Leaving group assistance in the La3+-catalyzed cleavage of dimethyl (o-methoxycarbonyl)aryl phosphate triesters in methanol

Article abstract of DOI:10.1021/ja904659e

The catalytic methanolysis of a series of dimethyl aryl phosphate triesters where the aryl groups contain an o-methoxycarbonyl (o-CO₂Me) substituent (4a-i) was studied at 25°C in methanol containing La ³⁺ at various concentrations and _s^spH. Determination of the second-order rate constant for La³⁺ ₂-catalyzed cleavage of substrate 4a (dimethyl (o-methoxycarbonyl) phenyl phosphate) as a function of _s^spH was assessed in terms of a speciation diagram that showed that the process was catalyzed by La³⁺ ₂(^-OCH₃)_x dimers, where x = 1-5, that exhibit only a 5-fold difference in activity between all the species. The second-order catalytic rate constants (k₂^La) for the catalyzed methanolysis of 4a-i at _s^spH 8.7 fit a Bronsted relationship of log k₂^La= (-0.82 ± 0.11)_s^spK_a^lg + (11.61 ± 1.48), where the gradient is shallower than that determined for a series of dimethyl aryl phosphates that do not contain the o-CO₂Me substituent, log k₂^La = (-1.25 ± 0.06)_s ^spK_a^lg + (16.23 ± 0.75). Two main observations are that (1) the o-CO₂Me group preferentially accelerates the cleavage of the phosphate triesters with poor leaving groups relative to those with good leaving groups and (2) it provides an increase in cleavage rate relative to those of comparable substrates that do not have that functional group, e.g., k₂^La(dimethyl o-(methoxycarbonyl) phenyl phosphate)/k₂^La(dimethyl phenyl phosphate) = 60. Activation parameters for the La³⁺₂-catalyzed methanolysis of 4a and dimethyl 4-nitrophenyl phosphate show respective ΔH? (ΔS?) values of 3.3 kcal/mol (-47 cal/mol·K) and 0.7 kcal/mol (-46.5 cal/mol·K). The data are analyzed in terms of a concerted reaction where the catalytic complex (La³⁺₂( ^-OCH₃)_x-1) binds to the three components of a rather loose transition state composed of a nucleophile CH₃O ^-, a nucleofuge ^-OAr, and a central (RO)₂P ²⁺-O^- in a way that provides leaving group assistance to the departing aryloxy group.

Full text of DOI:10.1021/ja904659e

Published on Web 09/08/2009
Leaving Group Assistance in the La3+-Catalyzed Cleavage of
Dimethyl (o-Methoxycarbonyl)aryl Phosphate Triesters in
Methanol
David R. Edwards, C. Tony Liu, Graham E. Garrett, Alexei A. Neverov, and
R. Stan Brown*
Department of Chemistry, Queen’s UniVersity, Kingston, Ontario, Canada K7L 3N6
Received June 18, 2009; E-mail: rsbrown@chem.queensu.ca
Abstract: The catalytic methanolysis of a series of dimethyl aryl phosphate triesters where the aryl groups
contain an o-methoxycarbonyl (o-CO2Me) substituent (4a-i) was studied at 25 °C in methanol containing
La3+ at various concentrations and sspH. Determination of the second-order rate constant for La3+2-catalyzed
cleavage of substrate 4a (dimethyl (o-methoxycarbonyl)phenyl phosphate) as a function of sspH was assessed
in terms of a speciation diagram that showed that the process was catalyzed by La3+2(-OCH3)x dimers,
where x ) 1-5, that exhibit only a 5-fold difference in activity between all the species. The second-order
s
catalytic rate constants (kL2a) for the catalyzed methanolysis of 4a-i at pH 8.7 fit a Brønsted relationship
s
of log kL2a ) (-0.82 ( 0.11)sspKlag + (11.61 ( 1.48), where the gradient is shallower than that determined for
a series of dimethyl aryl phosphates that do not contain the o-CO2Me substituent, log kL2a ) (-1.25 (
0.06)sspKlag + (16.23 ( 0.75). Two main observations are that (1) the o-CO2Me group preferentially accelerates
the cleavage of the phosphate triesters with poor leaving groups relative to those with good leaving groups
and (2) it provides an increase in cleavage rate relative to those of comparable substrates that do not have
that functional group, e.g., kL2a(dimethyl o-(methoxycarbonyl)phenyl phosphate)/kL2a(dimethyl phenyl phos-
phate) ) 60. Activation parameters for the La3+2-catalyzed methanolysis of 4a and dimethyl 4-nitrophenyl
phosphate show respective ∆Hq (∆Sq) values of 3.3 kcal/mol (-47 cal/mol ·K) and 0.7 kcal/mol (-46.5
cal/mol·K). The data are analyzed in terms of a concerted reaction where the catalytic complex
(La3+2(-OCH3)x-1) binds to the three components of a rather loose transition state composed of a nucleophile
CH3O-, a nucleofuge -OAr, and a central (RO)2P2+-O- in a way that provides leaving group assistance
to the departing aryloxy group.
conserve all the metal-binding ligands.3 While not exactly
established, the mechanism of action of these has been discussed
as involving intramolecular delivery of a Zn(II)-coordinated
hydroxide (possibly one bridged between the two Zn(II) ions)
to a phosphate ester substrate which is transiently activated by
PdO coordination to the second Zn(II) ion.4 This sort of
dinuclear core is characteristic of other enzymes that mediate
hydrolytic cleavage of phosphate diesters5 and monoesters6 and
gives the P. diminuta bacterium the ability to hydrolyze
1. Introduction
Phosphate triesters are not naturally occurring molecules and
so have no known natural biological function. However man-
made organophosphate (OP) pesticides and chemical weapons
(CW) of general structure 1 (Chart 1), where LG is a leaving
group with pKa e 8, are strong acetylcholinesterase inhibitors.1
The OP pesticides have important uses as animal and crop
protectants2 but are under increasing regulatory scrutiny due to
toxicity and environmental persistence. Phosphotriesterase (PTE)
enzymes produced by soil-dwelling bacteria, while having no
known naturally occurring substrates, are very effective at
destroying neutral OP pesticides, thereby protecting the organism
from poisoning. The two most studied PTEs, from Agrobacte-
rium radiobacter and Pseudomonas diminuta, are dinuclear
Zn(II) enzymes that have nearly 90% amino acid homology and
paraoxon, its preferred substrate, with an efficiency near the
7
diffusion limit, kcat/KM ≈ 108 M-1 s-1
.
(3) Yang, H.; Carr, P. D.; McLoughlin, S. Y.; Liu, J. W.; Horne, I.; Qiu,
X.; Jeffries, C. M. J.; Russell, R. J.; Oakeshott, J. G.; Ollis, D. L.
Protein Eng. 2003, 16, 135.
(4) (a) Jackson, C. J.; Foo, J.-L.; Kim, H.-K.; Carr, P. D.; Liu, J.-W.;
Salem, G.; Olis, D. L. J. Mol. Biol. 2008, 375, 1189, and references
therein. (b) Kim, J.; Tsai, P.-C.; Chen, S.-L.; Himo, F.; Almo, S. C.;
Raushel, F. M. Biochemistry 2008, 47, 9497, and references therein.
(5) (a) Lipscomb, W.; Stra¨ter, N. Chem. ReV. 1996, 96, 2375. (b) Collman,
J. E. Curr. Opin. Chem. Biol. 1998, 2, 222. (c) Cowan, J. A. Chem.
ReV. 1998, 98, 1067. (d) Davies, J. F.; Hostomska, Z.; Hostomsky,
Z.; Jordan, S. R.; Mathews, D. A. Science 1991, 252, 88. (e) Beese,
L. S.; Steitz, T. A. EMBO J. 1991, 10, 25. (f) Lahm, A.; Volbeda, S.;
Suck, D. J. Mol. Biol. 1990, 215, 207.
(1) (a) Main, R. A.; Iverson, F. Biochem. J. 1966, 100, 525. (b) Emsley,
J.; Hall, D. The Chemistry of Phosphorus; Wiley: New York, 1976; p
494.
(2) (a) Toy, A.; Walsh, E. N. Phosphorus Chemistry in EVeryday LiVing,
2nd ed.; American Chemical Society: Washington, DC, 1987; Chapters
18-20. (b) Quin, L. D. A Guide to Organophosphorus Chemistry;
Wiley: New York, 2000. (c) Gallo, M. A.; Lawryk, N. J. Organic
Phosphorus Pesticides. The Handbook of Pesticide Toxicology;
Academic Press: San Diego, CA, 1991.
(6) Gani, D.; Wilke, J. Chem. Soc. ReV. 1995, 24, 55.
(7) Omburo, G. A.; Kuo, J. M.; Mullens, L. S.; Raushel, F. M. J. Biol.
Chem. 1992, 267, 13278.
9
13738 J. AM. CHEM. SOC. 2009, 131, 13738–13748
10.1021/ja904659e CCC: $40.75  2009 American Chemical Society

La3+-Catalyzed Methanolysis of Dimethyl Aryl Phosphate Triesters
A R T I C L E S
Chart 1
There are several man-made metal-ion-containing catalysts
that promote the hydrolysis and alcoholysis of phosphate triesters
and related neutral OP esters.8-10 Studies of these systems have
helped define the modes by which metal ions can facilitate the
cleavage of neutral OP esters as well as phosphate esters in
general:8a (1) via Lewis acid activation of the phosphate ester
by metal ion coordination to the phosphoryl oxygen; (2) by
nucleophile activation via solvent coordination to the metal ion
to reduce its pKa and generate an active Mx+--OR species;
and (3) by providing leaving group assistance (LGA) via
coordination of the departing oxyanion to a metal ion, thereby
facilitating P-OLG bond cleavage (OLG ) leaving group).
Recently we reported a simple artificial phosphotriesterase
system comprising La3+ in methanol that is highly catalytic for
the cleavage of neutral phosphates and phosphorothioates,10
phosphonates,11 and phosphonothioates,12 the latter two OP
materials being taken as models for the G- and V-classes of
in PTE, where the ꢀlg is reported to be -1.8 to -2.2.16 The
extent of bond cleavage in the La3+-catalyzed reaction is far
greater than for the methoxide-catalyzed process, where the ꢀlg
was determined to be -0.70 ( 0.05.10c The results for both the
enzymatic and La3+ systems were rationalized by invoking the
first two proposed modes of metal ion catalysis, namely Lewis
acid activation of the bound substrate and intramolecular
delivery of a metal-ion-coordinated nucleophile. However, for
each system LGA was explicitly rejected as being inconsistent
with the large negative ꢀlg values.10c,17
Leaving group assistance by metal ions has been difficult to
demonstrate in the cleavage of phosphate diesters18 and more
so for phosphate triesters,19 although the closely related general
acid assistance of the cleavage of di-20 and triesters21 has been
amply documented. Our recent studies demonstrated LGA by
a dinuclear Zn(II):(-OCH3) complex during its catalytic cleavage
of some specific methyl aryl phosphates having o-methoxycar-
bonyl (o-CO2Me) and nitro groups in essentially neutral
methanol.22a Similarly, Yb3+ exhibits LGA of the methanolytic
cleavage of a series of methyl (o-methoxycarbonyl)aryl phos-
phate diesters (2a-h, Chart 1) in acidic methanol.22b The
o-CO2Me substituent was suggested to play the key role of
transiently coordinating to the metal ion, thereby positioning
the Yb3+ to provide a 1012 acceleration for the methanolytic
cleavage of the P-OAr bond.22b
nerve agents. The Brønsted plot determined for the La3+
catalyzed reaction of diethyl aryl phosphates at a nearly neutral
-
sspH of 9.1 in methanol13,14 fits a linear regression of log kL2a
)
(-1.43 ( 0.05)sspKaHOAr - (17.60 ( 1.07). The large gradient
of the line (ꢀlg) signifies that the catalyzed process has a very
loose transition state (TS)15 with a high degree of P-OAr bond
cleavage, which is very similar to the situation believed to occur
The ability of the o-CO2Me group to bring about such large
LGA for the Yb3+-promoted methanolysis of phosphate diesters
spurred this investigation of whether a similar effect would be
operative in the La3+-catalyzed cleavage of phosphate triesters
3a-e vs 4a-i (Chart 1). While one might wonder whether LGA
is even remotely required for a process where there is already
such significant P-OAr cleavage as in the La3+-catalyzed
methanolysis of diethyl aryl phosphates, herein we show that
the departure of the o-CO2Me-substituted aryloxy group is
subject to LGA in the La3+-promoted cleavage of 4. LGA is
(8) (a) Williams, N. H.; Takasaki, B.; Wall, M.; Chin, J. Acc. Chem. Res.
1999, 32, 485. (b) Weston, J. Chem. ReV. 2005, 105, 2152. A
compendium of works to 2005 is found in refs 5 and 6 of citation
10c.
(9) (a) Klinkel, K. L.; Kiemele, L. A.; Gin, D. L.; Hagadorn, J. R. J.
Mol. Catal. A 2007, 267, 173. (b) Aguilar-Perez, F.; Gomez-Tagle,
P.; Collado-Fregoso, E.; Yatsimirsky, A. K. Inorg. Chem. 2006, 45,
9502. (c) Sanchez-Lombardo, I.; Yatsimirsky, A. K. Inorg. Chem.
2008, 47, 2514. (d) Morrow, J. R.; Trogler, W. C. Inorg. Chem. 1988,
27, 3387.
(10) (a) Tsang, J. S.; Neverov, A. A.; Brown, R. S. J. Am. Chem. Soc.
2003, 125, 7602. (b) Tsang, J. S. W.; Neverov, A. A.; Brown, R. S.
Org. Biomol. Chem. 2004, 2, 3457. (c) Liu, T.; Neverov, A. A.; Tsang,
J. S. W.; Brown, R. S. Org. Biomol. Chem. 2005, 3, 1525.
(11) Lewis, R. E.; Neverov, A. A.; Brown, R. S. Org. Biomol. Chem. 2005,
3, 4082.
(16) (a) Caldwell, S. R.; Newcomb, J. R.; Schlecht, K. A.; Raushel, F. M.
Biochemistry 1991, 30, 7438. (b) Hong, S.-B.; Raushel, F. M.
Biochemistry 1996, 35, 1094.
(17) Aubert, S. D.; Li, Y.; Raushel, F. M. Biochemistry 2004, 43, 5707.
(18) (a) Browne, K. A.; Bruice, T. C. J. Am. Chem. Soc. 1992, 114, 4951.
(b) Dempcy, R. O.; Bruice, T. C. J. Am. Chem. Soc. 1994, 116, 4511.
(c) Bruice, T. C.; Tsubouchi, A.; Dempcy, R. O.; Olson, L. P. J. Am.
Chem. Soc. 1996, 118, 9867.
(12) Melnychuk, S. A.; Neverov, A. A.; Brown, R. S. Angew. Chem., Int.
Ed. 2006, 45, 1767.
(13) The nomenclatupre for pH in methanol is described by Bosch and
co-workers14 as per recommendations of the IUPAC: Compendium
of Analytical Nomenclature. DefinitiVe Rules 1997, 3rd ed.; Blackwell:
Oxford, UK, 1998. wwpH is used for the measured pH of an aqueous
solution when the measuring electrode is calibrated with aqueous
buffers; if the electrode is calibrated in water and the “pH” of the
neat methanol solution is measured, the term ws pH is used, and if a
correction factor of-2.24 is subtracted from the latter reading, then
the term sspH is used to designate the “pH” measured in the methanol
(19) Morrow, J. R.; Trogler, W. C. Inorg. Chem. 1989, 28, 2330.
(20) (a) Kirby, A. J.; Lima, M. F.; da Silva, D.; Roussev, C. D.; Nome, F.
J. Am. Chem. Soc. 2006, 128, 16944. (b) Mikkola, S.; Stenman, E.;
Nurmi, K.; Youseffi-Salakdeh, E.; Stromberg, R.; Lonnberg, H.
J. Chem. Soc., Perkin Trans. 2 1999, 1619. (c) Irisawa, M.; Takeda,
N.; Komiyama, M. J. Chem. Soc., Chem. Commun. 1995, 1221. (d)
Komiyama, M.; Matsumoto, Y.; Takahashi, H.; Shiiba, T.; Tsuzuki,
H.; Yajima, H.; Yashiro, M.; Sumaoka, J. J. Chem. Soc., Perkin Trans.
2 1999, 691.
solution and referenced to the same solvent.
s
(14) Since the Kauto of methanol is 10-16.77 M2, neutral pH is 8.4: Bosch,
s
E.; Rived, F.; Rose´s, M.; Sales, J. J. Chem. Soc., Perkin Trans. 2
1999, 1953.
(21) (a) Kirby, A. J.; Tondo, D. W.; Medeiros, M.; Souza, B. S.; Priebe,
J. S.; Lima, M. F.; Nome, F. J. Am. Chem. Soc. 2009, 131, 2023. (b)
Asaad, N.; Kirby, A. J. J. Chem. Soc., Perkin Trans. 2 2002, 1708.
(15) Hengge, A. C. AdV. Phys. Org. Chem. 2005, 40, 49.
9
J. AM. CHEM. SOC. VOL. 131, NO. 38, 2009 13739

A R T I C L E S
Edwards et al.
product. Second-order rate constants (k-2 OMe) were determined from
the slopes of the linear plots of kobs versus [NaOCH3].
evidenced by (1) a rate enhancement relative to phosphate esters
not having this ortho-substituent and (2) a reduction in the
Brønsted ꢀ value from -1.25 for substrates 3 to -0.82 for
substrates 4. Furthermore, we show that the catalytic TS with
4 has a degree of P-OAr cleavage similar to that observed for
the La3+-promoted cleavage of phosphates 3 and one that is
significantly looser than that for the corresponding methoxide-
promoted reaction.
d. Activation Parameters for the -OCH3-Catalyzed and
La3+-Catalyzed Methanolysis of 4a. The temperature dependence
of kL2a was investigated by determining the kobs values in duplicate
for the methanolysis of 4a at six temperatures from 11 to 52 °C
La
using 1.5 mM La(OTf)3, sspH 8.7. The k2 constants were
determined by dividing the kobs values by [La]total/2. Fitting of the
kL2a vs 1/T data to the Eyring equation gave ∆Hq and ∆Sq values of
3.3 ( 0.1 kcal/mol and -47.0 ( 0.4 cal/mol ·K, respectively. The
activation parameters for the methoxide-promoted reaction of 4a
were determined analogously from the k-2 OMe values at six temper-
atures between 12 and 50 °C, giving ∆Hq and ∆Sq values of 15.6
( 0.3 kcal/mol and -17.4 ( 0.95 cal/mol ·K, respectively.
e. Identification of Reaction Products of the La3+-Catalyzed
Cleavage of 4d. To a glass vial containing 10 mL of methanol
(93% deuterium content) were added 4d (0.01 mmol), La(OTf)3
(0.05 mmol), and NaOMe (0.05 mmol). The reaction mixture was
left at room temperature for 4 h, after which it was concentrated to
∼1 mL and the 1H NMR spectrum (600 MHz) obtained. The
spectrum indicated a mixture of 4-iodo methylsalicylate [δ 7.45
(1H, d, ArH, J ) 8.58 Hz), 7.05 (1H, d, ArH, J ) 1.41 Hz), 6.80
(1H, dd, ArH, J ) 1.41, 8.32 Hz), 3.93 (<3H, s, OCH3)] and
trimethylphosphate [δ 3.73 (6H, d, P-(OCH3)2, J ) 12 Hz)]
containing a single CD3O phosphoryl substituent. A 31P NMR
(243.06 MHz) spectrum, referenced to 70% phosphoric acid,
showed a single signal at 2.72 ppm.
2. Experimental Section
a. Materials. Methanol (99.8% anhydrous), La(CF3SO3)3, tet-
rabutylammonium methoxide (1.0 M solution in methanol, titrated
against N/50 certified standard aqueous HCl solution), HClO4 (70%
aqueous solution, titrated to be 11.40 M), sodium hydride (60%
dispersion in mineral oil), phenol (99+%), 4-nitrophenol, 3-nitro-
phenol (99%), 4-chlorophenol (99+%), 2,4,5-trichlorophenol (98%),
methyl salicylate (>99%), methyl 5-chlorosalicylate (97%), methyl
4-iodosalicylate (97%), methyl 5-fluoro-2-benzoate (97%), methyl
5-methoxysalicylate (98%), 2-chloro-4-nitrophenol (97%), 3-meth-
yl-4-nitrophenol (98%), 4-chlorophenol (99+%), 3-methoxyphenol,
dimethyl chlorophosphate (96%), diethyl chlorophosphate (97%),
methyl 4-methoxysalicylate (98%), methyl 4-methylsalicylate (98%),
methyl 2-hydroxy-5-nitrobenzoate (98%), methyl 5-iodosalicylate
(99%), and methyl thiosalicylate (97%) were commercial and used
as supplied. Phosphate esters 3 and 4 were synthesized according
to a general method23 (see Supporting Information). Each of 3a-e
and 4a-i had 1H NMR, 31P NMR, and exact MS spectra consistent
with the structure (see Supporting Information).
f. Determination of Binding Constants. A general procedure
was used to determine the conditional binding constant, Kcond, for
s
phenol 4a and La(OTf)3 at pH 8.7 whereby a UV cuvette was
s
b. Methods. The sspKa values in methanol for the phenol leaving
groups of 3b-d and 4a-e,g-i were obtained from previous work.22
charged with 18.4 mM N-iPr-morpholine buffer containing 0.5 equiv
of HClO4 along with phenol 4a (6.3 × 10-4 M) in 2.5 mL of
methanol, and the spectrum from 250 to 400 nm was obtained. A
5 µL aliquot from a 55.5 mM La(OTf)3 stock solution was added
to the cell along with 1 equiv of NBu4OCH3 and another scan
obtained. This procedure was repeated until the observed ∆Abs360
did not change, thereby indicating saturation binding. The change
in total volume during titration of 3a was approximately 2%. The
Kcond was obtained by fitting of the ∆Abs360 nm vs [La(OTf)3] data
to eq 5 (see Discussion).
The sspKa of the phenol of 3a in methanol was determined from the
reported relationship pKMa eOH ) (1.08 ( 0.03)sspKwa ater + (3.50 (
s
s
0.20).22a The pKa of the phenol of 4f was determined by half-
s
s
+
neutralization with NaOMe in methanol, and the CH3OH2
concentrations were determined potentiometrically using a combi-
nation glass electrode (Fisher Scientific Accumet electrode model
13-620-183A) calibrated with certified standard aqueous buffers
(pH ) 4.00 and 10.00) as described previously. The sspH values in
methanol were determined by subtracting the correction constant
of -2.24 from the electrode readings as described.10
g. Job Plot. To a series of UV cuvettes containing anhydrous
methanol buffered with N-iPr-morpholine (18.4 mM, containing
c. General UV/Visible Kinetics. La3+-catalyzed reactions were
monitored by UV/vis spectrophotometry under buffered conditions
(18.4 mM N-iPr-morpholine containing 0.5 equiv of HClO4) at sspH
s
0.5 equiv of HClO4) at pH 8.7 were added La(OTf)3 and methyl
s
5-chlorosalicylate (phenol of 4c) such that [La]total + [phenol of
4c] was maintained constant at 2.28 mM. A UV/vis spectrum was
obtained on each sample scanning from 250 to 400 nm and the
Abs375 nm plotted as a function of [phenol of 4c]/([La]total + [phenol
of 4c]) (see Supporting Information).
8.7. Reactions were initiated by the addition of 5 × 10-5
M
phosphate ester into standard 1 cm path length UV cuvettes
containing the buffered solution of La(OTf)3 in methanol. Reaction
progress was monitored for 4a,c,d,h at 343 nm, 4b at 330 nm, 4e,g,i
at 340 nm, 4f at 370 nm, 3a at 303 nm, 3b at 268 nm, 3d at 286
nm, and 3e at 290 nm to obtain pseudo-first-order rate constants
(kobs) at each [La(OTf)3]. Since previous work10 has demonstrated
that La3+(-OCH3)x dimers comprise the bulk of the active species
at the [La3+] concentrations used, the gradients of the kobs versus
[La]total/2 plots were used to give the second-order rate constants
(kL2a). The sspH/rate profile for the La3+-catalyzed cleavage of 4a (4
× 10-5 M) was determined at constant [La]total ) 1.1 mM with
varying amounts of added NBu4OCH3 (0.25-3.1 mM), and the sspH
of the methanol solutions was measured following complete
reaction. The base-catalyzed reactions of substrates 3 and 4a,b,
d,e,f,h were initiated by the addition of 0.02-0.06 M NaOCH3 to
1 cm UV cuvettes containing 1 × 10-4 M phosphate ester, and the
reaction progress was monitored at the λmax for the phenolate
3. Results
1
a. Identification of Reaction Products. After 4 h, H NMR
analysis of the reaction mixture comprising 0.9 mM 4d, 4.0
mM La(OTf)3, and 4.0 mM NaOMe in CD3OD/CH3OH (93%
deuterium content) indicated that the only observable products
were trimethyl phosphate and methyl 4-iodosalicylate. Com-
parison of the aryl and the methoxycarbonyl proton intensities
indicated that there was ∼28% exchange of CO2CH3 for
CO2CD3, which is due to the fact that the La3+ system
promotes the transesterifications of carboxylate esters.24
Phosphate triesters 4a-i all underwent catalytic cleavage at
sspH 8.7 to produce the metal-bound aryl oxide complexes as
the only identifiable products (>90% as determined by UV/
vis spectrophotometry).
(22) (a) Neverov, A. A.; Liu, C. T.; Bunn, S. E.; Edwards, D.; White, C. J.;
Melnychuk, S. A.; Brown, R. S. J. Am. Chem. Soc. 2008, 130, 6639.
(b) Edwards, D. R.; Neverov, A. A.; Brown, R. S. J. Am. Chem. Soc.
2009, 131, 368.
(23) Padovani, M.; Williams, N. H.; Wyman, P. J. Phys. Org. Chem. 2004,
17, 472.
(24) (a) Neverov, A. A.; Sunderland, N. E.; Brown, R. S. Org. Biomol.
Chem. 2005, 3, 65. (b) Neverov, A. A.; McDonald, T.; Gibson, G.;
Brown, R. S. Can. J. Chem. 2001, 79, 1704.
9
13740 J. AM. CHEM. SOC. VOL. 131, NO. 38, 2009

La3+-Catalyzed Methanolysis of Dimethyl Aryl Phosphate Triesters
A R T I C L E S
Figure 1. Plot of [La3+2(-OCH3)x] speciation (left axis) and kobs (right axis)
s
vs pH for the catalyzed methanolysis of 4a (4.0 × 10-5 M) determined
s
from the rate of appearance of product metal-bound phenoxide at 343 nm,
T ) 25.0 ( 0.1 °C, and [La(OTf)3]total ) 1.1 mM.
Table 1. Kinetic Constants (k22:x, M-1 s-1) for the Methanolysis of
4a Catalyzed by the Various La3+2(-OCH3)x Species Present at
[La3+
]
) 1.1 mM and T ) 25.0 ( 0.1 °C
total
Figure 2. Plot of kobs vs [La]total/2 for the catalyzed methanolysis of 4a (5
× 10-5 M) determined from the rate of appearance of product metal-bound
phenoxide at 343 nm, T ) 25.0 ( 0.1 °C, sspH 8.7. Linear regression of the
data gives the second-order rate constants (kL2a ) 1.10 ( 0.01 M-1 s-1, r2
) 0.9995).
La3+2(-OCH3)x
x ) 1
x ) 2
x ) 3
x ) 4
x ) 5
a
k22:x
2.74 (0.81) 1.11 (0.21) 5.81 (0.84) 1.10 (0.22) 0.75 (0.20)
sspKLa a
b
7.56c
10.11c
9.73c
12.00c
°C. The Figure 1 speciation diagram indicates that, above 1 mM
a
s
k22:x determined by fit of the kobs vs pH data to eq 1 as described in
s
s
concentrations, all the metal ions exist as dimers at pH 8.7,
b
s
the text. sspKLa a not calculable on the basis of the data at hand.
with 92% being La3+2(-OCH3)2 and the La3+2(-OCH3)1 and
La3+2(-OCH3)3 species each accounting for about 4%. The
reaction of 4a, however, is readily shown to proceed 73%
via La3+2(-OCH3)2 and 12 and 15% respectively via the
La3+2(-OCH3)1 and La3+2(-OCH3)3 complexes once one consid-
ers the rate constants for the species given in Table 1. We
assume this is also the situation for each of 4b-i. The linear
kobs vs [La3+]total/2 plot shown in Figure 2 for 4a has an apparent
x-intercept of (1.10 ( 0.07) × 10-4 M. Nonlinear behavior in
the plots of kobs versus [La3+] total/2 has previously been observed
at low concentrations for the La3+-catalyzed methanolysis of
neutral phosphate triesters10 and N-acetylimidazole.26 These
results were interpreted as being consistent with incomplete
formation of the catalytically active La3+2(-OCH3)x dimeric
c
sspKLa a
for
the
hypothetical
La3+2(-OCH3)x(HOCH3)n
T
La3+2(-OCH3)x+1(HOCH3)n-1 + H+ ionization computed on the basis of
potentiometric titration data given in ref 25.
b. pH/Rate Profile for the La3+-Catalyzed Cleavage of
s
s
Phosphate Ester 4a. In Figure 1 is presented the La3+2(-OCH3)x/
sspH/rate profile for methanolysis of 4a (4 × 10-5 M) promoted
by La(OTf)3 (1.1 mM) as a function of varying amounts of
added NBu4OCH3. Figure 1 also includes the concentration
dependencies of the various La3+2(-OCH3)x species10a,25 as a
s
function of pH, which shows that the maximum rate of the
s
catalyzed reaction loosely tracks the speciation profile for
La3+2(-OCH3)3 in solution. Closer analysis reveals that the other
La3+2(-OCH3)x dimers also possess catalytic activity for the
s
complexes at [La3+
]
total/2 < 1.5 × 10-4 M, which is assumed to
cleavage of 4a. The pH/kobs data were analyzed as a linear
combination of the individual rate constants (eq 1),
s
be the case in the current study. Plots of kobs versus [La3+
]total/2
for all substrates 4b-i are similar in appearance to that shown
in Figure 2. The second-order rate constants (kL2a) were
determined as the slopes of their linear plots and are listed in
kobs ) k22:1[La3+2(-OCH3)1] + k22:2[La3+2(-OCH3)2] + ...+
k22:x[La3+2(-OCH3)x] (1)
Table 2.
s
Given in Figure 3 are the Brønsted plots of log kL2a vs pKalg
s
where k22:1, k22:2, ..., k22:x are the second-order rate constants for
the various La2(-OCH3)1,2,...,n complexes. The k22:x constants listed
in Table 1 indicate there is <10-fold difference in reactivity
between all five complexes, with La3+2(-OCH3)3 being the most
for the La3+-promoted cleavage of 4a-i and, for comparison
purposes, the La3+-catalyzed methanolysis of substrates 3a,b,
d,e,where the two slopes are ꢀ4lg) -0.82 and ꢀl3g ) -1.25. The
Brønsted plot of log k-2 OMe vs sspKlag for the cleavage of
4a,b,d,e,f,h fits a linear regression of log k-2 OMe ) (-0.51 (
reactive, having a computed k22:3 ) 5.81 ( 0.84 M-1 s-1. Also
s
given in Table 1 are the various pKLa a values for the formal
s
0.04)sspKlag + (4.68 ( 0.56). Not shown in Figure 3 for reasons
La3+2(-OCH3)x(HOCH3)n T La3+2(-OCH3)x+1(HOCH3)n-1
+
of visual clarity is the Brønsted plot of log k-2 OMe vs pKalg for
s
s
H+ ionization computed from data given in ref 25. A Brønsted
the methoxide-catalyzed cleavage of 3a-e, which fits a standard
linear regression of log k-2 OMe) (-0.59 ( 0.07)sspKlag + (5.41
( 0.91). The substituents on the aryl ring of 4 influence the
reactivity by interactions with both the o-CO2Me group and the
bridging phenoxy oxygen during the La3+-promoted P-OAr
bond cleavage. Attempts to correlate the kL2a for methanolysis
plot (Figure 14S, Supporting Information) of the log k22:x data
s
in Table 1 vs the pKLa a is fairly scattered, with a slope of ꢀnuc
s
) -0.02 ( 0.15.
c. Substituent Effects on the Rate of Reaction for the
La3+-Catalyzed Cleavage of Phosphate Esters 4a-i. The rates
of methanolyses of phosphate esters 4a-i were investigated as
a function of increasing [La(OTf)3] at sspH 8.7, T ) 25.0 ( 0.1
(26) Neverov, A. A.; Montoya-Pelaez, P.; Brown, R. S. J. Am. Chem. Soc.
2001, 123, 210.
(25) Gibson, G. T. T.; Neverov, A. A.; Brown, R. S. Can. J. Chem. 2003,
81, 495.
(27) Bunn, S. E.; Liu, C. L.; Lu, Z.-L.; Neverov, A. A.; Brown, R. S. J. Am.
Chem. Soc. 2007, 129, 16238.
9
J. AM. CHEM. SOC. VOL. 131, NO. 38, 2009 13741

A R T I C L E S
Edwards et al.
Table 2. Second-Order Rate Constants kL2a and k2-OMe Determined
for the Methoxide- and La(OTf)3-Catalyzed Cleavage of 4a-i and
3a-e
sistent results were obtained which depended upon the method
of fitting employed (see Supporting Information for a complete
discussion).
d. Determination of Binding Constants (Kb) for Phenoxide
Leaving Groups 4a,c-i. Binding constants for the parent phenols
a
b
-1
-1
substrate
sspKpahenol
kL2a (M-1s
)
k-2 OMe (M-1s
)
4a
4b
4c
4d
4e
4f
4g
4h
4i
3a
3b
3c
3d
3e
14.20
10.64
12.57
12.79
13.20
13.24
14.50
14.65
14.82
10.73
11.18
12.41
13.59
14.33
1.10 ( 0.01
(3.20 ( 0.02) × 10-3
(6.98 ( 0.34) × 102
(1.29 ( 0.05) × 101
(5.50 ( 0.16) × 101
2.23 ( 0.04
(1.69 ( 0.04) × 10-1
of phosphates 4a,c-i with La(OTf)3 were determined by
s
spectrophotometric titrations at pH 8.7. In Figure 4a is the
s
(1.79 ( 0.07) × 10-2
(5.91 ( 0.02) × 10-3
(1.18 ( 0.04) × 10-2
(1.39 ( 0.06) × 10-3
(1.30 ( 0.01) × 10-1
(4.48 ( 0.03) × 10-2
(2.58 ( 0.03) × 10-2
(3.12 ( 0.06) × 10-3
(6.29 ( 0.14) × 10-4
∆Abs375 nm vs [La3+] plot for the titration of methyl 5-chloro-
salicylate (parent phenol of 4c), where the calculated line is
obtained by fitting the data to eq 3,
(1.42 ( 0.08) × 101
1.16 ( 0.03
(1.21 ( 0.03) × 10-1
(5.01 ( 0.17) × 10-1
(9.25 ( 0.20) × 102
(1.28 ( 0.92) × 102
Absobs ) Absmax(1 + Kcond[Sub] + [La]Kcond
-
X)/(2Kcond)/[Sub] (3)
(2.50 ( 0.42) × 10-1
where
1.86 × 10-2
X ) (1 + 2Kcond[Sub] + 2[La]Kcond + Kcond2[Sub]2 -
2Kcond2[La][Sub] + [La]2Kcond
a Determined from linear regression of the plots of kobs vs [La3+
s
]
total/2
at pH 8.7, T ) 25.0 ( 0.1 °C. b Determined from linear regression of
s
2 0.5
)
the plots of kobs vs [NaOCH3] at T ) 25.0 ( 0.1 °C.
which generates a conditional binding constant, Kcond, of (5.0
( 0.9) × 104 M-1 that corresponds to the process shown in
Scheme 1. Shown in Figure 4b is the Abs302 nm versus [La3+
]
plot for the titration of 4-chlorophenol. Its Kcond is (4.4 ( 0.3)
× 102 M-1, indicating that this phenol, without the o-CO2Me
group, binds to the La3+ complex(es) much more weakly than
does methyl 5-chlorosalicylate. The binding constants, Kb, for
all species are calculated using eq 4, as reported in a related
study,22b using the Kcond values, the sspKlag values for each phenol,
and the [H+] corresponding to the sspH at which the spectroscopic
titrations were performed.
[H+]Kcond
Kb )
(4)
Ka
Table 3 gives the Kcond and Kb values determined for the
parent phenols of phosphates 4a,c-i. A Brønsted plot (Figure
5) of log Kb vs sspKalg of the phenol gives a regression of log Kb
) (0.85 ( 0.07)sspKalg - (2.1 ( 1). Binding constants were also
determined for phenols not possessing the o-CO2Me substituent
in order to quantify the stabilization afforded to the binding of
La3+ by that group (see Table 3, entries 9-12). These data fit
a linear regression of log Kb ) (0.49 ( 0.06)sspKalg - (0.71 (
Figure 3. Brønsted plots for the La3+-catalyzed methanolysis of phosphate
triesters 4 (9, solid line) determined at sspH 8.7, T ) 25.0 ( 0.1 °C, where
log kL2a ) (-0.82 ( 0.11)ss pKlag + (11.61 ( 1.48), r2 ) 0.8878, nine data;
the La3+-catalyzed methanolysis of substrates 3 (O, solid line), where log
kL2a ) (-1.25 ( 0.06)ss pKalg + (16.23 ( 0.75), r2 ) 0.9954, four data; and
the methoxide-catalyzed methanolysis of phosphate triesters 4 (2, dashed
line), where log k-2 OMe ) (-0.51 ( 0.04)ss pKlag + (4.68 ( 0.56), r2 ) 0.9730,
six data. For visual clarity, the Brønsted plot for the methoxide-catalyzed
methanolysis of phosphate triesters 3 is not shown; however, a fit of the
data provided in Table 2 gives log k-2 OMe ) (-0.59 ( 0.07)ss pKalg + (5.41
( 0.91), r2 ) 0.9557, five data.
0.74).
nm
A Job plot of Abs375
versus [phenol of 4c]/([La3+
]
+
total
[phenol of 4c]) (total concentration ) 2.28 mM) is shown in
the Supporting Information, Figure S13. The plot has a plateau
maximum from 0.5 to 0.66, showing that at these concentrations
the La3+2 dimer system binds 2-3 molecules of the phenoxide.
4. Discussion
s
a. pH vs kL2 a Profile. The speciation/rate constant plot in
s
of 4a-i using a single substituent parameter as in a standard
Brønsted or Hammett plot resulted in the apparent scattering
of the data depicted in Figure 3. In addition, when the rate
constants pertaining to cleavage of 4a-i were fit to a two-
parameter model using the Jaffe´ equation,
Figure 1 for the cleavage of 4a is reminiscent of that for La3+
-
catalyzed methanolysis of paraoxon,10a where there is also a
non-integer slope in the log kL2a vs pH plot. This sort of plot
s
s
for metal-methoxide-catalyzed reactions results from the ef-
fectiveness of catalysis of the different metal-methoxy forms,
the speciation and concentrations of which change with sspH.28
All the La3+2(-OCH3)x (x ) 1-5) species formed over the sspH
range of this study show some catalytic activity, with the
log(kx2/kH2 ) ) F
σ
+ F
σ
C(O)OMe
(2)
phosphate
C(O)OMe
phosphate
where σphosphate and σC(O)OMe refer to the sigma substituent
constants as they relate to the bridging oxygen and o-CO2Me
substituents of the aryloxy leaving group, respectively, incon-
(28) Gibson, G. T. T.; Neverov, A. A.; Teng, A. C.-T.; Brown, R. S. Can.
J. Chem. 2005, 83, 1268.
9
13742 J. AM. CHEM. SOC. VOL. 131, NO. 38, 2009

La3+-Catalyzed Methanolysis of Dimethyl Aryl Phosphate Triesters
A R T I C L E S
Figure 4. (a) Plot of ∆Abs375 nm versus [La(OTf)3] for the spectrophotometric titration of methyl 5-chlorosalicylate at sspH 8.7 and 25.0 ( 0.1 °C; fit of the
data to eq 3 gives Kcond ) (5.0 ( 0.9) × 104 M-1. (b) Plot of ∆Abs302 nm versus [La(OTf)3] for the spectrophotometric titration of 4-chlorophenol at sspH 8.7
and 25.0 ( 0.1 °C; fit of the data to eq 4 gives Kcond ) (4.4 ( 0.3) × 102 M-1
.
Scheme 1. Thermodynamic Cycle for the La(OTf)3 Binding of the
Phenol Corresponding to 4a
Table 3. Kcond and Kb Binding Constants for the Complexation of
Phenols Corresponding to Phosphates 4a,c-i with La3+, Determined
by Spectrophotometric Titration at sspH 8.7 and T ) 25 °C
phenol corresponding to
parent phosphate
s
Figure 5. Plot of log Kb vs pKpahenol for the parent phenols of 4a,c-i (9)
s
b
Kcond (×104 M-1
)
Kbc (M-1
)
which fit a linear regression of log Kb ) (0.85 ( 0.07)sspKaphenol - (2.1 (
1), r2 ) 0.959, eight data, and for phenols lacking the o-CO2Me substituent
(4) which fit a linear regression of log Kb ) (0.49 ( 0.06)sspKaphenol - (0.71
( 0.74), r2 ) 0.970, four data. (b) Binding constant for methoxide
(sspKamethanol ) 18.16) coordination to La2(-OMe)1 forming the 2:2 complex
of La2(-OMe)2.
4a
4c
4d
4e
4f
4g
4h
4i
(2.6 ( 0.6) × 104
(5.0 ( 0.9) × 104
(6.6 ( 1.3) × 104
(4.1 ( 1.7) × 104
(8.6 ( 1.1) × 104
(2.2 ( 0.4) × 104
(4.4 ( 0.8) × 104
(3.6 ( 0.5) × 104
(3.6 ( 0.1) × 104
(3.8 ( 0.2) × 103
(4.4 ( 0.3) × 102
(1.1 ( 0.1) × 102
7.4 × 109
3.7 × 108
7.5 × 108
1.2 × 109
2.7 × 109
1.3 × 1010
3.9 × 1010
4.3 × 1010
1.6 × 105
2.4 × 106
3.4 × 107
1.9 × 107
the cleavage mechanism was claimed to be a general base one
in contrast to the more complicated nucleophilic one believed
operative here with the La3+ dimers.30 More pertinent to the
discussion is the finding that a homologous series of five
mononuclear Zn(II) triamine complexes catalyze the hydrolysis
of diethyl 2,4-dinitrophenyl phosphate with a ꢀnuc of -0.15 (
0.01 by a proposed nucleophilic mechanism involving Lewis
acid activation of the substrate and intramolecular attack of the
coordinated hydroxide.31 The low value of the ꢀnuc found in
the current study is reasonably suggested to result from a
complex combination of competing effects of increased nu-
2-chloro-4-nitrophenold
3-methyl-4-nitrophenold
4-chlorophenold
3-methoxyphenold
a
Protocols for the determination of pKpahenol in methanol described in
s
s
the Experimental Section. b Determined by fitting ∆Abs versus
[La(OTf)3] data to eq 4. c Calculated with eq 5 as described in the text.
d
s
The pKaphenol data can be found in refs 22a and 27.
s
La3+2(-OCH3)3 complex having the largest rate constant of k22:3
) 5.8 ( 0.8 M-1 s-1. However, the k2 values are relatively
2:x
cleophilicity for the methoxide attached to the species with the
highest pKa values, coupled with a concomitant decreasing
s
s
s
insensitive to the large variation in the pKa values for the
s
La3+2(-OCH3)x species, as is revealed in the Brønsted plot,
where the ꢀnuc ) -0.02. Low values for ꢀnuc are often observed
for metal-hydoxo complexes attacking CdO-containing spe-
cies.29 It was reported that the reaction of some mononuclear
metal-hydroxo complexes exhibited a ꢀnuc of 0.43 ( 0.13 in
their reaction with diethyl 2,4-dinitrophenyl phosphate, although
Lewis acidity/electrophilic activation and LGA attributable to
the reduced net positive charge on the two La3+ ions in the
complex.
b. Binding of Aryl Oxides to La3+. The general appearance
of the representative Abs vs [La3+] plots in Figure 4a,b
(30) Hay, R. W.; Govan, N. Trans. Met. Chem. 1998, 23, 133.
(31) Itoh, T.; Tada, T.; Yoshikawa, Y.; Hisada, H. Bull. Chem. Soc. Jpn.
1996, 1265.
(29) Martin, R. B. J. Inorg. Nucl. Chem. 1976, 38, 511.
9
J. AM. CHEM. SOC. VOL. 131, NO. 38, 2009 13743

A R T I C L E S
Edwards et al.
determined for the parent phenols of 3d and 4c indicates the
o-CO2Me substituent provides enhanced binding to the metal
complex relative to phenols without this group. The appearance
of the spectrophotometric titration plots with a slight x-intercept
and steep slope is consistent with a process wherein the aryl
oxides corresponding to phosphates 4a,c-i bind to one or more
dimeric La3+ forms. The Job plot (Supporting Information)
further indicates that 2-3 aryl oxides can be accommodated at
equilibrium within the coordination sphere of the dimeric
La3+2(-OCH3)x complex to give a bound form of stoichiometry
La3+2(-OAr)2-3(-OCH3)x.
The k-2 OMe values for the methoxide-catalyzed cleavage of 4
listed in Table 2 differ by only ∼102. The Brønsted plot of these
data fits a linear regression of log k-2 OMe ) (-0.51 ( 0.04)sspKlag
+ (4.68 ( 0.56), whereas that for the methoxide reaction of
phosphates 3 is log k-2 OMe ) (-0.59 ( 0.07)sspKlag + (5.41 (
0.91). Given the error limits, the two ꢀlg values of -0.51 and
-0.59 are close enough to suggest that the extent of P-OAr
bond rupture in the TSs for the two series is essentially identical.
The amount of bond rupture is calculated to be 27-32% based
on the Leffler parameter,32 R ) ꢀlg/ꢀeq, where the ꢀeq of -1.87
pertains to the equilibrium transfer of the diethoxyphosphoryl
group between oxyanion nucleophiles in water,33 which we
assume is similar to the situation in methanol.
d. Mechanistic Possibilities for the Catalyzed Reaction. The
mechanism of the La3+-catalyzed10c methanolysis and oxyanion-
promoted cleavage15,33 of phosphate triesters has been discussed
considering stepwise or concerted reactions. For acyclic phos-
photriesters with aryloxy leaving groups, linear free energy data
support concerted reactions for oxyanion nucleophiles, with the
TSs being looser for leaving groups with low pKa’s and tighter
with poorer leaving groups.33 Stepwise mechanisms for the
reaction of phosphate esters have also been proposed and
demonstrated convincingly in some instances, generally with
six-membered cyclic phosphates.34
The metal-bound aryl oxide complexes are the only observ-
able aryloxy products from the catalytic cleavage of 4a-i at
sspH 8.7 (>90% by UV/vis spectrophotometry). At the sspH where
these reactions were run, the complete binding of the aryl oxide
s
anion indicates that the pKa for the parent phenol of 4a is
s
reduced by at least 6.5 sspH units by complexation to
La3+2(-OCH3)x (Scheme 1). At sspH 8.7 and 1.1 mM La(OTf)3,
the speciation diagram of Figure 1 indicates that all the metal
ions exist as dimers, with 92% being La3+2(-OCH3)2, while
La3+2(-OCH3)1 and La3+2(-OCH3)3 account for about 4% each.
The gross binding constants (Kb) for the aryl oxides correspond-
ing to 4a,c-i with the La3+-containing species present at that
sspH are listed in Table 3: the values are large and range from
4 × 108 to 4 × 1010 M-1
.
For metal-ion-catalyzed cleavage of phosphate triesters, there
is little literature guidance about concerted or stepwise cleavage,
although we have previously favored concerted processes for
methanolyses of diethyl aryl phosphates promoted by both La3+
and a Zn(II) complex, primarily because of the large degree of
P-OAr cleavage (ꢀlg ) -1.43 and -1.12, respectively).10c At
one limit, the La3+-catalyzed methanolysis of phosphate esters
4a-i might follow a stepwise reaction mechanism (AN + DN)
as in Scheme 2, although the very large ꢀlg found with substrates
3 without a o-CO2Me group might require that the rate-limiting
step is breakdown of the intermediate.35,36 The Brønsted ꢀlg of
-0.82 for the o-CO2Me-containing phosphates 4 could also be
consistent with this process. A rate-limiting breakdown of the
intermediate would require a fast and reversible addition of the
nucleophile to 4 that might be detected by an isotopic exchange
experiment conducted in CD3OD. However, no incorporation
The slope of 0.85 for the plot of log Kb versus the sspKa values
of the phenols with the o-CO2Me groups (Figure 5) is
experimentally identical to that found for Yb3+ complexation
of these same phenoxides in methanol (0.84).22b This value
indicates that a formal charge of approximately -0.15 resides
on the oxyanion in the metal-bound complex and suggests that
85% of the negative charge is acquired by the metal complex
on binding the phenoxide to the La3+2(-OCH3)x. Aryl oxides
without the o-CO2Me substituent also bind to the La3+2(-OCH3)x
species, albeit with significantly smaller Kb constants (Table 3,
s
entries 9-12). The log Kb vs pKa plot in Figure 5 has a slope
s
of 0.49, indicating that these phenoxides experience a lesser
degree of charge neutralization upon La3+ complexation and
so have a formal charge of -0.5 residing on the coordinated
aryl oxides. An estimation of the enhanced stability afforded
the La3+-aryl oxide complexes by the o-CO2Me group is
possible based on an extrapolation of the lower line of Figure
5 to a sspKaof 14.20, which corresponds to a phenol having sspKa
identical to that of 2-(methoxycarbonyl)phenol but lacking the
ortho-substitution. The extrapolated Kb of 4.7 × 107 M-1
indicates that the o-CO2Me group provides an additional
stabilization of ∼3 kcal/mol.
of a CD3O- group into unreacted 4 was detected for the La3+
-
catalyzed reaction of 4d in methanol-d4, even though the
exclusive P-containing product is CD3O-P(O)(OCH3)2. This
experiment is suggestive, but the exchange via a phosphorane
intermediate must involve one OCH3 group at an apical position
(32) (a) Leffler, J. E.; Grunwald, E. Rates and equilibria of organic
reactions; Wiley: New York, 1963. (b) Williams, A. Acc. Chem. Res.
1984, 17, 425.
c. Linear Free Energy Treatment of the Rate Data. The k2La
values for La3+-catalyzed cleavage of 4a-i listed in Table 2
vary by nearly a factor of 6000. These give a rather scattered
Brønsted plot (Figure 3) with a best fit of log kL2a ) (-0.82 (
0.11)sspKalg + (11.31 ( 1.48). The ꢀlg of -0.82 is less than that
determined for the same sort of plot for the La3+-catalyzed
cleavage of phosphates 3 (ꢀlg ) -1.25). While at first glance
this may suggest that P-OAr cleavage in the reaction of 4a-i
is less extensive than with substrates 3, other evidence (Vide
infra) indicates that cleavage is well advanced with 4a-i, and
the less negative ꢀlg value is consistent with La23+(-OCH3)x
assistance of leaving group departure. In our previous study10c
on the La3+-catalyzed methanolysis of diethyl aryl phosphate
esters, a ꢀlg of -1.43 was determined which, in consideration
of the associated error limits, is experimentally identical to the
ꢀlg of -1.25 determined here for substrates 3.
(33) (a) Ba-Saif, S. A.; Williams, A. J. Org. Chem. 1988, 53, 2204. (b)
Ba-Saif, S. A.; Waring, M. A.; Williams, A. J. Am. Chem. Soc. 1990,
112, 8115. (c) Ba-Saif, S. A.; Waring, M. A.; Williams, A. J. Chem.
Soc. Perkin Trans 2 1991, 1653.
(34) (a) Bromilow, R. H.; Khan, S. A.; Kirby, A. J. J. Chem. Soc., Perkin
Trans. 2 1972, 911. (b) Assad, N.; Kirby, A. J. J. Chem. Soc., Perkin
Trans. 2 2002, 1708. (c) Hall, C. R.; Inch, T. D. Tetrahedron 1980,
36, 2059. (d) Khan, S. A.; Kirby, A. J. J. Chem. Soc. B 1970, 1172.
(e) Rowell, R.; Gorenstein, D. G. J. Am. Chem. Soc. 1981, 103, 5894.
(35) An analogous situation arises in the di-Co(III) complex promoted
cleavage of some methyl aryl phosphates in water, where a very large
ꢀlg of-1.38 for intramolecular attack of a di-Co(III)-coordinated oxide
on the P was originally interpreted as arising from a concerted process
with a late transition state,36a but later shown that the more likely
pathway had two steps with rate-limiting breakdown of the
intermediate.36b
(36) (a) Williams, N. H.; Cheung, W.; Chin, J. J. Am. Chem. Soc. 1998,
120, 8079. (b) Humphrey, T.; Forconi, M.; Williams, N. H.; Hengge,
A. C. J. Am. Chem. Soc. 2004, 126, 11864.
9
13744 J. AM. CHEM. SOC. VOL. 131, NO. 38, 2009

La3+-Catalyzed Methanolysis of Dimethyl Aryl Phosphate Triesters
A R T I C L E S
Scheme 2. Hypothetical Stepwise AN + DN and Concerted ANDN Mechanisms for La3+2(-OCH3)2-Catalyzed Cleavage of 3 and 4
Scheme 3. Charge Map for the La3+2(-OCH3)x-Catalyzed
Methanolysis of 4a
during its formation or, alternatively, pseudorotation15,37 of the
first-formed phosphorane and translocation of La3+ to provide
assistance of OCH3 departure. The pseudorotation, if required,
could be too slow to be permitted by a short-lived intermediate.38
A concerted (ANDN) reaction mechanism for the La3+
-
catalyzed cleavage of substrates 3 and 4 is also depicted in
Scheme 2 as the lower pathway and involves electrophilic
activation of the ester through PdO· · ·La3+ coordination,
delivery of a metal-coordinated methoxide, and extensive
departure of the leaving group. In an earlier study10c on the
cleavage of diethyl aryl phosphate esters, we suggested that it
was unlikely that an alkoxide (or hydroxide) bound between
two metal ions would be sufficiently nucleophilic to attack the
PdO unit, so that the La3+2(-OCH3)2 dimer opens to give a
terminally coordinated methoxide along with the PdO unit being
bound to the second La3+, such as in TS 5a. The only difference
between that TS and the hypothetical TS 5 shown in Scheme 2
involves a slight movement of the metal ions to provide double
coordination of the OdP during the concerted process. With
mol of stabilization for binding the La3+2(-OCH3)x to the parent
phenoxide of 4a relative to a phenol of identical sspKa but lacking
the o-CO2Me substituent. (2) Substrates bearing the o-CO2Me
substituent undergo La3+-catalyzed cleavage at a significantly
higher rate than substrates not having that group. In the case of
substrate 4a, this amounts to a 60-fold acceleration relative to
3e, which is similar to the acceleration seen with diesters cleaved
by the di-Zn(II) complex in methanol22a but not nearly as large
as the 1012 seen for the Yb3+-promoted reactions of diesters.22b
Interestingly, the additional catalysis achieved with substrates
4 relative to 3 is more efficient for poorer leaving groups, as
can be clearly seen upon visual inspection of the Brønsted plots
(Figure 3). The best leaving groups from each series, 4a and
phosphates 3 the departing group is not coordinated to La3+
,
but there could be a strong electrostatic interaction that does
not involve extensive charge transfer from -OAr to the La3+
but is consistent with the large observed ꢀlg. The reduced ꢀlg
observed with 4 might be explained in terms of TS 5b, where
the net charge buildup on the departing group is reduced by
some coordinative association with La3+. Additional evidence
for leaving group assistance in the La3+-catalyzed cleavage of
phosphate esters 4a-i is presented in the following section.
e. Catalysis Involving Leaving Group Assistance. Aspects of
the current study with 4a-i resemble those observed for the
Yb3+-catalyzed reaction of phosphate diesters with the same
series of leaving groups22b as well as with a di-Zn(II) complex-
catalyzed reaction of some methyl aryl phosphates.22a There
are three main similarities: (1) The reaction products for the
La3+-catalyzed cleavage of 4a-i are the metal-bound aryl
oxides, as is the case in the Yb3+ and dinuclear Zn(II) complex-
catalyzed reactions. The linear regressions of the log Kb vs
sspKpahenol plots shown in Figure 5 point to an additional ∼3 kcal/
3a, essentially lie at the point of intersection of the two Brønsted
s
plots, which suggests that with leaving groups of pKpahenol less
s
than ∼10.7, catalysis involving leaving group assistance is no
longer a requirement or even beneficial. (3) The Brønsted ꢀlg
observed for cleavage of 4 in this study (-0.82) is shallower
than the -1.25 determined for the analogous reaction with
substrates 3. This feature is also noted for the di-Zn(II) complex-
catalyzed cleavage of phosphate diesters having the o-CO2Me
group.22a We can construct the charge map shown in Scheme 3
for the reaction of 4a with La3+2(-OCH3)x, assuming that the
starting P-OAr bridging oxygen has a charge of +0.87.33
Insofar as the aqueous values are relevant to the situation in
methanol, the effective charge residing on the La3+2(-OCH3)x-
-
bound phenoxide is calculated as ECArOLa ) ECArO + ꢀbind
)
-1.0 + (0.85) ) -0.15, where the value of 0.85 comes from
the plot in Figure 5, log Kb ) (0.85 ( 0.07)sspKa - (2.1 ( 1).
The nascent cleaved phenoxy species cannot be exactly the same
as what is formed in the equilibrium binding experiments of
phenol plus La3+2(-OCH3)x used to determine the Kb values for
Figure 5. Thus, the computed ECArOLa value of -0.15 only
provides a lower estimate for the ꢀeq of -1.02 and an upper
estimate for the charge of 0.05 ( 0.11 on the leaving group in
(37) (a) Westheimer, F. H. Acc. Chem. Res. 1968, 1, 70. (b) Thatcher, G. R.;
Kluger, R. AdV. Phys. Org. Chem. 1989, 25, 99.
(38) The requirement that departure of the aryloxy leaving group is rate
limiting at first seems odd, given that its departure would be slower
s
than that of the methoxide nucleophile, even though the pKavalues
s
for all the ArylOH groups of 4 are considerably lower than that of
methanol (18.2). However, the metal-catalyzed reaction promoted by
s
La3+2(-OCH3)2 proceeds via a La3+-coordinated methoxide, the pKa
s
of which is 7.56 (Table 1), meaning that the metal-coordinated
methoxide is actually the better leaving group.
9
J. AM. CHEM. SOC. VOL. 131, NO. 38, 2009 13745

A R T I C L E S
Edwards et al.
Table 4. Activation Parameters for the Lyoxide- and
La3+-Promoted Cleavage of Dimethyl Paraoxon and 4a
reaction
∆Hq (kcal/mol)
∆Sq (cal/mol·K)
∆Gq (kcal/mol)
25
DMP + -OMe
DMP + -OHa
14.7 ( 0.7
11.6
11.6
0.70 ( 0.05
15.6 ( 0.3
3.3 ( 0.1
-16.0 ( 2
-26
19.5
19.3
21.1
14.6
20.8
17.3
DMP + -OH (calc)b
-31.9
c
(DMP + La3+
)
-46.5 ( 0.2
-17.4 ( 0.95
-47.0 ( 0.4
4a + -OMe
d
(4a + La3+
)
a Reference 40. b Reference 42. c Determined at [La(OTf)3]total ) 2.0
mM, pH) 9.1, ref 39. d Determined at [La(OTf)3]total ) 1.5 mM, pH )
s
s
s
s
8.7.
the TS. Although an estimate, the calculation indicates that the
Leffler index (R) of 0.80 ( 0.10 for the cleavage of the P-OAr
bond in the methanolyses of 4 is experimentally indistinguish-
able from that found for the La3+-catalyzed cleavages of 3,
where R ) ꢀlg/ꢀeq ) -1.25 ( 0.06/1.87 ) 0.67 ( 0.06. Thus,
the decreased ꢀlg determined for the reaction of substrates 4 is
not a result of less P-OAr bond rupture but rather a manifesta-
tion of the reduction in total charge development at the bridging
oxygen as a result of metal-ion-mediated leaving group assistance.
Figure6. Activationenergydiagramforthemethoxideandthe[La2(-OCH3)2]-
catalyzed cleavage of substrate 4a at standard state 1 M and 25 °C, showing
the calculated energy of binding methoxide by La2(-OCH3) and the
activation energies associated with k22:2 and k-2 OMe. Free energies of activation
are calculated as ∆Gq ) -RT ln(k2/(kT/h)) from the Eyring equation, where
(kT/h) ) 6 × 1012 s-1 at 298 K.
f. Activation
Parameters
for
the
La3+
-
and
-OCH3-Promoted Cleavage of Some Phosphate Triesters. The
activation parameters for the -OCH3 -and La3+-promoted
reactions of dimethyl paraoxon ((CH3O)2P(dO)-OC6H4NO2,
DMP)39 and 4a are given in Table 4. The data for the methoxide
reaction of DMP are similar to the activation parameters
obtained experimentally for the hydroxide-promoted reaction
in water,40,41 and a recent report gives computed values for the
hydroxide reaction of 11.6 kcal/mol, -31.9 cal/mol ·K, and 21.1
kcal/mol for the ∆Hq, ∆Sq, and ∆Gq for which the TS was
25
said to comprise an in-line concerted process with little cleavage
of the departing p-nitrophenoxy group.42 The ꢀlg value of -0.51
for the -OCH3 reaction of series 4 is also consistent with a
concerted reaction where the Leffler parameter indicates ∼30%
P-OAr bond cleavage in the TS.
Figure 7. Thermodynamic cycle depicting the formation of [La2(OMe)]
and [La2(OMe)2] from free 2La3+ and 2(-OMe). The stability constants
K2:2 and K2:1 were determined by fits of potentiometric data using the
Hyperquad program as described.25 Charges and associated counterions are
omitted for clarity.
The La3+-catalyzed methanolyses for DMP and 4a have very
low ∆Hq values that are compensated by large negative ∆Sq
terms, as might be expected for a TS where there is considerable
restriction of the degrees of freedom of the reaction partners.
The low ∆Hq values mean the catalytic reactions are particularly
insensitive to temperature changes so that the phosphate
cleavage reactions will proceed rapidly at low temperatures. The
large ∆Sq terms dominate the free energy calculations, with the
computed values for ∆Gq25 being about 4 and 3 kcal/mol lower
than the ∆Gq25 for the corresponding methoxide reactions. Since
there is no experimental difference between the ∆Sq terms for
DMP and 4a, one cannot say that the TS for the o-CO2Me-
bearing reactant is significantly more restricted than that for
DMP.
free energy diagram shown in Figure 6 shows the ∆Gq of
activation for the methoxide reaction of 4a and that of its
La2(-OCH3)2-catalyzed cleavage at the standard state of 1 M,
T ) 25 °C. In eq 5
∆∆Gqstab ) (∆Gbind + ∆Gcqat) - ∆G-q OR
)
(k22:2)(K-OR
)
-RT ln
(5)
k-2 OR
[
]
is the expression for determining the free energy of binding of
the catalyst to the methoxide plus substrate in the TS, ∆∆Gsqtab
.
The ∆Gq-ORfor the methoxide reaction was calculated to be 20.8
kcal/mol from the Eyring equation (eq 6) using k-2 OMe ) 3.20
× 10-3 M-1 s-1 (Table 1).
g. Catalytic Rate Acceleration and Transition State
Effects. The efficacy of catalyst-promoted reactions can be
evaluated by comparing the free energy of binding of the catalyst
to the TS of the presumed lyoxide-promoted reaction.43,44 The
∆Gq-OR)-RT ln(k-2 OR/(kT/h))
(6)
The K-OR binding constant of -OCH3 to the catalyst (here taken
to be La2(-OCH3)) is calculated in accordance with the
thermodynamic cycle in Figure 7, giving K-OMe ) 1020.86/1011.66
) 109.2 M-1, from which the free energy of binding is
determined to be ∆Gbind ) -RT ln(K-OMe) ) -12.5 kcal/mol.
(39) Liu, C. T.; Brown, R. S. Unpublished results.
(40) Ginjaar, L.; Vel, S. Recl. TraV. Chim. 1958, 77, 956.
(41) The reported activation parameters for the hydroxide-promoted reaction
of paraoxon are ∆Hq ) 12.69 ( 0.03 kcal/mol and ∆Sq )-24.81 (
0.05 cal/mol ·K (∆Gq ) 20 ( 0.3 kcal/mol): Purcell, J.; Hengge,
25
A. C. J. Org. Chem. 2005, 70, 8437.
(42) Iche´-Tarrat, N.; Barthelat, J.-C.; Rinaldi, D.; Vigroux, A. J. Phys.
Chem. B 2005, 109, 22570.
(44) For applications of this to phosphate cleavage and other reactions,
see: Yatsimirsky, A. K. Coord. Chem. ReV. 2005, 249, 1997, and
references therein.
(43) Wolfenden, R. Nature 1969, 223, 704.
9
13746 J. AM. CHEM. SOC. VOL. 131, NO. 38, 2009

La3+-Catalyzed Methanolysis of Dimethyl Aryl Phosphate Triesters
A R T I C L E S
Table 5. Computed Free Energies of Activation for the
[La2(OCH3)x]-Catalyzed Decomposition of 4a (∆GqLa), Free Energy
of Binding Methoxide to the Catalyst Precursor [La2(OCH3)x-1
]
(∆Gbind), Free Energy of Stabilization Afforded the Methoxide TS
by Binding to the Active Catalyst [La2(OCH3)x] (∆∆Gqstab), and
sspKNa uc for the Formal La2(OCH3)x-1(HOCH3) T La2(OCH3)x +
H2O+CH3 Ionization Processa
∆Gbind (kcal/mol) ∆GqLa (kcal/mol) ∆∆Gqstab (kcal/mol)
sspKNauc
[La2(OCH3)2]
[La2(OCH3)3]
[La2(OCH3)4]
[La2(OCH3)5]
12.54
9.08
9.59
6.49
17.35
16.37
17.36
17.58
15.99
13.51
13.03
9.71
7.56
10.11
9.73
12.00
a Stability constants and sspKaNuc data for the various La2(OCH3)x
complexes are found in ref 25.
Figure 9. A reaction coordinate diagram for the cleavage of phosphate
esters 4 depicting the trajectory taken by the reacting substrate through the
potential energy surface in the presence (blue line) and absence (red line)
of the La3+2(-OMe) catalyst.
La3+2(-OCH3)x parent in the TS is offset by a concomitant
decrease in the attractive energy between the transforming
substrate and La3+2(-OCH3)x-1
.
h. More O’Ferrall-Jencks Diagram45 for the Catalyzed
and Uncatalyzed Reactions. In accordance with linear free
energy15,33,46 and computational42,47 data that indicate that
oxyanion nucleophiles react via concerted processes with acyclic
triesters having aryloxy leaving groups, we favor a concerted
reaction for the reaction of methoxide with phosphate esters 3
and 4 in methanol. The noncatalyzed TS position (‡non) for this
reaction with 4 (and 3) is biased toward the phosphorane corner
of Figure 9, in agreement with the small Leffler parameter (R
) 0.27-0.32) and the general consensus that neutral triesters
react by tighter, associative TSs with the phosphoryl center
accepting negative charge.15 In the diagram, the catalyst
(formally La3+2(OCH3)1 in blue) is positioned suggestively
above the methoxide and phosphate reaction components (in
red) at each of the four corners to illustrate the proposed binding
of the catalyst. Both starting material and product are neutral
phosphate triesters and thus are not bound, to any great extent,
Figure 8. Plot of ∆∆Gsqtab versus sspKNa uc of a molecule of methanol
coordinated to the catalyst which fits a standard linear regression of ∆∆Gqstab
) (-1.39 ( 0.21)sspKaNuc + (26.7 ( 2.1) kcal/mol, r2 ) 0.9554, four data.
The free energy of activation for La3+-catalyzed cleavage of
4a is ∆Gqcat) 17.4 kcal/mol using the Eyring equation and the
second-order rate constant k22:2 ) 1.11 ( 0.21 M-1 s-1 (Table
1). It can be seen from Figure 6 that the binding of
La3+2(-OCH3) to the TS comprising 4a· · ·-OCH3 is ∆∆Gqstab
) 16 kcal/mol, which exceeds the stabilization afforded by
binding methoxide in the ground state by ∼3 kcal/mol.
by any of the La3+ species. Methoxide is bound to
2
La3+2(OCH3)1 with ∆Gbind ) 12.5 kcal/mol, but this is expected
to be partially counterbalanced by La3+2(OCH3)1 binding to the
aryl oxide in the product state. An exact ∆Gbind for the latter is
not available but should be intermediate between an upper limit
of 13.4 kcal/mol, as computed from the Kb for 4a by spectro-
photometric titration, and a lower limit of 10.5 kcal/mol, arrived
Using the above procedure ∆Gbind, ∆Gqcat, and ∆∆Gsqtab were
calculated for each of the catalytic [La2(OCH3)x] complexes,
where x ) 2-5, Table 5. It is notable that the Brønsted plot of
2:x
the k2 constants in Table 1 gives ꢀnuc ) -0.02 ( 0.15,
indicating that the anticipated increase in rate constant with
nucleophile sspKa is not evident for this reaction. This is probably
due to a counterbalancing electrophilic activation of the substrate
coupled to the nucleophilic delivery of a La3+2-bound methoxide.
The plot shown in Figure 8 of ∆∆Gqstab vs sspKaNuc for the last
methoxide associated with the La3+2(-OCH3)x-1 catalyst fits a
regression of ∆∆Gqstab ) (-1.39 ( 0.21)sspKNa uc + (26.7 ( 2.1)
kcal/mol, indicating that the catalytic proficiency of the system,
relative to the methoxide reaction, increases by some 1.39 kcal/
at by extrapolation of the linear regression (Figure 5) for regular
s
phenols to a pKa of 14.2, corresponding to a substrate of
s
identical sspKa to that of 4a but lacking the o-CO2Me substituent.
These considerations indicate that the catalyst should have a
minimal effect on the TS position along the reaction coordinate
in the classic Hammond sense.
However, the fact that the ꢀlg values for the La3+-catalyzed
cleavages of esters 3 and 4 are larger and more negative than
the respective methoxide reactions indicates that the catalyzed
TS (‡cat) has moved in the anti-Hammond sense toward the
upper left corner, with considerable development of negative
mol per unit decrease in sspKNa uc. The data in Table 5 show that
s
the ∆Gbind values increase regularly with pKNa uc as expected,
s
but there is only a small variation in the ∆GqLa. Taken together,
these suggest that the changes in ∆Gbind of methoxide for a given
La3+2(-OCH3)x complex are exactly compensated by changes
to the ∆∆G for binding of the entire TS by that complex, so
the (∆∆Gqstab - ∆Gbind) is essentially constant at ∼3.4 kcal/mol.
Put another way, whatever reduction in attractive energy occurs
in loosening the association of the attacking methoxide and its
(45) (a) More O’Ferrall, R. A. J. Chem. Soc. B 1970, 274. (b) Jencks, W. P.
Chem. ReV. 1972, 72, 705.
(46) Lewis, V. E.; Donarski, W. J.; Wild, J. R.; Raushel, F. M. Biochemistry
1988, 27, 1591.
(47) Zheng, F.; Zhan, C.-G.; Ornstein, R. L. J. Chem. Soc., Perkin Trans.
2 2001, 2355.
9
J. AM. CHEM. SOC. VOL. 131, NO. 38, 2009 13747

A R T I C L E S
Edwards et al.
charge on the leaving group, implying a far looser TS.
Additionally, the ꢀnuc of -0.02 ( 0.15 for the data in Table 1
and the TS stabilization (∆∆Gqstab in Figure 6) of 16 kcal/mol,
which exceeds the ∆Gbind of methoxide to La2(-OMe)1 by over
3 kcal/mol, demonstrate that the role of the catalyst in promoting
the cleavage of phosphate esters 4 is to bind all the components
of the loose TS, with delivery of a La3+-bound methoxide to a
phosphoryl group that is electrophilically activated by metal ion
coordination. In the catalyzed reactions of 3, not containing the
o-CO2Me group, the departing aryl oxide can be stabilized
electrostatically by the positively charged catalytic system.
However, esters 4, with the o-CO2Me group, exhibit an
additional coordinative interaction with the metal ions that
provides modest but real electrophillic assistance to the leaving
group departure.
(2) a faster rate of cleavage for o-CO2Me-containing substrates
relative to ones with the same leaving group sspKa but not having
the o-CO2Me group, and (3) cleavage products that are metal-
complex-bound phenoxides indicates that these specially de-
signed neutral OP triester substrates are subject to modest but
real LGA. Our preferred mechanism for this process is a
concerted one where the catalyst binds to all three components
of a rather exploded TS, namely the nucleophilic -OCH3, -OAr
nucleofuge, and a central (RO)2P2+-O- unit bridged between
a catalytic La3+2(-OCH3)x-1 entity. Enzymatic mechanisms for
phosphoryl transfer involving LGA through transition states
resembling that offered here in Scheme 2 have been proposed
for dinuclear M(II) phosphodiestereases,48 for alkaline phos-
phatase,49 and in an early paper16b for PTE, although there is
no general consensus for the latter’s mechanism of operation.50
Acknowledgment. The authors gratefully acknowledge the
financial assistance of the Natural Sciences and Engineering
Research Council of Canada. This project received support (grant
no. HDTRA1-08-1-0046) from the Defense Threat Reduction
Agency-Joint Science and Technology Office, Basic and Support-
ing Sciences Division. C.T.L. thanks NSERC Canada for a PGS-D
postgraduate scholarship (2006-2010).
5. Conclusions
The common feature between dinuclear Zn(II)-containing
PTEs and the La3+-catalyzed cleavage of phosphate triesters is
the large negative Brønsted ꢀlg, indicating substantial cleavage
of the P-OAr bond in the TS. A consequence of this is evident
from inspection of Figure 3, where a convergence of the La3+
-
s
and methoxide-catalyzed Brønsted plots at high pKa indicates
s
Supporting Information Available: Tables of observed pseudo-
first-order rate constants for the methoxide-promoted cleavage
of substrates 3 and 4 and plots of kobs vs [La3+]total/2; supporting
text describing procedures for the synthesis of all compounds
used in this study and their NMR data, attempted data treatment
with Jaffe´ equation, and experimental procedure for the Job plot
of 4c. This material is available free of charge via the Internet
at http://pubs.acs.org.
that substrates with good leaving groups are accelerated
disproportionately by the catalyst relative to substrates with poor
cat
-OR
leaving groups, (|ꢀ | > |ꢀ |). Leaving group assistance would
lg
lg
be optimized and generally desirable in situations where a
catalyst (enzyme) needs to select for substrates with poor leaving
cat
-OR
groups relative to those with good leaving groups (|ꢀ | < |ꢀ |).
lg
lg
Since the neutral OP phosphate esters that are toxic to organisms
are those having leaving groups of pKa e 8, there is no
evolutionary pressure for the PTEs to build in LGA to the
catalytic machinery for cleaving these types of OP materials.
To probe whether LGA is possible for our synthetic PTE
system comprising La3+-catalysis of phosphate triesters in
alcohol, we have employed a series of phosphate triesters having
aryloxy leaving groups with an o-CO2Me which were previously
shown to induce LGA to the metal-catalyzed cleavage of
phosphate diesters.22 Here, the three-fold demonstration of (1)
a less negative ꢀlg for the Brønsted plot of the catalyzed reaction,
JA904659E
(48) Steitz, T. S.; Steitz, J. A. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 6498.
(49) (a) Kim, E. E.; Wyckoff, H. W. J. Mol. Biol. 1991, 218, 449. (b)
Holtz, K. M.; Stec, B.; Kantrowitz, E. R. J. Biol. Chem. 1999, 274,
8351. (c) Zalatan, J. G.; Herschlag, D. J. Am. Chem. Soc. 2006, 128,
1293.
(50) (a) Wong, K.-Y.; Gao, J. Biochemistry 2007, 46, 13352. (b) Samples,
C. R.; Raushel, F. M.; DeRose, V. J. Biochemistry 2007, 46, 3435.
(c) Liu, J.-W.; Salem, G.; Ollis, D. L. J. Mol. Biol. 2008, 375, 1189.
9
13748 J. AM. CHEM. SOC. VOL. 131, NO. 38, 2009