Comparative characterization of experimental and calculated lipophilicity and anti-tumour activity of isochromanone derivatives

Article abstract of DOI:10.2174/092986710790192703

Compound lipophilicity connected to ADME(T)^a has great importance in drug development and it has to be evaluated by the generally used drug developmental process. In addition to the importance of lipophilicity in ADMET, recently it has been reported that lipophilicity of small molecules correlates with their antiproliferative activity because of certain specific hydrophobic and lipophilic interactions. Due to the complexity of ADME(T) parameters an efficient and fast method is needed to characterize the many promising candidate lead molecules as a preselection in order not to be rejected from the latter phase of drug development. In the present paper we provide an overview of the importance of lipophilicity of drug candidates for biological action and for ADME(T) and describe a novel approach for drug-likeness characterization of a molecular library using correlation study between lipophilicity and biological activity. Lipophilicity and molecular characteristics have been measured, predicted and optimized for a diverse library from which the best members have been selected to describe their biological, chemical and drug-likeness properties. Molecules were selected from the family of α,β-unsaturated ketones and thorough HPLC characterization for lipophilicity and morphological, antiproliferative and flow cytometric studies were carried out on them. Based on the results 17 member isochromanone library including E and Z geometric isomers were selected for further characterization. In this focused library linear correlation has been found between the calculated and measured lipophilicity and significant parabolic correlation was found between the antiproliferative effect and lipophilicity. Using our efficient and fast method, from a diverse library, we identified an outstandingly effective inhibitor of A431 tumour cell growth via a PARPa cleavage dependent apoptosis. In summary the optimized HPLC analyses of lipophilicity combined with the cell-culture assay, introduced above, resulted in the determination of an optimal lipophilicity range. This optimized lipophilicity range should be used in designing novel antiproliferative compounds.

Full text of DOI:10.2174/092986710790192703

Current Medicinal Chemistry, 2010, 17, 321-333
321
Comparative Characterization of Experimental and Calculated Lipophilic-
ity and Anti-Tumour Activity of Isochromanone Derivatives
,1
1
1
2
2
1
1
,1
M. Huszár* , A. Varga , A. Horváth , T. Loránd , A. Agócs , M. Idei , J. Mandl , T. Vántus* and
G. Kéri*
,1
1
Pathobiochemistry Research Group, Hungarian Academy of Sciences, Semmelweis University, Budapest, Hungary
Department of Biochemistry and Medical Chemistry, Faculty of Medicine, University of Pécs, Pécs, Hungary
2
a
Abstract: Compound lipophilicity connected to ADME(T) has great importance in drug development and it has to be
evaluated by the generally used drug developmental process. In addition to the importance of lipophilicity in ADMET, re-
cently it has been reported that lipophilicity of small molecules correlates with their antiproliferative activity because of
certain specific hydrophobic and lipophilic interactions. Due to the complexity of ADME(T) parameters an efficient and
fast method is needed to characterize the many promising candidate lead molecules as a preselection in order not to be re-
jected from the latter phase of drug development. In the present paper we provide an overview of the importance of lipo-
philicity of drug candidates for biological action and for ADME(T) and describe a novel approach for drug-likeness char-
acterization of a molecular library using correlation study between lipophilicity and biological activity. Lipophilicity and
molecular characteristics have been measured, predicted and optimized for a diverse library from which the best members
have been selected to describe their biological, chemical and drug-likeness properties. Molecules were selected from the
family of ꢀ,ꢁ-unsaturated ketones and thorough HPLC characterization for lipophilicity and morphological, antiprolifera-
tive and flow cytometric studies were carried out on them. Based on the results 17 member isochromanone library includ-
ing E and Z geometric isomers were selected for further characterization. In this focused library linear correlation has
been found between the calculated and measured lipophilicity and significant parabolic correlation was found between the
antiproliferative effect and lipophilicity. Using our efficient and fast method, from a diverse library, we identified an out-
a
standingly effective inhibitor of A431 tumour cell growth via a PARP cleavage dependent apoptosis. In summary the op-
timized HPLC analyses of lipophilicity combined with the cell-culture assay, introduced above, resulted in the determina-
tion of an optimal lipophilicity range. This optimized lipophilicity range should be used in designing novel antiprolifera-
tive compounds.
Keywords: Lipophilicity, apoptosis, isochromanone derivative, antiproliferative effect.
INTRODUCTION
It is a plausible alternative to use quick, automatizable
separation methods of high performance and low sample
demand (e.g. RP-HPLC, MEKC) as a substitute for the clas-
sical slow and uncomfortable shake-flask method to charac-
terize lipophilicity of a compound [13-17]. A great number
of publications on the efforts made to adjust HPLC methods
and to improve stationary phases to substitute Po/w measure-
ments are well reviewed by Testa [18] or Valkó [6].
Physico-chemical parameters are among the most impor-
tant properties to be evaluated in the early drug developmen-
tal phase [1-4]. Compounds having improper pharmacoki-
netic or pharmaco-chemical characteristics cannot be consid-
ered as drug candidate molecules [5]. One of the most impor-
tant physico-chemical parameters, in which case early pre-
screening is available, is lipophilicity [6]. In the last few
years it has been realized that the correlation analysis of
lipophilicity and biological activity, such as cytotoxicity,
antiproliferative effect or apoptosis, has a major impact on
drug development. If a precisely defined relationship can be
determined it can reduce the cost needed during the selection
of active drug candidates and can be used for prediction and
estimation of biological effectiveness [7-12].
Another choice to characterize lipophilicity of a molecule
is the computerized calculation frequently based on the
fragment approach [13, 19, 20] or on atomic-based (e.g. Crip-
pen’s, Viswanadhan’s, Broto’s method) [21] or molecular
a
lipop-hilicity potential (MLP ), developed by Testa et al.
methods [22].
Lipophilicity strongly depends on structural properties.
Adrien Albert was one of the first researchers who studied
the relationship between the structure and the activity of dif-
Lipophilicity of a non-ionic compound whose partition is
independent of the pH is commonly characterized by the n-
a
a
ferent drug molecules using QSAR studies. That was later
octanol/water (biphasic) partition coefficient (Po/w , and
extended to carry out relationship analyses for different
physico-chemical parameters, between these parameters and
the biological activity [23-26].
logPo/w). It has long been recognised that the retention of a
a
compound in different separation methods (RP-HPLC , mi-
a
cellar electrokinetic chromatography (MEKC ), micellar
liquid chromatography) is governed by its lipophilicity and
thus it correlates with the logPo/w measured in n-octanol-
water system.
Lipophilicity can be modified and easily determined by
characterizing molecules prepared with different moieties.
On the base of that correlation studies using lipophilicity
parameter as a basic factor could be used for prediction of
biological activity.
*
Address correspondence to these authors at the Pathobiochemistry Re-
The relationship between the antiproliferative activity
and lipophilic property has been established for different
search Group, Hungarian Academy of Sciences, Semmelweis University,
Budapest, Hungary; Tel: +36 1 4591500/60170; Fax: +36 1 4591500/60162;
E-mail: monika.huszar@eok.sote.hu, keri@eok.sote.hu
0
929-8673/10 $55.00+.00
© 2010 Bentham Science Publishers Ltd.

3
22 Current Medicinal Chemistry, 2010 Vol. 17, No. 4
Huszár et al.
molecular libraries such as platinum complexes [10], dehy-
droartemisinin derivatives [11], calcium antagonist drug
molecules [12] or isoquinoline-4,6-dione analogues where
the assumption was that the correlation between the lipophil-
icity and the antitumour activity on A549 non-small-cell
drug resistant lung carcinoma cells was caused by a change
in the receptor structure termed allosteric reaction [27-28].
These studies established linear correlation mainly but para-
bolic relationship could also be found in some cases.
the lipophilicity which generated the conclusion that the
lipophilic pocket interacted with the hydrophobic moiety of
the cyclic pseudopeptide [34]. Regan et al. [35-36] and Kim
et al. [37] in their structure-activity studies also emphasized
the connection between lipophilicity and the role of a lipo-
philic pocket. In the latter study influenza neuraminidase
inhibitors have been investigated and found that the inhibi-
tory activity depends on the presence of a hydrophobic do-
main. Regan et al. studied a specific compound, called 1-(5-
tert-butyl-2-p-tolyl-2H-pyrazol-3-yl)-3-[4-(2-morpholin-4-
yl-ethoxy)naphthalen-1-yl]urea against p38ꢀ MAP kinase.
Their aim was to find a method for the treatment of autoim-
mune diseases. According to their observation a lipophilic
domain of the kinase has major role in the inhibition, as the
tert. butyl group of the drug candidate fit into the activation
loop of the kinase completely.
In this paper, with the collection of data from several
publications and on the basis of our own data, we would like
to provide a full insight how the relationship can be de-
scribed between physico-chemical parameters, especially
lipophilicity, and biological parameters, how the molecule-
structure can affect the biological activity or how this struc-
tural property is in connection with the presence of activation
or hydrophobic pocket.
On the basis of the described studies, it seems that lipo-
philicity indeed has a great importance in pre-screening and
drug development.
Investigating inhibitors of choline kinase, a novel vali-
dated target kinase, Campos et al. proved the critical impor-
tance of the relationship between lipophilicity, structural
properties and antitumour activity resulting in novel
achievements in anticancer drug design [29].
In different biological processes the limitation of the
lipophilicity value is indispensable. Several studies have
been made to estimate the upper and lower level of this pa-
rameter and to estimate the optimal value of it [38-43].
Villar et al. and Encío et al. studied the effect of struc-
tural properties on the biological activity in case of
benzo[b]thiophene 1,1-dioxide and its derivatives. It was
found that these molecules have high cytotoxic and inhibi-
tory activity on tumour-associated NADH oxidase [8, 9] and
on human leukemic CCRF-CEM cells [7] confirming that
cytotoxic effect is in some cases in a clear correlation with
the lipophilicity (logP) due to structural phenomenon. Based
on that, the authors synthesised a new benzo[b]thiophene
The ideal range can be dependent on several key factors,
such as the cell or tissue types (where the molecule should
penetrate through) [5].
Coats et al. were among the first researchers who pointed
clearly out the need of an optimal lipophilicity range in drug
discovery. They used parabolic correlation analysis, plotting
the cytotoxic activity against the lipophilicity of copper
chelates, to determine an optimal range [44].
1
,1-dioxide derivative with relatively higher lipophilicity,
predicted that it has strong inhibitory effect on six human
tumour cell lines. They also found quantitative structure-
activity relationship when this drug candidate was tested
under reductive and oxidative conditions. In reductive condi-
tion they found high inhibitory activity on the tNADH oxi-
dase while in oxidative condition the compound was abso-
lutely inactive [8, 9].
Michael J. Waring recently published a paper where this
limitation was studied. For a very long time, only the upper
level of compound lipophilicity has been used as criteria in
drug design. Nowadays it is already determined that the too
polar molecules can also fail in clinical studies [45].
Upper limitation is required due to several expected drug
properties such as water solubility (which is very important
especially by orally applied drugs), easier and quicker excre-
tion and clearance through the kidney, faster transport by the
plasma and bigger concentration in it [3, 46]. It has recently
been stated that the risk for adverse toxicological behaviour
increases by drugs having high logP value [38-42, 45]. The
determination of a lower limit is also essential. Estimating
the optimal lipophilicity range it has to be concerned that
while too low lipophilicity reduces the ability of a drug can-
didate to penetrate into the hydrophobic membrane then too
high logP value increases the possibility of the drug mole-
cule sticking in the membrane [47-48]. Veldman et al.
showed an optimal lipophilicity range for the drug transport
through the cell membrane using parabolic correlation analy-
sis [49].
The relationship between lipophilicity and biological ac-
tivity was also shown by Toth et al. [30] In their effort to
find an inhibitor to the human neutrophil elastase, which has
a significant role in several disorders such as rheumatoid
arthritis or periodontitis, they tested several peptides with
increasing number of lipophilic amino acids coupled to them.
They found an increase in biological activity with the grow-
ing number of the lipophilic amino acid in the conjugate. It
was confirmed that the presence of a hydrophobic binding
site, a hydrophobic pocket, which was located near to the
active centre of the human neutrophil elastase, most likely
has an impact on the increased biological activity. This find-
ing also shows a clear correlation between lipophilicity and
biological activity [30-33].
Interaction between a hydrophobic pocket and an inhibi-
tor is strongly depending on the lipophilicity of the drug
candidate. Quartare et al. studied cyclic pseudopeptide NK-2
antagonists and determined that the potency of the antago-
nists increased with the lipophilic side chains. They also
found linear correlation between the biological activity and
The drugs penetration through the blood-brain barrier has
also great importance for molecules which are expected to
have an effect in the brain. Their interaction with the blood
brain barrier should be proper and the brain uptake of drug
molecules highly depends on the lipophilic property [50-53].
Too low lipophilicity prohibits the penetration through the
blood-brain barrier, high lipophilicity causes a limitless

Lipophilicity and Anti-Tumour Activity of Isochromanone Derivatives
Current Medicinal Chemistry, 2010 Vol. 17, No. 4
323
fluxuation through it and too high may cause trapping of the
compounds inside the membrane [54].
tural properties of potentially active drug candidate mole-
cules. A promising hit-lead molecule with drug-like charac-
ter and strong antitumour effect on A431 cancer cells was
selected during the determination of lipophilicity optimal
range.
Very often a parabolic relationship has been observed be-
tween the logP and the brain uptake. Timmermans found that
optimal lipophilicity value for the ideal concentration of the
drug in the blood is logP value of 2.16 [55].
MATERIALS AND METHODS
According to all of these, several predictions can be made
beware of the lipophilicity value of a drug candidate but one
of the most essential is to estimate biological effectiveness
such as cytotoxicity, antiproliferative activity or apoptotic
effect.
a
Triethylamine, ACN , orthophosphoric acid, methanol,
ethanol, piperidine, Methylene Blue dye and the chemicals
utilised in the synthesis (aldehydes) were purchased from
Fluka (Buchs, Switzerland). Solutions were prepared of
deionised, bacteria-free water made by Elgastat UHP system
In the course of drug development we investigated a wide
spectrum of molecules with different core structures. To our
lipophilicity studies we selected tetralones, chromanones,
benzylidencycloalkanones and isochromanones.
(
Elga Ltd. Bucks, England). A431 cells (ECACC
No.85090402, European Collection of Cell Cultures, Salis-
bury, Wiltshire, UK), PARP antibody (Cell Signalling Tech-
a
nology, Inc. Danvers, MA, USA), PI (Sigma Aldrich Inc ,
The naturally occurring isochromanone derivatives dis-
play various biological effects. Some ꢀ,ꢁ-unsaturated ke-
tones are also known as synthetic homoisoflavones. 3-
Isochromanones exhibited antibacterial, antifungal, insecti-
cide and phytotoxic effects (oosponol, fusarubin and
canescin) [56-60] and also play an important role in the utili-
zation of all fluorene carbons in fluorene metabolism by Ar-
throbacter sp. strain F101 [61]. Some homoisoflavones as
tetralone, chromanone or thiochromanone derivatives [62-
a
St. Louis, MO, USA), DMEM cell culture medium (Sigma
Aldrich Inc , St. Louis, MO, USA), Trypsine (Sigma Aldrich
a
Inc., St. Louis, MO, USA), FCS , cell culture flask (Corning
Incorporated, Corning, NY, USA), 24-well plate (Sarstedt
a
Inc, Newton, NC, USA), MB stain (Sigma Aldrich Inc , St.
Louis, MO, USA), anti-mouse secondary Ab (Amersham,
Buckinghamshire, U.K), X-ray film (KODAK X-OMAT AR
FILM, Cedex, France).
6
3] have also been screened against human pathogenic
Synthesis
yeasts (Candida spp). Some of them showed in these in vitro
tests as high activity (MIC values ) as 1.5-6 μg/mL [64].
a
A series of 4-(arylmethylene)-3-isochromanones (Fig.
(
1)) has been prepared by base catalysed Knoevenagel
condensation as it has been published in [70]. The synthetic
method involves solvent free condensation of 3-
Chromanone derivatives are also well known of their
strong antiproliferative activity. [65-66]. In our previous
studies we have already described molecular libraries, which
belong to the same molecular family as the isochromanones,
such as E-2-arylmethylene-1-tetralones, E-3-arylmethylene-
chroman-4-ones, E-3-arylmethylene-1-thiochroman- 4-ones
and aurones and their derivatives [67-68]. These compounds
have also been tested for their antiproliferative effect which
showed good correlation plotted against the measured lipo-
philicity. Moreover by the tetralone derivatives we were able
to determine an optimal lipophilicity value which belongs to
the best molecules with antiproliferative effect [67].
a
isochromanone and the appropriate aldehyde in equimolar
amounts in the presence of catalytic amount of piperidine at
o
1
40 C under argon atmosphere. The separation of the E-Z-
isomers and the purification of the reaction mixture were
performed by column chromatography and the title com-
pounds were recrystallized from ethanol.
8
1
8
a
2
7
6
O
3
O
4
a
O
O
To establish a system specific although in basics, gener-
ally applicable interrelationship between lipophilicity and
certain cellular functions, we further extended our studies
where we closely investigated the isomeric 4-arylmethylene-
5
4
a
R
R
Fig. (1). 4-(arylmethylene)-3-isochromanones (E- and Z-isomers).
3
-isochromanones [69-70] as a new type of antiproliferative
compound library. The type of the substituent on the 3-
isochromanone molecule was varied in order to investigate
the structure- lipophilicity - biological activity relationship.
We clearly showed how this correlation analysis can be ap-
plied to predict the efficiency of inhibitory activity on our
cancer cell line system of selected molecular libraries with
the same core-structure.
Stereochemical and conformational analysis were per-
1
13
formed by H and C NMR methods, and X-ray crystallog-
raphy has been used to complement the configurational as-
1
3
signment for a representative product [62]. The C NMR
spectroscopy was applied for the assignment of the configu-
ration of these ene-lactones.
HPLC Measurements
In harmony with the literature data and as an extension of
our previous studies [69-71], we created a molecular library
where the lipophilicity and the biological activity of the
molecules are in parabolic relationship. This established cor-
relation enabled us to determine an optimal lipophilicity
range for biological studies. This optimal range comprises
inhibitor molecules with the greatest biological activity and
using this area further prediction could be made for the struc-
For high performance liquid chromatographic analysis
stock solutions (0.5 mg/ml) of the samples in ACN: water
3:1) were prepared freshly before the analysis and filtered
through a 0.2 μm Millipore filter unit. HPLC analysis of the
samples was performed with Varian (Basel, Switzerland)
(
9
012 Solvent Delivery System, Varian 9065 Polychrom Di-

3
24 Current Medicinal Chemistry, 2010 Vol. 17, No. 4
Huszár et al.
ode Array Detector; column: Hypersil 5 MOS 5 μm , 250 x
.6 mm (BST, Hungary); injector: Rheodyne. In many cases
Flow Cytometric Studies
4
A431 cells were incubated at 37 °C for 24 h in DMEM
supplemented with 10 % FCS using 24-well plates, seeding
different alkyl-ammonium phosphates were used in the elu-
ent for buffering when the determination of the interaction
between the analyte and the hydrophobic alkyl chains of the
stationary phase is intended [16, 72-75]. Based on the former
5
0 000 cells per well. After treatment, the cells were washed,
fixed with 70 % ethanol, and were stored at -20 °C or used
for PI staining. The stained cells were subjected to flow cy-
a
practice triethyl-ammonium phosphate (TEAP ) was chosen
a
tometry (FACS Calibur, BD Biosciences) to detect and
as mobile phase additive. Eluents: A: 0.083 M TEAP (made
by weighing the calculated quantities of triethyl amine and
phosphoric acid), pH 2.25; B: 95 % ACN + 5 % A (TEAP).
quantify apoptosis. Cells with their DNA content less than
that of G1 phase cells (sub-G1) were assumed to be apop-
totic [79]. The analysis was performed with CellQuest soft-
ware.
Isocratic runs were performed in an eluent of 40 v/v%
o
ACN in A eluent; flow-rate: 1 ml/min; temperature: 20 C.
Injected volume 20 μl, the compounds investigated here
Immunoblotting Studies
were injected individually, number of intra-day repetitions
was 3 (n=3) for each of the compounds. The k values of the
2
a
For western blot experiments A431 cells in 25 cm cul-
ture flasks were treated for 24 hours with 8E applying the
doses of 1, 5 and 10 μM. Samples were lysed by adding a
Triton-based lysis buffer with protease and phosphatase in-
hibitors followed by a 10-min incubation on ice and spinning
down at 12 000 g for 10 min at 4°C. The Triton-soluble su-
pernatant was used in further immunoblot determinations.
The Triton-soluble lysate was mixed with sample buffer,
boiled for 5 min, run on 8 % SDS-PAGE, and blotted onto
nitrocellulose sheets. Blots were processed using mouse anti-
cleaved PARP Ab (Cell Signaling Technology Danvers, MA,
USA; Cat. no:9546) followed by a peroxidase-labeled anti-
mouse secondary Ab (Cell Signalling Technology Danvers,
MA, USA; Cat. no:7076). Blots were developed using Am-
ersham’s enhanced chemiluminescence system (ECL West-
ern Blotting Detection Reagents) and exposed to X-ray film
samples were calculated from the experimentally determined
a
retention data: (k = (t
R
-t
o
)/t
o
, where t
o
has been determined
by injection of water [76-77]). Correlation between the k and
a
CLOGP has been investigated.
Calculation of CLOGP Data
The software-predicted lipophilicity of the compounds
was calculated with the program called 3DNET4W (Vichem
Ltd., 1022 Budapest, Hermann O. u. 15., istvan.kovesdi@
eqnet.hu <mailto:istvan.kovesdi@eqnet.hu>, *2002*) based
on the number and the position of the fragment. Thus the
calculations set out from the base structure of the molecule
and the number, the type and the position of the substituent,
but they cannot distinguish between the different geometric
isomers.
(
KODAK X-OMAT AR FILM).
RESULTS AND DISCUSSION
Antiproliferative Assay
A431 human epidermoid carcinoma cell line was used
utilizing the MB test [78]. This colorimetric assay based on
the enzyme activity of various dehydrogenases of the living
cells, is suitable for testing the cytotoxic activity of anti-
tumour candidates in vitro. Human A431 epidermoid carci-
noma cells were cultured in DMEM supplemented with 10%
FCS (200 mM L-glutamine, 10000 U/ml penicillin and 10
A reliable, fast and accurate HPLC method was devel-
oped to investigate the 17-member molecular library. The
lipophilicity of the compounds was evaluated not only ex-
perimentally (HPLC) but by in silico calculation based on
their chemical structure, too. Antiproliferative effect of the
molecules was determined on A431 cells by the MB method
[78]. Both the relationship between the measured (k) and
calculated (CLOGP) lipophilicity data and that of the bio-
logical activity and the lipophilicity data obtained by HPLC
or by calculation have been investigated. Possibility of cell
death after treatment with the isochromanones has been de-
termined by light microscopic analysis based on the cell
morphology. Furthermore, more specific biological studies,
like PARP fragmentation by western blotting and flow cy-
tometry were applied to investigate the possibility of apopto-
sis.
2
mg/ml streptomycin (Gibco Life Sci) at 37˚C and 5% CO .
Cells were seeded into 96-well plates and incubated for 16
hours before serial dilutions of compounds were added (3
replicates have been made at every concentration for each of
the compounds investigated). Antiproliferative efficacy was
assessed after 48 hours: cells were fixed by 10% buffered
paraformaldehyde in 0.9% NaCl. Wells were then stained by
1
% MB, followed by thorough washing. Both apoptotic and
necrotic cells previously detached from the surface of wells
are thus removed. MB stain from cells entrapped on the plate
surface was dissolved by ethanol (100%): 0.1 M HCl 1:1 and
optical densities measured by a microtiter plate photometric
reader at 650 nm.
Chromatographic Separation
The RP-HPLC method applied in this work proved to be
applicable for fast analysis of the isochromanone molecular
library.
Microscopic Studies
Isocratic separation was achieved within 14 minutes, k
values are shown in Table 1. Our chromatographic method
was able to perceive the small structural differences resulting
in fine alteration of lipophilicity (in contrast of calculation
method). Ability of the method to separate members of the
A431 human epidermoid carcinoma cells maintained in
culture flasks or 24-well plates were studied and photo-
graphed using Zeiss Axiovert 200 microscope equipped with
photo camera at 200x magnification.

Lipophilicity and Anti-Tumour Activity of Isochromanone Derivatives
Current Medicinal Chemistry, 2010 Vol. 17, No. 4
325
library is demonstrated on a representative chromatogram:
baseline separation has been achieved for the samples 2E,
separated well with a difference of 0.4 in their k values.
These results confirm the importance of the spherical factor:
not only the chemical nature of the substituent, but its posi-
tion is also able to influence lipophilicity of the molecule as
it has been found earlier for the aurones, thioaurones, 2-
aryliden-tetralones and Mannich ketones, too [67-68, 80-81].
3
E, 5E, 7E, 11E and 12E (Fig. (2)).
The chromatographic method worked well for the separa-
tion of the structural isomers, too. Impact of the spherical
factor on the k value was shown by the structural isomers,
where for example two compounds, a monomethoxy and a
dimethoxy derivative (5E and 7E, respectively) eluted with a
The synthesis by Knoevenagel condensation may result
in both the E- and Z- isomers. The separation of isomers can
be performed only by chromatographic method. That is also
the reason why in correlation study the k value is more suit-
able than the CLOGP. The method reported here was able to
0
.3 difference in their retention coefficient (Table 1). Not
only the monomethoxy and dimethoxy compounds (5E, 7E),
but the isomeric dimethoxy compounds (7E, 8E) have been
Table 1. Calculated CLOGP and Measured k Values and Antiproliferative Activity of 17 Compounds
*
k
**
Compound
Substituent (R)
CLOGP
IC50 / μM
1
1
2
3
3
4
5
6
6
7
8
9
9
E
Z
E
E
Z
Z
E
Z
E
E
E
Z
E
Ph
1.7
1.6
0.4
1.2
1.1
3.3
1.9
2.4
2.4
1.6
2.0
1.4
1.2
1.3
1.0
2.6
0.6
3.8
3.8
2.3
2.5
2.5
5.3
3.8
4.5
4.5
3.5
3.8
3.6
3.6
3.0
3.0
4.7
3.2
6.3
20.6
20.6
86.7
100.0
100.0
8.1
Ph
3'-pyridyl
1-methyl-2’-pyrrolyl
1-methyl-2’-pyrrolyl
2 6 3
2',6'-Cl -C H
2'-CH O-C
3
6
H
4
2'-Cl-C
2'-Cl-C
6
6
H
4
34.8
7.1
H
4
2',3'-(CH
3
O)
2
-C
6
H
3
15.2
0.3
2',4'-(CH
3
O)
2 6 3
-C H
2'-O
2'-O
2
2
N-C
6
6
H
H
4
55.0
14.1
51.0
55.0
10.3
7.6
N-C
4
10Z
10E
11E
12E
2'-furyl
2'-furyl
2'-Br-C
6 4
H
3'-OH-C
6
H
4
*
*
R.S.D. of the k values was less, than 2%, the compounds investigated here were injected individually, number of intra-day repetitions was 3 (n=3) for each of the compounds.
* R.S.D. of the IC50 values were less, than 10 %, three replicates have been made for each of the compounds.
Fig. (2). Representative chromatogram of the separation of the compounds 2E, 3E, 5E, 7E, 11E, 12E.

3
26 Current Medicinal Chemistry, 2010 Vol. 17, No. 4
Huszár et al.
Fig. (3). Separation of the 9E/Z isomer pair.
distinguish between the E- and Z- isomers, it was able to
separate the isomers 1E – 1Z, 9E – 9Z and 10E – 10Z (Table
optimal lipophilicity range not only with optimal k values
but optimal CLOGP data as well.
1
). As a representative, Fig. (3) shows the elution profile of
the 9E and 9Z isomers.
Results of the Antiproliferative Assay
The A431 human epidermoid carcinoma cell line is well
known as epidermal growth factor overexpressing system
and indicates potential EGFR -related apoptotic effects.
CLOGP Values
a
Investigating the relationship between the experimental
and predicted lipophilicity values a good linear correlation
has been found:
Small molecules as inhibitors of protein kinases constitute
one of the most major class of the target-selective agents and
this system is a straightforward screening setup to test com-
pounds with EGFR inhibitory potential. As several protein
kinases comprise the above mentioned hydrophobic pocket
their inhibition highly depends on the compound lipophilic-
ity [35-36]. Thus determination of optimal lipophilicity
range can be used with higher impact for the prediction of
biological activity. The investigated compounds were evalu-
ated in this screening system to determine their cytotoxicity
features, to examine the impact of the various substituents on
the bioactivity regarding to their ability to alter lipophilicity
and to determine an optimal lipophilicity range where the
inhibitory activity is on maximum level.
CLOGP = A ꢀ log k + B
Correlation was determined for the whole set of com-
pounds investigated (Fig. (4) and Table 2).
IC50 -k Relationship
As lipophilicity can be a great predictor for biological ac-
tivity, it is a generally used method to compare the inhibitory
efficiency to the lipophilicity (logP or k value). For instance
Fish et al. studied the relationship between the selective in-
hibition of noradrenalin reuptake (NRI) and the CLOGP.
While they could not find an optimal lipophilicity range they
discovered that the inhibitory activity of NRI increases with
the lipophilicity [82].
Fig. (4). Calculated (CLOGP) vs. measured (logk) lipophilicity
values of the whole molecular library (Table 1). Abscissa: logk;
ordinate: CLOGP.
However calculation method is mainly insensitive for lit-
tle structural differences and thus the chromatographic
method is more suitable for the characterisation of the lipo-
philicity, CLOGP values proved to be also informative and
useful. As we found linear correlation between the measured
and calculated lipophilicity, biological activity indirectly can
also be predicted from the CLOGP parameter. Moreover
according to the good correlation we are able to describe the
Four of the 17 isochromanones investigated showed good
antiproliferative activity with low IC50 (antiproliferative ef-
fect) values in the μM range (Table 1 compounds: 5E, 6E,
8E, 12E). The E-isomers having phenyl-ring in the R –
substituent showed the greatest biological activity. The high-
est activity was shown by the compound 8E (R:2’,4’-
(OCH ) -C H ) with as low IC50 values as 0.3 (exactly 0.266
3 2 6 3
μM). Generally, the E-isomers showed higher antiprolifera-
tive activity than their Z-counterparts.

Lipophilicity and Anti-Tumour Activity of Isochromanone Derivatives
Current Medicinal Chemistry, 2010 Vol. 17, No. 4
327
Table 2. Correlation Parameters of the CLOGP– logk Relationship for the Whole Set of Molecular Library and Separately for the
E- and Z-Isomers
Parameters of the linear*
CLOGP vs. logk
CLOGP vs. logk by the E-isomers
CLOGP vs. logk by the Z-isomers
A
B
3.11560
3.09785
0.8724
0.41319
<0.0001
17
3.20941
2.54915
0.86425
0.39200
-0.46580
0.18444
0.98340
0.03797
R
SD
P
-
4
-4
6.0005*10
11
4.1128*10
N
6
*
A, B, R, SD, P and N are the parameters of the linear.
A: intercept value and its standard error, B: slope value and its standard error, R: correlation coefficient, SD: standard deviation of the fit, P: probability (that R is zero), N: number of
data points.
In our laboratory Hallgas et al. has already succeeded to
determine an optimal lipophilicity range, plotted IC50 versus
k, on the structurally similar tetralone library. Using para-
bolic regression a minimum range of IC50 data was deter-
mined by the set of optimal lipophilicity values [67].
Parabolic correlation and therefore optimal k value could
also be determined by the separately studied E- (Fig. (6)),
(Table 3) and Z-isomers.
With this new criterion, according to our successful stud-
ies both on the tetralones and on the isochromanones, we are
able to estimate biological activity, such as apoptotic or anti-
proliferative effect, for any other new library member. After
the determination of this optimal lipophilicity the most
promising member was studied in further biological assays
to get a more complete characterization of its antiprolifera-
tive effect.
Here we also found good parabolic correlation (with a
regression coefficient (R) of 0.8343) between the IC50 and k
value for 15 out of 17 molecules excluding two outliers (2E,
1
2E) found by the “leave one out” method of Allen et al.
83] (Fig. (5)), (Table 3). This lipophilicity range has the
optimal k value of 2.1.
[
Light Microscopic Morphological Studies
It is well known that A431 epidermal carcinoma cell line
is a fast-growing and a well-characterized cell target for sig-
nal transduction studies. Besides cell proliferation assays it is
also applicable for other biological observations, as well.
Therefore, morphological evaluation was continued in A431
cells. In the further biological investigations the focus was
on the most effective compound 8E with IC50 of 0.3 μM and
with the optimal k value of 2.0. Considering the very low
IC50 value of compound 8E to show significant differences
when compared to control, in further experiments at least 1
μM concentration of compound 8E was applied. As shown
in Fig. (7), even the simplest phase-contrast light micro-
scopic study could clearly show that the minimal concentra-
tion (1 μM) of compound 8E was perfectly enough to induce
robust cell (shape) morphology changes –looks like apopto-
sis- of A431 cells in culture after 24 hours comparing to the
normal-growing control cells.
*
ꢁ
where ꢀ symbols belong to the E-isomers, + to the Z-isomers and,
symbols show the two outliers.
Fig. (5). Antiproliferative activity (IC50) vs. measured lipophilicity
(k) of the whole molecular library. Abscissa: k, ordinate: IC50
(μmol/l).
Table 3. Correlation Parameters of the IC50 – k Relationship
*
*
Parameters of the linear
IC50 vs. k without outliers
IC50 vs. k by the E-isomers without outliers
A
268.01564
-249.72270
59.53753
0.69606
210.37100
-194.23249
45.19956
0.63287
20.17173
0.04948
9
B
B
R
1
2
2
SD
P
20.90323
-
4
7.88376*10
15
N
2
*
A, B
A, B
1
, B
2
, R , SD, P and N are the parameters of the polynomial.
2
1
1
, B : coefficient and the standard error for each order, R : square of the correlation coefficient, SD: standard deviation of the fit, P: probability (that R is zero), N: number of
data points.
*R.S.D. of the IC50 values were less, than 10 %, three replicates has been made for each of the compounds.
*

3
28 Current Medicinal Chemistry, 2010 Vol. 17, No. 4
Huszár et al.
*
where ꢀ symbols show the two outliers.
Fig. (6). Antiproliferative activity (IC50) vs. measured lipophilicity (k) of the E-isomers. Abscissa: k, ordinate: IC50 (μmol/l).
Fig. (7). Light microscopical study using A431 cells after compound 8E treatment in different concentrations.

Lipophilicity and Anti-Tumour Activity of Isochromanone Derivatives
Current Medicinal Chemistry, 2010 Vol. 17, No. 4
329
Fig. (8). Immunoblotting analysis of PARP fragmentation using the molecule 8E. The results shown are representative of three independent
experiments.
The majority of compound 8E treated cells were rounded
up, shrunk and retracted from their neighbouring cells, and
then eventually floated into the media which might be al-
ready an indicative of cell death, most probably apoptosis as
well.
PARP is a protein that detects DNA strand breaks and
functions in their repair. Once PARP is cleaved, this repair
function does not exist longer. Although PARP is not essen-
tial for apoptosis, the cleavage of PARP may contribute to
the commitment to apoptosis [84]. To investigate the effect
of compound 8E on the cleavage of PARP, SDS polyacry-
lamide gel electrophoresis was applied using 8 % gel and
specific antibody staining against the cleaved, approximately
These data (pictures) - in agreement with the antiprolif-
erative experiments - suggest that apoptosis induction might
be a major the mechanism of 8E-caused death of epidermal
carcinoma cells. Using other isochromanones, weaker or no
cell death induction has been observed compared to control
in A431 cells.
8
5 kD main fragment of PARP. Treating A431 cells with
solvent (control), 1 μM, 5 μM and 10μM doses of compound
8E for 24h, the appearance of the main PARP fragment,
which is a hallmark of apoptosis, was clearly detected at the
lowest concentration (1μM) we applied (Fig. (8)).
Immunoblotting and Flow Cytometric Studies Investigat-
ing the 8E Compound Induced Cell Death
Besides, as a further proof of apoptosis, the ratio of apop-
totic cells were assessed and quantified using FACS analysis,
where the apoptotic fraction was measured applying flow
cytometric analysis with the PI staining (Fig. (9)).
To determine whether the 8E-induced loss of the prolif-
eration capacity and cell viability of the epidermal carcinoma
cancer cells was associated with the induction of apoptosis,
A431 cells were treated with 8E for further apoptosis assays.
To examine whether cells with shrunken abnormal pheno-
type are apoptotic, immunoblot staining of PARP and FACS
analysis (both known as appropriate methods to check cer-
tain criteria of apoptosis) has been performed on cells.
Another event associated with the process of apoptosis is
DNA fragmentation, as a result of endonuclease activation.
During the fragmentation, DNA molecules become smaller
(
fragmented) with the appearance of more free DNA ends in
Fig. (9). The results of the flow cytometric analysis of the compounds 1E, 5E and the 8E (10 μM). The columns represent the percentage of
the apoptotic fraction.

3
30 Current Medicinal Chemistry, 2010 Vol. 17, No. 4
Huszár et al.
the cell. PI can be used to detect the later phase of apoptotic
cell death (data not shown, ref.: [79]).
the chromatographic method developed here was able to
distinguish the E and Z members of the isomer pairs having
the same calculated lipophilicity but different antiprolifera-
tive activity. On the basis of that we plotted the inhibitory
activity (IC50) rather against the more informative HPLC
determined lipophilicity (k) values to the computer calcu-
lated ones (CLOGP).
Since all our earlier measurements suggested that there
was no need to apply higher dose than 1 μM concentration of
compound 8E for a robust effect, the same amount was used
accompanying 8E with two more isomers of different behav-
iour, 1E and 5E showing the big variety in biological effect
of isochromanone family. Compound 1E is an almost inac-
tive isomer according to the microscopic studies – just like a
negative control, and 5E is a significantly active isomer but
still far behind the most active 8E isomer. Considering the
expectation based on the IC50 and other data of 8E, it was not
surprising to observe a very significant percentage of apop-
totic fraction, namely approximately 68 % after 24 hour
treatment using either 1 or 10 μM 8E. In this cellular system
the spontaneous apoptosis was around 5 % (data not shown).
As shown in Fig. (9), no significant apoptosis was observed
after treatment with 1E, even the more effective isomer 5E
could show much less increase of DNA fragmentation com-
paring to the effect of highly effective 8E, in full agreement
with our earlier results.
In the current paper we also summarized the most impor-
tant literature data about the impact of hydrophobic pocket
on biological activity [30-37] or about the optimalisation of
lipophilicity value and range connected to biological effec-
tiveness [38-43, 45]. We collected data how this optimal
range could be defined by parabolic correlation where this
range was determined between lipophilicity and some other,
very useful drug developmental parameter such as drug
transport through cell membrane [49] or brain uptake [55].
Following our previous observations by other molecular li-
braries [67-68] we succeeded to determine parabolic correla-
tion between lipophilicity (k) and cytotoxic activity (IC50).
The molecular family investigated here interferes with
cell proliferation mostly by apoptosis. Magnitude and type of
the effects depended on the nature of the substituents of the
Both assay types with completely different approaches
support our idea that 8E is an outstanding apoptosis inducer
even at low doses via PARP and DNA fragmentation. These
results suggest 8E may have potential anticancer effects
against epidermal carcinoma cells by inducing strong apop-
tosis.
core structure. Studying the antiproliferative activity (IC50
)
and experimental lipophilicity (k) the defined parabolic
correlation enabled us to determine an optimal lipophilicity
value (k = 2.1) and to predict a range where the biological
activity is the most effective. Also the biological activity can
be compared now not only to the k value but to the CLOGP
value as well, as good linear correlation was found between
the experimentally determined (logk) and the computer cal-
culated (CLOGP) lipophilicity parameters. With the deter-
mination of this criterion pre-selection of the biologically
effective molecules can be made within molecular libraries
containing chemically similar core structure as the isochro-
manones.
DISCUSSION
Since the elucidation of the critical role for the lipophilic-
ity parameter in drug development, it has been established
that prescreening for this value is essential early step in drug
development. In the present paper we provided an overview
of the importance of lipophilicity of drug candidates both for
biological action related efficacy and for ADME(T) and de-
scribed a novel approach for drug-likeness characterization
of a molecular library using correlation study between lipo-
philicity and biological activity. Our research has led to the
conclusion that a better knowledge of the relationship be-
tween lipophilicity and the compounds biological effect
could provide an easy but powerful screening platform to
select drug candidate molecules without expensive biological
studies.
Taking into account the determined optimal lipophilicity
range, the most promising members of the molecular library
has been selected. As generally the activity of the E isomers
was greater than that of the Z-ones, these molecules also
belonged to the E-isomers. The E-isomer of the 2’-methoxy-
phenil (5E), 2’-chloro-phenil (6E) and 3’-hydroxo-phenil
(
2
12E) derivatives showed good, while the molecule included
’,4’-dimethoxy-phenil substituent (8E) showed outstanding
antiproliferative activity. Compound 8E proved to be promi-
nently efficient in the proliferation and apoptosis assays,
showed a very robust and outstandingly efficient antiprolif-
erative effect in A431 cell culture model system with lower
than 1 μM IC50 value. The mechanism of action of 8E
proved to be apoptosis induction, supported by the morphol-
ogy and other more specific biological studies, adding extra
value to its antiproliferative effect since besides the blocking
of the tumour cell proliferation, it can eliminate cancer cells
with false cell proliferation signalling, too. Based on its
unique structure isomer 8E has very promising anti-tumour
characteristics for developing pre-clinical lead molecules.
In line with other recent publications where other differ-
ent types of molecular libraries have been investigated [67-
6
8, 10-12, 27-29], our study provides additional evidence on
the clear relationship between biological activity and lipo-
philicity characterizing an isochromanone molecular library
for logP, k and cell growth on A431 human tumour cells.
An applicable RP-HPLC method for fast analysis and
characterization of lipophilicity of the members 4-
(
arylmethylene)-3-isochromanones was also developed.
Separation of 17 4-(arylmethylene)-3-isochromanones pos-
sessing similar chemical structure could be obtained.
In summary, we found that analysing the interrelation-
ship between the optimal lipophilicity range and biological
effect would be crucial for further drug development in a
cost effective but still efficient way. Determining and apply-
ing this optimal lipophilicity range, using isochromanone
The chromatographic method developed here proved to
be able to distinguish between the small lipophilicity differ-
ences of the members of the library. While the prediction
gave the same calculated lipophilicity value for the E/Z pairs,

Lipophilicity and Anti-Tumour Activity of Isochromanone Derivatives
Current Medicinal Chemistry, 2010 Vol. 17, No. 4
331
molecular library, we could successfully develop a novel
potential isochromanone drug candidate (8E) with high anti-
tumour effect as a proof of concept considering our previous
results in this experimental system.
[7]
Alonso, M. M.; Asumendi, A.; Villar, J.; Gil, M. J.; Martínez-
Merino, V.; Encío, I.; Migliaccio, M. New benzo(b)thiophen-
esulphonamide 1,1-dioxide derivatives induce a reactive oxygen
species-mediated process of apoptosis in tumour cells. Oncogene,
2
003, 22, 3759-3769.
[
[
[
8]
Villar, R.; Encio, I.; Migliaccio, M.; Gila, M. J.; Martinez-Merino,
V. Synthesis and cytotoxic activity of lipophilic sulphonamide
derivatives of the benzo[b]thiophene 1,1-dioxide. Bioorgan. Med.
Chem., 2004, 12, 963-968.
Encío, I.; Morré, D.J.; Villar, R.; Gil, M. J.; Martínez-Merino, V.
Benzo[b]thiophenesulphonamide 1,1-dioxide derivatives inhibit
tNOX activity in a redox state-dependent manner. Br. J. Cancer,
ACKNOWLEDGEMENT
This work was supported by the following grants: OTKA
9]
6
4
0197, OTKA 49478, Stemkill OM00106/2008, ETT
48/2006 and CELLCOM-RET.
2
005, 92, 690 - 695.
10]
Yu, Y.; Lou, L; Liu, W.; Zhu, H.; Ye, Q.; Chen, X.; Gao W.; Hou
S. Synthesis and anticancer activity of lipophilic platinum(II) com-
plexes of 3,5-diisopropylsalicylate. Eur. J. Med. Chem., 2008, 43,
1438-1443.
Woerdenbag, H.; Moskal, T.; Pras, N.; Malingré, T. Cytotoxicity of
artemisinin-related endoperoxides to Ehrlich ascites tumor cells. J.
Nat. Prod., 1993, 56, 849-856.
John, G.; Shrivastava, R.; Chevalier, A.; Pognat J.; Massingham R.
An in vitro investigation of the relationships between potency,
lipophilicity, cytotoxicity and chemical class of representative
calcium antagonist drugs. Pharmacol. Res., 1993, 27, 253-262.
Xue, L.; Bajorath, J. Molecular descriptors in chemoinformatics,
computational combinatorial chemistry, and virtual screening.
Comb. Chem. High T. Scr., 2000, 3, 363.
ABBREVIATIONS:
ADMET
=
Absorption, Distribution, Metabolism, Ex-
cretion, and Toxicity
[
11]
12]
QSAR
PARP
=
=
=
Quantitative structure-activity relationship
Poly(ADP-ribose) polymerase
[
P
o/w
n-octanol/water (biphasic) partition coeffi-
cient
[
13]
14]
RP-HPLC
=
Reversed phase-high performance liquid
chromatography
[
Braumann, Th.; Weber, G.; Grimme, L. H. Quantitative structure—
activity relationships for herbicides: Reversed-phase liquid
MEKC
MLP
=
=
Micellar electrokinetic chromatography
Molecular lipophilicity potential
chromatographic retention parameter, log k
w
, versus liquid-liquid
partition coefficient as model of the hydrophobicity of
a
phenylureas, s-triazines and phenoxycarbonic acid derivatives. J.
Chromatogr., 1983, 261, 329-343.
Lambert, W. J. Modeling oil—water partitioning and membrane
permeation using reversed-phase chromatography. J. Chromatogr.,
MIC values = Minimum inhibitory concentration
[
15]
16]
CLOGP
k
=
=
=
=
=
=
=
=
=
=
Calculated logP
1
993, 656, 469-484.
Retention coefficient
acetonitrile
[
Valkó, K.; My Du, Ch.; Bevan, C.; Reynolds D. P; Abraham, M.
H. Rapid Method for the Estimation of Octanol/Water Partition
Coefficient (Log Poct) from Gradient RP-HPLC Retention and a
ACN
PI
Propidium iodide
Hydrogen Bond Acidity Term (ꢀꢁ
2
H). Curr. Med. Chem., 2001, 8,
1
137-1146.
DMEM
FCS
Dulbecco’s Mod Eagle Medium
foetal calf serum
[
17]
18]
Escuder-Gilabert, L.; Martinez-Pla, J. J.; Sagrado, S.; Villanueva-
Camanas, R. M.; Medina-Hernández, M. J. Biopartitioning micellar
separation methods: modelling drug absorption. J. Chromatogr. B,
2003, 797, 21-35.
Vallat, P.; Fan, W.; Tayar, N. El.; Carrupt, P. A.; Testa, B.
Solvatochromic analysis of the retention mechanism of two novel
stationary phases used for measuring lipophilicity by RP-HPLC. J.
Liquid Chromatogr., 1992, 15, 2133.
MB
Methylene blue
[
EGFR
Epidermal growth factor receptor
Retention time
t
R
FACS
Fluorescence-activated cell-sorting
[19]
Hansch, C.; Leo, A. Substituent Constants for Correlation Analysis
in Chemistry and Biology. Wiley: New York, 1979.
[
20]
21]
Takács-Novák, K. Computerized logP prediction using fragment
methods. Acta Pharm. Hung., 1998, 68, 39-48. (in Hungarian)
Djakovic-Sekulic, T.; Acanski, M.; Perisic-Janjic, N. Evaluation of
the predictive power of calculation procedure for molecular
hydrophobicity of some estradiol derivates. J. Chromatogr. B,
2002, 766, 67-75.
REFERENCES
[
[1]
Hansch, C.; Leo, A. Exploring QSAR Fundamentals and Applica-
tions in Chemistry and Biology, American Chemical Society:
Washington, 1995.
[
2]
Terrett, N. K.; Gardner, M.; Gordon, D. W.; Kobylecki, R. J.;
Steele, J. Combinatorial synthesis — the design of compound li-
braries and their application to drug discovery. Tetrahedron, 1995,
[22]
[23]
Carrupt, P. A.; Testa, B.; Gaillard, P. In Computational
Approaches to Lipophilicity: Methods and Applications; K. B.
Lipkowitz, B. D. Boyd, Eds.; Rev. Computat. Chem. New York,
1997; Vol. 11, pp. 241-315.
Albert, A. Relations between molecular structure and biological
activity: stages in the evolution of current concepts. Annu. Rev.
Pharmacol., 1971, 11, 13-36.
Burgen, A. S. Conformational changes and drug action. Fed. Proc.,
1981, 40, 2723-2728.
Chohan, K. K.; Paine, S. W.; Waters, N. J. Quantitative structure
activity relationships in drug metabolism. Curr. Top. Med. Chem.,
2006, 6, 1569-1578.
5
1, 8135-8173.
[3]
Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J. Ex-
perimental and computational approaches to estimate solubility and
permeability in drug discovery and development settings. Adv.
Drug Deliv. Rev., 1997, 23, 3-25.
Rishton, G. M. Nonleadlikeness and leadlikeness in biochemical
screening. Drug Discov. Today, 2003, 8, 86-96.
Van de Waterbeemd, H.; Smith, D. A.; Beaumont, K.; Walker D.
K. Property-based design: optimization of drug absorption and
pharmacokinetics. J. Med. Chem., 2001, 44, 1313-1333.
Valkó, K. Application of high-performance liquid chromatography
based measurements of lipophilicity to model biological distribu-
tion. J. Chromatogr. A, 2004, 1037, 299-310.
[24]
[25]
[
4]
5]
[
[26]
Barrett, I.; Meegan, M. J.; Hughes, R. B.; Carr, M.; Knox, A. J. S.;
Artemenko, N.; Golfis, G.; Zisterer, D. M.; Lloyd, D. G. Synthesis,
biological evaluation, structural-activity relationship, and docking
[6]
study for
a series of benzoxepin-derived estrogen receptor
modulators. Bioorg. Med. Chem., 2008, 16, 9554-9573.

3
32 Current Medicinal Chemistry, 2010 Vol. 17, No. 4
Huszár et al.
[
27]
Verma, R. P.; Hansch, C. Elucidation of structure-activity
relationships for 2- or 6-substituted-5,8-dimethoxy-1,4-
naphthoquinones. Bioorg. Med. Chem., 2004, 12, 5997-6009.
Matysiak, J. Evaluation of electronic, lipophilic and membrane
affinity effects on antiproliferative activity of 5-substituted-2-(2,4-
dihydroxyphenyl)- 1,3,4-thiadiazoles against various human cancer
cells. Eur. J. Med. Chem., 2007, 42, 940-947.
Campos, J.; Nunez, M.C.; Conejo-Garcia, A.; Sanchez-Martin,
R.M.; Hernandez-Alcoceba, R.; Rodriguez-Gonzalez, A.; Lacal,
J.C.; Gallo, M.A.; Espinosa, A. QSAR-derived choline kinase
inhibitors: how rational can antiproliferative drug design be? Curr.
Med. Chem., 2003, 10, 1095-1112.
[44]
Coats, E. A.; Milstein, S. R.; Pleiss, M. A.; Roesener J. A.
Comparative analysis of the cytotoxicity of substituted
[phenylglyoxal bis(4-methyl-3-thiosemicarbazone)]copper(II) chel-
ates. 2. Parabolic correlations and their implications for selective
toxicity. J. Med. Chem., 1978, 21, 804-809.
Waring, M. J. Defining optimum lipophilicity and molecular
weight ranges for drug candidates—Molecular weight dependent
lower logD limits based on permeability. Bioorg. Med. Chem. Lett.,
2009, 19, 2844-2851.
[28]
[45]
[46]
[29]
Takahashi, K.; Kunishiro, K.; Kasai, M.; Miike, T.; Kurahashi, K.;
Shirahase, H. Relationships between lipophilicity and biological
activities in
a series of indoline-based anti-oxidative acyl-
[30]
[31]
[32]
Toth, I.; Christodoulou, M.;Bankowsky, K.; Flinn, N.; Gibbons, W.
A.; Godeau, G.; Moczar, E.; Hornebeck, W. Design of potent
lipophilic-peptide inhibitors of human neutrophil elastase: In vitro
and in vivo studies. Int. J. Pharm., 1995, 125, 117-122.
Hornebeck, W.; Moczar, E.; Szecsi, J.; Robert, L. Fatty acid
peptide derivatives as model compounds to protect elastin against
degradation by elastases. Biochem. Pharmacol., 1985, 34, 3315-
CoA:cholesterol acyltransferase (ACAT) inhibitors. Arzneimit-
telforschung, 2008, 58(12), 666-72.
Pop, E.; Oniciu, D. C.; Pape, M. E.; Cramer, C. T.; Dasseux H. J.-
L. Lipophilicity parameters and biological activity in a series of
compounds with potential cardiovascular applications. Croat.
Chem. Acta, 2004, 77, 301-306.
[47]
[48]
[49]
Fahr, A.; van Hoogevest ꢀ, P.; Kuntsche, J.; Leigh L. S. M.ꢀ:
3
321.
Lipophilic drug transfer between liposomal and biological
membranes: what does it mean for parenteral and oral drug
delivery? J. Liposome Res., 2006, 16, 281-301.
Veldman, R. J.; Zerp, S.; Van Blitterswijk, W. J.; Verheij, M. N-
hexanoyl-sphingomyelin potentiates in vitro doxorubicin
cytotoxicity by enhancing its cellular influx. Br. J. Cancer, 2004,
90, 917 - 925.
Lafuma, C.; Frisdal, E.; Robert, L.; Moczar, E.; Lefrancier, P.;
Hornebeck, W. Lipopeptides as biofunctional inhibitors of elastase-
induced emphysema in mice by intratracheal pretreatment with
oleoyl-alanyl-alanyl-prolylvaline. In Aubry, A.; Marraud, M.;
Vitoux, B., Eds; Second Forum on Peptides, Colloque INSERM /
John Libbey Eurotext., 1989; Vol. 184, pp. 321-324.
[
33]
34]
Boduier, C.; Godeau, G.; Hornebeck, W.; Robert, L.; Bieth, J. G.
the elastolytic activity of cathepsin G: An ex vivo study with
dermal elastin. Am. J. Respir. Cell Mol. Biol., 1991, 4, 497-503.
Quartara, L.; Fabbri, G.; Ricci, R.; Patacchini, R.; Pestellini, V.;
Maggi,C. A.; Pavone, V.; Giachetti, A.; Arcamone, F. Influence of
lipophilicity on the biological activity of cyclic pseudopeptide NK-
[50]
[51]
Levin, VA. Relationship of octanol/water partition coefficient and
molecular weight to rat brain capillary permeability. J. Med.
Chem., 1980, 23, 682-684.
Atkinson, F.; Cole, S.; Green, C.; van de Waterbeemd, H.
Lipophilicity and other parameters affecting brain penetration.
Curr. Med. Chem.-Central Nervous System Agents, 2002, 2, 229-
240.
[
2
receptor antagonists. J. Med. Chem., 1994, 37, 3630-3638.
[
35]
Regan, J.; Capolino, A.; Cirillo, P. F.; Gilmore, T.; Graham, A. G.;
Hickey, E.; Kroe, R. K.; Madwed, J.; Moriak, M.; Nelson, R.;
Pargellis, C. A.; Swinamer, A.; Torcellini, C.; Tsang, M.; Moss, N.
Structure-activity relationships of the p38r MAP kinase inhibitor 1-
[52]
Ballet, S.; Misicka, A.; Kosson, P.; Lemieux, C.; Chung, N;
Schiller, P.W.; Lipkowski, A. W.; Tourwé, D. Blood-brain barrier
penetration by two dermorphin tetrapeptide analogues: role of
lipophilicity vs structural flexibility. J. Med. Chem., 2008, 51,
2571-2574.
De Lange, E. C.- M.; de Boer, B. A. G.; Breimer, D. D.
Microdialysis for pharmacokinetic analysis of drug transport to the
brain. Adv. Drug Deliv. Rev., 1999, 36, 211-227.
Gimenez, F.; Fernandez, C.; Mabondzo, A. Transport of HIV
protease inhibitors through the blood-brain barrier and interactions
with the efflux proteins, p-glycoprotein and multidrug resistance
proteins. J. Acq. Immun. Def. Synd., 2004, 36, 649-658.
Timmermans, P. B. M. W. M.; Brands, A.; van Zwieten, P.A.
Lipophilicity and brain disposition of clonidine and structurally
related imidazolidines. N-S. Arch. Pharmacol., 1977, 300, 217-226.
Dimmock, J. R.; Wong, M. L. C. Bioactivities and potential uses in
drug design of acyclic ꢀ ,ꢁ -unsaturated ketones. Can. J. Pharm.
Sci., 1976, 11, 35-53.
Geiger, H. In The Flavanoids; Harborne J. B, Ed.; Chapman & Hill
Ltd: London, 1988; pp. 389-397.
Ammar, M. S.; Gerber, N. N.; Mc Daniel, L. E. New antibiotic
pigments related to fusarubin from fusarium solani (MART.) sacc.
J. Antibiot., 1979, 32, 679-684.
Ravisé; Kirkiacharian, B. S. Effect of the structure of phenolic
compounds on the growth inhibition of Phytophthora parasitica and
the activity of parasitogenic enzymes. III. Homo-isoflavanones.
Phytopath. Zeitschrift, 1978, 92(1), 36-50.
Kern, H. The naphthazarins of fusarium. Ann. Phytopathol., 1978,
10, 327-345.
(
5-tert-Butyl-2-p-tolyl-2H-pyrazol-3-yl)-3-[4-(2-morpholin-4-yl-
ethoxy)naphthalen-1-yl]urea (BIRB 796). J. Med. Chem., 2003, 46,
676-4686.
[53]
[54]
4
[
36]
37]
Regan, J.; Breitfelder, S.; Cirillo, P.; Gilmore, T.; Graham, G. A.;
Hickey, E.; Klaus, B.; Madwed, J.; Moriak, M.; Moss, N.;
Pargellis, C.; Pav, S.; Proto, A.; Swinamer, A.; Tong, L.;
Torcellini, C. Pyrazole urea-based inhibitors of p38 MAP kinase:
from lead compound to clinical candidate. J. Med. Chem., 2002,
4
5, 2994-3008.
[55]
[56]
[
Kim, C. U.; Lew, W.; Williams, M. A.; Liu, H.; Zhang, L.;
Swaminathan, S.; Bischofberger, N.; Chen, M. S.; Mendel, D. B.;
Tai, C. Y.; Laver, W. G.; Stevens, R. C. Influenza neuraminidase
inhibitors possessing
a novel hydrophobic interaction in the
enzyme active site: design, synthesis, and structural analysis of
carbocyclic sialic acid analogues with potent anti-influenza
activity. J. Am. Chem. Soc., 1997, 119(4), 681-690.
[57]
[58]
[38]
Wenlock, M. C.; Austin, R. P.; Barton, P.; Davis, A. M.; Leeson, P.
D.
development and marketed oral drugs. J. Med. Chem., 2003, 46,
250-1256.
A comparison of physiochemical property profiles of
1
[59]
[
39]
40]
Leeson, P. D.; Springthorpe, B. The influence of drug-like concepts
on decision-making in medicinal chemistry. Nat. Rev. Drug Disov.,
2
007, 6, 881-890.
[
Hughes, J. D.; Blagg, J.; Price, D. A.; Bailey, S.; DeCrescenzo, G.
A.; Devraj, R. V.; Ellsworth, E.; Fobian, Y. M.; Gibbs, M. E.;
Gilles, R. W.; Greene, N.; Huang, E.; Krieger-Burke, T.; Loesel, J.;
Wager, T.; Whiteley, L.; Zhang, Y. Physiochemical drug properties
associated with in vivo toxicological outcomes. Bioorg. Med.
Chem. Lett., 2008, 18, 4872-4875.
[60]
[61]
Casellas, M.; Grifoll, M.; Bayona, J. M.; Solanas, A. M. New
metabolites in the degradation of fluorene by Arthrobacter sp.
strain F101. Appl. Environ. Microbiol., 1997, 63(3), 819-826.
Al-Nakib, T. M.; Bezjak, V.; Meegan, M. J.; Chandy, R. Synthesis
and antifungal activity of some 3-benzylidenechroman-4-ones, 3-
benzylidenethiochroman-4-ones and 2-benzylidene-1-tetralones.
Eur. J. Med. Chem., 1990, 25, 455-462.
Lóránd, T.; Al-Nakib, T. M.; Prókai, L. Synthesis, Antimycotic
Activity and QSAR Studies of Homoisoflavanone Analogues, 212th
National Meeting of the American Chemical Society: Orlando, FL,
USA, 1996.
[62]
[63]
[41]
Waring, M. J.; Johnstone, C. A quantitative assessment of hERG
liability as a function of lipophilicity. Bioorg. Med. Chem. Lett.,
2
007, 17, 1759-1764.
[
42]
43]
Gleeson, M. P. J. Generation of a set of simple, interpretable
ADMET rules of thumb. J. Med. Chem., 2008, 51, 817-834.
Waterhouse, R. N. Determination of lipophilicity and its use as a
predictor of blood-brain barrier penetration of molecular imaging
agents. Mol. Imaging Biol., 2003, 5(6), 376-389.
[

Lipophilicity and Anti-Tumour Activity of Isochromanone Derivatives
Current Medicinal Chemistry, 2010 Vol. 17, No. 4
333
[
64]
AL-Nakib, T. M.; Lóránd, T.; Földesi, A.; Varghese, R. The in
vitro antimycotic activity and acute toxicity of third-generation
benzylidenetetralones and heteroarylidenetetralones. Med. Prin.
Pract., 2001, 10, 191-196.
Lampronti, I.; Martello, D.; Bianchi, N.; Borgatti, M.; Lambertini,
E.; Piva, R.; Jabbar, S.; Shahabuddin Kabir Choudhuri, M.; Tareq
Hassan Khan, M.; Gambari, R. In vitro antiproliferative effects on
human tumor cell lines of extracts from the Bangladeshi
medicinalplant Aeglemarmelos Correa. Phytomedicine, 2003,
with pellicular and totally porous sorbents. J. Chromatogr. A, 1996,
728/1-2, 225-233.
[74]
[75]
Opperhuizer, A.; Sinnige, T. L.; van der Steen, J. M. D.; Hutzinger,
O. Differences between retentions of various classes of aromatic
hydrocarbons in reversed-phase high-performance liquid
chromatography: Implications of using retention data for
characterizing hydrophobicity. J. Chromatogr. B, 1987, 388, 51-64.
Melander, W. R.; Chen, B.-K.; Horváth, Cs. mobile phase effects
in reversed-phase chromatography: VII. Dependence of retention
on mobile phase composition and column temperature. J.
Chromatogr. A, 1985, 318, 1-10.
Snyder, L. R.; Kirkland, J. J. Introduction to Modern Liquid
Chromatography, John Wiley and Sons: New York, 1974.
Poole, C. F. The essence of chromatography. Elsevier: Amsterdam,
2003.
[
65]
1
0(4), 300-08.
Serenex, Inc., Kenneth, H.; Philip, H.; Thomas, B.; James, V.
Dihidropyridazine, Tetrahydropyridine, Chromanone and
[
66]
67]
[76]
[77]
[78]
Dihydronaphthalenone Derivatives as Heat-Shock Protein 90
Inhibitors. 2008
Hallgas, B.; Dobos, Zs.; ꢀsz, E.; Hollósy, F.; Schwab, R. E.;
Szabó, E. Z.; Erꢁs, D.; Idei, M.; Kéri G.; Lóránd, T.
Characterization of lipophilicity and antiproliferative activity of E-
[
Oliver, M. H.; Harrison, N. K.; Bishop, J. E.; Cole, P. J.; Laurent,
G. J. A rapid and convenient assay for counting cells cultured in
microwell plates: application for assessment of growth factors. J.
Cell. Sci., 1989, 92(3), 513-518.
2
-arylmethylene-1-tetralones and their heteroanalogues. J.
Chromatogr. B, 2005, 819(2), 283-91.
[
68]
Hallgas, B.; Dobos, Zs.; Agócs, A.; Idei, M.; Kéri, G.; Loránd, T.;
Mészáros, G. Lipophilicity and antiproliferative activity profiling
of 2-benzylidencycloalkanones. J. Chromatogr. B, 2007, 856, 148-
[79]
[80]
Herrmann, M.; Lorenz, H. M.; Vollm, R.; Grꢂnkem, M.; Woithm,
W.; Kaldenm, J. R. A rapid and simple method for the isolation of
apoptotic DNA fragments. Nucleic Acids Res., 1994, 22, 5506-
5507.
Hallgas, B.; Patonay, T.; Kiss-Szikszai, A.; Dobos, Zs.; Hollósy,
F.; Kéri, G.; Erꢁs, D.; ꢀrfi, L.; Idei, M. Comparison of measured
and calculated lipophilicity of substituted aurones and related
compounds. J. Chromatogr. B, 2004, 801, 229-235.
Hollósy, F.; Seprꢁdi, J.; ꢀrfi, L.; Erꢁs, D.; Kéri G.; Idei. M. Eva-
luation of lipophilicity and antitumour activity of parallel carboxa-
mide libraries. J. Chromatogr. B, 2002, 780, 355-363.
Fish, P. V.; Ryckmans, T.; Stobie, A.; Wakenhut, F. [4-
(Phenoxy)pyridin-3-yl]methylamines: A new class of selective
noradrenaline reuptake inhibitors. Bioorg. Med. Chem. Lett., 2008,
18, 1795-1798.
1
55.
[
69]
70]
Keresztury, G.; Holly, S.; Sundius, T.; Loránd, T. Analysis of
vibrational spectra of some new E- and Z-4-arylidene-3-
isochromanones. Vib. Spectrosc., 2002, 29, 53-59.
Lóránd, T.; Forgó, P.; Földesi, A.; ꢀsz, E.; Prókai, L. Improved
solvent-free synthesis and structure elucidation of (E)- and (Z)-4-
[
[81]
[82]
(
2
arylmethylene)-3-isochromanones. Eur. J. Org. Chem., 2002, 17,
996-3003.
[71]
Huszár, M.; Hallgas, B.; Idei, M.; Kiss-Szikszai, A.; Horváth, A.;
Patonay, T. Lipophilicity of substituted aurones and related
compounds measured on immobilized artificial membrane (IAM)
and conventional C8 (MOS) columns. J. Liq. Chromatogr., 2008,
3
1, 3143-4158.
[83]
[84]
Allen, D. M. The relationship between variable selection and data
augmentation and a method for prediction. Technometrics, 1974,
16, 125-127.
Seamus, J.; Martin and Green, D. R. Protease activation during
apoptosis: death by a thousand cuts? Cell, 1995, 82, 349-352.
[
72]
73]
Verzola, B.; Gelpi, C.; Righetti, P. G. Quantitative studies on the
adsorption of proteins to the bare silica wall in capillary
electrophoresis: II. effects of adsorbed, neutral polymers on
quenching the interaction. J. Chromatogr. A, 2000, 874, 293-303.
Corradini, D.; Kalghatgi, K.; Horváth, C.S. Effect of mobile phase
additives on peptide retention in reversed-phase chromatography
[
Received: October 14, 2009
Revised: January 18, 2010
Accepted: January 19, 2010