Migration of methyl groups between aliphatic amines in water

Article abstract of DOI:10.1021/ja021212u

Glycine undergoes spontaneous decarboxylation in dilute aqueous solution at elevated temperatures to form methylamine. During that process, we noticed the appearance of dimethylamine and trimethylamine in smaller amounts that increased gradually with time. These observations suggested the existence of disproportionation reactions of methylamines in water, for which there appears to be no direct precedent in the literature. Every member of the methylamine series is found to yield other members of the methylamine series. When the total concentration of amine was held constant and the rate of reaction was examined as a function of changing pH using the amine itself as the buffer, the initial rate of appearance of the products was found to reach a maximum when the conjugate acid and the conjugate base were present at equivalent concentrations. Near this equivalence point, the rate of reaction varied with pH as expected for a second-order reaction between the protonated and the unprotonated species. Under similar conditions, methyl groups were also found to migrate between the nitrogen atoms of N,N-dimethyl-1,3-propanediamine in a first-order process. With dimethylamine as a common acceptor, tetramethylammonium ion at ambient temperature. Copyright

Full text of DOI:10.1021/ja021212u

Published on Web 12/14/2002
Migration of Methyl Groups between Aliphatic Amines in Water
Brian P. Callahan and Richard Wolfenden*
Department of Biochemistry and Biophysics, UniVersity of North Carolina, Chapel Hill, North Carolina 27599-7260
Received September 26, 2002 ; E-mail: water @med.unc.edu
Glycine undergoes spontaneous decarboxylation in dilute aqueous
solution at elevated temperatures to form methylamine. During
Scheme 1
1
that process, we noticed the appearance of dimethylamine and
trimethylamine in smaller amounts that increased gradually with
time. These observations suggested the existence of disproportion-
ation reactions of methylamines in water, for which there appears
2
to be no direct precedent in the literature. Here, we show that
methyl groups migrate between aliphatic amines when they are
incubated with their conjugate acids at elevated temperatures and
that competition by water as a methyl acceptor is negligible. Half-
titrated dimethylamine (0.02 M), for example, yields trimethylamine
and methylamine in equal amounts (Scheme 1a), with no appearance
of methanol. In addition to the apparent novelty of these reactions,
these results are of interest in relation to the mechanism of methyl
transfer in biological systems.
To determine the rates of methyl transfer between amines, we
incubated aqueous amines, half-titrated with HCl, for various time
intervals at high temperatures in stainless-steel bombs lined with
PTFE After cooling, the reaction mixtures were removed and
analyzed by proton NMR, using added pyrazine as an integration
standard. Every member of the methylamine series was found to
yield other members of the methylamine series. At each of a series
of temperatures, the rate of disappearance of the starting material
followed satisfactory first-order kinetics and yielded a linear
Arrhenius plot when the logarithm of the initial rate of reaction
was plotted as a function of the reciprocal of the absolute
temperature. When the ratio of the protonated to the unprotonated
species of mono-, di-, or trimethylamine was held constant but the
total concentration of that amine varied, the initial rate of disap-
pearance of the starting material was found to vary in proportion
to the square of its concentration, as expected for a bimolecular
reaction involving two molecules of amine (see for example Figure
Table 1. Extrapolated Rate Constants and Activation Parameters
from Arrhenius Plots, Based on 10 or More Rate Constants
q
Obtained over a Range >50 °C, with Estimated Errors in ∆H and
q
T∆S of (1.5 kcal/mol
q
q
k (25 °C)
∆H
,
T∆S
,
(M- s-1
1
)
kcal/mol kcal/mol
-13
-12
dimethylamine + dimethylammonium
N,N′-dimethyl-1,3-propanediamine (s-1)
4 × 10
5 × 10
25.9
30.4
30.1
22.2
-8.5
-2.5
-3.4
-5.9
-
-
12
8
dimethylamine + tetramethylammonium 1.9 × 10
dimethylamine + trimethylsulfonium
1.5 × 10
became half-maximal at the pH value where dimethylamine is half-
converted to its uncharged form, approaching a constant value when
+
4
dimethylamine is fully converted to its uncharged form but Me N
retains its positive charge (Figure 1c). Activation parameters based
on the resulting Arrhenius plots are shown in Table 1.
Under similar conditions, methyl groups were also found to
migrate between the nitrogen atoms of N,N′-dimethyl-1,3-propane-
diamine in a first-order process, forming an equilibrium mixture
containing nearly equivalent concentrations of 3-(dimethylamino)-
propylamine and N,N′-dimethyl-1,3-propanediamine. At amine
concentrations up to ∼1 M, this reaction (and its reverse) proceeded
more rapidly than the bimolecular reactions described above, due
to a more favorable entropy of activation (Table 1). This intramo-
1a).
When the total concentration of amine was held constant, and
the rate of reaction was examined as a function of changing pH
3
using the amine itself as the buffer, the initial rate of appearance
of the products was found to reach a maximum when the conjugate
acid and the conjugate base were present at equivalent concentra-
tions. Near this equivalence point, the rate of reaction varied with
pH as expected for a second-order reaction between the protonated
and the unprotonated species, as shown by the solid line in Figure
q
lecular reaction exhibits a larger ∆H that may reflect ring strain
in the six-membered cyclic transition state.6
1
a. At pH values much below this equivalence point, product
Biological reactions that involve the transfer of acyl or glycosyl
groups often involve the formation of discrete intermediates that
differ substantially in structure from the substrates and products.
Such reactions can be accelerated by enzyme active sites that have
or can easily adopt structures with binding determinants that are
complementary to the structures of intermediate forms of the
substrate that approach the transition state in structure, forming new
H-bonds and other polar interactions that were not present in the
formation continued at a much slower rate (Figure 1b). The
persistence of reactivity at low pH, which has also been reported
for other transalkylation reactions,4 suggests that water may be
,5
acting as a general base catalyst.
+
In experiments in which the tetramethylammonium ion (Me
4
N )
was employed as the methyl donor and dimethylamine was the
acceptor, trimethylamine was formed at an initial rate proportional
to the concentrations of each of these two reactants. When this
reaction was followed as a function of increasing pH, using
7
ground-state enzyme-substrate complex. Lacking such discrete
intermediates, methyl-transfer reactions may be relatively difficult
3
8,9
5
dimethylamine itself as a buffer, its initial rate of disappearance
to catalyze except by desolvation and approximation effects that
310
9
J. AM. CHEM. SOC. 2003, 125, 310-311
10.1021/ja021212u CCC: $25.00 © 2003 American Chemical Society

C O MMU N I C A T I O N S
donors to a common acceptor in dilute aqueous solution. With
dimethylamine as the acceptor, trimethylsulfonium ion was found
4
to be ∼10 -fold more reactive than the tetramethylammonium ion
at ambient temperature (Table 1). When the second-order rate
-
8
-1 -1
constant for this trimethylsulfonium reaction (1.5 × 10
M
s )
is compared with the second-order rate constant (kcat/K ) for reaction
m
of histamine with the SAM complex of guinea pig histamine
5
-1 -1 12
N-methyl transferase (∼7 × 10 M
s ), this enzyme is seen to
enhance the rate of reaction with a nitrogen nucleophile by a factor
13 13
of 5 × 10 . Catechol O-methyltransferase has been shown to
enhance the rate of methyl transfer from SAM to an oxygen
1
7 5
nucleophile by an even greater factor, approximately 10 . Thus,
enzymes reinforce the inherently superior reactivity of sulfonium
compounds, making large contributions to the reactivity of SAM
as judged by the rate enhancements that they produce.
Acknowledgment. We thank Mark G. Snider for carrying out
preliminary experiments on methyl transfer between amines. This
work was supported by Grant GM-18325 from the National
Institutes of Health.
References
(
(
1) Snider, M. J.; Wolfenden, R. J. Am. Chem. Soc. 2000, 122, 11507.
2) Under driving conditions, in the absence of water, Raney nickel catalyzes
alkyl transfer between primary amines at temperatures between 130 and
200 °C (Winans, C. F.; Adkins, H. T. J. Am. Chem. Soc. 1932, 54, 306),
and the anion of benzylamine displaces ammonia from neat benzylamine
at 88 °C (Baltzly, R.; Blackman, S. W. J. Org. Chem. 1963, 28, 2405).
3) The horizontal coordinate of Figure 1, b and c, shows pH values
determined at room temperature with a glass electrode. The theoretical
lines in Figure 1 are based on the fraction of titratable amine that was
present in protonated form at the outset of the experiment. In no case did
the pH value of the reaction mixture, measured at room temperature before
and after the reaction, change by more than 0.4 units during the interval
over which the initial rate of reaction was determined.
(
(
(
(
4) Swain, C. G.; Kuh, D. A.; Schowen, R. L. J. Am. Chem. Soc. 1965, 87,
1553.
5) Mihel, I.; Knipe, R. G.; Coward, J. K.; Schowen, R. L. J. Am. Chem.
Soc. 1979, 101, 4349.
6) For a discussion of methyl transfer in larger ring systems, see King, J. F.;
McGarrity, M. J. J. Chem. Soc., Chem. Commun. 1982, 175.
Figure 1. (a) Initial rate of conversion of half-titrated dimethylamine to
methylamine and trimethylamine at 226 °C, plotted as a logarithmic function
of changing initial concentration of dimethylamine. The line (slope ) 2) is
calculated for a reaction that is of the second order with respect to the total
concentration of amine (Scheme 1a). (b) Initial rate of the same reaction at
(7) Wolfenden, R.; Snider, M. J. Acc. Chem. Res. 2001, 34, 938.
(8) Tang, K.-C.; Pegg, A. E.; Coward, J. K. Biochem. Biophys. Res. Commun.
1980, 96, 1371.
(
9) Craig, S. L.; Brauman, J. I. J. Am. Chem. Soc. 1999, 121, 6690. In the
biosynthesis of methionine, the methyl donor may be either betaine
2
26 °C, at a fixed total concentration of dimethylamine (0.5 M), plotted as
5
(
Garrow, T. A. J. Biol. Chem. 1996, 271, 41) or, more commonly, N -
3
a function of changing pH. The line represents the behavior expected for
the reaction shown in Scheme 1a. (c) Initial rate of reaction of dimethylamine
tetrahydrofolate (for a recent review, see Matthews, R. G. Acc. Chem.
Res. 2001 34, 681).
(
0.6 M) with tetramethylammonium chloride (0.25 M) at 171 °C, plotted
(10) Cantoni, G. Annu. ReV. Biochem. 1975, 44, 435.
(11) For reactions of the trimethylsulfonium ion with trimethylamine see
Hughes, E. D.; Wittingham, D. J. J. Chem. Soc. 1960, 806; with oxygen
nucleophiles see: Swain, C. G.; Taylor, L. J. J. Am. Chem. Soc. 1962,
4, 2456; Coward, J. K.; Sweet, W. D. J. Org. Chem. 1971, 36, 2337;
Knipe, J. A.; Coward, J. K. J. Am. Chem. Soc. 1979, 101, 4339).
12) This estimate is based on the specific activity and K value of histamine
3
as a function of effective pH. The line represents the behavior expected
for the reaction shown in Scheme 1c.
8
place the reactants in a position conducive to reaction. There are
probably limits to the rate enhancements that can be achieved by
these “physical” effects, so that the inherent reactivity of the
substrate is of special importance.
(
m
reported for guinea pig histamine N-methyltransferase (Brown, D. D.;
Tomchick, R.; Axelrod, J. J. Biol. Chem. 1959, 234, 2948) and a molecular
mass of 29 000 reported for the monomeric enzyme (Matuzewska, B.;
Borchardt, R. T. J. Neurochem. 1983, 271, 37, 22831), assuming that the
monomer contains a single active site.
If ammonium ions are capable of methyl transfer in water, then
why are reactions that involve methyl transfer from ammonium
ions9 less common than reactions using S-adenosylmethionine
(13) It may be worth noting that the rate enhancement produced by histamine
N-methyltransferase is similar, both numerically and in its detailed
thermodynamic origins, to the rate enhancement produced by yeast
hexokinase, whose action is probably based on similar principles (Sigmon,
K. A.; Wolfenden, R., unpublished observations).
10
(SAM) as a methyl donor? To make a direct comparison of the
inherent reactivities, we determined the rates of reaction of the
trimethylsulfonium ion11 with the tetramethylammonium ion as
JA021212U
J. AM. CHEM. SOC.
9
VOL. 125, NO. 2, 2003 311