Evidence for the Formation of α-Hydroxydialkylnitrosamines in the pH-1 dependent Solvolysis of α-(Acyloxy)dialkylnitrosamines

Article abstract of DOI:10.1021/jo962003t

A kinetic study of the decay of α-(acyloxy)dialkylnitrosamines in aqueous solutions, at 25°C, ionic strength 1 M (NaClO₄), and 4% acetonitrile by volume, is reported. Rate constants for disappearance of the N-NO chromophore or appearance of benzaldehyde product increase with the introduction of electron-withdrawing groups into the acyloxy moiety. For some compounds, in some regions of pH, the kinetics are non-first-order. For these compounds, in other regions of pH, the rate constants are coincident with those previously reported for the decay of the corresponding α-hydroxydialkyl- nitrosamines. These data provide direct evidence that α-hydroxydialkylnitrosamines are intermediates in the pH-independent decomposition of the α-(acyloxy)dialkylnitrosamines studied.

Full text of DOI:10.1021/jo962003t

2500
J . Org. Chem. 1997, 62, 2500-2504
Evid en ce for th e F or m a tion of r-Hyd r oxyd ia lk yln itr osa m in es in
th e p H-In d ep en d en t Solvolysis of r-(Acyloxy)d ia lk yln itr osa m in es
Latifa Chahoua, Milan Mesic´, Cynthia L. Revis, Alain Vigroux, and J ames C. Fishbein*
Department of Chemistry, Wake Forest University, Winston-Salem, North Carolina 27109
Received October 25, 1996X
A kinetic study of the decay of R-(acyloxy)dialkylnitrosamines in aqueous solutions, at 25 °C, ionic
strength 1 M (NaClO4), and 4% acetonitrile by volume, is reported. Rate constants for disappearance
of the N-NO chromophore or appearance of benzaldehyde product increase with the introduction
of electron-withdrawing groups into the acyloxy moiety. For some compounds, in some regions of
pH, the kinetics are non-first-order. For these compounds, in other regions of pH, the rate constants
are coincident with those previously reported for the decay of the corresponding R-hydroxydialkyl-
nitrosamines. These data provide direct evidence that R-hydroxydialkylnitrosamines are intermedi-
ates in the pH-independent decomposition of the R-(acyloxy)dialkylnitrosamines studied.
Sch em e 1
In tr od u ction
R-(Acyloxy)dialkylnitrosamines, 1 (Scheme 1), have
been widely employed in studies of nitrosamine carcino-
genesis.1 A number of these compounds have been shown
to be mutagenic or carcinogenic. They are believed to
be sources of R -hydroxydialkylnitrosamines, 2 (Scheme
1). As in Scheme 1, R-hydroxydialkylamines are the
likely products of enzymatic R-hydroxylation of dialkyl-
nitrosamines 3, and they decompose to yield electrophilic
species that are purportedly responsible for the alkylating
activity, and thus biological activity, of dialkylnitros-
amines.
The mechanism of solvolytic decay of simple R-(acyl-
oxy)dialkylnitrosamines was recently reexamined.2 In
the pH range between pH ) 3-13 it is dominated by the
two-term rate law of eq 1. It was concluded that the pH
independent reaction involves the formation of nitros-
the suggestion4 that R-(acyloxy)dialkylnitrosamines de-
compose with “anchimeric assistance” by the nitroso
group. There is evidence consistent with the formation
of R-hydroxydialkylnitrosamines in the base-catalyzed
reaction, kOH (eq 1), for which a carbonyl attack mecha-
nism was originally proposed, and the esterase-catalyzed
hydrolysis, another carbonyl attack mechanism.5 Thus,
it was shown that the stereochemical course of formation
of 1-phenylethanol was similar in the hydrolysis of (-)-
1-(acetoxymethyl)(1-phenylethyl)nitrosamine both in aque-
ous buffered solution, pH 8.5, and in the hog liver
esterase-catalyzed reaction. While the rate constants k1
and kOH are not known for this substrate, it seems likely
by comparison with similar systems that the nonenzy-
matic hydrolysis at pH ) 8.5 involves predominantly the
base-catalyzed reaction (kOH, eq 1).
kobsd ) k1 + kOH[OH-]
(1)
iminium ion intermediates, as in eq 2. It was also shown
that these can be trapped by added nucleophiles, and it
has been explicitly assumed that they react with solvent
A test for mechanisms that require the intermediacy
of R-hydroxydialkylnitrosamines involves the generation
of the R-hydroxydialkylnitrosamines from highly reactive
esters with good leaving groups. In the general mecha-
nism of eq 3, under the condition in which kA is fast
relative to kB, it is predicted that the esters should
water to yield R-hydroxydialkylnitrosamines, as in eq 2.3
There is in fact no evidence that R-hydroxy-N,N-
dialkylnitrosamines are intermediates in the pH inde-
pendent (ko, eq 1) solvolysis of R-(acyloxy)dialkylnitros-
amines. The notion that they might not be arises from
X Abstract published in Advance ACS Abstracts, March 15, 1997.
(1) Lawley, P. D. In Chemical Carcinogens; Searle, C. D., Ed.; ACS
Monograph 182, American Chemical Society: Washington, DC, 1984.
Lijinsky, W. Chemistry and Biology of N-Nitroso Compounds; Cam-
bridge University Press: Cambridge, 1992.
decompose more rapidly than the R-hydroxydialkylnit-
rosamines such that the latter become non-steady-state
intermediates. The lifetimes of a few R-hydroxydialkyl-
(2) Revis, C.; Rajama¨ki, M.; Fishbein, J . C. J . Org. Chem. 1995, 60
7733.
(3) Rajama¨ki, M.; Vigroux, A.; Chahoua, L.; Fishbein, J . C. J . Org.
Chem. 1995, 60, 2324.
(4) Wiessler, M. N-Nitrosamines; Anselme, J .-P., Ed.; American
Chemical Society: Washington, DC, 1979; p 57.
(5) Gold, B.; Linder, W. B. J . Am. Chem. Soc. 1979, 101, 6772.
S0022-3263(96)02003-8 CCC: $14.00 © 1997 American Chemical Society

pH-Independent Solvolysis of R-(Acyloxy)dialkylnitrosamines
J . Org. Chem., Vol. 62, No. 8, 1997 2501
nitrosamines were reported some years ago,6 and we have
recently reported the exalted reactivity of some others.7
In every case these compounds are substantially more
reactive than the acetate ester precursors so that they
are never more than steady state intermediates in the
decay of the R-acetoxydialkylnitrosamines. However,
with better leaving groups, for which kA increases while
kB remains constant, it should be possible to reach the
point at which kA > kB.
This report summarizes results of mainly kinetic
experiments involving the compounds 4-6. This study
indicates that there is a large dependence of R-(acyloxy)-
dialkylnitrosamine reactivity upon the pKa of the conju-
Results and Discussion sections. The pH values were obtained
after the kinetic runs using a Corning Model 501 pH meter
with attached combination electrode. Two point calibrations
were done before recording pH values. Calibrations were
carried out using commercially available standards or stan-
dards prescribed by the Merck Index, 8th edition.8
Syn th esis. r-Acetoxydialkyln itr osam in es 4a-6a. These
were prepared essentially as reported previously2 from the
appropriate imine (or cyclic trimer of the imine) and nitroso-
nium tetrafluoroborate, in the presence of triethylammonium
acetate. Physical data for 4a and 5a were reported previously.
6a : 1H NMR (CDCl3) δ (E) isomer (96%): 8.32 (1 H, s), 7.42
(5 H, s), 2.82 (3H, s), 2.26 (3H, s); (Z) isomer (4%): 3.56 (3H,
s), 2.19 (3H, s). Anal. Calcd: C, 57.69; H, 5.81; N, 13.45.
Found: C, 58.34; H, 5.85; N, 13.2.
r-(Ch lor oa cetoxy)d ia lk yln itr osa m in es 4b-6b. These
were prepared in a manner analogous to that for the acetates,
but the electrophilic addition was carried out at 0-5 °C.
Subsequent to addition of the NOBF4, the reaction was stirred
an additional 20 min and was washed in a separatory funnel
three times with an equal volume of water each time. The
organic layer was dried over MgSO4 and filtered, and the
methylene chloride was removed by evaporation with argon.
The crude material was freed of aldehyde by washing with
pentane. The compounds 4b and 5b were purified by chro-
matography on silica with ether/hexane (1/5) and ether,
respectively, used as solvents. The compound 6b was recrys-
tallized from pentane/methylene chloride (9/1). 4b: 1H NMR
(CDCl3) δ: (E) isomer (88%): 0.90 (t, 3H); 1.28, (m, 2H); 1.45,
(m, 2H); 3.55, (t, 2H); 4.10, (s, 2H); 6.25 (s, 2H); (Z) isomer
(12%): 1.00, (t, 3H); 1.45, (m, 2H); 1.78, (m, 2H); 4.30, (t, 2H);
4.05, (s, 2H); 5.42, (s, 2H). 5b: 1H NMR (CDCl3) δ 1.09 (3H,
t), 1.51 (3H, d), 3.59 (2H, m), 4.07 (2H, s), 7.21 (4 H, q). 6b:
1H NMR (CDCl3) δ (E) isomer (96%): 8.39 (1H, s), 7.45 (5 H,
s), 4.23 (2H, s), 2.83 (3H, s); (Z) isomer (4%): 3.60 (3H, s) ppm.
Anal. Calcd: C, 49.5; H, 4.6; N, 11.55. Found: C, 49.74; H,
4.63; N, 11.62.
r-(Dich lor oa cetoxy)d ia lk yln itr osa m in es 4c, 6c. A mix-
ture of dichloroacetic acid (3.5 mmol) and dry triethylamine
(3.5 mmol) in 5 mL of dry methylene chloride was added at
-10 °C to a stirred solution of imine (3.5 mmol) in 15 mL of
dry methylene chloride. NOBF4 (1.1 equiv) was added with a
spatula slowly. The solution bubbled and turned yellow-
orange. The clear solution was stirred for 10 min and was
washed rapidly three times with cold water. The organic layer
was dried over MgSO4 and filtered, and the methylene chloride
was evaporated off under an argon stream. The crude material
was washed with pentane to get rid of aldehyde and was stored
over dry ice. 4c: 1H NMR (CDCl3), (E) isomer (87%): δ 0.90
(t, 3H); 1.28 (m, 2H); 1.45 (m, 2H); 3.60 (t, 2H); 6.00 (s, 1H);
6.35 (s, 2H); (Z) isomer (13%): δ 1.00 (t, 3H); 1.45 (m, 2H);
1.78 (m, 2H); 4.30 (t, 2H); 5.50 (s, 2H); 5.94 (s, 1H). 6c: 1H
NMR (CDCl3) δ 8.36 (1H, s); 7.44 (5H, s); 6.12 (1H, s); 2.83
(3H, s).
gate acid of the leaving group acetate anion. Thus
R-(acyloxy)dialkylnitrosamines have been synthesized
that are more reactive than the reactivity reported for
the corresponding R-hydroxydialkylnitrosamines. A di-
rect test of the mechanism of eq 3 is afforded, and the
results require the intermediacy of R-hydroxydialkylni-
trosamines in the decay of R-(acyloxy)dialkylnitrosamines
in aqueous media.
Exp er im en ta l Section
Wa r n in g! Many R-acetoxydialkylnitrosamines are proven
to be powerful direct acting carcinogens! Procedures must be
carried out with due precautions. Manipulations were carried
out by personnel wearing frequently-changed double pairs of
disposable gloves and in a well-ventilated hood. Contaminated
and potentially-contaminated materials were treated with 50%
aqueous sulfuric acid containing the commercially available
oxidant “No Chromix” (Aldrich Chemical).
Ma ter ia ls. Organic solvents were purified by distillation
before use. Organic chemicals for synthesis were ACS grade
or better. Inorganic chemicals were ACS grade or better and
were used without further purification. Water was distilled
in glass.
Kin etics. All kinetic runs were carried out using a Hewlett
Packard 8452A diode array spectrophotometer or an Applied
Photophysics DX17MV stopped-flow spectrophotometer, both
thermostated at 25 °C by a circulating water bath. Kinetic
runs were initiated when solutions containing esters dissolved
in acetonitrile were injected into the cuvettes or injected
mechanically into the observation cell of the stopped-flow
instrument, to give final substrate concentrations of 1-2 ×
10-4 M. The final acetonitrile concentration was 4% by
volume. Routinely, values of kobsd were calculated from ab-
sorbance decay or appearance data using a commercially
available fitting program (Enzfitter) or the software routine
on the stopped-flow. In cases where non-first-order kinetics
were observed, the data were handled as described in the
Resu lts
For compounds 4a -c and 5a ,b, the kinetics of disap-
pearance of the NsNdO chromophore were monitored
at 230 nm and exhibited good first-order behavior over
most of the range of pH ) 2-13 (see exceptions below).
In general, changes in buffer concentration from 0.05-
0.30 M did not have much effect on the value of kobsd with
typical increases being 10-25% for variation over this
concentration range. Changes in bicarbonate buffer
concentrations (at pH > 9) effected the largest increases
in kobsd, these being as much as 50% increases, at ∼0.2
M buffer, above the value of kobsd extrapolated to a buffer
concentration equal to zero.
(6) Mochizuki, M.; Anjo, T.; Okada, M. Tetrahedron Lett. 1980, 21,
3693. Okada, M.; Mochizuki, M.; Anjo, T.; Sone, T.; Wakabayashi, Y.;
Suzuki, E. IARC Sci. Publ. 1980, 31, 71. Mochizuki, M.; Anjo, T.;
Takeda, K.; Suzuki, E.; Sekiguchi, N.; Huang, G. F.; Okada, M. IARC
Sci. Publ. 1982, 41, 553.
For compounds 6a -c, the kinetics of appearance of the
product benzaldehyde were monitored at 254 nm and
(7) Mesic´, M.; Revis, C.; Fishbein, J . C. J . Am. Chem. Soc. 1996,
118, 7412.
(8) The Merck Index, 8th ed.; Stecher, P. G., Ed., Merck & Co.:
Rahway, NJ , 1968.

2502 J . Org. Chem., Vol. 62, No. 8, 1997
Chahoua et al.
F igu r e 2. Plots of absorbance against time for the decay of
6c, in panel A, and 5b, in panel B, under the reaction
conditions described in Figure 1. Fits to the data were obtained
using eqs 6 or 7 and the parameters given in the text.
Ta ble 1. Yield s of Azid e Ad d u ct a n d Va lu es of kobsd in
th e Deca y of 6b a s a F u n ction of Azid e Ion
Con cen tr a tion in Aqu eou s Solu tion , 25 °C, 4%
Aceton itr ile, Ion ic Str en gth 1 M (Na ClO4), 0.05 M
Ca cod ylic Acid Bu ffer , 50% An ion
F igu r e 1. Plots of log ko, the buffer independent rate
constants for decay or increase of absorbance in the decom-
position of R-(acyloxy)dialkylnitrosamines (solid symbols) and
R-hydroxydialkylnitrosamines (open symbols), against pH in
aqueous solutions: 25 °C, ionic strength 1 M (NaClO4), 4%
acetonitrile by volume. Panel A, compounds 6: (b), a ; (9), b;
(2), c; and (4), the corresponding R-hydroxy derivative. Panel
B, compounds 5: (b), a ; (9), b; and (O), the corresponding
R-hydroxy derivative. Panel C, compounds 4: (b), a ; (9), b;
(2), c; and (O), the corresponding R-hydroxy derivative.
azide ion
concentration (M)
kobsd
(s-1
azide adduct
% yield
)
0
0.042
0.043
0.043
0.043
0.042
0
42
80
99
102
0.0002
0.001
0.005
0.008
In the case of 4c, 5b, and 6c the kinetics were non-
first-order in narrow regions of the pH range studied. The
ranges were 4c, pH ∼ 6-7.5; 5b, pH ∼ 5.5-7; and 6c,
pH ∼ 4.5-6. Examples of the data for such runs for
compounds 4c and 6c are given in Figure 2.
The yields of azide ion adduct as a function of azide
ion concentration in the reaction of 6b are summarized
in Table 1. Also in Table 1 are rate constants for the
first-order appearance of benzaldehyde in solutions of
varying azide ion concentration determined spectropho-
tometrically at 250 nm.
exhibited good first-order behavior over most of the range
of pH between pH ) 2-13 (see exceptions below). The
effects of changing buffer concentration on the value of
k
obsd were similar to those described for compounds 4 and
5 above. Slightly larger effects with bicarbonate buffers
were observed for compounds 6 with <150% increases,
at 0.2 M buffer, above the value of kobsd extrapolated to
zero buffer concentration being typical.
Values of ko, the buffer independent rate constant for
either disappearance of the NsNdO chromophore (com-
pounds 4 and 5) or the appearance of benzaldehyde
(compounds 6), were obtained as the y-intercept values
of the lines from plots of kobsd against buffer concentra-
tion. The slight effects of buffer concentration on the
values of kobsd (above) enabled values of ko to be obtained
with typical uncertainties of se ) (5%. Plots of log ko
against pH are presented in Figure 1.
Discu ssion
r-Hyd r oxyd ia lk yn itr osa m in e In ter m ed ia tes. We
previously studied the pH dependence of the buffer-
independent decay of some simple R-(acyloxy)dialkylni-
trosamines and observed that in the range of pH ) 3-13,

pH-Independent Solvolysis of R-(Acyloxy)dialkylnitrosamines
J . Org. Chem., Vol. 62, No. 8, 1997 2503
Ta ble 2. Ra te Con sta n ts for th e Deca y of
r-(Acyloxy)d ia lk yln itr osa m in es in Aqu eou s Solu tion s a t
25 °C, Ion ic Str en gth 1 M (Na ClO4), 4% Aceton itr ile by
Volu m e
increase in kobsd of greater than 5%. This data requires
that the azide adduct is formed in a reaction subsequent
to the rate-limiting step, presumably involving capture
of the N-nitrosiminium ion formed in the rate-limiting
step of the k1 process, as in eq 2.3,10
a
compound
k1,a s-1
kOH
,
M-1 s-1
4a b
4b
4c
2.0 × 10-6
2.1 × 10-4
0.017
0.00103
0.112
0.00044
0.053
4.4
2.4
320
Inspection of Figure 1 shows that in the case of the
chloroacetate 5b (Figure 1B, solid squares) and the
dichloroacetates 4c (Figure 1C, solid triangles) and 6c
(Figure 1A, solid triangles) the log ko-pH profiles are
more complex. The behavior is qualitatively similar for
all three compounds and is most clearly observed for 6c
(Figure 1A, solid triangles) using the five bars in the
figure, labeled a-e, that indicate pH ranges that exhibit
distinct pH dependences. The rate constants in regions
a and c are inversely proportional to hydrogen ion
concentration while those in b and d are pH independent
and those in region e are hydrogen ion dependent.
In the pH range between b and c the kinetics of
benzaldehyde formation are non-first-order, as illustrated
in Figure 2A. Non-first-order behavior in the disappear-
ance of the N-NO chromophore in the analogous regions
of pH is observed for compounds 4c and 5b. An example
of this behavior is indicated in the case of 5b in Figure
2B. The non-first-order behavior means that the appar-
ent downward break in the pH rate profiles between
regions b and c cannot be ascribed to a change in rate-
determining step involving a steady state intermediate
which would give clean first-order kinetic behavior at
every pH.
The observation of non-first-order behavior indicates
the accumulation of a non-steady-state intermediate that
is concluded to be the R-hydroxydialkylnitrosamines. This
conclusion is based on the coincidence of the pH-rate
profiles for decay of R-hydroxydialkylnitrosamines that
were recently reported7 with parts of the pH-rate profiles
for 4c, 5b, and 6c that are reported here. The pH-rate
profiles for decay of the R-hydroxydialkylnitrosamines are
included in Figures 1A-C (open symbols). The kinetic
behavior of the R-hydroxydialkylnitrosamines accounts
qualitatively for the kinetic behavior observed in regions
c, d and e of the pH-rate profiles for benzaldehyde
formation, in the case of 6c, and for the pH-rate profiles
for N-NO group disappearance, starting from 4c and 5b.
The behavior over the complete range of pH studied can
be understood by the kinetic scheme in eq 3 in which kA
and kB are further defined as in eq 4 and eq 5.11 In
regions a and b (Figure 1A), kB . kA so that the
R-hydroxydialkylnitrosamine does not build up to any
1.25 × 104
5a b
5b
6a
3.2
87
1.3
632
2.1 × 104
6b
6c
a
As defined in eq 1. Standard errors for all entries <(10%.
b
Previously reported in ref 2.
the rate law of eq 1 was obeyed.2 It was concluded that
the pH-independent term, k1, involved rate-limiting
formation of N-nitrosiminium ions, as in eq 2, on the
basis of the near-zero values of ∆Sq, structure-activity
relations, and the trapping of putative N-nitrosiminium
ions after the rate-limiting step by azide ion to yield
adducts such as in eq 2.2,3 Others16a have previously
characterized the reaction chemistry of 6a and obtained
qualitatively similar evidence for the involvement of an
N-nitrosiminium ion. It was also concluded, in accord
with earlier suggestions, that the pH-dependent term,
k
OH, involves hydroxide ion attack at the carbonyl group.
The kinetic data for the acetate esters 5a and 4a
(Figure 1, parts B and C, respectively) were presented
as part of the previous study,2 and the kinetic data for
the acetate ester 6a , Figure 1A, also shows the biphasic
dependence indicated for the rate law of eq 1 over the
limited range studied. Fits to the rate law of eq 1, using
values for the rate constants summarized in Table 2, are
indicated by the solid lines in Figure 1.
Substitution of electron-withdrawing groups into the
acetate moiety of the esters is expected to increase the
rate constants for both the pH dependent and pH
independent reactions, and this is in fact observed. The
increases in rate constants k1 and kOH are expected due
to the improved leaving group ability, for the k1 process,
and increased electrophilicity, in the kOH process, upon
introduction of the electron-withdrawing group. In the
case of the chloroacetates 6b and 4b, the data in Figure
1 (panels A and C, respectively) show the expected
increases in k1 and kOH. In the cases of the k1 process
for 4b and 6b, the observed reactions are too fast to
involve rate-limiting reactions of the carbonyl group.
Reported rate constants for such reactions involving
substrates which should have similar reactivity are too
small to account for the observed reactivity of 4b and 6b.
Thus, the pH independent decay of ethyl chloroacetate
kA ) k1 + kOH[OH-]
(4)
(5)
kB ) k°1 + k°2/[H+] + k°3[H+]
is ∼10-7 s-1 9
. Additional evidence that there is no change
in mechanism for the pH independent reaction in the case
of 6b is found in Table 1 which indicates a quantitative
yield of the azide adduct 7 when 6b is decomposed in
extent and good first-order kinetic behavior is observed.
The upward break in the pH-rate profile between region
a and b indicates a change in mechanism from hydroxide
ion attack of the carbonyl group (kOH, region a) to pH-
independent formation of the nitrosiminium ion (k1,
region b). Best fits of the data to such a kinetic scheme
using the parameters summarized in Table 2 are indi-
cated as solid lines in the Figure 1. Throughout regions
(10) Vigroux, A.; Kresge, A. J .; Fishbein, J . C. J . Am. Chem. Soc.
1995, 117, 4433.
(11) The rate constants k°1, k°2, and k°3 used here are, respectively,
identical with k1, k2, and k3 given in ref 7. The new symbols were
employed to avoid ambiguity with other rate constant names in the
present report.
the presence of 8 mM sodium azide in the absence of an
(9) J encks, W. P.; Carrioulo, J . J . Am. Chem. Soc. 1961, 83, 1743.

2504 J . Org. Chem., Vol. 62, No. 8, 1997
Chahoua et al.
c, d, and e, kA > kB so that good first-order kinetics of
the reactions involving the R-hydroxydialkylnitrosamines
are observed and the rate law is adhered to that was
previously determined in these cases.7,11
The non-first-order kinetic behavior between regions
b and c is expected for a system such as that in eq 3 when
kA ∼ kB, and the observed non-first-order kinetic behavior
is quantitatively consistent with the interpretation in
terms of the eq 3. Kinetic expressions for the appearance
of benzaldehyde from 6c and the disappearance of the
N-NO chromophore in the case of compounds 4c or 5b
are given in eqs 6 and 7, respectively.12
This absorbance subsequently decayed to a final value
of 0.046 units. The λmax at 230 nm of the transient
remaining after >99% decay of the ester is consistent
with that expected for the NsNdO chromophore of the
R-hydroxydialkylnitrosamine.6
Effects of Lea vin g Gr ou p . The values of k1 in Table
2 indicate that the N-nitrosiminium ion-forming reaction
is quite sensitive to the basicity of the carboxylate ion
leaving group. Plots of log k1 against the conjugate acid
pKa of the leaving group have slopes, âlg, of -1.13, -1.05,
and -1.15 for compounds 4a -c, 5a -b, and 6a -c,
respectively. The values of âlg indicate that the substit-
uents detect a build-up of a full negative charge in
proceeding from the ground state to the transition state.
Such large values âlg suggest an advanced transition
state with considerable C-O bond fission, but complete
scission of the bond is not required in the absence of
knowledge of the “effective charge” on the oxygen in the
ester, which could be considerable.14
At
Bt
Abs ) Ao[1 - (1/(kB - kA)[kBe-k - kAe-k ]] + Co
(6)
At
At
Bt
Abs ) Aoe-k + Ao[kA/(kB - kA)][e-k - e-k ] + C∞
(7)
Values of kB for the decay of the R-hydroxydialkylni-
trosamines were computed from the best fit of the data
in Figure 2, parts A and B, to eqs 6 and 7 using values
of kA for the decay of the R-acyloxy compounds equal to
the values of ko in Table 2, the observed absorbance at t
) 0 (Co) or t ) ∞ (C∞), and the total change in absorbance
(Ao) from the absorbance at t ) 0 and the absorbance at
t ) ∞. This analysis yields a rate constant kB for the
R-hydroxydialkylnitrosamine derivative of 6c of 2.4 s-1
(using eq 6) while the value calculated from the rate
constants determined from direct measurement of the
decay of the R-hydroxydialkylnitrosamine, reported pre-
viously,7 yields kB ) 2.3 s-1. A similar analysis using eq
7 and the data for 5b in Figure 2B gives kB ) 0.19 s-1
for the decay of the R-hydroxydialkylnitrosamine which
compares well with the value of kobsd ) 0.19 s-1 that was
measured directly from the R-hydroxy-N,N-diethylnitro-
samine under identical conditions of pH and buffer
concentration.13
The notion that there is considerable C-O bond
cleavage in the rate-limiting step is consistent with the
progress in the transition state that is indicated by other
measures. Substantial rehybridization at the central
carbon of the penultimate iminium ion is suggested by
the secondary R-deuterium kinetic isotope effects of kH/
kD ∼ 1.15 (per hydrogen) that has recently been deter-
mined.15
Biologica l Releva n ce. For most of the compounds
reported in this study, and in the case of a number of
others that have been studied in this laboratory, the
dominant mechanism of decomposition at physiological
pH involves the pH independent (k1) process. It is
frequently assumed that R-(acyloxy)dialkylnitrosamines
are converted to R-hydroxydialkylnitrosamines via es-
terase action in vivo; however, in a number of cases,
esterases fail to accelerate the decay of R-(acyloxy)-
dialkylnitrosamines.16 In these cases, the predominant
reaction likely involves the same pH independent (k1)
mechanism and the intermediacy of N-nitrosiminium
ions and R-hydroxydialkylnitrosamines.
Ultraviolet spectral analysis of the decay of 5b is
consistent with the non-steady-state formation of the
R-hydroxydialkylnitrosamine. A solution (0.10 M meth-
oxyacetic acid buffer, 50% anion, µ ) 1, 25 °C) of 5b with
an initial absorbance at λmax (232 nm) of 0.969 was
allowed to decay for 50 s (∼8 half-lives of ester decay)
after which the absorbance at λmax (230 nm) was 0.474.
Ack n ow led gm en t. This work was supported by the
National Cancer Institute (NIH) through grants RO1
CA 52881 and KO4 CA62124.
J O962003T
(12) Espenson, J . H. Chemical Kinetics and Reaction Mechanisms;
McGraw-Hill Book Co.: New York, 1981.
(14) Williams, A. Acc. Chem. Res. 1984, 17, 425.
(15) Cai, H.; Fishbein, J . C. Unpublished results.
(16) (a) Baldwin, J . E.; Scott, A.; Branz, S. E.; Tannenbaum, S. R.;
Green, L. J . Org. Chem. 1978, 43, 2427. (b) Hecht, S. S.; Young-Sciame,
R.; Chung, F.-L. Chem. Res. Toxicol. 1992, 5, 706.
(13) Allowing both rate constants to vary in the fit, as requested by
a referee, gives kA ) 0.11 s-1 and kB ) 0.20 s-1 for 5b. The former
value is in good agreement with the experimentally determined value
(Table 2) of k1 () kA, under these conditions) ) 0.11 s-1
.