Synthesis and antibacterial study of unsaturated Mannich ketones

Article abstract of DOI:10.1016/S0223-5234(01)01264-8

Several Mannich ketones of 2-arylmethylenecycloalkanones were synthesised using the classical acid-catalysed Mannich reaction. Antibacterial activity of these new water-soluble compounds was reported against Pseudomonas aeruginosa, Escherichia coli, Staphylococcus aureus, Staphylococcus saprophyticus, Micrococcus luteus and Bacillus subtilis standard strains. Human cell line cytotoxicity of our new compounds was evaluated against HeLa cell lines. Some compounds showed low cytotoxicity (41.52 nM mL^-1 for 14 and 46.60 nM mL^-1 for 18) and proved to be efficient antibacterial agents against the Gram-positive strains. Minimum inhibitory concentrations varied from 1.56 to 100 μg mL^-1. The mechanism of action was examined, too.

Full text of DOI:10.1016/S0223-5234(01)01264-8

Eur. J. Med. Chem. 36 (2001) 705–717
´
©
2001 Editions scientifiques et m e´ dicales Elsevier SAS. All rights reserved
PII: S0223-5234(01)01264-8/FLA
Original article
Synthesis and antibacterial study of unsaturated Mannich ketones
a,
b
c
d
e
f
Tam a´ s L o´ r a´ nd *, B e´ la Kocsis , P a´ l Soh a´ r , Gergely Nagy , Gyula Kisp a´ l , Hans-Georg Krane ,
g
h
Horst Schmitt , Edgar Weckert
a
Department of Medical Chemistry, Faculty of Medicine, University P e´ cs, H-7624 P e´ cs, Szigeti u´ t 12, Hungary
Department of Medical Microbiology and Immunology, Faculty of Medicine, University P e´ cs, H-7624 P e´ cs,
b
Szigeti u´ t 12, Hungary
c
Department of General and Inorganic Chemistry, L o´ r a´ nd E o¨ tv o¨ s University H-1518 Budapest, POB 32, Hungary
d
Department of Immunology and Biotechnology, Faculty of Medicine, University P e´ cs, H-7624 P e´ cs, Szigeti u´ t 12, Hungary
e
Department of Human Biochemistry, Faculty of Medicine, University P e´ cs, H-7624 P e´ cs, Szigeti u´ t 12, Hungary
f
Mineralogisch-Petrologisches Institut, Universit a¨ t Bonn, Poppelsdorfer Schloß, D-53115 Bonn, Germany
g
Mineralogisch-Petrographisches Institut, Universit a¨ t Hamburg, Grindelallee 48, D-20146 Hamburg, Germany
h
HASYLAB at DESY Notkestraße 85, D-22607 Hamburg, Germany
Received 26 April 2001; revised 20 July 2001; accepted 31 July 2001
Abstract – Several Mannich ketones of 2-arylmethylenecycloalkanones were synthesised using the classical acid-catalysed Mannich
reaction. Antibacterial activity of these new water-soluble compounds was reported against Pseudomonas aeruginosa, Escherichia
coli, Staphylococcus aureus, Staphylococcus saprophyticus, Micrococcus luteus and Bacillus subtilis standard strains. Human cell line
cytotoxicity of our new compounds was evaluated against HeLa cell lines. Some compounds showed low cytotoxicity (41.52
−
1
−1
nM mL for 14 and 46.60 nM mL for 18) and proved to be efficient antibacterial agents against the Gram-positive strains.
−
1
´
Minimum inhibitory concentrations varied from 1.56 to 100 mg mL . The mechanism of action was examined, too. © 2001 Editions
scientifiques et m e´ dicales Elsevier SAS
Mannich ketone / NMR / antibacterial activity / cytotoxicity
1
. Introduction
[4]. These compounds were screened against human
pathogenic yeasts showing marked antifungal effect.
In order to prepare more efficient water-soluble un-
saturated ketones as potential antimicrobial agents we
wanted to prepare some Mannich ketones of
arylidenecycloalkanones.
The increasing number of the resistant bacterial
strains, especially the highly resistant b-lactamase
producing Staphylococcus aureus and Gram-negative
strains require the development of new effective
chemotherapeutic agents of low toxicity.
The Mannich ketones could show more selective
toxicity toward microorganisms than the parent un-
saturated ketones. Their breakdown can afford either
in 1,2 elimination reactive vinyl ketones or they can
undergo reverse Mannich reaction [1]. The vinyl ke-
tones produced as intermediates have much higher
affinity toward thiols than hydroxy and amino groups
present in the nucleic acids, therefore they do not
show the mutagenic side effect of some alkylating
agents used in the therapy [5]. In addition the water
solubility of the Mannich ketones can alleviate their
The family of the a,b-unsaturated ketones is known
to possess antimicrobial effects [1]. The mode of ac-
tion of this class of compounds has been proposed to
be by their reaction with thiol groups of essential
enzymes [2, 3]. Previously we have reported on our
antifungal studies involving E-2-arylidene-1-te-
tralones, E-3-arylidenechroman-4-ones, and E-3-
arylidene-1-thiochroman-4-ones—homoisoflavones
*
Correspondence and reprints
E-mail address: tamas.lorand@aok.pte.hu (T. L o´ r a´ nd).

7
06
T. L o´ r a´ nd et al. / European Journal of Medicinal Chemistry 36 (2001) 705–717
transport to the site of action, too. Several Mannich
ketones are described having antibacterial, antifungal
and cytotoxic activity [6–9].
base-catalysed aldol condensation (figure 1). The title
compounds have been prepared from the correspond-
ing 2-arylidenecycloalkanones, secondary amines and
paraformaldehyde (figure 2). The products were iso-
lated as hydrochlorides. Our method was the classical
Mannich reaction applying ethanol as a solvent and
HCl as a catalyst. Under these reaction conditions the
1–13 unsaturated ketones are able to undergo iso-
merisation to the thermodynamically more stable en-
docyclic enones [10, 11]. This means the loss of the
starting ketones and the decrease of the possible yield.
In addition it results in a reaction mixture enriched in
the secondary amine component. In order to purify
our Mannich ketones a new method was introduced.
This very mild procedure liberates the Mannich bases
at 0 °C to remove the contaminating secondary
amines (see Section 6). Using this method it is possi-
ble to avoid the deamination of the Mannich bases to
methylene ketones. The physical data of the novel
compounds are shown in table I.
Our objective was to study the structure–antibacte-
rial activity relationship for this class of compounds.
On the other hand we wished to examine our Man-
nich ketones from stereochemical—configuration and
conformation—and electronical point of views as re-
gards the conjugated enone structure. These factors
can influence the possible reaction of these com-
pounds with biological nucleophiles, e.g. thiol en-
zymes. In addition some attempts were made to find
the mechanism of action, too.
2
. Chemistry
The strategy for the synthesis of unsaturated Man-
nich ketones (14–36) involves the preparation of 2-
arylidenecycloalkanones intermediates (1–13) by the
From the spectral data given in tables II and III the
1
postulated structures follow straightforwardly. Only
the remarks below are necessary (figure 3):
The bridging methylene hydrogens (8a) are chemi-
cally non-equivalent due to molecular asymmetry.
Their signals (two double doublets) are the proof of
the expected reaction leading to the aimed products.
The high difference in the chemical shifts of these
methylene hydrogens (ca. 0.5 ppm) is due to the
anisotropic neighbouring effect [12] of the near lying
carbonyl, which influences one of the hydrogens. This
fact suggests hindered rotation around the C-8ꢀC-8a
bond and a quasi-rigid structure for this part of the
molecule.
1
3
The IR carbonyl frequencies and also the C-
NMR chemical shifts of the carbonyl carbons depend
on the ring size and lie in the expected [16, 13] ranges.
Ring strain reveals in higher IR-frequencies for cy-
−
1
clopentanones [17]; wCꢁO: 1700–1709 cm for the
−
1
c-pentanones, 1670–1686 cm for all other com-
pounds with six- to eight-membered rings; lCꢁO:
2
08.1±0.5 ppm (c-pentanones), 203.5±0.5 ppm (c-
hexanones), 205.0 (33), 205.3 (34) and 208.8 ppm
35, 36). The slightly different values suggest similar
(
stereo structures (hardly different conformational
situations).
1
The NMR numbering is shown in figure 3. For easier com-
parison of analogous spectral data, the same numbering of
c-octanones was used for all compounds in the text and in the
tables. The correct (IUPAC) numbering is given in Section 6.
Figure 1. (a) NaOH/H O, 20 °C.
2

T. L o´ r a´ nd et al. / European Journal of Medicinal Chemistry 36 (2001) 705–717
707
favourable because of strong steric interaction be-
tween the aryl and the carbonyl groups) and was
proved earlier for very similar structures [14] by
DNOE measurements [15, 19] which demonstrated
the sterically close arrangement of the 3-methylene
and aryl groups. The similar downfield shifted (due to
anisotropy of the carbonyl [12]) H-8 signals in the
spectra of compounds described in the recent paper
confirm analogous configurations.
The chemical shifts of the H-2a signals (7.35±0.14
ppm) are, however, significantly smaller than the con-
densed benzocyclanones (7.84±0.29 ppm [18]). This
may be arising, besides the absence of a condensed
aromatic ring, from the higher probability of out-of-
plane conformations of the aryl group in the confor-
mational equilibrium. In contrast to benzocyclanones
[18], there is no difference between compounds with
c-pentanone ring on one hand and the ones contain-
ing six- to eight-membered cyclanone rings on the
other hand (indanones have fully planar molecular
skeletons [18]).
The narrow interval of the chemical shifts for H-2a
7.21–7.48 ppm) refers to similar dihedral angles
(
OꢀC(1)ꢀC(2)ꢀC(2a) in the compounds investigated.
The only exceptions are 23 and 24 (7.80 and 7.84
ppm) containing 2-methoxy substituted aryl groups.
The downfield shift of the H-2a signal can be inter-
preted as the anisotropic neighbouring effect of the
CꢀO bond [14]. This fact suggests the coplanar ar-
rangement of the H-2a atom and the methoxy group,
and beyond that, for 23 and 24, the predominance of
the rotamer containing the 3-methylene and methoxy
groups in S-trans position.
Non-coplanar position of the aryl ring to the enone
CꢁC double bond in the preferred conformation is
reflected in the hardly different chemical shifts of the
ring-hydrogens in the aryl group. Due to stronger
conjugation, higher differences in shifts should be
expectable. The commonly high polarisation of the
enone groups resulting in significant shift differences
2
(
ꢀ10 ppm) for the a and b sp carbon atoms to the
carbonyl [13] is not observable in our compounds
DlC-2, C-2a: 0.7–6.0 ppm). Because of the non-pla-
(
nar arrangement of the aryl group, the above fact
cannot be interpreted by the buffering effect of the
aromatic p-sextet. Thus, the suppressed polarisation
refers to non-planar enone conformation in accor-
dance with the X-ray measurement on 36. Rotamers
with non-planar aryl groups have somewhat more
Figure 2. (a) Ethanol, HCl, reflux.
The configuration E around the CꢁC double bond
is plausible chemically (configuration Z is highly un-

7
08
T. L o´ r a´ nd et al. / European Journal of Medicinal Chemistry 36 (2001) 705–717
Table I. Physical data of compounds 5, 7 and 14–36.
a
−1
Compound
General formula
M.p (°C)
Yield (%)
Time of heating (h)
IR (KBr, cm
)
5
7
C H O
77 (dec., hexane)
53
76
22
20
26
18
20
15
14
32
26
23
27
20
12
23
43
50
30
19
20
38
46
24
32
–
–
5
5
5
6
5
5
3
6
6
6
4
11
7
3
5
5
1709
1706
13
14
2
4
C H O
135–137 (methanol)
139 (dec., acetone)
135 (dec., acetone)
122 (dec., acetone)
130 (dec., acetone)
147 (dec., acetone)
147 (dec., acetone)
150 (dec., acetone)
128 (dec., acetone)
133 (dec., acetone)
117 (dec., acetone)
156 (dec., acetone)
144 (dec., acetone)
153 (dec., acetone)
127 (dec., acetone)
120 (dec., acetone)
119 (dec., acetone)
124 (dec., acetone)
136 (dec., acetone)
119 (dec., acetone)
185 (dec., acetone)
164 (dec., acetone)
166 (dec., acetone)
156 (dec., acetone)
15
18
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
C H ClNO
1716, 1626
1707, 1623
1709, 1626
1709, 1624
1705, 1624
1710, 1625
1700, 1622
1701, 1621
1707, 1626
1708, 1625
1708, 1624
1700, 1621
1701, 1617
1673
1675
1686, 1615
1673
1675, 1604
1670
1684, 1608
1678, 1608
1676
18
24
C H ClNO
17
22
2
C H ClNO
17
22
C H ClNO
22
24
C H ClNO
18
24
2
2
3
2
3
2
3
4
2
2
C H ClNO
19
26
C H ClNO
18
24
C H ClNO
19
26
C H ClNO
18
24
C H ClNO
19
26
C H ClNO
18
24
C H ClNO
21
30
C H ClNO
17
22
C H ClNO
18
24
C H ClNO
18
24
C H ClNO
20
28
C H ClNO
3
3
4
19
24
17
24
19
26
2
3
4
2
C H ClNO
19
26
C H ClNO
20
28
C H ClNO
19
26
C H ClNO
20
28
C H ClNO
20
28
2
C H ClNO
1676, 1613
21
30
a
The analytical values were within 90.4% of the theoretical values for C, H and N.
importance in the conformational equilibrium in
higher homologues 33–36 as shown by higher enone
polarisation (DlC-2, C-2a: 5.0–5.7 ppm) and nar-
rower range of ArH signals.
[21]. The torsion angles about the double bond are:
C1ꢀC2ꢀC2aꢀC11 177.01(9)°, C3ꢀC2ꢀC2aꢀC11
2.4(1)°, C1ꢀC2ꢀC2aꢀH2a 0.5(10)° and, C3ꢀC2ꢀ
C2aꢀH2a −174.1(10)°. Therefore, this is an almost
planar surrounding. The conformation of the central
eight-membered ring C1–C8 is crown-like. The tor-
sion angle about the C11ꢀC2a single bond C16ꢀC11ꢀ
C2aꢀC2 is 51.6(1)°. Therefore the double bond
C2ꢁC2a is inclined by this angle with respect to the
plane of the phenyl ring.
In the c-octanone homologues 35 and 36 H-8 lies in
the plane of the carbonyl group as proved by the
significant downfield shift of H-8 (3.42 and 3.51 ppm,
while these shifts are in the range 2.3–2.7 ppm for all
the other compounds). This conformation becomes
possible in the flexible eight-membered ring only.
1
Some H-NMR spectra contain two weak signals at
about 5.35±0.1 and 6.15±0.1 ppm, which originate
from the contamination of <1% concentration. These
signals can be assigned to a terminal methylidene
group in decomposition product formed via deamina-
tion [20]. This fact supports the proposed mechanism
of action of our compounds (see Section 4.3).
The X-ray structure of 36 is depicted in figure 4.
−
The Cl ion is weakly bound by a hydrogen bridge to
the nitrogen atom N1. The distance N1ꢀH1N1···Cl1 is
3
1
.048(1) A and the bond angle at the H-atom is
,
77(1)°. Taking into account the experimental errors,
all distances and angles are within the expected values
Figure 3. NMR numbering of compound 35.

1
a
b
Table II. H-NMR data for compounds 14–28 and 30–36 .
c
d
e
f
g
h
i
Compound CH (pos. 3–7) m (4–10H) H-8 m
CH (8a)
H-2a s NCH2 m (4H) CH2
m
CH2
(2H)
m
Aryl group 1–4 m CH3 s
2
2
(
1H)
2×dd (2H)
(1H)
(4H)
(2–5H)
(3H)
j
j
j
j
j
1
1
1
1
1
1
2
2
2
2
2
2
4
5
6
7
8
9
0
1
2
3
4
5
1.75, ꢁ2.35, ꢁ2.8, 2.97
2.55
ꢁ2.5
ꢁ2.35, 2.j83 7.25
ꢁ2.35, 2.45
1.55
3.63
ꢁ1.8
ꢁ2.9
1.41
–
–
ꢁ2.9
–
1.30
–
1.38
–
1.42
–
7.33, 7.39, 7.51
7.30, 7.33, 7.45
7.35, 7.39, 7.52
–
–
–
–
2.30
3.68
3.79
j
j
j
j
j
1.70, ꢁ2.28, ꢁ2.75, j2.95
2.35, 2.80
7.32
ꢁ2.32, 2.45
j
j
j
j
j
j
j
ꢁ1.8 2.41, 2.80, 3.00
ꢁ2.55
ꢁ2.55, 2.96 7.40
ꢁ2.55
j
j
j
j
j
j
j
1.85, 2.40, ꢁ2.7 ꢁ3.05
ꢁ2.7
2.63, ꢁ2.9
2.34, 2.80
ꢁ7.4
7.30
3.64, 3.74
7.37, 7.42, 7.56
j
j
j
j
j
j
1.70, ꢁ2.45, 2.74, 2.93
ꢁ2.3
ꢁ2.35, ꢁ2.45 3.64
7.14, 7.36
6.79, 7.35
6.89, 7.46
6.86, 7.01, 7.08, 7.27 3.77
6.89, 7.03, 7.10, 7.29 3.79
6.89, 6.96, 7.32, 7.46 3.86
6.93, 6.99, 7.36, 7.49 3.88
j
j
j
1.62, ꢁ2.2, 2.62, 2.82
2.40
ꢁ2.2, 2.72 7.21
ꢁ2.2, 2.35
1.45
j
j
j
j
1.74, 2.30, 2.76, 2.94
ꢁ2.5
2.52
ꢁ2.55
2.56
ꢁ2.55
2.55
ꢁ2.35, 2.82 7.32
ꢁ2.35, ꢁ2.5 3.66
j
j
j
j
1.72, ꢁ2.32, ꢁ2.8, 2.96
2.43, ꢁ2.8
7.30
7.33
ꢁ2.32, 2.45
1.55
3.67
1.57
j
j
j
j
j
j
j
1.75, 2.32, ꢁ2.78, 2.99
2.39, 2.83
ꢁ2.4, ꢁ2.5
j
j
j
j
j
1.71, ꢁ2.35, 2.76, 2.90
ꢁ2.35, 2.87 7.80
ꢁ2.35, 2.48
j
j
j
j
j
1.75, 2.35, 2.81, ꢁ2.95
2.44, 2j.95
7.84
7.28
ꢁ2.45, ꢁ2.55 3.73
j
j
j
1.80, ꢁ2.35, ꢁ2.8, 3.00
ꢁ2.35,
ꢁ2.35, 2.45
1.55
1.40
6.76
3.83
j
ꢁ
2.8
j
j
j
j
26
27
1.80, 2.35, 2.78, 2.98
1.57, 1.62, 1.78, 2.10, 2.61,
ꢁ2.55
2.48
ꢁ2.42, 2.86 7.32
ꢁ2.42, ꢁ2.55 3.71
–
–
6.85, 7.06
ꢁ7.2, ꢁ7.27
6.01
–
j
j
2.38, 2.72
7.31
2.28, 2.40
3.59
2.85
j
j
j
j
j
2
8
1.67, ꢁ1.75, 1.90, 2.25,
ꢁ2.6
2.75, 2.88
7.38
2.50, ꢁ2.6
ꢁ1.8
–
7.3–7.45
–
j
ꢁ
2.7, 2.95
j,k
j
j
j
j
30
31
32
ꢁ1.6, 1.81, 2.10, 2.65, 2.88 2.50
2.40, 2.75
2.41, 2.74
2. 50, 2.82 7.40
7.32
7.34
2.32, 2.45
2.30, 2.45
2.38, ꢁ2.55
3.62
3.62
3.70
–
–
–
7.10, 7.21
6.83, 7.30
6.88. 6.93, 7.02
2.29
3.75
3.88
j,k
j
j
j
ꢁ1.6, 1.83, 2.10, 2.66, 2.88 ꢁ2.48
j,k
j
j
ꢁ1.7, 1.90, 2.18, ꢁ2.75,
ꢁ2.57
2
.98
1.29, 1.38, 1.60, ꢁ2.0, 2.16, ꢁ2.4
.78, 3.00
1.29, 1.34, ꢁ1.6, 1.95, 2.00, ꢁ2.4
j,k
j
j
j
33
34
35
36
2.33, 3.10
7.48
7.47
7.43
ꢁ2.4, 2.50
3.67
–
7.25–7.37
7.25–7.35
ꢁ7.37
–
–
–
–
2
j
j
j
j
j
j
j
j
ꢁ2.4 3.08
ꢁ2.4, 2.5
ꢁ1.55
3.64
1.40
–
j
2
.15, 2.80, ꢁ3.1
j,k j,k
j,k
j
j
ꢁ1.47, ꢁ1.6, ꢁ1.85,
.66, 2.98
3.42
3.51
2.35, 2.89
ꢁ2.4
2
j,k
j,k
j
j
j
ꢁ1.55, ꢁ1.85, 2.68,
ꢁ2.35, 2.96 7.43
ꢁ2.35
ꢁ1.55
1.35
7.28, ꢁ7.34
j
ꢁ
2.92
a
Chemical shifts (in ppm, lTMS=0 ppm) and coupling constants (in Hz) at 500 MHz in CDCl solution.
Compound 29 is a ca. 2:1 mixture of diastereomers having overlapped signals. The only exactly determinable signal, CH , d (J: 6.5): 0.94\0.96. The
3
b
3
assignments were supported (except for 14, 17, 18, 31, 33, 34 and 36) by 2D-HSC (HMQC) and for 19 and 27 also 2D-COSY measurements.
c
4
×m (4H) for 14–26, 5/6×m (6H) for 27, 28, 32, 7/8×m (8H) for 33/34 and 5/4×m (10H) for 35/36.
d
e
f
J: 12.590.3, 8.790.2 and 4.590.3, for downfield dd (33): 15.0 and 6.4.
R group, two (2×2H) or one signal (4H). Pos. 1 for 17.
R group, CH (piperidyl, Pos. 3,5) for 14, 19, 21, 23, 25, 34 and 36, OCH (morpholyl) for 15, 18, 20, 22, 24, 26, 27, 30–33 and 35, CH (pyrrolidinyl,
2
2
2
j
Pos. 3,4) for 16 and 28, and NCH (Pos. 3) for 17 (1H) with a second m at 3.05 (1H).
2
g
R group, CH (piperidyl and compound 17, Pos. 4.
2
h
lH-2%,6% (ꢁd, 2H)\lH-3%,5% (ꢁt, 2H)\lH-4% (ꢁt, 1H) for 14–17, lH-2%,6% (ꢁd, 2H)\lH-3%,5% (ꢁd, 2H) for 18–20, 30 and 31, J: 8.1 (18, 30), 8.8 (19,
0, 31), lH-3% (ꢁt)\lH-6% (ꢁd)\lH-2% (ꢁs)\lH-4% (ꢁd) for 21 and 22, lH-2% (ꢁd)\lH-4% (ꢁt)\lH-5% (ꢁt)\lH-3% (ꢁd) for 23 and 24, lH-3%,5%
2
(
s) for 25, lH-6% (ꢁd)\lH-2% (ꢁs)\lH-5%: 7.03 ( for 26 and 32, lH-3%–6% (m, 4H)\lH-2% (m, 1H) for 27 and 36, coalesced multiplets (5H) for 28, 33
and 34, singlet-like signal for 35. Further aromatic signals (17), H-5%: 7.03 (ꢁd), H-6–8: ꢁ7.12 (coalesced ms, 3H).
i
Pos. 4 for 25, further signal (Pos. 3,5) at 3.86, s (6H), 2×s (2×3H) for 32, other signal at 3.90.
Overlapping signals.
Intensity: 2/4/10H.
j
k

13
a
b
Table III. C-NMR chemical shifts for compounds 14–36 .
Compound C-1 C-2 C-3 C-4 C-8 C-2a C-8a C-1%
C-2%
C-6%
C-3%
C-5%
C-4%
C-2¦
C-3¦ C-4¦ CH3
c
c
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
208.6 136.1
208.4 135.9
208.2 136.0
208.5 136.0
208.4 135.4
207.6 133.4
208.0 134.1
208.0 136.4
207.7 136.1
208.6 136.5
208.2 136.3
208.5 135.6
208.2 134.4
203.0 137.0
203.3 137.3
204.1 137.8
203.8 136.7
203.0 134.8
203.4 135.5
205.0 141.3
205.3 141.4
208.8 141.6
208.8 141.5
28.0 27.3 47.5 132.8
28.0 27.1 47.1 133.2
28.0 27.9 49.0 133.1
28.1 27.1 47.7 133.1
28.0 27.1 47.0 133.3
27.1 26.6 46.5 131.8
27.9 27.0 47.0 132.9
27.5 26.8 46.9 132.2
27.5 26.5 46.6 132.5
28.2 27.7 47.5 131.2
28.2 27.4 47.2 131.3
27.9 27.3 47.4 133.1
27.9 27.0 47.0 133.2
27.5 22.2 46.7 134.9
27.9 22.5 48.9 135.1
28.3 22.8 47.9 135.4
28.2 22.8 47.2 136.0
27.5 22.2 46.7 135.5
28.0 22.7 47.1 136.2
27.4 29.9 48.4 135.8
59.9
59.5
56.9
56.9
59.6
59.3
59.6
59.4
59.0
60.0
59.6
59.9
59.6
59.0
56.8
59.6
59.7
59.3
59.8
60.1
136.7 c
136.4
132.8
129.0
129.6
129.8
129.7
129.1
140.2
160.1
161.1
55.2
54.3
55.0
58.9
54.3
54.4
54.4
54.7
53.8
55.2
54.3
55.2
54.3
53.7
54.4
53.4
54.3
53.9
54.3
54.5
55.4
54.4
55.2
26.4
67.3
24.0
51.5
67.3
25.6
67.4
25.9
66.8
26.4
67.3
26.3
67.4
66.8
23.6
34.8
67.4
67.0
67.4
67.4
26.2
67.4
26.1
24.7
–
–
29.6
–
24.0
–
24.2
–
24.7
–
24.6
–
–
23.6
31.2
–
–
–
–
24.6
–
–
–
–
–
c
131.0
131.0
131.0
129.9
131.9
132.7
129.1
129.7 d
129.1
131.0
113.8
114.7
c
c
c
136.6 c
d
e
136.6
133.1
127.9
128.7 c
136.8
21.9
54.9
55.7
55.1
55.1
55.9
55.9 f
56.6
c
c
115.6 122.9 159.6 129.5 114.9
115.7 123.0 159.6 129.5 114.9
159.3 127.5 111.1 120.6 130.1
159.3 127.9 111.2 120.7 130.2
c
136.7
125.0
124.9
135.6
130.3
135.5c
135.8
136.2
133.3
128.3
129.0
136.5
136.6 d
136.6
136.7
108.4
153.6
131.5
g
110.0 126.9 148.5 109.1 149.2
101.9
–
–
130.0
130.2
130.6
130.7
132.1
128.1
128.4
128.7
129.5
113.8
128.2
128.3
128.3
139.1
159.8
c
h
h
–
21.8
55.2
56.3
–
–
–
d
111.2 124.2 149.0 113.9 149.9
129.8
129.8
130.0
130.1
128.8
128.7
128.8
128.8
128.5
128.4
128.7
128.7
27.3 29.8 48.7 135.7 d 60.4
26.2 26.2 44.2 136.6
26.3 26.8 44.2 136.5
61.9
62.1
24.5
–
a
In ppm (lTMS=0 ppm) at 125.7 MHz. Solvent: CDCl ; Further signals, OCH (25, Pos. 4): 61.3; CH (Pos. 5): 28.7 (27), 29.0 (28), 29.4 (29), 29.4 (30),
3
3
2
d
2
8.9 (31), 29.5(32), (Pos. 5–6): 30.7, 31.1 (33), 30.6, 31.7 (34), (Pos. 5–7): 26.8, 30.8, 35.3 (35), 26.8 , 31.0, 35.8 (36); isoquinoline ring (17), C-5¦: 126.0,
C-6¦,7¦: 127.0, 129.8, C-8¦: 126.6, C-4a¦: 134.7, C-8a¦: 135.2.
b
The assignments were supported by DEPT, 2D-HSC (HMQC, except for 14, 17, 18, 31, 33, 34 and 36), and for 28 also 2D-COLOC (HMBC)
measurements. Superscripts in C-1%–6% and C-2¦–4¦ refer to substituents Ar and R, respectively.
c
Interchangeable assignments.
Two overlapping lines.
C-1¦ (isoquinoline).
Pos. 3,5.
d
e
f
g
OCH O group.
2
h
Doubled signals with the second lines at 56.0 (C-2¦) and 34.9 (C-3¦) due to diastereomeric mixture.

T. L o´ r a´ nd et al. / European Journal of Medicinal Chemistry 36 (2001) 705–717
711
the size of the cycloalkanone ring, the secondary
amine and the quality of the aromatic substituent.
4.1. In vitro antibacterial activity
The antibacterial activity was determined (tables IV
and V) and compared to the activity of standard antibi-
otics (see table VI in Ref. [22]; MIC values from 0.20 to
−
1
1
00 mg mL ). Out of the unsaturated Mannich ketones
examined the seven- and eight-membered ketones were
the least active (30–36). The quality of the amine sub-
stituent generally did not influence the antibacterial
activity neither against the E. coli strains, nor against
the Gram-positive strains. The changes in the position
of the aromatic methoxy group did not affect the activ-
ity neither against E. coli strains (19, 21, and 23) nor
against the Gram-positive ones (20, 22, and 24). As for
the Gram-positive strains 14, 18, 27 and 28—unsatu-
rated Mannich ketones—proved to be the most effective
compounds. More than half of the strains had MIC
−
1
values of 6.25–12.5 mg mL (see table V). In compari-
son with standard commercial antibiotics [22] our com-
pounds were slightly less active. For these strains we did
not find any special structural characteristics required by
an active compound. As regards the Gram-negative
strains all our compounds were inactive against P.
aeruginosa. Mostly the Mannich ketones of cyclopen-
tanones showed activity against E. coli strain, com-
pound 15 produced the highest activity (MIC: 25
mg mL ). As for the effect of the electrondonating
aromatic substituents on the antibacterial activity, the
introduction of the methyl or methoxy groups decreases
the efficiency against E. coli strain (see 14, 15 and 18,
Figure 4. X-ray structure of 36 showing displacement ellip-
soids with 50% probability.
3
. Biology
In vitro antibacterial activity of the compounds
synthesised was examined on standard bacterial
strains: Pseudomonas aeruginosa NIH Hungary
−
1
1
70000, Escherichia coli ATCC 25922, Staphylococcus
saprophyticus NIH Hungary 12008, S. aureus NIH
Hungary 118003, Micrococcus luteus ATCC 9341 and
Bacillus subtilis ATCC 6633. The minimum inhibitory
concentration (MIC) and the minimum bactericidic
concentration (MBC) of our compounds were deter-
mined by test tube dilution method. In vitro cytotoxi-
city tests were carried out on HeLa cell line growing
on microplates. Finally we studied the connection
between the capability of thiol depletion and the
mechanism of action of the novel compounds (see
Section 6).
1
9). The MBC values were also same or near the
same—two or four times higher—as MIC values for
sensitive E. coli strains, so the compounds were mostly
bactericidic. The MBC values for Gram-positive strains
were 8, 16 or 32 times higher than the MIC values. So
the compounds proved to be bacteriostatic. We are sure
that the permeability of bacterial cell wall for our com-
pounds is also an important factor in their antibacterial
activity. The Gram-negative bacteria (P. aeruginosa, E.
coli, etc.) in comparison with Gram-positive ones (S.
aureus, M. luteus, etc.) have a more complicated cell
wall structure. Their thick outer membrane consists of
lipoproteins, lipopolysaccharides, etc. It is difficult for
the compounds to penetrate through this membrane into
the bacterial cytoplasm, to the site of action. We sup-
pose that some of our compounds cannot penetrate the
Gram-negative cell wall and sometimes that may explain
4
. Results and discussion
Our purpose was to find the structure–antibacterial
activity and cytotoxicity relationships in the series of
the new Mannich ketones. Therefore, we varied here

7
12
T. L o´ r a´ nd et al. / European Journal of Medicinal Chemistry 36 (2001) 705–717
their ineffectivity. Further structure–activity studies
e.g. QSAR calculations) are in progress.
efficient antibacterial activity against the Gram-positive
strains. On the other hand for the seven- and eight-
membered derivatives (33–36) the antibacterial effect
was generally diminished both against the Gram-posi-
tive and the Gram-negative strains, too. Out of the
above compounds several representatives are morpho-
line derivatives; while compounds 16, 17, 25, 27 and
especially 32 proved to be rather toxic to the HeLa cells.
These results demonstrate that the introduction of more
(
4.2. In vitro cytotoxicity tests
Compounds 14, 18, 20, 22, 24, 26, 29, and 33–36—
the Mannich ketones of seven- and eight-membered
ring—showed the lowest cytotoxic activity and in fact
this effect was not measurable in the case of 33 (table
VI). The low cytotoxicity of 14, 18 is coupled with an
OCH groups into the aromatic side chain dramatically
3
Table IV. In vitro antibacterial activity of Mannich ketones, expressed as minimum inhibitory concentration values (MIC,
−
1
mg mL ).
No. P. aeruginosa 170000 E. coli 25922 S. saprophyt. 120008 S. aureus 118003
M. luteus 9341
B. subtilis 8833
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
\200
\200
\200
\200
\200
\200
\200
\200
\200
\200
\200
\200
\200
\200
\200
\200
\200
\200
\200
\200
\200
\200
\200
100
25
100
200
\200
200
\200
200
\200
\200
200
\200
200
\200
200
\200
\200
\200
\200
200
\200
\200
\200
6.25
6.25
12.5
25
3.125
12.5
25
25
6.25
50
12.5
12.5
12.5
12.5
12.5
6.25
25
25
12.5
6.25
12.5
50
3.125
3.125
12.5
25
12.5
6.25
12.5
25
50
6.25
12.5
25
6.25
25
12.5
6.25
6.25
6.25
6.25
6.25
6.25
6.25
6.25
12.5
25
6.25
50
12.5
12.5
12.5
12.5
6.25
1.56
3.125
3.125
12.5
25
12.5
50
12.5
12.5
12.5
12.5
12.5
12.5
6.25
12.5
\200
12.5
12.5
25
6.25
3.125
3.125
6.25
12.5
3.125
12.5
25
25
50
100
50
100
100
25
100
50
12.5
100
Table V. Cumulative data of antibacterial sensitivity tests: number of compounds causing the given MIC value.
MIC
mg mL
P. aeruginosa
NIH H. 170000
E. coli ATCC S. saprophyticus NIH
S. aureus NIH M. luteus
B. subtilis
ATCC 6633
−
1
(
)
25922
H. 120008
H. 118003
ATCC 9341
\
200
00
00
23
–
–
–
–
–
–
–
13
7
2
–
1
–
–
–
–
–
–
–
2
5
8
4
4
–
–
2
3
1
11
3
2
1
–
–
2
1
6
3
11
–
–
1
–
1
2
3
11
3
2
–
2
1
5
2
1
6
3
1
0
5
2.5
.25
.125
.56
–
–
Gram-negative
Gram-positive

T. L o´ r a´ nd et al. / European Journal of Medicinal Chemistry 36 (2001) 705–717
713
Table VI. In vitro cytotoxicity of compounds 14–36 on HeLa
cell line, expressed as TD .
1
5>14 that do not show correlation with the expected
50
order; while for the six-membered Mannich ketones the
predicted rates of deamination are 27>29>28 that can
be correlated with the cytotoxicity of these compounds:
27>28>29.
−
1
−1
No.
IC /TD (mg mL
)
IC /TD (nM mL
)
50
50
50
50
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
12.7
9.8
4.6
4.05
15
7
17.2
9.2
35.3
11.7
14.1
7.1
32
0.7
11.6
15.4
7.55
41.52
31.86
15.76
11.44
46.6
20.84
50.91
27.39
104.48
34.83
41.78
19.75
90.95
2.17
4.3. Study of mechanism of action
All the compounds examined contained potential
thiol alkylating groups. The Mannich bases are consid-
ered latent thiol alkylators because under biological
conditions in 1,2 elimination they yield reactive vinyl
ketones that can undergo addition reaction with thiols
[9]. Our Mannich ketones possess two sites for alkyla-
tion (CꢁC and the latent position). Therefore, we exam-
ined the degree of thiol depletion caused by the selected
group of compounds administered at the MIC concen-
tration using the DTNB (5,5%-dithiobis-(2-nitrobenzoic
acid) method [26].
37.92
46.12
22.47
38.65
5a.49
13.6
2.1
Our results (figure 5) revealed that these compounds
can be divided into four classes. Those compounds (33,
3
a
16.5
16.5
25
48.67
49.41
71.85
4) belong to first group, which had no toxicity and
cause no thiol depletion. Most probably, they do not
enter the cells. The second group comprises mildly toxic
and mildly thiol depleting compounds (14, 16, and 24).
In this group the depletion of thiols may certainly
contribute to the toxicity. The third group (20, 18)
displayed strong alkylating effect. They decreased the
intracellular thiol content down to 20% of the control
value, hence they had no effect on the growth of the
cells showing, that living cells have a huge capacity to
protect the essential thiols even if the bulk thion, glu-
tathion, is greatly depleted. Finally, among these com-
pounds we also found a highly toxic one, which caused
no thiol depletion (15). Interestingly the free thiol con-
tent was significantly elevated when cells were grown in
the presence of 15.
a
Non-toxic/non measurable data.
Figure 5. Data of free thiol measurements (Cont.: control).
increases the cytotoxicity. According to the possible
mode of action—mentioned above—the Mannich ke-
tones afford in deamination reaction vinyl ketones that
react with the cellular thiols. The rate of deamination is
inversely proportional to the basicity of the amine side
5
. Conclusions
Some new unsaturated Mannich ketones of aryli-
chain [9]. Thus the pK values of morpholine, piperidine,
a
denecycloalkanones were prepared by using the classi-
cal Mannich reaction. Their structure was proved by
FT-IR and NMR spectroscopic methods. Generally
in the preferred conformation there is a non-coplanar
aryl ring to the enone CꢁC double bond. These results
were corroborated by the X-ray diffraction study
showing a crown-like conformation of the central
eight-membered ring C1–C8 in 36.
pyrrolidine, 1,2,3,4-tetrahydroisoquinoline and 4-
methyl-piperidine are 8.40 [23], 11.12 [23], 11.30 [23],
9
.41 [24] and 10.78 [25]. The predicted rates of deamina-
tion of the five-membered Mannich ketones are 15>17>
4>16. The rates of deamination (see above) can be
1
correlated with the cytotoxicity of these compounds.
The order of potencies against the HeLa cells is 17>16>

7
14
T. L o´ r a´ nd et al. / European Journal of Medicinal Chemistry 36 (2001) 705–717
The antibacterial activity (MIC) of these com-
Aldrich Chemical Co. and Fluka and were not further
purified. The majority of our starting unsaturated ke-
tones (1–4, 6 and 8–13) are known compounds synthe-
sised according to literature methods [27–34]. Thin-layer
chromatography (TLC) was performed on Merck silica
gel plates (60 F254), and as an eluent ethylacetate–ben-
zene (10:1 v/v) was applied. Melting points were deter-
mined in a Boetius apparatus and are uncorrected. The
analytical values were within ±0.4% of the theoretical
values for C, H and N.
pounds was examined both on Gram-positive and on
Gram-negative strains and compared to the activity of
standard antibiotics. All compounds were ineffective
against P. aeruginosa (minimum inhibitory concentra-
−
1
tion: MIC: >200 mg mL ). Thirteen compounds were
inactive and ten others were active (MIC: 25–200
−
1
mg mL ) against E. coli and were bactericidic. All the
Gram-positive strains were sensitive to all the com-
−
1
pounds (MIC: 1.56–200 mg mL ) and the compounds
were bacteriostatic. All the compounds contained two
alkylation sites a direct and a latent one. We observed
that these two sites are not of equal importance
regarding their biological action. This fact is proposed
by the antibacterial activity of the 19, 21 and 23, the
para-, meta-, and ortho-methoxy substituted com-
pounds. In this group the MIC values are rather
similar. In the latter one the ortho-methoxy group is
very close to the b-carbon making it less accessible
toward the attack of thiols. Since the antibacterial
activities are very similar for these p-, m- and o-substi-
tuted compounds, therefore, the direct alkylating ac-
tivity of these compounds does not correlate with the
antibacterial activity. This suggests a secondary im-
portance for the direct alkylating site in the toxicity.
The studies on HeLa cells demonstrated that some
highly active antibacterial agents (14, 18), displayed
low cytotoxicity. The latter fact could be important as
a base for a possible drug development.
1
13
The H- and C-NMR spectra were recorded in
CDCl solution in 5 mm tubes at room temperature, in
3
1
a Bruker DRX 500 spectrometer at 500.13 ( H) and
125.76 ( C) MHz, with the deuterium signal of the
solvent as the lock and TMS as an internal standard. The
13
standard Bruker microprogram NOEMULT.AU to gener-
ate NOE [35] was used with a selective pre-irradiation
time. DEPT spectra [36] were run in a standard manner
[37], using only the [=135° pulse to separate CH/CH3
and CH lines phased ‘up’ and ‘down’, respectively. The
2
2D-COSY, [38, 40], 2D-HSC (HMQC) [39, 41] and
2D-COLOC (HMBC) spectra [42, 43] were obtained by
using the standard Bruker pulse programs COSYGSSW
,
HXCO.AU (INV4GSSW) and HXXCO.AU (INV4GSLRNDSW),
respectively.
FT-IR spectra were taken in a Nicolet Impact 400
spectrophotometer in KBr pellets.
6.1.1. General method for the preparation of
All these compounds are potential alkylating
reagents; therefore their supposed mechanism of ac-
tion can be a depletion of free thiols. We examined
this issue by determining the free thiol content of the
cells grown in the presence of minimum inhibitory
concentration of these compounds. Some toxic com-
pounds depleted the cellular free thiol content, but
other non-toxic compounds decreased the free thiol
even stronger, showing that the thiol depletion cannot
be the only reason for the toxicity. The fact, that a
compound from the same family displayed high toxic-
ity, without thiol depletion gave further support to the
idea, that these compounds have other biological ef-
fects besides alkylation.
2-arylidenecyclopentanones (5 and 7)
The appropriate aldehyde (0.08 mmol) was added
dropwise to the mixture of cyclopentanone (0.4 mmol)
and 1000 mL aqueous sodium hydroxide (0.4%) under
stirring. After the addition was completed the reaction
mixture was stirred for an additional 4 h. Its pH was
adjusted to 7 using 10% acetic acid. The yellow oily
product was extracted with toluene and washed with
distilled water until neutral. The toluene solution was
dried over anhydrous magnesium sulphate for 12 h. The
solvent and the excess of cyclopentanone were evapo-
rated in vacuo and the residue was recrystallised from
hexane.
6.1.2. General procedure for the preparation of Mannich
ketones (14–36)
6
. Experimental protocols
The mixture of the appropriate unsaturated ketone (20
mmol), paraformadehyde (40 mmol) and the secondary
amine (20 mmol) in the presence of concentrated hydro-
chloric acid (0.3 mL) was refluxed in ethanol (50 mL).
The time of heating is indicated in table 1. The reaction
6.1. Chemistry
The reagents and solvents used were purchased from

T. L o´ r a´ nd et al. / European Journal of Medicinal Chemistry 36 (2001) 705–717
715
was monitored by TLC. The hot reaction mixture was
filtered and the solvent was evaporated. The residue was
treated with dry acetone yielding crystalline material.
The products were recrystallised from dry acetone–
methanol mixture. The following purification method
gave pure compounds. To the cold (0 °C) methanolic
solution (100 mL) of Mannich ketones as hydrochlo-
rides (10 mmol) an aqueous solution (50 mL) of anhy-
drous sodium carbonate (5 mmol) was added dropwise
at 0 °C (ice bath) under magnetic stirring. Thereafter,
ice and sodium chloride (20 g) were added to the
solution. The oily Mannich ketones were extracted with
cold chloroform (3×100 mL). The chloroform phase
was washed with cold brine (4×50 mL) and was dried
over anhydrous magnesium sulphate in refrigerator for 1
h. After filtration the solvent was removed in vacuo
using a bath temperature of 15–20 °C. The oily residue
was dissolved in dry acetone and cooled to 0 °C then
treated dropwise with cold methanolic hydrochloric acid
were calculated with an appropriate CꢀH-distance and a
hydrogen displacement parameter of 1.2 times that of
the corresponding C-atom. The shifts (Z/|)max of the
final least-square cycle were smaller than 0.001. R-val-
2
ues are R (ꢀFꢀ )=0.102 and R(ꢀFꢀ)=0.039 for 225
w
refined parameters. Goodness-of-fit is 1.041. The final
difference Fourier map is featureless (Dzmin= −0.28;
−
3
Dzmax=0.48 e A
,
).
6.2. Biology
6.2.1. Determination of minimum inhibitory
concentration and minimum bactericidic concentration
MIC values were determined by test tube dilution
method. The test strains were: P. aeruginosa NIH Hun-
gary 170000, E. coli ATCC 25922, S. saprophyticus NIH
Hungary 120008, S. aureus NIH Hungary 118003, M.
luteus ATCC 9341 and B. subtilis ATCC 6633. The
compounds investigated were dissolved in nutrient broth
−
1
(
4 mL, 2 N). The mixture was refrigerated and the
at a concentration of 200 mg mL . Double dilution
series of compounds were made in nutrient broth. A
nutrient broth starter culture (2 mL) of test bacterial
strains was added to each tube to achieve a final inocu-
crystals separated were filtered and dried. This proce-
dure was repeated if it was necessary. The NMR mea-
surements were done on the Mannich bases.
5
lum of ca. 5×10 colony forming units per millilitre.
6
.1.3. X-ray structure determination of 36
Control tubes without compounds were used to check
the inoculums. The cultures were incubated for 24 h at
37 °C. The MIC values were determined from the low-
est concentration of compounds where the tubes re-
mained clear, showing that the bacterial growth was
inhibited. Loopfuls (10 mL) of nutrient broth cultures
from each tube were plated on nutrient agar to check
the bacterial growth. MBC value was determined for the
lowest concentration of compound under investigation
where we could not detect any living bacterium. All
experiments were performed in triplicate [48].
C H ClNO, M =347.91, colourless crystal of size
2
1
30
r
0
.25×0.30×0.40 mm, triclinic, space group P1
(
(No. 2 of
, b=
h=99.14(3)°, i=
IT [44]). Lattice parameters are: a=7.099(1) A
1
1
,
0.065(2)
01.23(3)°,
A
,
,
c=13.901(3)
k=97.77(3)°
A,
,
and
3
V=947.6(3)
A
,
(
determined from ꢁ1000 reflections with 3<q<33.5°),
−
3
Z=2, Dcalc=1.219 g cm and v (synchrotron radia-
,
tion, k=0.71 A)=2.09 cm . Intensity data were col-
lected with a 2D-CCD detector (SMART 1K, Bruker)
using synchrotron radiation from the bending magnet of
station F1 at HASYLAB/Hamburg and a Si (1 1 1)
monochromator at T=120 K. Data collection was car-
ried out in the range −105h510; −155k513;
x
−
1
6.2.2. In vitro anticellular activity of Mannich ketones
against HeLa cell line
−
2051520 with (sin([)/u)max=0.755. The total num-
HCl, KCl, KH PO , NaCl, Na HPO , MTT (3-(4,5-
2
4
2
4
ber of reflections measured is 19 052, from which 5905
are unique (Rmerged=3.0%, completeness: 86.6%). Inten-
sity data were integrated, reduced and scaled using
SAINT V5 [45]. A total of 5466 reflections have I(h)>
dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide),
isopropanol, RPMI-1640, were obtained from Sigma
chemical Co. (St. Louis, MO, USA.). Heat-inactivated
foetal calf serum low IgG was obtained from Gibco
BRL™—Life Technology Ltd (UK). Streptomycin and
penicillin were obtained from Richter Ltd (Budapest,
Hungary).
2|(I(h)) and were retained for further analysis. The
structure was solved by direct methods (SHELXS [46]).
All non-hydrogen atoms were refined anisotropically
2
using full-matrix least-squares [47] based on ꢀFꢀ with
2
2
2
weights 1/(| (F )+(0.0555P) +0.2864P) using P=
6.2.2.1. Cellmicroculturing
In vitro cultured HeLa cell line (Flow-Labs. Ltd, UK)
was used. The cells were cultured in RPMI-1640 culture
o
2
2
(
F +2F )/3. Hydrogen atoms H1N1 and H2a were
o
c
refined isotropically. The positions of all other H-atoms

7
16
T. L o´ r a´ nd et al. / European Journal of Medicinal Chemistry 36 (2001) 705–717
medium containing 10% heat-inactivated foetal calf
determination of free thiol by the DTNB assay [26]. The
protein concentration was determined by the BCA
(bicinchoninic acid) assay [49]. The assay was repeated
using eight selected compounds in three parallel cul-
tures. Error bar means e.s.d. of three measurements.
−
1
serum with antibiotics streptomycin (25 mg mL ) and
−
1
penicillin (50 mg mL ) in 5% CO humidified atmo-
2
sphere at 37 °C (tissue culture incubator, Forma Scien-
tific, USA). The cells growing in the logarithmic phase,
had viability better than 98% as proved by Trypan blue
exclusion.
Acknowledgements
6.2.2.2. Modified tetrazolium assay
The cells growing in the logarithmic phase, showed
viability better than 98% as proved by Trypan blue
The authors are grateful to the OTKA program (cT
0
30261) for the financial support. We wish to thank Mrs
exclusion. Cells were plated out in 100 mL of medium at
G. N e´ meth for technical assistance.
3
concentration of 4–5×10 cells per flat-bottomed well in
96-wells microtiter plates (Costar Co. Cambridge, MA,
USA). Plates were incubated for 24 h at 37 °C in a
References
humidified atmosphere of 5% CO . Hundred microlitres
2
of media containing the drug dissolved in appropriate
solvent were added to each well and incubated for
further 48 h. Hundred microlitres of medium were then
removed from the wells. MTT (3-(4,5-dimethylthiazol-2-
yl)-2,5-diphenyl tetrazolium bromide) was prepared as a
[
1] Dimmock J.R., Wong M.L.C., Can. J. Pharm. Sci. 11 (1976)
35.
[2] Schraufstatter E., Deutsch S., Z. Naturforsch. Teil B 4 (1949)
76.
3] Barron E.S.G., Singer T.P., J. Biol. Chem. 147 (1945) 221.
2
[
−
1
[4] L o´ r a´ nd T., Al-Nakib T.M., Pr o´ kai L., Synthesis, antimycotic
activity and QSAR studies of homoisoflavanone analogues,
5
mg mL
PBS, pH adjusted to 7.2 with NaOH) containing in 5.4
mM Na HPO , 1.5 mM KH PO , 140 mM NaCl and
stock in 0.15 M phosphate-buffer saline
(
2
12th National Meeting of the American Chemical Society,
Orlando, FL, USA, August 24–29, 1996.
2
4
2
4
2.7 mM KCl. This stock solution can be stored for no
[5] Cairns J., Nature 286 (1980) 176.
more than 2 weeks in dark at 4 °C. Before use it was
filtered through a 0.22 mm filter to remove any blue
formazan products and diluted in RPMI-1640 contain-
[6] Dimmock J.R., Gureshi A.M., Noble L.M., Smith P.J., Baker
H.A., J. Pharm. Sci. 65 (1976) 38.
[
[
[
7] Dimmock J.R., Chamankhah M., Allen T.M., Halleran S.,
Pharmazie 50 (1995) 221.
−
1
ing antibiotics to 1 mg mL . Hundred microlitres of
this solution added to each well and incubated for 4 h.
All untransformed MTT (medium) was removed by
careful aspiration from the wells. The formazan crystals
were dissolved in 100 mL of isopropanol–1 N HCl
8] Dimmock J.R., Chamankhah M., Seniuk A., Allen T.M., Kao
G.Y., Halleran S., Pharmazie 50 (1995) 668.
9] Dimmock J.R., Sidhu K.K., Chen M., Reid R.S., Allen T.M.,
Kao G.Y., Truitt G.A., Eur. J. Med. Chem. 28 (1993) 313.
[10] Conia J.M., Amice P., Bull. Soc. Chim. Fr. (1968) 3327.
[11] Wiberg K.B., Furtek B.L., Olli K.L., J. Am. Chem. Soc. 101
(
24:1). Then the plate was shaken in order to ensure
(
1979) 7675.
solubilisation of the blue formazan crystals. The ab-
sorbance was recorded in an enzyme-linked immunosor-
bent assay plate reader (Dynatech MR 7000) at a 560
nm test wavelength and a 690 nm reference wavelength.
[
[
[
12] Soh a´ r P., Nuclear Magnetic Resonance Spectroscopy, vol. 1,
CRC Press, Boca Raton, FL, 1983, pp. 32, 33 (vol. 2, p. 51).
13] Soh a´ r P., Nuclear Magnetic Resonance Spectroscopy, vol. 2,
CRC Press, Boca Raton, FL, 1983, p. 181.
14] Soh a´ r P., Nuclear Magnetic Resonance Spectroscopy, vol. 2,
6.2.3. Determination of the free thiol content
CRC Press, Boca Raton, FL, 1983, p. 20.
Cells grown in shaker incubator (37 °C) to the mid
[15] Soh a´ r P., Nuclear Magnetic Resonance Spectroscopy, vol. 1,
CRC Press, Boca Raton, FL, 1983, pp. 194–196.
log phase were treated with the different compounds at
−
1
[16] Holly S., Soh a´ r P., in: L a´ ng L., Prichard W.H. (Eds.), Theoret-
ical and Technical Introduction to the Series Absorption Spec-
tra in the Infrared Region, Akad e´ miai Kiad o´ , Budapest, 1975,
p. 93.
the MIC concentration or at 200 mg mL in case of
−
1
ineffective substance (MIC>200 mg mL ) for 1 h. The
treated cells were collected by centrifugation (+4 °C),
washed with 150 mM NaCl solution and stored at
[17] Holly S., Soh a´ r P., in: L a´ ng L., Prichard W.H. (Eds.), Theoret-
ical and Technical Introduction to the Series Absorption Spec-
tra in the Infrared Region, Akad e´ miai Kiad o´ , Budapest, 1975,
p. 97.
−
80 °C as a pellet. These cell pellets were resuspended
in a buffer containing 50 mM Tris–HCl, 0.1% SDS of
pH 7.4, and lysed by sonication. The cell lysates were
clarified by centrifugation and used immediately for the
[
18] Perj e´ si P., Nusser T., Tarczay Gy., Soh a´ r P., J. Mol. Struct.
13–19 (1999) 479.

T. L o´ r a´ nd et al. / European Journal of Medicinal Chemistry 36 (2001) 705–717
717
[
[
[
19] Sanders J.K.M., Mersch J.D., Prog. Nucl. Magn. Reson. 15
1982) 353.
[36] Pegg D.T., Doddrell D.M., Bendall M.R., J. Chem. Phys. 77
(1982) 2745.
(
[
[
[
[
37] Bendall M.R., Doddrell D.M., Pegg D.T., Hull W.E., High
Resolution Multipulse NMR Spectrum Editing and DEPT,
Bruker, Karlsruhe, 1982.
20] Ward F.E., Garling D.L., Buckler R.T., Lawler D.M., Cum-
mings D.P., J. Med. Chem. 24 (1981) 1073.
21] Allen F.H., Kennard O., Watson D.J., Brammen L., Orpen
A.G., Taylor R., in: Wilson A.J.C. (Ed.), International Tables
for X-Ray Crystallography, vol. C, Kluwer Academic, London,
38] Ernst R.R., Bodenhausen G., Wokaun A., Principles of Nu-
clear Magnetic Resonance in One and Two Dimensions,
Clarendon Press, Oxford, UK, 1987, pp. 400–448.
1
992.
22] L o´ r a´ nd T., Kocsis B., Em oˆ dy L., Soh a´ r P., Eur. J. Med. Chem.
4 (1999) 1009.
39] Ernst R.R., Bodenhausen G., Wokaun A., Principles of Nu-
clear Magnetic Resonance in One and Two Dimensions,
Clarendon Press, Oxford, UK, 1987, pp. 471–479.
[
[
[
3
23] Stuart R., The Proton, Applications to Organic Chemistry. In:
Organic Chemistry, vol. 46, Academic Press, New York, 1985.
40] Sanders J.K.M., Hunter B.K., Modern NMR Spectroscopy, A
Guide for Chemists, University Press, Oxford, UK, 1987, pp.
1
08–113.
24] Perrin D.D., Dissociation Constants of Organic Bases in
Aqueous Solution, Butterworths, London, 1972 (Supplement to
Pure and Applied Chemistry).
[41] Sanders J.K.M., Hunter B.K., Modern NMR Spectroscopy, A
Guide for Chemists, University Press, Oxford, UK, 1987, pp.
9
4–97 and 100–108.
[
25] Katritzky A.R., Oeksne H., Boulton A.J., Tetrahedron 18
[
[
42] Bax A., Morris G., J. Magn. Res. 42 (1981) 501.
(1962) 777.
43] Kessler H., Griesinger C., Zarboch J., Loosli H., J. Magn. Res.
[
[
26] Ellman G.L., Arch. Biochem. Biophys. 74 (1958) 443.
5
7 (1984) 331.
27] Perj e´ si P., F o¨ ldesi A., Szab o´ D., Zschunke A., M a´ k M., Chem.
[
44] International Tables for X-Ray Crystallography, Kynoch Press,
Birmingham, England, 1979 (present distributor: Kluwer Aca-
demic, Dordrecht, Holland).
Ber. 120 (1987) 1449.
[
[
[
[
[
28] Elphimoff-Felkin I., Sarda P., Tetrahedron 31 (1975) 2781.
29] Mayer R., Gebhardt B., Chem. Ber. 93 (1960) 1212.
30] Sarkar T.K., J. Chem. Soc. Perkin Trans. I. (1973) 2454.
31] Walton H.M., J. Org. Chem. 22 (1957) 1161.
[45] Program SAINT, Bruker AXS Inc., Madison, WI, USA, 1994–
996.
1
[
[
46] Sheldrick G.M., Acta Crystallogr. A 46 (1990) 467.
47] Sheldrick G.M., Schneider T.R., in: Carter C.W., Sweet R.M.
32] Perj e´ si P., F o¨ ldesi A., Batta Gy., Tam a´ s J., Chem. Ber. 122
(
Eds.), SHELXL: High Resolution Refinement. In: Methods in
(1989) 651.
Enzymology, vol. 277, Academic Press, San Diego, 1997.
[
33] Emerson W.S., Birum G.H., Longley R.I. Jr., J. Am. Chem.
Soc. 75 (1952) 1312.
[
48] Shm D.F., Washington J.A., Antibacterial susceptibility tests:
dilution methods, in: Ballows A., Hausler W.J. Jr., Herrmann
K.L., Isenberg H.D., Shadomy H.J. (Eds.), Manual of Clinical
Microbiology, 5th ed., ASM, Washington, DC, 1991, pp.
[
[
34] Braude E.A., Forbes W.F., Gofton B.F., Houghton R.P.,
Waight E.S., J. Chem. Soc. (1957) 4711.
35] Noggle H., Schirmer R.E., The Nuclear Overhauser Effect,
Academic Press, New York, 1971.
1
105–1116.
[49] Smith P.K., Anal. Biochem. 150 (1985) 76.