Synthesis, pharmacological activity and hydrolytic behavior of ethylenediamine and benzathine conjugates of ibuprofen

Article abstract of DOI:10.1016/j.ejmech.2007.10.028

For reducing the gastrointestinal toxicity associated with ibuprofen, its carboxylic group was masked by synthesizing its amide conjugates with ethylenediamine and benzathine (4a, 4b, respectively) by carbodiimide assisted coupling method. In vitro hydrolysis of conjugates showed that they were stable in HCl buffer (pH 1.2) indicating that the prodrugs did not break in stomach and there was no release of ibuprofen at gastric pH, whereas in phosphate buffer (pH 7.4) they undergo significant hydrolysis and thus release ibuprofen in adequate amounts following first order kinetics. The ibuprofen conjugates 4a, 4b were retaining anti-inflammatory activity intact and exhibited better analgesic activity along with much reduced ulcerogenicity. These findings suggested that both the conjugates are better in action as compared to parent drug and are advantageous in having less gastrointestinal side effects. Compound 4b however showed better analgesic activity and longer action (t_1/2) than 4a, and hence it could be considered as a better candidate for prodrug among the two.

Full text of DOI:10.1016/j.ejmech.2007.10.028

Available online at www.sciencedirect.com
European Journal of Medicinal Chemistry 43 (2008) 2819e2823
http://www.elsevier.com/locate/ejmech
Original article
Synthesis, pharmacological activity and hydrolytic behavior of
ethylenediamine and benzathine conjugates of ibuprofen
*
Nithar R. Chatterjee , Amol A. Kulkarni, Santosh P. Ghulekar
Dr. D.Y. Patil Institute of Pharmaceutical Sciences and Research, Pharmaceutical Chemistry, Pimpri, Pune 411018, Maharashtra, India
Received 29 November 2006; received in revised form 5 October 2007; accepted 5 October 2007
Available online 18 December 2007
Abstract
For reducing the gastrointestinal toxicity associated with ibuprofen, its carboxylic group was masked by synthesizing its amide conjugates
with ethylenediamine and benzathine (4a, 4b, respectively) by carbodiimide assisted coupling method. In vitro hydrolysis of conjugates showed
that they were stable in HCl buffer (pH 1.2) indicating that the prodrugs did not break in stomach and there was no release of ibuprofen at gastric
pH, whereas in phosphate buffer (pH 7.4) they undergo significant hydrolysis and thus release ibuprofen in adequate amounts following first
order kinetics. The ibuprofen conjugates 4a, 4b were retaining anti-inflammatory activity intact and exhibited better analgesic activity along
with much reduced ulcerogenicity. These findings suggested that both the conjugates are better in action as compared to parent drug and are
advantageous in having less gastrointestinal side effects. Compound 4b however showed better analgesic activity and longer action (t1/2
than 4a, and hence it could be considered as a better candidate for prodrug among the two.
Ó 2007 Elsevier Masson SAS. All rights reserved.
)
Keywords: NSAIDs; Ibuprofen amides; Diamine; In vitro hydrolysis; Analgesic; Ulcerogenicity
1. Introduction
a generalized systemic effect, which may be manifested after
absorption following intravenous dosing [5]. This type of dam-
age could be prevented if the carboxylic acid functionality be
masked and therefore the use of prodrug has been postulated
as an approach to decrease the GI toxicity due to the direct
contact effect. Some amide conjugates and a few ester deriva-
tives of ibuprofen have been reported with reduced ulcero-
genic tendency [6e11], but the search for a better prodrug
with reduced side effects still continues.
The purpose of this investigation was to synthesize the
diamide conjugates of ibuprofen with ethylenediamine and
benzathine (N,N0-dibenzylethylenediamine) through N,N0-
1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDAC) or
N,N0-dicyclohexyl carbodiimide (DCC) assisted coupling
method and characterize by physicochemical, spectral (UV,
IR, NMR and MS) and elemental analyses in order to establish
their assigned structures. Furthermore, this report also deals
with preliminary pharmacological screening of the above
said conjugates along with their in vitro hydrolysis studies
for determination of half life (t1/2) as well as to evaluate
Gastrointestinal (GI) side effects constitute the most fre-
quent of all the adverse reactions of nonsteroidal anti-
inflammatory drugs (NSAIDs) [1]. Even though ibuprofen is
very potent and widely used among other clinically used
NSAIDs, literature is abundant with its gastric and other
side effects because of the presence of free carboxylic group.
These reactions range in both severity and frequency leading
to GI bleeding, ulceration and hemorrhage [2,3]. The major
factor in the development of GI ulceration and hemorrhage in-
duced by NSAIDs is the inhibition of prostaglandin synthesis,
as the endogenous prostaglandins are known to have cytopro-
tective action on the gastric mucosa [4]. It has also been
accepted that GI lesions produced by NSAIDs are the result
of two different mechanisms: a direct contact effect and
* Corresponding author.
E-mail
amolkulkarni89@rediffmail.com (A.A. Kulkarni).
addresses:
nitharc@rediffmail.com
(N.R.
Chatterjee),
0223-5234/$ - see front matter Ó 2007 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2007.10.028

2820
N.R. Chatterjee et al. / European Journal of Medicinal Chemistry 43 (2008) 2819e2823
effectivity of the diamines employed herein as linker in the
prodrug approach. The rationale behind the use of ethylenedi-
amine and benzathine to mask the free carboxylic group of the
drug temporarily could be substantiated from the recent report
of the effectivity of two carbon chain, i.e. e(CH2)2e as a linker
in case of polymer-drug conjugates for developing novel drug
delivery system because of its structural spacing [12]. Benza-
thine possessing limited water solubility has been used effec-
tively as depot form for providing effective blood level over
longer period of time, as demonstrated by the wide acceptabil-
ity of long acting benzathine penicillin till date.
Winter et al. [14]; analgesic activity was carried out on Swiss
albino mice by following the method of Seigmund et al. [15]
and ulcerogenic studies were carried out by the method of
Cioli et al. [16] on Wistar rats. In vitro hydrolysis studies
of conjugates were carried out by UV method at pH 1.2
and 7.4.
4. Results and discussion
2. Chemistry
4.1. Hydrolysis studies
The synthesis of the title compounds is illustrated in Fig. 1.
The ethylenediamine conjugate of ibuprofen 4a and benza-
thine conjugate of ibuprofen 4b were synthesized by carbodii-
mide assisted coupling method. This involves condensation of
the racemic carboxylic acid 1 and diamines 2a, 2b in the pres-
ence of an activating agent such as EDAC 3a or DCC 3b to
produce the amide conjugates of the acid 4a, 4b from their re-
spective diamines, through formation of the key intermediate
O-acylisourea [13]. The structures of the synthesized com-
pounds were established by their spectral (UV, IR, NMR and
MS) data and elemental analysis. However, the conjugates
appeared to be an enantiomeric mixture, since the parent
drug employed herein was an enantiomeric mixture.
Hydrolysis studies were carried out in aqueous buffer so as
to study whether the prodrugs hydrolyze in aqueous medium
and to what extent, or not, suggesting the fate of the prodrugs
in the system. Hydrolysis kinetics of the synthesized diamide
conjugates 4a and 4b were studied in aqueous buffer solution
at pH 7.4. Under the experimental conditions the target com-
pounds hydrolyzed to release the parent drug as evident by UV
analysis. At constant pH and temperature the reaction dis-
played strict first order kinetics as the kobs was found to be
fairly constant. The data are given in Table 1.
To examine the degradation of diamide conjugates in pH as
that in stomach, pH 1.2 was selected. An assay time of 2 h was
selected, after which time stomach emptying would normally be
effectively complete. The diamide conjugates did not hydrolyze
to release the parent compound suggesting that they are stable at
the gastric pH. At pH 7.4 the diamide conjugates 4a and 4b
hydrolyzed to parent compound indicating that the prodrugs
will undergo hydrolysis in the system easily [17].
3. Biological investigations
The anti-inflammatory activity of the conjugates 4a, 4b
was carried out on Wistar rats by following the method of
RCOOH
+
R'
N
C
N
R''
1
3 a or b
CH2-C6H5
NH
HN (CH2)2
H2N-(CH2)2-NH2
CH2-C6H5
2a
2b
O
R
CH2-C6H5
O
R
N
N
(CH2)2
O
NH
HN (CH2)2
CH2-C6H5
R
R
O
4a
4b
H3C
H3C
N
CH2-CH2-CH2-
3a
R'' = -CH2-CH3
R' =
R=
CH3
H3C
H3C
3b
R', R'' = cyclo C6H11
Fig. 1.

N.R. Chatterjee et al. / European Journal of Medicinal Chemistry 43 (2008) 2819e2823
2821
Table 1
Hydrolysis of ethylenediamine and benzathine conjugates of ibuprofen
conjugates showed analgesic activity comparable to ibuprofen.
The percent protection was calculated using equation
a
Compound
kobs ꢃ SD
Phosphate buffer
(pH 7.4)
t1/2 (min)
Hydrochloric acid
buffer (pH 1.2)
Phosphate buffer
(pH 7.4)
Protection % ¼100 ꢀ ½number of writhings in test=
number of writhings in control ꢁ 100ꢂ
4a
4b
a
0.01 ꢃ 0.007
0.008 ꢃ 0.005
e
e
69.5
82.5
Mean of three sets of experiments.
4.4. Ulcerogenic study
The ulcerogenic effect of ibuprofen and diamide conjugates
4a and 4b was studied at a dose of 100 mg kgꢀ1. It was ob-
served that the ulcerogenic dose for the conjugates was almost
approximately five times the dose of ibuprofen. Less number
of ulcers were seen in animals treated with ethylenediamine
conjugates of ibuprofen compared with the animals treated
with ibuprofen. All the animals treated with benzathine conju-
gates of ibuprofen did not develop ulcers as they did not hy-
drolyze in gastric pH. These findings suggested successful
masking of the carboxylic function of ibuprofen.
4.2. Anti-inflammatory activity
The inhibition of swelling in carrageenan-induced edema in
rat paw brought about by oral administration of the drugs is
shown in Table 2. The percentages of swelling inhibition
were calculated using the equation
Inhibition %
ÈÂ
Ãꢀ
É
¼
ðVt ꢀ V0ÞcontrolꢀðVt ꢀ V0Þtreated ðVt ꢀ V0Þcontrol ꢁ 100
5. Conclusions
wherein, V0 and Vt relate to the average volume in the hind
paw of the rats (n ¼ 6) before any treatment and after anti-
inflammatory agent treatment, respectively.
The diamide conjugates of ibuprofen employing ethylene-
diamine and benzathine were successfully synthesized by car-
bodiimide assisted coupling method and characterized by
spectral (UV, IR, NMR and MS) data and elemental analysis.
The diamide conjugates released ibuprofen quantitatively at
pH 7.4, but were resistant to hydrolysis at pH 1.2 indicating
that the conjugates are resistant to acidic condition and both
were found to be significantly less ulcerogenic showing en-
hanced anti-inflammatory and analgesic activities than the par-
ent drug. Hence diamides having two carbon atom chain as
a linker could be used as effective prodrug moieties for ibupro-
fen. Furthermore, the benzathine conjugate of ibuprofen
showed longer time of action (t1/2) than the corresponding eth-
ylenediamine conjugate, and hence it could be a better prodrug
candidate for ibuprofen than the other one.
The diamide conjugates 4a and 4b showed better activity
compared to the free parent drug. The maximum anti-inflam-
matory activity of diamide conjugates 4a and 4b was observed
at 3 h and remained practically constant upto 6e8 h. The anti-
inflammatory activity of free ibuprofen decreased with time.
Statistical significance testing using one way analysis of vari-
ance showed that the anti-inflammatory activities of the parent
drug and its diamide conjugate were effective in comparison
with the control group. However, differences in the potency
of anti-inflammatory activity of the diamide conjugates com-
pared to the free ibuprofen were observed over a long period.
Thus diamide conjugates were proved to be suitable promoi-
eties for ibuprofen prodrugs.
6. Materials and methods
4.3. Analgesic activity
All reagents were obtained from Loba Chem (India) Ltd.
Ibuprofen was procured from Sun Pharmaceuticals. All the
solvents used in these studies were dried and distilled before
use. Wistar rats of either sex weighing between 150 and
200 g and Swiss albino mice of either sex weighing between
25 and 30 g were procured from Animal House of Dr. D.Y. Pa-
til Institute of Pharmaceutical Sciences and Research, Pune.
The percent protection in mice brought about by adminis-
tration of the drug is shown in Table 2. The diamide
Table 2
Pharmacological profile of ethylenediamine and benzathine conjugates of
ibuprofen
Compound Oral
dose,
Anti-inflammatory activity Analgesic
(% inhibition of edema)a activity
Ulcer indexb
6.1. Characterization of compounds
mg kgꢀ1
(% analgesia)a
3 h
24 h
Melting points (mp): Veego VMP-PM digital melting point
apparatus, uncorrected. UV spectra: Shimadzu Pharmspec
1700, UVeVIS spectrophotometer. IR spectra: Shimadzu
8400 S FT-IR. H NMR spectra: 300 MHz JEOL NMR Spec-
trophotometer. The EI mass spectra were obtained from Pune
University, Maharashtra, India.
Ibuprofen 20
52.15
48.03
53.48
35.65
38.03
32.37
23.42
29.63
35.84
13.51 ꢃ 0.47
4.63 ꢃ 0.41
Nil
ECI
BCI
a
21.15
29.98
1
Statistical analysis was performed with ANOVA followed by t-test,
P < 0.001.
b
Dose: 100 mg kgꢀ1 for ulcerogenic activity.

2822
N.R. Chatterjee et al. / European Journal of Medicinal Chemistry 43 (2008) 2819e2823
6.2. 2-(4-Isobutylphenyl)-N-{2-[2-(4-isobutylphenyl)
propionylamino]ethyl}propionamide 4a
and phosphate buffers. The total buffer concentration was
0.05 M and constant ionic strength of 0.5 M for each sample
was maintained by adding KCl. The total buffer volume was
900 ml. The mixture was equilibrated at 37 ꢄC for 1 h. To
this mixture 100 mg of each sample dissolved in alcohol
was added separately and the mixture agitated by an overhead
stirrer. At selected time intervals of 15, 30, 45, 60, 75, 90, 105
and 120 min, 5 ml of the solution (at pH 1.2) was removed and
transferred to a separating funnel containing chloroform. Free
parent drug supposed to be released after hydrolysis was esti-
mated by UV from the chloroform extract by the following
usual procedure. The samples were withdrawn at time inter-
vals of 0.5, 1, 2, 3, 4, 5, 6, 7, 8, 9 and 10 h at pH 7.4 and
free drug released was estimated. The rate of reaction and hy-
drolysis constant were calculated with the help of the equation,
k ¼ 2.303/t ꢁ log(a/a ꢀ x), where, ‘k’ is the hydrolysis con-
stant, ‘t’ is time in minutes, ‘a’ is initial concentration of the
drug, ‘x’ is the amount of drug hydrolyzed and ‘a ꢀ x’ is the
amount of drug remaining.
A solution of 3a (2.67 g, 14 mmol) in 8 ml dimethylforma-
mide was added to cooled stirring solution of 1 (2.88 g,
14 mmol) in 50 ml chloroform at 0e2 ꢄC and kept stirring for
2 h. To the above stirring solution, 2a (0.46 ml, 7 mmol) was
then slowly added maintaining the temperature 0e2 ꢄC and
the reaction mixture was then stirred for a further period of
2 h during which it attains room temperature. After completion
of reaction (TLC), it was successively washed twice with 5%
NaHCO3 (10 ml ꢁ 2) followed by water wash (15 ml ꢁ 2).
The organic layer was separated and dried over anhydrous
Na2SO4 followed by usual work-up to give a semisolid mass,
which on trituration with pet. ether gave off-white solid product
in about 61% yield. A portion of this product on recrystallization
from acetone and ethanol mixture gave 4a as white solid melting
at 137e139 ꢄC; Rf: 0.40 in benzene:ethyl acetate (2:1). UV
(CHCl3): lmax at 244 nm (log 3 3.35). IR (KBr): 3315, 2952,
1645, 1537, 1365, 1232, 848 cmꢀ1
;
1H NMR (300 MHz,
The initial concentration of each of the diamide conjugates
was 100 mg/900 ml. The half life was calculated using the for-
mula, t1/2 ¼ 0.693/k for first order reaction.
CDCl3) d 0.9 (s, 12H, 4CH3), 1.42 (d, 6H, 2CH3), 1.73 (m, 2H,
2CeH), 2.57 (d, 4H, 2CH2), 3.42 (s, 4H, CH2eCH2), 3.66 (s,
2H, 2COeCH), 4.46 (br s, 2H, 2NH, D2O exch.), 6.9e7.4 (m,
8H, Ar-H) ppm. MS (EI): m/z 437 (m þ 1), 276, 248, 188, 161
(100%), 91, 58, 28; calc. for C28H40N2O2: C-77.02%, H-9.23%
and N-6.42%, found: C-76.63%, H-9.16% and N-6.35%.
6.5. Anti-inflammatory activity
The anti-inflammatory activity was evaluated using carra-
geenan-induced paw edema method on rat [14]. Wistar rats
(150e200 g) were divided into four groups of six animals
each. Group I served control group without using the drug,
group II received ibuprofen (20 mg kgꢀ1), groups III and IV
received diamide conjugates 4a and 4b (21.15 and
29.98 mg kgꢀ1, respectively) where the dose was molecularly
equivalent to ibuprofen. A stock solution of 4, 4.23 and
5.99 mg mlꢀ1 was prepared as a homogeneous suspension in
aqueous solution of sodium CMC (0.5% w/v) and each animal
received 0.75e1.0 ml of the respective drugs orally. After
30 min of administration of drug, each rat received a sub
planter injection of 0.1 ml of 1% carrageenan solution in its
left hind paw. The measurement of the hind paw volume
was carried out using an Ugo Basile Plethysmometer before
any treatment (V0) and in any interval (Vt) after the administra-
tion of the drugs. All the results are expressed as mean -
ꢃ S.E.M. Statistical evaluation was performed using analysis
of variance followed by t-test for sub group comparison [18].
6.3. N-Benzyl-N-(2-{benzyl[2-(4-isobutylphenyl)
propionyl]amino}ethyl)-2-(4-isobutylphenyl)-
propionamide 4b
To a cooled stirring solution of 1 (2.88 g, 14 mmol) in 50 ml
chloroform was slowly added 3a (14 mmol) in DMF (8 ml) and
the mixture was kept stirring for 2 h at 0e2 ꢄC. Benzathine
base (prepared from its diacetate 3.75 g by treatment with
dil. NaOH followed by extraction with CHCl3 and usual
work-up) was added to the above stirring solution at 0e2 ꢄC
and stirring continued for 3 h more followed by 5% NaHCO3
and water wash. After usual work-up as mentioned in the ear-
lier experiment, the product was obtained as semisolid mass,
which on trituration with hexane was kept overnight in freezer
to afford off-white solid in 58% yield. A portion of the above
solid was recrystallized from acetone and ethanol to give 4b
as white solid melting at 89e91 ꢄC; Rf: 0.50 in benzene:ethyl
acetate (2:1). UV (CHCl3): lmax at 260 nm (log 3 2.94). IR
(KBr): 2960, 1637, 1460, 1282, 875 cmꢀ1
;
1H NMR
6.6. Analgesic activity
(300 MHz, CDCl3) d 0.9 (s, 12H, 4CH3), 1.2 (d, 6H, 2CH3),
1.8 (m, 2H, 2CeH), 2.3 (d, 4H, 2CH2), 2.4 (s, 4H), 3.5 (br s,
2H, 2COeCH), 2.6 (t, 4H), 6.9e7.4 (m, 18H, Ar-H) ppm;
MS (EI): m/z 617 (m þ 1), 526, 435, 246, 161, 91 (100%),
57, 28; calc. for C42H52N2O2: C-81.78%, H-8.50% and N-
4.54%, found: C-81.70%, H-8.45% and N-4.50%.
Analgesic activity was carried out by using acetic acid in-
duced writhing method [15] in Swiss albino mice (25e30 g)
of either sex. A 1% v/v solution of acetic acid was used to in-
duce writhing. Test compounds were administered orally 3 h
prior to acetic acid injection. Number of writhings for 10 min
in control and test compounds were counted and compared. An-
algesic activity was measured as percent decrease in writhings
in comparison to control. Mice were divided into four groups of
six animals each. Group I served as a control group receiving
6.4. Hydrolysis study in aqueous buffers
The hydrolysis kinetics of diamide conjugates 4a and 4b
was studied at pH 1.2 and 7.4 using hydrochloric acid buffers
vehicle 5 ml kgꢀ1
, while group II received ibuprofen

N.R. Chatterjee et al. / European Journal of Medicinal Chemistry 43 (2008) 2819e2823
2823
(20 mg kgꢀ1), groups III and IV received diamide conjugates
4a and 4b (21.15 and 29.98 mg kgꢀ1, respectively), where the
dose of conjugate was molecularly equivalent to ibuprofen. A
stock solution of 1, 1.05 and 1.49 mg mlꢀ1 was prepared as
a homogeneous suspension in aqueous solution of sodium
CMC (0.5% w/v) and each animal received 0.5e0.6 ml of the
respective drugs orally. Acetic acid was administered intraper-
itoneally 1 ml/100 g body weight of the animal. All the results
are expressed as mean ꢃ S.E.M. Statistical evaluation was
performed using analysis of variance followed by t-test for
sub group comparison (level of significance is P < 0.001).
gastric damage (level of significance is P < 0.05 with respect
to control).
Acknowledgements
The authors are thankful to Director Principal Dr. A.D.
Deshpande for providing the facilities to carry out this
research work. The authors are thankful to Chemistry Depart-
ment of Pune University, Pune, Maharashtra, India, for provid-
ing all necessary spectral data.
References
6.7. Ulcerogenicity
[1] A.L. Blower, C.P. Armstrong, Br. J. Surg. 74 (1987) 759e761.
[2] K.D. Rainsford, Toxicol. Pathol. 16 (1998) 251e254.
[3] D. Bhosale, S. Bharambe, N. Gairola, S.S. Dhaneshware, Indian J.
Pharm. Sci. 3 (2006) 286e294.
Gastrointestinal toxicity was determined by the method of
Cioli et al. [16]. The studies were carried out on adult Wistar
rats (150e200 g) at the dose of 100 mg kgꢀ1. The animals
were divided into four groups of six each, group I served as
control and received vehicle only. Group II received pure ibu-
profen 100 mg kgꢀ1. Group III received test compound 4a and
group IV received test compound 4b in a dose molecularly
equivalent to 100 mg kgꢀ1 of ibuprofen. The animals were
fasted 8 h prior to administration of each of the control and
test compounds. The animals were sacrificed 17 h after drug
administration and food and water were available ad libitum.
The gastric mucosa of the rats was examined by means of
a 4ꢁ binocular magnifier. The lesions were counted and
divided into large (greater than 2 mm in diameter), small
(1e2 mm in diameter) and punctiform (less than 1 mm in
diameter). For each stomach, the severity of mucosal damage
was assessed according to the following scoring system:
[4] I.G. Otterness, M.L. Bliven, J.G. Lombardino, Nonsteroidal Anti-
inflammatory Drugs, vol. 11, Wiley, New York, 1985.
[5] H. Bundgaard, N.M. Nielson, A. Butt, Int. J. Pharm. 44 (1988) 154e157.
[6] V.R. Shanbag, M.A. Crinder, R. Gokhale, A. Harpalini, R.M. Dick, J.
Pharm. Sci. 81 (1991) 149e158.
[7] W.S. Saari, M.B. Freedam, R.D. Hartman, S.W. King, A.W. Raab,
W.C. Randa, R. Hitschmann, J. Med. Chem. 21 (1978) 746e750.
[8] M.S.Y. Khan, M. Akther, Eur. J. Med. Chem. 40 (2005) 371e376.
[9] M.T. Cocco, C. Congiu, V. Onnis, M. Morelli, O. Cauli, Eur. J. Med.
Chem. 38 (2003) 513e518.
[10] N.M. Mahfouz, F.A. Omar, T.A. Fadl, Eur. J. Med. Chem. 34 (1999)
551e562.
[11] N.L. Moller, K.E. Dietzel, H.T. Schneider, K. Brune, Eur. J. Med. Chem.
24 (1989) 463e469.
[12] R.J. Bergeron, G. Haung, J. Mcmanis, H. Yao, J. Nguyen, J. Med. Chem.
(2005) 3099.
[13] N.J. Delgado, W.A. Remers, Wilson and Gisvold’s A Textbook of
Organic Medicinal and Pharmaceutical Chemistry, eleventh ed.,
2004, pp. 753e755.
- 0 e no lesions or upto five punctiform lesions;
- 1 e more than five punctiform lesions;
- 2 e 1e5 small ulcers;
- 3 e more than five small ulcers or one large ulcer;
- 4 e more than one large ulcer.
[14] C.A. Winter, E.A. Risley, G.W. Nuss, Proc. Soc. Exp. Biol. Med. (1962)
544e549.
[15] E. Seigmund, R. Cadmus, G. Lu, Proc. Soc. Exp. Biol. Med. 95 (1957)
729e733.
[16] V. Cioli, P. Putzolu, S. Rotzolu, V. Rossi, P.S. Barcellona, C. Corradino,
Toxicol. Appl. Pharmacol. 50 (1979) 283e287.
[17] N.M. Nielson, H. Bundgaard, J. Pharm. Sci. 77 (1988) 77e85.
[18] G.H. Vogel, W.H. Vogel, Drug Discovery and Evaluation of Pharmaco-
logical Assays, Springer-Verlag, Berlin, 1997, pp. 445e449.
The mean score of each treated group minus the mean score
of the control group was considered as the ‘severity index’ of