Quinone methide phosphodiester alkylations under aqueous conditions

Article abstract of DOI:10.1021/jo015792+

A detailed analysis of the alkylation of phosphodiesters with a p-quinone methide under aqueous conditions has been accomplished. The relative rates of phosphodiester alkylation and hydrolysis have been examined by ¹H NMR analysis of the reacti

Full text of DOI:10.1021/jo015792+

7072
J . Org. Chem. 2001, 66, 7072-7077
Qu in on e Meth id e P h osp h od iester Alk yla tion s u n d er Aqu eou s
Con d ition s
Qibing Zhou† and Kenneth D. Turnbull*
Department of Chemistry and Biochemistry, University of Arkansas, Fayetteville, Arkansas 72701
kturnbul@uark.edu
Received May 24, 2001
A detailed analysis of the alkylation of phosphodiesters with a p-quinone methide under aqueous
conditions has been accomplished. The relative rates of phosphodiester alkylation and hydrolysis
1
have been examined by H NMR analysis of the reaction of 2,6-dimethyl-p-quinone methide in a
buffered diethyl phosphate/acetonitrile solution (1:9 v/v, pH 4.0). The rate of hydrolysis of the
quinone methide was confirmed by UV analysis in 28.5% solutions of aqueous inorganic phosphate
in acetonitrile at pH 4.0 and 7.0. Similarly, the rate of phosphodiester alkylations by the quinone
methide was also confirmed by UV analysis in 28.5% solutions of aqueous dibenzyl, dibutyl, or
diethyl phosphate in acetonitrile at pH 4.0 and 7.0. These kinetic studies further establish that
the phosphodiester alkylation reactions are acid-catalyzed, second-order processes. The rate constant
for phosphodiester alkylation was found to range from approximately 370-3700 times the rate
constant of quinone methide hydrolysis with diethyl and dibenzyl phosphate, respectively (pH 4.0,
28.5% aqueous acetonitrile).
Sch em e 1
In tr od u ction
A fundamental determinant of the biological effective-
ness of most drugs is their hydrolytic stability.1 Hydroly-
sis is a prominent concern in the development of drugs
involving quinone methides as the reactive intermediate.2
Hydrolysis has been observed as the major competitive
reaction in alkylation studies of nucleic acids by quinone
methide derivatives.3 Skibo and co-workers recently
demonstrated selective phosphate alkylation of nucleic
acids with aziridinium derivatives, yet found that hy-
drolysis of the active intermediate and products was
predominant.4 We previously reported that phosphodi-
ester alkylation with 2,6-dimethyl-p-quinone methide 1
to produce phosphotriester 2 is readily achieved under
anhydrous conditions (CDCl3/CD3CN, 1:1) when pro-
moted by a Brønsted acid (Scheme 1).5 We now examine
this alkylation process under aqueous conditions to
determine the competitive rate of quinone methide hy-
drolysis versus phosphodiester alkylation.
Sch em e 2
Several studies have included the hydrolysis of a
variety of simple p-quinone methides 3 to afford benzyl
alcohols 4 under aqueous conditions (Scheme 2).6,7 The
substituents at the 2- and 6-positions of p-quinone
methides have a dramatic effect on the rate of hydrolysis.
The half-life in a phosphate buffer (pH 7.4, 25 °C) ranged
from 1.3 s for 3a 6b to 3060 s for 3c.6d The acid- and base-
catalyzed nature of the hydrolysis of quinone methides
has also been demonstrated.7 Bolton and co-workers
examined the pH-rate profile of the hydrolysis of qui-
† Current address: Department of Chemistry and Biochemistry,
University of Maryland, College Park, MD 20742.
(1) (a) Streng, W. H. Drug Discovery Today 1997, 2 (10), 415-426.
(b) Stella, V. J . Parenteral Sci. Technol. 1986, 142-163. (c) Connors,
K. A.; Amidon, G. L.; Stella, V. J . In Chemical Stability of Pharma-
ceuticals: A Handbook for Pharmacists, 2nd ed.; J ohn Wiley & Son:
New York, 1986.
(2) Cabaret, D., Adediran, S. A. Gonzalez, M. J . G.; Pratt, R. F.;
Wakselman, M. J . Org. Chem. 1999, 64, 713-720. (b) Myers, J . K.;
Cohen, J . D.; Widlanski, T. S. J . Am. Chem. Soc. 1995, 117, 11049-
11054. (c) Wakselman, M. Nouv. J . Chim. 1983, 7, 439-447.
(3) For recent examples, see: (a) Pande, P.; Shearer, J .; Yang, J .;
Greenberg, W. A.; Rokita, S. E. J . Am. Chem. Soc. 1999, 121, 6773-
6779. (b) Ouyang, A.; Skibo, E. B. J . Org. Chem. 1998, 63, 1893-1900.
(c) Angle, S. R.; Rainier, J . D.; Woytowicz, C. J . Org. Chem. 1997, 62,
5884-5892. (d) Lewis, M. A.; Yoerg, D. G.; Bolton, J . L.; Thompson, J .
A. Chem. Res. Toxicol. 1996, 9, 1368-1374.
(6) (a) Bolton, J . L.; Turnipseed, S. B.; Thompson, J . A. Chem.-Biol.
Interact. 1997, 107, 185-200. (b) Bolton, J . L.; Comeau, E.; Vuko-
manovic, V. Chem.-Biol. Interact. 1995, 95, 279-290. (c) Thompson,
D. C.; Perera, K. Biochem. Biophys. Res. Commun. 1995, 209, 6-11.
(d) Bolton, J . L.; Valerio, J r., L. G.; Thompson, J . A. Chem. Res. Toxicol.
1992, 5, 816-822. (e) Richard, J . P. J . Am. Chem. Soc. 1991, 113,
4588-4595.
(7) (a) McCracken, P. G.; Bolton, J . L.; Thatcher, R. J . J . Org. Chem.
1997, 62, 1820-1825. (b) Hemmingson, J . A.; Leary, G. J . Chem. Soc.,
Perkin Trans. 2 1975, 1584-1587.
(4) (a) Skibo, E. B.; Islam, I.; Schulz, W. G.; Zhou, R.; Bess, L.;
Boruah, R. Synlett 1996, 297-309. (b) Skibo, E. B.; Xing, C. Biochem-
istry 1998, 37, 15199-15213.
(5) Zhou, Q.; Turnbull, K. D. J . Org. Chem. 1999, 64, 2847-2851.
10.1021/jo015792+ CCC: $20.00 © 2001 American Chemical Society
Published on Web 09/20/2001

Quinone Methide Phosphodiester Alkylations
J . Org. Chem., Vol. 66, No. 21, 2001 7073
Sch em e 3
none methide 3b in aqueous buffers at 25 °C and found
the rate of hydrolysis was constant (2.7 × 10-4 M-1‚s-1
)
over the pH range of 4.7-9.5.7a
We report the investigation of phosphodiester alkyla-
tion with 2,6-dimethyl-p-quinone methide 1 in various
aqueous buffered phosphodiester/acetonitrile solutions.
These investigations have enhanced our understanding
of the phosphodiester alkylation mechanism and guided
the development of more advanced alkylating reagents8
for future applications.
F igu r e 1. Reaction plot showing the loss of quinone methide
1 (2) through phosphodiester alkylation to phosphotriester 2d
(9) and hydrolysis to benzyl alcohol 5 (b) in a buffered diethyl
phosphate/acetonitrile solution (1:9 v/v) at pH 4.0. The reaction
1
was monitored by H NMR analysis at 20.0 °C. Only one set
of the triplicated 1H NMR derived data is shown in the plot.
The plotted curves were derived from simulation of the
reaction based on the proposed mechanism (see text). All of
the experimental data points derived from the three runs were
within 3% of the simulated curves.
Resu lts a n d Discu ssion s
Our initial investigations of hydrolysis versus phos-
phodiester alkylation by quinone methide 1 were based
on 1H NMR analysis. The half-life of 2,6-dimethyl qui-
none methide 1 in a phosphate buffer (pH 7.40, 25.0 °C)
has been determined by Bolton and co-workers to be 26
s.6d To sufficiently extend the half-life of quinone methide
1 for H NMR analysis, initial studies were carried out
in solutions of only 10% aqueous buffered diethyl phos-
phate in acetonitrile.
formation of phosphotriester 2d was evident in the
reaction by the benzylic proton resonances. These reso-
nance signals appeared as a doublet (3J HP ) 8.2 Hz) at
4.84 ppm characteristic of phosphorus coupling in accord
with our previous report.5 The formation of benzyl alcohol
5 as the hydrolysis product of quinone methide 1 was
confirmed by appropriate resonance signal enhancement
upon addition of authentic benzyl alcohol 5. The percent
conversions to phosphotriester 2d and benzyl alcohol 5
from quinone methide 1 were calculated on the basis of
the average area integration of the corresponding char-
1
A stock solution of quinone methide 1 in CD3CN was
obtained from oxidation of 2,4,6-trimethylphenol with
silver(I) oxide over 40 min. A stock solution of diethyl
phosphate buffer in D2O was prepared by initial neu-
tralization of diethyl phosphoric acid with a solution of
KOD in D2O (40%), followed by pH adjustment with
methanesulfonic acid. The use of methanesulfonic acid
was based on prior determination that its conjugate base
did not undergo nucleophilic addition to the quinone
methide.5,9 The resulting buffer solution (100 µL) was
then added to quinone methide 1 (900 µL) in CD3CN. The
final concentrations of quinone methide 1 and diethyl
phosphate were 6.60 and 25.0 mM, respectively. The
concentration of water (5.56 M) in the reaction solution
was approximately 220 times higher than that of diethyl
phosphate. A preliminary analysis of the alkylation
reaction of diethyl phosphate with quinone methide 1 at
pH 2.0, 4.0, and 5.7 revealed the optimal pH for phos-
phodiester alkylation was at pH 4.0 (buffered aqueous
diethyl phosphate/acetonitrile solution, 1:9 v/v). The
reaction was monitored by 1H NMR analysis over 40 min
at 20.0 °C in triplicate. The pH of the reaction solution
remained unchanged throughout the reaction.
1
acteristic resonance signals in the H NMR spectra. The
half-life of quinone methide 1 was approximately 4.5 min
under the reaction conditions (Figure 1). The percent
conversion to phosphotriester 2d reached a maximum of
16% in approximately 9 min, followed by a slow decrease
to 10% within 40 min. The percent benzyl alcohol 5 in
1
the reaction reached 75% after 10 min. These H NMR
studies clearly demonstrate that phosphodiester alkyla-
tion by quinone methide 1 occurs in an aqueous buffered
dialkyl phosphate/acetonitrile mixture (1:9 v/v) at pH 4.0
along with the hydrolysis of quinone methide 1.
1
Further analysis of the triplicated H NMR data during
the initial 4 min (Figure 1) afforded the rates of quinone
methide reactions. These are 1.54 × 10-5 M‚s-1 for
quinone methide 1 disappearance, 1.28 × 10-5 M‚s-1 for
benzyl alcohol 5 formation and 0.26 × 10-5 M‚s-1 for
phosphotriester 2d formation. This indicates that no side
reactions are occurring as the rate of the quinone methide
disappearance is completely accounted for by the com-
bined rates of benzyl alcohol 5 and phosphotriester 2d
formations according to eq 1. Considering the second-
order nature of both phosphodiester alkylation and
quinone methide hydrolysis reactions, the loss of quinone
methide 1 over time can be expressed by eq 2. This
provides the rate constant for diethyl phosphate alkyla-
tion (k1) and quinone methide hydrolysis (k2) of 1.6 × 10-2
and 3.5 × 10-4 M-1‚s-1, respectively. The second-order
rate constant for hydrolysis of quinone methide 1 is only
2.2% of that for phosphodiester alkylation.
1
As indicated by H NMR analysis, quinone methide 1
in the buffered diethyl phosphate/acetonitrile solution at
pH 4.0 underwent two reactions: (i) phosphodiester
alkylation to 2d and (ii) hydrolysis to 5 (Scheme 3). The
(8) (a) Dyer, R. G.; Turnbull, K. D. J . Org. Chem. 1999, 64, 7988-
7995. (b) Zhou, Q.; Turnbull, K. D. J . Org. Chem. 2000, 65, 2022-
2029.
(9) Leary, G.; Miller, I. J .; Thomas, W.; Woolhouse, A. D. J . Chem.
Soc., Perkin Trans. 2 1977, 1737-1739.

7074 J . Org. Chem., Vol. 66, No. 21, 2001
Zhou and Turnbull
Ta ble 1. Ra tes (kobs) of Qu in on e Meth id e Disa p p ea r a n ce in Bu ffer ed P h osp h a te/CH3CN Solu tion (28.5%, p H 4.0)a
inorganic phosphate
dibenzyl phosphate
dibutyl phosphate
diethyl phosphate
entry concn (mM) rate (×10-3 s-1
)
concn (mM) rate (×10-3 s-1
)
concn (mM) rate (×10-3 s-1
)
concn (mM) rate (×10-3 s-1
)
1
2
3
4
9.5
19
29
38
6.05 ( 0.21
5.99 ( 0.06
6.13 ( 0.05
6.11 ( 0.09
0.95
1.9
7.38 ( 0.11
8.41 ( 0.15
9.96 ( 0.22
11.0 ( 0.30
2.4
4.8
9.5
6.42 ( 0.05
6.81 ( 0.03
8.42 ( 0.15
9.16 ( 0.02
9.5
19
6.95 ( 0.31
8.03 ( 0.30
9.51 ( 0.16
10.6 ( 0.20
2.9
29
38
3.8
14
a
All experiments were monitored at 288 nm by UV spectroscopy in triplicate at 25.0 °C. The rates were calculated as the slope of the
natural log of absorbance over time (ln A/t).
-d[1]/dt ) d[2]/dt + d[5]/dt
(1)
) k1[phosphodiester][1] + k2[H2O][1] (2)
) kobs[1]
(3)
(4)
kobs ) k1[phosphodiester] + k2[H2O]
1
Confirmation of the reaction kinetics derived from H
NMR analysis was accomplished by UV spectroscopic
analysis at 288 nm. The first-order rates of the disap-
pearance of quinone methide 1 (kobs) in 28.5% aqueous
buffered inorganic phosphate in acetonitrile at pH 4.0
was initially determined to afford the rate of hydrolysis
of quinone methide 1. A similar analysis was then
conducted in 28.5% aqueous buffered dibenzyl, dibutyl,
or diethyl phosphate in acetonitrile at pH 4.0 to afford
the rate of phosphodiester alkylation.
A stock solution of quinone methide 1 (1.00 mM in
acetonitrile) was obtained from oxidation of 2,4,6-trimeth-
yl phenol with silver(I) oxide over 45 min. Stock solutions
(10, 20, 30, or 40 mM) of 30% aqueous buffered inorganic
phosphate in acetonitrile at pH 4.0 were prepared by pH
adjustment of the acid with diluted NaOH solutions. The
quinone methide solution (1.00 mM, 150 µL) was added
to the appropriate concentration of inorganic phosphate
buffer solution (2.85 mL) at 25.0 °C. The final concentra-
tion of quinone methide 1 in the reaction solution was
50.0 µM, and the concentrations of aqueous buffered
inorganic phosphate/acetonitrile solutions were 95% of
their stock concentrations. The percent water in the final
reaction solution was 28.5% and approximately 15.8 M.
The pH of the final reaction solution remained unchanged
throughout the reaction. The observed rates of disap-
pearance of quinone methide 1 (kobs) were calculated as
the slope of the natural log of absorbance over time. All
reactions were run in triplicate.
The first-order rate of disappearance of quinone meth-
ide 1 was found to be independent of the inorganic
phosphate concentration (Table 1). The kobs was ap-
proximately 6.05 × 10-3 s-1. A water concentration of 15.8
M in the 28.5% buffer/acetonitrile mixture affords the
second-order hydrolysis rate constant (k2) of 3.8 × 10-4
M-1‚s-1. Our results confirm that the quinone methide
hydrolysis is a specific acid catalyzed process, as previ-
ously demonstrated.7a,10
The observed rate of disappearance of the quinone
methide 1 (kobs) in 28.5% aqueous buffered dibenzyl,
dibutyl, or diethyl phosphate in acetonitrile at pH 4.0
was similarly determined. Stock solutions of 30% aqueous
buffered dibenzyl, dibutyl, or diethyl phosphate in ac-
etonitrile at pH 4.0 were prepared by pH adjustment of
the dialkyl phosphoric acids with dilute NaOH solutions.
The solubility dependent concentrations of these stock
F igu r e 2. Dependence of the rate of quinone methide 1
disappearance on the concentration of dialkyl phosphate at
pH 4.0. (b) Dibenzyl phosphate, slope: 1.3 × 10-3 mM-1‚s-1
(R2 ) 0.993). (9) Dibutyl phosphate, slope: 0.24 × 10-3
mM-1‚s-1 (R2 ) 0.974). (2) Diethyl phosphate, slope: 0.13 ×
10-3 mM-1‚s-1 (R2 ) 0.996). (×) inorganic phosphate.
solutions were 10, 20, 30, or 40 mM for diethyl phosphate;
1.0, 2.0, 3.0, or 4.0 mM for dibenzyl phosphate; and 2.5,
5.0, 10, or 15 mM for dibutyl phosphate. The quinone
methide solution (1.00 mM, 150 µL) was added to the
appropriate phosphodiester buffer solution (2.85 mL) at
25.0 °C.11,12 The observed rates of disappearance of the
quinone methide 1 (kobs) were calculated as the slope of
the natural log of absorbance over time.
The first-order rate of quinone methide 1 disappear-
ance (kobs) in phosphodiester/acetonitrile solutions at pH
4.0 was found to increase with increased concentrations
of the phosphodiesters (Table 1).11 The rate increased
from 7.38 × 10-3 to 11.0 × 10-3 s-1 when the concentra-
tion of dibenzyl phosphate was increased from 0.95 to
3.8 mM. The kobs in dibutyl phosphate increased from
6.42 × 10-3 to 9.16 × 10-3 s-1 when the concentration
was increased from 2.4 to 14 mM. Similarly, kobs of the
disappearance of quinone methide 1 in diethyl phosphate
also increased from 6.95 × 10-3 to 10.6 × 10-3 s-1 with
increased concentrations from 9.5 to 38 mM.
A plot of kobs versus the phosphate concentrations from
Table 1 is shown in Figure 2. The linear dependencies
with each dialkyl phosphate confirms phosphodiester
(11) The ionic strength of the reactions was not normalized due to
problems in consistent solubility in all of the reactions analyzed and
due to competitive nucleophilic reactions with various salts. On the
basis of the relative increase in the observed rate of the loss of 1 with
the three different dialkyl phosphates at nearly equivalent salt
concentrations (e.g., Table 1, 9.5 mM), it appears that changes in ionic
strength do not account for the relative rate differences in this reaction
series at the concentration range studied. It has been observed with
hydrolysis of highly stabilized benzylic carbenium ions (similar to our
protonated quinone methide) that ionic strength effects are less
important than the nature of the anion component.12
(10) Velek, J .; Koutek, B.; Soucek, M. Collect. Czech. Chem. Com-
mun. 1979, 44, 110-122.
(12) Crugeiras, J .; Maskill, H. Can. J . Chem. 1999, 77, 530-536.

Quinone Methide Phosphodiester Alkylations
J . Org. Chem., Vol. 66, No. 21, 2001 7075
Ta ble 2. Ra tes (kobs) of Qu in on e Meth id e Disa p p ea r a n ce in Bu ffer ed P h osp h a te/CH3CN Solu tion (28.5%, p H 7.0)a
inorganic phosphateb
dibenzyl phosphatec
dibutyl phosphatec
diethyl phosphatec
entry concn (mM) rate (×10-4 s-1
1
2
3
4
)
concn (mM) rate (×10-4 s-1
)
concn (mM) rate (×10-4 s-1
)
concn (mM) rate (×10-4 s-1
)
9.5
9.24 ( 0.03
0.95
1.9
9.92 ( 0.26
10.0 ( 0.38
10.0 ( 0.31
9.81 ( 0.14
2.4
4.8
9.5
9.62 ( 0.26
9.48 ( 0.20
9.64 ( 0.16
9.07 ( 0.21
9.5
19
9.26 ( 0.36
9.49 ( 0.22
9.31 ( 0.23
9.05 ( 0.05
2.9
29
38
3.8
14
a
All experiments were monitored at 288 nm by UV spectroscopy in triplicate at 25.0 °C. The rates were calculated as the slope of ln
A/t. b The rate was determined at only one concentration on the basis that hydrolysis at pH 4.0 was independent of phosphate concentration.
c All the dialkyl phosphate buffer acetonitrile mixtures contained 10.0 mM inorganic phosphate to effectively buffer the pH.
Sch em e 4
alkylation is a second-order process. The rates (k1) of
phosphodiester alkylation determined from the slope of
the line are 1.3 M-1‚s-1 for dibenzyl phosphate, 0.24
M-1‚s-1 for dibutyl phosphate, and 0.13 M-1‚s-1 for
diethyl phosphate at pH 4.0 in 28.5% aqueous acetonitrile
solutions at 25.0 °C. The rate (k1) of dibenzyl phosphate
alkylation is approximately 5 times faster than that of
dibutyl phosphate and 10 times that of diethyl phosphate.
The difference in the rates of phosphodiester alkylation
is likely due to the different nucleophilicities of these
dialkyl phosphates. The variation in nucleophilicities
may be the result of solvation effects.13
The relative second-order rate constant analysis sug-
gests that dibenzyl phosphate (k1 ) 1.3 M-1‚s-1) is
approximately 3700 times more reactive than water
(k2 ) 3.5 × 10-4 M-1‚s-1) in reacting with quinone
methide 1 under these reaction conditions at pH 4.0. The
least reactive phosphodiester, diethyl phosphate, is still
∼370 times more reactive than water.
experiments were run in triplicate and the pH of the
reaction solution remained unchanged throughout the
reaction. The rates of quinone methide 1 disappearance
in buffered dialkyl phosphate/acetonitrile solutions at
pH 7.0 were nearly identical to the rate of the hydrolysis
(kobs ≈ 9. 24 × 10-4 s-1) (Table 2). This reveals that
hydrolysis predominates at pH 7.0. This was confirmed
by 1H NMR analysis in 10% D2O/CD3CN at pH 7.0 where
only hydrolysis product 5 was observed. This emphasized
the acid-catalyzed nature of phosphodiester alkylation by
quinone methide 1.5
Comparison of the reaction monitored by 1H NMR
analysis to those monitored by UV analysis reveals that
the rate constant for diethyl phosphate alkylation (k1) is
approximately 8-fold larger by UV analysis. The reaction
1
monitored by H NMR was conducted in 10% D2O/CD3-
CN at 20.0 °C and provided a k1 of 1.6 × 10-2 M-1‚s-1
.
The reaction monitored by UV analysis was conducted
in 28.5% H2O/CH3CN at 25.0 °C and provided a k1 of 0.13
M-1‚s-1. This 8-fold rate constant enhancement for di-
ethyl phosphate alkylation in going from 10% D2O to
28.5% H2O in acetonitrile is not observed in the hydroly-
sis rate constants, which remained relatively constant
(3.5 × 10-4 versus 3.8 × 10-4 M-1‚s-1, respectively). This
difference is attributed to a combination of temperature,
solvent polarity,14 and solvent isotope effects.6e,10
Based on these experimental results, the following
mechanism for the reaction of quinone methide 1 in an
aqueous buffered phosphodiester/acetonitrile solution at
pH 4.0 is proposed (Scheme 4). Two fast equilibriums
exist in the reaction: the protonation of quinone methide
1, resulting in the low steady-state concentration of
reactive intermediate 6 (K1), and ionization of dialkyl
phosphoric acid (K2). Protonated quinone methide inter-
mediate 6 reacts with dialkyl phosphate to form phos-
photriester 2 (k1). The equilibrium nature of this reaction
is primarily based on our previous studies,5 and the
accuracy afforded to the simulation of this mechanism
when this reversibility (k-1) is incorporated. Benzyl
alcohol 5 is produced from both the direct hydrolysis of
intermediate 6 (k2) and the slow hydrolysis of phospho-
triester 2 (k3). The hydrolysis of phosphotriester 2d to
benzyl alcohol 5 was clearly observed by 1H NMR
analysis as the slow decrease of 2d coincided with the
slow increase of 5 after approximately 15 min in the
reaction (see Figure 1). This mechanism accounts for the
rate of quinone methide disappearance by the combined
rates of phosphodiester alkylation and quinone methide
hydrolysis. The proposed mechanism indicates that phos-
phodiester alkylation is an acid-catalyzed, second-order
process and quinone methide hydrolysis is a specific acid
The rate of hydrolysis of quinone methide 1 at pH 7.0
was determined in a manner similar to those described
at pH 4.0. The rate of hydrolysis of quinone methide 1
(kobs) was determined at a single aqueous buffered
inorganic phosphate/acetonitrile concentration (9.5 mM)
based on the finding that the hydrolysis rate was
independent of the inorganic phosphate buffer concentra-
tion. The rate kobs in inorganic phosphate at pH 7.0 was
determined to be 9.24 × 10-4 s-1 at 25.0 °C (Table 2).
This was approximately 15% of the rate at pH 4.0
(6.05 × 10-3 s-1). This pH dependent rate of hydrolysis
agrees with previous reported results.7,10
The rates of quinone methide 1 disappearance (kobs
)
in aqueous buffered phosphodiester/acetonitrile solution
at pH 7.0 were similarly determined. All dialkyl phos-
phate stock solutions at pH 7.0 contained 10 mM
inorganic phosphate to effectively buffer the pH. All
(13) Courtemanche, P.; Merlin, J .-C. Bull. Soc. Chim. Fr. 1967, 10,
3911-3919.
(14) Bunton, C. A.; Mhala, M. M.; Moffatt, J . R. J . Org. Chem. 1984,
49, 3639-3641.

7076 J . Org. Chem., Vol. 66, No. 21, 2001
Zhou and Turnbull
phosphate solution (100 µL) was added to a solution of quinone
methide 1 in CD3CN (7.3 mM, 900 µL), prepared as described
as above. The pH of the final reaction solution was measured
as 4.0 and the final concentrations of quinone methide 1 and
diethyl phosphate were 6.6 and 25 mM, respectively. The
reaction was monitored by 1H NMR analysis over 40 min
approximately 9 half-lives) at 20.0 ( 0.1 °C. The percent
conversion to trialkyl phosphate 2d was calculated on the basis
of area integration of the resonance of benzylic protons of 2d
(at 4.84 ppm) relative to the resonance of alkylidene protons
of quinone methide 1 (at 5.86 ppm). The percent of benzyl
alcohol 5 was calculated on the basis of the area integration
of the resonance of the benzylic protons of 5 (at 4.36 ppm).
Formation of 5 was further confirmed by adding authentic
benzyl alcohol 5 to the reaction. The pH of the reaction solution
remained unchanged throughout the reaction. The experiment
was repeated at least three times. Trialkyl phosphate 2d
(while not sufficiently stable to allow isolation, the following
solution characterization was in accord with previously re-
ported analysis):5 1H NMR (10% D2O in CD3CN, 270 MHz) δ
catalyzed process. This proposed mechanism is fully
correlated by both H NMR and UV analysis results.
1
A complete simulation of the 1H NMR experimental
results according to this proposed mechanism afforded
the plotted curves shown in Figure 1. All of the triplicated
experimental data were within 3% of the simulated
reaction curves. This further supports the proposed
mechanism of p-quinone methide phosphodiester alky-
lation and hydrolysis according to Scheme 4.
Con clu sion
We have extended our investigation of phosphodiester
alkylation with a p-quinone methide in aqueous buffered
phosphodiester/acetonitrile solutions in order to deter-
mine the competitive hydrolysis during phosphodiester
alkylation. Our 1H NMR studies revealed that phos-
phodiester alkylation occurs with a faster rate constant
relative to hydrolysis at pH 4.0. This is attributed to the
higher nucleophilicity of the phosphodiester relative to
water. Further investigation of quinone methide 1 reac-
tions by UV analysis in various aqueous buffered phos-
phate/acetonitrile solutions confirmed this result as an
acid-catalyzed, second-order process.
These investigations continue to direct our design
process for development of a DNA phosphodiester alky-
lating reagent for potential in vivo applications. On the
basis of these investigations, the phosphodiester alky-
lating reagent will incorporate appropriate functionality
to accomplish quinone methide activation at neutral pH,
increase the effective concentration of the phosphodiester
target to outcompete hydrolysis, and trap the phospho-
triester product to prevent conversion to the hydrolysis
products.8b
3
6.95 (s, 2H), 4.84 (d, J HP ) 8.4 Hz, 2H), 3.96-4.07 (m, 4H),
2.14 (s, 6H), 1.23-1.28 (m, 6H).
Stu d y of Qu in on e Meth id e (1) Rea ction in Bu ffer ed
P h osp h a te/Aceton itr ile Solu tion (3:7 v/v, p H 4.0, 25 °C).
A stock solution (10.0 mM) of 2,4,6-trimethylphenol was
prepared by dissolving the phenol (13.6 mg, 0.100 mmol) in
CH3CN (10.0 mL). A 1.00 mM solution (10.0 mL) of 2,4,6-
trimethylphenol was obtained by diluting the stock solution
(10.0 mM, 1.00 mL) with CH3CN. The resulting 1.00 mM
trimethylphenol solution was oxidized to quinone methide 1
by stirring with silver(I) oxide (250 mg) at room temperature
for 45 min. The suspension was filtered with an Acrodisc filter
(13 CR, 0.45 µm) to afford the desired quinone methide solution
(1.00 mM, 10.0 mL).
A stock solution (240 mM, 6.0 mL) of inorganic phosphoric
acid was prepared by diluting the acid (85%, 166.0 mg, 1.44
mmol) with H2O. Inorganic phosphate solutions (5.4 mL) of
various concentrations were prepared by diluting the stock
solution (240 mM; 250, 500, 750 or 1000 µL) with CH3CN (4.20
mL) and H2O (950, 700, 450 or 200 µL, correspondingly). The
resulting solutions were adjusted to pH 4.0 with 2, 0.2 and
0.02 M NaOH aqueous solution and H2O to afford the final
buffered inorganic phosphate/acetonitrile solutions (3:7 v/v, 6.0
mL of 10, 20, 30, or 40 mM, respectively).
A stock solution (24.0 mM, 6.0 mL) of dibenzyl phosphoric
acid was prepared by dissolving the acid (40.0 mg, 0.144 mmol)
with CH3CN. Dibenzyl phosphate solutions (5.40 mL) of
various concentrations were prepared by diluting the stock
solution (24.0 mM; 250, 500, 750, or 1000 µL) with CH3CN
(3.95, 3.70, 3.45 or 3.20 mL, correspondingly) and H2O (1.20
mL). The resulting solutions were adjusted to pH 4.0 with 2,
0.2 and 0.02 M NaOH aqueous solution and H2O to afford the
final buffered dibenzyl phosphate/acetonitrile solutions (3:7 v/v,
6.0 mL of 1.0, 2.0, 3.0, or 4.0 mM, respectively).
A stock solution (120 mM, 6.0 mL) of dibutyl phosphoric acid
was prepared by diluting the acid (151.3 mg, 0.720 mmol) with
CH3CN. Dibutyl phosphate solutions (5.40 mL) of various
concentrations were prepared by diluting the stock solution
(120 mM; 125, 250, 500, or 750 µL) with CH3CN (4.07, 3.95,
3.70, or 3.45 mL, correspondingly) and H2O (1.20 mL). The
resulting solutions were adjusted to pH 4.0 with 2, 0.2, and
0.02 M NaOH aqueous solution and H2O to afford the final
buffered dibutyl phosphate/acetonitrile solutions (3:7 v/v, 6.0
mL of 2.5, 5.0, 10, or 15 mM, respectively).
Exp er im en ta l Section
All commercially available compounds were purchased from
Aldrich Chemical Co. (Milwaukee, WI), Lancaster Synthesis,
Inc. (Windham, NH), or Acros Organics (Fisher Scientific) and
used without further purification unless noted otherwise.
Deuterium solvents were purchased from Cambridge Isotope
Laboratories, Inc. (Andover, MA) and Merck Sharp & Dohme
Limited (Montreal, Canada). Diethyl phosphoric acid was
vacuum distilled at 0.3 Torr prior to use. All H2O was distilled
and deionized through a Milli-Q RG purification system.
Acetonitrile was passed through a Solv-Tek ST-002 Solvent
Purification System (Solv-Tek Inc., Berryville, VA). pH was
measured with a Ag/AgCl electrode/Accumet 910 at 24.0 °C.
1H NMR analysis was carried out on a J EOL 270 NMR
spectrometer. UV data were recorded at 288 nm using a
Hitachi U-2000 spectrometer with a temperature-controlled
cell at 25.0 ( 0.1 °C.
Qu in on e Met h id e (1) Solu t ion . A solution of 2,4,6-
trimethylphenol (1.1 mg, 8.1 µmol) in CD3CN (1.10 mL) was
oxidized with silver(I) oxide (150 mg) by stirring at room
temperature for 40 min. The suspension was filtered with glass
wool to give known quinone methide 1 in solution:15 1H NMR
(CD3CN, 270 MHz) δ 7.10 (s, 2H), 5.86 (s, 2H), 2.10 (s, 6H).
Stu d y of Qu in on e Meth id e (1) Rea ction in a Bu ffer ed
Dieth yl P h osp h a te/Aceton itr ile Solu tion (1:9 v/v, p H 4.0,
20.0 °C). To a solution of diethylphosphoric acid (39.2 mg,
0.250 mmol in 800 µL D2O) was added a solution of KOD (40%
in D2O, 40 µL). The pH of the resulting solution was adjusted
with a solution of methanesulfonic acid (141.8 mg in 1.00 mL
D2O). The final volume of buffered diethyl phosphate solution
was adjusted to 1.0 mL with D2O. The resulting diethyl
A stock solution (240 mM, 6.0 mL) of diethyl phosphoric acid
was prepared by diluting the acid (221.9 mg, 1.44 mmol) with
H2O. Diethyl phosphate solutions (5.40 mL) of various con-
centrations were prepared by diluting the stock solution (240
mM; 250, 500, 750, or 1000 µL) with CH3CN (4.20 mL) and
H2O (950, 700. 450, or 200 µL, correspondingly). The resulting
solutions were adjusted to pH 4.0 with 2, 0.2, and 0.02 M
NaOH aqueous solution and H2O to afford the final buffered
diethyl phosphate/acetonitrile solutions (3:7 v/v, 6.0 mL of 10,
20, 30, or 40 mM, respectively).
(15) Dyall, L. K.; Winstein, S. J . Am. Chem. Soc. 1972, 94, 2196-
2199.

Quinone Methide Phosphodiester Alkylations
J . Org. Chem., Vol. 66, No. 21, 2001 7077
Quinone methide 1 in CH3CN (1.00 mM, 150 µL), prepared
as described as above, was added to the desired inorganic
phosphate or dialkyl phosphate solution (2.85 mL) described
above. The disappearance of quinone methide 1 in the reaction
solution was monitored at 288 nm by UV spectroscopy at
25.0 ( 0.1 °C. All experiments were repeated at least three
times. The pH of all the reaction solutions remained un-
changed throughout the reaction. The rates (kobs) of quinone
methide 1 disappearance were calculated as the slope of ln
A/t. Data are shown in Table 1.
Stu d y of Qu in on e Meth id e (1) Rea ction in Bu ffer ed
P h osp h a te/Aceton itr ile Solu tion s (3:7 v/v, p H 7.0, 25 °C).
A diluted inorganic phosphate solution (5.40 mL) was prepared
by diluting the inorganic phosphoric acid stock solution (240
mM, 250 µL) with CH3CN (4.20 µL) and H2O (950 µL). The
resulting phosphate solution was adjusted to pH 7.0 with 2,
0.2, and 0.02 M NaOH aqueous solution and H2O to afford
the final buffered inorganic phosphate/acetonitrile solutions
(3:7 v/v, 6.0 mL, 10 mM).
Dibenzyl phosphate solutions (5.40 mL) of various concen-
trations were prepared by diluting the dibenzyl phosphoric acid
stock solution (24.0 mM; 250, 500, 750, or 1000 µL) and
inorganic phosphoric acid solution (240 mM, 250 µL) with CH3-
CN (3.95, 3.70, 3.45, or 3.20 mL, correspondingly) and H2O
(950 µL). The resulting solutions were adjusted to pH 7.0 with
2, 0.2, and 0.02 M NaOH aqueous solution and H2O to afford
the final buffered dibenzyl phosphate/acetonitrile solutions (3:7
v/v, 6.0 mL of 1.0, 2.0, 3.0, or 4.0 mM, respectively).
Dibutyl phosphate solutions (5.40 mL) of various concentra-
tions were prepared by diluting the dibutyl phosphoric acid
stock solution (120 mM; 125, 250, 500, or 750 µL) and inorganic
phosphoric acid solution (240 mM, 250 µL) with CH3CN (4.075,
3.95, 3.70, or 3.45 mL, correspondingly) and H2O (950 µL). The
resulting solutions were adjusted to pH 7.0 with 2, 0.2, and
0.02 M NaOH aqueous solution and H2O to afford the final
buffered dibutyl phosphate/acetonitrile solutions (3:7 v/v, 6.0
mL of 2.5, 5.0, 10, 15 mM, respectively).
Diethyl phosphate solutions (5.50 mL) of various concentra-
tions were prepared by diluting the diethyl phosphoric acid
stock solution (240 mM; 250, 500, 750, or 1000 µL) and
inorganic phosphoric acid solution (240 mM, 250 µL) with CH3-
CN (4.20 mL) and H2O (750, 500, 250, or 0 µL, correspond-
ingly). The resulting solutions were adjusted to pH 7.0 with
2, 0.2, and 0.02 M NaOH aqueous solution and H2O to afford
the final buffered diethyl phosphate/acetonitrile solutions (3:7
v/v, 6.0 mL of 10, 20, 30, or 40 mM, respectively).
Quinone methide 1 solution in CH3CN (1.00 mM, 150 µL),
prepared as described as above, was added to the desired
inorganic phosphate or dialkyl phosphate solution (2.85 mL)
described above. The disappearance of quinone methide 1 in
the reaction solution was monitored at 288 nm by UV
spectroscopy at 25.0 ( 0.1 °C. All the experiments were
repeated at least three times. The pH of all the reaction
solutions remained unchanged throughout the reaction. The
rates (kobs) of quinone methide 1 disappearance were calculated
as the slope of ln A/t. Data are shown in Table 2.
Ack n ow led gm en t. We gratefully acknowledge the
financial support of the National Science Foundation
(EHR-9108762 and CHE-9734187) and the Arkansas
Science and Technology Authority (98-B-03).
J O015792+