Investigation into improving the aqueous solubility of the thieno[2,3-b]pyridine anti-proliferative agents

Article abstract of DOI:10.3390/molecules23010145

It is now established that the thieno[2,3-b]pyridines are a potent class of antiproliferatives. One of the main issues encountered for their clinical application is their low water solubility. In order to improve this, two strategies were pursued. First,

Full text of DOI:10.3390/molecules23010145

molecules
Article
Investigation into Improving the Aqueous Solubility
of the Thieno[2,3-b]pyridine Anti-Proliferative Agents
Ayesha Zafar 1, Lisa I. Pilkington 1, Natalie A. Haverkate 1, Michelle van Rensburg 1,
Euphemia Leung 2, Sisira Kumara 2, William A. Denny 2, David Barker 1, Ali Alsuraifi 3,
Clare Hoskins 3
and Jóhannes Reynisson 1,
*
ID
ID
1
School of Chemical Sciences, University of Auckland, 23 Symonds Street, 1142 Auckland, New Zealand;
ash_imran@hotmail.com (A.Z.); lisa.pilkington@auckland.ac.nz (L.I.P.);
nhav422@aucklanduni.ac.nz (N.A.H.); m.vanrensburg@auckland.ac.nz (M.v.R.);
d.barker@auckland.ac.nz (D.B.)
Auckland Cancer Society Research Centre and Department of Molecular Medicine and Pathology,
University of Auckland, 1142 Auckland, New Zealand; e.leung@auckland.ac.nz (E.L.);
s.kumara@auckland.ac.nz (S.K.); b.denny@auckland.ac.nz (W.A.D.)
2
3
Institute for Science and Technology in Medicine, Keele University, Stoke-on-Trent ST4 7QB, UK;
a.t.y.alsuraifi@keele.ac.uk (A.A.); c.hoskins@keele.ac.uk (C.H.)
*
Correspondence: j.reynisson@auckland.ac.nz; Tel.: +64-9-373-7599 (ext. 83746)
Received: 18 December 2017; Accepted: 8 January 2018; Published: 11 January 2018
Abstract: It is now established that the thieno[2,3-b]pyridines are a potent class of antiproliferatives.
One of the main issues encountered for their clinical application is their low water solubility. In order
to improve this, two strategies were pursued. First, a morpholine moiety was tethered to the molecular
scaffold by substituting the sulphur atom with nitrogen, resulting in a 1H-pyrrolo[2,3-b]pyridine
core structure. The water solubility was increased by three orders of magnitude, from 1.2
-thieno[2,3-b]pyridine) to 1.3 mg/mL ( -pyrrolo[2,3-b]pyridine), however, it was only marginally
active against cancer cells. The second strategy involved loading a very potent thieno[2,3-b]pyridine
derivative ( ) into a cholesteryl-poly(allylamine) polymer matrix for water solubilisation. Suppression
of human pancreatic adenocarcinoma (BxPC-3) viability was observed to an IC50 value of 0.5 g/mL
(1.30 M) in conjunction with the polymer, which is a five-fold ( 5) increase in potency as compared
to the free drug alone, demonstrating the utility of this formulation approach.
µg/mL
(1
3
2
µ
µ
×
Keywords: morpholine substitution; 1H-pyrrolo[2,3-b]pyridine; molecular modelling; nano aggregates;
polymer formulation
1. Introduction
The class of thieno[2,3-b]pyridines are potent anti-proliferative agents against a number of
tumour cell lines [
the thieno[2,3-b]pyridines was discovered by virtual high throughput screen (vHTS) against the
phospholipase C- 2 (PLC- 2) isoform, a promising anticancer target [ ]. The administration of
1–3] and their basic molecular structure is shown in Figure 1. The efficacy of
γ
γ
4
thieno[2,3-b]pyridines causes the breast cancer cell line MDA-MB-231 to be severely growth-restricted,
rounding and blebbing of the plasma membrane, G2/M phase population to increase in the cell
cycle, and a decrease in motility [
inhibition of the PLC- 1 and 2 isoforms, making them the most plausible target for this class
of compounds [ ]. However, it has been shown that other bio-molecular targets are modulated,
contributing to the overall efficacy of the thieno[2,3-b]pyridines: tyrosyl-DNA phosphodiesterase I
(TDP1) [ ], the A2A receptor (A2AAR) a G-protein coupled receptor (GPCR)[ ], copper trafficking
5,6]. These effects on MDA-MB-231 are more in line with the
δ
δ
6
7
8
protein Atox [9], and tubulin [10,11].
Molecules 2018, 23, 145; doi:10.3390/molecules23010145
www.mdpi.com/journal/molecules

Molecules 2018, 23, 145
2 of 14
Figure 1. The structure of the anticancer thieno[2,3-b]pyridine
tetrahydrothieno[2,3-b]quinoline-2-carboxamide).
1 (3-amino-5-oxo-N-phenyl-5,6,7,8-
A major obstacle in developing the thieno[2,3-b]pyridines into effective anticancer drugs is
their lack of aqueous solubility. In order to derive a meaningful mouse pharmacokinetic profile
the compound was required to be solubilised using a cyclodextrin (HP-
solubilising method [ ]. Furthermore, in a mouse xenograft study on the effect of thieno[2,3-b]pyridine
administration, water solubility was the dose-limiting factor [ ]. It is estimated that ~40% of
drug discovery programmes are discontinued due to limited water solubility and associated poor
pharmacokinetics of the lead compounds [12 13]. It is, therefore, evident that the water solubility of
β CD), a well-known
6
7
,
thieno[2,3-b]pyridines needs to be improved for meaningful clinical application.
The aim of this study was to investigate two strategies to improve the water solubility of the
thieno[2,3-b]pyridines. First, by adding a solubilising group to the molecular scaffold and, second,
to load an active derivative in a formulation polymer.
2. Results and Discussion
Seven compounds were chosen for investigation into the improved aqueous solubility of the
thieno[2,3-b]pyridines. Compounds 1 and 2 were selected due to their established anticancer profiles [1,2].
In order to attach a solubilising group to the thieno[2,3-b]pyridine scaffold, the sulphur atom was
replaced with nitrogen creating the 1H-pyrrolo[2,3-b]pyridine molecular frame as seen for derivatives
3
–
7
in Figure 2. It has been previously shown that substitution of the sulphur atom for oxygen gave
derivatives with comparable activity [ ]. The standard solubilising group morpholine was attached
resulting in derivative , as a direct comparison a comparable phenyl derivative ( ) was synthesised,
and derivatives had a methyl substitution on R2, but with different ring sizes and substitutions to
3
3
4
5–7
gauge the effect of substituting sulphur with a nitrogen atom, in general.
R1
NH2
O
X2
X3
X1
R2
HN
N
n
: X1 = S, X2 = X3 = H, R1 = O, n = 1
: X1 = S, X2 = Me, X3 = Cl, R1 = O, n = 1
: X1 = N, X2 = X3 = H, R1 = H/H, R2 = Morpholinopropyl, n = 1
: X1 = N, X2 = X3 = H, R1 = H/H, R2 = Phenylpropyl, n = 1
: X1 = N, X2 = X3 = H, -naphthyl, R1 = H/H, R2 = Me, n = 1
: X1 = N, X2 = X3 = Me, R1 = H/H, R2 = Me, n = 1
: X1 = N, X2 = X3 = H, R1 = H/H, R2 = Me, n = 3
Figure 2. The molecular structures of the thieno[2,3-b]pyridines (1, 2) and 1H-pyrrolo[2,3-b]pyridines (3–7).

Molecules 2018, 23, 145
3 of 14
2.1. Synthesis of Pyrrolo[2,3-b]pyridine Derivatives
Derivatives
the pyrrolopyridines (
pyrrolopyridines and
1
and
2
were synthesised using previouslydescribed methods [
) were synthesised as described by Pilkington et al. [14]. As an example,
3
]. Furthermore,
3
–
7
3
4
were made by coupling amines 8a with carbonitrile (Scheme 1). Initially,
,b
9
aniline 10 was converted to bromophenylacetamide using bromoacetyl bromide 11 in the presence of
triethylamine, which was then reacted with various amines to give N-substituted phenylacetamides
8a,b. Carbonitrile 9 was prepared from cyclohexanone 12 in three steps; firstly, a methylenehydroxy
salt was prepared by adding ethyl formate and cyclohexanone 12 to a solution of freshly prepared
sodium methoxide. The pyridone was formed through addition of cyanoacetamide to the prepared salt,
giving 13, which was followed by chlorination using POCl3, providing carbonitrile
9
. N-substituted
phenylacetamides 8a
,
b
were coupled with carbonitrile
9
in DMSO using KF to give nitriles 14a,b,
which then underwent cyclisation using KOtBu to give the desired analogues
yields (90–100%).
3 and 4 in excellent
Scheme 1. Synthesis of
3 and 4. (i) Et3N (1.1 equiv.), CH2Cl2, r.t., 24 h, 88%; (ii) RNH2, EtOH, r.t.,
24 h, 8a 68%-quant.; (iii) Na (1 equiv.), MeOH (5.6 equiv.), ethyl formate (1 equiv.), r.t., 24 h, 42%;
,b
(iv) cyanoacetamide (1 equiv.), piperidinium acetate, H2O, reflux, 24 h, 29%; (v) POCl3 (1.3 equiv.),
◦
reflux, 24 h, 49%; (vi) KF (2.4 equiv.), DMSO, 120 C, 24 h, 14a
,b
10–51%; (vii) KOtBu (1.2 equiv.), THF,
reflux, 30 min, 3 and 4 90–100%.
2.2. Water Solubility Testing
The aqueous thermodynamic solubility [15] of derivatives 1–7 were measured using a standard
HPLC (High Performance Liquid Chromatography) procedure [16]. Solubility is defined as the amount
of substance that passes into solution in order to establish the equilibrium at constant temperature and
pressure and, thus, produce a saturated solution. The water solubility (log S) and log P were calculated
for comparison. The results are given in Table 1.
According to the experimental results, it is clear that compounds
1 and 2 are relatively insoluble
with only single-digit g/mL. This modest solubility can explain the low plasma concentration in
µ
mice of thieno[2,3-b]pyridines, i.e., only ~90 nM maximum concentration (Cmax) was observed after
10 mg/Kg intra-peritoneal injection [6].

Molecules 2018, 23, 145
4 of 14
Table 1. The experimental and calculated aqueous solubility(S) of derivatives 1–7, as well as their
predicted log P.
Experimental S
Calculated S
log S a
µg/mL
log S a
µg/mL
log P
1
2
3
4
5
6
7
a
1.21
3.53
−5.4
−5.0
−2.5 c
−6.3
−6.3
−5.7
−5.8
13.4
96.9
1089.1
1.1
−4.4
−3.6
−2.6
−5.6
−4.8
−4.1
−4.4
2.9
2.2
2.9
5.3
4.0
3.3
3.5
1339.8/1336.7 b
0.20
0.18
0.76
0.61
5.9
27.7
13.9
c
Units are in mol/L. b Two measurements made. Avarage of the two measurements.
Compound
of a magnitude giving ~1.3 mg/mL as compared to
for a phenyl ring led to the lower solubility of derivative
very poor solubility even though they were only substituted with a methyl group.
The calculated solubilities follow the trend of their measured counterparts and for derivative
3
, with a tethered morpholine moiety resulted in a solubility increase by three orders
. As expected, substituting the morpholine moiety
. Interestingly, compounds also have
1
4
5–7
3
a
reasonable correlation is seen (18.6%). For compounds
large by approximately one order of magnitude.
1,
2
,
4,
5
,
6, and
7
the predicted values are too
Interestingly, the log P values do not follow the solubility trends well except for derivative
4
with
a high log P at 5.3, which is outside of drug-like chemical space. The other mainstream descriptors:
molecular weight (MW), hydrogen bond donors (HD), hydrogen bond acceptors (HA) polar surface
area (PSA) and rotatable bonds (RB) (see Table S1 in the Supplementary Materials) are all within the
drug-like chemical space. It has been previously noted that the currently-used molecular descriptors
like log P and log S do not entirely capture the required water solubility profiles of successful drugs [17].
2.3. Cell Proliferation
Thymidine uptake was used for viability testing on the MDA-MB-231 (breast) and HCT116
(colon) human tumour cell lines. This measurement reflects the anti-proliferative/cytotoxic effects
of the compounds. A concentration of 1
1H-pyrrolo[2,3-b]pyridine derivatives (see Table S2 in the Supplementary Materials). In comparison,
derivative has a measured IC50 of 163.5 nM against the MDA-MB-231 cell line using the thymidine
uptake assay [ ]. To corroborate these findings, derivatives , and were tested against the
National Cancer Institute’s human tumour cell line panel (NCI60) [18]. Again, very modest or no
activity was found (see NCI data in the Supplementary Materials). Derivative was the most potent
µM was used. Very little activity was observed for the
3–7
2
6
3,
4
6
6
with 77% overall relative growth for the 60 cell lines as compared untreated cells (100% growth).
The leukaemia cell lines K-562 (45.5% growth) and SR (47.7% growth) were the most affected followed
by the melanoma MDA-MB-435 (53.9% growth) cell line. This is an interesting observation since
the thieno[2,3-b]pyridines are often active against leukaemia cells and the MDA-MB-435 cell line, in
particular [2,7,19,20]. It can, therefore, be concluded that the 1H-pyrrolo[2,3-b]pyridines are nearly
inactive against cancer when compared to their thieno[2,3-b]pyridine counterparts. Again, this is in
stark contrast to the thieno[2,3-b]pyridines, in particular for derivative 2, with reported GI50 = 141 nM
for MDA-MB-231 and GI50 = 67 nM for HCT116 [1,2,6].
2.4. Molecular Modelling
Compounds
Atox [ ], and tubulin-colchicine site [10
Reasonable scores were observed for the four scoring functions used for all of the targets, which are
1
–
7
were docked into the binding pockets of the PLC-
δ1 [6], TDP1 [7], A2AAR [8],
9
,
11], the reported biological targets for the thieno[2,3-b]pyridines.

Molecules 2018, 23, 145
5 of 14
given in Tables S3–S7 in the Supplementary Materials. Based on the scores it may be deduced that
derivatives 1–7 all interact with these targets, i.e., no obvious reduction in scores was seen.
For PLC the thieno[2,3-b]pyridines and 1H-pyrrolo[2,3-b]pyridine scaffolds are predicted to fit
in a similar fashion into the binding pocket (see Table S3 in the Supplementary Materials) and these
conformations are in line with previous work [
derivatives and occupy a cleft on the protein surface as shown for
the morpholine moiety is exposed to the aqueous phase.
2
,
5
]. The morpholine and phenyl side chains on
3
4
3
in Figure 3. Importantly,
A
B
Figure 3. The docked configuration of 3: (A) to the binding site of PLC-δ1 using the ChemPLP
scoring function. The morpholine occupies a cleft on the protein surface, but has access to the water
environment. The protein surface is rendered. Red depicts a positive partial charge on the surface,
blue depicts negative partial charge and grey shows neutral/lipophilic areas; (B) Hydrogen bonds are
depicted as green lines between the ligand and the amino acid residues Lys348, Arg549, and Glu341.
For TDP1 it is proposed that both histidine amino acid residues (His263 and His493) play a role
in its biological function and the binding pocket was defined there [21]. Upon docking, compounds
showed varied hydrogen bonding patterns (Table S4 in the Supplementary Materials), but a similar fit
was predicted for all of them.
A common binding motif is reported for Atox involving polar interactions with side chains of
arginine (Arg21), lysine (Lys60), and glutamic acid (Glu17) in order to block protein-protein interface
for copper delivery [
9
]. Similar binding poses are predicted for
1–7, e.g., for 2 the phenyl ring has a
lipophilic contact with arginine (Arg21) and a
π-π
stacking with lysine (Lys60) is possible (see Table S5).
The A2ARA has reported binding motifs of stacking between aromatic moieties of the ligands
and the Phe168 side chain of the receptor, as well as polar interactions with Asn253 and Glu169 side
chains [
8]. These interactions were predicted for compounds 1–7 and similar binding configurations
were seen for the derivatives (see Table S6 in Supplementary Materials). Finally, the tubulin-colchicine
site was investigated and
(see Table S7).
1–7 displayed similar stacking and H-bonding interaction as colchicine
In all of the bio-molecular targets investigated, the chemical modifications made to the
thieno[2,3-b]pyridine scaffold can be accommodated by the binding sites. These results suggest
that other factors reduce the potency of the 1H-pyrrolo[2,3-b]pyridine derivatives
to the established targets. This might include the very high solubility of , preventing it passing into
cell membranes, thus reducing its permeation. The very low solubility of analogues prevents
3–7 than binding
3
4–7
concentration build up in the cancer cells presenting a plausible explanation for their inactivity. In both
scenarios offered the ligands do not reach the targets in high enough concentration to adversely affect
the cancer cells. An alternative explanation is that the thieno[2,3-b]pyridines exert their anticancer
efficacy by modulating an unknown molecular target.

Molecules 2018, 23, 145
6 of 14
2.5. Polymer Encapsulation
A common strategy to facilitate water solubility for bioactive compounds is their formulation
in a solubilising amphiphilic polymer (see, e.g., [22 23] and references therein). The most potent
derivative, , was chosen to be formulated with cholesteryl-poly(allylamine) (Ch5) polymer [24].
,
2
The hydrodynamic size of Ch5 (1 mg/mL) formed in aqueous solution was 189 nm (see Table 2).
The low polydispersity index indicated that fairly uniform single sized populations were produced.
The polydispersity index is calculated between 0 and 1, the closer the value is to 0, the more
monodispersity is reached. It is desirable to have such a value with <0.3 for pharmaceutical products.
Interestingly, core expansion was observed when the nano-aggregates were loaded with
2 and the size
increased to 286 nm but the polydispersity index only marginally increased. Additionally, the surface
charge of the aggregates formed did not change after drug loading (Ch5+2) compared with the
unloaded aggregate (Ch5) as reflected in the Zeta potential (Table 2).
Table 2. The hydrodynamic diameter with the corresponding polydispersity index and the zeta
potential measurement of the aggregates in water (n = 3, ±SD).
Hydrodynamic Diameter (nm)
Polydispersity Index
Zeta Potential, mV
Ch5
Ch5+2
189 ± 1.800
286 ± 5.839
0.246 ± 0.010
0.267 ± 0.011
+39.6 (2.242)
+36.5 (1.635)
Loading studies showed that at increased polymer concentrations, an enhanced incorporation of
into the hydrophobic core was achieved. Furthermore, increasing the feed ratio between compound
2
2
and the polymer resulted in enhanced water solubility, as shown in Figure 4. The maximum
compound solubilisation observed was 6 mg/mL at a 10:1 ratio, with 0.59 mg/mL drug encapsulation
being achieved.
Figure 4. Drug encapsulation at three different feed ratios, 1:1, 5:1, and 10:1, between
Ch5. Three initial concentrations of 2 were used: 1, 3, and 6 mg/mL (n = 3, ±SD).
2 and the polymer
Release studies showed steady leaching of
2 from the polymer ensemble into the aqueous phase,
plateauing after ~40 min as shown in Figure 5.
Finally the viability was measured with the MTT assay [25] in human pancreatic adenocarcinoma
(BxPC-3) cells using with and without the polymer encapsulation; the results are shown in Figure 6.
2

Molecules 2018, 23, 145
7 of 14
25
20
15
10
5
0
0
10
20
30
40
50
60
Time / min
◦
Figure 5. Compound 2 release from the polymer matrix into PBS at 37 C for 1 h (n = 3, ±SD).
100
Ch5
2
Ch5+2
80
60
40
20
0
0.0001
0.001
0.01
0.1
1
10
100
1000
Total drug concentration (μg/mL)
Figure 6. The viability of BxPC-3 cells after exposure to the polymer (Ch5),
drug (Ch5+2) for 24 h (n = 6, ±SD).
2 (drug) and encapsulated
As can be seen in Figure 6 the cancer cell viability for the free compound
down to 5 µg/mL (12.95 µM) concentration with ~30% activity remaining. However, at the 1 µg/mL
(2.59 M) concentration the cells are growing without any restrictions. Thus, the IC50 value for free
can be estimated at 2.5 g/mL (6.48 M). Interestingly, when is encapsulated into the Ch5-polymer
2 is effectively suppressed
µ
2
µ
µ
2
there is a notable reduction in cell viability compared to the free compound. At higher concentration a
complete cytotoxic effect is seen, which gradually grows to 30–35% viability as the concentration is
lowered and reduces further to zero inhibition at concentrations lower than 0.29
µg/mL (0.75
µM).
The IC50 value of Ch5+2 was observed at 0.5 µg/mL (1.30 µM).
These results show that encapsulating
thereby vindicating this approach for the thieno[2,3-b]pyridines. The most direct explanation for the
2
in the polymer matrix increases its efficacy five-fold (
×
5),

Molecules 2018, 23, 145
8 of 14
enhanced efficacy is simply the increased presence of
2
in the aqueous phase, leading to more material
reaching the cancer cells. Ideally, the polymer should be non-cytotoxic however, as Ch5 has an IC50 of
~50 g/mL a synergistic effect between and Ch5 is possible. An effective viability suppression is
seen for the Ch5 complex at 0.1 g/mL, considering the mild cytotoxic effect of the Ch5 polymer,
µ
2
+
2
µ
it can be argued that it is unlikely that the polymer contributes to the anticancer efficacy. Finally,
it is known that after encapsulation into nanoparticles, drug molecules are capable of more rapid
cellular internalisation due to endocytosis when compared to free drugs [26]. Free drugs often enter
through the cellular tight junctions or other transport systems, which can be more time-consuming
than an active transport process [27]. This phenomenon can be observed in the drug uptake studies,
whereby increased intracellular drug concentrations were apparent from cells incubated with the
Ch5+2 compared with the free drug as demonstrated in Figure 7. This trend was consistent at both
concentrations tested.
0.012
0.01
2
0.008
0.006
0.004
0.002
0
Ch5+2
25
50
Total drug incubation concentration (µg/mL)
Figure 7. Drug uptake into BxPC-3 cells after 4 h exposure of 25
formulation (n = 3, ±SD).
µg/mL and 50
µg/mL drug and
3. Materials and Methods
3.1. Drug Solubility
Intrinsic aqueous drug solubility was determined by forming a supersaturated solution of the drug
compound in water. The suspension was sonicated for 15 min and centrifuged at 13,000 rpm for 6 min.
After 15 min the supernatant was decanted and the total drug content solubilised was analysed using
reverse phase HPLC. The solution was analysed using an Agilent 1260 Infinity/DAD (100–900 nm)
(Waldbronn, Germany) with a Zorbax eclipse XDB (8.5
column. The mobile phase comprised of 80:20 acetonitrile: H2O containing 45 mM ammonium formate
µ
m, 4.6 mm
×
150 mm) (Waldbronn, Germany)
buffered to pH 3.5 with formic acid. Flow rate was 1 mL/min with 5 µL injection volume.
3.2. Cell Proliferation Assay
As described in detail previously [28] cell proliferation was measured using a thymidine
incorporation assay by seeding 3000 cells in each well using 96-well plates with varying concentrations
of inhibitors for three days. Experiments were performed in triplicate with a minimum of two
experimental repeats. Briefly, 0.04
five hours, after which the cells were gathered onto glass fibre filters using an automated TomTec
µ
Ci of 3H-thymidine was added to each well and incubated for

Molecules 2018, 23, 145
9 of 14
harvester. Filters were incubated with Betaplate Scint and thymidine incorporation determined with
Trilux/Betaplate counter showing the percentage of cells incorporated with 3H-thymidine into the
DNA helix.
Furthermore, derivatives
3, 4, and 6 were sent to the National Cancer Institute’s Developmental
Therapeutic Program (DTP) where they were screened against a panel of sixty human tumour cell
lines (NCI60, for further information see [18,29,30] and a complete description in the Supplementary
Materials section).
3.3. Preparation of Cholesteryl-Poly(allylamine) (Ch5)
Cholesteryl chloride was grafted in 5% molar ratio to PAA as shown in Figure 8.
Cl
C
O
O
CH2
CH
CH2
NH
C
CH2
CH
CH2
CH
a
(C2H5)3N
b
n
CH2
NH2
CH2
NH2
O
O
Figure 8. Chemical synthesis of cholesterol-PAA (Ch5).
1
3.3.1. H NMR Spectra of Ch5
The Ch5 spectrum (see Figure S1) had peak assignments at: δ0.75, δ1.10, δ1.15, and δ1.25 = CH3
(cholesteryl), δ1.4–2.0 = CH2 (cholesteryl, PAA), δ2.60 = CH2 (cholesteryl), δ2.75–3.00= CH2 (attached to
–NH and –NH2 groups on PAA), δ4.5 = CH=O (carbonyl bond), δ5.5 = C=CH (cholesteryl). The peaks at
δ3.25 and δ4.75 were due to water and MEOD respectively. The degree of hydrophobic modification
could be calculated by comparing the peak at δ0.75 (cholesteryl) with the peaks at δ2.75–3.20 (–CH2
attached to –NH and –NH2 groups) on PAA backbone as follows: (Integration at δ0.75/3)/(Integration
at δ2.75–3.00/2) × 100% = 3.6% hydrophobic substitution.
3.3.2. Elemental Analysis of Ch5 Polymer
Elemental analysis. Modified polymer samples (1 mg) were analysed for abundance of carbon,
hydrogen, and nitrogen (and halogen content) using a Perkin Elmer series 2 elemental analyser
(Perkin Elmer, Oswestry, UK).
The elemental analysis results indicated that the actual % mole hydrophobic grafting per PAA
monomer with Ch5 were in good agreement with the initial molar feed ratio (Table S8).
3.3.3. FTIR of Ch5 Polymer
FTIR analysis of all the polymer was carried out with a Perkin Elmer (Spectrum BX, Oswestry,
UK), fitted with a diamond powder tip. The polymers (5 mg) were placed under the diamond tip and
64 scans were run following a background correction. The FTIR spectrum for Ch5 (Figure S2) shows
−1
the presence of absorbance for the bonds present within the polymer (Table S9). The peak at 2357 cm

Molecules 2018, 23, 145
10 of 14
is due to C–O arising from background CO2 and the broad peak at 3293 cm−1 is due to O–H of water,
which has been absorbed by the hygroscopic polymer. The presence of C=O and C–O peaks due to
absorbance at 1464 cm−1 and 1141 cm−1, respectively, indicating the presence of the carbonyl group
between the backbone and the cholesteryl moiety.
Photon correlation spectroscopy: Polymeric self-assemblies were formed by probe sonication
of the polymers (1 mg/mL) in doubly distilled water before filtering with 0.45 µm syringe filter.
Hydrodynamic diameters and polydispersity index measurements were carried out using a photon
correlation spectrometer (Zetasizer Nano-ZS, Malvern Instruments, Malvern, UK). All measurements
were conducted in triplicate at 25 ◦C and an average value was determined.
3.4. Drug Loading
Polymer solutions of 1 mg/mL, 3 mg/mL, and 6 mg/mL were made up in deionised water and
probe sonicated for 10 min to ensure aggregation of the self-assemblies was achieved. Compound
2
was added in 1:1, 5:1, and 10:1 initial drug: excipient weight ratios and the drug-polymer solutions
were probe sonicated for a further 10 min to ensure maximum encapsulation was achieved. After
cooling to room temperature, the solutions were filtered using 0.45 µm syringe filters (with prefilters)
to ensure any excess un-encapsulated drugs were removed. Drug quantification was carried out
using UV-VIS spectroscopy with the samples diluted in DMSO. Absorbance was measured at 304
nm. Calibration was run from 0.0039–0.0313 mg/mL with R2 = 0.992 (see Supplementary Materials
Figure S3 for the calibration plot).
3.5. In Vitro Drug Release Studies
The formulation (1 mL) was pipetted inside dialysis tubing (MW cut off = 12–14 kDa) and dialysed
against PBS in sink (100 mL, 0.2 M) at 37 ◦C with stirring. At various time points 1 mL of the external
PBS was extracted and replaced with 1 mL of fresh PBS. The presence of drug in the collected PBS was
determined using UV-VIS spectroscopy as before.
3.6. Cytotoxicity of Formulation Using MTT Assay
An MTT assay was carried out using human pancreatic adenocarcinoma (BxPC-3) cells. Cells were
cultured in RPMI medium containing 10% FBS and 1% penicillin streptomycin (Penstrep). Cytotoxicity
assay for compound
2
alone (1–1
×
10−4 mg/mL) and formulations of the Ch5
+2 polymer was carried
out. The experiments were performed in triplicate (n = 3) using eight wells per plate for each data
point. The polymers were fixed at a concentration of 0.005 mg/mL where the cell viability is 90% (IC90)
based on the MTT assay. Polymeric self-assemblies were formed by probe sonicating the polymers
in sterile water (5 mg/mL). The polymer stock solutions were diluted to 0.005 mg/mL with media.
A drug stock solution (20 mg/mL) was prepared by diluting compound
2 in DMSO. The formulations
were prepared by addition of compound 2 into 0.005 mg/mL polymer solution.
3.7. Drug Uptake into BxPC-3 Cells
Cells (3 mL, 150,000 cells/well) were seeded into six-well plates and incubated for 24 h at 37 ◦C
with 5% CO2. The media was replaced with 25
µg/mL and 50 µg/mL of either 2 or Ch5+2 and
incubated for 4 h. The medium was removed and each well was washed with 1 mL PBS before the
addition of 185 µL trypsin into each well. Cells re-suspended in 1 mL media and viable cells were
counted using an automated cell counter (Invitrogen Countess®, Loughborough, UK). Cells (100,000)
were transferred into Eppendorf tubes and centrifuged (800 rpm, 5 min). The supernatant was removed
and cells were resuspended in DMSO and was quantified as previously described.

Molecules 2018, 23, 145
11 of 14
3.8. Molecular Modelling
The compounds were docked to the crystal structure of PLC-
δ1 (PDB ID: 1DJX, resolution
2.3 Å) [31], TDP1 (PDB ID: 1MU7, resolution 2.0 Å) [32], Atox1(PDB ID:1FEE, resolution 1.8 Å) [33],
tubulin-colchicine complex (PDB ID: 4O2B, resolution 2.3 Å) [34], and A2AAR (PDB ID: 3EML,
resolution 2.6 Å) [35], which was obtained from the Protein Data Bank (PDB) [36
,37] using the
GoldScore (GS) [38], ChemScore (CS) [39 40], ChemPLP [41], and ASP [42] scoring functions were
,
implemented to validate the predicted binding modes and relative energies of the ligands using
the GOLD v5.2 software suite. Fifty docking runs were allowed for each ligand with default search
efficiency (100%) with 10 Å radiuses. The basic amino acid residues lysine and arginine were defined as
protonated. Furthermore, aspartic and glutamic acids were assumed to be deprotonated. The Scigress
Ultra version F.J 2.6 program [43] was used to prepare the crystal structures for docking, i.e., hydrogen
atoms were added, the co-crystallised heavy ions and ligands as well as crystallographic water
molecules were removed. The Scigress software suite was also used to build the inhibitors and the
MM2 [44] force field was used to optimise the structures. For PLC-
pocket was defined as the position of the Ca2+ ion (x = 78.799, y = 38.878, z =
the colchicine pocket is used as the binding centre (x = 13.222, y = 8.371, z = 23.331), whilst A2AAR
obtained good consensus upon re-docking of the co-crystalized ligand using (x = 9.42, y = 9.544,
δ1, the centre of the binding
−
19.069), for tubulin
−
−
−
z = 56.644) coordinates. For TDP1, the centre of the binding pocket was defined as the position of the
tungsten (W) of the co-crystallised ligand (x = 8.312, y = 12.660, z = 34.452), while, for Atox1, the centre
of the binding pockets were defined as the position of the sulphur of Cys12 (x = 66.121, y =
z = 36.866).
−
55.634,
The QikProp 3.2 [45] software package was used to calculate the molecular descriptors of the
compounds. The reliability of the prediction power of QikProp is established for the molecular
descriptors used in this study [46].
4. Conclusions
In this work, two strategies were explored in order to enhance the water solubility of the
thieno[2,3-b]pyridine anticancer compounds. First, a morpholine moiety was attached to the molecular
scaffold by substituting the sulphur atom for a nitrogen, leading to the 1H-pyrrolo[2,3-b]pyridine
molecular framework. An excellent increase in solubility was achieved but, concurrently, the anticancer
efficacy was severely diminished. Molecular modelling against the putative targets of the
thieno[2,3-b]pyridines did not give a conclusive result. The lack of efficacy can be hypothesised by
poor cell permeability of the 1H-pyrrolo[2,3-b]pyridine ligands. Alternatively, an unknown molecular
target could be responsible for the efficacy of the thieno[2,3-b]pyridines. Second, as an alternative
strategy, thieno[2,3-b]pyridine derivative
forming nanoparticles. The complex was tested against the BxPC-3 pancreatic cancer cell lines and
showed excellent potency with a measured IC50 value of 0.5 g/mL (1.30 M) as compared with
2 was loaded into the cholesteryl-poly(allylamine) polymer,
µ
µ
2.5 µg/mL (6.48 µM), which is five times more potent than the free derivative, 2.
Supplementary Materials: Supplementary materials are available online.
Acknowledgments: The work was partly supported by the grant from the Auckland Medical Research Foundation
(Ref: 115004).
Author Contributions: A.Z. performed the molecular modelling; L.I.P., N.A.H., M.v.R., and D.B. did the synthetic
work; E.L. conducted the cell proliferation assay, S.K. did the drug solubility experiments; W.A.D. gave advice on
the direction of the project; A.A. and C.H. did the polymer experiments; and J.R. directed the work and produced
the manuscript. All the authors read the paper.
Conflicts of Interest: The authors declare no conflict of interest.

Molecules 2018, 23, 145
12 of 14
References
1.
2.
3.
Feng, L.; Reynisdóttir, I.; Reynisson, J. The effect of PLC-gamma2 inhibitors on the growth of human tumour
cells. Eur. J. Med. Chem. 2012, 54, 463–469. [CrossRef] [PubMed]
Arabshahi, H.J.; Leung, E.; Barker, D.; Reynisson, J. The development of thieno[2,3-b]pyridine analogues as
anticancer agents applying in silico methods. Med. Chem. Commun. 2014, 5, 186–191. [CrossRef]
Hung, J.M.; Arabshahi, H.J.; Leung, E.; Reynisson, J.; Barker, D. Synthesis and cytotoxicity of
thieno[2,3-b]pyridine and furo[2,3-b]pyridine derivatives. Eur. J. Med. Chem. 2014, 86, 420–437. [CrossRef]
[PubMed]
4.
5.
6.
7.
8.
9.
Reynisson, J.; Court, W.; O’Neill, C.; Day, J.; Patterson, L.; McDonald, E.; Workman, P.; Katan, M.; Eccles, S.A.
The identification of novel PLC-c inhibitors using virtual high throughput screening. Bioorg. Med. Chem.
2009, 17, 3169–3176. [CrossRef] [PubMed]
Leung, E.; Hung, J.M.; Barker, D.; Reynisson, J. The effect of a thieno[2,3-b]pyridine PLC-γ inhibitor on DNA
synthesis, morphology, migration and cell cycle of breast cancer cells. Med. Chem. Commun. 2014, 5, 99–106.
[CrossRef]
Reynisson, J.; Jaiswal, J.K.; Barker, D.; D’mello, S.A.N.; Denny, W.A.; Baguley, B.; Leung, E. Evidence that
phospholipase C is involved in the antitumour action of NSC768313, a new thieno[2,3-b]pyridine derivative.
Cancer Cell Int. 2016, 16, 18. [CrossRef] [PubMed]
Arabshahi, H.J.; van Rensburg, M.; Pilkington, L.I.; Jeon, C.Y.; Song, M.; Gridel, L.-M.; Leung, E.; Barker, D.;
Vuica-Ross, M.; Volcho, K.P.; et al. A synthesis, in silico, in vitro and in vivo study of thieno[2,3-b]pyridine
anticancer analogues. Med. Chem. Commun. 2015, 6, 1987–1997. [CrossRef]
Katritch, V.; Jaakola, V.; Lane, J.R.; Lin, J.; IJzerman, A.P.; Yeager, M.; Kufareva, I.; Stevens, R.C.; Abagyan, R.
Structure-Based Discovery of Novel Chemotypes for Adenosine A2A Receptor Antagonists. J. Med. Chem.
2010, 53, 1799–1809. [CrossRef] [PubMed]
Wang, J.; Luo, C.; Shan, C.; You, Q.; Lu, J.; Elf, S.; Zhou, Y.; Wen, Y.; Vinkenborg, J.L.; Fan, J.; et al. Inhibition
of human copper trafficking by a small molecule significantly attenuates cancer cell proliferation. Nat. Chem.
2015, 7, 968–979. [CrossRef] [PubMed]
10. Eurtivong, C.; Semenov, V.; Semenova, M.; Konyushkin, L.; Atamanenko, O.; Reynisson, J.; Kiselyov, A.
3-Amino-thieno[2,3-b]pyridines as microtubule-destabilising agents: Molecular modelling and biological
evaluation in the sea urchin embryo and human cancer cells. Bioorg. Med. Chem. 2017, 25, 658–664. [CrossRef]
[PubMed]
11. Romagnoli, R.; Baraldi, P.G.; Salvador, M.K.; Preti, D.; Tabrizi, M.A.; Bassetto, M.; Brancale, A.; Hamel, E.;
Castagliuolo, I.; Bortolozzi, R.; et al. Synthesis and Biological Evaluation of 2-(Alkoxycarbonyl)-3-
Anilinobenzo[b]thiophenes and Thieno[2,3-b]pyridines as New Potent Anticancer Agents. J. Med. Chem.
2013, 56, 2606–2618. [CrossRef] [PubMed]
12. Huuskonen, J.; Salo, M.; Taskinen, J. Aqueous Solubility Prediction of Drugs Based on Molecular Topology
and Neural Network Modeling. J. Chem. Inf. Comput. Sci. 1998, 38, 450–456. [CrossRef] [PubMed]
13. Llinàs, A.; Glen, R.C.; Goodman, J.M. Solubility Challenge: Can You Predict Solubilities of 32 Molecules Using a
Database of 100 Reliable Measurements? J. Chem. Inf. Model. 2008, 48, 1289–1303. [CrossRef] [PubMed]
14. Pilkington, L.I.; Haverkate, N.A.; van Rensburg, M.; Reynisson, J.; Leung, E.; Barker, D. Synthesis of
3-Amino-2-carboxamide Tetrahydropyrrolo[2,3-b]quinolines. Synlett 2016, 27, 2811–2814. [CrossRef]
15. Apley, M.; Crist, G.B.; Fellner, V.; Gonzalez, M.A.; Hunter, R.P.; Martinez, M.N.; Messenheimer, J.R.; Modric, S.;
Papich, M.G.; Parr, A.F.; et al. Determination of Thermodynamic Solubility of Active Pharmaceutical Ingredients
for Veterinary Species: A New USP General Chapter. Pharm. Forum 2015, 41. [CrossRef]
16. Loftsson, T.; Vogensen, S.B.; Desbos, C.; Jansook, P. Carvedilol: Solubilization and cyclodextrin complexation.
A technical note. AAPS PharmSciTech 2008, 9. [CrossRef] [PubMed]
17. Zafar, A.; Reynisson, J. Hydration Free Energy as a Molecular Descriptor in Drug Design: A Feasibility Study.
Mol. Inf. 2016, 35, 207–217. [CrossRef] [PubMed]
18. Shoemaker, R.H. The NCI60 Human Tumour Cell line Anticancer Drug Screen. Nat. Rev. Drug Dis. 2006
6, 813–823. [CrossRef] [PubMed]
,

Molecules 2018, 23, 145
13 of 14
19. Leung, E.; Pilkington, L.I.; van Rensburg, M.; Jeon, C.Y.; Song, M.; Arabshahi, H.J.; De Zoysa, G.H.; Sarojini, V.;
Denny, W.A.; Reynisson, J.; et al. Synthesis and cytotoxicity of thieno[2,3-b]quinoline-2-carboxamide and
cycloalkyl[b]thieno [3,2-e]pyridine-2-carboxamide derivatives. Bioorg. Med. Chem. 2016, 24, 1142–1154.
[CrossRef] [PubMed]
20. Van Rensburg, M.; Leung, E.; Haverkate, N.A.; Eurtivong, C.; Pilkington, L.I.; Reynisson, J.; Barker, D.
Synthesis and antiproliferative activity of 2-chlorophenyl carboxamide thienopyridines. Bioorg. Med.
Chem. Lett. 2017, 27, 135–138. [CrossRef] [PubMed]
21. Huang, S.N.; Pommier, Y.; Marchand, C. Tyrosyl-DNA Phosphodiesterase 1 (Tdp1) inhibitors. Expert Opin.
Ther. Pat. 2011, 21, 1285–1292. [CrossRef] [PubMed]
22. Gillies, E.R.; Fréchet, J.M.J. Dendrimers and dendritic polymers in drug delivery. Drug Discov. Today 2005, 10,
35–43. [CrossRef]
23. Aliabadi, H.M.; Lavasanifar, A. Polymeric micelles for drug delivery. Expert Opin. Drug Deliv. 2006, 3,
139–162. [CrossRef] [PubMed]
24. Hoskins, C.; Ouaissi, M.; Lima, S.C.; Cheng, W.P.; Loureirio, I.; Mas, E.; Lombardo, D.; Cordeiro-da-Silva, A.;
Ouaissi, A.; Kong Thoo Lin, P. In vitro and in vivo anticancer activity of a novel nano-sized formulation
based on self-assembling polymers against pancreatic cancer. Pharm. Res. 2010, 27, 2694–2703. [CrossRef]
[PubMed]
25. Mosmann, T. Rapid colorimetric assay for cellular growth and survival: Application to proliferation and
cytotoxicity assays. J. Immunol. Meth. 1983, 65, 55–63. [CrossRef]
26. Barnett, C.M.; Gueorguieva, M.; Lees, M.R.; McGarvey, D.J.; Hoskins, C. Physical stability, biocompatibility
and potential use of hybrid iron oxide-gold nanoparticles as drug carriers. J. Nanopart. Res. 2013, 15, 1706.
[CrossRef]
27. Cartiera, M.S.; Johnson, K.M.; Rajendran, V.; Caplan, M.J.; Saltzman, W.M. The uptake and intracellular fate
of PLGA nanoparticles in epithelial cells. Biomaterials 2009, 30, 2790–2798. [CrossRef] [PubMed]
28. Leung, E.Y.; Kim, J.E.; Askarian-Amiri, M.; Rewcastle, G.W.; Finlay, G.J.; Baguley, B.C. Relationships between
Signaling Pathway Usage and Sensitivity to a Pathway Inhibitor: Examination of Trametinib Responses in
Cultured Breast Cancer Lines. PLoS ONE 2014, 9, e105792. [CrossRef] [PubMed]
29. Alley, M.C.; Scudiero, D.A.; Monks, P.A.; Hursey, M.L.; Czerwinski, M.J.; Fine, D.L.; Abbott, B.J.; Mayo, J.G.;
Shoemaker, R.H.; Boyd, M.R. Feasibility of Drug Screening with Panels of Human Tumor Cell Lines Using a
Microculture Tetrazolium Assay. Cancer Res. 1988, 48, 589–601. [PubMed]
30. Boyd, M.R.; Paull, K.D. Some Practical Considerations and Applications of the National Cancer Institute In
Vitro Anticancer Drug Discovery Screen. Drug Dev. Res. 1995, 34, 91–109. [CrossRef]
31. Essen, L.O.; Perisic, O.; Katan, M.; Wu, Y.; Roberts, M.F.; Williams, R.L. Structural mapping of the catalytic
mechanism for a mammalian phosphoinositide-specific phospholipase C. Biochemistry 1997, 36, 1704–1718.
[CrossRef] [PubMed]
32. Davies, D.R.; Interthal, H.; Champoux, J.J.; Hol, W.G.J. Crystal Structure of a Human Tyrosyl-DNA
Phosphodiesterase (Tdp1)-Tungstate Complex. J. Mol. Biol. 2003, 324, 917–932. [CrossRef]
33. Wernimont, A.K.; Huffman, D.L.; Lamb, A.L.; O’Halloran, T.V.; Rosenzweig, A.C. Structural basis for copper
transfer by the metallochaperone for the Menkes/Wilson disease proteins. Nat. Struct. Biol. 2000, 7, 766–771.
[CrossRef] [PubMed]
34. Prota, A.E.; Danel, F.; Bachmann, F.; Bargsten, K.; Buey, R.M.; Pohlmann, J.; Reinelt, S.; Lane, H.;
Steinmetz, M.O. The Novel Microtubule-Destabilizing Drug BAL27862 Binds to the Colchicine Site of
Tubulin with Distinct Effects on Microtubule Organization. J. Mol. Biol. 2014, 426, 1848–1860. [CrossRef]
[PubMed]
35. Jaakola, V.P.; Griffith, M.T.; Hanson, M.A.; Cherezov, V.; Chien, E.Y.; Lane, J.R.; Ijzerman, A.P.; Stevens, R.C.
The 2.6 Angstrom Crystal Structure of a Human A2A Adenosine Receptor Bound to an Antagonist. Science
2008, 322, 1211–1217. [CrossRef] [PubMed]
36. Berman, H.; Henrick, K.; Nakamura, H. Announcing the worldwide Protein Data Bank. Nat. Struct. Biol.
2003, 10, 980. [CrossRef] [PubMed]
37. Berman, H.M.; Westbrook, J.; Feng, Z.; Gilliland, G.; Bhat, T.N.; Weissig, H.; Shindyalov, I.N.; Bourne, P.E.
The Protein Data Bank. Nucleic Acids Res. 2000, 28, 235–242. [CrossRef] [PubMed]
38. Jones, G.; Willet, P.; Glen, R.C.; Leach, A.R.; Taylor, R. Development and Validation of a Genetic Algorithm
for Flexible Docking. J. Mol. Biol. 1997, 267, 727–748. [CrossRef] [PubMed]

Molecules 2018, 23, 145
14 of 14
39. Eldridge, M.D.; Murray, C.; Auton, T.R.; Paolini, G.V.; Mee, P.M. Empirical scoring functions: I.
The development of a fast empirical scoring function to estimate the binding affinity of ligands in receptor
complexes. J. Comput.-Aided Mol. Des. 1997, 11, 425–445. [CrossRef] [PubMed]
40. Verdonk, M.L.; Cole, J.C.; Hartshorn, M.J.; Murray, C.W.; Taylor, R.D. Improved protein-ligand docking
using GOLD. Proteins 2003, 52, 609–623. [CrossRef] [PubMed]
41. Korb, O.; Stützle, T.; Exner, T.E. Empirical Scoring Functions for Advanced Protein Ligand Docking with
−
PLANTS. J. Chem. Inf. Model. 2009, 49, 84–96. [CrossRef] [PubMed]
42. Mooij, W.T.M.; Verdonk, M.L. General and targeted statistical potentials for protein–ligand interactions.
Proteins 2005, 61, 272–287. [CrossRef] [PubMed]
43. Scigress Ultra v. F.J 2.6 FujitsuLimited; Fujistu: Tokyo, Japan, 2000.
44. Allinger, N.L. Conformational analysis. 130. MM2. A hydrocarbon force field utilizing V1 and V2 torsional
terms. J. Am. Chem. Soc. 1977, 99, 8127–8134. [CrossRef]
45. QikProp; v3.2; Schrödinger: New York, NY, USA, 2009.
46. Ioakimidis, L.; Thoukydidis, L.; Naeem, S.; Mirza, A.; Reynisson, J. Benchmarking the Reliability of QikProp.
Correlation between Experimental and Predicted Values. QSAR Comb. Sci. 2008, 27, 445–456. [CrossRef]
Sample Availability: Sample Availability: Samples of the compounds are available from the authors.
©
2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).