R. A. Moss, P. K. Gong / Tetrahedron Letters 44 (2003) 7845–7848
Table 2. Kinetics of aryl alkyl phosphonoacetate reactions
7847
Substrate
Ar
R
104 kexch (s−1 a
)
104 kOD (s−1 b
)
104 kTh (s−1 c
)
104 kZr (s−1 d
)
e
g
g
i
f
4a
4b
4c
4d
Ph
Ph
PNP
PNP
Me
Et
Me
Et
4.00
1.95
8.73
4.70
5.79
2.28
7.81
4.79
h
h
j
a D exchange of spacer CH2, pD 10.4, 37°C.
b Hydrolysis of C-OR with NaOD, pD 13.2, 25°C.
c Only C-OR cleavage at pD 3.1, 37°C.
d At pD 1.7, 37°C.
e Only C-OMe cleavage was observed; see text.
f After 2 d at 37°C, 48% of P-O and 52% of C-O cleavage were observed.
g Precipitation occurred.
h Not studied.
i C-OEt was completely cleaved after 5 h at 37°C.
j Quantitative P-O cleavage was observed upon mixing 20 mM Zr4+ and 6 mM 4d at 37°C.
plexes. These could be redissolved upon gentle heating
(and sonication), but rate constants were unobtainable.
Nevertheless, some conclusions can be drawn. As with
dialkyl PA diesters 3, and DMPF,6 Th4+ induces C-OR,
not P-OR or POAr scission; cf., 4a and 4d in Table 2.
pD 1.7, substrate 3c requires 60°C to afford kZr=0.30×
10−4 s−1 for C-OEt cleavage. P-OMe cleavage is not
competitive in this case, whereas DMPF undergoes
Zr4+-mediated P-OMe hydrolysis with kZr=4.4×10−4 s−1
at pD 1.7, 25°C.6 The hydrolytic reactivity and P-OR
chemoselectivity of Zr4+ toward phosphonoformate
diesters are lost with the corresponding PA substrates.17
On the other hand, Zr4+ slowly cleaves C-OMe and
P-OPh competitively (substrate 4a), while P-OPNP
cleaves in preference to C-OEt (substrate 4d). Here we
find an echo of the P-OR chemoselectivity that Zr4+
exerts in phosphonoformate diester hydrolysis.6 How-
ever, the Th4+/C-OR, Zr4+/P-OR chemoselectivity
dichotomy is not as clearly expressed with PA diesters
as with DMPF; the PA substrates require a P-OAr
leaving group to compete against C-OR, even with Zr4+
mediation.
In order to observe P-O esterolysis with PA diesters, we
require aryl alkyl substrates 4a–d, where the phosphoryl
site features a more reactive aryloxide leaving group,
while the carbonyl site remains an alkyl ester. Results
for these substrates appear in Table 2.
One first notes that methylene proton exchange is accel-
erated relative to the dialkyl PA esters of Table 1.
Replacement of the Me or Et phosphoryl ester residues
of substrates 3 by the inductively withdrawing Ph or
PNP residues of substrates 4 enhances base catalyzed
D/H exchange by factors up to 10.8 (4c versus 3c), with
larger enhancements induced by the more strongly
withdrawing PNP residue (4c,4d versus 4a,4b).
PA diesters are less hydrolytically reactive toward M4+
than their phosphonoformate relatives: under otherwise
similar conditions, the former require reaction tempera-
tures of 37°C to produce hydrolytic rates comparable to
those attained by phosphonoformate analogues at
25°C. Conceivably, electrostatic repulsions between the
adjacent PꢀO and CꢀO moieties of the phosphonofor-
mate diesters activates them toward esterolytic attack
by OH (supplied either as free OH− or as M-OH in
metal ion mediated attacks6,7), relative to PA diesters,
where PꢀO and CꢀO are separated by a methylene
spacer.
Base-catalyzed hydrolyses of 4 at pD 13.2, 25°C, occur
with preferential C-OR scission instead of P-OPh or
even P-OPNP cleavage. The recalcitrance of phospho-
rolysis, relative to C-OR esterolysis, is reminiscent of
phosphonoformate chemistry, where basic hydrolysis of
1 (R1=Ph, R2=Me) at pH 8.4 occurs only with
C-OMe cleavage (k=3×10−7 s−1), rather than P-OPh
cleavage.5 Similarly, with 1 (R1=R2=Ph), C-OPh
hydrolysis is observed at pH 11.7–13.4.5
In summary, phosphonoacetate diesters 3 and 4 exhibit
base catalyzed D/H proton exchange and C-OR esterol-
ysis, as well as (acidic) hydrolyses mediated by Th4+
and Zr4+. However, esterolytic chemoselectivity on the
part of the M4+ cations is not as clearly expressed with
the phosphonoacetates as with comparable phosphono-
formate substrates.
Although PA substrates 4 undergo C-OR hydrolysis in
preference to P-OAr cleavage, the Ar groups do exert
an activating effect. Comparison of Tables 1 and 2
reveals that the PNP and Ph substituents of 4 accelerate
C-OR hydrolysis by factors up to ꢀ3 (4d versus 3d);
the electron withdrawing inductive properties of PNP
or Ph mildly activate the carbonyl esters of 4 toward
hydroxide attack.
Acknowledgements
We are grateful to the US Army Research Office for
financial support. P.K.G. acknowledges support from a
NIH Biotechnology Training Grant (GM 08339).
Studies of Th4+ or Zr4+ mediated hydrolyses of 4 were
hindered by precipitation of substrate–metal ion com-