Cytotoxic 1,4-bis(2-oxo-1-cycloalkylmethylene)benzenes and related compounds

Article abstract of DOI:10.1016/S0223-5234(01)01294-6

A series of 1,4-bis(2-oxo-1-cycloalkylmethylene)benzenes 2a-c and 4 and a related acyclic analogue 6a were synthesised and converted to the corresponding Mannich bases 3a-c, 5 and 6b. Evaluation of these compounds against murine P388 and L1210 cells as well as human Molt 4/C8 and CEM T-lymphocytes revealed that the Mannich bases were more cytotoxic than the corresponding unsaturated ketones. 1,4-bis(3-Dimethylaminomethyl-2-oxo-1-cyclohexylmethylene)benzene dihydrochloride (3a) had lower IC₅₀ values than melphalan against the four cell lines and was 15 times more potent than this drug when examined against a panel of human tumours.

Full text of DOI:10.1016/S0223-5234(01)01294-6

Eur. J. Med. Chem. 37 (2002) 35–44
Original article
Cytotoxic 1,4-bis(2-oxo-1-cycloalkylmethylene)benzenes and related
compounds
Jonathan R. Dimmock a,*, Amitabh Jha a, Praveen Kumar a, Gordon A. Zello a,
J. Wilson Quail b, Eliud O. Oloo b, Jennifer J. Oucharek b, Mohammed K. Pasha c,
Dallas Seitz c, Rajendra K. Sharma c, Theresa M. Allen d, Cheryl L. Santos d,
Elias K. Manavathu e, Erik De Clercq f, Jan Balzarini f, James P. Stables g
a College of Pharmacy and Nutrition, Uni6ersity of Saskatchewan, 110 Science Place, Saskatoon, Saskatchewan, Canada S7N 5C9
b Department of Chemistry, Uni6ersity of Saskatchewan, Saskatoon, Saskatchewan, Canada S7N 5C9
c Department of Pathology, College of Medicine and Cancer Research Unit, Health Research Di6ision, Saskatchewan Cancer Agency,
Uni6ersity of Saskatchewan, Saskatoon, Saskatchewan, Canada S7N 4H4
d Department of Pharmacology, Uni6ersity of Alberta, Edmonton, Alberta, Canada T6G 2H7
e Department of Medicine, Wayne State Uni6ersity, Detroit, MI 48201-1908, USA
f Rega Institute for Medical Research, Katholieke Uni6ersiteit Leu6en, B-3000 Leu6en, Belgium
g National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, MD 20892-9020, USA
Received 20 July 2001; received in revised form 12 November 2001; accepted 14 November 2001
Abstract
A series of 1,4-bis(2-oxo-1-cycloalkylmethylene)benzenes 2a–c and 4 and a related acyclic analogue 6a were synthesised and
converted to the corresponding Mannich bases 3a–c, 5 and 6b. Evaluation of these compounds against murine P388 and L1210
cells as well as human Molt 4/C8 and CEM T-lymphocytes revealed that the Mannich bases were more cytotoxic than the
corresponding unsaturated ketones. 1,4-bis(3-Dimethylaminomethyl-2-oxo-1-cyclohexylmethylene)benzene dihydrochloride (3a)
had lower IC50 values than melphalan against the four cell lines and was 15 times more potent than this drug when examined
´
against a panel of human tumours. © 2002 Editions scientifiques et me´dicales Elsevier SAS. All rights reserved.
Keywords: cytotoxicity; a,b-unsaturated ketones; Mannich bases; N-myristoyltransferase; X-ray crystallography
1. Introduction
ketones was proposed on the basis of their more effec-
tive thiol alkylating properties compared to the precur-
sor ketones [2]. Second, compounds containing two or
more alkylation sites may demonstrate a selective toxi-
city towards neoplastic cells in contrast to the corre-
sponding normal tissues [3]. This possibility is based
upon the observation that various tumours are more
susceptible to chemical insult than the corresponding
normal cells [4,5], making the neoplasms more vulnera-
ble to a second chemical attack. In fact, various studies
have revealed the increased tumour sensitivity to anti-
neoplastic drugs by potentiators [6–8]. Therefore, in the
present cluster of compounds, two sites for thiolation
were incorporated into the molecules. Third, myristoyl-
The aim of the present investigation was to prepare a
small number of novel prototypic molecules for evalua-
tion as candidate cytotoxic agents. The specific com-
pounds 2–6 were designed for the following reasons.
First, the compounds chosen are a,b-unsaturated ke-
tones which is a class of molecules having a marked
preference or exclusivity for thiols rather than the
amino or hydroxy groups found in nucleic acids [1].
The synthesis of the Mannich bases of the unsaturated
* Correspondence and reprints.
E-mail address: dimmock@skyway.usask.ca (J.R. Dimmock).
´
0223-5234/02/$ - see front matter © 2002 Editions scientifiques et me´dicales Elsevier SAS. All rights reserved.
PII: S0223-5234(01)01294-6

36
J.R. Dimmock et al. / European Journal of Medicinal Chemistry 37 (2002) 35–44
[13,14]. Of relevance to the present investigation is the
hypothesis of Peseckis and Resh that interaction of
NMT with a substrate involves covalent bonding with
the cysteine 169 mercapto group of the enzyme [15].
Consequently, the compounds under consideration may
be effective inhibitors of NMT.
CoA: protein N-myristoyltransferase (NMT) catalyzes
the co-translational transfer of myristic acid from
myristoyl-CoA to the amino terminal glycine residue of
various proteins [9]. Its importance as a molecular
target in developing cytotoxic agents includes the fol-
lowing considerations. Significantly elevated levels of
NMT have been found in colon neoplasias compared to
normal colon cells [10]. This study demonstrated that
an increase in NMT activity appeared at an early stage
in colonic carcinogenesis [10]. Furthermore, immuno-
histochemical studies also showed increased staining for
NMT in colorectal tumours compared to normal mu-
cosa and the staining appeared to be cytoplasmic rather
than nuclear [11]. In addition, gallbladder carcinoma
showed moderate to strong cytoplasmic positivity for
NMT with increased intensity in the invasive compo-
nent whereas the normal gallbladder mucosa showed
weak to negative cytoplasmic staining [12]. In addition,
higher concentrations of various proteins that require
myristoylation have been found in certain preneoplastic
lesions and neoplasms compared to normal tissues
In addition, the importance of the shapes of the
molecules on bioactivity was addressed by incorpora-
ting part of the a,b-unsaturated keto group into ali-
cyclic rings (series 2–5). Changes in the sizes of the
rings would be expected to alter the relative positions of
the alkylating sites which may influence potency con-
siderably. The importance of the shapes of molecules
was also considered in the design of the acyclic ana-
logues 6. It is conceivable that any marked variation in
bioactivity could be attributed, at least in part, to
differences in the topology of the molecules 2–6.
In summary, this study focussed on the preparation
of the compounds in series 2–6 with a view to deter-
mining those structural features which contribute to
cytotoxicity and inhibition of the enzyme NMT.
2. Chemistry
The synthetic routes for preparing the target com-
pounds 2–6 are portrayed in Fig. 1. Reaction of tereph-
thaldehyde with either cyclohexanone, 2-methylcyclo-
hexanone or cycloheptanone in the presence of aqeous
sodium hydroxide solution led to the aldols 1a–c.
X-ray crystallography of 1a and 1b not only confirmed
the aldol structures but revealed different orientations
of the two alicyclic rings relative to the aryl ring. In the
case of 1a for example, the oxygen atoms 1 and 2 were
on opposite sides of the centre of symmetry than the
O1% and O2% atoms (Fig. 2). Treatment of 1a–c with
hydrochloric acid led to the formation of the unsatu-
rated ketones 2a–c. X-ray crystallography of a repre-
sentative compound 2c also indicated that the O1 and
O1% atoms were on different sides of the centre of
symmetry (not shown). It is likely therefore that the
molecules described herein have this spatial arrange-
ment, a view that was strengthened by the molecular
modelling of compound 4. In this case, the energies of
4 modelled with the two oxygen atoms on opposite or
the same sides of the centre of symmetry were 28.03
and 40.13 kcal mol−1, respectively. The crystal struc-
ture of 2c revealed that the olefinic double bonds
adopted the E configuration which is consistent with
the X-ray crystallographic and 1H-NMR studies of
related compounds [16,17]. Reaction of 2a–c with N,N-
dimethylmethyleneammonium chloride led to the for-
mation of the corresponding Mannich bases 3a–c. An
attempt was made to prepare the enone 4 by the
method of synthesising the compounds in series 2 via
Fig. 1. Synthetic scheme leading to compounds 1–5 and structure of
6. The reagents used in the synthetic route were as follows: (i)
terephthaldicarboxaldehyde, aqueous sodium hydroxide solution; (ii)
hydrochloric acid; (iii) N,N-dimethylmethyleneammonium chloride;
(iv) 4-toluenesulphonic acid–morpholine; (v) terephthaldicarboxalde-
hyde.
Fig. 2. ORTEP diagram of 1a (50% of elipsoids).

J.R. Dimmock et al. / European Journal of Medicinal Chemistry 37 (2002) 35–44
37
Table 1
Cytotoxicity of 2a–c, 3a–c, 4, 5, 6a,b and melphalan against murine P388 and L1210 cells and human Molt 4/C8 and CEM T-lymphocytes.
Compound
IC50 (mM)
Potency score a
P388
L1210
Molt 4/C8
CEM
2a
2b
2c
3a
3b
3c
4
5
15.8491.0
\5091.2
15.9091.4
0.0490.01
1.2390.04
1.0690.02
9.0590.7
0.05190.01
3.2790.07
0.2190.01
0.2290.01
218912
]500
56.4914.6
22098
402993
153939
26594
9
4
96.9951.0
1.7290.11
1.6590.13
3.4092.54
53.196.8
8.6990.55
7.4490.91
1.8490.37
3.2490.79
87.2957.7
1.4890.13
1.9390.28
2.2390.68
35.390.14
7.7691.15
7.3090.54
1.8090.17
2.4790.30
12
39
31
30
15
25
22
33
–
1.0890.11
6.1491.41
1.5690.22
59.594.7
7.6590.83
10.390.5
1.7890.21
2.1390.03
6a
6b
Melphalan b
a Potency score reflects the relative cytotoxicities towards all four cell lines (see Section 4 for details).
b These data are taken from Eur. J. Med. Chem. 35 (2000) 970 and are reproduced with the permission of the copyright owner.
the aldols 1 but the isolated product was shown to
be 2-cyclopentylidenecyclopentanone. Thus an al-
ternate strategy was undertaken using an enamine inter-
mediate leading to the desired product 4 which was
converted in to the b-aminoketone 5. Condensation
between terephthalaldehyde and acetone afforded 6a
which was successfully transformed into the Mannich
base 6b.
4. Results and discussion
Initially the cytotoxicity of the compounds in series
2–6 against P388, L1210, Molt 4/C8 and CEM cells
(Table 1) will be discussed. The choice of cell lines was
determined on the basis that the murine cell lines have
been claimed to be good predictors of clinically useful
anticancer agents [18] while activity towards the T-
lymphocytes would indicate a capacity to inhibit the
growth of human tumours. Since melphalan is a clini-
cally useful antineoplastic agent which acts by alkyla-
tion, the potencies of the compounds listed in Table 1
were compared with this reference drug. The Mannich
bases were either more potent than (3a–c, 5 and 6b) or
equipotent with (3b,c and 6b) melphalan against some
or all of the cell lines. Of particular note were 3a and 6b
which were more active than melphalan in all screens
with the exception of the equiactivity of 6b against
P388 cells. These two compounds are clearly important
lead molecules for further development.
In order to compare the bioactivities of the com-
pounds in series 2–6, the following approach was
adopted. A comparison was made in each cell line of all
10 compounds and a score of 10 was assigned to the
most potent compound, 9 to the analogue possessing
the next highest activity and so on while the least active
compound was assigned a rating of 1. The scores for
each of the four cell lines were combined and the results
are presented in Table 1. The order of cytotoxicity was
found to be 3a\6b\3b\3c\5\6a\4\2c\2a\
2b, indicating that the Mannich bases were more cyto-
toxic than the corresponding unsaturated ketones. The
similar cytotoxicity of 3a and 3b suggests that bioactiv-
ity is due to the Mannich bases per se and not to any
deamination process which is blocked in the case of 3b.
In an attempt to discern the physicochemical proper-
ties of these molecules that influence cytotoxicity, the
The log P values, surface areas and volumes of 2a–c,
3a–c, 4, 5, 6a and 6b were obtained by computational
means; these results are portrayed in Table 2.
3. Biological evaluations
All of the compounds in series 2–6 were evaluated
against murine P388 and L1210 cells as well as human
Molt 4/C8 and CEM T-lymphocytes. Bioevaluation
was undertaken using a maximum concentration of 50
mM of the compounds in the case of the most sensitive
cell line, i.e. the P388 cells. The maximum concentra-
tion employed for the three other cell lines was 500 mM.
These data are summarised in Table 1. Compounds 2c
and 3a were examined against a panel of human tu-
mours and the results are given in Table 3. Compounds
3b,c, 5 and 6b inhibited NMT activity in the presence of
cAMP-dependent protein kinase derived peptide, while
all four compounds stimulated enzyme activity when
pp60src was used as the substrate. Four representative
compounds, namely 2b,c and 3b,c, had minimum in-
hibitory concentration (MIC) figures in excess of 25 mg
mL−1 when evaluated against different strains of As-
pergillus fumigatus and Candida albicans. Administra-
tion of 2b,c, 3b,c, 4, 5, 6a and 6b to mice revealed that
anticonvulsant activity was displayed in the cases of 3b
and 5 and the Mannich bases 3b,c, 5 and 6b were lethal
in the 30–300 mg kg−1 dose range.

38
J.R. Dimmock et al. / European Journal of Medicinal Chemistry 37 (2002) 35–44
Table 2
lipophilic groups into this series of compounds may
lead to increased cytotoxicity. In contrast to the enones,
the only relationship noted with 3a–c, 5 and 6b was a
positive correlation using linear but not logarithmic
plots between the log P values of the Mannich bases
(both unprotonated and protonated species) and po-
tency in the P388 screen. These data indicated that the
significant activity of the Mannich bases is not corre-
lated with their surface areas and volumes.
Log P values, surface areas and volumes of 2a–c, 3a–c, 4, 5 and 6a,b.
2
3
,
,
Compound
Log P
Surface areas (A )
Volume (A )
2a
2b
2c
3.87
5.00
4.67
3.84
2.39
5.17
3.73
4.63
3.19
3.08
3.05
1.93
2.28
2.37
0.93
544
590
586
736
740
753
755
781
781
509
710
684
445
667
678
908
993
991
3a (base)
3a (salt)
3b (base)
3b (salt)
3c (base)
3c (salt)
4
5 (base)
5 (salt)
6a
1289
1298
1349
1356
1375
1329
826
1210
1175
708
Several compounds were considered for inclusion in a
screen utilizing a panel of human tumours. In this
process, a prescreen is conducted initially using NCI-
H460, MCF7 and SF-628 cells which are lung, breast
and central nervous system tumours, respectively. The
criterion for activity in this prescreen is that the growth
of at least one cell line should be inhibited by 68% at a
concentration of 10−4 M. Compounds 2a,c and 3a
were examined in the prescreen and the specific figures
for the inhibition, or lack thereof, of growth of the
three cell lines are summarised in Section 6. The pre-
screen revealed that maximum activity was displayed by
3a which inhibited completely the growth of all three
cell lines. The enone 2c and the Mannich base 3a met
the criterion of the prescreen and were evaluated in
vitro against ca. Fifty six human tumours from nine
different neoplastic diseases, namely leukemia,
melanoma, non-small cell lung, colon, central nervous
system, ovarian, renal, prostate and breast cancers. The
results of the evaluation of 2c and 3a as well as three
reference compounds in the human tumour panel are
presented in Table 3. In this screen concentrations of
10−8–10−4 M were used and the amount of compound
required to inhibit the growth of the cells by 50% was
noted. However, when the maximum concentration of
compound, namely 10−4 M, did not inhibit the growth
of a cell line by 50%, the figure of 10−4 M was still
incorporated into the calculation of average cytotoxi-
city. Hence the designation of mean graph midpoint
(MG MID) rather than IC50 values were used in pre-
senting these data.
6b (base)
6b (salt)
1111
1124
log P values, surface areas and volumes of 2a–c, 3a–c,
4, 5, 6a and 6b were determined computationally. The
pKa of a Mannich base of a conjugated styryl ketone,
namely 5-diethylamino-4,4-dimethyl-1-phenyl-1-penten-
3-one hydrobromide is 7.19 [19] and if the basicities of
the Mannich bases described in this study are compara-
ble, then the ratio of free bases to ionised species will be
ca. 3:2 at pH 7.4 [20]. Thus the physical properties of
both molecular species were obtained and these results
are presented in Table 2. Both linear and logarithmic
plots were made between the individual physicochemi-
cal constants and the IC50 values in each of the four cell
lines with firstly the enones, 2a–c, 4 and 6a, secondly
the Mannich bases, 3a–c, 5 and 6b (free bases) and,
finally, with these five Mannich bases as the protonated
species. The P values listed as either B0.05 or B0.1
are given in the Section 6 of this report. A positive
correlation was noted between the cytotoxicity of the
unsaturated ketones 2a–c, 4 and 6a in the P388, Molt
4/C8 and CEM screens and their log P values, surface
areas and volumes. Thus the insertion of large
Table 3
Evaluation of 2c, 3a and reference compounds against various human tumour cell lines.
Compound
All cell lines
Colon cancer cells
Leukemic cells
MG MID a (mM)
MG MID a (mM)
SI b
MG MID a (mM)
SI b
2c
3a
63.1
1.55
23.5
32.6
1.45
53.2
0.683
44.6
7.90
1.42
1.2
2.3
0.5
4.1
1.0
48.1
0.447
4.65
27.6
0.620
1.3
3.5
5.1
1.2
2.3
Melphalan c
5-Fluorouracil c
Helenalin c
a The letters MG MID indicate the mean graph midpoint which is explained in the text.
b The letters SI refer to the selectivity index which is calculated by dividing the MG MID figures for all cell lines by either the MG MID values
of the colon cancer or leukemic cells.
c The data are taken from Eur. J. Med. Chem. 35 (2000) 971 and are reproduced with the permission of the copyright owner.

J.R. Dimmock et al. / European Journal of Medicinal Chemistry 37 (2002) 35–44
39
The data presented in Table 3 revealed that 3a was
15 times more potent than melphalan when cytotoxicity
to all cell lines was considered. In view of the interest of
this laboratory in discovering compounds which are
effective towards colon cancers [19] and leukemic cells
[21], the MG MID values against these specific tumours
were obtained and compared to the average figures
against all cell lines in order to produce selectivity index
(SI) figures. A SI value of 1.5 was arbitrarily chosen as
an indicator of selectivity, i.e. the colon cancers or the
leukemic cells were 50% more vulnerable to the com-
pound than the average cytotoxicity towards all cell
lines. The SI figures are also presented for 5-fluorou-
racil and melphalan which are useful drugs in treating
colon neoplasms [22] and leukemia [23], respectively. In
addition, the figures are also given for helenalin which
is a cytotoxic thiol alkylator [24]. Compound 3a
demonstrated selectivity for colon cancer and leukemic
cells. In particular, the SI figures for 3a were 56 and
69% of the established drugs, 5-fluorouracil and mel-
phalan, respectively. Helenalin was markedly cytotoxic
and demonstrated selectivity to leukemic cells although
its SI figure was lower than those obtained by 3a. The
results of evaluation of selected compounds against a
panel of human tumours revealed the emergence of a
significant lead compound 3a which can be considered
a prototypic molecule for subsequent development.
An attempt was made to determine whether represen-
tative compounds in series 2–6 inhibited NMT. The
amino terminal glycine residue of various proteins is
myristoylated as noted for the catalytic subunit of
cAMP-dependent protein kinase [9,25] and various ty-
rosine kinases including pp60src [26]. In the concentra-
tion range of 15–40 mM, the unsaturated ketones
2a–c, 4 and 6a were virtually insoluble in both water
and dimethylsulphoxide or only sparingly soluble in
dimethylsulphoxide. Hence the efficacy of these com-
pounds to inhibit NMT was not determined. The Man-
nich bases 3b,c, 5 and 6b however were water soluble
and a study of their effect on NMT using both sub-
strates was determined. In the presence of cAMP-de-
pendent protein kinase derived peptide, the IC50 figures
of 3b,c, 5 and 6b towards NMT were 1.5, 25.0, 4.0 and
10.0 mM, respectively. Inhibition therefore was fa-
voured by the compounds with five and six membered
alicyclic rings (5 and 3b) or 6b which has no aliphatic
ring. This effect may have been due to 3b, 5 and 6b
causing less steric impedance to interaction with the
enzyme than 3c. The potency scores listed in Table 1
were in the order 6b\3b\3c\5 while the relative
NMT-inhibiting cytotoxicity properties were 3b\5\
6b\3c indicating that there was no correlation be-
tween cytotoxicity and inhibition of this enzyme. While
some effect on NMT occurs, the concentrations used
are considerably higher than those causing cytotoxicity
to P388, L1210, Molt 4/C8 and CEM cells. The average
of IC50 values of 3b,c, 5 and 6b against these four cell
lines were 2.74, 2.06, 6.04 and 1.41 mM, respectively.
Thus the IC50 values of these Mannich bases towards
NMT were 548, 12 136, 662 and 7092 times, respec-
tively, higher than those required to generate IC50
figures against malignant cells.
In contrast, when the pp60src substrate was em-
ployed, no inhibition was noted using concentrations
up to and including 40 mM except 6b when 75% of the
activity of the enzyme was inhibited using 20 mM of
the Mannich base. However, all of the compounds
caused stimulation of the enzyme, with a 50% increase
in enzyme activity was noted for 3b,c, 5 and 6b at
concentrations of 2, 10, 0.3 and 1.5 mM, respectively.
Similar to the enzyme-inhibiting properties of these
compounds, the greatest potencies were noted with
those molecules possessing small alicyclic rings (3b and
5) and the acylic Mannich base 6b. As the concentra-
tions of compounds were elevated, stimulation of en-
zyme activity increased; the fold increases in
stimulation for 3b,c, 5 and 6b (concentrations in mM)
were 2.8 (5), 5 (20), 2.5 (20) and 3.3 (10), respectively.
Recent results from this laboratory have demonstrated
stimulation of NMT activity by dimethylsulphoxide
and related solvents in the presence of different sub-
strates.1 The postulate was made that hydrogen bon-
ding between the oxygen atoms of the solvents and
protons of the substrate, particularly the hydroxy hy-
drogen atoms of serine and threonine residues, accounts
for this phenomenon. In the current investigation, sup-
port for this hypothesis was obtained since there was no
stimulation noted using cAMP-dependent protein ki-
nase derived peptide, which contains no serine or
threonine residues. However, stimulation was found
using pp60src which is a nonamer with three serine
residues. The stimulation noted is likely due to inter-
molecular bonding between the oxygen atoms of the
Mannich bases and the hydrogen atom of the serine
hydroxyl groups.
The conclusions that may be drawn from the NMT
studies are that it is unlikely that the Mannich bases
interact with the sulfhydryl group of NMT to an appre-
ciable extent, the inhibition-stimulation phenomena do
not contribute significantly to cytotoxicity and there
must be other molecular targets that are primarily
responsible for the marked cytotoxicities of these
compounds.
A number of conjugated styryl ketones and related
Mannich bases possess antifungal activity [27,28]. The
two most common fungal pathogens are A. fumigatus
and C. albicans. Consequently four representative com-
pounds, namely 2b,c and 3b,c, were evaluated against
three strains of both of these fungi. The MIC figures of
1 Pasha, M.K., Dimmock, J.R., Hollenberg, M.D., Sharma, R.K.,
unpublished results.

40
J.R. Dimmock et al. / European Journal of Medicinal Chemistry 37 (2002) 35–44
all four compounds were in excess of 25 mg mL−1 in
contrast to the reference drugs amphotericin B and
itraconazole, which have MIC values of less than 1 mg
mL−1. This observation is of interest because 3b and
3c, in particular, demonstrated marked cytotoxicity to
neoplastic cells, so that their lack of appreciable anti-
fungal activity may indicate that the compounds are
not general biocidal agents but have preferential toxi-
city to neoplastic tissue although further work in this
regard is required.
6. Experimental
6.1. Chemistry
Melting points are in Celsius degrees and are uncor-
rected. Elemental analyses were undertaken by K.
Thoms, Department of Chemistry, University of
Saskatchewan, on the compounds 1a–c, 2a–c, 3a–c, 4,
5, 6a and 6b, and were within 0.4% of the calculated
values. The following compounds were obtained in
hydrated forms (moles of water of crystallisation),
namely 1a (0.25), 2a (0.33), 2c (0.2), 3a (0.5) and 3b
Finally, an estimate of the behaviour of representa-
tive compounds in vivo was made. A useful method is
to undertake anticonvulsant and toxicity screens which
afford not only evidence of the tolerance of mice to the
compounds but also provides information as to
whether penetration of the CNS takes place. Activity in
the anticonvulsant screen may be beneficial in revealing
compounds of potential use in treating CNS tumours;
on the other hand such properties may be dystherapeu-
tic by causing neurotoxicity (NT). Compounds 2b,c,
3b,c, 4, 5, 6a and 6b were administered to mice and
examined for protection in the maximal electroshock
(MES) and subcutaneous pentylenetetrazole (scPTZ)
screens as well as for neurological deficit [29]. In gene-
ral, doses of 30, 100 and 300 mg kg−1 were employed
in these tests. The a,b-unsaturated ketones 2b,c, 4 and
6a displayed no anticonvulsant activity. Neither NT
nor other signs of impairment were noted with these
compounds with the exception of minimal NT dis-
played by 4. In the case of the Mannich bases 3b,c, 5
and 6b, marginal protection was afforded by 3b in the
MES screen and activity was demonstrated by 5 in the
scPTZ test. However, in contrast to the enones, deaths
due to respiratory depression were noted for all of the
Mannich bases. Specific details of these and other
pathological symptoms are given within the Section 6
of this report.
1
(0.5). H-NMR spectra were recorded using a Bruker
AM 500 FT-NMR machine (500 MHz). Some of the
protons on the alicyclic rings are designated by the
subscripts a and b since they were not unequivocally
assigned to the equatorial and axial positions. Silica gel
(60–120 mesh) was used in column chromatography.
TLC utilised silica gel 60 F254-precoated plastic sheets.
6.1.1. Synthesis of series 1
The preparation of 1a was accomplished as follows.
A solution of NaOH (0.1 mol) in water (50 mL) was
added slowly to an ice-cold mixture of cyclohexanone
(0.4 mol), terephthaldicarboxaldehyde (0.04 mol) and
water (100 mL). The reaction mixture was stirred with
ice cooling for 6 h. The resultant precipitate was col-
lected, washed with HCl (3.75% w/v, 3×50 mL), dried
and recrystallised from Et2O–CHCl3 to give 1a, m.p.
198–202 °C in 42% yield.
Compound 1b was prepared in a similar fashion as
1a except the molar concentrations of NaOH, 2-methyl-
cyclohexanone and terephthaldicarboxaldehyde were
0.05, 0.135 and 0.015, respectively. In addition, the
reaction mixture was stirred at room temperature (r.t.)
for 3 days. After acidification, an oil formed which was
separated and extracted with CHCl3. The organic ex-
tract was dried and the solution was concentrated to 15
mL. After standing at r.t. for 1 h, a colourless precipi-
tate was collected and washed with MeOH (3×20 mL).
Recrystallisation from Et2O–CHCl3 gave 1b, m.p. 184–
188 °C in 34% yield.
5. Conclusions
This study has revealed a synthetic chemical route to
various novel enones and the corresponding Mannich
bases. The latter group of compounds demonstrated
marked cytotoxicity. In particular 3a is a prototypic
molecule having considerably greater potency than mel-
phalan towards a number of human tumour cell lines.
Selective toxicity was displayed by 3a towards human
colon and leukemic cells. While some effects on NMT
were demonstrated by four of the Mannich bases, it is
unlikely that their mode of action is mediated via this
enzyme. Molecular modification will need to be con-
ducted in conjunction with murine toxicity studies with
a view to enhancing the differential between selective
cytotoxicity and mammalian general toxicities.
The aldol 1c was synthesised in the same manner as
1b except the molar concentrations of cycloheptanone
and terephthaldicarboxyaldehyde were 0.1 and 0.02,
respectively, and the time of stirring at r.t. was 24 h.
The crude product was chromatographed on silica gel
and eluted with a mixture of EtOAc and C6H14 (5:95)
to yield material which was recrystallised from CHCl3–
MeOH to give 1c, m.p. 139–144 °C in 22% yield.
Compounds 1a–c were obtained as a mixture of
diastereoisomers as revealed by TLC using a solvent
system of C6H14–EtOAc (4:1) and 1H-NMR spec-
troscopy. The assignments made, where possible, of a
representative compound 1c were as follows.
l
(CDCl3): 1.02–1.03 (6H, dd, J=3.0 Hz, 2×CH3),

J.R. Dimmock et al. / European Journal of Medicinal Chemistry 37 (2002) 35–44
41
1.25–1.40 (4H, m, 2×4-CH2), 1.51–1.80 (8H, m, 2×
3-CH2, 2×5-CH2), 2.05–2.11 (2H, m), 2.40–2.52 (2H,
m, 2×6-CH), 2.53–2.61 (2H, m, 2×2-CH), 4.77 (1H,
d, J=9.0 Hz, CHOH), 5.34 (1H, d, J=2 Hz, CHOH),
7.23–7.28 (4H, m, aryl H).
apparatus for 9 h. Terephthaldicarboxaldehyde (0.05
mol) was added and the reaction mixture was heated
under reflux for a further 12 h. On cooling to r.t., HCl
(37% w/v, 20 mL) was added and the mixture stirred
vigorously for 3 h at r.t. The product was extracted
with CHCl3, the organic extracts dried and removal of
the solvents led to the isolation of a brown semisolid
which was chromatographed using silica gel and an
eluting solvent of EtOAc–C6H14 (1:4) to give 4, m.p.
185 °C in 39% yield. 1H-NMR (CDCl3): l 2.04 (4H, m,
2×C4ꢀH2), 2.40 (4H, t, J=7.9 Hz, 2×C5ꢀH2), 2.96–
3.00 (4H, m, 2×C3ꢀH2), 7.35 (2H, t, J=2.57 Hz,
2×vinylic), 7.56 (4H, s, aryl H).
6.1.2. Synthesis of series 2
Hydrochloric acid (37% w/v, 25 mL) was added
dropwise over a period of 10 min to a stirred mixture of
the hydroxyketone 1 (0.02 mol) and MeOH (50 mL) at
r.t. After heating at 80 °C for 3 h, the mixture was
cooled and the precipitates were collected. Compound
2a, m.p. 168–173 °C (dec.), was obtained in 62% yield
upon recrystallisation from Et2O–CHCl3 while 2b, m.p.
193–196 °C, was isolated in 72% yield by recrystallisa-
tion from EtOAc–CHCl3. The unsaturated ketone 2c,
m.p. 129 °C, was purified by chromatography using a
column of silica gel and elution was accomplished using
a gradient solvent of C6H14 and EtOAc. Finally 2c was
recrystallised from EtOAc–CHCl3 to give the ketone in
An initial attempt to prepare 4 by using the method
employed for synthesising series 2 via the intermediates
1 led to the isolation of a product which was purified by
column chromatography using silica gel and an eluting
solvent of a mixture of EtOAc and C6H14 (1:9) to give
2-cyclopentylidenecyclopentanone [31] as a colourless
1
oil in 37% yield. H-NMR (CDCl3): l 1.16–1.71 (4H,
1
m, 8,9-CH2), 1.84–1.90 (2H, m, 4-CH2), 2.26 (4H, m,
7,10-CH2, J=8.0 Hz), 2.48–2.51 (2H, m, 3-CH2),
2.60–2.85 (2H, m, 5-CH2). The carbon atoms in the
cyclopentanone ring were designated 1–5 and the cy-
clopentylidene carbon atoms were assigned numbers
6–10. In this numbering sequence, carbon atom 2 was
attached to carbon atom 6.
51% yield. The H-NMR spectrum of a representative
compound 2b was as follows: l (CDCl3): 1.06 (6H, d,
J=6.8 Hz, 2×CH3), 1.42–1.61 (4H, m, 2×C5ꢀHa,b),
1.72–1.80 (2H, m, 2×C4ꢀHa), 1.91–1.98 (2H, m, 2×
C4ꢀHb), 2.28–2.36 (2H, m, 2×C3ꢀHa), 2.50–2.58 (2H,
m, 2×C3ꢀHb), 2.82–2.90 (2H, m, 2×C6ꢀH), 7.24
(6H, s, 4 aryl H, 2 vinylic H).
Compound 4 was converted into the corresponding
Mannich bases 5 by the same procedure used for the
preparation of the compounds in series 3. It was
purified by digestion in MeOH in the same manner as
3b to give the desired product m.p. 200 °C (dec.) in
6.1.3. Synthesis of series 3
A
mixture of N,N-dimethylmethyleneammonium
chloride (0.009 mol), which had been freshly prepared
by a literature procedure [30], the appropriate unsatu-
rated ketone 2 (0.003 mol) and dry MeCN (100 mL)
was heated at 80 °C for 9 h. The progress of the
reaction was monitored by TLC using a solvent system
of MeOH–CHCl3 (1:4). Approximately one third of the
solvent was removed in vacuo and the precipitate was
collected, washed with dry MeCN and dried. Com-
pound 3a, m.p. \285 °C, was obtained in 22% yield
by recrystallisation from Et2O–MeOH. The other two
Mannich bases were digested in dry MeCN (20 mL) at
60° for 3 h to give 3b, m.p. 227 °C and 3c, m.p. 213 °C
in yields of 79 and 76%, respectively.
1
71% yield. H-NMR (D2O): l 1.56–1.62 (2H, m, 2×4-
CHa), 2.37–2.43 (2H, m, 2×4-CHb), 2.75–2.92 (18H,
m, 4×NCH3), 2×CH2N, 2×5-CH), 3.14–3.18 (2H,
m, 3-CHa), 3.40–3.44 (2H, m, 3-CHb), 7.17 (2H, s,
2×vinylic H), 7.48 (4H, s, 4×aryl H). In this com-
pound, the olefinic and dimethylaminomethyl sub-
stituents were assigned positions 2 and 5, respectively,
on the alicyclic ring.
6.1.5. Synthesis of series 6
A literature procedure [32] was employed for the
synthesis of 6a. Purification was accomplished by chro-
matography on silica gel and using C6H14–EtOAc as
the gradient solvent led to the isolation of the desired
compound m.p. 159 °C (lit. [33] m.p. 156 °C) in 42%
yield.
Conversion of 6a into the Mannich base 6b was
undertaken using the general procedure for preparing
the analogues 3a–c. The crude product was purified by
recrystallisation from MeOH to give 6b, m.p. 234–
235 °C (dec.) in 49% yield. 1H-NMR (D2O): l 2.83
(12H, s, 4×NCH3), 3.24 (4H, t, 2×COCH2, J=6.3
Hz), 3.38 (4H, t, 2×NCH2, J=6.3 Hz), 6.78 (2H, d,
2×olefinic H, J=16.4 Hz), 7.58–7.62 (6H, m, 4×aryl
H, 2×olefinic H).
The 1H-NMR spectrum of a representative com-
pound 3b was as follows: l (DMSO-d6): 1.14 (6H, s,
2×CH3), 1.74–1.84 (4H, m, 2×C4ꢀHa,b), 2.24–2.36
(2H, m, 2×C5ꢀHa), 2.72–2.84 (16H, m, 4×NCH3,
2×NCH2), 2.96–3.20 (2H, m, 2×C5ꢀHb), 3.12–3.18
(2H, m, 2×C3ꢀHa), 3.56–3.60 (2H, m, 2×C3ꢀHb),
7.32 (2H, s, vinylic H), 7.53 (4H, s, aryl H), 10.25 (2H,
(+)
bs, N H).
6.1.4. Synthesis of 4 and 5
A solution of cyclopentanone (0.1 mol), morpholine
(0.11 mol), 4-toluenesulfonic acid (20 mg) and C6H6
(100 mL) was heated under reflux in a Dean–Stark

42
J.R. Dimmock et al. / European Journal of Medicinal Chemistry 37 (2002) 35–44
6.1.6. X-ray crystallographic determination of 1a,b and
2c
literature method [43]. Some of these data are presented
in Table 3.
The data were collected using an Enraf–Nonius
CAD-4 diffractometer with an ꢀ scan. The structure
was solved by NRCVAX [34] and refined using SHELXL
[35]. Fig. 1 was generated by ORTEP [36] as described in
XTAL 3.6 [37]. Atomic scattering factors and anomolous
dispersion corrections were taken from the literature
[38]. Non-hydrogen atoms were found on the E map
and were refined anisotropically. Hydrogen atoms were
not refined and were placed on atoms by geometry.
Specific details of the X-ray crystallography of 1b,c and
2c are available from the authors on request.
6.2.2. E6aluation using NMT
Escherichia
pTrcHisC.hNMT was grown to stationery phase in LB
medium at 37 °C containing ampicillin (50 mg mL−1
and isopropyl-b- -thiogalactopyranoside (1 mM) for 2
coli
DH5a
with
recombinant
)
D
h. The cells were harvested by centrifugation at
10,000×g for 30 min and suspended in tris(hydroxy-
methyl)aminomethane hydrochloride (50 mM, pH 8.0)
containing ethylenediaminetetracetic acid (2 mM),
phenylmethanesulphonyl fluoride (1 mM), leupeptin
(0.5 mg mL−1) and polyoxyethylene (10) isooctylphenyl
ether (0.1%). The suspension was sonicated on ice for
10 s. After centrifugation of 15,000×g for 20 min, the
supernatent was removed for the purification of the
expressed protein. The recombinant hNMT was
purified by a literature procedure [44]. The assay using
NMT has been described previously [45,46].
6.1.7. Molecular modelling and statistical analyses
The structures of the compounds in series 2–6 were
built and minimised using the HyperChem molecular
modelling programme [39] which also generated the
log P values, surface areas and volumes of these
molecules. The linear and logarithmic plots between the
IC50 values of various compounds in the P388, L1210,
Molt 4/C8 and CEM screens and the log P figures,
surface areas and volumes were generated using a com-
mercial software package [40]. The cytotoxicity of 2a–
c, 4 and 6a was positively correlated with the log P
values, surface areas and volumes using linear (lin) or
logarithmic (log) plots in the following cases (P values
in parentheses): P388lin: log P (B0.1), surface area
(B0.05), volume (B0.05); P388log: log P (B0.05), sur-
6.2.3. Antifungal e6aluation
The antifungal evaluations of 2b,c, 3b,c amphotericin
B and itraconazole was accomplished using a broth
dilution method [47] and three isolates of A. fumigatus
(ATCC 208995, ATCC 208996 and ATCC 208997) and
C. albicans (ATCC 90028, ATCC 32354 and ATCC
28367). The highest concentrations (in mM) used were
as follows: 2b: 77.5; 2c: 76.7; 3b: 48.2; and 3c: 49.1. The
MIC values of amphotericin B and itraconazole against
the six organisms were 0.101–0.812 and 0.19–0.39 mM,
respectively.
face area (B0.05), volume (B0.05); Molt 4/C8log
:
log P (B0.05), surface area (B0.05), volume (B0.05);
CEMlin: log P (B0.1) and CEMlog: log P (B0.05),
surface area (B0.05), volume (B0.05). Linear plots
between the IC50 values of the Mannich bases 3a–c, 5
and 6b (both free bases and dihydrochloride salts) in
the P388 screen were positively correlated with the
log P figures (PB0.1). In all other cases the P values
were \0.1.
6.2.4. Anticon6ulsant and NT screens
The compounds were evaluated in the murine MES,
scPTZ and NT screens using a literature procedure [29].
The doses employed were 30, 100 and 300 mg kg−1
except in the case of 3b and 6b, 30 and 100 mg kg−1
doses only were used in the MES and scPTZ tests. The
animals were observed 0.5 and 4 h after administration
of the compounds. No results were available in the
following screens (test, dose in mg kg−1, time in hours):
3c: (scPTZ, 300, 0.5); 5: (MES, 300, 4); 5: (scPTZ, 100
and 300, 4); 6b: (MES, 100, 4); 6b (scPTZ, 100, 4); and
6b: (NT, 300, 4). Protection was noted in the MES
screen with 3b (1/3 mice, 100 mg kg−1, 4 h) and in the
scPTZ test (1/1 mouse, 300 mg kg−1, 0.5 h).
At the end of 0.5 h, NT was demonstrated in the
following cases (dose in mg kg−1, toxicity/number of
animals used): 3b: (30, 1/4), (100, 3/8), (300, 4/4); 3c:
(100, 2/8), (300, 3/4); 4: (100, 1/8); 5: (30, 2/4), (100,
6/8), (300, 4/4); and 6b: (30, 1/4), (100, 5/8), (300, 4/4).
After 4 h, the following NT was observed: 3c: (100,
1/4); 5: (30, 1/2), (100, 4/4), (300, 2/2); and 6b: (30, 2/2),
(100, 1/1).
6.2. Biological e6aluations
6.2.1. Cytotoxicity screening
Evaluation of the compounds in the P388D1 screen
was carried out using a literature procedure [41], while
the L1210, Molt 4/C8 and CEM tests were undertaken
by a method described previously [42].
In the case of the human tumour panel assay, com-
pounds 2a,c, and 3a were examined against the NCI-
H460, MCF7 and SF-268 cell lines at a concentration
of 10−4 M. After incubation for 48 h, the viability of
the cells was determined using the protein-binding dye
sulforhodamine B. The percentage inhibition against
these three cell lines was as follows namely 2a: 18, 64,
37; 2c: 71, 56, 43; and 3a: 100, 100, 100. The Mannich
bases 2c and 3a were subsequently evaluated using ca.
56 cell lines from different neoplastic diseases by a

J.R. Dimmock et al. / European Journal of Medicinal Chemistry 37 (2002) 35–44
43
[13] J. Park, A.I. Meisler, C.A. Cartwright, Oncogene 8 (1993) 2627–
2635.
[14] C.A. Cartwright, M.P. Kamps, A.I. Meiser, J.M. Pipas, W.
Eckhart, J. Clin. Invest. 83 (1989) 2025–2033.
[15] S.M. Peseckis, M.D. Resh, J. Biol. Chem. 269 (1994) 30888–
30892.
[16] J.R. Dimmock, N.M. Kandepu, A.J. Nazarali, T.P. Kowalchuk,
N. Motaganahalli, J.W. Quail, P.A. Mykytiuk, G.F. Audette, L.
Prasad, P. Perjesi, T.M. Allen, C.L. Santos, J. Szydlowski, E. De
Clercq, J. Balzarini, J. Med. Chem. 42 (1999) 1358–1366.
[17] J.R. Dimmock, M.P. Padmanilayam, R.N. Puthucode, A.J.
Nazarali, N.L. Motaganahalli, G.A. Zello, J.W. Quail, E.O.
Oloo, H.-B. Kraatz, J.S. Prisciak, T.M. Allen, C.L. Santos, J.
Balzarini, E. De Clercq, E.K. Manavathu, J. Med. Chem. 44
(2001) 586–593.
The following compounds caused death and/or other
pathological symptoms (dose in mg kg−1, time in hours
and observation). In the scPTZ screen, the adverse
effects were as follows: 3b: (30, 0.5 and 4, death follow-
ing tonic extension); 3b: (100, 4, tonic extension); 5:
(300, 0.5, myoclonic jerks); and 6b: (100, 0.5, death
following clonic seizure). In the NT test, deaths were
caused by the following compounds (dose in mg kg−1
,
time in hours, mortality/number of animals used): 3b:
(300, 0.5, 4/4); 3c: (300, 0.5, 2/4); 5: (100, 4, 4/4); and
6b: (100, 0.5, 5/8), (300, 0.5, 4/4), (100, 4, 1/1).
[18] M. Suffness, J. Douros, in: V.T. De Vita, H. Busch (Eds.),
Methods in Cancer Research, vol. XVI, Academic Press, New
York, 1979, p. 84 Part A.
Acknowledgements
[19] J.R. Dimmock, O.A. Phillips, S.L. Wonko, R.A. Hickie, R.G.
Tuer, S.J. Ambrose, R.S. Reid, B. Mutus, C.J. Talpas, Eur. J.
Med. Chem. 24 (1989) 217–226.
[20] A. Albert, Selective Toxicity, 7th ed., Chapman and Hall, Lon-
don, 1985, p. 642.
[21] J.R. Dimmock, S.C. Vashishtha, S.A. Patil, N. Udupa, S.B.
Dinesh, P.U. Devi, R. Kamath, Pharmazie 53 (1998) 702–706.
[22] W.A. Remers, in: J.N. Delgado, W.A. Remers (Eds.), Wilson
and Gisvold’s Textbook of Organic Medicinal and Pharmaceuti-
cal Chemistry, 10th ed., Lippincott-Raven, Philadelphia, 1998, p.
367.
[23] E.M. Greenspan, H.W. Bruckner, in: E.M. Greenspan (Ed.),
Clinical Cancer Chemotherapy, Raven Press, New York, 1977,
p. 37.
The authors thank the following organisations for
financial support for the study, namely Purdue Neuro-
science Company, USA (J.R.D., A.J., P.K.), Natural
Sciences and Engineering Research Council of Canada
(J.W.Q.), University of Saskatchewan (E.O.O.), Cana-
dian Institutes of Health Research (R.K.S.), National
Cancer Institute of Canada (T.M.A.), Belgian Fonds
voor Geneeskundig Wetenschappelijk Onderzoek
(E.D.C., J.B.) and the National Institute of Neurologi-
cal Disorders and Stroke, USA (J.P.S.). In addition, the
National Cancer Institute, USA, kindly undertook the
bioevaluations using a panel of human tumour cell
lines.
[24] I.H. Hall, K.-H. Lee, E.C. Mar, C.O. Starnes, T.G. Wadddell, J.
Med. Chem. 20 (1977) 333–337.
[25] S.A. Carr, K. Biemann, S. Shozo, D.C. Pamelee, K. Titani, Proc.
Natl. Acad. Sci. USA 79 (1982) 6128–6131.
[26] M.D. Resh, Biochim. Biophys. Acta 1451 (1999) 1–16.
[27] J.R. Dimmock, C.B. Nyathi, P.J. Smith, J. Pharm. Sci. 68 (1979)
1216–1221.
References
[28] J.R. Dimmock, L.M. Smith, J. Pharm. Sci. 69 (1980) 575–580.
[29] J.P. Stables, H.J. Kupferberg, in: G. Avanzini, P. Tanganelli, M.
Avoli (Eds.), Molecular and Cellular Targets for Antiepileptic
Drugs, John Libbey, London, 1997, pp. 191–198.
[30] H. Bohme, D. Hartke, Chem. Ber. 93 (1960) 1305–1309.
[31] D.M. Newitt, R.P. Linstead, J. Chem. Soc. (1937) 876–883.
[32] C.B. Nyathi, M.A. Benhura, T. Nyanzunda, Pharmazie 41
(1986) 430–431.
[1] J.R. Dimmock, P. Kumar, Curr. Med. Chem. 4 (1997) 1–22.
[2] J.R. Dimmock, L.M. Smith, P.J. Smith, Can. J. Chem. 58 (1980)
984–991.
[3] J.R. Dimmock, N.M. Kandepu, A.J. Nazarali, N.L. Motagana-
halli, T.P. Kowalchuk, U. Pugazhenthi, J.S. Prisciak, J.W. Quail,
T.M. Allen, R. LeClerc, C.L. Santos, E. De Clercq, J. Balzarini,
J. Med. Chem. 43 (2000) 3933–3940.
[4] K. Tsutsui, C. Komuro, K. Ono, T. Nishidai, Y. Shibamoto, M.
Takahashi, M. Abe, Int. J. Radiat. Oncol. Biol. Phys. 12 (1986)
1183–1186.
[5] J.B. Mitchell, A. Russo, Br. J. Cancer 55 (1987) 96–104.
[6] L.J. Brandes, G.M. Queen, F.S. LaBella, Cancer Chemother.
Pharmacol. 45 (2000) 298–304.
[7] R.J. Griffin, C.E. Arris, C. Bleasdale, F.T. Boyle, A.H. Calvert,
N.J. Curtin, C. Dalby, S. Kanugula, N.K. Lembicz, D.R.
Newell, A.E. Pegg, B.T. Golding, J. Med. Chem. 43 (2000)
4071–4083.
[8] A.L. Vahrmeijer, J.H. van Dierendonck, J. Schutrups, C.J. van
de Velde, G.J. Mulder, Cancer Chemother. Pharmacol. 44 (1999)
111–116.
[33] W. Lo¨w, Annalen 231 (1885) 361–384.
[34] E.J. Gabe, Y. LePage, J.-P. Charland, F.L. Lee, P.S. White, J.
Appl. Cryst. 22 (1989) 384–387.
[35] G.M. Sheldrick, SHELXL 97. Programme for Refinement of
Crystal Structures, University of Gottingen, Germany, 1997.
[36] C.K. Johnson, ORTEP II Report ORNL-5138, Oak Ridge Na-
tional Laboratory, TN, USA, 1976.
[37] S.R. Hall, G.S.D. King, J.M. Stewart, Xtal 3.6 System, Univer-
sity of Western Australia, Perth, Australia, 1999.
[38] International Tables for X-ray Crystallography, vol. IV, Kynoch
Press, Birmingham, UK, 1974.
[39] HYPERCHEM Release 6.02 for Windows (2001), Hypercube Inc.,
Waterloo, ON, Canada.
[9] R.V.S. Rajala, R.S.S. Datla, T.N. Moyana, R. Kakkar, S.A.
Carlsen, R.K. Sharma, Mol. Cell. Biochem. 204 (2000) 135–155.
[10] B.A. Magnuson, R.V.S. Raju, T.N. Moyana, R.K. Sharma, J.
Natl. Cancer Inst. 87 (1995) 1630–1635.
[40] Statistical Package for Social Sciences, SPSS for Windows, Stan-
dard Version, Release 10.0.5, SPSS Inc., Chicago, IL, USA,
1999.
[11] R.V.S. Raju, T.N. Moyana, R.K. Sharma, Exp. Cell. Res. 235
(1997) 145–154.
[41] O.A. Phillips, L.A. Nelson, E.E. Knaus, T.M. Allen, R. Fathi-
Afshar, Drug Des. Deliv. 4 (1989) 121–127.
[12] R.V.S. Rajala, J.M. Radhi, R. Kakkar, R.S.S. Datla, R.K.
Sharma, Cancer 88 (2000) 1992–1999.
[42] J. Balzarini, E. De Clercq, M.P. Mertes, D. Shugar, P.F. Tor-
rence, Biochem. Pharmacol. 31 (1982) 3673–3682.

44
J.R. Dimmock et al. / European Journal of Medicinal Chemistry 37 (2002) 35–44
[43] M.R. Boyd, K.D. Paull, Drug Dev. Res. 34 (1995) 91–109.
[44] R.V.S. Raju, R.S.S. Datla, R.K. Sharma, Mol. Cell. Biochem.
155 (1996) 69–76.
[46] R.V.S. Raju, R.K. Sharma, Methods Mol. Biol. 116 (1999)
193–211.
[47] E.K. Manavathu, S.C. Vashishtha, G.J. Alangaden, J.R. Dim-
mock, Can. J. Microbiol. 44 (1998) 74–79.
[45] M.J. King, R.K. Sharma, Anal. Biochem. 119 (1991) 149–153.