GG 1
IHI~EVE~WBLE ENZYME INHIBITWS.CXXVI
July 196s
TABLE V PHYSICAL PROPERTIES OF
y&
t!$I!$b H
NO.
Xfetliod
1I
'?c yield
_-_M u , " C dec
l'ormula
.lnalyaes
Xlllns, inP----
PH 1
pH 13
C H ~ C ~ H I ( S H C O C ~ I I I S ~ ~ FFb - ~ ) - ?62' ~ 250-252 C1sITisFN603S. 1320 C, 11, N 270 272 237,d305 C6Ha(NHCOCsHaSO%F-p)-m Fb 58C >340 C~EH~~FK~O~S.O C,. H, ~ HN~ O 226,294 9 C6Ha(NHCOC6HdSO2F-m)-p Fb NG >340 CigH13FN6OaS.0.5H20 C, H, N 320 252,d 323 10 C ~ H ~ ( K H C O C ~ H ~ S O ~ F - P ) - ~Fb 63. >340 C18Hi3FrlToOYS.H20 C, H ; F' 322 231,d 325 2337,304 C, H, N 224,288 E 78C >35O CI~HSNGOP 40 CsHaNOa-m 247,265,326 259,369 CIIHSN~O~ C. H, N E 47c >350 41 CaHaNO2-p 266 272 C, H, K E 70' 307-309 C12H10K602 42 CH~C~H~KO~-~~C 43 CeHaSHz-m C 858 >340 CiiHiuN~.O.5I120 C, H, N 226,294 239,d 306 254,318 C 76" >340 C1iHioN6.0.25H20 C, H, S 226,205 44 CsH4NHy-p 45 CH~C~H~NH~WZ C 53Q 242-245 C~~HI~N~.H~O 'J 267 276 a I n IO%, EtOH. See method A, ref 2. c Recrystallized from bIeOEtOII-H20. d Inflection. e IlIF-H,O gave 1.73 g (32yc)of pure product, mp >3.i0°. See Table IF7 for additional data and other compounds prepared by this method. S-(ni-Nitrophenyl)adenine(40)(Method E).-To a mixture of 1.20 g of 34 and 12.5 g of P2Os cooled in an ice bath was added 9 ml of 85% H3P04. The mixture mas heated in a bath a t 165170" for 1.5 hr, then cooled and poured into 20 ml of iced HjO with stirring. The solution was adjusted to pH 8-0 with 4 S NaOH. The product n-as collected on a filter and mashed xith H20, then IIeOH. Recrystallization from I)lIF-H20 gave 0.60 g ( 5 . 5 % ) of pure product, mp >330°. See Table V for additional data and other compounds prepared by this method.
4-5-Diamino-6-(o-nitroanilino)pyrimidine(38).-A mixture of 500 mg (1.6.5 mmoles) of 32, and 15 ml of 4 NaOH was refluxed for 10 hr. The cooled suspension was filtered and the product was N-ashed with H1O. The solid was dissolved in 3 '1' H2S04,then spin evaporated to a syrup in vacuo. The sulfate salt was collected on a filter and washed with EtaO; yield 410 mg, mp 226-226", that moved as a single spot on tlc. The salt was dissolved in H20 and the free base precipitated by addition of 2 A' KaOH. The product was collected on a filter and thoroughly washed with H20, then LleOH, and finally EtzO; yield 260 mg (67%); mp 249-252" dec; Amax ( m r ) pH 1, 267, 401; p H 13, 280 (infl), 409. Anal. (CIOHIUN~OZ) C, €1, S: The para isomer (39) was prepared similarly except that the sulfate salt was insoluble in cold 3 LVH2SO4. The sulfate salt was collected by filtration and recrystallized from DMF-Et,O; yield 270 mg (937,), mp 303-305' dec. Anal. (C~UHION&.O.~H2S04)C, H. The free base was recrystallized from IZleOEtOH-H20; yield 166 mg (64%); mp 328-331" dec; Amax ( m l ) pH 1,261,368; p H 13, 236, 382. Anal. (CioHioN602) C, H ; N: calcd, 34.1; found, 33.6.
Irreversible Enzyme Inhibitors. CXXVI.1r2 Hydrocarbon Interaction with Xanthine Oxidase by Phenyl Substituents on Purines and Pyrazolo[3,4-d]pyrimidines
DepartwLent of Chemistry, l'niversily of Calzfomia at Santa Barbara, Sanfa Barbara, California 95106
Received February 26, 1968 A hydrophobic bonding region exists on xanthine oxidaie just adjacent to the active site that can complex aryl groups attached to purines and pyrazolo [3,4-d]pyrimidines. Inhibition by 57 purines and pyrazolo [3,4-d]pyrimidines bearing polar groups or both polar and hydrophobic bonding groups was measured; no unifying theory emerged on the mode of binding of these heterocycles to xanthine oxidase, although it was established by several parameters that the heterocycles could bind in one of a number of rotomeric configurations depending upon the positions of polar and phenyl groups on the heterocycle. The three best reversible inhibitors of xanthine oxidase found in this study were S-phenylhypoxanthine (S), 6-(mnitropheny1)adenine (15),and 6-(mnitropheny1)pyrazolo [3,4-d]pyrimidine (42), which were complexed 100-500-fold better than the substrate hypoxanthine (5)and 12-5Pfold better than 4-hydroxypyrazolo [3,Pd] pyrimidine.
The Bergmann school has made extensive studies on the influence of substituents on the xanthine oxidase catalyzed oxidation of purines in order to elucidate the mode of binding of purines and the mechanism of action of the e n ~ y m e . ~From their studies i t was apparent (1) This work was generously supported b y Grant CA-08695 from the National Cancer Institute, U. S. Public Health Service. (2) For the previous paper of this series see B. R. Baker and J. A. Koama, J. M e d . Chem., 11, 656 (1968).
that there were multiple modes of binding of purines to the enzyme depending upon the purine substituents. For example, hypoxanthine and 8-hydroxypurine were oxidized a t the 2 position but 2-hydroxypurine was oxidized a t the 5 position; in contrast, adenine was oxidized a t the 8 position, but 2-amino- and 8-amino(3) F.Bergmann, G . Levin, H. Kmietny-Gorvin, and H. Ungar, Biochim. Bzophys. Acta, 47, 1 (18611, and references therein.
SH.
0
4
R 1, I: = I I
2. I: 3, I:
= =
NIICOCII,BI SHCOC,T~,SO~I’-~
sit d i r e c t e d irreversible irihibitor~”’ of s:intliiiie oxid:iw. It \va* etivisioned that these three I):iranieter\ might be combined to try to rationalize the mode of purine binding to xanthine oxidase in the following manner. ( a ) If oire purine such as hypoxanthine is oxidized a t thc 2 Imsition, but another purine such as adenine i.: oxidized a t the S position, then the modes of binding of the two purines are different; such a statement requires the reasonable assumption” that there is onlj. one catalytic subsite per active site on xanthine oxidase. (b) If a phenyl group on one position of a purine such :IC guanine (1) give5 a strong hydrophobic interaction, but does not give hydrophobic interaction if the purine subqtituents other than phenyl are changed, such a‘ 9- (p-ch1orophenyl)xanthine (34), then the two purineq do not bind iii the same fashion. (c) If one active-sitedirected irreversible inhibitor such as 3 inactivates the eiizynie, but does not inactivate the enzyme when the other substituents o n the purine or the purine or both :ire chsnged, ~ c l :iis in the case of 4, then it, can b(, ccluivocally stated that the iulfonyl fluoride group doe. not reside ill an identical position within the respectiw ii 111i bi tor et izymc reversible complexes.l o Sunierou.: a t t r i q ) t s to use these three parameters to r:ttioii:ilize ilie modc of binding of purines to xanthine oxidaie still f:iilcd w e t i though more compounds were atldcd iii c w l i attempt to clarify the situation; part of tlw difficulty iri interpretation may be due to the ol-).erv:ition of Hofitcle12a that excess substrate citii irihihi t xnnthinc oxid:i-e, the 1;inetics arid interpreta-
IRREVERSIBLE ENZYME INHIBITOI~S. CXXVI
July 196s
TABLEI INHIBIPION' O F
x INTHIXE 0YIL)ASE BY PURINES A N D PYR IZOILJ [3,4-d]PYRI\IIDIKES
Cornpd Position no. 5 2 6 8 8 8 9 9 9 10 9 11 9 12 9 13
14 15 16 17 18 19 20 21 22 23 24 25 26 2i 28 29 30 31 32 33 34
8 8 8 9 9 'J
I&O,h Substituerit Kone CsHsCH?Y CsHsCH2S CsHa CHI CSH5 m-xO,CbEb p-XO?CsHa p-CHsOCsHa Sone m-NOzCsH4 p-NO?CsHa p-NH>CsHa CHI CsHn p-NO.CsHa
p.M
8 1c 0 75d 2 8d 0 062 1-10 13 31 63 6 3 5.6d 0 016 0 21
23 2800 Y 00 450
YourceP NBC
CCNSCe / Exptl Exptl Exptl h h /k
J I7
ESP11 Sigma
40d 7.4k
?iUC
9 9 9 9 9
None CsHn CsHs(CHzj3 CH3 CeHa CsHsCH. p-ClCsHI m-NHzCsHa p-XHzCsHa p-CHaOCsHa
9
Kone p-CICsHa
8 9 9
8.4k 5Sk 0.41k 23k
1. 8k 0.601 3.71 0.50'
ccxsci
Robins Robins Robins Robins
2 , 5"L
240
9
Xone C6H5
r.ll i.7d 1.1
SourceP
Sigma Robins
3i 38 39
NUC Robins Robins
Y
NO
NBC
102 69
8
35 36
I5",h
Substituent
0
Sone CsHj
9
WITH R Y D K O C I R B O h $UBSTITUENTS
Compd Position no.
Robins
10 41 42 43 44 45 4ti
6 ti 6
6 6 1
Xone CsHn ni-NOrCsHa p-NOnCsHa m-KfIKsHa p-Pi HKsHa CSH5
0.s i n 6.5
0,iO 0.83 18 4,O 260
h h h 11
h Robins
3.P 2.P 3Zk 1100 88
Rohins Taylor Taylor Taylor Taylor Robins Robins
1
?;one C H3 p-X01CsHa p-CHsOCsHa P-XHPCGHP CHI CsHn
55 56
1 1
None CHI p-CICsH4
2 3" 370 0.5;
Robins Robins Robins
57 58 59 60
1 1 6
6
70 330 12 0 58 14
Robins Robins Robins
61
None CHs CsHn m-NOKsHa m-NHzCsHa
47
48
6
49
6 6 6 1
50 51 52 53 .;I
0.20 13ok
0
0
The technical assistance of Maween Baker and Pepper Caseria with there assays is acknowledged. I;, = concentration of inhibitor necessary for 50% inhibition when assayed wit'h 8.1 p.lf hypoxanthine in p H 7.4 Tris buffer containing lo%, DMSO as previously described;" K , = 8.5 p i l l [J. B. Wyngaarden, J . Biol. Chenz., 224, 463 (1957)]. Substrate concentration. Data from ref 11. e SSC-88412. f Prepared according to R. K. Robins and H. H. Lin, J . Am. Chem. SOC.,79, 490 (1967). g Prepared according to S.11.Greenberg, L. 0. Ross, and R. K. Robins, J . Org. Chem., 24, 1314 (1959). See ref 2 for synthesis. Previously Calcd from K , = 20 p M ; reported to be an inhibitor with Ki = 18 p M . 1 5 f NSC-26293. k Data from ref 3. 1 Data from ref 8. H. Gutfriend and J. AI. Sturtevant, Biochem. J., 73, 1 (1959). nPreviously reported as an inhibitor.2%2' See ref 2. p NBC: Nutritional Biochemical Corp.; CCNSC: Cancer Chemot'herapy Natioiial Service Center; Sigma: Sigma Chemical Co.; Robins: Professor R. K. Robins; Taylor: Professor E. C. Taylor. a
hibitor in Table I. When the nitro group was moved to para position (16), the increment in binding was considerably less, 16 being complexed 27-fold better than aderiirie; reduction of the nitro group of 1 6 to amino gave a compound (17) which was complexed only twofold better than adenine (14). Since the S H ?group is both more polar and more electron donating than SOz, further studies would be worthwhile to see if these substituent effects are due to the relative hydrophobicity of the groups or to their electronegativity or both, l 4 as previously explored with the binding of 9-phenylguanine to xanthine oxidase.6,7 Little can be said about hydrophobic bonding to xanthine oxidase by 9-phenyl-6-mercaptopurine (22) other than a sevenfold loss in binding occurs compared to ti-mercaptopurine (21) ; the latter complexes to xanthine oxidase nearly as well as h y p ~ x a n t h i n e , ' ~ and is a slow substrate.'6 More compounds in this series would be needed to state whether or not some hydrophobic bonding occurs by the phenyl group of 22 (11) T. Fujita, ,J. IwaPa. and C. Hansch, .I. A m . Chem. Soc., 86, 51i5 (1964). (15) €I. R . SillJerman and .I. I3. \T'ynaaarden, Biochim. B i o p h u s . A c t a , 47, 178 (1961). ( 1 6 ) (a) G . 13. Elion, 8. Bieher, a n d G . H. Hitcliings, A n n . S. Y . Acad. Sci., 60,297 (1951); (b) T.L.Loo, M. E. Michael, A. J. Garceau, a n d J. C. Reid, J . A m . Chem. Yoc., 81, 3039 (1969).
as was done in the hypoxanthine series (5, 9, 10); such an undertalting is probably not worthwhile. From other studies in this laboratory on substituted 9-phenylguanines as inhibitors of guanine deamir1ase,6-~3~7 a large number of these compounds were available. Since 9-phenylguanine was also a good inhibitor of xanthine o x i d a ~ e these ,~ compounds were also investigated as inhibitors of this e n ~ y m e . 6 ~ ~ Selected examples are shown in Table I that clearly indicate a difference in binding of derivatives of guanine and hypoxanthine to xanthine oxidase. Guanine (23) binds fivefold less effectively than hypoxanthine ( 5 ) in our assa) method; guanine ha5 also been reported to be a slow bubstrate of xanthine oxidase.18 Introduction of the 9-methyl group (26) 011 guanine (23) gave little change in binding, in contrast to hypoxanthine (5 us. 9). However. introduction of the 9-phenyl group (27) gave a 140-fold increment in binding over 9-methyl (26), considerably larger than the increment in the hypoxanthine series (9 us. 10). Further studies showed the 9-phenyl group (27) on guanine was hydrophobically bonded and substituent effects were correlatablc with h y d r o ~ ) h o b i c i t y ;how~~~ (17) (a) B. R . Baker and D. V. Santi, J . .Wed. Chem., 10,62 (IQOi),paper LXXIV of this series: (bj ref 10, p p 101-109. (18) J. U.JVyngaarden, .I. B i d . Chem., 224, 453 ( l S 5 i ) .
ever, coIiij)ouiids correspotiditig to 29-32 crc tiot avail:tble iri the h) poxniithiiie series for conip:irisoii. Iiitroductioii of :in S-phenyl groul) (24) oii guiiiiiic (23) gave oiily :I bixfold iricreiiicrit iii 1)iiidiiig. iii c o l i trast t o the hyl)osaiithiiie series nIierc1 i t 1:iO-fold iiicrcniciit \vas ohscrved (5 P S . 8). T h t :t 9-:ir) 1x:inthiiic did iiot coinl)lcx tlic 5:~iiie:I' :I !~-:rrylliy~~osnrithiric or :L 9-arylguxninc was seeii in thv coiiil)ariwii of Ir-(p-chloroj)hCiiyl)saiitlli~ie(34): the 1:itter \\:is coinplesed 100-fold 1e5s cffectivcly than writhinc (33) in contrast to a 122-fold gain in biiidiiig i i i the guanine series (23 1's. 29) and little change in I)iiidiiig iii the hypoxanthine series (5 2's. 10). 0-l'hcnyl-~i-thioguaniiie (35) JVLL- about half :ts cff'c~rtivc:IS 9 - p h e n ~1gu:inirir (27) as :in inhibitor; hou cvcr, the iricremeiits in 1)iiiding crc~not strictlj coniI):irablc sincc t hioguaninr (35) u x s :L fivefold txtter irihihitor th:iii guaiiiiic (23), but the iiicrciiicwt i i i 1)licwJ I hindirig \ \ a s 14 tiiiics ngcr in thc guuiiiicb wrie5 (roiiipired 35 os. 36 IT ith r , 1hc 1:iht member of tho purine series iiivestigated way L',(i-diaiiiiiiol,urirlc (37) ; thii M :la hii-nilar to the aderiiiic seriez 11 h c w :L l o s ~in binding occurrcd when the 0~ ) l i e i i suhstituciit ~~I (38) IT a' iiitroduccd. ~)yr:izoIo[ 3 , l - d J p jiiniidiiic (40) (alloreported t o he :ti1 c~acc~lleritinhibitor of a:inthiric oxid:ihe-" t h a t is k i d i w iii ni:m for derrc:tsiiig uric, acid cxcrc6oii. 2 ' Therefore, :I micb of p j r:Lzolo[ : 2 , 1 - t / ] p ~ riinidiiie. I\ erc' iiivwtigiitcd for 11) drophobic lioiidiiig. In our test tcm, :~llopiiriiiol (40) 11a\ coniplcxed about tenfold hct ter thnii 11) i)oxantliiric~ ( 5 ) . However. in contraat to hypoxaiithiric ( 5 L S . lo), introduction of :L1-phtiiiyl group (46) on nllopurinol g ~ v ca 300-fold 105, iii binding : thus 9-phenylhy1)oxniithiiie (10) :tiid the corrc.spoiidirig 1)) rnzolo deriv:rtivc (46) cnriiiot bind to x:iiitliiric o d : w iii thc s:me fashion. Introduction of (j-phenyl group (41) o i i :illol)urinol (40) g : ~ w:I sevenfold loss in 1)iridiiig. IIovcver, it i. itill probable t h a t this 1)henyI groul) iiiter:irta hydro~11iobic:tllywith the enz) me 11 ith the heterocyclic ring t:tl.riiig :L different rotonieric biiidirig conforniatioii t h i i 40 for t\vo re:tsons. First, the ability of the (i~~1ietiyl:illopurillol (41) t o bind is influenced strong11 by +ubstitueiits; the ?n-nitrophciiyl derivative (42) i\ :t 12-foltl better inhibitor than :dlol)urinol (40) arid tliv , m i i t rol)henyl dcrivatiye (43) is the ~ n i :I\e :dlq)uriiioI. Ho\\ cver, reduction t o :Iniino 1c:tds to :t "GO-fold lo+ i i i hintling iii thc niefn serieb (42 z's. 44) : r i d a fiwfold 10'. i i i ljiiidirig in tho pura hwics (43 z's. 45). Thc s ~ c o ~ i t l line of evidence, is 11ast.d on tlit: liindiiig of tlerivalivcs of the thio analog (47) to dimissed later. Tllc nitro niid ntiiirio p t t c r i i of 42-45 is siniilar to that olwxvcd in thc 4-amino series (60, 61), the 4-thio series (49-51), :tnd the S-p2iciiyl:idenine scric+ (15-17), hut cluitc diffcreiit from t h e I)attcrii oliservcd in tlw 9 - ~ ) l i ( ~I1iypox:uitliinc n> (10-13) or !)-l)licii2'lguaiiiiii, wits (27. 29-32). 1-~Ierc:ii)tol)S.r:ieolo[3,4-d11)) riniidiiiv (47) \ \ a b :I fourfold hetter inhibitor t huii :Jlopuriiiol (40) a i d t lie substitution cfl'wts were simi1:u in the two systemz.
IRREVE~~SIBLE ENZYME ISmwroits. CXXVI
July 196s
TABLEI1 PHYSICAL PROPEKTIES OF 3,R = guanin-9-yl 4 , R = adenin-8-yl
62,R = guanin-%,yl 63,R = adenin-8-yl 64,R = 4-hydroxypyrazolo[3,4-d]pyrimidin- 6-yl
65,R = 4-aminopyrazolo[3,4-d> pyrimidin-6-yl
MIL
xCl. 11 12 13 20
73 74
-io
NHCO' 66,R = guanin-g-yl 67,R= adenin-8-yl 68,R = 4-hydroxypyrazolo[3,4.dl. pyrimidin-6-yl
69,R = 4-aminopyrazoIo[3,4dlpyrimidin6yl
bearing the identical hydrophobic and leaving groups such as 3 and 4 are complexed to an enzyme, but only one such as 3 can inactivate the enzyme by a neighboring-group reaction within the enzyme-inhibitor complex, it can be stated unequivocally that the leaving group on the two molecules is positioned differently v ithin the complex.?? Since the guanine derivative (3) can rapidly inactivate the e n ~ y m ebut , ~ the adenine derivative (4) does not,?it is clear that the phenyl groups of two molecules are not complexed to xanthine oxidase in the same way. Similarly, the 4-hydroxypyrazolo[3,+d]pyrimidine (64) can inactivate xanthine oxidnse,2 but the corresponding derivatives of guanine (62),9 adenine (63),? and 4-aminopyrazolo [3,4-d]1)yrimidine (65)* do not inactivate the enzyme; therefore 64 complexes to the enzyme in a different mode than 62, 63, and 65. The same conclusion can be reached by comparing the irreversible inhibitor 68 with 66, 67, and 69 which are not irreversible inhibit o r ~ . ~ ~ ~ S o unifying theory for the mode of binding to heterocycles to xanthine oxidase has emerged from this study, even though a number of practical objectives have been met. Although it would be aesthetically pleasing to have a unifying theory of binding such as that achieved n ith dihydrofolic reductase,22csuch a theory is helpful, but not essential to the design of active-site-directed irreversible enzyme inhibitors. lo From these studies on the mode of binding to xanthine oxidase have emerged a number of potent reversible inhibitors derived from purines arid pyrazolo [3,1-d]pyrimidines that bear an oversized group tolerated n ithin the enzyme-inhibitor. complex. By placement of a leaving group on the oversized side chain a number of potent irreversible inhibitors such as 3,9 64, and 6 V have emerged. Similarly, irreversible inhibitors have been constructed (22) (a) R. R . Baker a n d J. H. Jordaan, J . Heterocycl. Chem., 4 , 31 (1Y67), paper L S S X I I I of this series; (b) R. R. Raker and R. 13. hleyer, Jr., J . Pharm. Set.. 66, 570 (1967). paper LXXXIV of this series; (c) ref 10,
pp 231-248.
1 1 ~
R
Aletllod'l
OH OH 011 NH?
~n-h-02 p-NO1
B
CI
m-XOa
CI
p-NO1 p-OCHa
CI
p-OCHs p-NOa
B B
(&
oc
yield 73 64
dec
75
C 4 A
80 80
A
64
97
>800" >3OOc >300d
278-279 211-212 226-227 201-202
F o r . i i 1 11 l e
CIIH~NSO~ C11H1Ss0~0.2511rO CISHIOK~O? CIIHSNGO? CiiH6CINsOr
CIIH~CIN~O? CirKsNaO
All samples Fere crystallized from EIOII-1120. Amnx ( m p ) p H 1, 229, 250 (infl); pII 13, 233, 234 (infl). Amnx (mp) p,H 1, ( m p ) pfI 1, 238, 250 (infl); 237; p H 13, 260, 290 (infl). , , ,A p H 13, 235. e All compounds were analyzed for C, H, N. 11
from 6 and 7.23 Further studies on variation of the length and stereochemical nature of the bridge between the heterocycle and sulfonyl fluoride of 3 , 4 , and 62-69 are under investigation in order to find additional types of irreversible inhibitors and to apply the bridge principle of ~pecificity?~ for tissue-specific inhibitors of xanthine oxidase. S e w classes of candidate irreveraible inhibitors of this enzyme derived from 8, 10, 17, and 56 are worthy of investigation. Chemistry.-All of the compounds in Table I except 11-13 and 20 have been previously described in earlier papers of this series, were in the literature, or \\ere generously donated by Professor Roland I
