Synthesis and evaluation of 6-nitro-7-(1-piperazino)quinazolines: dual-acting compounds with inhibitory activities toward both tumor necrosis factor-alpha (TNF-alpha) production and T cell proliferation.

Article abstract of DOI:10.1248/cpb.51.1109

We investigated the chemical modifications of the nitroquinazoline derivative (1) through the replacement of the NH group at the C(4)-position with several N-alkyl groups to increase the lipophilicity at the C(4)-position. Among them, we found that the N-methyl analogue (5a) showed a 2-fold loss in the inhibitory activity toward tumor necrosis factor-alpha (TNF-alpha) production in vitro as compared with the NH analogue (1); however, 5a exhibited an oral inhibitory activity on TNF-alpha production with an ED50 value of 26 mg/kg, whereas 1 did not. Moreover, the oral bioavailability of 5a was higher than that of 1 (1, F=1%; 5a, F=21%), and the calculated ClogP value for 5a was higher than that for 1. These results suggest that the improved lipophilicity of 5a compared with that of 1 reflects its greater inhibitory activity on TNF-alpha production in vivo as well as oral bioavailability.

Full text of DOI:10.1248/cpb.51.1109

September 2003
Notes
Chem. Pharm. Bull. 51(9) 1109—1112 (2003)
1109
Synthesis and Evaluation of 6-Nitro-7-(1-piperazino)quinazolines:
Dual-Acting Compounds with Inhibitory Activities toward Both
a
a
Tumor Necrosis Factor- (TNF- ) Production and T Cell Proliferation
Masanori TOBE, Yoshiaki ISOBE, Hideyuki TOMIZAWA, Takahiro NAGASAKI, Masamitsu AOKI,
Toru NEGISHI, and Hideya HAYASHI*
Research Division, Sumitomo Pharmaceuticals Co., Ltd.; Konohana-ku, Osaka 554–0022, Japan.
Received May 12, 2003; accepted June 16, 2003
We investigated the chemical modifications of the nitroquinazoline derivative (1) through the replacement of
the NH group at the C(4)-position with several N-alkyl groups to increase the lipophilicity at the C(4)-position.
Among them, we found that the N-methyl analogue (5a) showed a 2-fold loss in the inhibitory activity toward
tumor necrosis factor-a (TNF-a) production in vitro as compared with the NH analogue (1); however, 5a exhib-
ited an oral inhibitory activity on TNF-a production with an ED50 value of 26 mg/kg, whereas 1 did not. More-
over, the oral bioavailability of 5a was higher than that of 1 (1, F؍1%; 5a, F؍21%), and the calculated ClogP
value for 5a was higher than that for 1. These results suggest that the improved lipophilicity of 5a compared with
that of 1 reflects its greater inhibitory activity on TNF-a production in vivo as well as oral bioavailability.
Key words tumor necrosis factor-a production; T cell proliferation; 6-nitroquinazoline; lipophilicity
Tumor necrosis factor-a (TNF-a) exerts a key role in the framework at the C(7)-position of the quinazoline ring was
cytokine network with regard to the pathogenesis of autoim- the simple piperazine ring. The optimal length was 1 methyl-
mune diseases such as rheumatoid arthritis (RA),1,2) septic ene unit as a spacer between the quinazoline core and the
shock,3) and Crohn’s disease.4) To date, two anti-TNF-a phenyl ring at the C(4)-position. The substitution with the
agents (RemicadeTM and EnbrelTM) are available for the treat- 3,4-methylenedioxy group on the phenyl ring at the C(4)-po-
ment of patients with RA and Crohn’s disease.5—10) Addition- sition was suitable for the TNF-a inhibitory activity, whereas
ally, autoimmune diseases are considered to be caused by ab- the hydrophilic substituents, such as the amide group and the
normalities of T cell immune responses, and the activation of carboxyl one, were unfavorable.
T cells has been shown to be a possible mechanism by which
Based on the above-mentioned SAR data, we concluded
inflammation is enhanced in these diseases.11—14) Based on that the enhancement of the lipophilicity at the C(4)-position
the above information, we speculated that ideal anti-autoim- of the quinazoline ring would be desirable for potent in-
mune diseases agents should possess the ability to block ab- hibitory activity. In the present study, we examined further
normal T cell-mediated immune responses as well as to in- chemical modifications on the basis of the replacement of
hibit TNF-a activity.
the NH group at the C(4)-position of 1 with several N-alkyl
Recently, we have reported that a series of quinazoline groups to increase the lipophilicity at the C(4)-position.
derivatives showed inhibitory activities toward both TNF-a Herein we wish to describe the discovery of a new series of
production and T cell proliferation.15) Among them, com- 6-nitroquinazolines that show the inhibition of TNF-a pro-
pound 1 exerted potent inhibitory activities toward both para- duction by oral administration.
meters (Fig. 1). However, 1 did not show an inhibitory activ-
Chemistry The synthetic pathway to the quinazoline de-
ity toward TNF-a production by oral administration.15) Al- rivatives is shown in Chart 1. The key intermediate 4,7-
though 1 showed metabolic stability against rat liver S9 frac- dichloro-6-nitroquinazoline (3) was prepared by the chlorina-
tion in vitro (data not shown), its oral bioavailability was tion of 7-chloro-6-nitro-4-quinazolone (2)15) with thionyl
poor (Fϭ1%). These data suggest that poor oral bioavailabil- chloride. Compound 3 was treated with the corresponding
ity of 1 reflects no inhibitory activity toward TNF-a produc- amines to give 4-substituted-7-chloro-6-nitroquinazolines
tion in vivo. To overcome the above mentioned problem, we (4a—e). With the exception of 5e, the target compounds
considered the hypothesis that the enhancement of lipophilic- 5a—d were prepared by the reaction of 4a—d with piper-
ity of 1 would lead to an orally-active compound, along with azine in the presence of Hunig’s base, respectively. Reaction
good oral bioavailability.
of 4e with piperazine provided the mixture of the acid (5e)
In the previous papers, we delineated the structure activity and its ester analogue, which was then hydrolyzed by treat-
relationship (SAR) for the 6-nitroquinazolines.15,16) The best ment with 5 N NaOH to yield 5e only.
Results and Discussion
The compounds listed in Table 1 were evaluated for their
abilities to inhibit both TNF-a production and T cell prolif-
eration as previously reported.15) Also included in Table 1 are
the results of cytotoxicity experiments using the MTS assay
in human PBMCs.17) As shown in Table 1, the replacement
of the NH group at the C(4)-position with the N-methyl one
(5a) led to a 2-fold loss in the TNF-a inhibitory activity as
Fig. 1. Structure of Compound 1
To whom correspondence should be addressed. e-mail: hayashih@sumitomopharm.co.jp
© 2003 Pharmaceutical Society of Japan

1110
Vol. 51, No. 9
Chart 1. Synthesis of Compounds 5a—e
Table 1. Inhibition of TNF-a Production and T Cell Proliferation by Compounds 5a—e
IC50 (nM)
Con Ab)
Compound
R4
ClogPd)
TNF-aa)
MTSc)
1
H
Me
Et
78
158
244
321
535
2743
2432
4242
4136
4023
4240
Ͼ30000
Ͼ30000
8650
4427
Ͼ30000
2.21
2.61
3.14
3.67
4.21
5a
5b
5c
5d
5e
n-Pr
n-Bu
CH2CO2H
Ͼ10000
Ͼ10000
Ϫ0.24
a) IC50 for inhibition of TNF-a production from human PBMCs stimulated by LPS. b) IC50 for inhibition of Con A-induced proliferation of mice spleen cells. c) IC50
for the growth inhibition of human PBMCs stimulated by LPS. d) The ClogP values were calculated using the program ClogP for Windows, Version 4.0, a product of BioByte
Corp.
Table 2. Effects of 1 and 5a on LPS-Induced TNF-a Production in Mice, and Pharmacokinetic Profile in Rats
Pharmacokinetic profileb)
ED50 or % of TNF-a
Compound
inhibition (mg/kg/p.o.)a)
Cmax (nM)
Tmax (h)
AUC (mM ·min)
F (%)
1
5a
18% @ 30
26
39
462
2
2
40
414
1
21
a) ED50 value was determined from dose–response curve of TNF-a inhibition. Compounds were evaluated as their corresponding hydrochroride salts, and administered orally
to mice 1 h prior to the LPS challenge. b) Compounds were examined for pharmacokinetics when orally administered to rats in water (30 mg/kg).
compared with the activity of 1. However, 5a improved ity in this model. Also included in Table 2 were the results of
against the cytotoxicity; i.e., the IC50 value for the cell the pharmacokinetic studies on compounds 1 and 5a con-
growth inhibition of 5a was over 30 mM, whereas that of 1 ducted in rats, when orally administered at a dose of
was 4.2 mM. Elongation of the alkyl chain length (5b—d) re- 30 mg/kg. Compound 5a showed the Cmax value of 462 nM
sulted in the reduction of both inhibitory activities with in- and oral bioavailability of 21%, whereas 1 resulted in low
creasing the cytotoxicity. On the other hand, compound 5e, Cmax value (39 nM) and poor oral bioavailability (1%). Fur-
which was the hydrophilic analogue (ClogPϭϪ0.24), led to thermore, as we expected, the lipophilicity of 5a was higher
almost complete loss of both inhibitory activities.
than that of 1 (5a, ClogPϭ2.61; 1, ClogPϭ2.21). These re-
On the basis of the above-mentioned in vitro study, we se- sults suggest that the higher lipophilicity of 5a compared
lected compound 5a and performed the comparative evalua- with that of 1 reflects its greater oral inhibitory activity as
tion of 1 and 5a toward the inhibitory effects on TNF-a pro- well as oral bioavailability.
duction in vivo. As shown in Table 2 and Fig. 2, compound
5a inhibited TNF-a production with an ED50 value of Conclusions
26 mg/kg, whereas compound 1 showed no inhibitory activ-
In order to find a novel quinazoline derivative possessing

September 2003
1111
m), 6.94—6.82 (3H, m), 6.01 (2H, s), 4.90 (2H, s), 3.31 (3H, s). TOF-MS
m/z: 373 (Mϩ). Anal. Calcd for C17H13ClN4O4: C, 54.78; H, 3.52; N, 15.03.
Found: C, 54.71; H, 3.63; N, 14.93.
Similarly to the procedure described for 4a, compounds 4b—e were pre-
pared from 2, respectively.
7-Chloro-4-(N,N-ethylpiperonylamino)-6-nitroquinazoline (4b) Light
1
yellow solid (67% yield for 2 steps from 2). mp 123—125 °C (i-Pr2O). H-
NMR (DMSO-d6) d: 8.65 (1H, s), 8.64 (1H, s), 8.01 (1H, s), 6.95—6.83
(3H, m), 6.02 (2H, s), 4.99 (2H, s), 3.78 (2H, q, Jϭ7.0 Hz), 1.35 (3H, t,
Jϭ7.0 Hz). TOF-MS m/z: 387 (Mϩ). Anal. Calcd for C18H15ClN4O4: C,
55.89; H, 3.91; N, 14.49. Found: C, 55.86; H, 3.97; N, 14.52.
7-Chloro-4-(N,N-n-propylpiperonylamino)-6-nitroquinazoline (4c)
Light yellow oil (58% yield for 2 steps from 2). 1H-NMR (DMSO-d6) d:
8.65 (1H, s), 8.64 (1H, s), 8.02 (1H, s), 6.93—6.82 (3H, m), 6.01 (2H, s),
5.01 (2H, s), 3.71—3.65 (2H, m), 1.91—1.76 (2H, m), 0.93 (3H, t,
Jϭ7.3 Hz). TOF-MS m/z: 401 (Mϩ). Anal. Calcd for C19H17ClN4O4: C,
56.93; H, 4.28; N, 13.98. Found: C, 56.73; H, 4.44; N, 13.92.
7-Chloro-4-(N,N-n-butylpiperonylamino)-6-nitroquinazoline (4d)
Light yellow oil (40% yield for 2 steps from 2). 1H-NMR (DMSO-d6) d:
8.65 (2H, s), 8.02 (1H, s), 6.93—6.82 (3H, m), 6.01 (2H, s), 5.00 (2H, s),
3.74—3.68 (2H, m), 1.86—1.75 (2H, m), 1.42—1.29 (2H, m), 0.92 (3H, t,
Jϭ7.3 Hz). TOF-MS m/z: 415 (Mϩ). Anal. Calcd for C20H19ClN4O4: C,
57.90; H, 4.62; N, 13.51. Found: C, 58.15; H, 4.54; N, 13.42.
Fig. 2. Effects of 1 and 5a on LPS-Induced TNF-a Production in Mice
Compounds 1 and 5a were evaluated as their corresponding hydrochloride salts. The
results are expressed as the meansϮS.E.M. of 4 mice per group.
pϽ0.001 versus
vehicle control (Dunnett’s test). Veh., vehicle control; Dex., dexamethasone.
7-Chloro-4-[N,N-(ethoxycarbonylmethyl)piperonylamino]-6-nitro-
quinazoline (4e) Light yellow oil (46% yield for 2 steps from 2). 1H-NMR
(DMSO-d6) d: 8.70 (1H, s), 8.58 (1H, s), 8.09 (1H, s), 7.04 (1H, s), 6.96
(2H, s), 6.04 (2H, s), 5.08 (2H, s), 4.50 (2H, s), 4.15 (2H, q, Jϭ7.2 Hz), 1.20
(3H, t, Jϭ7.2 Hz). TOF-MS m/z: 445 (Mϩ). Anal. Calcd for C20H17ClN4O6:
C, 54.00; H, 3.85; N, 12.60. Found: C, 54.28; H, 3.90; N, 12.56.
an oral inhibitory activity on TNF-a production, we exam-
ined further chemical modifications on the basis of the re-
placement of the NH group at the C(4)-position of 1 with
several N-alkyl groups. Among them, we found that the N-
methyl analogue (5a) showed a 2-fold loss in the inhibitory
activity toward TNF-a production in vitro as compared with
the NH analogue (1); however, 5a exhibited an oral in-
hibitory activity on TNF-a production with an ED50 value of
26 mg/kg, whereas 1 did not. Moreover, the oral bioavailabil-
ity of 5a was higher than that of 1 (1, Fϭ1%; 5a, Fϭ21%),
and the calculated ClogP value for 5a was higher than that
for 1. Therefore, we concluded that the improved lipophilic-
ity of 5a compared with that of 1 reflects its greater in-
hibitory activity on TNF-a production in vivo as well as oral
bioavailability. Compound 5a is currently being prepared for
the pharmacological evaluation and safety as a candidate
which may be used clinically as an anti-autoimmune disease
agent.
4-(N,NЈ-Methylpiperonylamino)-6-nitro-7-(1-piperazino)quinazoline
Hydrochloride (5a) To a suspension of 4a (150 mg, 0.402 mmol) and
N,NЈ-diisopropylethylamine (210 ml, 1.21 mmol) in n-BuOH (8 ml) was
added piperazine (104 mg, 1.21 mmol). The reaction mixture was stirred at
110 °C for 10 h under a nitrogen atmosphere. After the mixture was cooled,
the solvent was evaporated in vacuo, and then the residue was partitioned
between CH2Cl2 and 5% citric acid aqueous solution. The aqueous layer was
adjusted to pH 9 with 5 N NaOH, and extracted with CH2Cl2. The organic
layer was washed with water and brine, dried over Na2SO4 and evaporated
in vacuo to give crude 5a. To a suspension of 5a in EtOH (7 ml) was added
12 N HCl (84 ml). The mixture was stirred at ambient temperature for 3 h,
and then evaporated in vacuo. The residue was triturated with Et2O, and pre-
cipitated solid was collected by filtration. The obtained solid was dried in
vacuo to give the hydrochloride salt as a light yellow powder (123 mg, 66%
yield for 2 steps from 4a). mp 263—265 °C (Et2O). 1H-NMR (DMSO-d6) d:
9.58 (2H, br s), 8.83 (2H, s), 7.57 (1H, s), 7.00—6.88 (3H, m), 6.02 (2H, s),
5.15 (2H, s), 3.57 (3H, s), 3.43—3.25 (8H, m). TOF-MS m/z: 423 (Mϩ).
Anal. Calcd for C21H23ClN6O4·2.0H2O: C, 50.96; H, 5.09; N, 16.98. Found:
Experimental
Chemistry All reagents and solvents were obtained from commercial C, 50.75; H, 4.90; N, 16.65.
suppliers and were used without further purification. Melting points were
Similarly to the procedure described for 5a, compounds 5b—d were pre-
measured with a BÜCHI 535 melting point apparatus and were uncorrected. pared from 4b—d, respectively.
Proton NMR spectra were recorded on a JEOL GSX270 FT NMR spectrom-
4-(N,N-Ethylpiperonylamino)-6-nitro-7-(1-piperazino)quinazoline
eter. Time-of-flight mass spectrometry (TOF-MS) was recorded on a KOM- Hydrochloride (5b) Light yellow solid (78% yield for 2 steps from 4b).
PACT MALDI III spectrometer. High-resolution mass spectra were obtained mp 86—88 °C (Et2O). 1H-NMR (DMSO-d6) d: 9.57 (2H, br s), 8.85 (1H, s),
on a JEOL JMS-700 mass spectrometer. Elemental analyses were performed 8.57 (1H, s), 7.57 (1H, s), 6.99—6.87 (3H, m), 6.04 (2H, s), 5.15 (2H, s),
at the Toray Research Center. Monitoring of reactions was carried out using 3.45 (2H, q, Jϭ6.8 Hz), 3.40—3.24 (8H, m), 1.39 (3H, t, Jϭ6.8 Hz). TOF-
Merck 60 F254 silica gel, glass-supported TLC plates, and visualization with MS m/z: 437 (Mϩ). Anal. Calcd for C22H25ClN6O4·3.5H2O: C, 49.30; H,
UV light (254, 365 nm). Following abbreviations are used for solvents: 5.36; N, 15.68. Found: C, 49.55; H, 5.60; N, 15.31.
DMF (N,N-dimethylformamide), EtOH (ethanol), i-PrOH (2-propanol), n-
BuOH (1-butanol), Et2O (diethyl ether), i-Pr2O (diisopropyl ether).
6-Nitro-4-(N,N-n-propylpiperonylamino)-7-(1-piperazino)quinazoline
Hydrochloride (5c) Light yellow solid (76% yield for 2 steps from 4c).
7-Chloro-4-(N,N-methylpiperonylamino)-6-nitroquinazoline (4a) To mp 85—87 °C (Et2O). 1H-NMR (DMSO-d6) d: 9.48 (2H, br s), 8.83 (1H, s),
a suspension of 215) (250 mg, 1.11 mmol) in thionyl chloride (6 ml) was 8.55 (1H, s), 7.53 (1H, s), 6.97—6.85 (3H, m), 6.03 (2H, s), 5.16 (2H, s),
added 1 drop of DMF at ambient temperature, and the mixture was refluxed
3.81—3.76 (2H, m), 3.37—3.25 (8H, m), 1.91—1.86 (2H, m), 0.97 (3H, t,
for 2 h. After the reaction mixture was cooled, excess thionyl chloride was Jϭ7.3 Hz). TOF-MS m/z: 451 (Mϩ). Anal. Calcd for C23H27ClN6O4·4.5H2O:
removed under reduced pressure to give crude 4,7-dichloro-6-nitroquinazo- C, 48.63; H, 5.59; N, 14.80. Found: C, 48.94; H, 5.58; N, 14.82.
line (3), which was used directly. Subsequently, to a mixture of 3 and triethy-
4-(N,N-n-Butylpiperonylamino)-6-nitro-7-(1-piperazino)quinazoline
lamine (233 ml, 1.67 mmol) in i-PrOH (12 ml) was added N-methylpiperony- Hydrochloride (5d) Light yellow solid (80% yield for 2 steps from 4d).
lamine (367 mg, 2.22 mmol). The reaction mixture was stirred at ambient mp 87—89 °C (Et2O). 1H-NMR (DMSO-d6) d: 9.56 (2H, br s), 8.84 (1H, s),
temperature for 1 h and evaporated in vacuo, and partitioned between 8.56 (1H, s), 7.56 (1H, s), 6.98—6.85 (3H, m), 6.03 (2H, s), 5.16 (2H, s),
CH2Cl2 and 5% citric acid aqueous solution; the organic layer was washed 3.85—3.79 (2H, m), 3.44—3.25 (8H, m), 1.84 (2H, br s), 1.44—1.36 (2H,
with 1 N NaOH, water, brine, and dried over Na2SO4. The solution was
m), 0.94 (3H, t, Jϭ7.3 Hz). TOF-MS m/z: 465 (Mϩ). Anal. Calcd for
evaporated in vacuo and the residue was triturated with CH2Cl2–hexane C24H29ClN6O4·3.5H2O: C, 51.11; H, 5.81; N, 14.90. Found: C, 51.36; H,
(1 : 1, v/v). The precipitated solid was washed with i-Pr2O, and was filtered 6.03; N, 14.60.
to give 4a (246 mg, 59% yield for 2 steps from 2). mp 138—139 °C (i-Pr2O).
4-[N,N-(Carboxymethyl)piperonylamino]-6-nitro-7-(1-piperazino)-
quinazoline (5e) To a suspension of 4e (217 mg, 0.488 mmol) and N,NЈ-
1H-NMR (DMSO-d6) d: 8.54 (1H, s), 8.13—8.05 (1H, m), 7.81—7.79 (1H,

1112
Vol. 51, No. 9
Feldmann M., Eur. J. Immunol., 22, 1907—1912 (1992).
3) Lechner A. J., Tredway T. L., Brink D. S., Klein C. A., Matuschak G.
M., Am. J. Physiol., 263, L526—L535 (1992).
4) Van Dullemen H. M., Van Deventer S. J. H., Hommes D. W., Bijl H.
A., Jansen J., Tytgat G. N. J., Woody J., Gastroenterology, 109, 129—
135 (1995).
5) Maini R. N., St. Clair E. W., Breedveld F. C., Furst D. E., Kalden J. R.,
Weisman M., Smolen J. S., Emery P., Harriman G. R., Feldmann M.,
Lipsky P. E., Lancet, 354, 1932—1939 (1999).
6) Lipsky P. E., Van Der Heijde D. M. F. M., St. Clair E. W., Furst D. E.,
Breedveld F. C., Kalden J. R., Smolen J. S., Weisman M., Emery P.,
Feldmann M., Harriman G. R., Maini R. N., N. Engl. J. Med., 343,
1594—1602 (2000).
7) Markham A., Lamb H. M., Drugs, 59, 1341—1359 (2000).
8) Weinblatt M. E., Kremer J. M., Bankhurst A. D., Bulpitt K. J., Fleis-
chmann R. M., Fox R. I., Jackson C. G., Lange M. L. M., Burge D. J.,
N. Engl. J. Med., 340, 253—259 (1999).
9) Moreland L. W., Schiff M. H., Baumgartner S. W., Tindall E. A., Fleis-
chmann R. M., Bulpitt K. J., Weaver A. L., Keystone E. C., Furst D.
E., Mease P. J., Ruderman E. M., Horwitz D. A., Arkfeld D. G., Garri-
son L., Burge D. J., Blosch C. M., Lange M. L. M., McDonnell N. D.,
Weinblatt M. E., Ann. Intern. Med., 130, 478—486 (1999).
10) Bathon J. M., Martin R. W., Fleischmann R. M., Tesser J. R., Schiff M.
H., Keystone E. C., Genovese M. C., Wasko M. C., Moreland L. W.,
Weaver A. L., Markenson J., Finck B. K., N. Engl. J. Med., 343,
1586—1593 (2000).
11) Quayle A. J., Chomarat P., Miossec P., Kjeldsen-Kragh J., Førre Ø.,
Natvig J. B., Scand. J. Immunol., 38, 75—82 (1993).
12) Simon A. K., Seipelt E., Sieper J., Proc. Natl. Acad. Sci. U.S.A., 91,
8562—8566 (1994).
13) Schulze-Koops H., Lipsky P. E., Kavanaugh A. F., Davis L. S., J. Im-
munol., 155, 5029—5037 (1995).
14) Van Roon J. A. G., Van Roy J. L. A. M., Duits A., Lafeber F. P. J. G.,
Bijlsma J. W. J., Ann. Rheum. Dis., 54, 836—840 (1995).
15) Tobe M., Isobe Y., Tomizawa H., Nagasaki T., Obara F., Matsumoto
M., Hayashi H., Chem. Pharm. Bull., 50, 1073—1080 (2002).
16) Tobe M., Isobe Y., Tomizawa H., Matsumoto M., Obara F., Nagasaki
T., Hayashi H., Bioorg. Med. Chem. Lett., 11, 545—548 (2001).
17) For the purpose of evaluating the cytotoxicity of the compounds, we
assessed cell viability by the MTS assay when we assayed TNF-a pro-
duction.
diisopropylethylamine (261 ml, 1.46 mmol) in n-BuOH (10 ml) was added
piperazine (126 mg, 1.46 mmol). The reaction mixture was stirred at 110 °C
for 5 h under a nitrogen atmosphere. After the mixture was cooled, 5 N
NaOH was added. The mixture was stirred at ambient temperature for 3 h,
and then evaporated in vacuo. The resulting residue was poured into water
(3 ml), and then was adjusted to pH 5 with 5% citric acid aqueous solution,
and the precipitated solid was washed with EtOH. The light yellow powder
was filtered to give 5e (140 mg, 61% yield for 2 steps from 4e). mp 217—
219 °C (EtOH). 1H-NMR (DMSO-d6) d: 12.02 (1H, br s), 8.54 (1H, s), 8.49
(1H, s), 7.21 (1H, s), 7.00 (1H, s), 6.92—6.85 (2H, m), 6.01 (2H, s), 4.97
(2H, s), 4.17 (2H, s), 3.08 (4H, br s), 2.98 (4H, br s). HR-MS (FAB) m/z:
Calcd for C22H22N6O6 (Mϩ): 467.1679. Found: 467.1641.
Biology. LPS-Induced TNF-a Production in Human PBMCs Inhi-
bition of TNF-a production was measured by ELISA using human PBMCs
stimulated by LPS, as previously reported.15)
T Cell Proliferation Assay T cell proliferation was determined by the
MTS assay using Con A-stimulated mice spleen cells according to a previ-
ous report.15)
LPS-Induced TNF-a Production in Mice Inhibitory effects of 1 and
5a toward TNF-a production in LPS treated mice were evaluated according
to a previously reported method.15) Briefly, BALB/c mice (Charles River
Japan Inc.) were used at 10 weeks of age. LPS from Escherichia coli
(serotype 026:B6) was purchased from Difco Laboratories (Detroit, U.S.A.).
A solution of a test compound in water was orally administered to mice 1 h
prior to i.v. injection of the LPS (25 mg/mouse). Blood samples were ob-
tained 2 h after the LPS injection. Amounts of TNF-a in the blood were de-
termined by a specific ELISA kit (Genzyme Techne, U.S.A.).
Pharmacokinetic Analysis The pharmacokinetic parameters of 1 and
5a were studied in rats. Briefly, in the rat i.v. or p.o. administration studies, 3
male SD rats received either a 10 mg/kg intravenous dose or a 30 mg/kg oral
dose. Blood samples were obtained after dosing at appropriate times and an-
alyzed by reverse phase HPLC. The pharmacokinetic parameters for the
compounds were estimated by a non-compartmental method.
Acknowledgements The authors are grateful to Dr. Satoru Misawa for
useful suggestions and encouragement during the course of this research.
References and Notes
1) Macnaul K. L., Hutchinson N. I., Parsons J. N., Bayne E. K., Tocci M.
J., J. Immunol., 145, 4154—4166 (1990).
2) Brennan F. M., Gibbons D. L., Mitchell T., Cope A. P., Maini R. N.,