- Email: [email protected]
Synthesis and Structural, Conformational, Biochemical, and Pharmacological Study of New Compounds Derived from Tropane-3-spiro-4’(5’)-imidazoline as Potential 5-HT3 Receptor Antagonists
Synthesis and Structural, Conformational, Biochemical, and Pharmacological Study of New Compounds Derived from Tropane-3-spiro-4’(5’)-imidazoline as Potential 5-HT3 Receptor Antagonists B. A. WHELAN’, 1. IRIEPA’, E. G A L V E A. ~ ~ORJALES~, ~, A. BERISA~, L. LABEAGA~, A. G.GARCIA§,G. UCEDA~, J. SANZ-APARICIO~~, AND I. FONSECA~~ Received September 23, 1993, from the *Do. de Quimica Organica, Universidad de Alcala de Henares, Madrid, Spain, FAES S.A. Apartado 555, Bilbao, Spain, Qto. de Farmacologia, Faculfad de Medicina, U.A.M. Madrid, Spain, and Vnstituto Rocasolano E. U.1. Accepted for publication August 24, 1994@. Crisfalografia C.S.I.C., Serrano 119, 28006 Madrid, Spain. Abstract 0 A series of tropane-3-spiro-4’(5’)-imidazolines was synthesized and studied by ’H and 13C NMR spectroscopy, and the crystal structure of 2’-( 1Hindol-3-yl)tropane-3-spiro-4‘(5’)-imidazoline hydrochloride 5(6)f was determined by X-ray diffraction. In CD30D solution, compounds 5(6)a-f display the same preferred conformation. The pyrrolidine and piperidine rings adopt an envelope conformation flattened at N8 and a distorted chair conformation puckered at N8 and flattened at C3, respectively, with the N-substituent in the equatorial position with respect to the piperidine ring. This conformation is similar to that observed for compound 5(6)f in the solid state. From binding studies on the compounds synthesized, compound 5(6)d demonstrated the ability to efficiently displace the binding of [3H]GR65630 to bovine brain area postrema membranes to an extent comparable to MDL 72222. In the von Bezold-Jarisch reflex, compound 5(6)d was equipotent with metoclopramide. It is, therefore, likely that the imidazoline ring may provide a useful bioisosteric replacement for the carbonyl group in 5-HT3 antagonists.
H3C”$p+.m
3a
I a C6HS
d 3.5-diCIC6H3
b 3-MeOC6H,
e 4-Pyridin-N-oxide
c 3,5-diMeOC6H3
f 3-indolyl
6
The 5-HT3 receptor subtype, although originally thought to be confined t o the peripheral nervous system,l is now known to be present in the central nervous ~ y s t e m . Several ~,~ highly selective antagonists of the 5-HT3 receptor have been described4 and a number of these are the subject of clinical studies. In particular, Ondansetron, Granisetron, and ICS 205-930 have been shown to be highly effective inhibitors of cytotoxic drug induced emesis in man5r6and are currently being marketed for this purpose. Animal s t ~ d i e s ~suggest z~,~ that this class of compounds may also possess anxiolytic, antipsychotic, or cognition-enhancing activity. Preliminary clinical studies have provided support for these psychiatric uses in m a n g Many of the currently reported 5-HT3 antagonists are aromatic esters or amides of appropriate bicyclic alcohols or amines,lOJ1though some heterocyclics (such as oxadiazoles and oxazolines)which were shown to be isosteric with respect t o the ester group have recently been successfully introdu~ed.’~J~ By considering that the C=N group of the imidazoline ring could possibly act as a bioisosteric replacement for esters or amides, we have designed compounds 5(6)a-f as potential 5-HT3 antagonists. We report here the synthesis and structural analysis, carried out with the aid of ‘H and 13CNMR spectroscopy, of compounds 5(6)a-f, derived from tropane-3-spiro-4‘-imidazoline.In order to determine their preferred conformations, both in solution and in the solid state, the crystal structure of compound 5(6)f has also been determined. @Abstractpublished in Advance ACS Abstracts, October 1, 1994.
0 1995, American Chemical Sociely and American Pharmaceutical Association
1
H3c”T I
Introduction
h
LiAIHI
RC:NH$XOMe) 29-52 Ya
5
Scheme 1
Results Chemistry-Compounds 5(6)a-f were prepared as shown in Scheme 1. Reaction of tropinone 1 with ammonium chloride and potassium cyanide in water gave the aminonitrile 3a. The most stable conformation of 1 in CDC4 is as represented in Scheme l.14 By assuming the same preferred conformation for the ketimine 2 in solution, the more favorable p attack of the cyanide should be expected, thus leading to the 3/3 isomer. Therefore, the formation of the more stable 3 a aminonitrile [in the case of 3a and 3/3,the most severe steric interaction is exerted by the NH2 group on H6(7)n]can be attributed to the thermodynamic control of the process. With an eye on forming the imidazoline derivative of the 3p cyanide intermediate, exhaustive attempts a t its synthesis were made but proved futile in all cases, the a isomer invariably being the aminonitrile isolated. By treating compound 3 with lithium aluminium hydride in ether, the amine 4 was obtained. Compounds 5(6)a-f were obtained by treatment of 4 with the corresponding methyl imidate hydrochloride. Structural Study-Description of the Structure of 5(6)pTable 115-21 summarizes the experimental and structural displays the title comdetermination procedures. Figure lz2 pound and the numbering used in the crystallographic study (supplementary material is available from the authors).
0022-3549/95/3184-0101$09.00/0
Journal of Pharmaceutical Sciences / 101 Vol. 84, No. I, January 1995
Table 1-Experimental Data and Structural Refinement Procedures Value
Parameter Crystal data Formula Symmetry Unit cell determinations Unit cell dimensions (A) Packing: V(A3), Z Dc (g ~ m - ~M, ) ,F(OO0) P (cm-9 Experimental data Technique
CI~HZ~N~OCIZ Monoclinical, Rl/n Least squares fit from 60 reflections (0< 15") 7.661(1), 26.432(4), 9.986(2), 90.0, 98.43(1), 90.0 2000.3(6), 4 1.3261,399.36,848 3.389 Four circle diff ractometer: Enraf-Nonius, Cad-4, Bisecting geometry. Grafite-oriented monochromator:Mo Ka. w scans; scan width, 1" up to a max of 28"
Number of reflections Measured Observed Range of hkl Value of Rn, Max-min transmission factors Solution and refinement Solution Refinement H atoms w scheme Final AFpeaks Final Rand R, Computer and programs Scattering factors Anomalous dispersion
1C14 c2
4773 1026 (241) criterion) 0 10,O 34, -13 13 0.01 1.244,0.555 Direct methods and Fourier synthesis Full-matrix least squares on Fobs.Anisotropic thermal parameters of C6 fixed Difference synthesis. H52, H92, H93, and H262 not refined Empirical as to give no trends in (WAF)vs (Fobs)and (sin el^) 0.3 e/A3 0.054,0.013 Vax 6410, D i f a b ~ ,Multan80,16 '~ Dirdif,i7 Xray System,'* Pesos,lg Parstzo lnfemafional Tables for X-Ray Crysfallographfl lntemational Tables for X-ray Crysfal/ographfl
b
AC15
?2O Figure 1-PLUTO view of the molecule 5(6)f showing the atomic numbering.22
The bicyclic system is in the chair-envelope conformation commonly found in these kind of compounds. The chair is flattened at the C4 atom to release steric hindrance, C4 and N1 being 0.49(1) and 0.86(1) A, respectively, from the mean plane defined by the other four atoms. N1, on the other hand, is in the flap of the envelope, 0.70(1) A away from the plane defined by the rest of the atoms in this ring. The molecule presents a pseudo-mirror plane, passing through the N1, C4, and C9 atoms. The imidazoline ring is almost in this plane with an interplanar angle of only 5". The indole moiety, on the contrary, deviates from this symmetry, defining an interplanar angle of 24". C123 and C124 are in opposite positions with res ect to the pseudo-mirror plane, being 0.639(4) and 1.472(3) respectively, out of this plane. As shown by the bond lengths, the electrons are delocalized along NlO-Cll-Nl2 and the indole moiety, which maintains the interplanar angle of 24" between the indole and the imidazoline ring. Consequently, there is no difference between N10 and N12 or between the CllN12 and the CllNlO bond lengths. Both N atoms of the imidazoline ring form hydrogen bonds with the two chlorine atoms. However, one of them acts as the counter-chlorideion (C124), while the other seems to be present as a hydrochloride of crystallization and
1,
102 / Journal of Pharmaceutical Sciences Vol. 84, No. 1, January 1995
Figure 2-Unit cell packing for compound 5(6)f. also interacts with the CH30 group (see Table 2) of the solvent, which in turn interacts with the bicyclic N1 atom. The hydrogen-bonding pattern is completed by a hydrogen bond between the indole N16 and the C124 atom from a different asymmetric unit. Two molecules related by a center of symmetry form a dimer through two hydrogen bonds: N12-Hl2-Cl24 and Nl6-Hl6-Cl24. These dimers form isolated units stacked along the "a" axis. The first hydrogen bond links the C124 atom to the N12 one in the same molecule ( x , y , z ) and the second one links the C124 to the N16 of a
KO~ 1 . 3 9 t 0.5 10 80 7
-'M
Table 3-Selected Interatomic Distances of 5(6)f in the Solid State
70 -
60 50 -
10
2
1
4
3
[ WGR65630 ] (nM)
Figure 3-Saturable specific binding of [3H]GR65630 to membranes of bovine area postrema (2 mg wet weight, 174.62 p g protein) using filtration binding methodology. Assays were performed in duplicate; error bars represent SEM of three experiments performed on separate batches of membranes. Table 2-Interatomic Bond Distances and Angles of 5(6)f N1-HI
I .o(i) A
C123-H23 1.7(2) 8, N10-H1 0 1.0(1) A N12-Hl2 1.0(1) A N16-HI 6 0.7(2) A 1 - x , 1 -y,-z
Atom 2
Interatomic Distance, 8,
Atom 1
Atom 2
Interatomic Distance, A
N8 N8 N8 N8
Xa Yb N3' N1'
9.2 7.6 4.1 5.3
N3' NI' N3' N1'
X X Y Y
5.2 4.4 3.5 3.6
= Gravimetric center of the indole six-membered aromatic ring. Gravimetric center of the indole five-membered ring.
40 -
0
Atom 1
Em,= 84.32 k 21.9 h o t . mg prot-'
H1...025 1.7(1) A H23...025 2.5(1) 8, H IO-CI23 2.3(1) A H12-CI24 2.2(1) A H16-CI24' 2.5(2) A
N1...025 2.72(1) A C123...025 3.04(1) A N1O - m 3 3.24(1) A N12-CI24 3.15(1) A N16..C124' 3.18(1) A
NI-H1.-025 164(l)O C123-H23...025 92(1)" N10-HI O-CI23 168(1)0 N12-H 12-CI24 165(I)" N16-H16..C124' 165(1)O
different asymmetric unit (-z, 1 - y, -21, the C1 atom acting as the nexus between the two molecules. NMR Spectra-The 'H and 13C NMR data of compounds 5(6)a-f show great similarity. The assignment of proton and carbon resonances has been made on the basis of double resonance experiments on compound 5(6)cand our previous studies of tropane compounds.23-25In CD30D all the proton resonances could be assigned. The Hl(5) signal appears as a
=
nonresolvable broad singlet. The H2(4)a and H2(4)b signals were assigned on the basis of the values of the corresponding couplings with Hl(5) protons for related systems. The H1, H2a, and H2P (or H5, H4a and H4P) atoms form a three-spin AMX system whose analysis leads to the establishment of their proton magnetic parameters. The assignment of the H6(7)n and H6(7)x signals was carried out from their shapes: the simpler signal being attributed to H6(7)n since 3JH6(7)nHl(5) is very small (0.5Hz) in tropane derivative^^^-^^ i.e. practically unobserved. The 'H and 13C NMR data of 5(6)f and 5(6)c are summarized in Tables 4 and 5 respectively, as representative compounds of the series studied. Conformational Study-From the lH and 13C NMR data, we postulate that compounds 5(6)a-f adopt a preferred conformation in solution similar t o that observed for compound 5(6)f in the crystalline state: (a) The pyrrolidine and piperidine rings adopt a flattened N8 envelope and distorted chair conformation puckered at N8 and flattened at C3 as evidenced by the lH NMR spectra where the Wllz values for the Hl(5) signals of 9-11 Hz correspond to a tropane system with the piperidine ring in a chair conformation. In com) greater than 3 J ~ ~ ( 4 ) a - ~ 1 ( 5 ) , pounds S(6)a-C 3 J ~ 2 ( 4 ) p ~ 1 ( 5 is consequently, the dihedral angle H2(4)a-C-C-H1(5) is greater than H2(4)/3-H1(5). In the 13C NMR spectra, the chair conformation adopted by the piperidine ring is confirmed by the 6 C2(4) value^.^^-^^ (b) The 6 of the N-substituent is in agreement with an equatorial positon for this (c) The groups linked t o the imidazoline ring are nearly coplanar with respect t o this ring, i.e. partial conjugation between these groups is observed as indicated by the 6 lH and I3C values for the aromatic groups in 5(6)a-f. Biological Studies-Effects of Compounds 5(6)a-f on the Binding of [3H]GR65630 to Brain Area Postrema Membranes-To test the affinities of compounds Fi(6)a-f for 5-HT3 receptors, their ability to displace the binding of [3Hl-
Table 4-Proton and Magnetic Parameters for 5(6)c and 5(6)f (CD3OD)
Coupling Constantsb(J, Hz)
Chemical Shiftsa(ppm)
HI(5) (brs) HZ(4)a (dd) H2(41B (dd) H6(7)x (m). H6(7)n (my CWW H4' (s) H2" H4" H5" H 6 H7" C h-O (S) ..
3.32
4.08
Wi/2= 9.79 HZ 1.86 2.08 2.08 1.83 2.37 3.88 6.95 (d) 6.58 (t)
Wi/2 11.78 HZ 2.55 2.65 2.42 2.28 2.86 4.33 8.83 (s) 7.93 (d) 7.34 (m) 7.34 (m) 7.57 (d)
6.95 (d)
H2(4)a-H2(4)/3 H2(4)a-H1 (5) H2(4)/3-H1(5) H2'-H4' H6'-H7' H5'-H4'
5(W
5(6)f
-14.40 1.97 3.42 2.20
--16.30 1.70 3.00 8.47 7.01
3.80
Abbreviations: brs, broad singlet; d, doublet; dd, doublet of doublets; t, triplet, m, multiplet, s, singlet; HI, HZa, and H2/3 (or H5, H4a, and H4p) form a three-spin AMX system; error f 0.05 ppm. Error, fl.3 Hz. CTabulatedchemical shifts correspond to the center of the multiplet. a
Journal of Pharmaceutical Sciences / 103 Vol. 84, No, 1, January 1995
Table 5-13C Chemical Shifts (a) for Compounds 5(f)c and 5(6)f Shift, ppm
Table 6-Ability of Various Compounds To Block the von BezoldJarisch Reflex Induced by Serotonin in Mice
Shift, ppm
Position
5(6)c
5(6)f
Position
5(6)c
5(6Y
Cornpd
Cl(5) C2(4) c3 W7) CH3N CY(CH2) C2'(C) C1"
61.86 42.54 62.47 26.26 38.80 67.26 164.22 132.79
59.85 41.00
C2" C3" C3a"
106.26 162.30
24.04
C4"
64.34 161.91
104.07 162.30 106.26
5(6)a 5(6)b 5(6)c 5(6)d
C5" C6" C7" C7a"
132.3 99.34 134.50 120.20 123.90 125.11 114.05 138.44
Dose, mg/kg
% lnhibn
Cornpd
Dose, rng/kg
% lnhibn
25 25 25 25 10 5
18.27NSa -7.51 NS 7.55 NS 63.97 Sb (p
5 m 5P)f Metoclopramide Zacopride MDL 72222
25 25 10 1 1
-3.05NS 11.34NS 66.61 S (pe 0.001) 76.74 S (p< 0.001) 63.03 S (p< 0.001)
~
120 100
1
j
T
I
5(6)1
5(6)d
5(6)e
5(6)c
5(6)a
5(6)b
Meto.
T
Zaco.
MOL72222
DRUG Figure 4-Competition experiments at two concentrations of compounds 5(6)af, MDL 72222, Zacopride, and metoclopramide using [3H]GR65630at 1 nM. Data are normalized as percentage specific binding; the nonspecific binding was obtained in the presence of 30pm MDL 72222; they are the means (SEM) of three triplicate experiments. metoclopramide (n=2) 120
A 0
(3
I a
z
m I!
u.
8P v)
E
-9
-8
-7
-6
-5
-4
Log ([Drugl) (MI Figure 5-Representative competition curves for MDL72222, metoclopramide, Zacopride, and compound 5(6)d, using [3H]GR65630(1 nM) as radioligand. Data are means (SEM) of the number of experiments shown in parentheses.
GR65630 to membranes from the brain stem area postremaZ6 was measured. In three duplicate experiments, increasing concentrations of [3H]GR65630produced a saturation equilibrium isotherm. The Scatchard analysis of the results gave a dissociation constant, KDof 1.39 f 0.5 nM and a maximum number of binding sites, Bma, of 84.3 f 0.5 fmol/mg of protein. For displacement experiments, 1nM C3H1GR65630was used. 104 /Journal of Pharmaceutical Sciences Vol. 84, No. 1, January 1995
NS = not significant. S = significant.
In the initial screening, displacement of [3HlGR65630 binding was studied a t two concentrations of each compound. At 3 pM, compounds 5(6)e and 5(6)b, and 5(6)a did not displace the binding of [3H]GR65630. Compounds 5(6)f and 5(6)c displaced the binding by 20%, and compound 5(6)dby 50%. This last value was similar t o that exhibited by the selective 5-HT3 receptor antagonist MDL 72222. Metoclopramide displaced [3H]GR65630binding by 20% and Zacopride by 90%. To corroborate this, a full concentration-displacement curve was performed on compounds 5(6)d, MDL 72222, metoclopramide, and Zacopride. The IC50for compound 5(6)d was 3.03 f 0.6 x loe6 M and the Ki = 1.76 f 0.4 x M; for MDL 72222, the IC50 was 6.06 f 5.1 x M and the Ki = 3.52 i 2.9 x lop7 M. For metoclopramide, the IC5,, was 10.8 x 10-5 M and the Ki = 5.7 x M while Zacopride M and a Ki = 3.92 f 1.9 x had an IC50of 5.97 f 2.9 x M. Inhibition of the von Bezold-Jarisch Reflex-From Table 6 it can be clearly seen that compound 5(6)d displays the ability t o antagonize the von Bezold-Jarisch reflex in rats to an extent similar to that of metoclopramide. At a dose of 10 mgkg, 5(6)d causes 63.8% inhibition of the reflex compared with 66.6% for metoclopramide at a similar dosage.
MDL 72222 (n=5) 5@)d (n=6)
0 racopride (n=3)
100
a
Discussion The selective 5-HT3 receptor antagonists reported to date may be represented by the structure Ar-(carbony1)-N, in which an aromatic group is linked by a carbonyl-containing moiety to a basic amine. This simple model can be further refined by considering the three-dimensional aspects of the pharmacophore. The carbonyl group is coplanar to an aromatic group, and the interatomic distances in the 5-HT3 receptor antagonist pharmacophore are in adequate ranges (carbonyl oxygen-an aromatic center, ca. 3-4 A; carbonyl oxygen-basic nitrogen center, ca. 5 8;basic nitrogen center-aromatic center, ca. 7-8 A.27-30 One of the most important structural factors is the coplanarity between the aromatic ring and the carbonyl moiety. However, according to molecular modeling studies on l-alkyl-2-oxo-l,2-dihydroquinoline-4-carboxylic acid derivatives, which possess high affinity for the 5-HT3 receptors, their minimum energy did not have such c ~ p l a n a r i t y .Bearing ~~ in mind that the more stable conformation of compounds 5(6)a-f in solution is similar to that found for 5(6)f in the solid state, and by considering the interatomic distances (Table 3) deduced from the X-ray data in compound 5(6)f,it can be seen that the C=N group of the imidazoline ring closely approximates the position of the ester or amide carbonyl group of the pharmacophore model. Given that compound 5(6)d demonstrates 5-HT3 receptor antagonism comparable to that of MDL72222 and metoclopramide, it is likely that the imidazoline ring may provide a useful bioisosteric replacement for the carbonyl group in 5-HT3 antagonists.
Experimental Section
2-~3,5-Dichlorophenyl)-8-methyl-8-azabicyclo[3.2.lloctane3-spiro-4'(5')-imidazoline,B(6)d-Yield 38%. Mp 139.0-140 "C (from acetone). MS: mlz 324 (M+). Anal.-Calcd for C16H19Except where otherwise stated, the following procedures were ClzN3.Hz0: C, 56.14; H, 6.14; N, 12.28. Found: C, 56.02; H, 6.10; N, adopted: 'H (600 MHz) were recorded on a Bruker AM-600 spec12.18. trometer in perdeuteriomethanol. Spectral parameters included 2'-(N-oxido-4-pyridy1)-8-methyl-8-azabicyclo[3.2.l]octane-3sweep widths of 7000 Hz in 16K memory and acquisition times of 1 spiro-4(5')-imidazoline dihydrochloride, 5(6)e-Yield 29%. Mp min and 32 s over 32 transients. Chemical shifts are reported in ppm 232-234 "C (from methanol). MS: rnlz 272 (M+). Anal.-Calcd for relative to methanol. Coupling constants were evaluated by firstC15HzzClzN40: C, 52.18; H, 6.42; N, 16.23. Found: C, 52.18; H, 6.48; order rules with a n estimated accuracy of 0.5 Hz. The 13C NMR N, 16.44. spectra were obtained at 75.429 MHz on a Varian UNITY 300 2-~~--indol-3-yl)-8-methyl-8-azabicyclo[3.2.lloct~e-3-spiro spectrometer in perdeuteriomethanol. The spectra were obtained 4'(5')-imidazoline dihydrochloride, 5(6)f-Yield 52%. Mp 255with a 16501 Hz spectral window in 64K memory, with acquisition 257 "C (from methanol). MS: rnlz 294 (M+). And-Calcd for time of 1s and delay of 1s. Mass spectra were recorded on a HewlettPackard 5890 mass spectrometer. Elemental analyses were perC ~ ~ H Z ~ C ~ ~ N ~C,- 57.14 M ~ O H, H :7.04; N, 14.03. Found: C, 57.39, H, 7.14; N, 14.06. formed on a Perkin-Elmer model 24093 elemental analyzer. The IR spectra were recorded on a Perkin-Elmer 883 spectrophotometer in Pharmacological Methods-Serotonin-Znduced uon Bezoldthe solid state (KBr). Organic solvents were purified when necessary Jarisch Reflex in Mice-Compounds 5(6)a-f were evaluated for by the methods described by Perrin or were purchased from Aldrich antagonism of the von Bezold-Jarisch reflex evoked by serotonin (5Chemical Co. Thin layer and preparative chromatographies were HT) in anesthetized mice by a modification of the method of Saxena performed on basic alumina plates and gravity columns. Melting Swiss CD-1 mice (23-28 g, Charles River) and L a ~ a n g .Female ~~ points were taken in open capillary tubes on a n Electrothermal were fasted for 15 h before the experiment, but water was freely IA6304 apparatus and are uncorrected. available. Compounds (20 mgkg) suspended in 0.25% aqueous Synthesis of 8-Methyl-8-azabicyclo[3.2.l]octan-3-one-This xanthan gum (Sanofi) solution were given orally. Metoclopramide compound was obtained according to literature method^^^-^^ o r was (Sigma), Zacopride (Glaxo), and MDL 72222 (Research Biochemicals commercially available. Inc.) were used as reference compounds. One group received only Synthesis of 3~-Amino-8-methyl-8-azabicyclo[3.2.l]octane-vehicle and was used as a control. 3a-carbonitrile-A mixture of potassium cyanide (4.48 g, 68.9 mmol), Forty-five minutes later, the mice were anesthetized with urethane ammonium chloride (3.69 g, 68.9 mmol), and N-methyltropinone (10.0 (Aldrich) (1.25 gkg, ip) and electrocardiogram and heart rate were g, 68.9 mmol) was added to 14 mL of HzO and stirred at 20 "C for 48 continuously monitored and recorded (Hugo Sachs Electronik, HSEh in a stoppered flask. The suspension was then cooled to 0-5 "C 571, HSE-567, HSE-WR3310). Fifteen minutes later (1h after oral for 1 h and filtered. The resulting solid was taken up into ethyl treatment) serotonin (Sigma) (0.25 mgkg) was given intravenously acetate, filtered, and dried over MgS04. The aminonitrile was then and changes in heart rate were quantified. Results were computed precipitated from solution with petroleum ether (9.90 g, 83.4%) mp to show the percent von Bezold-Jarisch reflex inhibition in compari72-74 "C (lit.25mp 72-74 "C). son with the control group (Table 6). Statistical significance was Synthesis of 3a-(aminomethyl)-8-rnethyl-8-azabicyclo[3.2.1]- determined by Student's t test for nonpaired data. Significance level octyl-3B-amine-The aminonitrile (4.5 g, 27.3 mmol) was added in was considered at P < 0.05%. portions over 0.5 h to a suspension of LiAlH4 (2.5 g, 67.5 mmol) in pHlGR65630 Binding-t3H]GR65630 (63.74 Cit'mmol) was obtained anhydrous EtzO (30 mL) at 0-10 "C under a Nz atmosphere. The from DuPont. MDL 72222, Zacopride, and metoclopramide were reaction was stirred at reflux for 48 h and then quenched a t 0-10 "C obtained from the sources indicated above. Calf brains were obtained with 2.5 mL of H20, followed by 2.5 mL of 4 N NaOH and then 2.5 in ice from a local slaughterhause and dissected on ice, using a mL of HzO. The inorganic solid was filtered and then extracted with procedure adapted from the method of Glowinski and I v e r ~ e n The .~~ EtzO using a Soxhlet extractor. The organic solutions were then area postrema was scraped away from the surrounding tissues and combined, dried, and stripped to dryness. The residue was refluxed placed in cold buffer. Approximately 100 mg of wet weight tissue in methanol containing a drop of diluted HCI for 24 h. The methanol was dissected per brain. The tissue was homogenized in ice-cold 50 solution was filtered and stripped to dryness (yield 80%) MS: mlz mM Tris-HC1,0.5 mM EDTA, 10 mM MgS04, pH 7.4, and centrifuged 170 (M 1). at 3000g for 15 min. The pellet was resuspended in buffer (in a Synthesis of Carboximidate Hydrochlorides: G e n e r a l Prohomogenizer), incubated a t 37 "C for 15 min, and then recentrifuged cedure-Dry hydrogen chloride gas was bubbled through a solution twice a t 30000g for 15 min (with a resuspension between each centrifuging). The final pellet was resuspended in 50 mM Tris-HC1, of the appropriate nitrile (14.0 mmol) in dry methanol (40 mL). The solution was allowed to stand at 0-5 "C for 24 h. Addition of dry 0.5 mM EDTA, 10 mM MgS04, 0.1% ascorbate, M paragyline, ether (30 mL) precipitated the corresponding imidate hydrochloride. and 140 mM NaCl, aliquoted, and frozen. For the indole imidate the solution was left at room temperature for Assays were performed three times, with sets of tubes in triplicate 24 h. in competition experiments and in duplicate in saturation experiSynthesisof Imidazolines ti(6)a-f: G e n e r a l Procedure-The ments, in a 2.0 mL volume containing 2 mg of wet weight tissue appropriate carboximidate hydrochloride (4.79 mmol) and the tropane(added last). 1,2-ethylenediamine 4 (6.0 mmol) were added t o dry methanol (25.0 Tubes were incubated a t room temperature for 30 min, filtered on mL) and stirred a t room temperature for 24 h under a Nz atmosphere. glass microfibre filters (Whatman GFlC), and washed three times with The reaction was heated to reflux for 3 h and then cooled to 20 "C. 7 mL of ice-cold buffer. The filters were counted in a liquid The solvent was removed under reduced pressure and the residue scintillation analyzer (Packard Tri-carb 1500) in 4 mL of aqueous purified by column chromatography using basic alumina and CHzcounting scintillation fluid (Beckman Ready Micro), following 12 h of C12:EtOH as eluent. Trituration of the resulting oil with EtzO equilibration. afforded a white solid which was recrystallized from acetone. TreatKinetic saturation constants and competition experiments were ment of the bases with methanolic HCl furnished the dihydrochloride analyzed by a computer program. Protein determinations were salts which were recrystallized from methanol. performed following the method a Lowry et al.34using bovine serum 2'-Phenyl-8-methyl-8-azabicyclo[3.2.lloctane-3-spiro-4'(5')albumin (BSA) as a standard. imidazoline, B(6)a-Yield 43%. Mp 175- 177 "C (from acetone). MS: mlz 255 (M+). And-Calcd for C16H~iN3:C, 75.26; H, 8.29; N, 16.46. Found: C , 75.46; H, 8.49; N, 16.51. References and Notes 2-(3-Methoxyphenyl)-8-methyl-8-azabicyclo-[3.2.lloctane-3spiro-4(5')-imidazoline, B(6)b-Yield 41%. Mp 152-154 "C (from 1. Gaddum, J . M.; Picarelli, Z. P. Br. J . Pharmucol. Chemoter. 1957, acetone). MS: rnlz 285 (M+). Anal.-Calcd for C17Hz3N30: C, 71.55; 12, 323. H, 8.12; N, 14.72. Found: C, 71.46; H, 8.32; N, 14.67. 2. Kilpatrick, G. J.; Jones, B. J.; Tyers, M. B. Eur. J . Pharmacol. 2-(3,5-D~etho~henyl)-8-methyl-8-~abicyclo[3.2.ll~~e- 1989,159,127. 3-spiro-4'(5')-imidazoline, B(6)c-Yield 48%. Mp 167- 168 "C (from 3. Barnes, J. M.; Barnes, N. M.; Costall, B.; Deakin, J. F. W.; Ironside, J . W.; Kilpatrick, G. J.; Naylor, R. J.; Rudd, J. A.; acetone). MS: m l . 315 (M+). And-Calcd for C~jH23N302:C, 68.54; Simson, M. D. Neurosci. Lett. 1990, 111, 80. H, 7.99; N, 13.20. Found: C, 68.30; H, 8.30; N, 13.21.
+
Journal of Pharmaceutical Sciences / 105 Vol. 84, No. 1, January 1995
4. Kilpatrick, G. J.; Bunce, K. T.; Tyers, M. B. Med. Res. Rev. 1990, 10, 441. 5. Bunce, K.;Tyers, M.; Beraneck, P. Trends Pharmacol. Sci. 1991, 12, 46. 6. Hesketh, P. J.; Gandara, D. R. J . Natl. Cancer Znst. 1991,83, 613. 7. Costall, B.; Naylor, R. J.;Tyers, M. B. Pharmacol. Ther. 1990, 47,181. 8. Jones, B. Drugs News Prospects 1990,3, 106. 9. Lader, M. H.; Crook, T. H.; Meltzer, H. Y.; Sellers, E. M. 5th World Coneress o f Bioloeical Psvchiatn, Satellite Symposium, - Florence, Ztaly, 1991. 10. Fozard, J. R. Arch. Pharmacol. 1984,326,36. 11. Smith, W. W.; Sancilio, L. F.; Ower-Atepo, J. B.; Naylor, R. J.; Lambert, L. J . J . Pharm. Pharmacol. 1988,40,301. 12. Swain, C. J.; Baker, R.; Kneen, C.; Moseley, J.; Saunders, J.; Seward, E. M.; Stevenson, G.; Beer, M.; Stanton, J. J . Med. Chem. 1991,34,140-151. 13. Swain, C. J.; Baker, R.; Kneen, C.; Herbert, R.; Moseley, J.; Saunders, J.; Seward, E. M.; Stevenson, G.; Beer, M.; Stanton, J.; Watling, K.; Ball, R. G. J . Med. Chem. 1992,35,1019-1031. 14. Arias, M. S.;Galvez, E.; Izquierdo, M. L.; Burgos, C. J . Mol. Struct. 1986,147,381 and references therein. 15. Walker, N.; Stuart, D. Acta Crystallogr. 1983,A39, 158-166. 16. Main, P., Fiske, S. J., Hull, S. E., Lessinger, L., Germain, G., Declercq, J. P., Woolfson, M. M. MULTANSO. A System of Computer Programs for Automatic Solution of Crystal Structures from X-ray Diffraction Data. University of York, England, and University of Louvain, Belgium, 1980. 17. Beurslens, P. R.;Bosman, W. P.; Doesburg, H. M.; Gould, R. 0.;van der Hark, TH.E.M.; Haltiwanger, R. C. DIRDIF system of Computer Programs, Technical Report 19983/1 CrystallomaDhv Laboratorv: ” , Toernooiveld 6525 ED Niimeaen. The Nztherhnds, 1983. 18. Stewart, J. M., Kundell, F. A.; Baldwin, J. C. The X-RAY80 System of crystallographic Programs, Computer Saence, University of Maryland: College Park, MD, 1980. 19. Martinez-Riaoll. M.: Cano. F. H. PESOS. A Comnuter Program for the Autdmatic Treatment of Weighing Schemes. InsGtuto Rocasolano, CSIC, Madrid, Spain, 1975.
-
I
106 / Journal of Pharmaceutical Sciences Vol. 84, No. 1, January 1995
-
,
20. Nardelli, M. Comput. Chem. 1983,7,95-98. 21. International Tables for X-ray Crystallography 1974; Birmingham: Kynoch Press. (Present distributor, C. Reidel: Dordrecht), 1974,Vol. IV. 22. Motherwill, W. D. S.; Clegg, W. Pluto Program for Plotting Molecular and Crystal Structures. University of Cambridge, England, 1978. 23. Arias, M.S.;Galvez, E.; Izquierdo, M. L.; Burgos, C. J . MoZ. Struct. 1986,147,381 and references therein. 24. Izquierdo, M. L.; Galvez., E.; Burgos, C.; Arias, M. S.; SanzAparicio, J.; Fonseca, I.; Bellanato, J. J . Mol. Struct. 1993,291, 1. 25. Galvez, E.; Izquierdo, M. L.; Burgos, C.; Arias, M. S.; SanzAparicio, J.; Fonseca, I.; Gago, F.; Baldominos, G.; Lopez, P.; Prieto, J. C. J . Pharm. Sci. 1993,82,794. 26. Miller, K.; Weisberg, E.; Fletcher, P.; Teitler, M. Synapse 1992, 11,58. 27. Schmidt, A. W.; Peroutka, S. J. Mol. Pharmacol. 1989,36,505. 28. Hibert, M. F.; Hofmann, R.; Miller, R. C.; Carr, A. A. J . Med. Chem. 1990,33,1594. 29. Smidt, A. W.; Peroutka, S. J . Mol. Pharmacol. 1990,38,511. 30. Evans, S.M.; Galdes, A.; Gall, M. Pharmacol. Biochern. Behav. 1991,40,1033. 31. Hayashi, H.;Miwa, Y.; Miki, I.; Ichikawa, S.; Yoda, N.; Ishii, A.; Kono, M.; Suzuki, F. J . Med. Chem. 1992,35,4893. 32. Saxena, P. R.; Lawang, A. Arch. Znt. Pharmacodyn. 1985,277, 235. 33. Glowinsky, J.; Iversen, L. L. J. Neurochem. 1966,13,655. 34. Lowry, 0.H.; Rosenbrough, N. J.; Farr, A. L.; Randall, R. J. J . Biol. Chem. 1951,193,265.
Acknowledgments We thank the Commission of the EuroDean Communities for the funding of this research under the SCIENCE program. (Grant no. ERBSCl*CT915120). Supplementary Material Available-Data from the crystallographic study of compound 5(6)f are available from the authors.
H3C”$p+.m
3a
I a C6HS
d 3.5-diCIC6H3
b 3-MeOC6H,
e 4-Pyridin-N-oxide
c 3,5-diMeOC6H3
f 3-indolyl
6
The 5-HT3 receptor subtype, although originally thought to be confined t o the peripheral nervous system,l is now known to be present in the central nervous ~ y s t e m . Several ~,~ highly selective antagonists of the 5-HT3 receptor have been described4 and a number of these are the subject of clinical studies. In particular, Ondansetron, Granisetron, and ICS 205-930 have been shown to be highly effective inhibitors of cytotoxic drug induced emesis in man5r6and are currently being marketed for this purpose. Animal s t ~ d i e s ~suggest z~,~ that this class of compounds may also possess anxiolytic, antipsychotic, or cognition-enhancing activity. Preliminary clinical studies have provided support for these psychiatric uses in m a n g Many of the currently reported 5-HT3 antagonists are aromatic esters or amides of appropriate bicyclic alcohols or amines,lOJ1though some heterocyclics (such as oxadiazoles and oxazolines)which were shown to be isosteric with respect t o the ester group have recently been successfully introdu~ed.’~J~ By considering that the C=N group of the imidazoline ring could possibly act as a bioisosteric replacement for esters or amides, we have designed compounds 5(6)a-f as potential 5-HT3 antagonists. We report here the synthesis and structural analysis, carried out with the aid of ‘H and 13CNMR spectroscopy, of compounds 5(6)a-f, derived from tropane-3-spiro-4‘-imidazoline.In order to determine their preferred conformations, both in solution and in the solid state, the crystal structure of compound 5(6)f has also been determined. @Abstractpublished in Advance ACS Abstracts, October 1, 1994.
0 1995, American Chemical Sociely and American Pharmaceutical Association
1
H3c”T I
Introduction
h
LiAIHI
RC:NH$XOMe) 29-52 Ya
5
Scheme 1
Results Chemistry-Compounds 5(6)a-f were prepared as shown in Scheme 1. Reaction of tropinone 1 with ammonium chloride and potassium cyanide in water gave the aminonitrile 3a. The most stable conformation of 1 in CDC4 is as represented in Scheme l.14 By assuming the same preferred conformation for the ketimine 2 in solution, the more favorable p attack of the cyanide should be expected, thus leading to the 3/3 isomer. Therefore, the formation of the more stable 3 a aminonitrile [in the case of 3a and 3/3,the most severe steric interaction is exerted by the NH2 group on H6(7)n]can be attributed to the thermodynamic control of the process. With an eye on forming the imidazoline derivative of the 3p cyanide intermediate, exhaustive attempts a t its synthesis were made but proved futile in all cases, the a isomer invariably being the aminonitrile isolated. By treating compound 3 with lithium aluminium hydride in ether, the amine 4 was obtained. Compounds 5(6)a-f were obtained by treatment of 4 with the corresponding methyl imidate hydrochloride. Structural Study-Description of the Structure of 5(6)pTable 115-21 summarizes the experimental and structural displays the title comdetermination procedures. Figure lz2 pound and the numbering used in the crystallographic study (supplementary material is available from the authors).
0022-3549/95/3184-0101$09.00/0
Journal of Pharmaceutical Sciences / 101 Vol. 84, No. I, January 1995
Table 1-Experimental Data and Structural Refinement Procedures Value
Parameter Crystal data Formula Symmetry Unit cell determinations Unit cell dimensions (A) Packing: V(A3), Z Dc (g ~ m - ~M, ) ,F(OO0) P (cm-9 Experimental data Technique
CI~HZ~N~OCIZ Monoclinical, Rl/n Least squares fit from 60 reflections (0< 15") 7.661(1), 26.432(4), 9.986(2), 90.0, 98.43(1), 90.0 2000.3(6), 4 1.3261,399.36,848 3.389 Four circle diff ractometer: Enraf-Nonius, Cad-4, Bisecting geometry. Grafite-oriented monochromator:Mo Ka. w scans; scan width, 1" up to a max of 28"
Number of reflections Measured Observed Range of hkl Value of Rn, Max-min transmission factors Solution and refinement Solution Refinement H atoms w scheme Final AFpeaks Final Rand R, Computer and programs Scattering factors Anomalous dispersion
1C14 c2
4773 1026 (241) criterion) 0 10,O 34, -13 13 0.01 1.244,0.555 Direct methods and Fourier synthesis Full-matrix least squares on Fobs.Anisotropic thermal parameters of C6 fixed Difference synthesis. H52, H92, H93, and H262 not refined Empirical as to give no trends in (WAF)vs (Fobs)and (sin el^) 0.3 e/A3 0.054,0.013 Vax 6410, D i f a b ~ ,Multan80,16 '~ Dirdif,i7 Xray System,'* Pesos,lg Parstzo lnfemafional Tables for X-Ray Crysfallographfl lntemational Tables for X-ray Crysfal/ographfl
b
AC15
?2O Figure 1-PLUTO view of the molecule 5(6)f showing the atomic numbering.22
The bicyclic system is in the chair-envelope conformation commonly found in these kind of compounds. The chair is flattened at the C4 atom to release steric hindrance, C4 and N1 being 0.49(1) and 0.86(1) A, respectively, from the mean plane defined by the other four atoms. N1, on the other hand, is in the flap of the envelope, 0.70(1) A away from the plane defined by the rest of the atoms in this ring. The molecule presents a pseudo-mirror plane, passing through the N1, C4, and C9 atoms. The imidazoline ring is almost in this plane with an interplanar angle of only 5". The indole moiety, on the contrary, deviates from this symmetry, defining an interplanar angle of 24". C123 and C124 are in opposite positions with res ect to the pseudo-mirror plane, being 0.639(4) and 1.472(3) respectively, out of this plane. As shown by the bond lengths, the electrons are delocalized along NlO-Cll-Nl2 and the indole moiety, which maintains the interplanar angle of 24" between the indole and the imidazoline ring. Consequently, there is no difference between N10 and N12 or between the CllN12 and the CllNlO bond lengths. Both N atoms of the imidazoline ring form hydrogen bonds with the two chlorine atoms. However, one of them acts as the counter-chlorideion (C124), while the other seems to be present as a hydrochloride of crystallization and
1,
102 / Journal of Pharmaceutical Sciences Vol. 84, No. 1, January 1995
Figure 2-Unit cell packing for compound 5(6)f. also interacts with the CH30 group (see Table 2) of the solvent, which in turn interacts with the bicyclic N1 atom. The hydrogen-bonding pattern is completed by a hydrogen bond between the indole N16 and the C124 atom from a different asymmetric unit. Two molecules related by a center of symmetry form a dimer through two hydrogen bonds: N12-Hl2-Cl24 and Nl6-Hl6-Cl24. These dimers form isolated units stacked along the "a" axis. The first hydrogen bond links the C124 atom to the N12 one in the same molecule ( x , y , z ) and the second one links the C124 to the N16 of a
KO~ 1 . 3 9 t 0.5 10 80 7
-'M
Table 3-Selected Interatomic Distances of 5(6)f in the Solid State
70 -
60 50 -
10
2
1
4
3
[ WGR65630 ] (nM)
Figure 3-Saturable specific binding of [3H]GR65630 to membranes of bovine area postrema (2 mg wet weight, 174.62 p g protein) using filtration binding methodology. Assays were performed in duplicate; error bars represent SEM of three experiments performed on separate batches of membranes. Table 2-Interatomic Bond Distances and Angles of 5(6)f N1-HI
I .o(i) A
C123-H23 1.7(2) 8, N10-H1 0 1.0(1) A N12-Hl2 1.0(1) A N16-HI 6 0.7(2) A 1 - x , 1 -y,-z
Atom 2
Interatomic Distance, 8,
Atom 1
Atom 2
Interatomic Distance, A
N8 N8 N8 N8
Xa Yb N3' N1'
9.2 7.6 4.1 5.3
N3' NI' N3' N1'
X X Y Y
5.2 4.4 3.5 3.6
= Gravimetric center of the indole six-membered aromatic ring. Gravimetric center of the indole five-membered ring.
40 -
0
Atom 1
Em,= 84.32 k 21.9 h o t . mg prot-'
H1...025 1.7(1) A H23...025 2.5(1) 8, H IO-CI23 2.3(1) A H12-CI24 2.2(1) A H16-CI24' 2.5(2) A
N1...025 2.72(1) A C123...025 3.04(1) A N1O - m 3 3.24(1) A N12-CI24 3.15(1) A N16..C124' 3.18(1) A
NI-H1.-025 164(l)O C123-H23...025 92(1)" N10-HI O-CI23 168(1)0 N12-H 12-CI24 165(I)" N16-H16..C124' 165(1)O
different asymmetric unit (-z, 1 - y, -21, the C1 atom acting as the nexus between the two molecules. NMR Spectra-The 'H and 13C NMR data of compounds 5(6)a-f show great similarity. The assignment of proton and carbon resonances has been made on the basis of double resonance experiments on compound 5(6)cand our previous studies of tropane compounds.23-25In CD30D all the proton resonances could be assigned. The Hl(5) signal appears as a
=
nonresolvable broad singlet. The H2(4)a and H2(4)b signals were assigned on the basis of the values of the corresponding couplings with Hl(5) protons for related systems. The H1, H2a, and H2P (or H5, H4a and H4P) atoms form a three-spin AMX system whose analysis leads to the establishment of their proton magnetic parameters. The assignment of the H6(7)n and H6(7)x signals was carried out from their shapes: the simpler signal being attributed to H6(7)n since 3JH6(7)nHl(5) is very small (0.5Hz) in tropane derivative^^^-^^ i.e. practically unobserved. The 'H and 13C NMR data of 5(6)f and 5(6)c are summarized in Tables 4 and 5 respectively, as representative compounds of the series studied. Conformational Study-From the lH and 13C NMR data, we postulate that compounds 5(6)a-f adopt a preferred conformation in solution similar t o that observed for compound 5(6)f in the crystalline state: (a) The pyrrolidine and piperidine rings adopt a flattened N8 envelope and distorted chair conformation puckered at N8 and flattened at C3 as evidenced by the lH NMR spectra where the Wllz values for the Hl(5) signals of 9-11 Hz correspond to a tropane system with the piperidine ring in a chair conformation. In com) greater than 3 J ~ ~ ( 4 ) a - ~ 1 ( 5 ) , pounds S(6)a-C 3 J ~ 2 ( 4 ) p ~ 1 ( 5 is consequently, the dihedral angle H2(4)a-C-C-H1(5) is greater than H2(4)/3-H1(5). In the 13C NMR spectra, the chair conformation adopted by the piperidine ring is confirmed by the 6 C2(4) value^.^^-^^ (b) The 6 of the N-substituent is in agreement with an equatorial positon for this (c) The groups linked t o the imidazoline ring are nearly coplanar with respect t o this ring, i.e. partial conjugation between these groups is observed as indicated by the 6 lH and I3C values for the aromatic groups in 5(6)a-f. Biological Studies-Effects of Compounds 5(6)a-f on the Binding of [3H]GR65630 to Brain Area Postrema Membranes-To test the affinities of compounds Fi(6)a-f for 5-HT3 receptors, their ability to displace the binding of [3Hl-
Table 4-Proton and Magnetic Parameters for 5(6)c and 5(6)f (CD3OD)
Coupling Constantsb(J, Hz)
Chemical Shiftsa(ppm)
HI(5) (brs) HZ(4)a (dd) H2(41B (dd) H6(7)x (m). H6(7)n (my CWW H4' (s) H2" H4" H5" H 6 H7" C h-O (S) ..
3.32
4.08
Wi/2= 9.79 HZ 1.86 2.08 2.08 1.83 2.37 3.88 6.95 (d) 6.58 (t)
Wi/2 11.78 HZ 2.55 2.65 2.42 2.28 2.86 4.33 8.83 (s) 7.93 (d) 7.34 (m) 7.34 (m) 7.57 (d)
6.95 (d)
H2(4)a-H2(4)/3 H2(4)a-H1 (5) H2(4)/3-H1(5) H2'-H4' H6'-H7' H5'-H4'
5(W
5(6)f
-14.40 1.97 3.42 2.20
--16.30 1.70 3.00 8.47 7.01
3.80
Abbreviations: brs, broad singlet; d, doublet; dd, doublet of doublets; t, triplet, m, multiplet, s, singlet; HI, HZa, and H2/3 (or H5, H4a, and H4p) form a three-spin AMX system; error f 0.05 ppm. Error, fl.3 Hz. CTabulatedchemical shifts correspond to the center of the multiplet. a
Journal of Pharmaceutical Sciences / 103 Vol. 84, No, 1, January 1995
Table 5-13C Chemical Shifts (a) for Compounds 5(f)c and 5(6)f Shift, ppm
Table 6-Ability of Various Compounds To Block the von BezoldJarisch Reflex Induced by Serotonin in Mice
Shift, ppm
Position
5(6)c
5(6)f
Position
5(6)c
5(6Y
Cornpd
Cl(5) C2(4) c3 W7) CH3N CY(CH2) C2'(C) C1"
61.86 42.54 62.47 26.26 38.80 67.26 164.22 132.79
59.85 41.00
C2" C3" C3a"
106.26 162.30
24.04
C4"
64.34 161.91
104.07 162.30 106.26
5(6)a 5(6)b 5(6)c 5(6)d
C5" C6" C7" C7a"
132.3 99.34 134.50 120.20 123.90 125.11 114.05 138.44
Dose, mg/kg
% lnhibn
Cornpd
Dose, rng/kg
% lnhibn
25 25 25 25 10 5
18.27NSa -7.51 NS 7.55 NS 63.97 Sb (p
5 m 5P)f Metoclopramide Zacopride MDL 72222
25 25 10 1 1
-3.05NS 11.34NS 66.61 S (pe 0.001) 76.74 S (p< 0.001) 63.03 S (p< 0.001)
~
120 100
1
j
T
I
5(6)1
5(6)d
5(6)e
5(6)c
5(6)a
5(6)b
Meto.
T
Zaco.
MOL72222
DRUG Figure 4-Competition experiments at two concentrations of compounds 5(6)af, MDL 72222, Zacopride, and metoclopramide using [3H]GR65630at 1 nM. Data are normalized as percentage specific binding; the nonspecific binding was obtained in the presence of 30pm MDL 72222; they are the means (SEM) of three triplicate experiments. metoclopramide (n=2) 120
A 0
(3
I a
z
m I!
u.
8P v)
E
-9
-8
-7
-6
-5
-4
Log ([Drugl) (MI Figure 5-Representative competition curves for MDL72222, metoclopramide, Zacopride, and compound 5(6)d, using [3H]GR65630(1 nM) as radioligand. Data are means (SEM) of the number of experiments shown in parentheses.
GR65630 to membranes from the brain stem area postremaZ6 was measured. In three duplicate experiments, increasing concentrations of [3H]GR65630produced a saturation equilibrium isotherm. The Scatchard analysis of the results gave a dissociation constant, KDof 1.39 f 0.5 nM and a maximum number of binding sites, Bma, of 84.3 f 0.5 fmol/mg of protein. For displacement experiments, 1nM C3H1GR65630was used. 104 /Journal of Pharmaceutical Sciences Vol. 84, No. 1, January 1995
NS = not significant. S = significant.
In the initial screening, displacement of [3HlGR65630 binding was studied a t two concentrations of each compound. At 3 pM, compounds 5(6)e and 5(6)b, and 5(6)a did not displace the binding of [3H]GR65630. Compounds 5(6)f and 5(6)c displaced the binding by 20%, and compound 5(6)dby 50%. This last value was similar t o that exhibited by the selective 5-HT3 receptor antagonist MDL 72222. Metoclopramide displaced [3H]GR65630binding by 20% and Zacopride by 90%. To corroborate this, a full concentration-displacement curve was performed on compounds 5(6)d, MDL 72222, metoclopramide, and Zacopride. The IC50for compound 5(6)d was 3.03 f 0.6 x loe6 M and the Ki = 1.76 f 0.4 x M; for MDL 72222, the IC50 was 6.06 f 5.1 x M and the Ki = 3.52 i 2.9 x lop7 M. For metoclopramide, the IC5,, was 10.8 x 10-5 M and the Ki = 5.7 x M while Zacopride M and a Ki = 3.92 f 1.9 x had an IC50of 5.97 f 2.9 x M. Inhibition of the von Bezold-Jarisch Reflex-From Table 6 it can be clearly seen that compound 5(6)d displays the ability t o antagonize the von Bezold-Jarisch reflex in rats to an extent similar to that of metoclopramide. At a dose of 10 mgkg, 5(6)d causes 63.8% inhibition of the reflex compared with 66.6% for metoclopramide at a similar dosage.
MDL 72222 (n=5) 5@)d (n=6)
0 racopride (n=3)
100
a
Discussion The selective 5-HT3 receptor antagonists reported to date may be represented by the structure Ar-(carbony1)-N, in which an aromatic group is linked by a carbonyl-containing moiety to a basic amine. This simple model can be further refined by considering the three-dimensional aspects of the pharmacophore. The carbonyl group is coplanar to an aromatic group, and the interatomic distances in the 5-HT3 receptor antagonist pharmacophore are in adequate ranges (carbonyl oxygen-an aromatic center, ca. 3-4 A; carbonyl oxygen-basic nitrogen center, ca. 5 8;basic nitrogen center-aromatic center, ca. 7-8 A.27-30 One of the most important structural factors is the coplanarity between the aromatic ring and the carbonyl moiety. However, according to molecular modeling studies on l-alkyl-2-oxo-l,2-dihydroquinoline-4-carboxylic acid derivatives, which possess high affinity for the 5-HT3 receptors, their minimum energy did not have such c ~ p l a n a r i t y .Bearing ~~ in mind that the more stable conformation of compounds 5(6)a-f in solution is similar to that found for 5(6)f in the solid state, and by considering the interatomic distances (Table 3) deduced from the X-ray data in compound 5(6)f,it can be seen that the C=N group of the imidazoline ring closely approximates the position of the ester or amide carbonyl group of the pharmacophore model. Given that compound 5(6)d demonstrates 5-HT3 receptor antagonism comparable to that of MDL72222 and metoclopramide, it is likely that the imidazoline ring may provide a useful bioisosteric replacement for the carbonyl group in 5-HT3 antagonists.
Experimental Section
2-~3,5-Dichlorophenyl)-8-methyl-8-azabicyclo[3.2.lloctane3-spiro-4'(5')-imidazoline,B(6)d-Yield 38%. Mp 139.0-140 "C (from acetone). MS: mlz 324 (M+). Anal.-Calcd for C16H19Except where otherwise stated, the following procedures were ClzN3.Hz0: C, 56.14; H, 6.14; N, 12.28. Found: C, 56.02; H, 6.10; N, adopted: 'H (600 MHz) were recorded on a Bruker AM-600 spec12.18. trometer in perdeuteriomethanol. Spectral parameters included 2'-(N-oxido-4-pyridy1)-8-methyl-8-azabicyclo[3.2.l]octane-3sweep widths of 7000 Hz in 16K memory and acquisition times of 1 spiro-4(5')-imidazoline dihydrochloride, 5(6)e-Yield 29%. Mp min and 32 s over 32 transients. Chemical shifts are reported in ppm 232-234 "C (from methanol). MS: rnlz 272 (M+). Anal.-Calcd for relative to methanol. Coupling constants were evaluated by firstC15HzzClzN40: C, 52.18; H, 6.42; N, 16.23. Found: C, 52.18; H, 6.48; order rules with a n estimated accuracy of 0.5 Hz. The 13C NMR N, 16.44. spectra were obtained at 75.429 MHz on a Varian UNITY 300 2-~~--indol-3-yl)-8-methyl-8-azabicyclo[3.2.lloct~e-3-spiro spectrometer in perdeuteriomethanol. The spectra were obtained 4'(5')-imidazoline dihydrochloride, 5(6)f-Yield 52%. Mp 255with a 16501 Hz spectral window in 64K memory, with acquisition 257 "C (from methanol). MS: rnlz 294 (M+). And-Calcd for time of 1s and delay of 1s. Mass spectra were recorded on a HewlettPackard 5890 mass spectrometer. Elemental analyses were perC ~ ~ H Z ~ C ~ ~ N ~C,- 57.14 M ~ O H, H :7.04; N, 14.03. Found: C, 57.39, H, 7.14; N, 14.06. formed on a Perkin-Elmer model 24093 elemental analyzer. The IR spectra were recorded on a Perkin-Elmer 883 spectrophotometer in Pharmacological Methods-Serotonin-Znduced uon Bezoldthe solid state (KBr). Organic solvents were purified when necessary Jarisch Reflex in Mice-Compounds 5(6)a-f were evaluated for by the methods described by Perrin or were purchased from Aldrich antagonism of the von Bezold-Jarisch reflex evoked by serotonin (5Chemical Co. Thin layer and preparative chromatographies were HT) in anesthetized mice by a modification of the method of Saxena performed on basic alumina plates and gravity columns. Melting Swiss CD-1 mice (23-28 g, Charles River) and L a ~ a n g .Female ~~ points were taken in open capillary tubes on a n Electrothermal were fasted for 15 h before the experiment, but water was freely IA6304 apparatus and are uncorrected. available. Compounds (20 mgkg) suspended in 0.25% aqueous Synthesis of 8-Methyl-8-azabicyclo[3.2.l]octan-3-one-This xanthan gum (Sanofi) solution were given orally. Metoclopramide compound was obtained according to literature method^^^-^^ o r was (Sigma), Zacopride (Glaxo), and MDL 72222 (Research Biochemicals commercially available. Inc.) were used as reference compounds. One group received only Synthesis of 3~-Amino-8-methyl-8-azabicyclo[3.2.l]octane-vehicle and was used as a control. 3a-carbonitrile-A mixture of potassium cyanide (4.48 g, 68.9 mmol), Forty-five minutes later, the mice were anesthetized with urethane ammonium chloride (3.69 g, 68.9 mmol), and N-methyltropinone (10.0 (Aldrich) (1.25 gkg, ip) and electrocardiogram and heart rate were g, 68.9 mmol) was added to 14 mL of HzO and stirred at 20 "C for 48 continuously monitored and recorded (Hugo Sachs Electronik, HSEh in a stoppered flask. The suspension was then cooled to 0-5 "C 571, HSE-567, HSE-WR3310). Fifteen minutes later (1h after oral for 1 h and filtered. The resulting solid was taken up into ethyl treatment) serotonin (Sigma) (0.25 mgkg) was given intravenously acetate, filtered, and dried over MgS04. The aminonitrile was then and changes in heart rate were quantified. Results were computed precipitated from solution with petroleum ether (9.90 g, 83.4%) mp to show the percent von Bezold-Jarisch reflex inhibition in compari72-74 "C (lit.25mp 72-74 "C). son with the control group (Table 6). Statistical significance was Synthesis of 3a-(aminomethyl)-8-rnethyl-8-azabicyclo[3.2.1]- determined by Student's t test for nonpaired data. Significance level octyl-3B-amine-The aminonitrile (4.5 g, 27.3 mmol) was added in was considered at P < 0.05%. portions over 0.5 h to a suspension of LiAlH4 (2.5 g, 67.5 mmol) in pHlGR65630 Binding-t3H]GR65630 (63.74 Cit'mmol) was obtained anhydrous EtzO (30 mL) at 0-10 "C under a Nz atmosphere. The from DuPont. MDL 72222, Zacopride, and metoclopramide were reaction was stirred at reflux for 48 h and then quenched a t 0-10 "C obtained from the sources indicated above. Calf brains were obtained with 2.5 mL of H20, followed by 2.5 mL of 4 N NaOH and then 2.5 in ice from a local slaughterhause and dissected on ice, using a mL of HzO. The inorganic solid was filtered and then extracted with procedure adapted from the method of Glowinski and I v e r ~ e n The .~~ EtzO using a Soxhlet extractor. The organic solutions were then area postrema was scraped away from the surrounding tissues and combined, dried, and stripped to dryness. The residue was refluxed placed in cold buffer. Approximately 100 mg of wet weight tissue in methanol containing a drop of diluted HCI for 24 h. The methanol was dissected per brain. The tissue was homogenized in ice-cold 50 solution was filtered and stripped to dryness (yield 80%) MS: mlz mM Tris-HC1,0.5 mM EDTA, 10 mM MgS04, pH 7.4, and centrifuged 170 (M 1). at 3000g for 15 min. The pellet was resuspended in buffer (in a Synthesis of Carboximidate Hydrochlorides: G e n e r a l Prohomogenizer), incubated a t 37 "C for 15 min, and then recentrifuged cedure-Dry hydrogen chloride gas was bubbled through a solution twice a t 30000g for 15 min (with a resuspension between each centrifuging). The final pellet was resuspended in 50 mM Tris-HC1, of the appropriate nitrile (14.0 mmol) in dry methanol (40 mL). The solution was allowed to stand at 0-5 "C for 24 h. Addition of dry 0.5 mM EDTA, 10 mM MgS04, 0.1% ascorbate, M paragyline, ether (30 mL) precipitated the corresponding imidate hydrochloride. and 140 mM NaCl, aliquoted, and frozen. For the indole imidate the solution was left at room temperature for Assays were performed three times, with sets of tubes in triplicate 24 h. in competition experiments and in duplicate in saturation experiSynthesisof Imidazolines ti(6)a-f: G e n e r a l Procedure-The ments, in a 2.0 mL volume containing 2 mg of wet weight tissue appropriate carboximidate hydrochloride (4.79 mmol) and the tropane(added last). 1,2-ethylenediamine 4 (6.0 mmol) were added t o dry methanol (25.0 Tubes were incubated a t room temperature for 30 min, filtered on mL) and stirred a t room temperature for 24 h under a Nz atmosphere. glass microfibre filters (Whatman GFlC), and washed three times with The reaction was heated to reflux for 3 h and then cooled to 20 "C. 7 mL of ice-cold buffer. The filters were counted in a liquid The solvent was removed under reduced pressure and the residue scintillation analyzer (Packard Tri-carb 1500) in 4 mL of aqueous purified by column chromatography using basic alumina and CHzcounting scintillation fluid (Beckman Ready Micro), following 12 h of C12:EtOH as eluent. Trituration of the resulting oil with EtzO equilibration. afforded a white solid which was recrystallized from acetone. TreatKinetic saturation constants and competition experiments were ment of the bases with methanolic HCl furnished the dihydrochloride analyzed by a computer program. Protein determinations were salts which were recrystallized from methanol. performed following the method a Lowry et al.34using bovine serum 2'-Phenyl-8-methyl-8-azabicyclo[3.2.lloctane-3-spiro-4'(5')albumin (BSA) as a standard. imidazoline, B(6)a-Yield 43%. Mp 175- 177 "C (from acetone). MS: mlz 255 (M+). And-Calcd for C16H~iN3:C, 75.26; H, 8.29; N, 16.46. Found: C , 75.46; H, 8.49; N, 16.51. References and Notes 2-(3-Methoxyphenyl)-8-methyl-8-azabicyclo-[3.2.lloctane-3spiro-4(5')-imidazoline, B(6)b-Yield 41%. Mp 152-154 "C (from 1. Gaddum, J . M.; Picarelli, Z. P. Br. J . Pharmucol. Chemoter. 1957, acetone). MS: rnlz 285 (M+). Anal.-Calcd for C17Hz3N30: C, 71.55; 12, 323. H, 8.12; N, 14.72. Found: C, 71.46; H, 8.32; N, 14.67. 2. Kilpatrick, G. J.; Jones, B. J.; Tyers, M. B. Eur. J . Pharmacol. 2-(3,5-D~etho~henyl)-8-methyl-8-~abicyclo[3.2.ll~~e- 1989,159,127. 3-spiro-4'(5')-imidazoline, B(6)c-Yield 48%. Mp 167- 168 "C (from 3. Barnes, J. M.; Barnes, N. M.; Costall, B.; Deakin, J. F. W.; Ironside, J . W.; Kilpatrick, G. J.; Naylor, R. J.; Rudd, J. A.; acetone). MS: m l . 315 (M+). And-Calcd for C~jH23N302:C, 68.54; Simson, M. D. Neurosci. Lett. 1990, 111, 80. H, 7.99; N, 13.20. Found: C, 68.30; H, 8.30; N, 13.21.
+
Journal of Pharmaceutical Sciences / 105 Vol. 84, No. 1, January 1995
4. Kilpatrick, G. J.; Bunce, K. T.; Tyers, M. B. Med. Res. Rev. 1990, 10, 441. 5. Bunce, K.;Tyers, M.; Beraneck, P. Trends Pharmacol. Sci. 1991, 12, 46. 6. Hesketh, P. J.; Gandara, D. R. J . Natl. Cancer Znst. 1991,83, 613. 7. Costall, B.; Naylor, R. J.;Tyers, M. B. Pharmacol. Ther. 1990, 47,181. 8. Jones, B. Drugs News Prospects 1990,3, 106. 9. Lader, M. H.; Crook, T. H.; Meltzer, H. Y.; Sellers, E. M. 5th World Coneress o f Bioloeical Psvchiatn, Satellite Symposium, - Florence, Ztaly, 1991. 10. Fozard, J. R. Arch. Pharmacol. 1984,326,36. 11. Smith, W. W.; Sancilio, L. F.; Ower-Atepo, J. B.; Naylor, R. J.; Lambert, L. J . J . Pharm. Pharmacol. 1988,40,301. 12. Swain, C. J.; Baker, R.; Kneen, C.; Moseley, J.; Saunders, J.; Seward, E. M.; Stevenson, G.; Beer, M.; Stanton, J. J . Med. Chem. 1991,34,140-151. 13. Swain, C. J.; Baker, R.; Kneen, C.; Herbert, R.; Moseley, J.; Saunders, J.; Seward, E. M.; Stevenson, G.; Beer, M.; Stanton, J.; Watling, K.; Ball, R. G. J . Med. Chem. 1992,35,1019-1031. 14. Arias, M. S.;Galvez, E.; Izquierdo, M. L.; Burgos, C. J . Mol. Struct. 1986,147,381 and references therein. 15. Walker, N.; Stuart, D. Acta Crystallogr. 1983,A39, 158-166. 16. Main, P., Fiske, S. J., Hull, S. E., Lessinger, L., Germain, G., Declercq, J. P., Woolfson, M. M. MULTANSO. A System of Computer Programs for Automatic Solution of Crystal Structures from X-ray Diffraction Data. University of York, England, and University of Louvain, Belgium, 1980. 17. Beurslens, P. R.;Bosman, W. P.; Doesburg, H. M.; Gould, R. 0.;van der Hark, TH.E.M.; Haltiwanger, R. C. DIRDIF system of Computer Programs, Technical Report 19983/1 CrystallomaDhv Laboratorv: ” , Toernooiveld 6525 ED Niimeaen. The Nztherhnds, 1983. 18. Stewart, J. M., Kundell, F. A.; Baldwin, J. C. The X-RAY80 System of crystallographic Programs, Computer Saence, University of Maryland: College Park, MD, 1980. 19. Martinez-Riaoll. M.: Cano. F. H. PESOS. A Comnuter Program for the Autdmatic Treatment of Weighing Schemes. InsGtuto Rocasolano, CSIC, Madrid, Spain, 1975.
-
I
106 / Journal of Pharmaceutical Sciences Vol. 84, No. 1, January 1995
-
,
20. Nardelli, M. Comput. Chem. 1983,7,95-98. 21. International Tables for X-ray Crystallography 1974; Birmingham: Kynoch Press. (Present distributor, C. Reidel: Dordrecht), 1974,Vol. IV. 22. Motherwill, W. D. S.; Clegg, W. Pluto Program for Plotting Molecular and Crystal Structures. University of Cambridge, England, 1978. 23. Arias, M.S.;Galvez, E.; Izquierdo, M. L.; Burgos, C. J . MoZ. Struct. 1986,147,381 and references therein. 24. Izquierdo, M. L.; Galvez., E.; Burgos, C.; Arias, M. S.; SanzAparicio, J.; Fonseca, I.; Bellanato, J. J . Mol. Struct. 1993,291, 1. 25. Galvez, E.; Izquierdo, M. L.; Burgos, C.; Arias, M. S.; SanzAparicio, J.; Fonseca, I.; Gago, F.; Baldominos, G.; Lopez, P.; Prieto, J. C. J . Pharm. Sci. 1993,82,794. 26. Miller, K.; Weisberg, E.; Fletcher, P.; Teitler, M. Synapse 1992, 11,58. 27. Schmidt, A. W.; Peroutka, S. J. Mol. Pharmacol. 1989,36,505. 28. Hibert, M. F.; Hofmann, R.; Miller, R. C.; Carr, A. A. J . Med. Chem. 1990,33,1594. 29. Smidt, A. W.; Peroutka, S. J . Mol. Pharmacol. 1990,38,511. 30. Evans, S.M.; Galdes, A.; Gall, M. Pharmacol. Biochern. Behav. 1991,40,1033. 31. Hayashi, H.;Miwa, Y.; Miki, I.; Ichikawa, S.; Yoda, N.; Ishii, A.; Kono, M.; Suzuki, F. J . Med. Chem. 1992,35,4893. 32. Saxena, P. R.; Lawang, A. Arch. Znt. Pharmacodyn. 1985,277, 235. 33. Glowinsky, J.; Iversen, L. L. J. Neurochem. 1966,13,655. 34. Lowry, 0.H.; Rosenbrough, N. J.; Farr, A. L.; Randall, R. J. J . Biol. Chem. 1951,193,265.
Acknowledgments We thank the Commission of the EuroDean Communities for the funding of this research under the SCIENCE program. (Grant no. ERBSCl*CT915120). Supplementary Material Available-Data from the crystallographic study of compound 5(6)f are available from the authors.