Esters, amides and substituted derivatives of cinnamic acid: Synthesis, antimicrobial activity and QSAR investigations

Article abstract of DOI:10.1016/j.ejmech.2004.06.013

A series of esters (I_a-k), substituted derivatives (II _a-d) and amides (III_a-q) of cinnamic acid were synthesized and evaluated as antibacterial and antifungal agents. All the derivatives belonging to the series I, II and III showed antimicrobial activity comparable to the standard. Compounds I_f and II_c proved to be the most effective compounds. Quantitative structure-activity relationship (QSAR) investigation with multiple linear regression analysis was applied to find a correlation between different calculated physicochemical parameters of the compounds and biological activity. The quantitative models relating the structural features of cinnamic acid derivatives I_a-k, II _a-d and III_a-q and their antimicrobial activity showed that Gram negative Escherichia coli and Candida albicans (fungus) were the most sensitive microorganisms.

Full text of DOI:10.1016/j.ejmech.2004.06.013

European Journal of Medicinal Chemistry 39 (2004) 827–834
www.elsevier.com/locate/ejmech
Original article
Esters, amides and substituted derivatives of cinnamic acid: synthesis,
antimicrobial activity and QSAR investigations
Balasubramanian Narasimhan a, Deepak Belsare b, Devayani Pharande b,
Vishnukant Mourya c, Avinash Dhake a,*
a Department of Pharmaceutical Sciences, Guru Jambheshwar University, Hisar 125001, India
b NDMVPS College of Pharmacy, Nashik 422002, India
c Government College of Pharmacy, Aurangabad 431005, India
Received 18 September 2003; received in revised form 3 June 2004; accepted 9 June 2004
Available online 11 September 2004
Abstract
A series of esters (Ia–k), substituted derivatives (IIa–d) and amides (IIIa–q) of cinnamic acid were synthesized and evaluated as antibacterial
and antifungal agents. All the derivatives belonging to the series I, II and III showed antimicrobial activity comparable to the standard.
Compounds If and IIc proved to be the most effective compounds. Quantitative structure–activity relationship (QSAR) investigation with
multiple linear regression analysis was applied to find a correlation between different calculated physicochemical parameters of the
compounds and biological activity. The quantitative models relating the structural features of cinnamic acid derivatives Ia–k, IIa–d and IIIa–q
and their antimicrobial activity showed that Gram negative Escherichia coli and Candida albicans (fungus) were the most sensitive
microorganisms.
© 2004 Elsevier SAS. All rights reserved.
Keywords: Cinnamic acid; Antibacterial activity; Antifungal activity; QSAR
1. Introduction
ment of antimicrobial agents [12–14] and in particular it has
already been suggested [15] that QSAR study of mono-, di-
and tri-substituted cinnamic acids against Listeria monocy-
togenes indicated the importance of lipophilicity parameter
on activity. The mode of action [16,17] concerning antibac-
terial activity involves an unspecific interaction (strictly re-
lated to the hydrophobic character of the molecules) on
protein thiol groups. Moreover, concerning antifungal acti-
vity, many compounds exert a membrane action, strictly
related to lipophilicity [18,19]. QSAR study of benzene sul-
phonamide flouroquinolones [12] showed a linear correla-
tion of antibacterial activity with electronic distribution
along with steric parameters. From the above consideration,
it was decided to find a correlation between lipophilicity
parameters along with steric and electronic parameters of
cinnamic acid derivatives and their antimicrobial activity.
Following general antibacterial screening (series Ia–I), indi-
cating sufficiently encouraging effects [7], three series (I, II,
III) were selected for an extensive evaluation of their antimi-
crobial effects against representative bacterial and fungal
microorganisms.
Pursuing the studies on cinnamic acid derivatives with
antibacterial and antifungal activity [1–6], as a part of our
research project on sorbic, cinnamic and ricinoleic acid deri-
vatives with potential antimicrobial activity, we recently re-
ported [7] the synthesis and the correlation between physico-
chemical properties and biological activity of compounds
belonging to the series Ia–i (Table 1).
Since it is well known that cinnamic acid plays an impor-
tant role for the antimicrobial activity [8–11], in the present
paper the evaluation of the in vitro antimicrobial activity of
compounds belonging to the series I–III (Table 1) and the
investigation of the relationship between their physicochemi-
cal properties and microbiological effects have been dis-
cussed.
Several studies have pointed out the major role of quanti-
tative structure–activity relationship (QSAR) in the develop-
* Corresponding author. Tel.: +91-1662-263162; fax: +91-1662-276240.
E-mail address: asdhake@yahoo.co.in (A. Dhake).
0223-5234/$ - see front matter © 2004 Elsevier SAS. All rights reserved.
doi:10.1016/j.ejmech.2004.06.013

828
B. Narasimhan et al. / European Journal of Medicinal Chemistry 39 (2004) 827–834
Table 1
Physicochemical properties of cinnamic acid derivatives
C6H5–CH=CH–COOR(Ia–Ik)
C6H4(R)–CH=CH–COOH(IIa–b) C6H5–CH(R)–CH(R)–COOH(IIc–d)
C6H5–CH=CH–CO–R(IIIa–q)
Compound R
code
Molecular
formula
Molecular
weight
148
162
176
190
190
204
204
218
260
238
275
224
193
178
308
182
147
223
268
313
257
257
257
253
237
175
193
225
217
215
189
203
162
m.p./b.p.
Rf value
Yield
(%)
C(Calc.)
(%)
Anal.
H(Calc.) (%) (%)
–
N (Calc.)
I
H
C9H8O2
133
–
0.76 a
0.71 a
0.75 a
0.87 a
0.921 a
0.901 a
0.68 a
–
–
–
–
–
–
–
–
–
–
–
–
–
Ia
Me
C
C
C
C
C
C
10H10O2
11H12O2
12H14O2
12H14O2
13H16O2
14H18O2
35
86.9
73.2
75.2
69.4
85.9
88.0
67.0
74.0
75.0
43.6
29.0
73.1
90.8
47.0
96.4
60.7
30.0
44.8
30.6
61.5
40.4
81.7
69.6
40.5
46.9
20.7
46.2
15.7
59.5
39.7
61.1
30.2
74.02 (74.06) 6.21 (6.21)
74.88 (74.97) 6.80 (6.86)
75.77 (75.76) 7.38 (7.41)
75.69 (75.76) 7.35 (7.41)
76.40 (76.44) 7.81 (7.89)
76.37 (76.44) 7.81 (7.89)
76.91 (77.03) 8.26 (8.30)
78.38 (78.40) 9.28 (9.29)
80.59 (80.64) 5.92 (5.92)
78.49 (78.52) 4.74 (4.76)
80.37 (80.33) 5.36 (5.40)
55.91 (55.95) 3.62 (3.66)
67.36 (67.40) 5.64 (5.66)
35.26 (35.06) 2.58 (2.62)
59.19 (59.29) 5.52 (5.54)
73.40 (73.44) 6.13 (6.17)
80.59 (80.68) 5.90 (5.87)
67.11 (67.65) 4.49 (4.51)
57.46 (57.50) 3.46(3.54)
70.20 (70.04) 4.63 (4.71)
70.19 (70.04) 4.74(4.71)
70.10 (70.04) 4.64 (4.71)
75.73 (75.86) 5.87 (5.97)
80.82 (80.98) 6.28 (6.38)
75.48 (75.39) 7.46 (7.48)
76.78 (76.80) 8.24 (8.44)
66.52 (66.36) 7.21 (7.29)
71.90 (71.86) 6.98 (6.96)
78.23 (78.09) 7.86 (7.97)
76.41 (76.15) 7.87 (8.00)
76.62 (76.80) 8.29 (8.44)
66.45 (66.64) 6.32 (6.18)
Ib
Et
269 b
280 b
261 b
146 b
137 b
216 b
Ic
n-Pr
Id
I-Pr
Ie
n-Bu
If
I-Bu
Ig
Ih
Ii
CH2–CH2–(CH3)2
CH2–(CH2)6–CH3
CH2-Ph
C9H8O2
C17H24O2
C16H14O2
C18H13O2
C15H12O2
Above 300 b 0.74 a
75
0.91 a
0.61 a
0.91 a
0.62 a
0.22 c
0.53 a
0.45 a
0.62 d
0.24 e
0.89 a
0.23 a
0.11 a
0.90 a
0.47 c
0.68 a
0.21 a
0.33 a
0.11 a
0.10 a
0.32 a
0.36 a
0.60 a
0.32 a
0.25 a
Ij
Ik
8-Hydroxy Quinolinyl
Ph
79
73
–
IIa
IIb
IIc
IId
IIIa
IIIb
IIIc
IIId
IIIe
IIIf
IIIg
IIIh
IIIi
IIIj
IIIk
IIIl
IIIm
IIIn
IIIo
IIIp
IIIq
m-NO2
C9H7O4
10H10O3
194
170
111
Above 300
92
7.23 (7.25)
p-Ome
C
–
Dibromo
Dihydroxy
NH2
C9H8O2Br2
C9H10O4
C9H9NO
–
–
9.48 (9.52)
6.31 (6.28)
Ph-NH
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
15H13NO
15H12N2O3
15H11N3O5
15H12NOCl
15H12NOCl
15H12NOCl
16H15NO2
16H15NO
11H13NO
13H17NO
13H17NO3
13H15O2
112
60
o-NO2-Ph-NH
2,4-(NO2)2-Ph-NH
p-Cl-Ph-NH
o-Cl-Ph-NH
m-Cl-Ph-NH
p-OMe-Ph-NH
o-Me-Ph-NH
N(CH3)2
10.32 (10.45)
13.31 (13.42)
5.36 (5.45)
5.33 (5.45)
5.38 (5.45)
5.49 (5.53)
5.81 (5.90)
7.68 (7.80)
6.81 (6.89)
5.91 (5.95)
6.31 (6.45)
6.47 (6.51)
7.28 (7.40)
6.86 (6.89)
17.16 (17.28)
110
170
95
145
85
170
120
72
N(C2H5)2
N(C2H4OH)2
Morpholinyl
Piperidinyl
I-Pr-NH
49
65
14H17NO
12H15NO
13H17NO
102
65
n-Bu-NH
45
NH–NH2
C9H10N2O
126
bBoiling point.
TLC MOBILE PHASE:
aBenzene.
cBenzene:ethylacetate(1:1).
dEthyl acetate.
eBenzene:hexane(1:1).
Therefore to complete our study, we have performed a
quantitative structure–activity relationship (QSAR) investi-
gation in order to reach a better understanding of the different
physicochemical properties. In principle, QSAR analysis is
based on the assumption that the structurally similar com-
pounds are similarly oriented and bind to the same biological
site. The different QSAR techniques differ in the way of
describing compounds and detecting the relationship
between their structure and activity. The aim of multiple
linear regression analysis is to find a statistically significant
correlation between different physicochemical parameters of
the compounds and their biological activity.
activity. Besides the explanatory ability of these models, they
could be used to predict potency of compounds not yet tested
and to suggest ideas for further synthesis of new molecules
with enhanced activity.
2. Chemistry
The esters of Ia–i were synthesized by refluxing cinnamic
acid with corresponding alcohols as reported in our previous
work [7]. Esters Ij and Ik were prepared via the reaction of
corresponding alcohols with acid chloride (Fig. 1). The subs-
tituents on cinnamic acid were chosen by introducing diffe-
rent groups, such as –NO2, –OCH3 that can be assumed to
confer antimicrobial activity (Fig. 2). To study the effect of
removal of double bond on antimicrobial activity, compound
In the present study, multiple linear regression analysis
method was applied to derive statistically significant quanti-
tative models relating the structural features of cinnamic acid
derivatives Ia–k, IIa–d and IIIa–q and their antimicrobial

B. Narasimhan et al. / European Journal of Medicinal Chemistry 39 (2004) 827–834
829
Many of the tested cinnamic acid derivatives showed anti-
fungal properties, displaying log(1/MIC) values of 2.55–
3.10 against Candida albicans and Aspergillus niger. Com-
pounds If and IIc showed activity equal to the reference
compound (log 1/MIC = 3.10) against both the fungal strains
tested. Compound IIIk showed good activity against A. niger
(log 1/MIC = 3.05). All the other compounds tested showed
activity comparable to that of salicylic acid, the reference
drug.
Fig. 1. General scheme for the synthesis of esters (Ia–k).
Fig. 2
(IIa–b).
. General scheme for the synthesis of ring substituted cinnamic acids
Compounds If and IIc were found to be the most effective
antimicrobial substances displaying growth inhibition of all
the tested microorganisms.
IIc and IId were prepared [20,21] (Fig. 3). The amides
4.2. Quantitative structure–activity relationship
The antimicrobial activities of the compounds studied,
presented as log 1/MIC values (Table 3), were used in QSAR
studies. For the purpose of QSAR, the chemical structure of
each compound was described by three groups of parame-
ters: steric, electronic and hydrophobic which were selected
due to their encouraging effect in describing antimicrobial
activity in our previous study [7]. Kier and Hall [23] sugges-
ted the valence zero order molecular connectivity, a structu-
ral parameter that can be calculated by the following equa-
tion.
Fig. 3. General scheme for the side chain substituted cinnamic acids (IIc–d).
Fig. 4. General scheme for the synthesis of amides (IIIa–q).
(IIIa–q) of cinnamic acid were synthesized by the reaction of
cinnamoyl chloride with the corresponding primary or se-
condary amines [22] (Fig. 4).
A
͚
−1⁄2
͒
0vV =
div
͑
The compounds Ia–k, IIa–d and IIIa–q were generally
obtained in good yield and purified by recrystallization. Phy-
sicochemical data of these compounds and elemental analy-
ses, for C, H, N, are shown in Table 1. Structural IR data is
where A is the number of hydrogen atoms and div is the
valence vertex degree of atom ‘i’ in hydrogen suppressed
molecule. This connectivity index signifies the degree of
branching, connectivity of atoms, and the unsaturation in the
molecule, which accounts for the variation in the activity
[24]. Topological indices are 2D descriptors [23], based on
graph theory concepts which have been widely used in QSPR
and most recently in QSAR studies, which help to differen-
tiate the molecules according to their size, degree of bran-
ching, flexibility and overall shape. Global topological
charge index [25,26], a recently developed topological indi-
ces can be calculated as
1
shown in Table 2. Structural H-NMR data have only been
given in experimental protocols for compounds Ia, IIb and
IIIa.
3. Biology
The in vitro antimicrobial activity of the synthesized com-
pounds was assayed on bacterial and fungal strains and the
minimum inhibitory concentration (MIC) was determined by
serial dilution using salicylic acid as a reference compound.
Global topological charge index =
Ch.Jk
͚
͑ ͒
k−1
where Ch.Jk is the topological charge index of order k. The
hydrophobic parameter log P was computed by using
CLOGP software. The other parameters like unsaturation
index (Ui), hydrophilic factor (Hy), mean atomic Van der
Waals volume (Mv), valance zero order molecular connecti-
vity (0vV) and global topological charge index (JGT) were
computed by using DRAGON software. Total energy (Te)
was computed from Hyperchem version 6. The values of the
above parameters are presented in Table 4. The multiple
linear regression analysis was performed by using QSAR-PC
software. Table 5 presents the correlation matrix of the anti-
microbial activity and Table 6 presents the comparison
log(1/MIC) against different parameters. The equations ob-
tained by multiple linear regression analysis are presented in
Table 7.
4. Results
4.1. Antimicrobial activity
The log(1/MIC) values obtained for cinnamic acid deriva-
tives Ia–k, IIa–d and IIIa–q as well as the parent compound,
cinnamic acid (I) and the reference compound, salicylic acid
are reported in Table 3. All the compounds reported in Ta-
ble 3 showed a good antibacterial activity against Gram
negative Escherichia coli than Gram positive Staphylococ-
cus aureus and Bacillus subtilis. Compounds If and IIc
proved to be the most effective, having a log(1/MIC) value of
2.73 against E. coli. In all other cases the assayed substances
showed activity comparable to the standard salicylic acid.

830
B. Narasimhan et al. / European Journal of Medicinal Chemistry 39 (2004) 827–834
Table 2
IR spectra of cinnamic acid derivatives
Compound IR (Nujol, m cm–1
)
code
Ia
Ib
Ic
Id
1590 (C–C, aromatic), 1640–1650 (C=C, alkene disubstituted trans), 1730 (C=O, ester), 1280 (C–O, COOH), 730 (C–H, aromatic)
1590 (C–C, aromatic), 1640–1650 (C=C, alkene disubstituted trans), 1710 (C=O, ester), 1320 (C–O, COOH), 780 (C–H, aromatic)
1590 (C–C, aromatic), 1630–1650 (C=C, alkene disubstituted trans), 1690–1730 (C=O, ester), 1250–1340 (C–O, COOH), 780 (C–H, aromatic)
1590 (C–C, aromatic), 1620–1730 (C=O, ester), 770–780 (C–H, aromatic)
I
If
Ig
Ih
1590 (C–C, aromatic), 1630–1650 (C=C, alkene disubstituted trans), 1680–1730 (C=O, ester), 770–780 (C–H, aromatic)
1590 (C–C, aromatic), 1650 (C=C, alkene disubstituted trans), 1720 (C=O, ester), 1320 (C–O, COOH), 780 (C–H, aromatic)
1590 (C–C, aromatic), 1640–1650 (C=C, alkene disubstituted trans), 1690–1720 (C=O, ester), 780 (C–H, aromatic)
1590 (C–C, aromatic), 1640–1650 (C=C, alkene disubstituted trans), 1690–1720 (C=O, ester), 1310–1320 (C–O, COOH), 780 (C–H, aromatic)
1640 (C=C, aromatic), 1700 (C=C, alkene disubstituted trans), 1730 (C=O, ester), 1320 (C=N, aromatic), 730–740 (C–H, aromatic)
1610 (C–C, aromatic), 1650 (C=C, alkene disubstituted trans), 1740–1750 (C=O, ester), 1310–1320 (C–O, COOH), 730 (C–H, aromatic)
1640 (C=C, alkene disubstituted trans), 1700 (C=O, COOH), 1310 (C–N, C–NO2 aromatic), 730 (C–H, aromatic)
1590–1610 (C–C, aromatic), 1620–1640 (C=C, alkene disubstituted trans), 1700 (C=O, COOH), 730 (C–H, aromatic)
1700 (C=O, COOH), 1600–1640 (C=C, alkene disubstituted trans), 730 (C–H, aromatic)
Ij
Ik
IIa
IIb
IIc
IId
IIIa
IIIb
3350–3450 (C–OH, 3° alcohol), 1640 (C=O), 1550–1560 (C=C, aromatic), 1310 (OH, 3° alcohol), 730 (C–H, aromatic)
3400 (N–H, 1° amide), 1660–1770 (C=O, 1° amide), 1610 (C=C, alkene disubstituted trans), 1120 (C–N, aliphatic), 730–740 (C–H, aromatic)
1720 (C=O, 2° amide), 1700 (C=C, alkene disubstituted trans), 1630–1640 (C=C, aromatic), 1170–1180 (C–N, aliphatic), 720–730 (C–H,
aromatic)
IIIc
IIId
IIIe
1590 (N–H, 2° amide), 1690 (C=O, 2° amide), 1640 (C=C, alkene disubstituted trans), 1320 (C–N, C–NO2 aromatic), 1170–1180 (C–N,
aliphatic), 730–740 (C–H, aromatic)
1730 (C=O, 2° amide), 1680–1710 (C=C, alkene disubstituted trans), 1640 (C=C, aromatic), 1320 (C–N, C–NO2 aromatic), 1180 (C–N,
aliphatic), 730–740 (C–H, aromatic)
1600–1610 (N–H, 2° amide), 1670–1680 (C=O, 2° amide), 1630–1640 (C=C, alkene disubstituted trans), 1160–1190 (C–N, aliphatic), 740
(C–H, aromatic)
IIIf
IIIg
IIIh
IIIi
IIIJ
IIIK
IIIl
IIIm
IIIn
IIIo
1600 (N–H, 2° amide), 1670–1680 (C=O, 2° amide), 1650 (C=C, alkene disubstituted trans), 1170 (C–N, aliphatic), 740 (C–H, aromatic)
1580–1610 (N–H, 2° amide), 1670–1680 (C=O, 2° amide), 1170 (C–N, aliphatic), 740 (C–H, aromatic)
1590 (N–H, 2° amide), 1700 (C=O, 2° amide), 1640 (C=C, alkene disubstituted trans), 1160–1170 (C–N, aliphatic), 740 (C–H, aromatic)
1600 (N–H, 2° amide), 1670–1680 (C=O, 2° amide), 1630–1640 (C=C, alkene disubstituted trans), 1170 (C–N, aliphatic), 740 (C–H, aromatic)
1700 (C=O, 3° amide), 1640 (C=C, alkene disubstituted trans), 1170 (C–N, aliphatic), 740 (C–H, aromatic)
1670–1680 (C=O, 3° amide), 1630–1640 (C=C, alkene disubstituted trans), 1170 (C–N, aliphatic), 740 (C–H, aromatic)
1700 (C=O, 3° amide), 1640 (C=C, alkene disubstituted trans), 1170 (C–N, aliphatic), 740 (C–H, aromatic)
1660 (C=O, 3° amide), 1610 (C=C, alkene disubstituted trans), 1130–1140 (C–N, aliphatic), 740 (C–H, aromatic)
1160–1670 (C=O, 3° amide), 1640 (C=C, alkene disubstituted trans), 1170 (C–N, aliphatic), 740 (C–H, aromatic)
3350 (NH), 1560 (N–H, 2° amide), 1660–1670 (C=O, 2° amide), 1630 (C=C, alkene disubstituted trans), 1170 (C–N, aliphatic), 740 (C–H,
aromatic)
IIIp
IIIq
1590 (NH, 2° amide), 1700 (C=O, 2° amide), 1640 (C=C, alkene disubstituted trans), 1160–1170 (C–N, aliphatic), 730–740 (C–H, aromatic)
1680–1710 (C=O, 2° amide), 1630–1650 (C=C, alkene disubstituted trans), 1160–1170 (C–N, aliphatic), 730–740 (C–H, aromatic)
5. Discussion
5.2. QSAR studies
The QSAR studies were performed by considering
log(BA) as dependent variable, where BA = 1/MIC1. A cor-
relation matrix (Table 5) was constructed to check the inter-
relationship amongst the different parameters, which indica-
ted that none of the parameters is closely related to any other
parameter, r < 0.7 in all the cases.
Preliminary analyses were carried out in terms of correla-
tion between log(BA) and each of the seven physicochemical
parameters independently and the results are presented in
Table 6.Amongst the bacterial species, all independent varia-
bles showed a poor correlation with a maximum r value
being 0.42 (log(BAs) with Mv). A similar trend was observed
5.1. Antimicrobial activity
From the results of the microbiological studies, it is evi-
dent that both isobutyl cinnamate (If) and dibromo cinnamic
acid (IIc) exhibited strong antibacterial activity against Gram
positive and Gram negative bacteria and good antifungal
properties. It is worthwhile observing that the removal of
double bond in side chain of cinnamic acid was effected with
–OH and –Br group. The results showed that addition of
halogens to the side chain (IIc) caused remarkable increase in
growth inhibitory effect of cinnamic acid whereas addition of
hydroxy groups to the side chain (IId) double bond did not
remarkably enhance the antimicrobial activity.
1 BA= (1/MIC)

B. Narasimhan et al. / European Journal of Medicinal Chemistry 39 (2004) 827–834
831
Table 3
Table 4
Antimicrobial activity of cinnamic acid derivatives, expressed as
Physicochemical parameters of cinnamic acid derivatives
log(1/MIC) µg M ml–1
Compound Log P Ui
Hy
Mv
0vV
JGT
Te
Compound
code
Log(1/MIC)
S. aureus B. subtilis E. coli
code
C. albicans A. niger
I
Ia
Ib
Ic
Id
Ie
If
Ig
Ih
Ii
Ij
Ik
IIa
IIb
IIc
IId
IIIa
IIIb
IIIc
IIId
IIIe
IIIf
IIIg
IIIh
IIIi
IIIj
IIIk
IIIl
IIIm
IIIn
IIIo
IIIp
IIIq
2.09 3.17
2.47 3.17
3.00 3.17
2.52 3.17
3.30 3.17
4.05 3.17
3.92 3.17
4.45 3.17
6.16 3.17
4.23 3.90
4.58 4.32
4.31 3.90
1.83 3.46
2.01 3.17
2.88 3.00
–0.39 3.00
1.43 3.17
3.61 3.90
3.45 4.09
3.36 4.25
4.79 3.90
3.94 3.90
4.79 3.90
3.89 3.90
3.81 3.90
1.73 3.17
2.61 3.17
0.78 3.17
1.57 3.17
2.74 3.17
2.59 3.17
3.34 3.17
2.52 3.17
–0.20
–0.81
–0.82
–0.84
0.65
5.90
0.35 –0.45
I
Ia
Ib
Ic
Id
Ie
If
Ig
Ih
Ii
Ij
Ik
IIa
IIb
IIc
IId
IIIa
IIIb
IIIc
IIId
IIIe
IIIf
IIIg
IIIh
IIIi
IIIj
IIIk
IIIl
IIIm
IIIn
IIIo
IIIp
IIIq
S
2.34
2.29
2.29
2.25
2.43
2.34
2.69
2.34
2.34
2.29
2.34
2.39
2.29
2.34
2.73
2.13
2.29
2.34
2.43
2.39
2.25
2.25
2.34
2.39
2.43
2.43
2.29
2.25
2.34
2.43
2.29
2.39
2.34
2.39
2.34
2.25
2.34
2.34
2.39
2.34
2.73
2.34
2.39
2.34
2.34
2.29
2.34
2.34
2.73
2.43
2.29
2.25
2.39
2.39
2.25
2.25
2.34
2.34
2.48
2.43
2.34
2.25
2.39
2.39
2.39
2.34
2.39
2.39
2.39
2.39
2.34
2.34
2.43
2.34
2.73
2.39
2.29
2.34
2.39
2.39
2.29
2.39
2.73
2.39
2.29
2.25
2.34
2.39
2.25
2.25
2.39
2.34
2.43
2.43
2.34
2.25
2.34
2.39
2.39
2.34
2.34
2.39
2.69
2.69
2.69
2.65
2.73
2.64
3.05
2.73
2.73
2.73
2.59
2.64
2.73
2.69
3.10
2.55
2.64
2.55
2.59
2.55
2.64
2.64
2.64
2.69
2.55
2.73
2.78
2.55
2.64
2.69
2.69
2.73
2.59
3.10
2.69
2.64
2.64
2.73
2.73
2.78
3.10
2.73
2.64
2.69
2.64
2.69
2.73
2.69
3.10
2.59
2.64
2.59
2.55
2.69
2.69
2.69
2.73
2.69
2.59
2.69
3.05
2.64
2.64
2.73
2.64
2.73
2.59
3.1
0.64
0.62
0.61
6.85
7.56
8.72
8.43
8.97
9.41
9.85
11.8
9.95
0.34
0.32
0.31
0.41
0.29
0.40
0.37
0.23
0.29
4.25
4.89
5.51
5.43
6.26
6.06
8.24
8.70
1.65
–0.84 0.691
–0.85
–0.85
–0.86
–0.88
–0.88
–0.84
–0.92
1.47
0.61
0.61
0.60
0.59
0.66
0.69 11.27
0.38 15.32
0.32 15.07
0.68
0.63
0.63
0.72
0.61
0.64
0.67
8.84
7.16
7.23
9.83
6.79
6.03
9.34
0.53
0.46
0.52
0.52
0.35
2.66
6.50
8.12
9.28
9.90
–0.19
–0.12
1.48
–0.20
–0.37
–0.27
–0.19
–0.33
–0.33
–0.32
–0.35
–0.38
–0.82
–0.85
0.42
0.31 17.97
0.67 10.52
0.67 11.71
0.69 10.39
0.44
0.55
4.65
4.70
0.41 22.16
0.69 10.39 0.401 19.18
0.69 10.39
0.65 10.67
0.66 10.26
0.41 21.84
0.39 12.47
0.40 –9.77
0.42 10.03
0.41 20.36
0.62
0.60
0.60
0.62
0.61
0.61
0.61
0.63
7.90
9.31
9.62
9.13
9.43
8.53
8.36
6.53
0.38
0.32
0.32
3.62
1.89
8.15
–0.80
–0.86
–0.30
–0.30
0.60
0.42 –2.84
0.31
0.34
7.83
1.44
C. albicans (Eqs. (16)–(20)). For C. albicans, the regression
with single parameter indicated the importance of constitu-
tional parameter Ui towards activity (P < 0.05, Eq. (16)). The
inclusion of liphophilic parameter log P to Ui highly impro-
ved the statistical significance from P < 0.05 to P < 0.01 (Eq.
(17)). This indicated the importance of log P in contribution
to the activity. A similar trend was observed when JGT was
added as a third parameter (P < 0.01, Eq. (18)).
The results of regression analysis showed the strongest
dependence of activity were on constitutional parameter Ui,
topological parameter JGT followed by lipophilicity parame-
ter log P. The effect of steric parameter 0vV on activity was
seen rarely as in case of A. niger, which may be considered as
a chance correlation. Similarly the effect of electronic para-
meter Te was seen rarely as it was added as fourth and fifth
parameters with a marginal increase in r value. The impor-
tance of liphophilic parameter was well supported by the
QSAR study of mono-, di-, tri-substituted cinnamic acids
against Listeria monocytogenes, which indicated the impor-
tance of lipophilicity parameter and ionisation constant on
activity [15].
in case of fungal species with a maximum r value being
0.39 with Ui.
The multiple regression analysis showed a good correla-
tion between log(BA) and Ui, JGT and log P. The regression
equations (Table 7) obtained for the bacterial species showed
the importance of constitutional parameter Ui, global topolo-
gical charge index JGT followed by the lipophilic parameter
log P in contribution to antibacterial activity (Eqs. (1)–(15)).
The equations obtained by regression analysis were statisti-
cally significant in particular for E. coli.
For E. coli, the activity was strongly dependent on Hy and
JGT and the relationships obtained were significant statisti-
cally (r = 0.552, P < 0.01, Eq. (12)). The addition of other
parameters like Ui, log P, Te, respectively, to the above
parameters showed a gradual increase in r value which was
significant statistically (P < 0.01, Eqs. (13)–(15)), which
indicated the importance of these parameters in contribution
to activity.
The regression equations obtained for fungal species were
significant statistically (Eqs. (16)–(26)) in particular for

832
B. Narasimhan et al. / European Journal of Medicinal Chemistry 39 (2004) 827–834
Table 5
Correlation matrix for cinnamic acid derivatives against S. aureus
Log(BAs)
1
Log P
Ui
Hy
Mv
0vV
JGT
Te
Log(BAs)
Log P
Ui
0.31
1
0.51
–0.00
–0.28
0.42
1
Hy
–0.56
0.53
–0.08
0.40
0.61
0.14
0.29
1
Mv
0vV
–0.48
–0.36
0.58
1
0.27
0.69
0.33
–0.20
0.08
1
JGT
Te
0.17
–0.36
0.31
–0.003
0.29
1
–0.05
–0.15
0.004
1
Table 6
Comparison of log(BA) vs. parameters
Log(BAs)
Log(BAB)
0.02
Log(BAE)
Log(BAC)
0.17
Log(BAA)
0.15
Log P
Ui
0.31
–0.00
–0.28
0.42
0.05
–0.21
–0.15
0.03
–0.24
0.001
–0.07
–0.013
0.35
–0.39
–0.22
0.19
–0.27
–0.17
0.18
Hy
Mv
0vV
JGT
Te
0.27
0.04
0.02
0.12
0.17
0.35
0.12
0.22
–0.05
–0.24
–0.15
0.06
0.14
In conclusion, some interesting compounds (If, IIc) en-
dowed with high antimicrobial activity emerged from the
present study. The cinnamic acid moiety is necessary for the
studied activity. The constitutional parameter Ui, topological
parameter JGT and the lipophilicity parameter log P can be
considered as major factors of determining the differences in
antimicrobial activity, but the other physicochemical indices
are helpful for understanding the microbiological results, as
shown by the QSAR analysis.
As the process of antimicrobial activity is not clearly
defined in terms of target biomolecules, the QSAR results
could not be addressed to a concrete drug–receptor interac-
tion. However, they can reveal trends in the relationship
between ligand structures and their activity for the set of
antimicrobial agents selected. This will be useful for the
future work in antimicrobial research and development.
phuric acid (2 ml). The solution was refluxed for 4 h. After
completion of reaction, the reaction mixture was added in
200 ml of ice cold water and the oily layer was separated. The
product was extracted with ether and evaporation of solvent
resulted in product. The product was recrystallized from
methanol. For the compounds Ie–Ii, the excess of alcohol
from the product was removed by distillation at its boiling
point.
Methyl cinnamate (Ia) IR:m cm–1 = 1590 (C–C str., aroma-
tic) 1640–1650 (C=C str., alkene disubstituted trans), 1730
(C=O ester), 1280 (C–O str.), 730 (C–H, aromatic). 1H-NMR
(300 MHz, CDCl3):d ppm = 3.8 (s, 3H, CH3), 7.5 (m, 5H,
Ar–H), 6.42–6.47, 7.67–7.73 (dd, H, CH=CH, J = 16.5 Hz).
6.1.2. General procedure for synthesis of p-methoxy
cinnamic acid (IIa–b
)
The appropriate aldehyde (0.015 mol) and malonic acid
(0.035 mol) were dissolved in a mixture of 7.5 ml of pyridine
and 1–2 drops of piperidine contained in a round bottom
flask. The mixture was heated under reflux for 5 h. After the
evolution of carbon dioxide ceased, the reaction mixture was
poured into excess of water containing hydrochloric acid to
remove excess of pyridine. The mixture was filtered and the
product was recrystallized from alcohol.
6. Experimental
6.1. Chemistry
Melting points degree Celsius were determined with Elico
melting point apparatus and are uncorrected. IR (Nujol)
spectra were measured on Shimadzu IR 408 spectrophoto-
meter. 1H-NMR was recorded on FX-90Q FT-NMR spectro-
photometer by using CDCl3 as solvent and TMS as an inter-
nal standard (chemical shift in d ppm). The purity of the
compounds was checked by thin layer chromatography
(TLC) on silica gel plates. Elemental analyses of (C, H, N)
were obtained on a Vario-EL instrument.
p-Methoxy cinnamic acid (IIb) IR:m cm–1 = 1590–1610
(C–C str., aromatic), 1620–1640 (C=C str., alkene disubsti-
tuted trans), 1700 (C=O str.), 730 (C–H, aromatic) 1H-NMR
(300 MHz, CDCl3):d ppm = 3.7 (s, 3H, CH3), 7.5 (m, 4H,
Ar–H), 6.82–6.88, 7.67–7.72 (dd, H, CH=CH, J = 0.051).
6.1.3. General procedure for synthesis of compounds
IIIa–q
The solution of corresponding amine (0.1 mol) in ether
(50 ml) was added dropwise to a solution of cinnamoyl
6.1.1. General procedure for synthesis of compounds Ia–k
The appropriate alcohol (0.74 mol) was poured into a
round bottom flask with cinnamic acid (0.08 mol) and sul-

B. Narasimhan et al. / European Journal of Medicinal Chemistry 39 (2004) 827–834
833
Table 7
Equations derived by regression analysis for bacterial and fungal species
Equation Equation
no
Statistics
n
R
s
F
For S. aureus
(1)
(2)
(3)
(4)
Log
͑
BAs
͒
= 2.081
= 2.027
= 1.848
͑
0.108
͒
+ 0.445
− 0.106
− 0.074
͑
0.172
0.035
0.038
0.044
͒
Mv
Hy + 0.760
Hy + 0.341
0.017 log P 0.144
33
33
0.422
0.504
0.578
0.669
0.103
6.73 a
5.11 a
4.85 b
5.68 b
log
͑
BAs
͒
͑
0.120
͒
͑
͒
͑
0.288
͒
JGT
Mv + 0.070
0.050 UI
0.100
0.096
0.089
Log
͑
͑
BAs
BAs
͒
͒
͒
͑
0.150
͒
͑
͒
͑
0.183
͒
͑
0.278
͒
JGT
0.173
33
Log
͒
=
2.113
͑
0.163
͒
+
͑
͒
͑
͒
+
0.484
͑
͒
M v
Hy
+
+
0.070 33
͑
0.278
BAs
JGT
(5)
log
͑
=
2.073
͑
0.164
͒
+
0.037
͑
0.018
͒
log P
−
0.132
͑
0.050
͒
Ui
−
0.047
͑
0.037
͒
0.408 33
0.692
0.088
4.96 b
͑
0.86
͒
Mv + 0.927
͑
0.268
͒
JGT
For B. subtilis
(6)
(7)
(8)
(9)
Log
͑
͑
BAB
BAB
͒
͒
= 2.179
͑
͑
0.093
0.163
͒
͒
+ 0.502
͑
͑
0.238
0.043
͒
͒
JGT
Ui + 0.557
33
33
33
0.354
0.455
0.622
0.675
0.101
0.098
0.088
0.084
4.43 a
3.92 a
6.09 b
5.85 b
Log
= 2.419
− 0.075
͑
0.233
0.048
͒
͒
JGT
log
͑
BAB
͒
= 2.433
͑
0.146
͒
+ 0.046
͑
0.016
͒
log P − 0.161
͑
Ui + 0.913
͑
0.241
͒
JGT
Log
͑
BAB
͒
=
2.397
͑
0.141
͒
+
0.052
͑
0.016
͒
log
P
−
0.150
͑
0.047 Ui
͒
+
0.945
͑ ͒
0.232 33
JGT − 0.004
Log BAB
0.225
͑
0.002
͒
͑
Te
0.149
0.002
(10)
͑
͒
͒
=
2.340
͒
+
0.044
͑
0.017
͒
log P
−
0.144
͑
0.047
͒
Ui
−
0.038
͑
0.034 Hy
͒
+
1.068 33
0.693
0.083
4.99 b
͑
JGT − 0.004
͑
͒
Te
For E. coli
(11)
Log
͑
͑
͑
͑
BAE
BAE
͒
͒
͒
= 2.193
= 1.988
= 2.259
͑
͑
͑
0.091
0.107
0.143
͒
͒
͒
+ 0.447
− 0.089
− 0.093
͑
͑
͑
0.233
0.032
0.037
͒
͒
͒
JGT
Hy + 0.898
Ui − 0.104
33
33
0.346
0.552
0.656
0.701
0.099
0.089
0.082
0.079
4.22 a
6.56 b
7.29 b
6.76 b
(12)
(13)
(14)
Log
͑
0.258
0.030
0.141
͒
JGT
Log
Log
BAE
BAE
͒
JGT
͑
͒
Uy + 1.035
͑
0.243
͒
JGT
0.032
33
͒
=
2.304
͑
0.140
͒
+
0.029
͑
0.016
͒
log P
−
͑
0.044
0.044
͒
Ui
−
0.078
͑
͒
Hy
+
+
1.138 33
͑
0.240
Log BAE
0.237
(15)
͑
͒
=
2.277
͑
0.139
͒
+
0.033
͑
0.016
͒
log P
−
0.133
͑
͒
Ui
−
0.081
͑
0.031
͒
Hy
1.167 33
0.720
0.077
5.98 b
͑
͒
JGT − 0.003
͑
0.002
͒
Te
For C. albicans
(16)
(17)
(18)
(19)
Log
Log
Log
Log
͑
͑
͑
͑
BAC
͒
͒
͒
͒
= 3.085
͑
͑
0.168
0.157
͒
͒
− 0.116
͑
͑
0.049
0.016
͒
Ui
log P − 0.190
log P − 0.258
33
33
0.393
0.580
0.737
0.781
0.112
0.101
0.085
0.080
5.66 a
BAC
BAC
BAC
= 3.021
+ 0.045
͒
͑
0.051
͒
Ui
Ui + 0.848
7.62 b
= 2.889
͑
0.146
͒
+ 0.062
͑
0.015
͒
͑
0.047
͒
͑
0.234
0.351
͒
JGT
33
11.50 b
10.96 b
=
2.889
͑
0.146
͒
+
0.062
͑
0.015
͒
log P
−
0.277
͑
0.045
͒
Ui
+
͑
0.160
͑
Mv
Hy
+
+
0.891 33
͑
0.221
Log BAC
0.168
͒
JGT
(20)
͑
͒
=
2.855
͑
0.148
͒
+
0.056
͑
0.016
͒
log P
−
0.268
͑
0.045
͒
Ui
−
0.038
͑
0.034
͒
0.288 33
0.793
0.079
9.13 b
͑
͒
Mv + 1.008
͑
0.242
͒
JGT
For A. niger
(21)
Log
͑
͑
͑
͑
͑
BAA
BAA
͒
͒
͒
͒
͒
= 2.972
= 2.944
= 3.063
͑
͑
͑
0.172
0.163
0.170
0.153
0.162
͒
͒
͒
− 0.079
− 0.158
+ 0.035
͑
͑
͑
0.050
0.059
0.017
͒
͒
͒
Ui
Ui + 0.034
log P − 0.136
log P − 0.209
0.017 log P 0.227
33
33
0.273
0.452
0.435
0.662
0.705
0.115
0.18
2.50
(22)
(23)
(24)
(25)
Log
͑
0.0150XV
0.055
0.051
0.050
͒
3.86 a
3.50 a
7.54 b
6.93 b
Log
Log
Log
BAA
BAA
BAA
͑
͒
Ui
33
0.109
0.092
0.088
= 2.871
͑
͒
+ 0.065
͑
0.017
͒
͑
͒
Ui + 0.910
͑
0.254
͒
JGT
33
=
2.748
͑
͒
+
0.055
͑
͒
−
͑
͒
Ui
+
0.322
͑
0.178
͒
Mv
Hy
+
+
0.951 33
͑
0.245
Log BAA
0.187
͒
JGT
(26)
͑
͒
=
2.711
͑
0.165
0.269
͒
+
0.048
͑
0.018
͒
log P
−
0.216
͑
0.050
͒
Ui
−
0.044
͑
0.037
͒
0.251 33
0.722
0.088
5.80 b
͑
͒
Mv + 1.082
͑
͒
JGT
a P < 0.05.
b P < 0.01.
chloride (0.05 mol) in ether (50 ml). The solution was stirred
for 30 min. The precipitated amide was separated and recrys-
tallized from alcohol.
B. subtilis, Gram negative E. coli and also against fungi C.
albicans and A. niger.
The MIC (µM ml–1) was determined by using serial dilu-
tion technique [27,28] at different concentrations (2.5, 1.25...
0.078 µM ml–1). Salicylic acid was used as reference. The
compounds were dissolved in dimethyl sulfoxide. To the
culture tubes containing 1 ml of medium, 1 ml of test solution
was added under sterile conditions. To all the tubes including
standard and controls, 0.1 ml of the fresh inoculum was
added. The cells were incubated at 37 °C for 24 h (bacteria),
at 30° for 48 h (C. albicans) and at 30 °C for 7 days (A.
niger). The MIC was recorded in each case as the minimum
concentration of compound, which inhibited the growth of
Cinnamide (IIIa) IR:m cm–1 = 3400 (N–H str., 1° amide),
1660–1670 (C=O str., 1° amide), 1610 (C=C str., alkene
disubstituted trans), 1120 (C–N str., aliphatic), 730–740
(C–H bond, aromatic). 1H-NMR (300 MHz, CDCl3):d
ppm = 7.05 (s, 2H, –NH2), 7.5 (m, 5H, Ar–H), 6.62–6.67,
7.54–7.61 (dd, H, CH=CH, J = 0.060).
6.2. Microbiology
The synthesized compounds were evaluated for their in
vitro antimicrobial activity against Gram positive S. aureus,

834
B. Narasimhan et al. / European Journal of Medicinal Chemistry 39 (2004) 827–834
[2] S.P. Ahmed, M. Ahmed, S.U. Kazmi, M. Shahid, S.I. Ahmed, Pak. J.
Pharmacol. 12 (1995) 31–36.
[3] H.S. Lee, Y.J. Ahn, J. Agr. Food Chem. 46 (1998) 8–12.
[4] P.N. Ramanan, M.N. Rao, Indian J. Exp. Biol. 25 (1987) 42–43.
[5] J.M. Ovale, D. Landman, M.M. Zaman, S. Burney, S.S. Sathe, Amer.
J. Chin. Med. 24 (1996) 103–109.
[6] A. Srivastava, Y.N. Shukla, S. Kumar, J. Ethnopharmacol. 67 (1999)
241–243.
[7] B. Narasimhan, U.R. Kothawade, D.S. Pharande, V.K. Mourya,
A.S. Dhake, Indian J. Chem. 42 (B) (2003) 2828–2834.
[8] S.V. Christine, K.G. Rohan, B.R. Ian, J. Gen. Microbiol. 130 (1984)
2845–2850.
tested microorganism. From the MIC values observed, the
intermediate concentrations between MIC values were prepa-
red by serial dilution by using 0.1 µM ml–1 test solution and
the accurate MIC values were determined. The antibacterial
activity was tested by using double strength nutrient broth and
antifungal activity was tested by using Sabouraud Liquid
Medium. 2 All experiments were performed in triplicate.
6.3. Data analysis
[9] R.J. Cremlyn, K. Thandi, R. Wilson, Indian J. Chem. 23 (B) (1984)
94–96.
[10] R.J. Cremlyn, O. Obioran, G. Singh, Indian J. Chem. 25 (B) (1986)
559–561.
[11] V.K. Ahluwalia, N. Kaila, S. Bala, Indian J. Chem. 25 (B) (1986)
663–664.
[12] M.J. Nicto, F.D. Alovero, A.H. Manzo, M.R. Mazzieri, Eur. J. Med.
Chem. 34 (1999) 209–214.
[13] K.E. Bergamann, M.H. Cynaman, J.T. Welch, J. Med. Chem. 39
(1996) 3394–3400.
[14] P. Vicini, F. Zani, P. Cozzini, I. Doytchinova, Eur. J. Med. Chem. 37
(2002) 553–564.
[15] R.M.E. Nino, M.N. Gifford, M.R. Adams, J. Appl. Bacteriol. 80
(1996) 303–310.
6.3.1. QSAR
The QSAR parameter log P was generated from CLOGP
program for windows software (Biobyte version 4.0, Clare-
mont, CA). The other parameters like unsaturation index
(Ui), hydrophilic factor (Hy), mean atomic Van der Waals
volume (Mv), valence zero order molecular connectivity
(0vV) and global topological charge index (JGT) were gene-
3
rated from DRAGON software. Total energy (Te) was
computed by using Hyperchem version 6. 4 The MLR analy-
sis was performed by QSAR-PC (medicinal chemistry re-
gression program) software. 5
[16] S.J. Fuller, S.P. Denyer, W.B. Hugo, D. Pemberton, P.M. Woodcock,
A.J. Buckley, Lett. Appl. Microbiol. 1 (1985) 13–15.
[17] M.L. Carmellino, G. Pagani, M. Pregnolato, M. Terreni, F. Pastoni,
Eur. J. Med. Chem. 29 (1994) 743–751.
Acknowledgements
[18] T. Hashiguchi, T. Itoyama, A. Kodama, K. Uchida, H. Yamaguchi,
Arzneim-Forsch/Drug Res. 47 (1997) 1218–1221.
[19] Y. Wahbi, C. Tournarie, R. Caujolle, M. Payard, M.D. Linas,
J.P. Segula, Eur. J. Med. Chem. 29 (1994) 701–706.
[20] J.W. Lehman (Ed.), Operational Organic Chemistry—A Laboratory
Course, Allyn and Bacon Inc, London, 1981.
The authors thank the Head, Analytical Department, Uni-
versity of Pune, for recording 1H-NMR and the Head, Ana-
lytical Department, C-MET Electronics Ltd., Pune for C, H,
N analysis. One of the authors (B.N.) is grateful to All India
Council of Technical Education, New Delhi for the award of
Junior Research Fellowship.
[21] W. Carruthers (Ed.), Some Modern Methods of Organic Synthesis,
Cambridge University Press, New Delhi, 1993.
[22] B.S. Furniss, B.A. Hannaford, P.W.G. Smith, A.R. Tatchel (Eds.),
Vogels Textbook of Practical Organic Chemistry, ELBS, Singapore,
1989.
[23] L.B. Kier, L.H. Hall, Molecular Connectivity in Chemistry and Drug
Research, Academic Press, New York, 1976.
References
[1] S. Tawata, S. Taira, N. Kobamoto, J. Zhu, M. Ishihara, S. Toyama,
Biosci. Biotech. Biochem. 60 (1996) 909–910.
[24] S.P. Gupta, A.N. Kumar, A.N. Nagazppa, D. Kumar, S. Kumaran, Eur.
J. Med. Chem. 38 (2003) 867–873.
[25] J. Galvez, R. Garcia, M.T. Sababert, R. Soler, J. Chem. Inf. Comput.
Sci. 34 (1995) 520–525.
[26] J. Galvez, R.G. Domenech, D.J. Ortiz, R. Soler, J. Chem. Inf. Comput.
Sci. 35 (1995) 272–284.
[27] C.H. Collins, P.M. Lyne, Microbiological Methods, Butterworth,
London, 1970.
[28] B.G. Mullen, R.T. Decory, T.J. Mitchell, D.S. Allen, C.R. Kinsolving
Vassill St. Georgiev, J. Med. Chem. 31 (1988) 2008–2014.
2 Pharmacopoeia of India, Ministry of Health Department, Government of
India, New Delhi, 1996.
3 DRAGON 5 Evaluation Version, Talete SRL, 13-20124, Milano, Italy.
4 Hyperchem Version 6, Hypercube Inc., Gainesville, FL 32601, USA.
5 SYSTAT 5.02 for Windows. SYSTAT Inc., 1800 Sherman Ave., Evans-
ton, IL 60201.