503
J.Med. Chem. 1984,27, 503-509 28717-34-2; 14, 7361-61-7; 15, 1082-57-1; 16, 526-36-3; 17, 38941-33-2; 18, 36067-72-8; 19, 84-22-0; 20, 76833-41-5; 21, 65955-46-6; 22,715-83-3; 23,59-42-7;24,57101-49-2;25,37571-84-9; 26,82013-55-6; 27, 68593-96-4; 28,74938-11-7;29,64309-39-3;30, 83964-56-1; 31, 59939-16-1; 32, 15327-38-5; 33, 39478-90-5; 34,
5051-62-7; 35, 1491-59-4; prazosin, 19216-56-9; phentolamine, 50-60-2; dihydroergotamine, 511-12-6; clozapine, 5786-21-0; cornynanthine, 483-10-3; azapetine, 146-36-1;yohimbine, 146-48-5; piperoxane, 59-39-2; tolazoline, 59-98-3; mianserin, 24219-97-4; rauwolscine, 131-03-3.
,&Selective Adrenoceptor Antagonists. 3. 4-Azolyl-Linked P henoxypropanolamines Peter J. Machin,*??David N. Hurst,? Rachel M. Bradshaw,? Leslie C. Blaber,t David T. Burden,t and Rosemary A. Melaranget Chemistry and Pharmacology Departments, Roche Products Limited Welwyn Garden City, Herts AL7 3AY, United Kingdom. Receioed July 1, 1983
A series of 4-substituted phenoxypropanolamines has been prepared and examined for P-adrenoceptor activity. The 4-substituenta, di- and triazole ring systems connected to the phenoxy ring by different length chains, were chosen as a means of introducing cardioselectivity. This has been achieved, especially in the l-[4-[(4-chloropyrazol-lyl)methoxy]phenoxy]-3-(isopropylamino)-2-propanol(l l), the 4-[(2H-1,2,3-triazol-2-yl)methoxy] analogue (21)) and analogue (22), which show potent &-blockade with selectivity ratios in excess the 4-[2-(2H-1,2,3-triazol-2-yl)ethoxy] of 1OO:l. Structureactivity relationships are discussed, and the optimum position of the heteroatom in the 4-substituent is defined.
The preceding papers1I2in this series describe the synthesis and @-blockingactivities of variously substituted (ary1oxy)propanolamines and support the c0ntention~9~ that a heteroatom suitably positioned in a 4-substituent is necessary to produce both potent and cardioselective fiblockade. Although, in general, the more potent agents are those with oxygen functions in the 4-substituent, we wished to find an alternative functionality that could achieve the same interaction with the &-receptor. Since this interaction is likely to involve the lone pairs of electrons on the oxygen atom, a situation was sought in which a nitrogen atom lone pair was available in similar fashion. Having excluded simple amines because of protonation at physiological pH and amides because of delocalization over the carbonyl system, we considered the possibility that the pyridinic nitrogen atoms in diazoles and triazoles closely fulfilled the requirement. Consequently, we have synthesized a series of phenoxypropanolamines (1-26)* sub-
PH
OCHzCHCHzNHR
Scheme I. Method A ?H
?H
CH20H
CH2-heterocycle
27 Scheme 11. Method B OCHZPh
I
OCH2Ph
I
H z , P d / C or HOAc-HBr
I
X(CHz),Y
28, Y = OMS; n = 2 29, X = 0 ; Y = Br; n=2 30, X = 0; Y = Br; n=3
I
X(CH2),-heterocycle
31, n = 2 32, X = O ; n = 2 33, X = 0; n = 3
I
OH
Q
A
X(CH2)n-Het
1-26, R = i-Pr, t-Bu; X = 0, S; n triazoles, benzoazoles
=
1-3; Het
=
stituted in the 4-position by groups incorporating di- and triazole ringsS5 The use of one of the ring nitrogen atoms as the point of attachment to the chain afforded easy synthetic access to molecules with variable heteroatom position and basicity. This paper describes their synthesis and evaluation as P-adrenoceptor antagonists. Chemistry. All but two of the oxypropanolamines listed in Table I were obtained by the classical phenol-epoxide-amino alcohol sequence using the conditions de+ Chemistry f
X(CH2),-heterocycle
pyrazoles,
Department. Pharmacology Department.
34,n = 2 35, X = 0; n = 2 36, X = 0; n = 3
scribed previously.2 The phenol starting materials were prepared by the following general methods (A-E). (1) Kierstead, R. W.; Faraone, A.; Mennona, F.; Mullin, J.; Guthrie, R. W.; Crowley, H.; Simko, B.; Blaber, L. C. J.Med. Chern. 1983, 26, 1561. (2) Machin, P.; Hurst, D. N.; Bradshaw, R. M.; Blaber, L. C.; Burden, D. T.; Fryer, A. D.; Melarange, R. A.; Shivdasani,C. J. Med. Chem. 1983, 26, 1570. (3) Smith, L. H. J. Appl. Chem. Biotechnol. 1978, 28, 201. (4)
I
,
.,
.._ _. . ~
Machin, P. British Patent Application 8035997, 1980.
0 1984 American Chemical Society
504 Journal of Medicinal Chemistry, 1984, Vol. 27, No. 4
Scheme 111. Method C
Scheme V. Method E
c
OCHZPh
-
n - B u O H , 100 "C pyro2a'e
I
NoBH4
"3 "-
HOCH2-N
H2°
-
C,C H 2
6
5
NOH, D V F
N
OH
I
-0
45
H
O
-e
O
-@
Hz,Pd/C
tH p h
O C &
N
46a, 1-isomer b, 2-isomer
Scheme IV. Method D OH
41, X = C1; n = 1 42, X = TsO; n = 2 43a, n = 1 b,n=2
Thermal coupling6 of heterocycles with 4-hydroxybenzyl alcohol gave the methylene-linked phenols (27) (method A, Scheme I). The ethylene (34), oxyethylene (35), and oxypropylene (36) linked phenols were prepared by alkylation of the requisite heterocycle with the appropriate mesylate or bromide, followed by debenzylation (method B, Scheme 11). Benzotriazole, indazole, and 1,2,3-triazole all gave both possible isomers, which were separated by conventional means; isomer assignment was made on the basis of spectral or physical characteristics (see Experimental Section). Interestingly, debenzylation of the benzotriazoles (32kand 321) by hydrogenation also gave some saturation of the benzotriazole moiety itself. Selective debenzylation was effected with HBr-acetic acid. The propylene-linked phenol (40) was prepared from the Mannich base (37)7 by displacement with pyrazole, reduction, and double hydrogenolysis (method C, Scheme 111). Reaction of 1-(chloromethy1)pyrazole (41) and 1[ 2-(tosyloxy)ethyl] pyrazole (42) with 4-merceptophenol afforded the sulfur-linked phenols (43) (method D, Scheme IV). Treatment of 1,2,3-triazole with formalin gave the 2-hydroxymethyl derivative (44) recently reporteds by Vereshchagin et al. However, conversion to the chloromethyl derivative (45) must have involved some rearrangement, because subsequent reaction with (benzyloxy)phenol gave both triazole isomers (46). Separation and hydrogenolysis afforded the desired (triazoly1methoxy)( 5 ) A different approach has led to a series of cardioselective
agents with somewhat related structural features. See: Baldwin, J. J.; Denny, G. H.; Hirschmann, R.; Freedman, M. B.; Ponticello, G. S.; Gross, D. M.; Sweet, C. S . J. Med. Chem. 1983,26, 950, and references therein. (6) Wakselman, M.; Robert, J.-C.; Decodts, G.; Vilkas, M. Bull. SOC.Chim. Fr. 1973, 3, 1179. (7) Palekar,A. D.; Desai, P. D.; Kulkami,R. A. Indian J. Phurm. 1973, 35, 135. (8) Vereshchagin, L. I.; Maksikova, A. V.; Tikhonova, L. G.; Buzilova, S. R.; Sakovich, G. V. Chem. Heterocycl. Compd. (Engl. Transl.) 1981, 5, 510.
$ OH
OCH2Ph - N a
HCI
1
N H
38
OCHrPh
SH
(CHzOL
44
37
OH
-
NN ;]
cI oc H2 c H 2
COCH~CHZN(CH,),
I
M u c h i n et al.
OCH2& N
47a, 1-isomer b, 2-isomer
phenols (47) (method E, Scheme V). The benzyl ether and phenol intermediates are listed in Tables I1 and 111, respectively. The two remaining analogues (10 and 11) were prepared directly from l-(isopropylamino)-3-(4-hydroxyphenoxy)2-propanol by alkylation with the appropriate (chloromethy1)pyrazole (41 and 49), respectively. Pharmacology. Compounds were tested for P-adrenoceptor blocking and partial agonist activities in anesthetized rats as described previously.2 The results, shown in Table I, are expressed as ED,, values (in micrograms per kilogram intravenously) for PI- and &-blockade and ED,, values (in micrograms per kilogram intravenously) for partial agonist activity. Discussion In common with the 4-substituted phenoxypropanolamines described in the earlier papers, most of the analogues in Table I showed little or no &-blockade at doses as high as 2 mg kg-l iv, clearly demonstrating the steric limitations of the &-receptor. Potency a t the &-receptor can be related primarily to the distance of the heteroatom from the phenoxy ring. Examination of the groups of compounds with different ring systems reveals that the most potent member in each group, Le., 2, 6, 11,21, incorporates a pyridinic nitrogen atom linked to the phenoxy ring by three or four atoms. This compares well with the three-atom link of the potent ether type of P1-antagonist.2 Further refinement of the position parameter is possible by comparison of the equipotent 2-substituted 1,2,3-triazoles (21 and 22). Computer modeling of these structures indicates that despite an extra methylene in the chain of 22, different conformations of the connecting chains could allow the triazole 1-nitrogen atom in both molecules to occupy virtually the same position in space, 6 A from the center of the phenoxy ring. Other structure-activity relationships can be discerned. For example, there is some correlation between activity and heterocycle basicity. Comparison of analogues in which the phenoxy ring-heteroatom separation is kept constant but in which the type of heterocycle is varied reveals, in general, increasing activity with decreasing basicity (Table IV). Thus, the very weakly basic chloropyrazole 11 and triazoles 21 and 22 show the highest potencies at the &-receptor. Another factor that is apparent from the biological activities is the positive effect of the oxygen link on potency. For example, the oxy-linked compounds 10 and 13 are considerably more potent than both carbon-linked compounds 9 and 19 and sulfur-linked analogues 12 and 18. Although possibly due to differences in lipophilicity and in vivo distribution, an advantageous binding of the oxygen atom to the &-receptor is an alternative explanation.
Journal of Medicinal Chemistry, 1984, Vol. 27, No. 4 505
4-Azolyl-Linked Phenoxypropanolamines
000000 000000 000000
0 0 0 0 0 0 0 0 0
0000t-00**00000a*00 0000 o o m m o o o o o r i r i o o 0 0 0 0 O a r l ~ O O O O O 00
rlrirlrlrirl
rid!+
A A A A A A
A A A
000000 000000 000000
rlrlmmmcv A A A A A A
o o o o o o o w m o o o o o o o m o o o o o m o o m m o o o o o o o m o 0 0 0 0 * 0 0 0 m 0 0 0 0 0 0 0 a 0 m m m m m m m r l mmmmC.1mC.I m
0 0 0
A A A A
A
A A
A A A A A A A
o m 0
orlo mmm
A
owoc-mo mmc-mmcwrlmwwd
rid0
t-am
m
l-l
m
2E-
6
z4: rJi
0"
*W 01 V h
1
2
z
N
1
"
a I 0 1
A
E &9Z
+
0,4t
01 S I 9
W
0--0
t-wmmmm
oo*t-c-m
e -
?? e
0
0
GOrnGOOOorn
00000
0
E
m
506 Journal of Medicinal Chemistry, 1984, Vol. 27, No. 4
Machin et al.
Table 11. Benzyl Ether Intermediates OCHZPh
I
Q I
X(CHz),,-heterocycle
no. X n heterocycle mp, "C 31a 2 1-imidazolyl 104-106 31b 2 lH-l,2,4-triazol-l-y1 103-104 31c 2 1-pyrazolyl 94-95 0 32a 2 1-imidazolyl 137-138 0 2 1-benzimidazolyl 115-117 32b 32c 0 2 lH-l,2,4-triazol-l-y1 104-105 32d 0 2 1-pyrazol 1 84-85 32e 0 2 4-chloropyrazol-1-yl 92-93 32f 0 2 4-phenylpyrazol-1-yl 111-112 32g 0 2 1H-indazol-1-yl 81-83 0 2 2H-indazol-2-yl 110 32h 32i 0 2 1H-1,2,3-triazol-l-y1 127-129 2H-1,2,3-triazol-2-yl 107-108 32j 0 2 32k 0 2 1H-benzotriazol-1-yl 108-110 321 0 2 2H-benzotriazol-2-yl 105-107 33a 0 3 lH-1,2,4-triazol-l-y1 83-84 33b 0 3 1-pyrazolyl 81-82 38 C=O 2 1-pyrazolyl 111-114 39 CHOH 2 1-pyrazolyl 84-86 46a 0 1 lH-l,2,3-triazol-l-y1 78-80 46b 0 1 2H-1,2,3-triazol-2-y1 a 'oil; purified by chromatography with chloroform-hexane.
crystn solvent
meth- yield, od %
MeCN i-PrOH i-PrOH EtOH i-PrOH i-PrOH EtOH EtOH EtOH hexane EtOH EtOH EtOH MeOH EtOH acetone-hexane acetone-hexane EtOAc EtOAc-hexane i-PrOH
B B B B B B B B B
B B C E E
44 65 72 71 78 62 70 72 77 52 21 32 45 53 36 51
35 70 98 35 25
emp formula
anal.
C,,H,,N,O C17H,,h,0 C,,H,,N,O Ci ,Hi BNzOz C,,H,,N,O, C17H17N302
Ci ,Hi aNzOz C,,H,,N,O,Cl C24H'22N202
C,,H,oN,O, C,,H,oN,O, C17H17N302 C17H17N302
C21H19N302 C21H19N302
C,,H,,N,O, C,,H,,N,O, C19H18N202
C,,H,,N20, Cl6Hl,N3O, C16H1SN302
Table 111. Phenol Intermediates OH
I
X (CH2 ).-heterocycle
no.
n
heterocycle
crystn solvent
mp, "C
206-208' 1-imidazolyl 113-115 1-pyrazolyl 158-161 1-imidazolyl 165-167 1H-l,2,4-triazol-l-yl 94-95 1-pyrazolyl 131-133 1-imidazolyl 0 1 9 1-1 9 2 1-benzimidazolyl 0 148-149 lH-l,2,4-triazol-l-y1 0 103-105 1-pyrazolyl 0 4-chloropyrazol-1-yl 105 0 155-157 0 4-phenylpyrazol-1-yl 124-125 1H-indazol-1-yl 0 63-66 2H-1,2,3-triazol-2-~1 35h 0 1H-benzotriazol-1-yl 110-111 35i 0 98 35j 0 2H-benzotriazol-2-yl 93 35k 0 4,5,6,7-tetrahydro-2Hbenzotriazol-2-yl 36a 0 3 lH-1,2,4-triazol-l-y1 145-146 36b 0 3 1-pyrazolyl 108-110 40 3 1-pyrazolyl 177-180 43a S 1 1-pyrazolyl 119-122 43b S 2 1-pyrazolyl 87-91 47a 0 1 lH-1,2,3-triazol-l-y1 168-171 47b 0 1 2H-1,2,3-triazol-2-yl 68-70 Characterized as the hydrochloride. Literatures mp 211 "C. 27a 21b 34a 34b 34c 35a 35b 35c 35d 35e 35f 35g
a
X
1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2
Increasing the size of the heterocycle either with benzo derivatives or bulky substituents, in general, reduces activity. With regard to partial agonist activity, no clear relationships are evident. An oxygen link is mandatory, as was found previously,2 but not all compounds with the
EtOH EtOAc EtOH i-PrOH toluene EtOH EtOH i-PrOH toluene
cc1,
toluene toluene Et,O MeCN i-PrOH toluene EtOAc EtOAc EtOH toluene toluene i-PrOH cc1,
meth- yield, od %
emp formula
anal.
B B B B
75 47 75 85 68 96 80 85 55 53 83 58 81 50 55 69
C,,H,,N,O C,,H,,N,O C,,H,,N,O C,,H,,N,O C,,H,,N,O, ClsH1,N,O, C,,H,,N,O, C,,H,,N,O, C,,H,,N,O,C1 C,,H,,N,O, C,,H,,N,O, C,,H,,N,O, C,,H,,N,O, C,,H,,N,O, C,,H17N,0,
C, H, N C, H, N C, H, N C, H, N C,H,N C, H , N C, H, N C,H, N C, H , N C,H,N C,H, N C, H , N C, H , N C, H, N C, H, N
B B C D D E E
61 60 70 81 47 91 60
C,,H,,N,O, C,,H,,N,O, C,,H,,N,OCI C,,H,,N,OS C,,H,,N,OS C,H,N,O, C,H,N,O,
C,H, N C,H,N C, H, N C, H, N C,H,N C, H, N C, H, N
A A B B B B B B B B B B
oxygen link show agonist activity, e.g., 10 and 23. In summary, a number of 4-substituted phenoxypropanolamines have been prepared in which the oxygen functionality of the 4-substituent, usually required to endow potent cardioselectivity on the @blockingactivity, has
Journal of Medicinal Chemistry, 1984, Vol. 27, No. 4 507
4-Azolyl-Linked Phenoxypropanolamines Table IV. Comparison of Biological Activity with pK, Values OH
I
Y-heterocycle
4,
EDSO & kg-' iv pKaa 9
no.
Y
heterocycle
620 7.2 1-imidazolyl 1 CH, 3.2 5 CH,CH, lH-1,2,4-triazol-l-y1 233 1-pyrazolyl 209 2.0 9 CH,CH, lH-l,2,4-triazol-l-y1 1 7 0 3.2 6 OCH,CH, 1-pyrazolyl 50 2.0 13 OCH,CH, 2H-1,2,3-triazol 2-yl 1 4
