J. Med. Chem. 1993,36, 162-165
162
Synthesis and Structure-Activity Relationship of Some 6-[ [[(Dialkylamino)alkyl]-l-piperidinyl]acetyl]-10,1 l-dihydroSdl-benzo[ b,e][ 1,4]diazepin-ll-onesas M2-Selective Antimuscarinics Victor I. Cohen,' Jesse Baumgold, Biyun Jin, Rosanna De La Cruz, Waclaw J. Rzeezotarski, and Richard C. Reba Section of Radiophwmaceutical Chemistry, George Washington University Medical Center, Wauhington,D.C. 20097 Received July 20,1992
A series of 5- [[[(dialkylamino)alkyll-l-piperidinyllacetyll-10,1l-dihydro-SH-dibenzo[b,e] [1,4]diazepin-11-ones were prepared as potential Mpselective ligands. The compounds were evaluated for their affinity and selectivity for the muscarinic cholinergic receptor. The best M2-selective antimuscarinic agent studied is 5 4[4-[4-(diethylamino)butyll-l-piperidinyllacetyl] -10,ll-dihydro5H-dibenzo[b,el[1,4ldiazepin-ll-one, which is approximately 10times more potent at M2 receptom than previously known compounds such as 11-[[4-[4-(diethylamino)butyll-l-piperidinyllacetyl]5,1l-dihydr0-6H-pyrido[2,3-b] [1,4]benzodiazepin-6-one (AQ-RA 741). Our group has developed l-azabicyclo[2.2.21oct-3-y1 a-hydroxy-a-(4-iodophenyl)-a-phenylacetate (4IQNB),a potent m-AChR antagonist with high affinity but nonselectivity for In order to generate m-AChR radioligands with improved pharmacokinetics, we have I I c=o c=o developed a group of MI-selective AChR antagonists I I y 2 y 2 posseeeinglipophilicitylower thanthat of IQNB.S6 Since 4IQNB binds to each of receptor subtype with a similar affhity, and sinceonlythe M2 subtype is lost in Alzheimer's 2 E'2 disease, we have focused on developing an Mz-selective (CH2)4 radioligand. I The importance of developing subtype ligands is illusNEt2 1 trated by the following findings. When assayed with nonselective radioligands such as r3H1QNB, the density of m-AChR in brain from Alzheimer patients has been reported to be unchanged compared to age-matched contr~ls.~ However, when these studies were repeated wing radioligands which are selective for the Mz subtype such as 11-[[2-[(diethylamino>methyl-l-piperidiny 2 5,ll-dihydr~6H-pyrido[2,3-bl[1,4lbenzodiazepin-6-one([3HlAF-DX 1161, it became clear that M2 receptors are selectivelylost by as much as 61?6 .%loSincebrain contains .. a high density of MI receptors and a relatively low density I of Ma receptors, even a 61?6 lose of Mz receptors was not CH3 3 detected using nonselective ligands. Therefore, in the present work a significant effort was devoted to developing Mrselective ligands suitable for PET and SPECT studies benzodiazepin-6-one(AF-DX 116)(2). Wethen combined of the central nervous system (CNS). them with the lipophilic dibenzodiazepinone bulk group from the MI-selective 6-chloro-5-[(l-methyl-4-piperidiAlthough a large number of compounds have been ny1)acetyll-10,ll-dihydro-6H-dibenu[b,el[l,4ldiazepinreported to possess a high selectivity toward the MI 11-one (UH-AH 37) (3) to obtain the compounds with muscarinic receptor subtype," the structure-activity general formula 4. relationship (SARI of other subtype-selectivecompounds has barely begun to emerge, and so far has produced only Cbmirtry a few compounds that are 10-fold selective for either M2 or Ms muscarinic receptors.12 The required 11-oxo-10,ll-dihydro-SH-dibenzo[b,elAccording to Eberlein et al.>2the selectivityof tricyclic [1,4ldiazepin (9) was prepared from the reaction between antagonists for a particular muscarinic receptor subtype 2-chlorobenzoic acid and o-phenylenediamine. The reis vested primarily in the side chain carrying the amino action between 2- and 4-vinylpyridines(Sa,b)and dialltylgroup, with only slight contribution by the tricyclic ring amines6a-e provided2-(dialkylamino)ethylpyridine~ 70system. Therefore, in developing Mrselective ligands we (Table I). 4-Picolyllithium (Sc) reacted with 3-(dimethylused the cationic heads from the Mrselective compounds or 3-(diethylamino)propylchloride (61,g) to provide 444like 114[4-[4-(diethylamino)butyll-l-piperidinyllacetyll- dimethyl- or (diethylamino)butyllpyridines (71j). [(Di5,11-dihydro-6H-pyrido[2,3-bl[1,4lbenzodiazepin-6-one alkylamino)alkyllpyridines7a-j were converted into the (AQ-RA 741) (1) and 11412-[(diethylamino)methyll-lrespective piperidinea 8a-j by catalytic reduction (platpiperidinyll acetyl]-6,ll-dihydro-6H-pyrido 12,341[1,41inum oxide). Condensation of 9 with chloroacetylchloride
0
clr"
0
0022-2623/93/1836-0162$O4.00/0 (6 1993 American Chemical Society
Dibenzo[b,e][ll,4]diazepinones as Antimuscarinics Table I. Data on [(N~-Dialkylamino)alkyllpyridines7a-h and -piperidines Sa-h
I 7
product 70
7b 7c 7d 7e 7f 7g 7h 8a 8b 8c
8d 80 8f
se 8h
R
H
8
n position % yield bp,OC (preeeure,mm) 75-76 (0.2) 17 2 2 93 (0.1) 2 2 41 70-79 (0.1) 2 2 38 41 40 (0.1) 2 4 72 (0.3) 2 4 36 42 7 6 7 7 (0.08) 2 4 92 (0.1) 2 4 16 84-90 (0.1) 37 2 4 9698 (0.95) 2 2 86 91-94 (0.15) 90 2 2 75 (0.1) 70 2 2 60 (0.65) 91 2 4 70 (0.4) 2 4 89 66 (0.05) 2 4 92 104-110 (0.4) 89 2 4 81-82 (0.08) 2 4 85
Journal of Medicinal Chemistry, 1999, Vol. 36,No. 1 16s
imately 10 times more potent at Mz receptors than previously known compounds such as AQ-RA 741. Using Rekker's tables of fragmental partition coefficients we have calculated that 43 should be approximately 15 times more lipophilic than AQ-RA 741. Furthermore, we determined the CNS penetrability of this compound. For these studies, mice were injected (ip) with this compound, and after 3 h, their brains were homogenized and the number of unoccupied receptors was determined by I3H1NMS binding. We found that 3 h following injection of 0.05 mg/mouee of this compound, 0.042% of the dose penetrated the brain. In contrast, upon injection of the same doses, no detectable amount of AQ-RA 741 penetrated the brain. This result indicates that 4j crosses the blood-brain barrier better than AQ-RA 741 but not enough for evaluation as a potential muscarinic receptor radiotracer. Experimental Section
The melting points were obtained on a Fisher-Johnapparatus. The IR spectra of the compounds were obtained on a PerkinElmer 1710 infrared Fourier transform spectrometer. 1H NMR spectra were recorded on a Bruker AC-300 instrument and Scheme I expressed as parts per million (6). HPLC was performed on an Altex Model llOA. 2-Chlorobenzoic acid, o-phenylenediamine, 2- and 4-vinylpyridines,Cpicoline, dialkylamines, chloroacetyl chloride, and platinum oxide were obtained from Aldrich. 11-0xo-10,ll-dih~dro-5H-dibenzo[b,e][l,4]diazepine (9) was prepared by condensation of o-phenylenediamine with I 2-chlorobenzoic acid in chlorobenzene by method reported by H Giani et al.13 It was recrystallized from n-butanol. 9 N 5-(Cbloroacetyl)-lO,ll-dihydrodibenzo[b,e][ 1,4]diazepin5a,b 11-one (10). A solution of 9 (16.4 g, 0.078 mol), NJV-dimethylaniline (5.5 g), and chloroacetyl chloride (15.6 mL) in THF (350 mL) was refluxed with stirring for 5 h.14 The mixture was washed with 5% potassium bicarbonate and water. The solvent was removed under reduced pressure. The crude product was recrystallized from n-butanol (91%). Mp: 241-242 OC. TLC 5c [silica gel, EtOAc/hexane (1:l)l: Rf 0.50; IR (KBr): 1685,1661 cm-l. Anal. (ClsHlIN20~Cl):C, H, C1, N. Typical Procedure for the Preparation of 2- and 4-[2(Dialkylamino)ethyl]pyridine~(Table I, 7a-h). 2-[2-(Mpropylamino)ethyl]pyridine (la). The procedure described by Reich et al.lSwas followed. A suspension of 2-vinylpyridine(Sa, 21 g, 0.2 mol), dipropylamine (6c, 20.25 g, 0.2 mol), and acetic acid (10 mL) was refluxed with stirring overnight. The mixture was made strongly basic with 25% sodium hydroxide solution, saturated with anhydrous potaesium carbonate, and e x t r a d with ether. The combined extracts were dried over magnesium I O=C--CH*-N ~ ( C H ~ I ~ N R ~ sulfate, and the solvent was removed under reduced pressure. The residue distilled in vacuum to give 7.2 g of 7a (Bp: 75-76 4ci O U O . 2 mm). x = CI n=2or4 4-[4- (Dimethyl- and (Diethylamino)butyl] pyridine8 R = CH,, CzH5,nC3H7, nC4HQ,rC4H9 (SchemeI,7iJ)and-piperidines(SchemeI,8iJ). Toasolution of 4-picolyllithium (5c) in ether wae added 3-dimethyl- or 3-diethylpropylchloride (6f,g) at 0 OC. The resulting solution provided chloroacetylcompounds 10. Subsequentreaction was then stirred at 0 "C for 30 min. The mixture was poured into of [(dialkylamino)alkyllpiperidines8a-j with 10 yielded water and extracted with ether. The ethereal extract was washed compounds 4a-j (Scheme I and Table 11). with water, dried, and distilled to give 7i (Bp: 50 W0.03 mm) and 7j (Bp: 66-68 'C/0.03 mm). Results and Discussion General Procedure for the Preparation of [ ( D i a l k y W no)alkyl]piperidines (Table I, 8a-I). A mixture of 0.1 mol of Compounds 4a-g,i,j and two reference compounds [(dialkylamino)alkyl]pyridine (71,100 mL of methanol, 15 mL (Table 111)were tested for their apparent affinities at MI of concentrated hydrochloric acid, and 0.3 g of platinum dioxide and Mz subtypes. The Ki values (means f SD) are from was hydrogenated to obtain product 8. General Procedure for the Preparation of 6-[[[(DhlLyldisplacement binding studies of PHINMS binding to amin0)~1]-1-pipe~~1]ecety1]-10,11~~~wmembranes from brain (mostly MI receptors), from MI[b,e][ 1,4]diazepin-ll-oner (Table 11,4a-j). T h e chloroacetyl transfected qlls, or from heart tissue (for M2). These derivative 10 (2.87 g, 0.01 mol), [(dialkylamino)alkyllpiperidine data demonstrate that 4j is the most potent compound in (8; 0.01 mol), and potaseium carbonate (1 g) in 30 mL of the series. This compound is selective for M2 receptors acetonitrilewere refluxed for 5 h. The solventwas removed under over MIreceptors (the most abundant in brain) by almost reduced pressure and the residue WBB partitioned between ethyl %fold. These results also demonstrate that 4j is approxacetate and water. The organic layer was separated, waahed
tu"
tu-
164 Joumul of Medicinal Chemistry, 1993, Vol. 36,No.1
Cohen et 01.
O=d--CHz-N3(CHz),NR~ 4
relative lipophilicity,( position % yield mp,OC RP tR, min formula 34 71-72 4. n-CaH1 2 0.67 11.4 cdd40zd 28 67-68 0.75 16.4 cao&diOs' 4b n-CJ-Ig 2 31 82-83 4c i-CJ-Ig 2 0.3gb 13.6 cao&&od 27 101-106 M cH8 4 0.40 8.8 C&d%of 42 82-85 0.46 8.8 C8eHdJdh 40 cas 4 41 n-C& 4 21 73-76 0.73 9.8 cae&d%oo 41 58-61 4 0.76b 12.8 cao&diOlh 4e n-C& 45 86-89 4h i-CJ-Ie 4 0.69 11.5 C&N401' 4 59 157-158 ai CH3 0.35 8.5 clbHd4oz 71 70-72 4i CzHs 4 4 0.35 9.0 C d d & 0 M e O H / m O H (98:2). CHCWMeOH (101). Compounds were chromatographed on a Waters 8 M B CIS10 Radial Pak column eluted at 2 ml/min with methanol/water/acetonitrile(50:40:10) conthing 1 g/L of eodium l - o c t a n ~ ~ l f and ~ ~ t1e.2 mL/L of formic acid. The compounds were detected by absorption at 280 nm. C: calcd, 72.69; found, 72.16. * C: calcd, 73.43; found, 72.87. f C calcd, 73.43; found, 72.74. nC: calcd, 70.91; found, 70.48.h C: calcd, 73.43, found, 72.93. C: calcd, 73.43,found, 72.90. IC: calcd, 72.69; found, 71.97. compound
R
n 2 2 2 2 2 2 2 2
*
Table 111. Affiity Constaata of 6-[[[(Dialkylamino)alkyl]-1-piperidmyl] acetyl]-10,ll-dihydro6H-dibenzo[b,e][l,4]diazepin-ll-ones(a), AQ-RA 741 (I), and AF-DX 116 (2)' compound R n position brain heart 2 2 56.6 6.2 34.3 5.2 40 2 2 177.0 88.0 4b 74.0 2 2 99.0 4c 2 4 51.0 6 19.5 6.7 M 35.0 7 16.0 3 2 4 40 2 4 97.0 19 47.0 2 41 2 47.0 2 4 100.0 35 4e 4 4 11.0' 3.7 4i 4.0 1+ 0.3 0.1 4 4 4j 34.0 5+ 3.7 0.4 I (AQ-RA 741) 740.0 & 87' 73.0 2 2 (AF-DX 116) v&es for displacement of [SH] NMS binding to brain (mostly muecarinic receptors. Varying concentration MI) or to heart (MZ) of compounds were incubated with 0.6nM [SHINMSand membranes from either rat brain, rat heart, or (*) A9 L cella transfected with ml receptors for 60min at 37 OC. ICs0 values were converted to Ki using the Cheng-Prueoff equation. Kd valuea for [SHINMS binidng to brain, heart, and Mptransfected cell membranes were 360,420.and 235 pM, respectively. Unless otherwiee indicated, data are means SD of triplicate determinations.
*
** * *
** * *
with water,and dried over magnesium sulfate. The ethyl acetate was removed under reduced pressureand the residue purifiedby flaeh chromatography on silica gel usingmethanoVNEtOH (100: 2) as eluent. Biological Methods. Membranes from MI-transfected A9 L cells, from rat heart and from rat brain were obtained aspreviously described.16 Displacement studies were performed by incubating (37 OC for 60min) varying concentrations of compound, 0.5 n M [ ' H I M , and 100-300 g of membrane in a total volume of 1 mL of phoephate-buffered saline (pH 7.2). The reaction was terminatedby rapid filtrationover GF/B fiiters, and the fiiterswere counted in a scintillation counter following equilibration with scintillation fluid. ICs0 values were obtained by nonlinear regreasion analysis of the resulting data, as performed by the Graphpad computer program (San Diego, CAI, and were converted to Ki's using the Cheng-PNeoff equation.ll
Acknowledgment. This work was supported in part by Grant NS22215from the National Institutes of Health, DHEW, and by contract DE FG05 MER60649 from the
Department of Energy. We thank Dr. Mask Brann for providing us with the MI-transfected A9 L cella. Supplementary Material Available: Tables listing IH NMR data of compounds 4a-3, 7a-h, and 8a-h (2 pages). Ordering information is given on any current masthad pnge.
References (1) Rzeazotarnki,W.J.;Eckelman,W.C.;Francie,B.E.;Si,D.A.; Gibson, R.E.;Jagoda, E. M.; Griseom, M. P.; Eng, R R.; Conklin, J. J.; Reba, R. C. Synthesis and evaluation of Radioiodiuatd
Derivatives of 1-Azabicyclo[2.2.21oct3-y1 a-Hydroxy-a-(rl-iodophenyl)-a-phenylacetateas Potetial Radiopharmaceuticals. J. Med. Chem. 1984,27,156-160. (2) Eckelman, W.C.; Eng, R. R.;Rzeszotareki, W.J.; Gibson,R. E.; Francis, B. E.; Reba, R.C. Use of 3-Quinuclidinyl CIodobendate as a Receptor Binding Radiotracer. J. Nucl. Med. lSl,ZS,637642.
(3) Holman, B. L.;Gibson, R.E.;Hill, T. C.; Eckelman,W. C.; Albert, M.; Reba, R. C. In Vivo Imaging of M d c Acetylcholine Receptors with 1-1234IQNB and Emhion Tomography in Alzheimer's dieease. J. Am. Med. hsoc. 1986,264,3083-3086. (4) Cohen, V. I.; Rzeszotareki, W.J.; Gibson, R. E.;Fan, L. H.; Reba, R C. Preparation and Propertiee of (R)-l-Azabicycl0[2.2.21& 3-yl(R)-a-Hydro.y-a-(C['9liodophsnyl-a-phenylacetate and (R)1-Azabicyclo~2.2.2loct3-yl~S~-a-Hydro~a-(~[~~Iliodop~nyl)a-phenylacetateas potential radiopharmaceuticals.J.P h m . Sei. 1989,78,833-836. (5) Cohen, V. I.; Gibson, R. E.; Fan, L. H.; de la CNZ,R;Gitler, M. 5.;Hariman, E.;Reba, R. C. Synthesisand Muscarirric Cholinergic Receptor Affiiitiee of 3-Quinuclidinyl a-(Akoxyalkyl)-a-arylryl-aHydroxyacetates. J. Med. Chem. 1991,34,2988.2993. (6) Cohen, V. I.; Gibson, R. E.;Fan, L. H.; de la CNZ,R.; Gitler, M. 5.: Hariman. E.:Reba. R. C. Svnthesis and ReceDtor Affiitiea of 3guinuclidihyla-He&aryl-a&yl-a-hydroxy A&tatm. J.P b m . Sci. 1092,81,326-329. Whitahow, P. J.; Au, K.S.Cholinergic Receptom in Aging and Alzheimer's Di". Prog. Neuro-Psycholpharmacol. Biol. Psychiatry 1986,10,665-676. Mash, D. C.; Flynn, D. D.; Potter, L. T. Loer of Mn M d e Receptors in the Cerebral Cortex in Alzheimer's Dlwcw, and Experimental Cholinergic Denervation. Science 1986,2P8,lllb " 1114. (B) Quirion, R.;Aubert, I.; Lapchak, P. A.; Schaum, R.P.; Teolir, S.; Gauthier, S.; Araujo, D. M. M d c h p t o r Subtypes in Human Neurodegenerativedisorders: Focur on Alzheimer's Dine-. Trends Pharm. Sci. 1989,Suppl. IV, M. (10)Aubert, I.; Araujo, D. M.;Gauthier, 9.; Quiron, €2. M d Receptor AlterationinHumanCoenitivoD~rd~. !bendsPhOnn. Sci. 1989,Suppl. IV,B8. (11) Rzedz~tamki,W.J; McPhenon, D. W.; Ferkany, J. W.; Kinniw, W. J.; Noronha-Blob, L.; Kirkien-Reerzotsnki, A. Affinity and Selectivityof the O p t i d h m e r s of 3-QuinuclidinylBendata md Related M d c Antagoniota. J. Med. Chem. 1980,91,148s. 1466.
...
Dibenzo~,e][ll,4]diazepinonesas Antimuscarinics (12) Eberlein, W. G.;Engel, W.; M h ,G.; Rudolf, K.;Wetzel, B.;
Entzeroth, M.;Mayer, N.;Doode, H.N. Structure-Activity Relationship and Pharmalogical Profile of Selective Tricyclic Antimuncarinics. Trendd Pharm. Sci. 1989, Suppl. IV,50-54. (13) Giani, R. P.; Borsa, M.;Parini, E.; Tonon, G. C. A New Facil Syntheeia of 11-0.~10,1 l-dihydmbHdibendb,el[ 1,41diazephm. Synthesis 1985,550-552. (14) Moaro, A. M.;Quinton, R. M.;Wrigley, T. I. Some Analogs of Imipramine. J. Med. Chem. 1962,6,255-261.
Journal of Medicinal Chemistry, 1993, Vot. 36, No.1 161
(15) Reich, H. E.; Levine, R. T h e pyridylethylation of Active Hydrogen Compounds. 111. The Reaction of 2-Viylpyridiuewith Secondary Amines. J. Am. Chem. SOC.1965, 77,49134915. (16) Karton, Y.;Bradbury, B. J.; Baumgold, J.; Psek, R.; Jacokon, K. A. F u n c t i o d i Congener Approach to MuscarinicAntagonis& Analogues of Piremapine. J. Med. Chem. 1991,34, 2133-2146. (17) Cheng, Y.C.; Prueoff, W. H. Relatio~hipbetween the Inhibition Comtant (Ki)and the Concentration of Inhibitor which Caused W per cent Inhibition ( E m ) of an Enzymatic Reaction. Mol. Pharmacol. 1975,22,3099.
