J . Am. Chem. Soc 1983, 105, 135-137
in hexane and ethanol, show a rough linear correlation with In ( 1 / ~ )(Figure 2),'* consistent with our mechanism. When diazenyl radicals are formed from cis-azoalkanes, we believe that they serve as isomerization intermediates (Scheme I).19 However, not all cis-azoalkanes undergo deazatization (radical formation) competitively with is~merization.''~~In those cases, such as cis-azo-1-bicyclo[2.2. llheptane and cis-azo- l-bicyclo[2.1.l]hexane, we agree that isomerization occurs by inversion at nitrogen. We also agree with Engel and Timberlake that increasing steric bulk of the R group increases the inversion rate: However, we believe that isomerization via diazenyl radicals is a lower energy process for 1,2, and the [2.2.2] isomer, for example, than isomerization via inversion.20$21
I 0
A
c 'A
0
0
1 0
i
A
0
3
A
135
c 0
Acknowledgment. We are grateful for Paul Engel's initial inspiration and subsequent help and encouragement. This work was supported by a grant to R.C.N. from the National Science Foundation (CHE-7905954).
C
Registry No. 1, 59388-63-5; 2, 63561-19-3 0 0
(18) (a) Since solvent internal pressure and viscosity are related, this supports earlier proposals of internal pressure effects by Snyder.17b(b) Olsen, H.; Snyder, J. P. J . Am. Chem. SOC.1978,100,285. (19) However, if ( k +~ kd) >> k I t(e.g. azo-2-methyl-2-propane), isomerization could be ~ndetectable.~-' (20) See ref 4, Figure 3; the log krelvs. E, correlation could be fit with two lines, one through 1 and [2.2.2] and the other through [2.2.1] and [2.1.1]. (21) (a) Dannenberg21b calculates that cis-azoethane decomposes by onebond scission: (b) Dannenberg, J., private communication.
n
0
Figure 2. Viscosity dependence of In (kN2/ki);hexane (U), toluene (A), 2-propanol (a),acetone ( O ) , variable t e m p e r a t ~ r ehexane ;~ ( O ) , ethanol (0),variable pressure.
scission,I0 recombination to give the stable trans isomers (kit) should also be competitive." The smaller positive values of AK* (Table I) are also consistent with this mechanism. On the basis of Scheme I, AC* depends both on AV,* and the pressure dependences of the ratios k-lc/k-l, and kN2/ki(eq 1 ) . l 2 The latter ratio decreases with increasing AV,* = AVl* + R T (3 In (1 + k-Ic/k-lt + k N J k i ) / d P ) (1) pressure (Figure 1) while the ratio k-lc/k-lt is expected to remain constant or, perhaps, increase with p r e s s ~ r e , 'causing ~ the differential term in eq 1 to be small. Thus, AVi* should be comparable to AVl* (ca. +5 ~ m ~ / m o l 'and ~ ) ,this agrees with the data (Table I).I4 In contrast, AsanoI5 has found that nonradical cis trans isomerizations of azobenzenes give negative values of A Vi*. Decreases in kN2/kiwith increasing solvent polarity and decreasing temperature4 have been explained in part by polar eff e c t ~ . ~ ,On ' ~ the basis of Scheme I, k N 2 / kis, equal to ( k , k d ) / k l t ,which is expected to decrease with solvent viscosity due to its effect on kd. In fact, with the exception of the acetone data," the values of In (kN,/ki), whether derived from temperature variation4in hexane, toluene, and 2-propanol or pressure variation
-
+
(10) See: Neuman, R. C., Jr.; Amrich, M. J., Jr. J. Am. Chem. SOC.1972, 94,2730. (11) While recognized as a possible mechanism: lack of evidence for diazenyl radical intermediates made it unattractive. (1 2) Equation 1 is derived from Scheme I by recognizing that (a) A K * = -RT(B In ki/BP); (b) AV,* = -RT(a In k , / a P ) ; (c) ki = kl[k-l,/(k-le + kl, f kd + k ~ ) l ;(d) k~~ + k , [ ( k , + kd)/(k-ic f k-it f kd f k&l3and ( e ) kx2/ki = (kp + kd)/k-it. (13) Geminate recombination to give cis-1 or -2should be pressure accelerated. Formation of t r a w l or -2(kit) involves bond formation but also demands pressure-retarded rotational diffusion of the caged radicals. (14) A V ' N ~= AVl* + RT(B In (1 f k-ic/(k# + kd) 4- k i / k ~ , ) / a P ) ;both I~ rate constant ratios in this differential term increase with pressure causing AV'
>> AVII."
(l?) (a) Asano, T.; Okada, T.; Shinkai, S.; Shigematsu, K.; Kusano, Y.; Manabe, 0. J . Am. Chem. SOC.1981,103, 5161. (b) Asano, T.; Yano, T.;
Okada, T. Ibid. 1982,104,4900. (16) (a) Schulz, A.; Ruchardt, C. Tetrahedron Lett. 1976, 3883. (b) Duismann, W.; Riichardt, C. Chem. Ber. 1978,111, 596. (1 7) The acetone data4 are less accurate than those for the other solvents.
0002-7863/83/ 1505-0135$01.50/0
An Incremental Approach to Hosts That Mimic Serine Proteases' Donald J. Cram* and Howard Edan Katz Department of Chemistry and Biochemistry University of California, Los Angeles Los Angeles, California 90024 Received July 30, 1982 The design and synthesis of enzyme-mimicking host compounds remains one of the challenging and stimulating problem of organic chemistry. We chose to study serine protease mimics because the structure and mechanism of action of these enzymes have been so thoroughly studied. Their active sites contain a complexing cavity, an acyl-receiving and -releasing hydroxymethylene group, and a proton-transfer system that is organized to complement the structures of certain amide and ester substrates2 The naturally occurring cyclodextrins neatly combine a complexing cavity with primary hydroxyl groups (nucleophiles), and they have been successfully modified to provide systems that exhibit some of the features of the serine protease^.^ The structures of two totally unknown systems, 1 and 2 (Chart I), have been designed with CPK molecular models to combine in a cooperative arrangement similar to that of the proteases a binding site, a primary hydroxyl, an imidazole, and a carboxyl group. These two "ultimate target" hosts have in common with simpler host 3 the same organization of binding site and hydroxyl nucleophile. We report here the synthesis of 3, its binding ( I ) We thank the Public Health Service for Grant GM 12640, which supported this research. (2) (a) Blow, D. M.; Birktoft, J. J.; Hartley, B. S.Nature (London) 1969, 103,337-340. (b) Hamilton, S. E.; Zerner, B. J . Am. Chem. SOC.1981,103, 1827-1831 and references therein. (3) (a) Trainor, G. L.; Breslow, R. J . Am. Chem. SOC.1981,103,154-158 and references quoted therein. (b) Bender, M. L.; Komiyama, M . 'Cyclodextrin Chemistry"; Springer-Verlag: New York, 1977; pp 1-79, and references quoted therein.
0 1983 American Chemical Society
136 J . Am. Chem. SOC.,Vol. 105, No. 1. 1983
Communications to the Editor Table I. Rate Factors for Acyl Transfer from a alanyl p-Nitrophenyl Ester Perchlorate t o U(AU)BOH (3) vs. 6'
Chart I
run no. 1 2
3 4
5 6
U
7 8 gf
nucleophile kind
concn, M
6 6 3 3 3 3 3 3 3
0.016 0.032 0.0014 0.0014 0.0014 0.0014 0.0014
R3N concn, M
103k,
0.006 0.006 0,001 0.003d
kl, eq 5 reduces to eq 6. The rate factor due to all effects of complexation is k,/kb, whose values at constant [MI can be estimated through eq 7. We warmly thank Professors R. L. Schowen and F. A. L. Anet for very helpful suggestions regarding this kinetic treatment. (12) Jencks, W. P. 'Catalysis in Chemistry and Enzymology"; McGrawHill: New York, 1969; pp 1-242.
0002-7863/83/ 1505-0137$01.50/0
'John Simon Guggenheim Memorial Fellow, 1981-1982. ( I ) Padwa, A. Org. Photochem. 1979, 4, 261. (2) Weigert, F. J.; Barid, R. L.; Shapley, J. R. J . Am. Chem. SOC.1970, 92, 6630. (3) Padwa, A.; Reiker, W. J . Am. Chem. SOC.1981, 103, 1859. (4) Stechl, H . H., Chem. Ber. 1964, 97, 2681. (5) Padwa, A. Acc. Chem. Res. 1979, 12, 310. (6) Greenberg, A.; Liebman, J. F. "Strained Organic Molecules"; Academic Press; New York, 95, 861. (7) DeBoer, C. D.; Wadsworth, D. H.; Perkins, W. C. J . Am. Chem. SOC. 1973, 95, 861.
0 1983 American Chemical Society
