Evidence that metaphosphate monoanion is not an intermediate in

-0.98 to -0.79 as the pATa of the oxygen nucleophile increases from 3.6 to 15.7 (aqueoussolution; 25 °C; ionic strength, 1.5). This coupling between ...
0 downloads 0 Views 1MB Size
J . Am. Chem. SOC.1989, 1 1 1 , 7519-7586

1519

Evidence That Metaphosphate Monoanion Is Not an Intermediate in Solvolysis Reactions in Aqueous Solution' Daniel Herschlag and William P. Jencks* Contribution No. 1683 from the Graduate Department of Biochemistry, Brandeis University, Waltham, Massachusetts 02254. Received February 27, 1989

Abstract: The slope of plots of log k2 against the pKa of the leaving group for reactions of uncharged, monoanionic, and dianionic oxygen nucleophiles with three phosphorylated pyridine monoanions changes from PI, = -0.98 to -0.79 as the pK, of the oxygen nucleophile increases from 3.6 to 15.7 (aqueous solution; 25 "C; ionic strength, 1.5). This coupling between the strength of the nucleophile and the measure of bond cleavage is described by an interaction coefficient, pxy= dp,,/dpK,,, = dp,,,/dpK,, = 0.013, and provides evidence for concerted phosphoryl transfer between pyridine and oxygen bases. The solvolysis of phosphorylated pyridines and of acetyl phosphate in aqueous solution also shows the behavior expected for a bimolecular substitution reaction with no metaphosphate intermediate: (1) The values of log k2 for reactions of oxygen nucleophiles, including water, with phosphorylated y-picoline monoanion follow a single line of slope 0.51 in a plot against log k2 for the corresponding bimolecular reactions with methyl 2,4-dinitrophenyl phosphate monoanion (Kirby, A . J.; Younas, M. J . Chem. SOC.B 1970, 1165). (2) Water behaves as expected for a nucleophile of its pKa in a Br~nsted-typeplot of log k2 for reaction with phosphorylated y-picoline against the pK, of oxygen nucleophiles. (3) The value of PI, = -1.02 for the hydrolysis of phosphorylated pyridines is less negative than the value of Ow' = -1.25 for complete breaking of the P-N bond. (4) The value of Pi, = -1.02 for the reaction of water with phosphorylated pyridines fits the correlation of PI, against the pKa of oxygen nucleophiles for concerted, bimolecular reactions, with a slope ofp, = a~l,/dpKn,, = 0.013. (5) The reactions of acetyl phosphate monoanion and dianion with aqueous alcohols show selectivity that depends on the pKa of the alcohol, with a larger selectivity for the less reactive dianion [50% aqueous alcohol (v/v), 55 "C].

Metaphosphate monoanion has been discussed as a possible reaction intermediate since 1955 when eq 1 was proposed in order to explain the rapid hydrolysis of phosphate monoester monoan-

+

HOR

Scheme I 0

I/

H20

0

0

OH

ions.2 Phosphoryl-transfer reactions in aqueous solution exhibit a small dependence on the pKa of the nucleophile, p,,, and a large dependence on the pKa of the leaving group, -PI,, which suggests that there is a small amount of bond making and a large amount of bond breaking in the transition state. They also exhibit near-zero values of the entropy and volume of activation and a '~ large isotope effect with lSOin the leaving g r o ~ p . ~ -Although (1) This research was supported in part by grants from the National

Institutes of Health (GM20888 and 4-61271) and the National Science Foundation (PCM-8117816). D.H. was also supported by a fellowship from the Gillette Foundation. (2) Butcher, W. W.; Westheimer, F. H. J . Am. Chem. SOC.1955, 7 7 , 2420-2424. Kumamoto, J.; Westheimer, F. H. J . Am. Chem. SOC.1955, 77, 2515-2518. Barnard, P. W. C.; Bunton, C. A,; Llewellyn, D. R.; Oldham, K. G.; Silver, B. L.; Vernon, C. A. Chem. I n d . (London) 1955, 760-763. Bunton, C. A.; Llewellyn, D. R.; Oldham, K. G.; Vernon, C. A. J . Chem. SOC. 1958,3574-3587. Jencks, W. P. Brookhauen Symp. 1962, No. 15, 134-153. Bruice, T. C.; Benkovic, S. J. Bioorganic Mechanisms; Benjamin: New York, 1966; Vol. 2, pp 1-109. Westheimer, F. H. Chem. Rev. 1981, 81, 313-326. (3) (a) Whalley, E. Can. J . Chem. 1962,40, 1220-1224. (b) Chanley, J. D.; Feageson, E. J . Am. Chem. SOC.1963, 85, 1181-1190. (c) Cox, J. R., Jr.; Ramsay, 0. B. Chem. Rea. 1964, 64, 317-352. (d) Jencks, W. P., Gilchrist, M. J . Am. Chem. SOC.1965, 87, 3199-3209. (e) Kirby, A. J.; Jencks, W. P.J . Am. Chem. SOC.1965,87, 3209-3216. (f) Bunton, C. A,; Fendler, E. J.; Fendler, J. H. J . Am. Chem. SOC.1967, 89, 1221-1230. (9) Milstein, S.; Fife, T. H. J . Am. Chem. SOC.1967,89,5820-5826. (h) Phillips, D. R.; Fife, T. H. J . Org. Chem. 1969, 34, 2710-2714. (i) Benkovic, S. J.; Sampson, E. J. J . Am. Chem. SOC.1971, 93, 4009-4016. (j)Osterheld, R. K. Top. Phosphorus Chem. 1972, 7, 103-254. (k) Gorenstein, D. G.; Lee, Y . - G . ,Kar, D.J. J . Am. Chem. SOC.1977, 99, 2264-2267. (I) Ramirez, F.; Marecek, J.; Minore, J.; Srivastava, S.;le Noble, W. J . Am. Chem. SOC.1986, 108, 348-349. (4) Kirby, A. J.; Varvoglis, A. G. J . Chem. SOC.E 1968, 135-141. ( 5 ) DiSabato, G.; Jencks, W. P. J . Am. Chem. SOC.1961,83, 4400-4405. (6) Jameson, G. W.; Lawlor, J. M . J . Chem. SOC.B. 1970, 53-57. (7) Bourne, N.; Williams, A. J . Am. Chem. SOC.1984, 106, 7591-7596. (8) Skoog, 41.T.; Jencks, W. P. J . Am. Chem. SOC.1984, 106, 7597-7606.

0002-7863/89/ 15 11-7579$01.50/0

11 -0-P-Y I -0

+

0

\ \/ 1 k .

A

Y

0-

xY

-

+

x+-P-0I I

X*P.Y

k2

0 XI-P-0-

0-

+

Y

B

0-

C

0-

0-

these data characterize the transition state as metaphosphate-like, they do not provide evidence for a metaphosphate intermediate. Indeed, we are aware of no evidence that requires the formation of a metaphosphate monoanion in aqueous solution."~12 There are three types of evidence that a metaphosphate intermediate is not formed in the bimolecular reactions of phosphorylated pyridines with added amine nucleophiles in aqueous soIution:'J (1) Three possible routes for phosphoryl transfer are shown in Scheme I: A is a concerted pathway; B is a stepwise pathway through a preassociation mechanism, in which a metaphosphate intermediate is formed but does not have a lifetime sufficient to allow diffusion; and C is a stepwise pathway involving a freely diffusing metaphosphate intermediate. The fact that reactions of phosphorylated pyridines 1 (X = H , 4-CH3, 4-morpholino, 3-CH30)with added pyridine and amine nucleophiles are second order is consistent with concerted (A) or stepwise preassociation (B) mechanisms, but not with a free metaphosphate intermediate (C).6-8 With a stepwise preassociation mechanism (B), (9) Jencks, W. P.; Haber, M. T.; Herschlag, D.; Nazaretian, K. L. J . Am. Chem. SOC.1986, 108, 479-483. (IO) Kirby, A. J.; Varvoglis, A. G. J . Am. Chem. Soc. 1967,89,415-423. (11) Jencks, W. P. Chem. SOC.Rec. 1981, IO, 345-375. (12) Herschlag, D.; Jencks, W. P. J . Am. Chem. SOC. 1986, 108, 7938-7946. and references therein.

C2 1989 American Chemical Society

1580 J. A m . Chem. Soc., Vol. 111, No. 19, 1989

Herschlag and Jencks

01

Bransted-type plots of log k2 against pK, are expected t o give a break where t h e pK, of t h e nucleophile and leaving g r o u p are equal, from a c h a n g e in rate-limiting step. However, no break is observed in these plots for phosphoryl transfer between pyridines.'~~ (2) An increase in Pnucwith poorer leaving groups in these reactions can be described by an interaction coefficientI3 of pxy = 0.014 (eq 2).

This interaction is evidence for coupling between bond formation t o the nucleophile and bond breaking t o t h e leaving group in a concerted reaction (A) .8 (3) The r a t e constant for diffusion a p a r t of an encounter pair is -lO1os-l. T h i s sets a lower limit on t h e r a t e constant for reaction of a nucleophile with a metaphosphate intermediate by the stepwise preassociation mechanism B because t h e reaction occurs, by definition, prior t o diffusional separation of the metaphosphate intermediate and the nucleophile. The upper limit for this rate constant is the frequency of a bond vibration, l O I 3 s-I, because collapse of an intermediate to form the product can b e no faster t h a n vibrational motion. Therefore, t h e most t h a t t h e observed r a t e constant can vary with t h e preassociation ' ~ = lo3; this is less t h a n t h e mechanism B is l O I 3 ~ - ] / 1 0 s-l observed variation8 of lo5. The mono- and dianion of acetyl phosphate react with inorganic phosphate monoanion in concentrated aqueous sodium perchlorate t o give different yields of pyrophosphate; this proves t h a t there is not a common free metaphosphate intermediate.12 The second-order reactions of phosphorylated pyridine monoanions with inorganic phosphate anions in t h e absence of high concentrations of salt are also consistent with concerted phosphoryl transfer t o

-

ph0~phate.I~ However, t h e intermediacy of metaphosphate monoanion in reactions with solvent is not excluded by these results because there could be a barrier for t h e reaction of metaphosphate with a weak nucleophile such as water; Le., metaphosphate could have a significant lifetime in t h e presence of water, b u t not in the presence of the stronger pyridine nucleophiles. The inversion of configuration in solvolysis reactions of phosphoryl compounds in aqueous solution15 and t h e absence of positional isotope exchange of t h e a-P bridge oxygen a t o m of [P-I8O4]ADPwith t h e nonbridge a-oxygen a t o m s concurrent with hydrolysisI6 show t h a t there is not an intermediate with a lifetime long enough t o allow diffusion or rotation, but do not distinguish between stepwise and concerted preassociation mechanisms. In this and t h e accompanying paper we describe experiments that were designed t o characterize t h e reactions of phosphorylating agents with anionic oxygen nucleophiles a n d water. This paper is concerned with t h e mechanism of these reactions; t h e following paper provides an evaluation of t h e role of electrostatic repulsion in reactions with anions a n d additional characterization of t h e n a t u r e of t h e transition s t a t e for phosphoryl transfer. In this report we first provide evidence that phosphoryl transfer from pyridines to oxygen nucleophiles is concerted, like phosphoryl transfer between t w o pyridines. We t h e n show t h a t (1) t h e behavior of water in structure-reactivity correlations is like t h a t of ( 1 3 ) Jencks, D. A.; Jencks, W. P. J . Am. Chem. SOC.1977,99,7948-7960. Jencks, W. P. Chem. Rev. 1985, 85, 511-527. (14) Herschlag, D.; Jencks, W. P., in preparation. (15) Buchwald, S. L.; Knowles, J . R. J . Am. Chem. SOC.1982, 104, 1438-1440. Buchwald, S. L.; Friedman, J. M.; Knowles, J. R. J . Am. Chem. SOC. 1984, 106, 491 1-4916. (16) Lowe, G.;Tuck, S. P. J . Am. Chem. SOC.1986, 108, 1300-1301.

other oxygen nucleophiles t h a t react through concerted, secondorder processes, (2) positive r a t e deviations t h a t arise from solvolysis by an additional p a t h w a y through a metaphosphate intermediate are not observed, and (3) the d a t a are inconsistent with a stepwise preassociation mechanism with rate-limiting addition of water to a metaphosphate intermediate. Evidence for concerted phosphoryl transfer from acetyl phosphate t o alcohols in aqueous solution is also described.

Experimental Section Materials. y-Picoline, pyridine, 2-methoxyethanol, 3-hydroxypropionitrile, acetohydroxamic acid, succinic acid, cacodylic acid, and trimethylamine N-oxide were purified by distillation or recrystallization. Aqueous solutions of phosphorylated y-picoline (PicP),I7 phosphorylated 4-morpholinopyridine, phosphorylated pyridine, and acetyl phosphate were prepared as described previously.8*'z.184-Morpholinopyridine was a gift from Dr. Mark Skoog. Reactions of Phosphorylated Pyridines. Reactions of 2 X lo4 M PicP, M phosphorylated 4-morpholinopyridine, and 5 X lo4 M phosphorylated pyridine at 25.1 f 0.1 OC were followed spectrophotometrically at 256-258, 303, and 262 nm, respectively. These reactions were first order for >3tlj2; end points were determined after 210fl/2. The ionic strength was maintained at 1.5 with potassium chloride, and the pH was determined at the end of each reaction. Reactions of Acetyl Phosphate. Initial concentrations of acetyl phosphate in reaction mixtures were determined colorimetrically after conversion to acetohydroxamic acid by reaction with hydroxylamine by the method of Lipmann and Tuttle" with minor modification: 1 mL of 1:1 4 M NH20H.HC1/3.5 M N a O H was added to a 1-mL aliquot; after 2 1 5 min at room temperature 4 mL of 10% FeC13.6H20 in 0.7 M HCI was added and the absorbance at 540 nm was compared to that with acetohydroxamic acid. Reactions of acetyl phosphate in 50% aqueous alcoholic solutions (V/Vat 25 "C) were studied by product analysis after incubation overnight at 55 OC. In general, the alkyl phosphate product was assayed as follows. Inorganic phosphate was precipitated by the addition of 2 mL of a solution of 0.27 M MgCI2, 1.8 M NH,CI, and 1.5 M N H 4 0 H to a 2-mL aliquot of the reaction mixture and incubation at 4 "C for 2 3 h. After centrifugation, aliquots of the supernatant solution were removed to assay for small amounts (