Oxidation of NADH by Hydroperoxyi Radical
The Journal of Physical Chemlstry, Vol. 83, No, 15, 1979
M. S. K. Pabst, H. S. Tan, and J. L. Franklin, Int. J. Mass Spectrom. Ion Phys., 20, 191 (1976). A. G. Marshall and S. E. Butrili, Jr., J. Chem. Phys., 52, 2752 (1970). (a) J. L. Beauchamp, D. Holtz, S. D. Woodgate, and S. L. Patt, J . Am. Chem. Soc., 94, 2798 (1972); (b) R. J. Blint, T. B. MacMahon, and J. L. Beauchamp, ibid., 96, 1269 (1974). J. H. J. Dawson, W. G. Handerson, R. M. O’Malley, and R. Jennings, Int. J . Mass Spectrom. Ion Phys., 11, 61 (1973). J. M. S. Henis, M. D. Loberg, and M. J. Welch, J . Am. Cbem. Soc., 96, 1665 (1974). M. Colosimo and R. Bucci, J . Phys. Cbem., 82, 2061 (1976). (a) F. Cacace, A&. Phys. Org. Chem., 8, 79, (1970); (b) “fi-Decay of Tritlated Molecules as a Tool for Studylng Ion-Molecule Reactlons” in “Interactions between Ions and Molecules”, P. Ausloos Ed., Plenum Press, London, 1975, p 527. M. Coiosimo and R. Bucci, 5th International Symposium on Gas Kinetics, UMIST, Manchester, U.K., July 11-14, 1977. R. Yamadgni and P. Kebarle, J. Am. Chem. SOC.,96, 1320 (1976). D. Hok,J. L. Beauchamp, and S. D. Woodgate, J . Am. Chem. Soc., 92, 7464 (1970). F. W. McLafferty, Anal. Chem., 34, 2 (1962). For reviews of the radiolytic method to generate unsolvated species at high pressures see the following: (a) P. Ausloos, Prog. React. Klnet., 5, 113 (1959); (b) “Ion-Molecule Reactions”, J. L. Franklin, Ed., Plenum Press, New York, 1970; (c) P. Ausloos, S. G. Lias, and A. A. Scala, “Ion-Molecule Reactions in the Gas-Phase”, American Chemical Society, New York, N.Y., 1970, p 264; (d) S. G. Lias, “Ion-Molecule Reactlons in Radiation Chemistry” in “Interactions between Ions and Molecules”, P. Ausloos, Ed., Plenum Press, London, 1975, p 541. D. C. McGilvery and J. D. Morrison, J. Chem. Phys., 67, 368 (1977). R. Clpollini, N. Pepe, and M. Speranza, Gazz. Cbim. Ital., 108, 33 (1978). Z. Luczynski, W. Malicki, and H. Wincel, Int. J . Mass Spectrom. Ion Pbys., 15, 321 (1974). (a) R. Cipollini, M. Colosimo, N. Pepe, and M. Speranza, ORMDG (Organic Reactions Mechanism Discussion Group), Bangor, U.K., July 4-6, 1978; (b) M. Colosimo and R. Bucci, to be submitted for Dublication. W. T. Huntress, Jr., and D. D. Elleman, J . Am. Cbem. Soc., 92, 3565 (1960). N. V. Kir’yalkov, M. I. Markin, and V. L. Tal’roze, Kbim. Vys. Energ., 7, 189 (1973). G. P. K. Smith, M. Saunders, and R. J. Cross, Jr., J . Am. Chem. Soc., 98, 1324 (1976).
1957
(22) J. L. Beauchamp, ”Reaction Mechanlsm of Organic and Inorganic Ions in the Gas-Phase” in “Interactions between Ions and Molecules”, P. Ausloos, Ed., Plenum Press, London, 1975, p 413. (23) From MCA(CH,OH) = 66 kcal mol-’, calculated from PA(CH,OCH,) = 190 kcal mol-‘ (ref 11 and K. Hiraoka, E. P. Grismund, and P. Kebarle, J . Am. Chem. SOC.,96, 3359 (1974)), and MCA(CH,F) = 44 kcal mol-’ (ref 5a), and taking into account that methyl Ions, formed in reaction 2 and participating in reaction 7, might possess some vibratlonal energy (C.A. McDowell, Trans. Faraday Soc., 50, 423 (1970) and J. E. Williams, Jr., V. Buss, L. C. Allen, P. Y. R. Schleyer, W. A. Lothan, W. L. Hehre, and J. H. Pople, J. Am. Chem. Sac., 92, 2141 (1970)), AHlP N -44 kcal mol-’ Is calculated. (24) R. D. Smlth and J. H. Futrell, Chem. Phys. Lett., 41, 64 (1978). (25) To our knowledge, the formation of the condensed product [SCHR] , where S is a n-base, has been reported only by Beauchamp in ref 22 concerning the reaction
CH,lFt t ROH -+ CH,=O+-R
t HF
CH,FOH t R + We note that u s u a ~ Beauchamp ~~ produces CH,F+ ions in the presence of large quantities of [CH,FCH,]’ ions, and, remarkably, he has not observed the reaction
[CH,FCH,It
(26) (27) (28) (29) (30) (31)
+ ROH
-+
[
+
t CH,
that is a reaction between the dimethylfluoronium ion and a n-base, during which the CHzFe group is transferred to the n-base. V. D. Nefedov, E. N. Sinotova, G. P. Akulov, and V. P. Sass, Radiokimlya, 10, 761 (1968). R. Cipollini, N. Pepe, and M. Speranza, to be submitted for publication. From MCA(NH3) = 111 kcal mor’ (ref 12) and MCA(CH,F) = 44 kcal mol-‘ (ref 5a), we calculate AHi1 = MCA(CH3F) - MCA(NH,) = -67 kcal mol-’ when S = NH,. F. H. Field and J. L. Franklin, “Electron Impact phenomena”, Academic Press, London, 1970. G. A. Olah, D. J. Donovan, and H. C. Lin, J . Am. Chem. SOC.,96, 2661 (1978). (a) J.-Y. Calves and R. J. Gillesple, J . Chem. Soc., Chem. Commun., 506 (‘1976); (b) J. Am. Chem. Sac., 99, 1768 (1977).
Oxidation of Nicotinamide Adenine Dinucleotide by Hydroperoxyl Radical. A Flash Photolysis Study A. Naderhdin and
H. B. Dunford*
Department of Chemlstry, University of Alberta, Edmonton, Alberta, Canada TBG 2G2 (Received February 5, 1979) Publication costs assisted by the Department of Chemistry, University of Alberta
T h e one-electron o x i d a t i o n o f nicotinamide adenine dinucleotide (NADH) by hydroperoxyl radical was studied over a b r o a d range o f reagent concentrations and pH values by t h e m e t h o d o f flash photolysis. Of t h e acid-base f o r m s o f t h e radical, H02.a n d 02-., only t h e f i r s t one i s an active oxidant. T h e r a t e constant o f t h e r e a c t i o n H02. NADH H20z NAD. i s equal t o (1.8 f 0.2) X M-’ s-’ at 20 “C. T h e temperature dependencies o f t h e reaction r a t e a t d i f f e r e n t pH values gave the a c t i v a t i o n energy o f t h i s reaction, E , = 5.5 f 0.5 kcal, as ==02-.He, A”’ = 2.7 f 0.7 kcal. w e l l as t h e s t a n d a r d e n t h a l p y o f t h e e q u i l i b r i u m H02-
+
-
+
lo5
+
Introduction The m e c h a n i s m of r e a c t i o n of t h e coenzyme or substrate molecule, n i c o t i n a m i d e adenine d i n u c l e o t i d e (NADH), and i t s o x i d i z e d form NAD+ have a t t r a c t e d w i d e s p r e a d att e n t i o n . T h u s , in e l e g a n t s t e r e o c h e m i c a l e x p e r i m e n t s , it was demonstrated that w h e n NAD+ acts as a coenzyme for dehydrogenases’ a hydride i o n i s t r a n s f e r r e d stereospec i f i c a l l y f r o m s u b s t r a t e t o NAD+. H o w e v e r , as a substrate in p e r o x i d a s e reaction^,^!^ NADH i s o x i d i z e d in a one-
electron s t e p t o form the free r a d i c a l NAD. w h i c h h a s t h e ability t o undergo t h e r e a c t i o n
NAD.
+0 2
-
NAD+ + 02-m
The d e s t r u c t i o n of 02-.,t h e s u p e r o x i d e anion, by superoxide dismutase was d e m o n s t r a t e d in 1969.4 Since that time c o n s i d e r a b l e controversy has arisen over the imp o r t a n c e of 02-.as an oxidizing agent.5p6 This paper sheds further light on the relative importance of reactions of
0022-365417912083-1957$01.00/0 0 1979 American Chemical Society
1958
The Journal of Physical Chemistry, Vol. 83,
No. 15, 1979
superoxide anion and its protonated form, the hydroperoxyl radical HOz-,using NADH as the primary reducing agent. The flash photolysis of HzOzwas used as a source of O p or HOz-. Irradiation of H z 0 2with ultraviolet light leads to 0-0 bond rupture producing .OH radicals with a quantum yield of 0.82.7 Hydroxyl radicals react with H20z with a rate constant of 1.7 X lo7 M-l s-l to generate hydroperoxyl radicals? .OH + HzOz HOz. + HzO
A. Nadezhdin and H. B. Dunford
7
1
x
3
[ti20210= 0.96M
/
+
In our experiments [H2O2lO ranged from 0.15 to 0.9 M and [NADHIofrom 1 X to 6 X 10" M. Therefore, virtually every -OH was converted into HOz.. Even if the reaction of .OH with NADH is diffusion controlled, its importance is nullified by keeping [Hz02]/[NADH] > lo3. Experimental Section Materials and Methods. All the chemicals used were of reagent grade. Hydrogen peroxide, 30%, was from Fisher Scientific Go. and NADH, grade 111,was obtained from Sigma. Buffer solutions for pH regions 4.2-5.3 and 6.2-7.5, respectively, were acetic acid-sodium acetate and potassium dihydrogen phosphate-sodium hydroxide. The ionic strength of the reaction solutions was 0.03 M. HzOZ and NADH concentrations were determined spectrophotometrically by using known molar absorptivities = 1M-' cm-l and 6253.7 = 19.6 M-l cm-' for Hz029and 6340 = 6.2 X lo3M-l cm-', 6% = 3.3 X lo3 M-l cm-l for NADH.1° Experiments were performed on rapid reaction analyzer RA-601, obtained from Union Giken Co., 4018-1, Tsuda, Hirakata, Osaka 573-01, Japan, equipped with the flash photolysis mode and by using monochromatic flash light with X = 330 nm. Path lengths of the analyzing and photolyzing light were 1 cm each. Changes of absorbance at X = 340 and 366 nm vs. time after a flash were recorded. The ratio of absorbance changes at both wavelengths was equal to the ratio of absorbances of NADH itself, showing that no product with absorption maxima at 340 nm was formed and, in agreement with Land and Swallow,ll NAD. radical recombination reactions did not play any noticeable role under our experimental conditions. The concentration of peroxide radicals formed per flash was measured by using the Oz-. scavenger C(NOz)4.12The product of the one-electron reduction of tetranitromethane by Oz-. is the nitroform anion C(NOJ3- with a molar absorptivity at 350 nm13 of 1.54 X lo4M-l cm-l. A solution containing H202and C(N0J4, at a pH low enough (pH =3) for suppressing their direct reaction in the dark, was exposed to flash light and the absorbance change at X = 350 nm was measured. Since the stoichiometry of the reaction 02-. C(N0J4 is 1:114the HOz. (or Of.)initial concentration after the flash would be measured. This value was compared with that obtained by using the exwhere €330 is the molar pression [HO,.], = 2.31,,~~~~1[H~O~]$, absorptivity of H 2 0 za t the wavelength of the flash light (330 nm), 1 the optical path length in the direction of flash and lights beam, 4 the quantum yield of H202phot~lysis,~ Io the number of quanta per unit of volume per flash, measured by the standard Hatchard and Parker's procedure.15 Both methods gave consistent results, the first one being more accurate and convenient. Calibration plots, i.e., dependencies of [HO,.], on [H20210and voltage applied to the flash lamp are shown in Figure 1. The molar absorptivity of NADH at 330 nm is 5.8 X lo3 M-' cm-l at 20 "C and that of H20zis 0.11 M-' cm-l. Therefore under some of our experimental conditions more light is being absorbed by NADH than by Hz02. The NADH is known to fluoresce strongly16 with a lifetime of 4.5 ns.17 To the
+
0'
I
6
I
a
I
I
I
10
Voltage,
I
12
I
kV
Figure 1. Calibration plot. The initial concentrations of superoxide radicals formed at different [H202] vs. voltages applied to flash lamp.
E
sm
'1
I
0.201
c
1 O
I 1
I
1
I
1 i 2
time, s Figure 2. Absorbance change at h = 340 nm vs. time after flash in the system NADH H202. The procedure for obtaining the value of T, is schematically shown. The broken line (a) corresponds to the initial slope of the kinetic curve, line (b) to the slope onshalf that of (a). The value of T~ corresponds to the point of maximum distance of the kinetic curve from line (b) inside the angle between (a) and (b). At this point (d[NADH]ldt),o = 'I, (d[NADH]/dt)o. [NADH], = 4.1 X M, [H,O*]o = 0.32 M, 10.5 kV, pH 5.6, t = 20 O C .
+
best of our knowledge fluorescence is the sole result of the light absorption process by NADH. Flash experiments in the presence of NADH but absence of H 2 0 z led to recordings of absorbance vs. time at 340 nm which were constant within experimental error. Under the worst experimental conditions the signal-to-noise ratio was estimated to be 3, under the best 10. Results a n d Discussion A set of flash photolysis experiments was performed at different concentrations of H202, NADH, and H'. A typical run is illustrated by Figure 2. A table of results is available as supplementary material (see paragraph at end of text regarding supplementary material). The decrease of absorption at 340 nm corresponded to consumption of NADH in solution due to the photoinduced reaction. It could not be due to the action of primary hydroxyl radicals as was explained above. The dark interaction of NADH with H202at pH >4 was sufficiently slow to be neglected compared to the light induced reaction. From the following consideration one can see that
The Journal of Physical Chemistry, Vol. 83, No. 15, 1979
Oxidation of NADH by Hydroperoxyl Radical
the interpretation of the entire trace as illustrated in Figure 2 is not simple. However, there is a simple interpretation of the initial slope and a quantity r0, defined below. The NAD. radical, product of the one-electron oxidation of NADH, is known as an extremely strong reducing agent, and in the presence of molecular oxygen dissolved in solution, it is oxidized by the reaction NAD. + Oz NAD' 02-.with the rate constant equal to 2 X lo9 M-l s-l.ll In addition H 2 0 zmust also be a good oxidant for NAD. as was suggested el~ewhere.~ Most of the hydroxyl radicals, produced by the reaction NAD. HzOZ NAD+ -OH + OH-, have to react with H 2 0 2under our experimental conditions (see above) giving H02. (or Oz-.). Taking into consideration the recombination of HOz. radicals, we obtain the following set of reactions (H02.here is used to designate whichever form of the superoxide radical which is reactive):
+
1959
+
+
-
+
22.OH
Hz02 *OH HOy
+ HzOz
+ NADH
NAD. NAD.
+ 02
-
"
2
(i)
+ HzO HzOz + NAD. NAD+ + HOy HOz.
(1)
4
6
[NADHIo x lo5 ,M Figure 3. Initial rate of NADH oxidation plotted against [NADH], at two different [HO,.],; pH 4.9, t = 20 O C .
(2)
+ -+
I
I
I
1
2
1
I
(3)
+ HzOz *OH NAD' + OHHOz. + HOz* HzOz + 02 +
(4)
(5)
The rate constants characterizing NAD. + Ozand .OH + H 2 0 2reactions are several orders of magnitude larger than that expected for HOz. + NADH3)" (k 5 lo5 M-l s-l a t pH 15.6). Therefore [.OH] and [NAD.] are much smaller than [HO,.], so reaction 5 is the only termination step of importance. Using the quasi-steady-state approximation,18 one obtains d[NAD.]/dt = 0 = kz[HOy][NADH] k3[021 [NAD.] - k4~HzOzl [NAD.]
0
[HO;lox
d[.OH]/dt = 0 = h4[H202][NAD.] - kl[.OH] [HZOZ] d[HOz.]/dt = k3[02][NAD.]
3
10,'
M
+ kl[.OH][HZO,] 2k5 [HOy]
Adding these equations, we have d[HOy]/dt = -2h5[H0y]2 2
which upon integration yields
4
6
[NADHIo x
where [HOz-]ois the initial concentration of HOz. radicals, which can be obtained from the data of Figure 1. Consumption of NADH corresponds to reaction 2 of the above scheme and is described by -
d [NADH] = hz[NADH][HOz.] = dt
8
lo5,M
Flgure 4. (A) The dependence of T ~ - 'on [HO,.],; [NADH], = 3.8 X M, pH 4.9, 20 OC. (8)7, plotted vs. [NADH] at constant [HO,.],; [HO,.], = 1.5 X IO-' M, pH 4.9, 20 O C .
so that k 2 can be calculated from Figure 3. The time, 70, in which the rate of disappearance of NADH is reduced by a factor of 2 from its initial value is given by 1
The overall change of [NADH] during a single experiment corresponded to only a few percent of the total [NADH] so the latter can be considered as a constant in the interpretation of experimental data (Figure 2). The initial rate of NADH disappearance is given by a reduced form of eq 6 ( t = 0 )
Thus T~ must not depend on [NADH] and has to be inversely proportional to [HO,.],. Our results in Figure 4 are in agreement with both statements, and the value of k5 is readily obtained from Figure 4A. Values of k z and k5 obtained at different pH values are listed in Table I. It is known that the HOz-recombination process (eq 5) is pH dependent because of acid-base equilibrium:
-{d[NADHl /dtlinitial = h,[NADHI [H0,*10
H02- e H+ + 0 2 - s
(7)
(9)
1960
The Journal of Physical Chemistry, Vol. 83, No. 15, 1979
TABLE I: Values of k , , k,, and kscalcd at Different pH
Values ( t =
20 " C )
17 + 2 15f.2 131 2 11 1 1.5 8.5 f. 1.5 5.5 1 1 . 5 2 % 1
4.4 4.6 4.9 5.25 5.6 5.8 6.25
I
1513 17 1 3 221 3 1 9 f. 2 14+2 7.5 f. 1 . 5 3 f. 0.5 I
I
A. Nadezhdin and H.
1
5.5
B. Dunford
' pH
4.36
16.5 20.1 21.9 18.5 13.7 8.4 3.5 I
I
L 33 3.4 3.5 3.6
4.5
(VT)
N
& 5
-D0
A
4
6
5
PH Flgure 5. The dependence of log k, on pH at 20 O C , where k , is the apparent rate constant for the reaction of hydroperoxyl radical with NADH. The results indicate that 02-. is unreactive.
with pKg = 4.88.19 There are three possible ways of recombination HO2. + HO2. HzO2 0 2 (a) HOy + 0 2 - . HOz- + 02 (b) -+
-
+
+
+ 02-.
+
02- 02 with rate constants ha = 0.76 X lo6 M-l s-', hb = 8.7 X lo7 M-l s-l, and h, 5 0.3 M-l s-1.19,20A t the pH's of our experiments (c) is insignificant and the rate of superoxide radical recombination is given by
(4
x 103
Figure 6. The Arrhenius plots of log k , vs. reciprocal temperature at different pH values.
02--
where [H02-]designates total radical concentration, in both acidic and basic forms, and
Using formula (10) and known values of ha, kb and Kg, one can calculate h5 for various pH values (see kbcdcdin the last column of Table I). Calculated and experimentally obtained values of k5 practically coincided which is additional support for the above mechanism outlined in eq 1-5. Figure 5 is a semilogarithmic plot of h2vs. pH. One can see that the rate constant is independent of pH at pH 1 5 and is proportional to [H+] at higher pH. Since there are no acidic groups with pK's close to 5 in the NADH molecule, the pK for the acidic dissociation of H02-(4.88)19 is indicated by the results. Therefore the acidic form of the superoxide radical only is active in the oxidation of NADH. The elementary rate constant of the H02- + NADH reaction, obtained from Figure 5, is equal to (1.8
f 0.2) X lo5 M-l s-'. It is interesting to note that the effective rate constant for NADH oxidation by superoxide radical at pH 5.6, predicted by Yokota and Yamazaki from enzymatic ~ t u d i e s6, ~X lo4 M-l s-l, is quite close to our value of (8.5 f 1.5) X lo4 M-l s-' Similarly, the value of Land and Swallow,ll 1 2 7 M-l s-; at pH 8.4, is roughly in accord with our results since [HO,.] is small at higher pH. The fact that only the HOz. form of superoxide radical is active as oxidant can be explained by the better thermodynamic stability of Hz02,the product of hydrogen atom addition to H02-,compared to H02-, the product of 02-* reduction. The temperature dependence of k2 at different pH values yields the following information. Since NADH reacts with the acidic form of superoxide radicals only, the apparent second-order rate constant is given by the expression bapp
=
kZ'[H+l K9
+ [H+l
where k,' is the elementary rate constant of the NADH + HOz. reaction and K9 is defined above. At pH C pKg, hzapp= h2' and Eapp= E'. A t pH > pKg, haspp= h,'[H+]/Kg and Eapp= E ' - AH,o, where E designates activation energies and AHg" is the standard enthalpy of equilibrium 9. Figure 6 shows standard Arrhenius plots of log kzappvs. 1/T at pH 4.36 and 5.8. From the slopes of these straight lines E' and AHgoare estimated to be (5.5 f 0.5) and (2.7 f 0.7) kcal. Considering the important role of the molecular oxygen reduction products in various processes, their kinetic and thermodynamic proper tie^^^-^^ may be of some interest. In conclusion, our major finding is that although superoxide anion is unreactive as an oxidant in the presence of NADH, its protonated form, the hydroperoxyl radical, is highly reactive. This finding, if it is also valid for other systems, may help eliminate much of the controversy which currently exists in the literature concerning the oxidizing ability and hence ease of reduction of 0 2 - s and HOZ.. Supplementary Material Available: Tables A and B containing concentration and pH dependencies of ro and -[d(NADH)/dt], and temperature dependency data (2 pages). Ordering information is available on any current masthead page. References and Notes (1) H. F. Fisher, E. E. Conn, B. Vennesland, and F. H. Westheimer, J . Biol. Chem., 202, 681 (1953); F. A. Loewus, P. Ofner, H. F. Fisher, F. H. Westheimer, and B. Vennesland, ibid., 202, 699 (1953).
Evaluation of Solute-Solvent Interactions
(2) I. Yamazaki, K.-N. Yokota, and R. Nakajima in "Biochemical and
Medical Aspects of Active Oxygen", 0. Hayaishi and A. Asada, Ed., University of Tokyo Press, Tokyo, 1977,pp 91-100. (3) K. Yokota and I. Yamazaki, Biochemistry, 16, 1913 (1977). (4) J. M. McCord and I. Fridovich, J. Bid. Chem., 244,6049 (1969). (5) J. A. Fee and J. S. Valentine in "Superoxide and Superoxide Dismutase", A. M. Michelson, J. M. McCord, and I. Fridovich, Ed., Academic Press, New York, 1977,pp 19-60. (6) I. Fridovich, Science, 201,875 (1978). (7) Y. N. Kozlov and V. M. Berdnikov, Zh. Fir. Khim., 47,593 (1973). (8) J. H. Baxendale and A. A. Khan, J. Radiat. phys. Chem., 1, 1 1 (1969). (9) J. H. Baxendak and J. A. Wilson, Trans. Faracby Scc.,53,344(1957). (10)J. M. Siegel, B. A. Montgomery, and R. M. Bock, Arch. Biochem. Biophys., 82, 288 (1959). (1 1) E. J. Land and A. J. Swallow, Bbchim. Bbphys. Acta, 234,34 (1971). (12) J. Rabani, W. A. Mulac, and M. S.Matheson, J. Phys. Chem., 69,
53 (1965). (13) S.A. Chandry and K. D. Asmus, J . Chem. SOC.,Faraday Trans 7,
The Journal of Physical Chemisity, Vol. 83, No. 15, 1979
1961
68, 1010 (1972). (14)A. D. Nadezhdin, Yu. N. Kozlov, and A. P. Purmal, Zh. Fir. Khim., 49,2263 (1975). (15) C. C. Hatchard and C. A. Parker, Proc. R . SOC.London, Ser. A , 235,518 (1956). (16) S.Udenfriend, "Fluorescence Assay in Biology and Medicine", Vol. 1, Academic Press, New York, 1962,pp 245-249. (17) R. F. Chen, G. G. Vurek, and N. Alexander, Science, 156,949(1967). (18) N. M. Emanuel and D. G. Knorre, "Chemical Kinetics: Homogeneous Reactions". Wilev. New York. 1973. D 328. (19) D. Behar, G.Chapski, J. Rabani, L. M. Dorfman, and H. A. Schwarz, J . Phys. Chem., 74,3209 (1970). (20) B. H. Bielski and A. 0. Allen, J . Phys. Chem., 81, 1048 (1977). (21) Y. Sawada, T. Iyanagi, and I. Yamazaki, Biochemistry, 14,3761 (19751. (22) k. A. Holroyd and B. H. J. Bielski, J. Am. Chem. Soc., 100,5796 (1978). (23) P. S.Rao and E. Hayon, J . Phys. Chem., 79,397 (1975).
Quantitative Evaluation of Solute-Solvent Interactions. Environmental Effects on the n-Butylaminolysis of Tetrachloro-N-n-butylphthalimide Ottd B.Nagy,* Mukana wa Muanda," and JInos f3.Nagylb Laboratoire de Chimie Ggn6rale et Organique, Universit.4 Catholique de Louvain, Bitiment Lavoisier, Place Louis Pasteur, I, 1348-Louvain-la-Neuve,Belgium (Received September 26, 1977; Revised Manuscript Received February 20, 1979)
Tetrachloro-N-n-butylphthalimide reacts readily with n-butylamine in aprotic media to form the corresponding diamide. The reaction is third order with respect to butylamine both at low and high amine concentrations. The substrate and the nucleophile form a n-a type complex whose catalyzed transformation leads to the final product. Compared to cyclohexane as a standard, all other solvents used (cyclohexene,benzene, toluene, rn-xylene, p-xylene, mesitylene, and a-methylnaphthalene) solvate the substrate preferentially and partially inhibit the reaction. Since the dielectric constant has been fixed detailed analysis shows that specific solvent effects (P--R interactions) are responsible for the observed effects. Therefore the reaction performed in cyclohexane with added aromatic solvent represents a case where an external complexing agent influences the reactivity. The use of liquid complexing agents made it possible to extend their concentration up to 100% in a series of stripping experiments where cyclohexane has been gradually replaced by a-donor solvents. Thus a connection between complexation and solvation has been realized. The results obtained in these mixed solvents cannot be interpreted along traditional lines. Therefore a new model, called the competitive preferential solvation theory, is proposed which allows the interpretation of the observed kinetic anomalies and throws new light on the solvatingcomplexing role of the solvent. For the first time a linear free energy relationship is established which connects the interacting power of the solvent to its catalytic activity.
It is now well established that molecular complexes may play a catalytic role in chemical transformations,2-5 On the other hand, solvents are known to considerably modify chemical reactivityn6Therefore the question arises whether these two apparently different catalytic activities are really of a different nature or whether they can be considered as just two particular aspects of the same phenomenon. The problem is particularly important in the case of weak interactions since relatively high complexing agent concentrations are required to observe any noticeable kinetic effect. As a result the mechanistic assignment of complexing and/or solvating role played by the complexing molecule is difficult. At high complexing agent concentrations the reaction medium is considerably changed and we have to deal with the influence of mixed solvents on chemical reactivity. There is a large amount of work published on the kinetic effects of mixed solvents. They concern mainly ionic, solvolytic, or ion-forming reactions in protic or aproticprotic solvent mixtures where general solvent effects dominate. No work seems to exist on nonprotic and 0 0 2 2 - 3 6 5 4 / 7 9 ~ 2 0 ~ 3 - 1 9 6 i,0010 ~oi
nonionic systems in the full concentration range where specific effects play an important role. Therefore, it is appropriate to examine this problem in more detail and in the present paper we attempt to apply a new unified model for complexation-solvation phenomena. For this purpose we have examined the n-butylaminolysis of tetrachloro-N-n-butylphthalimidein aprotic media. This substrate has been chosen for two reasons. First, due to the particular geometry of the molecule, complexing agents may simultaneously influence both the leaving group (substituted benzamide) and the substrate itself (substituted benzoyl). Second, phthalimide and its derivatives have a considerable pharmacological interest on account of their teratogenic activity. Imidazole and its derivatives (e.g., histamine) partially inhibit the hydrolysis of phthalimide-containing corn pound^.^-^ On the other hand, histamine is a very important factor in controlling the fetal development of the embryo.'O Therefore it is believed that it is the formation of this nonproductive histamine-phthalimide drug complex which is responsible for the observed teratogenic a ~ t i v i t y .The ~ aprotic media
0 1979 American
Chemical Society
