Iyer and Hengge
34
Kinetics Studies of S-3′mNB. Kinetic studies were carried out
at 80 °C from pH 9.0-13.0 buffered with CHES (pH 9.0) and
CAPS (pH 10.0, 10.5, 11.0, 11.5). For pH 12.0 and 13.0, 0.01 and
0.1 N NaOH were used, respectively. Constant ionic strength (I )
1.0) was maintained using KCl. The reaction progress was
monitored using 31P NMR. In a typical experiment, 100 µmol of
S-3′mNB and an equivalent amount of integration standard (benzyl
phosphonic acid) were dissolved in 500 µL of buffer and allowed
to react in an NMR tube immersed in an oil bath maintained at 80
°C. The NMR tube was cooled, and 31P NMR was recorded at
appropriate intervals. Reaction was allowed to progress until three
half-lives. At pH 12 and 13, the experiment was carried out in the
NMR with the probe temperature set to 80 °C and spectra
automatically recorded at fixed time intervals. The rate constant
(k) for the disappearance of S-3′mNB was determined at each pH
from the first order loss of the reactant. A graph of log k as a
function of pH appears in the Supporting Information.
mechanism might result in a very small klg; since bonds to
phosphorus in pentacoordinate phosphoranes are slightly longer
than phosphate esters, modest weakening of the P-S bond is
expected. If pseudorotation is needed to bring sulfur into an
apical position this should not have a large effect on the leaving
group KIE, since whether pseudorotation is required or not, the
KIEs on the reaction in such a mechanism will be those on
phosphorane formation as long as all subsequent steps are rapid.
While the data point to a highly associative transition state, the
data from this study are not sufficient to distinguish between a
concerted reaction with a phosphorane-like transition state and
a two-step mechanism for S-5′mNB.
Conclusions
Sulfur at the 2′ position in the conformationally flexible model
S-2′pNP is completely dysfunctional as a nucleophile toward
the adjacent phosphorus center. In spite of the presence of the
labile p-nitrophenyl leaving group, the thiolate reacts exclusively
with the neighboring carbon. Thus, the ribose ring is crucial to
position a 2′ thiolate in a suitable geometry to force nucleophilic
attack on the phosphorus center in 2′-thio nucleotides.
The combination of aryl leaving groups and sulfur at either
bridging position proved unfavorable to the stability of these
compounds. Aryl leaving groups activate the phosphorus center
such that these compounds undergo either spontaneous isomer-
ization or hydrolytic cleavage, depending on the location of the
sulfur atom. Sulfur at the bridging positions in models with alkyl
leaving groups exhibited much greater stability compared to their
aryl analogues. The overall cleavage mechanisms of these model
compounds are essentially similar to those of the ribose-
containing systems. The conformationally flexible models in this
study react substantially slower than ribose compounds with
sulfur substitution at equivalent positions. Kinetic isotope effects
reveal that when sulfur is in the 3′ position, the transition state
is very early, with bond formation to the nucleophile and leaving
group bond fission both only modestly advanced. With sulfur
in the scissile 5′ position, KIEs indicate a highly associative
transition state.
Kinetic Studies of S-5′mNB. Compound S-5′mNB was allowed
to react over the pH range 8.0-12.0 buffered with TRIS-HCl (pH
8.0), CHES (pH 9.0), and CAPS (pH 10.0, 10.5, 11.0, 11.5). For
pH 12.0, 0.01N NaOH was used. Constant ionic strength (I ) 1.0)
was maintained using KCl. The reaction progress was monitored
using 31P NMR. In a typical experiment, 100 µmol of S-5′mNB
and an equivalent amount of integration standard (methyl phos-
phonic acid) were dissolved in 500 µL of buffer and allowed to
react in an NMR tube immersed in an oil bath maintained at 80
°C. The NMR tube was cooled, and 31P NMR was recorded at
appropriate intervals. Reaction was allowed to progress until 3 half-
lives. At pH 11.0, 11.5, and 12.0, the experiment was carried out
in the NMR with the probe temperature set to 80 °C and spectra
automatically recorded at fixed time intervals. The rate constant
(k) for the disappearance of S-5′mNB was determined at each pH
from the first order loss of the reactant. A graph of log k as a
function of pH appears in the Supporting Information.
Kinetic Isotope Effect Studies. The isotope effects were
measured by the competitive method using an isotope ratio mass
spectrometer (IRMS) to measure the change in isotopic composition
over the course of the reaction. The 34S isotope effect was measured
using natural abundance compound. The 18O KIEs were measured
using the double label method,42 in which a nitrogen atom in the
departing ester group was used as a reporter for 18O ratios at the
positions of interest. This methodology has been used in numerous
previous KIE studies involving phosphate esters.10 A description
of the isotopic isomers needed for the measurement of the 18O KIEs
and their synthesis can be found in the Supporting Information.
Measurement of Kinetic Isotope Effects. The isotope effects
for S-3′mNB and S-5′mNB were measured under similar conditions
at pH 10.5. The absence of an effect of buffer concentration on the
rate demonstrated that the reactions proceed by specific base
catalysis (data not shown). The leaving group in S-3′mNB is
mNBA, and in the case of S-5′mNB it is the m-nitrobenzylthiolate
anion, which subsequently oxidizes to the symmetrical dimer,
m-nitrobenzyl disulfide, under the reaction conditions. Thus, for
KIEs on reactions of S-3′mNB, isotopic analysis was carried out
on isolated mNBA, while for S-5′mNB the disulfide product was
isolated and analyzed for either sulfur isotopic analysis or nitrogen
isotopic analysis.
Experimental Section
Reactivity of S-2′pNP as Determined by 31P NMR. Compound
S-2′pNP was allowed to react over the pH range 7.0-10.5 buffered
with MOPS (3-(N-morpholino)propanesulfonic acid) (pH 7.0, 7.5),
TRIS (trishydroxymethylaminomethane)-HCl (pH 8.0, 8.5), CHES
(3-cyclohexylaminoethylsulfonic acid) (pH 9.0, 9.5), and CAPS (3-
[cyclohexylamino]-1-propanesulfonic acid) (pH 10.0, 10.5). A 100
µmol portion of S-2′pNP was dissolved in 500 µL of buffer and
allowed to react at rt. The reaction progress was monitored by 31
P
NMR at 162 MHz. The reactant (δ ) -4.7) was converted cleanly
into a single phosphorus-containing product that had a 31P chemical
shift identical to pNPP. The identity of the product was confirmed
by spiking the NMR tube after reaction was complete with authentic
pNPP. The externally added pNPP overlapped with the 31P and
KIE Measurements for S-3′mNB. S-3′mNB (100 µmol) was
dissolved in 0.25 M CAPS buffer at pH ) 10.5. The solution
temperature was maintained at 80 °C in a block heater. Reactions
were run in triplicate and were allowed to proceed to 50%
completion. The reaction progress was monitored by 31P NMR.
Once the desired amount of hydrolysis was reached, the progress
of reaction was stopped by cooling in an ice bath. The solution
was extracted three times with diethyl ether (20 mL) to separate
the mNBA from the remaining reaction mixture. The ether from
all three extractions was pooled together and dried over MgSO4.
1
the H NMR signals of the reaction product.
To obtain a quantitative estimate of S-2′pNP reactivity, the half-
life for the conversion of S-2′pNP to pNPP was determined at pH
7.0 by 31P NMR. A 100 µmol portion of S-2′pNP was dissolved
in 500 µL of 0.5 M MOPS buffer along with an equivalent amount
of integration standard (methylphosphonic acid) and allowed to react
at rt. The 31P NMR was recorded at an interval of 2.5 min until the
peak for S-2′pNP had disappeared completely. A t1/2 of ap-
proximately 15 min was obtained for the conversion of S-2′pNP
to pNPP under these conditions.
4828 J. Org. Chem. Vol. 73, No. 13, 2008