Journal of Medicinal Chemistry, 1975, Vol. 18, No. 5
0-Arylamidoacrylic Acids (2) J. B. Stothers, “Carbon-13 NMR Spectroscopy”, Academic Press, New York, N.Y., 1972; G. C. Levy and G. L. Nelson,
“Carbon-13 Nuclear Magnetic Resonance for Organic Chemists”, Wiley-Interscience, New York, N.Y., 1972; L. F. Johnson and W. C. Jankowski, “Carbon-13 NMR Spectra”, WileyInterscience, New York, N.Y., 1972. (3) See, for example (a) H. G. Floss, Lloydia, 35, 399 (1972); (b) J. B. Grutzner, ibid., 35, 375 (1972); (c) L. L. Martin, C.-j. Chang, H. G. Floss, J. A. Mabe, E. W. Hagaman, and E. Wenkert, J . Am. Chem. SOC.,94,8942 (1972); (d) R. D. Johnson, A. Haber, and K. L. Rinehart, Jr., ibid., 96, 3316 (1974); (e) T. C. Feline, G. Mellows, R. B. Jones, and L. Phillips, J. Chem. SOC.,Chem. Commun., 63 (1974); (f) G. McInnes, D. G. Smith, J. A. Walter, L. C. Vining, and J. J. C. Wright, ibid., 282 (1974); (8) M. Tanabe and K. T. Suzuki, ibid., 445 (1974); (h) A. R. Battersby, M. Ihara, E. McDonald, and J. R. Stephenson, ibid., 458 (1974); (i) J. w. Westley, R. H. Evans, Jr., G. Harvey, R. G. Pitcher, D. L. Pruess, A. Stempel, and J. Berger, J . Antibiot., 27,288 (1974); and (K) J. R. Hanson, T. Marten, and M. Siverns, J. Chem. SOC.,Perkin Trans. 1, 1033 (1974). (4) See, for example (a) E. Wenkert, C.-j. Chang, D. W. Cochran, and R. Pellicciari, Experientia, 28, 377 (1972); (b) E. Wenkert, C.-j. Chang, A. 0. Clouse, and D. W. Cochran, J. Chem. SOC.,Chem. Commun., 961 (1972); (c) A. G. McInnes, D. G. Smith, C. K. Wat, L. C. Vining, and J. J. C. Wright, ibid., 281 (1974); (d) K. Nakanishi, V. P. Gullo, I. Miura, T. R. Govindachari, and N. Viswanathan, J. Am. Chem. Soc., 95,6473 (1973); (e) F. J. Schmitz and F. J. McDonald, Tetrahedron Lett., 2541 (1974); and (f) E. Wenkert, J. S. Bindra, C.-j. Chang, D. W. Cochran, and F. M. Schell, Acc. Chem. Res., 7, 46 (1974). (5) (a) N. J. Bach, H. E. Boaz, E. C. Kornfeld, C.-j. Chang, H. G. Floss, E. W. Hagaman, and E. Wenkert, J . Org. Chem., 39, 1272 (1974); (b) F. I. Carroll and C. G. Moreland, J . Chem.
509
SOC.,Perkin Trans. 1, 374 (1974); (c) T. E. Walker, H. P. C. Hogenkamp, T. E. Needham, and N. A. Matwiyoff, J . Chem. SOC.,Chem. Commun., 85 (1974); (d) A. A. Gallo and H. Z. Sable, J . Biol. Chem., 249, 1382 (1974); (e) L. Simeral and G. E. Maciel, Org. Magn. Reson., 6, 226 (1974); (f) N. S. Bhacca, D. D. Giannini, W. S. Jankowski, and M. E. Wolff, J . Am. Chem. SOC.,95, 842 (1973); and (g) J. Okada and T. Esaki, Yakugaku Zasshi, 93,1014 (1973). (6) (a) P. Lazzeretti and F. Taddei, Org. Magn. Reson., 3, 283 (1971); (b) G. Miyajima, Y. Sasaki, and M. Suzuki, Chem. Pharm. Bull., 19, 230 (1971); (c) G. Miyajima, Y. Sasaki, and M. Suzuki, ibid., 20, 429 (1972); and (d) Y. Sasaki, ibid., 21, 1901 (1973). (7) (a) R. R. Ernst, J . Chem. Phys., 45, 3845 (1966); (b) H. J. Reich, M. Jautelat, M. T. Messe, F. J. Weigert, and J. D. Roberts, J . Am. Chem. SOC.,91,7445 (1969). (8) (a) A. R. Tarpley, Jr., and J. H. Goldstein, J . Mol. Spectrosc., 39, 275 (1971); (b) A. R. Tarpley, Jr., and J. H. Goldstein, J. Am. Chem. SOC.,93, 3573 (1971); and (c) F. J. Weigert and J. D. Roberts, ibid., 89,2967 (1967). (9) (a) P. C. Lauterbur, J . Chem. Phys., 43, 360 (1965); (b) A. Mathias and V. M. S. Gil, Tetrahedron Lett., 3163 (1965); and (c) R. J. Pugmire and D. M. Grant, J . Am. Chem. SOC., 90,697 (1968). (10) H. L. Retcofsky and R. A. Friedel, J . Phys. Chem., 72, 2619 (1968). (11) C.-j. Chang, H. G. Floss, and G. E. Peck, unpublished data. (12) R. Hagen and J. D. Roberts, J . Am. Chem. SOC.,91, 4504 (1969). (13) W. J. Horsley and H. Sternlicht, J . Am. Chem. SOC.,90, 3738 (1968). (14) W. Adam, A. Grimison, and G. Rodriguez, J . Chem. Phys., 50, 645 (1969). (15) R. J. Pugmire and D. M. Grant, J . Am. Chem. SOC.,90, 697 (1968).
Antiinflammatory P-Arylamidoacrylic Acids Robert T. Buckler,* Harold E. Hartzler, Medicinal Chemistry Department
a n d Barrie M. Phillips Toxicology Department, The Research Division, Miles Laboratories, Inc., Elkhart, Indiana 46514. Received June 27, 1974
A series of 34 0-arylamidoacrylic acids was prepared and examined for antiinflammatory activity. These compounds are vinylogous carbamic acids, and several displayed activity equal to phenylbutazone in the rat pleural effusion model. Highest activity was associated with structures bearing halogen and cyano substituents. Amides were inactive.
We became interested in the title compounds for two reasons. They appear as partial structures in a number of antiinflammatory compounds such as the fenamates and the related benzamido benzoic acids.1,2 And they satisfy the rudimentary structural requirements suggested for certain nonsteroidal antiinflammatory acids: an aryl group attached to a nonbasic nitrogen atom that is separated from the carboxyl group by one or two a t o m s . 3 ~Because ~ of its simplicity, we felt that the structure of these compounds might be close to the minimum geometry and constitution necessary for such activity. Carbamic acids are unstable, they decompose spontaneously with loss of carbon dioxide. T h e few that are known are extensively stabilized by hydrogen b ~ n d i n g . ~ Since double bonds transmit the electrical effects causing this instability, there was some doubt t h a t vinylogous carbamic acids would be stable enough to be useful, even though one of them had been casually characterized earlier.6 Nevertheless, when prepared these acids turned out t o be stable to all but high temperatures and strong acids.
They all decarboxylate a t the melting point. Chemistry. The arylamidoacrylic acids I11 were prepared as outlined in Scheme I (see Table I for details). The starting @-aminoacrylate esters I were made from the appropriate @-ketoesters in the usual way.7 Ethyl esters were preferred but methyl and tert- butyl esters were used when necessary. The acylation of the @-aminoesters to give the @-amidoesters I1 (see Table I1 for details) was uneventful except when the starting @-amino ester possessed an ahydrogen atom (I, R1 = H). In these cases the reaction proceeded to give both an N-acylated product (11, RZ = H ) and a C-acylated product (IV). T h e isomer ratio varied according to the reaction conditions b u t it was roughly unity for reactions in ether a t room temperature. The isomer mixtures were easily separated by fractional crystallization or column chromatography on alumina or Florisil. T h e IH NMR spectra of the N-acylated isomers displayed the vinyl proton a t Rz as a singlet in the range 290310 Hz (CDCls). This peak was absent from the spectra of the C-acylated isomers. T h e infrared spectra of the C-acyl-
J i i u r n a l of Medicinal
510
Chemistry, 1975, Vol. 18, No. 5
Nui~kler,Hurtzler, Phillip,,
Table I. $-Arylamidoacrylic Acids
_____
.
No .
R
R,
Mp,"C"
7; yield
Formula
Recrystn solvent
Analyses
AF
H) N H. N H, N 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 C. H . N C. H, N C , H, N C , H. N H . N: C" C. H. S
3 .O 7.1 3.4 9.1
33 57
C,H,-ligroine MeOH-H?O C RH CIH C rH c ,H MeOH CIH CH Llgl o1ne Me ,CO-ligroine MeOH C,H MeOH FtOAc C H CH I - PrOH CH MeOH R.leOHf MeOHf MeOH EtOXc 2IeOH- H - 0 Ct04c %le CO- ligroine EtOAc C,H ligroine MeOH-H,O
50
EXOAc
45 50 45 2
EtOAc D M F H,O DI'vIF-H 0 D M F MeOH
~-
1 2 3
4 5
6 7 8 9 10
11 12 13
14 15 16 17 18 19 20 21 22 23
24 25
26 27 28 29
30 31 32 33 34
H 4- F 3-C1 4-C1 3,4-C1, 3-Br 4- Br 4- CH, 4-C:H4-I-Pr 4-f-BU 4- CN 4-CH30 4-NMe, H 4-C1 3.4-Cl9 3-Br 3,5-Br, 3-1 4-1 4-CH30 4- Pyridyl 4- / - P r 4-/-Bu H 4- Br 4-I-Pr
14OC 155 146 149 167 149 164 150 111 114 136 177 147 142 155 171 183 172 182
171 184 150 170 141 167 127 150 112
152 174 169 180 198 197
11 H
H 4- Br 4-CN 4-Pyridyl
__
48 35 23 5 1I d 3 5d 7 37 11 12 15 2 13 37 47 21 44 37 42 52 57 58 19
36 40 38 3
C, C, C, C.
c. H,N C. H . C, H . H , N: C, H, H , N: C. H. C , H, C, H , C , H.
N N
Ch N C'
N N
N N
e
1.9 7.1 9.1 6 .'i 9.2 5.2 9 .o 1.3 1.o
3.7
10 .o
4.8 5 .O 1.6 5.8 5.2 4.5 2.11 1.?
10.1 7.1 11.0 4.2 1.5 1.o
4.3 5.2 14.2 4.3
-~
flThesecompounds effervesce upon melting. *Activity index = ten times the ratio of the % reduction of the mean pleural exudate volumes of the test compound at 1 mmol/kg over 1 mmol/kg of aspirin. AI for phenylbutazone = 14.3. (Reference 6 reports m p 139" dFrom cleavage of t h e tert-butyl ester. eNot tested, insufficient compound. fWashed with methanol. S C : calcci. 59.99: found, 59.23. h C : calcd. 69.80; lound. 70.25. ~ C calcd, C : 57.83; found, 57.12.
Scheme I
.4rCOX"C=CCOOC,Hj
I
I
KOH
ArCONHC=CCOOH
I
Ri R, I1
R,
I
R?
III OR
rv
I
COAr V
ated isomers displayed the typical NH2 stretching absorptions at 3480 (s) and 3300 cm-l (m) while t h e N-acylated
compounds exhibited a moderately strong N H band at 3250 cm-1.8 Both E and Z isomers are possible in structures I1 and 111. However, only one geometrical isomer is observed, the strongly hydrogen bonded 2 isomer V. This assignment is supported by the 'H NMR spectra of the N-acylated 6amino esters. A very broad, one-proton multiplet is found centered a t about 723 Hz in t h e esters (CDCl3) and about 630 Hz in the acids (DMSO-de) indicative of bonded N-H. No peak for free Tu'-H is seen. Also, t h e parent ester of t h e series (11, Ar = phenyl; RI = CH3; R2 = H) had A,, 295 nm [ e 15,900 (CHC1:3!]. This is precisely as calculated for the Z isomer from prior studies on t h e uv spectra of vinylogous amide^.^ As a kind of structure proof, one of the C-acylated isomers was hydrolyzed, tert-butyl $-amino-n-(3-bromobenzoy1)crotonate (68). When treated with CH2C12-HC1, it gave ij-amino-3-bromocrotonophenone (69) via the spontaneous decarboxylation of the intermediate free acid. 'The saponification of the N-acylated a-aminoacrylates gave the expected $-arylamidoacrylic acids I11 in fair
Journal of Medicinal Chemistry, 1975, Vol. 18, No. 5
P-Arylamidoacrylic Acids
511
Table 11. B-Arvlamidoacrvlate Esters
No.
R
35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66
4-F 3-C1 4-C1 3,4- c12 3-Br 4- Br 4-CH3 4-CZHs 4-i-Pr 4-t-BU 4- CN 4-CH3O 4-NMe2 H 4-C1 3,4-c1, 3-Br 3 ,5-Br2 3-1 4-1 4-CH3O 4- Pyridyl 4-i- Pr 4-t-Bu H 4-Br 4-i-Pr H H H 4- Br 4- CN 4- Pyridyl
67
Rl
R,
E s t e r a Mp, "C %yield
A A A C C B B A A B A B A B B B B B B B B B B B B B B B B B B B B
102 87 129 102 103 104 67 63d 69 74 173 111 117 96 128 146 105 99 105 128 97 139 61 92 75 83 Oil i 79 104 119 81 118
67 42 63 7 29 37 42 57 58 42 47 26 27 59 80 65 60 68 65 72 51 73 59 66 70 89 44 46 25 70 35 48 50
Formula
Recrystn solvent
Analyses
MeOH Ligroine C,H6-ligroine MeOH-H,O MeOH MeOH Ligroine MeOH Ligroine MeOH-H,O MeOH C ,H,-ligroine MeOH Ligroine EtOH EtOH EtOH EtOH EtOH EtOH EtOH EtOH-H,O EtOH-H,O EtOH EtOH-H 2O EtOH h Ligroine Et OH-HZO EtOH EtOH EtOH
a A = methyl; B = ethyl; C = tert-butyl. bC: calcd, 56.80; found, 57.31. cC: calcd, 67.99; found, 67.50. dBp 153" (0.5 mm). "C: calcd, 67.99; found, 68.96. 'C: calcd, 51.55; found, 50.89. gC: calcd, 64.96; found, 65.57. hChromatographed on Florisil, eluted with benzene-ligroine, 'Bp 185-195" (0.7 mm).
Br
Sr 69
yields. A previous report had mentioned t h a t t h e saponification of ethyl P-benzamidocrotonate with KOH-ethanol gave P-benzamidocrotonic acid in only 7% yield. T h e major products were potassium benzoate a n d ethyl benzoate in 27 and 47% yield, respectively.6 Now under these conditions, ethyl benzoate can arise only by the attack of t h e lyate ion of t h e solvent on t h e amide carbonyl group of t h e enamido ester 11. We repeated this reaction using 2-propanol as the solvent since this alcohol has little tendency to form the lyate ion. With these conditions t h e yield of P-benzamidocrotonic acid 1 was increased t o 48%. This KOH-2-propa-
no1 method was then used to make most of t h e acids listed in Table I. However, esters with deactivating aromatic substitutents still gave very poor yields of acids when saponified; hydrolysis of t h e enamide group became t h e major reaction. We avoided this side reaction by preparing t h e analogous tert- butyl esters and cleaving them with CH2C12-HC1.10 I n one case we used phenyl isocyanate as the acylating agent. When reacted with ethyl p-aminocrotonate, it too gave both a C- and a N-acylated product, 70 and 71. The N isomer was saponified as usual, but when acidified, t h e @phenylureidocrotonic acid t h a t resulted cyclized spontaneously t o 6-methyl-3-phenyluracil (72)." Pharmacology. T h e antiinflammatory activity of these compounds was measured by their ability t o protect rats against pleurisy induced by the irritants Evans Blue and carrageenan. The introduction of a mixture of these irritants (0.05 ml of a 0.316% aqueous solution) into t h e r a t pleural cavity causes a n inflammation which results in a n increase in t h e volume of pleural fluid in t h e following 6 hr. This increase can be inhibited by a wide variety of antiinflammatory compounds including steroids and gold salts, and the inhibition is dose dependent.l*
512
.Journal of Medicinal Chemistry, 1975, Vol 18, No. 5
Experimental Section
II c H,N'
\CH,
70
--
C,,H,NHCONHC=CHCOOCzH,
I
CH 9
0
mu,
/ C6H5
HK
J-h 72
71
Each test compound was administered orally t o six rats 1 hr before t h e irritants a t a dose of 1 mmol/kg. Six hours after t h e irritants were given, the animals were killed and the volumes of their pleural fluid measured. These values were compared with those of control groups as well as those from animals treated with 1 mmol/kg doses of aspirin. T h e results are expressed as an activity index (AI) which is a value equal to ten times the ratio of the percent reduction of the mean pleural exudate volumes of t h e rats treated with t h e test compound a t 1 mmol/kg divided by t h e percent reduction of the volumes for the animals treated with aspirin
Discussion Geometry, size, electronic state, and lipophilicity are t h e four principle factors t h a t influence biological activity in a drug series where the basic functional elements are kept constant. This series was designed in such a way as to keep the overall geometry and chief functional elements unchanged while varying t h e aromatic substitutents and t h e substitutents about the double bond There is no obvious structure-activity pattern in this series, although some structural features d o seem t o be better than others. Halogens were the best aromatic substitutents and were more effective in the para than in the meta position T h e equally high activity of t h e 4-cyano analogs 12 and 33 suggests t h a t the biological response might be directlv proportional t o t h e degree of electron deficiency of the aromatic ring Even so, two 4-alkyl-substituted compounds, 10 and 25, were also somewhat active. T h e substitution of alkyl groups for hydrogen a t the Lk position (Kzin Table I) had a slight but positive effect on the biological activity. Large alkyl groups such as C ~ H and S (CH2)4 appear best. M o s t of t h e compounds studied have a methyl group at the $-carbon atom (R1 in Table I). Its replacement by phenyl in 29 gave an inactive compound. There is no clear cut dependence of antiinflammator) activity on lipophilicity in this series. Still, in spite of exceptions like 19, 21, 24, and 31, activity is greater for t h e more lipophilic members of t h e set. T h e most active compound. the 4cyanocyclohexyl analog 33, does combine high lipophilicity with deactivation of t h e aromatic ring. T h e activity of 33 seems t o indicate t h a t our original hypothesis may be correct: structures like I11 may well represent a kind of minimum structural requirement for the antiinflammatory activity associated with arylalkanoic acids. Three amides were tested. They are t h e 4-flUOrO 73, the 4-ethyl 71, and the 4-chloro analog 7.5 of N,N-diethyl-fibenzamidocrotonamide VI. A11 were inactive. I-\
CHJ
VI
All melting points are uncorrected and were determined with n Buchi capillary melting point apparatus. Gnalyses indicatrd !I,. symbols were within 10.4%of t h e calculated izaiues. 8-Aminoacrylate Esters, I. 'These were made by refluxing 1 he corresponding 3. keto esters and equal v o l u n i e ~of pentane Lvith ;O mg of' p-toiuenesulfonic acid while bubbling i n a s l o w stream of ammonia with rapid stirring. A water separator was attached. Higher boiling hydrocarhon solvents were less efficient.. prohahlv due t o lower soluhilii y ( i f ammonia. Thi- proc.ec!ure gave :In 80":3 yield on a ?-mol scale of tcrt-butyl d-aminocrotonate: m p :37@;i ) p 60' 10.4 m m ) (lit.I3 m p 3 L 3 9 ' i . Heating f i l r :?fi hr was required. Two of the amino ester.; made in this way were new compound+: ethyl 3-amino-2-ethylcroto~~ate [mp 57' recrystallized f r o m E t O H . Anal. Calcd for CsHl&JOs: N. 8.91. Fiiund: h'. 8.841 and diethyl :{ aminoglutaconate [bp 1 l o - 1 12" (0.:; trim) ..tna!. Calcd f ( i r CgH15NOs: neut equiv 217. E'liund: 2211. Acylation of the 8-Aminoacrylate Esters, I. T h e amino esters, and 1 equiv of triethylamine or pyridine were heated to reflux in ether or tetrahydrofuran. 2 l./mol of ester. T h e acid chlorides, 1.0,5 equiv, were added dropwise over 1 hr. Heating was continued for a total of 24 hr. When ciiol. the solutions Lvere filtered, washed with 1 S HCL and 5% NaHCO3 solution, dried over anhydrous MgSOi. a n d evaporated. When mixtures of acylat.ion isomers were p r o duced. the residues uere chromatographed o n nrntral alumina o r Florisil eluting with benzene-ethyl acetate mixtures. Other they were directly recrystallized from benzene or ethanol. Saponification of the Methyl or Ethyl @-Arylamidoacrylates, 11. T h e 3-amido e'tern I1 were finely ground and suspended in 2-propanol, 200 nil1'0.1 mol. A solutior: of 1.5 equiv 11f 15 :V KOH was added with stirrinp over a 90-min period. The temperature was maintained a t 20". After 16 hr, most of the 2-propanoI was rcmoved under reduced presiure and the iiincentrate diluted with -1 vol of' H2O. This: was wahhed twice with cther and then carelully acidified with dilute H r i . T h e resulting .!-orylainidoacryli(, acid.< were usually recrystallized from MeOH or benzene. b-Amino-3-bromoerotonophenone (69). A ioliition 0 : ' 60 g (0.38 mol) of t ~ r t - b u t y l,i-atninoirotonate in 500 nil of dry T H F was cooled t o 0' and conihined with .38t: (0.38 mol) of 2,G.iutidinc and 32 g (0.38 mol) ot :!-.i,ronlr)benzoyl rhloride. T h e temperatrirta was held a t 0' fi)r 5 h r and then allowed t o cclnie to room temperature overnight. lV;ork-up and chromatography as described above gave 26 g of t h r (' acy!ated ester, t o r t - h a t y l .I-arninci-o -(Xhroinc benzoy1)crotonate ( 6 8 ) . as white p l a t r l x - t ~from XIeOH: i n p i j Y O Anal. (C15Hl~RrNO:jiC. H. Y . A solution of ?6 g r0.056mol) of chi> ester in 5Ob nll III C H 2 ( ' 1 2 was cooled tc, -10" m d saturated wit,h HCl gas. After t hr tiit, reaction was allowed til warm to lO0 and stand overnight. Thcs lution was then washed with HaO and 6% NaHCO:i solution, d r i d over anhydrous 4IgSO4. and evaporated. Recrystallization from ether-pentane gave i.'i p I ) of the crotonophenone 69 as i'inr white plates: m p i19' .Sn;il loH1oBrNO) C . H. S . Ethyl p-Amino-a-( N-pheny1carboxamido)crotonate ( 7 0 ) and Ethyl 9-Phenylureidocrotonate (71). A (0.23 moll of ethyi ,i-arninocrotonate and 27 g IO. isocyanate in 200 nil of ether \vas refluxed for 1 hr. 1iIien C I W I the ether was evaporated alii! the solid residue digested with 4011 mi i ) l ' pentane. T h e pentane-ii:i~rlublematerial was twice frum benzene-pentane t i , give I 1 g of t h e C'-acyIated isomc~r;IS t i t i c , white needles: m p 12S5". Anal. ~ l . ' ~ : ~ H i & C. ~ OH. : ~N ~ 'Yhcs peir tane-soluble isomer was twice recrystallized from x n t a n e t,o gii. t> 9 g of t h e N-acylated isiinipr 71 as Fine \vt:itz nercll (Ci3HI6N2O5j(-, El. ~' availahle .l'._l.'-dieth\.l-~:-aniin(~cr~)t~~~ namide in t h e same. manner as that used f o r t h e aminri esters I.
References K . (.'ah, h n i i Rep. M c d . C h ~ m .226 , (1970). 11. .I. Drain. bl. .J. Daly, B. Davy. M. Horlington, J . G. H. Howes. .I. &I. Scrunton. and R. A. Selway. J . Pharm. P h a r m a C I J / . , 2 2 , 684 (1970). .J. G.1,ombardino and E. H. [Viseman, J . .Wed. Chern.. 15, 848 i 197.2 I . 'T.1..Shen. T'fip ,LIc,ct. i ' h ~ m . 1,. Si?(1967). ( i l l 1'. F. F'ascoe. .lirviu.\ Liebigs A n n . ('hem.. 705, 109 (19671, , .I. %f. Pryce. and P. J. Taylor, Chem. Com(*
Phenprocoumon a n d Monohydroxylated Derivatives (6) C. C. Barker, J . Chem. Soc., 317 (1954). (7) R. L. Augustine, R. F. Bellina, and A. J. Gustavsen, J . Org. Chem., 33,1287 (1968). (8) C. N. R. Rao, “Chemical Applications of Infrared Spectroscopy”, Academic Press, New York, N.Y., 1963, p 245. (9) D. L. Ostercamp, J . Org. Chem., 35,1632 (1970).
J o u r n a l of Medicinal Chemistry, 1975, Vol. 18, No. 5
513
(10) D. A. Cornforth, A. E. Opara, and G. Read, J . Chem. SOC., 2799 (1969). (11) R. N. Lacey, J. Chem. SOC.,839 (1954). Exp. Biol. Med., 127,597 (1968); (12) (a) L. F. Sancilio, Proc. SOC. (b) L. F. Sancilio, J . Pharmacol. Exp. Ther., 188,199 (1969). (13) R. Littell and G. R. Allen, Jr., US.Patent 3,449,363.
Synthesis and Thin-Layer Chromatographic, Ultraviolet, and Mass Spectral Properties of the Anticoagulant Phenprocoumon and Its Monohydroxylated Derivatives? Lance R. Pohl, Department of Pharmaceutical Chemistry, School of Pharmacy, University of California, S a n Francisco, California 94122
Rodney Haddock, William A. Garland, and William F. Trager* Department ofPharrnaceutica1 Chemistry, School of Pharmacy, University of Washington, Seattle, Washington 98195. Received April 8, 1974 Phenprocoumon and all of its aromatic monohydroxylated derivatives have been synthesized and analyzed by TLC, uv, and chemical ionization mass spectroscopy. By utilization of various combinations of these analytical techniques all of the titled compounds can be uniquely identified.
As a continuation of our studies on the biotransformation of warfarin (1)l and the relationship between druginduced interactions and its m e t a b o l i ~ m , l ~we * ~ have ~~ begun a study of the metabolism of the closely related oral anticoagulant phenprocoumon (2). Although the two drugs are structurally quite similar, significant differences in their pharmacologic properties exist. For example, in an extensive clinical study phenprocoumon has been reported t o elicit a more stable and reliable hypoprothrombinemic response compared to warfarin.4 In addition, it is significantly more active than warfarin and has a much .longer biologic half-life.5 The reasons for these differences are a t present obscure. Since warfarin is known to be monohydroxylated in the 6, 7, 8, and 4’ positions in the rat2+6and in the 6 and 7 positions in man,1g7,8 it seemed reasonable to anticipate that phenprocoumon would also be susceptible to aromatic hydroxylation. Indeed, preliminary work (mass spectrometry) in our laboratory had indicated the presence of such species in the urine of rats who had been injected with the drug. Hence, in order to facilitate the unambiguous identification of potential metabolites and to provide standards we embarked upon a synthetic program concurrent with our metabolic studies to characterize all the possible aromatic monohydroxylated derivatives of phenprocoumon. Synthesis. Phenprocoumon (2) can be synthesized by alkylating 4-hydroxycoumarin (3) with l-phenyl-l-propanol using HC19J0 or H2S0411 as catalysts or by using POC1312 as a condensing agent. Alternatively, it can be synthesized by direct alkylation with I-phenyl-l-br~mopropane.~~ As a consequence, these reaction schemes were initially explored for the synthesis of 6-hydroxyphenprocoumon from 4,6dihydroxycoumarin but were unsuccessful as only intractable tars were obtained. T o circumvent the problem of 0- vs. C-alkylation in the condensation reaction the 5-, 6-, 7 - , and 8-hydroxy positions were initially protected with benzyl
groups which were subsequently removed by hydrogenolysis.
A method similar to that of Hermodson, Barker, and Link14 was utilized for the synthesis of 5-, 6-, 7 - ,and 8-benzyloxy-4-hydroxycoumarin and involved the selective* monobenzylation of the isomeric dihydroxyacetophenones. T h e acetophenones were then allowed t o react with diethyl carbonate by the method of Dickenson16 t o yield the benzyloxy-4-hydroxycoumarins. Condensation of these materials with 1-phenylpropanol or 1-phenyl-1-bromopropane also led to intractable tars. Since coumarin 3 can exist in two other tautomeric forms,l7 the chromone structure 4 and the diketo structure 5, it seemed that alkylation of 4-hydroxycoumarin5 and its benzyloxy derivatives might be possible using conditions that are commonly employed for the alkylation of /3-keto esters.18 Sodium acetate was chosen as the base, since it was sufficiently strong to abstract the 4-hydroxy protonl9 but also weak enough to prevent significant dehydrobro-
tThis investigation was supported in part by NIH Training Grant No, 5-TO1-GM 00728, by a Washington State Heart Association Grant, and by an American Foundation for Pharmaceutical Education Fellowship Grant. Funds for the modification of the MS-9 to permit chemical ionization were provided by a grant from the Seattle Foundation.
$The ’H NMR absorption of the intramolecular hydrogen-bonded ortho groups15 at approximately 11 ppm clearly indicated the selectivity of benzylation. $For a review of the reactions of 4-hydroxycoumarin, see Zanten and Nauta.18
H 0 ’
s
1, R = CH,COCH, 2, R = CH,CH,
3
4
5
