Synthesis of new pyrrolo[1,2-d][1,2,4]triazines and thiazolo[3,4- d][1,2,4]triazines as immunostimulating agents

Article abstract of DOI:10.1016/S0968-0896(97)10052-9

Four pyrrolo[1,2-d][1,2,4]triazines and four thiazolo[3,4- d][1,2,4]triazines were synthesized from trans-4-hydroxy-L-proline and L- thiaproline, respectively. The synthetic route involved formation of hydrazides followed by cyclization with orthoesters. The proliferative response to human lymphocyte mitogen (phytohemagglutinin) revealed significant immunostimulant activity for all test drugs. Furthermore, some triazine derivatives were effective to activate production of free oxygen radical by phagocytes in response to stimulation by opsonized zymosan.

Full text of DOI:10.1016/S0968-0896(97)10052-9

BIOORGANIC &
MEDICINAL
CHEMISTRY
Bioorganic & Medicinal Chemistry 6 (1998) 349±354
Synthesis of New Pyrrolo[1,2-d][1,2,4]triazines and
Thiazolo[3,4-d][1,2,4]triazines as Immunostimulating Agents
Veronique Issartel,a Veronique Spehner,b Pascal Coudert,a Estelle Seillesb
and Jacques Couqueleta*
a
Â
Groupe de Recherche en Pharmacochimie, Faculte de Pharmacie, 28 Place Henri Dunant, 63001 Clermont-Ferrand, France
b
  Â
Equipe de Chimie Therapeutique, Faculte de Medecine et de Pharmacie, Place St Jacques, 25030 Besancon, France
,
Received 22 May 1997; accepted 27 October 1997
AbstractÐFour pyrrolo[1,2-d][1,2,4]triazines and four thiazolo[3,4-d][1,2,4]triazines were synthesized from trans-4-
hydroxy-l-proline and l-thiaproline, respectively. The synthetic route involved formation of hydrazides followed by
cyclization with orthoesters. The proliferative response to human lymphocyte mitogen (phytohemagglutinin) revealed
signi®cant immunostimulant activity for all test drugs. Furthermore, some triazine derivatives were e€ective to activate
production of free oxygen radical by phagocytes in response to stimulation by opsonized zymosan. # 1998 Elsevier
Science Ltd. All rights reserved.
Introduction
sole).7,8 This one is ecacious in immunode®cient indi-
viduals presumably through an action of its sulphur
moiety to induce a thymic hormone-like factor.1 The
toxicity of levamisole, however, reduced the maximum
prescribed dose of this agent giving rise to the develop-
ment of a research program for the synthesis of chemi-
cally-related compounds.5,8±12 In view of the
immunoregulatory actions on human T lymphocytes of
previously synthesised pyrrolopyrimidinediones and
thiazolohydantoins,13,14 we decided to extend our
investigations to some structurally related derivatives.
We prepared a series of pyrrolo and thiazolotriazines in
order to evaluate the e€ect on activity of replacing the
pyrimidine or the hydantoin ring by a triazine nucleus,
while maintaining the bridgehead atom present in the
aforementioned immunoagents.
The clinical need for therapeutic agents able to restore a
normal immune response in immunocompromised
patients has led to the discovery of a number of sub-
stances collectively de®ned as immunomodulators.1,2
Stimulation of the nonspeci®c immune response may be
expected to signi®cantly assist patients a€ected with
infectious diseases and cancers. Presently, whereas nat-
ural macromolecular immunostimulants (bacterial,
extracts, cytokines, grow factors) are being increasingly
used in human medicine, only a small number of syn-
thetic immunostimulants are ecient in clinical prac-
tice.3 Microbial products and cytokines tend to be
pleiotropic, in¯uencing several parts of functions of the
immune system. In contrast chemical immunomodula-
tors could theoretically be prepared with selective e€ects
acting at speci®c points in the immune response. One
promising issue is to discover compounds acting on T
lymphocytes either by enhancing T cell mediated acti-
vation or by favoring T cell di€erentiation.4,6 The ®rst
synthetic compound acting by stimulation of T cell
function in vivo was an imidazothiazole derivative, the
levogyre enantiomeric form of tetramisole (i.e., levami-
Chemistry
The e€ects of modi®cation of the diazabicyclooctane
ring system on biological potency were examined
through two series of compounds prepared as described
in Scheme 1. Esters 1a and 1b were obtained from trans-
4-hydroxy-l-proline and l-thiaproline, respectively,
according to previously reported methods.15,16 The
action of hydrazine hydrate on 1 furnished hydrazides 2
*Corresponding author.
0968-0896/98/$19.00 # 1998 Elsevier Science Ltd. All rights reserved
PII: S0968-0896(97)10052-9

350
V. Issartel et al./Bioorg. Med. Chem. 6 (1998) 349±354
in good yields. Cyclization of 2 with an appropriate
orthoester a€orted the target pyrrolotriazines 3 and
thiazolotriazines 4. Stereochemistry of compounds 3
and 4 was established from NOE experiments. Irradia-
tion of the H7 proton caused an increase of the intensity
of both the H6b and H8b protons of 3. In addition irra-
diation of H8a concomitantly caused an increase of
e€ects on free oxygen radical (FOR) production by
phagocytes in response to stimulation by opsonized
yeast stimulus (zymosan).18±20
Proliferative response: MTT assay
Brie¯y, a mitogenic stimulation assay using T cell mito-
gen phytohemagglutinin (PHA) was performed with
peripheral blood leukocytes taken from healthy donors.
Proliferating cells were evaluated using a quantitative
colorimetric assay with 3-(4,5-dimethyl-thiazol-2-yl)-
2,5-diphenyl tetrazolium bromide (MTT). This colori-
metric assay converts yellow tetrazolium salt to its blue
formazan derivative via the mitochondrial enzymes in
viable cells. Assays for compounds and for levamisole,
used as reference, were performed on leukocytes from
ten blood donors. The responses to PHA are shown in
Figure 1. Proliferative activity induced by levamisole,
was signi®cantly increased in all concentrations tested
when compared to that of control (p=0.004). Similarly
all compounds, whatever the concentrations used, had
signi®cant positive e€ects on the proliferation of lym-
phocytes when compared to control (p<0.05). In addi-
tion several compounds signi®cantly (*) increased
proliferation as compared to levamisole (p<0.05).
These e€ects were particularly remarkable for the fol-
lowing derivatives: 3a, 4a, 4b at 50 mg/mL and 3a, 3b, 4a
at 100 mg/mL. At the concentrations used (between 1
H8b, H8a, and OH. At last when H8b was irradiated,
NOE was observed at the signals of H7 and H8a. Based
on the foregoing evidence, the stereochemistry of tria-
zines 3 was concluded to be 7R and 8aS (Scheme 1).
Similar NOE e€ects were observed in the case of
thiazolo derivatives 4 supporting a con®guration R for
the carbon 8a.
Physical constants and spectral data of 3 and 4 are
reported in Tables 1 and 2, respectively. For 13C NMR,
all spectra of bicyclic compounds showed a signal about
162 ppm attributable to the C=0 group.
Biological Results
The biological activity of the triazine derivatives 3 and 4
was ®rst determined by the ability of these compounds
to stimulate the proliferative response of lymphocytes to
a mitogen.17 In a second time, compounds 3 and 4 were
tested for immunostimulant activity by checking the
X
X
COOCH3
X
COOH
CONHNH2
SOCl2
NH2NH2
N
H
N
H
N
H
CH3OH
1a: X = CHOH
1b: X = S
2a,b
O
Hβ
H
Hα
H
H
1
8
2N
3N
8a
7
5
6
4
HO
Hβ
Hα
R
3a−d
R-C(OEt)3
R = H, Me, Et, Ph
O
Hβ
H
Hα
H
1
8
2N
3N
8a
7
S
5
6
4
Hβ
Hα
R
4a−d
Scheme 1.

V. Issartel et al./Bioorg. Med. Chem. 6 (1998) 349±354
351
and 100 mg/mL), the stimulating e€ects on the pro-
liferative response varied only slightly (p>0.05). None
of the compounds activities was dose-dependent.
only three compounds had signi®cant e€ects: 3b at
100 mg/mL (p=0.009), 4b at 10 mg/mL (p=0.02), and 4d
at 10 mg/mL (p=0.01). No compound had greater e€ect
than Biostim1 on FOR production.
Chemiluminescence assay
Assays for compounds and Biostim1, used as the refer-
ence, were performed on blood phagocytes from nine
donors. The oxidative response measured by evaluating
FOR production and induced only by zymosan (con-
trol) or by zymosan and compounds is shown in Figure
2. Only the peak value of CL at the concentration with
the greatest oxidative response is shown as the median.
Discussion
Results on the proliferative response of blood lympho-
cytes showed that test drugs 3 and 4 had greater
immunostimulating activity than did the reference drug,
levamisole at the same doses. They all provided potent
stimulation indexes in this model with minimum e€ec-
tive doses of 1 mg/mL. Although derivatives 3 repre-
sented an attempt to assess the contribution to activity
of the sulphur atom of the thiazolotriazines, there was
no signi®cant di€erence between the two chemical series.
Moreover, the nature of the substituent in the 4-position
FOR production induced by Biostim1 was signi®cantly
increased in all concentrations tested when compared to
that of control (p<0.001). Except for 4a, all compounds
increased FOR production as compared to control. But
Table 1. Physical constants of pyrrolo-triazines 3a±d and thiazolo-triazines 4a±d
Compound
R
Formula
M.W.
Yield (%)
mp‹1 (ꢀC)
3a
3b
3c
3d
4a
4b
4c
4d
H
C6H9O2N3
C7H11O2N3
C8H13O2N3
C12H13O2N3
C5H7ON3S
C6H9ON3S
C7H11ON3S
C11H11ON3S
155.15
169.18
183.21
231.25
157.19
171.21
185.24
233.28
33
54
25
12
15
54
33
9
191
186
182
226
126
130
117
135
CH3
C2H5
C6H5
H
CH3
C2H5
C6H5
Table 2. Spectral data of pyrrolo-triazines 3a±d and thiazolo-triazines 4a±d
1
Compound
IR n (cm
C=O; C=N
)
1H NMR (DMSO-d6)-d (ppm)
3a
1665; 1635
1.85 (1H, ddd, J=13, 11, 5 Hz, C8-H), 2.0 (1H, dd, J=13, 6 Hz, C8-H), 3.2 (1H, dd, J=11, 2 Hz,
C6-H), 3.75 (1H, dd, J=11, 5 Hz, C6-H), 4.0 (1H, dd, J=11, 6 Hz, C8a-H), 4.3 (1H, m, C7-H), 5.1
(1H, s, OH), 7.0 (1H, s, C4-H), 10.2 (1H, s, NH)
3b
3c
1670; 1625
1660; 1610
1.85 (1H, m, C8-H), 1.9 (3H, s, CH3), 2.0 (1H, dd, J=13, 6 Hz, C8-H), 3.2 (1H, d, J=11 Hz, C6-
H), 3.8 (1H, dd, J=11, 5 Hz, C6-H), 3.95 (1H, dd, J=11, 6 Hz, C8a-H), 4.3 (1H, m, C7-H), 5.1 (1H,
s, OH), 10.1 (1H, s, NH)
1.1 (3H, t, J=7 Hz, CH3), 1.9 (1H, ddd, J=13, 11, 5 Hz, C8-H), 2.0 (1H, dd, J=13, 6 Hz, C8-H),
2.2 (2H, m, CH2), 3.2 (1H, d, J=11 Hz, C6-H), 3.8 (1H, dd, J=11, 5 Hz, C6-H), 3.95 (1H, dd,
J=11, 6 Hz, C8a-H), 4.3 (1H, m, C7-H), 5.1 (1H, s, OH), 10.1 (1H, s, NH)
2.2 (2H, m, C8-H), 3.3±3.5 (2H, m, C6-H), 4.2 (2H, m, C8a-H and C7-H), 5.2 (1H, s, OH), 7.4±7.6
(5H, m, C6H5), 10.6 (1H, s, NH)
3d
4a
4b
4c
1665; 1610
1660; 1630
1670; 1635
1670; 1630
2.8 (1H, t, J=10 Hz, C8-H), 3.4 (1H, dd, J=10, 7 Hz, C8-H), 3.9 (1H, dd, J=10, 7 Hz, C8a-H), 4.3
(1H, d, J=9 Hz, C6-H), 4.9 (1H, d, J=9 Hz, C6-H), 7.0 (1H, s, C4-H), 10.7 (1H, s, NH)
1.95 (3H, s, CH3), 2.9 (1H, t, J=10 Hz, C8-H), 3.4 (1H, dd, J=10, 7 Hz, C8-H), 3.8 (1H, dd,
J=10, 7 Hz, C8a-H), 4.2 (1H, d, J=9 Hz, C6-H), 4.9 (1H, d, J=9 Hz, C6-H), 10.5 (1H, s, NH)
1.0 (3H, t, J=7Hz, CH3), 2.3 (2H, m, CH2), 2.8 (1H, t, J=10 Hz, C8-H), 3.3 (1H, dd, J=10, 7 Hz,
C8-H), 3.9 (1H, dd, J=10, 7 Hz, C8a-H), 4.2 (1H, d, J=9 Hz, C6-H), 4.9 (1H, d, J=9 Hz, C6-H),
10.5 (1H, s, NH)
4d
1665; 1610
3.4 (2H, m, C8-H), 4.2 (1H, m, C8a-H), 4.45 (1H, d, J=9 Hz, C6-H), 4.55 (1H, d, J=9 Hz, C6-H),
7.5 (5H, m, C6H5), 11.0 (1H, s, NH)

352
V. Issartel et al./Bioorg. Med. Chem. 6 (1998) 349±354
of the triazine ring did not signi®cantly in¯uence biolo-
gical e€ects; therefore compounds 3c and 4c were not
tested.
compound 4b only in its lack of substituent on the tri-
azine ring, exhibited a markedly di€erent e€ect since it
induced a decrease of FOR production when compared
to control.
In contrast, results for the oxidative response of blood
phagocytes demonstrated a di€erence between the two
series of derivatives. In the case of pyrrolotriazines, 3b
with a methyl group induced a signi®cant increase when
compared to the activity of the unsubstituted analogue
3a and to that of 3d, possessing a phenyl nucleus. Con-
sidering thiazolotriazine, methyl compound 4b and
phenyl derivative 4d increased signi®cantly the production
of FOR. Surprisingly, 4a di€ering from the parent
More precise interpretation of the structure activity
relationships is dicult on account of the limited num-
ber of molecules tested. In previously reported
works,13,14 it was postulated that a dioxo-group as well
as two ®ve-membered rings in the bicyclic system were
necessary to improve the immunomodulating activity.
In view of the last data, it appears that the bicyclic
skeleton tolerates much more variations than we
Figure 1. Proliferative response: stimulation index (SI) of human lymphocytes stimulated by PHA and incubated with and without
(control) compounds (median values of 10 assays).
Figure 2. Oxidative response: chemiluminescence index (CLI) of human phagocytes stimulated by opsonized zymosan and induced
with and without (control) compounds.

V. Issartel et al./Bioorg. Med. Chem. 6 (1998) 349±354
353
supposed, which justi®ed the synthesis and the investi-
gation of additional series to specify the structural features
contributing to potent immunostimulating activity.
1.7 (1H, ddd, J=13, 8.5, 5 Hz, C3-H), 1.95 (1H, dd,
J=13, 8 Hz, C3-H), 2.7 (1H, d, J=12 Hz, C5-H), 2.9
(1H, dd, J=12, 4 Hz, C5-H), 3.7 (1H, t, J=8 Hz, C2-H),
3.7±4.1 (4H, b, OH, NH, NH2), 4.2 (1H, m, C4-H), 8.9
(1H, br s, NH-NH2).
None of the compounds activities was dose-dependent
for the two tests. However, it is possible to state that the
two series of compounds were immunostimulants for
the proliferation of lymphocytes and for FOR produc-
tion of phagocytes whatever dose used. This result is
particularly attractive, considering that in closely related
chemical series it was observed simultaneous immuno-
suppressive and immunostimulating properties accord-
ing to concentration of products.9,11 Due to their potent
activity in the two tests used, triazinones 3b and 4b
appeared as the most promising compounds for further
pharmacological and toxicological studies. In addition
to the uncertainty relative to in vitro studies, in vivo
experiments are required to con®rm interest of these
derivatives. Obviously the in vivo activity will be a
re¯ection not only of the inherent activity of the com-
pounds but pharmacokinetic/pharmacodynamic para-
meters as well.
4-Hydrazido-thiazolidine (2b). A mixture of 1b (2.5 g,
17 mmol), hydrazine hydrate (1 g, 20 mmol) and ethanol
(10 mL) was stirred for 24 h and then evaporated. The
crude material was chilled in an ice bath. The resulting
crystals were collected and puri®ed by recrystallization
from hexane to a€ord 2.17 g (87%) of 2b; bp 77ꢀC. IR
1
(KBr) 3280, 1650 cm
;
1H NMR (DMSO-d6) d 2.8
(1H, dd, J=10, 7 Hz, C5-H), 2.95 (1H, dd, J=10, 7 Hz,
C5-H), 3.15 (1H, br s, NH), 3.65 (1H, t, J=7Hz, C4-H),
4.0 (1H, d, J=9 Hz, C2-H), 4.3 (1H, d, J=9 Hz, C2-H),
4.25 (2H, b, NH2), 9.2 (1H, s, NH-NH2).
7 -Hydroxy-4 -methyl-1-oxo-6,7,8,8a -tetrahydro-2H-
pyrrolo[1,2-d][1,2,4]triazine (3a). To a solution of 2a
(1.02 g, 7 mmol) in DMF (10 mL) was added triethyl
orthoformiate (1.25 g, 7.7 mmol). The mixture was
re¯uxed for 8 h and evaporated to dryness. The resulting
solid was recrystallized in methanol to obtain 3a as
white plates.
Experimental
Chemistry
Compounds 3b, 3c, and 3d were prepared similarly to 3a
with appropriate triethyl orthoesters, but derivatives 3c
and 3d were puri®ed by column chromatography
(eluent:ethylacetate:methanol, 9:1).
Melting points were determined on a Ko¯er apparatus
without correction. Infrared (IR) spectra were recorded
on a Beckman 4240 spectrophotometer in KBr disks.
1
All H and 13C spectra were recorded on a Bruker AC
1-Oxo-5,6,8,8a-tetrahydro-2H-thiazolo[3,4-d][1,2,4]tri-
azines (4). Compounds 4 were prepared in the same
manner as described for 3, using 2b as starting material
and heating for 6 h. They were puri®ed by column
chromatography with ethyl acetate as eluent.
400 (400 MHz) spectrophotometer using tetra-
methylsilane as an internal standard. The coupling con-
stants are in Hertz (Hz) and the chemical shifts are
reported in parts per million (d, ppm). The following
abbreviations are used : s=singlet, d=doublet, t=tri-
plet, b=band, m=multiplet, br s=broad singlet. Ele-
mental analyses were performed at the Service Central
d'Analyses, Centre National de la Recherche Scienti®-
que, 69390 Vernaison, France. Analytical results
obtained for all compounds were within ‹0.4% of the-
oretical values. For analytical TLC, plastic sheets
coated with a 0.2 mm layer of silica-gel 60 F254 (Merck)
were used and spots were visualized with UV light.
When required, the separation of crude reaction pro-
ducts was achieved by chromatography on a silica gel
column (35±70m).
Immunopharmacology
Proliferative response to mitogenic stimulation (PHA):
MTT assay. Peripheral blood leukocytes from healthy
donors were separated by centrifugation on a ®coll-
hypaque gradient. Cells were cultured in RPMI 1640
(Gibco BRL, Cergy Pontoise, France) containing 10%
foetal calf serum (DAP, Volgebrun, France). The cell
suspension was diluted to 106 cells/mL, and 100 mL
put into each well of the 96-well plates. Phytohe-
magglutinin (PHA) was used at ®nal concentration of
25 mL/mL. Incubation was carried out with and without
(without=control) potential immunomodulating com-
pounds at various concentrations (1, 10, 25, 50, and
100 mg/mL), for 48 h. All cells were counted using the
blue trypan (0.25%) exclusion test, and viability was
always over 95%. A stock solution of MTT (Sigma, St
Quentin, France) was dissolved in saline phosphate
bu€er at 5 mg/mL and sterilized by ®ltering. 100 mL was
2-Hydrazido-4-hydroxy-pyrrolidine (2a). To a solution
of 1a (2.47 g, 17 mmol), in ethanol (10 mL) was added
hydrazine hydrate (1 g, 20 mmol) and the resulting solu-
tion was stirred at room temperature during 18 h. After
evaporation, the crude product was recrystallized in
ethanol to provide 2 g (81%) of 2a; mp 139 ꢀC. IR (KBr)
1
3300, 2900, 1650, 1550 cm
;
1H NMR (DMSO-d6) d

354
V. Issartel et al./Bioorg. Med. Chem. 6 (1998) 349±354
added to each well, and plates were incubated at 37 ꢀC
for 3 h. Plates were then centrifuged and dimethylsulf-
oxyde was added to all wells. After a few minutes at
room temperature, the plates were read with a Titertek
Multiskan Plus (a microplate reader) using a test wave-
length of 570 nm, and a reference wavelength of 630 nm.
Results calculated from triplicates, were expressed as a
stimulation index (SI) using the following formula:
SI=OD (T cells stimulated by PHA+compound)/O.D
(T cells stimulated by PHA), (OD: optic density). For
control, the value of SI is 100%, indicating the absence
of proliferative response. Below this value, results indi-
cate immunosuppressing activity, and above immuno-
stimulating activity.
They also wish to thank Professor P. Tronche for the
interest he took in our work, Dr F. Pouthier (ETS
Besancon) for providing the blood samples, S. Biichle
,
and F. Leal for technical assistance.
References
1. Hadden, J. W. Immunol. Today 1993, 14, 275.
2. Turowski, R. C.; Triozzi, R. Ph.; Triozzi, P. L. Cancer
Invest. 1994, 12, 620.
3. Georgiev, V. S. Med. Res. Reviews 1990, 10, 371.
4. Goldstein, A. L.; Guha, A.; Zatz, M. M.; Hardy, M. A.;
White, A. Proc. Nat. Acad. Sci. U.S.A. 1972, 69, 1800.
5. Birouk, M.; Harraga, S.; Panouse-Perrin, J.; Robert, J. F.;
Damelincourt, M.; Theobald, F.; Mercier, R.; Panouse, J. J.
Eur. J. Med. Chem. 1991, 26, 91.
Oxidative response: chemiluminescence assay. Peripheral
blood, containing phagocytes, was directly transferred
to tubes and diluted at 1/10 in Hanks for basal mea-
surements. We then added appropriately diluted stimu-
6. Mailland, F.; Signorelli, G.; Coppi, G. Drugs Fut. 1991, 16,
1096.
7. Renoux, G. Drugs 1980, 19, 89.
lating agents
:
opsonised zymosan-A from
8. Devlin, J. P.; Hargrave, K. D. Tetrahedron 1989, 45, 4327.
Saccharomyces cerevisiae (20 mg/mL) (Sigma Chemical,
St. Louis, MO, USA) for the control, and opsonised
zymosan-A and compounds at di€erent concentrations.
All were incubated with 50 mL of luminol (5-
9. Bennett, G. A.; Radov, L. A.; Trusso, L. A.; Georgiev, V. S.
J. Pharm. Sci. 1987, 76, 633.
10. Jeanneau-Nicolle, E.; Benoit-Guyot, M.; Namil, A.;
Leclerc, G. Eur. J. Med. Chem. 1992, 27, 115.
amino-2,3-dihydro-1,4-phtalazine-dione)
(Boehringer
11. Blanc, M.; Cussac, M.; Boucherle, A.; Leclerc, G. Eur. J.
Med. Chem. 1992, 27, 839.
Mannheim, Germany) for 30 min. The kinetic measure-
ment of chemiluminescence, performed every 2 min by a
biolumat LB 9500 C photometer (Berthold, Wilbad,
Germany) was expressed in Relative Light Units
(RLU). For chemiluminescence response (CL) evalua-
tion, we established a chemiluminescence index (CLI) as
follows: CLI=RLU (stimulated cells) RLU (unstimu-
lated cells)/RLU (unstimulated cells). Only the peaks of
CL were utilized. Biostim1 (RU41740) was used as
reference, and kindly supplied by Cassenne±Roussel
UCLAF Laboratories (Paris, France). RU41740 was
dissolved in proportions of 5 mg/mL lyophilized powder
in sterile saline solution.
12. Harraga, S.; Nicod, L.; Drouhin, J. P.; Xicluna, A.;
Panouse, J. J.; Seilles, E.; Robert, J. F. Eur J. Med. Chem.
1994, 29, 309.
13. Bahaji, El H.; Tronche, P.; Couquelet, J.; Harraga, S.;
Panouse-Perrin, J.; Rubat, C. Chem. Pharm. Bull. 1991, 39,
2126.
14. Refouvelet, B.; Harraga, S.; Nicod, L.; Robert, J. F.;
Seilles, E.; Couquelet, J.; Tronche, P. Chem. Pharm. Bull. 1994,
42, 1076.
15. Margoum, N.; Tronche, P.; Bastide, P.; Bastide, J.; Rubat,
C. Eur. J. Med. Chem. 1984, 19, 415.
16. Dia®, L.; Couquelet, J.; Tronche, P. J. Het. Chem. 1990,
27, 2181.
Statistical analysis. Because of the small sample size,
results were expressed as median values. Statistical
comparisons were made using the non-parametric Mann±
Whitney U test.21
17. Mosmann, T. J. Immunol. Methods 1983, 65, 55.
18. Easmon, C. S. F.; Cole, P. J.; Williams, A. J.; Hasting, M.
Immunol. 1980, 41, 67.
19. Descamps-Latscha, B.; Nguyen, A. T.; Golub, R. M.;
Feuillet-Fieux, M. N. Ann. Immunol. 1982, 133C, 349.
20. Griscelli, C.; Grospierre, B.; Montreuil, J.; Fournet, B.;
Bruvier, C.; Lang, J. N.; Marchani, C.; Zalisz, K.; Edelstein, R.
In Immunomodulation by Microbial Products and Related Syn-
thetic Compounds; Yamamura, Y.; Kotani, S., Eds; Excerpta
Medica: Amsterdam, 1982; pp 261±265.
Acknowledgements
This research was supported by the Ligue Nationale
Contre le Cancer and partially funded by a grant from
Vetoquinol Pharmaceutical Laboratory (70204 Lure,
France). The authors wish to thank Cassenne±Roussel
UCLAF, Paris, for kindly supplying the RU41740.
21. Siegel, S. Nonparametric Statistics for the Behavorial Sci-
ences; McGraw±Hill: New York, 1956; pp 95±156.