni, D. Mezzanzanica, S. Miotti, P. Alberti, S. Cecchetti, E. Iorio, V.
Dolo, S. Canevari, F. Podo, Phosphatidylcholine-Specific Phospholipase C Activation in
Epithelial Ovarian Cancer Cells, Cancer Research. 68 (2008) 6541-6549.
[4] L. Abalsamo, F. Spadaro, G. Bozzuto, L. Paris, S. Cecchetti, L. Lugini, E. Iorio, A.
Molinari, C. Ramoni, F. Podo, Inhibition of phosphatidylcholine-specific phospholipase
C results in loss of mesenchymal traits in metastatic breast cancer cells, Breast Cancer
Res. 14 (2012) R50.
[5] L. Paris, S. Cecchetti, F. Spadaro, L. Abalsamo, L. Lugini, M.E. Pisanu, E. Iorio, P.G.
Natali, C. Ramoni, F. Podo, Inhibition of phosphatidylcholine-specific phospholipase C
downregulates HER2 overexpression on plasma membrane of breast cancer cells, Breast
cancer research : BCR. 12 (2010) R27.
22

[6] W. Xingzhong, H. Lu, L. Zhou, Y. Huang, H. Chen, Changes of Phosphatidylcholine-
Specific Phospholipase C in Hepatocarcinogenesis and in the Proliferation and
Differentiation of Rat Liver Cancer Cells. Cell Bio. Int. 21 (1997) 375-381.
[7] I. Plo, D. Lautier, T. Levade, H. Sekouri, J.P. Jaffrézou, G. Laurent, A. Bettaïeb,
Phosphatidylcholine-specific phospholipase C and phospholipase D are respectively
implicated in mitogen-activated protein kinase and nuclear factor kappaB activation in
tumour-necrosis-factor-alpha-treated immature acute-myeloid-leukaemia cells, Biochem.
J. 351 (2000) 459-467.
[8] H.Y. Li, L. Zhang, D.L. Yin, Y. Zhang, J. Y. Miao, Targeting Phosphatidylcholine-
Specific Phospholipase C for Atherogenesis Therapy, Trends Cardiovas. Med. 20 (2010)
172-176.
[9] P. González-Bulnes, A. González-Roura, D. Canals, A. Delgado, J. Casas, A. Llebaria, 2-
Aminohydroxamic acid derivatives as inhibitors of Bacillus cereus phosphatidylcholine
preferred phospholipase C PC-PLCBc, Bioorg. Med. Chem. 18 (2010) 8549-8555.
[10] S.E. Aakre, C. Little, Inhibition of Bacillus cereus phospholipase C by univalent anions,
Biochem. J. 203 (1982) 799-801.
[11] S.F. Martin, B.C. Follows, P.J. Hergenrother, C.L. Franklin, A novel class of zinc-
binding inhibitors for the phosphatidylcholine-preferring phospholipase C from Bacillus
cereus, J. Org. Chem. 65 (2000) 4509-4514.
[12] S.F. Martin, Y. Wong, A.S. Wagman, Design, Synthesis, and Evaluation of
Phospholipid Analogs as Inhibitors of the Bacterial Phospholipase C from Bacillus
cereus, J. Org. Chem. 59 (1994) 4821-4831.
[13] R. Adibhatla, J. Hatcher, A. Gusain, Tricyclodecan-9-yl-Xanthogenate (D609)
Mechanism of Actions: A Mini-Review of Literature, Neurochem. Res. 37 (2012) 671-
679.
[14] E. Amtmann, The antiviral, antitumoural xanthate D609 is a competitive inhibitor of
phosphatidylcholine-specific phospholipase C, Drugs Exp. Clin. Res. 22 (1996) 287-294.
[15] M. Wu, M. Takemoto, H. Luo, J.J. Xu, M. H. Lu, M. Kameyama, T. Takumi, W.J. Song,
A novel role of the antitumor agent tricyclodecan-9-yl-xanthogenate as an open channel
blocker of KCNQ1/KCNE1, Eur. J. Pharmacol. 824 (2018) 99-107.
[16] R.M. Adibhatla, J.F. Hatcher, A. Gusain, Tricyclodecan-9-yl-xanthogenate (D609)
mechanism of actions: a mini-review of literature. Neurochem. Res. 37 (2012) 671-679.
[17] K. Jynge, E. Dodson, Z. Derewenda, C. Little, L.K. Hansen, E. Hough, B. Birknes, S.
Hansen, A. Hordvik, High-resolution (1.5 Å) crystal structure of phospholipase C from
Bacillus cereus, Nature. 338 (1989) 357-360.
[18] S. Subramaniam, M. Mehrotra, D. Gupta, Virtual high throughput screening (vHTS) - A
perspective, Bioinformation. 3 (2008) 14-17.
23

[19] R.V. Guido, G. Oliva, A.D. Andricopulo, Virtual screening and its integration with
modern drug design technologies, Curr. Med. Chem. 15 (2008) 37-46.
[20] T. Zhu, S. Cao, P.C. Su, R. Patel, D. Shah, H.B. Chokshi, R. Szukala, M.E. Johnson,
K.E. Hevener, Hit identification and optimization in virtual screening: practical
recommendations based on a critical literature analysis, J. Med. Chem. 12 (2013) 6560-
6572.
[21] T. Scior, A. Bender, G. Tresadern, J. Medina-Franco, K. Martínez-Mayorga, T. Langer,
K. Cuanalo-Contreras, D.K. Agrafiotis, Recognizing Pitfalls in Virtual Screening: A
Critical Review, J. Chem. Inf. Model. 52 (2012) 867-881.
[22] M.A. Clark, R.G.L. Shorr, J.S. Bomalaski, Antibodies prepared to Bacillus cereus
phospholipase C crossreact with a phosphatidylcholine preferring phosoholipase C in
mammalian cells, Biochem. Biophys. Res. Comm. 140 (1986) 114-119.
[23] L. Levine, D.M. Xiao, C. Little, Increased arachidonic acid metabolites from cells in
culture after treatment with the phosphatidylcholine-hydrolyzing phospholipase C from
Bacillus cereus, Prostaglandins. 34 (1987) 633-642.
[24] ChemBridge, DIVERSet™ Diverse Screening Libraries, (2018).
[25] C.A. Lipinski, F. Lombardo, B.W. Dominy, P.J. Feeney, Experimental and
computational approaches to estimate solubility and permeability in drug discovery and
development setting, Adv. Drug Deliv. Rev. 23 (1997) 3-25.
[26] S.F. Martin, P.J. Hergenrother, Catalytic cycle of the phosphatidylcholine-preferring
phospholipase C from Bacillus cereus. Solvent viscosity, deuterium isotope effects, and
proton inventory studies, Biochemistry. 38 (1999) 4403-4408.
[27] G. Jones, P. Willet, R.C. Glen, A.R. Leach, R. Taylor, Development and Validation of a
Genetic Algorithm for Flexible Docking, J. Mol. Biol. 267 (1997) 727-748.
[28] M.D. Eldridge, C. Murray, T.R. Auton, G.V. Paolini, P.M. Mee, Empirical scoring
functions: I. The development of a fast empirical scoring function to estimate the binding
affinity of ligands in receptor complexes, J. Comp. Aid.Mol. Design. 11 (1997) 425-445.
[29] M.L. Verdonk, J.C. Cole, M.J. Hartshorn, C.W. Murray, R.D. Taylor, Improved protein-
ligand docking using GOLD, Proteins. 52 (2003) 609-623.
[30] O. Korb, T. Stützle, T.E. Exner, Empirical Scoring Functions for Advanced
Protein−Ligand Docking with PLANTS, J. Chem. Inf. Model. 49 (2009) 84-96.
[31] W.T.M. Mooij, M.L. Verdonk, General and targeted statistical potentials for protein–
ligand interactions, Proteins. 61 (2005) 272-287.
[32] C.G. Bologa, T.I. Oprea, Compound collection preparation for virtual screening,
Methods in molecular biology (Clifton, N.J.). 910 (2012) 125-143.
24

[33] A.P. Benfield, N.M. Goodey, L.T. Phillips, S.F. Martin, Structural studies examining the
substrate specificity profiles of PC-PLC(Bc) protein variants, Arch. Biochem. Biophys.
460 (2007) 41-47.
[34] eMolecules, www.emolecules.com. (2018).
[35] ChemDiv, www.chemdiv.com, (2018).
[36] InterBioScreen, www.ibscreen.com, (2018).
[37] R.M. Coward, C.C. Carson, Tadalafil in the treatment of erectile dysfunction, Ther. Clin.
Risk Man. 4 (2008) 1315-1329.
[38] R. Wang, S. Wang, How Does Consensus Scoring Work for Virtual Library Screening?
An Idealized COmputer Experiment, J. Chem. Inf. Comput. Sci. 41 (2001) 1422-1426.
[39] J. Yang, Y. Chen, T. Shen, B.S. Kristal, D.F. Hsu, Consensus Scoring Criteria for
Improving Enrichment in Virtual Screening, J. Chem. Info. Modeling. 45 (2005) 1134-
1146.
[40] Y. Zhao, Auxin Biosynthesis: A Simple Two-Step Pathway Converts Tryptophan to
Indole-3-Acetic Acid in Plants, Molecular Plant. 5 (2012) 334-338.
[41] S.Y. Kim, J.S. Ryu, H. Li, W. Park, H. Yun, K.J. Baek, N.S. Kwon, U.D. Sohn, D. Kim,
UVB-activated indole-3-acetic acid induces apoptosis of PC-3 prostate cancer cells,
Anticancer Res. 30 (2010) 4607-4612.
[42] Peter Wardman, Indole-3-Acetic Acids and Horseradish Peroxidase: A New Prodrug /
Enzyme Combination for Targeted Cancer Therapy, Curr. Pharm. Design. 8 (2002)
1363-1347.
[43] S. Kim, D. Kim, S. Jeon, Y. Jeong, S. Kwon, K. Park, Hydrogen peroxide is a mediator
of indole-3-acetic acid/horseradish peroxidase-induced apoptosis, FEBS Letters. 580
(2006) 1439-1446.
[44] D. Kim, S. Jeon, K. Park, Oxidation of indole-3-acetic acid by horseradish peroxidase
induces apoptosis in G361 human melanoma cells, Cell. Signal. 16 (2004) 81-88.
[45] D.F. Veber, S.R. Johnson, H. Cheng, B.R. Smith, K.W. Ward, K.D. Kopple, Molecular
Properties that Influence the Oral Bioavailability of Drug Candidates, J. Med. Chem. 45
(2002) 2615-2623.
[46] C. Eurtivong, J. Reynisson, The Development of a Weighted Index to Optimise
Compound Libraries for High Throughput Screening, Mol. Inf. 38 (2019) e180006-n/a.
[47] H.M. Berman, J. Westbrook, Z. Feng, G. Gilliland, T.N. Bhat, H. Weissig, I.N.
Shindyalov, P.E. Bourne, The Protein Data Bank, Nuc. Acids Res. 28 (2000) 235-242.
[48] Scigress: Version FJ 2.6 (EU 3.1.7), Fujitsu Limited 2008-2016.
25

[49] J. Reynisson, W. Court, C. O’Neill, J. Day, L. Patterson, E. McDonald, P. Workman, M.
Katan, S.A. Eccles, The identification of novel PLC-c inhibitors using virtual high
throughput screening, Bioorg. Med. Chem. 17 (2009) 3169-3176.
[50] E. Robinson, E. Leung, A.M. Matuszek, N. Krogsgaard-Larsen, D.P. Furkert, M.A.
Brimble, A. Richardson, J. Reynisson, Virtual screening for novel Atg5–Atg16 complex
inhibitors for autophagy modulation, Med. Chem. Comm. 6 (2015) 239-246.
[51] T. Khomenko, A. Zakharenko, T. Odarchenko, H.J. Arabshahi, V. Sannikova, O.
Zakharova, D. Korchagina, J. Reynisson, K. Volcho, N. Salakhutdinov, O. Lavrik, New
inhibitors of tyrosyl-DNA phosphodiesterase I (Tdp 1) combining 7-hydroxycoumarin
and monoterpenoid moieties, Bioorg. Med. Chem. 24 (2016) 5573–5581.
[52] R. Huang, D. Ayine-Tora, M.N.M. Rosdi, Y. Li, J. Reynisson, I.K.H. Leung, Virtual
screening and biophysical studies lead to HSP90 inhibitors, Bioorg. Med. Chem. Lett. 27
(2017) 277-281.
[53] A.R. Leach, V.J. Gillet, 1. Representation and manipulation of 2D molecular structures,
An Introduction to Chemoinformatics. (2003) 1-26.
[54] R. Todeschini, M. Lasagni, E. Marengo, New molecular descriptors for 2D and 3D
structures. Theory, J. Chemometrics, 8 (1994) 263-272.
[55] C.A. Tan, M.J. Hehir, M.F. Roberts, Cloning, Overexpression, Refolding, and
Purification of the Nonspecific Phospholipase C from Bacillus cereus, Protein Expres.
Pur. 10 (1997) 365-372.
[56] P. Ravasi, M. Braia, F. Eberhardt, C. Elena, S. Cerminati, S. Peirú, M.E. Castelli, H.G.
Menzella, High-level production of Bacillus cereus phospholipase C in Corynebacterium
glutamicum, J. Biotech. 216 (2015) 142-148.
26

Highlights
Virtual screen resulted in inhibitors for Phosphatidylcholine–specific phospholipase C
Pyrido[3,4-b]indoles and morpholinobenzoic acid drug-like series identified
Single digit micro-molar potency reached

Declaration of interests
☒ The authors declare that they have no known competing financial interests or personal relationships
that could have appeared to influence the work reported in this paper.
☒The authors declare the following financial interests/personal relationships which may be considered
as potential competing interests:
On behalf of all the authors
Jóhannes Reynisson
11/11 2019