Stereochemical preference for heterochiral coupling controls selectivity in competitive peptide synthesis

Article abstract of DOI:10.1039/a901730e

Competitive activated couplings of N-phthaloyl amino acids with amino acid dimethylamides show little selectivity among substrates in respect of their sidechains, but a consistent and significant preference for heterochiral outcomes.

Full text of DOI:10.1039/a901730e

Stereochemical preference for heterochiral coupling controls selectivity in
competitive peptide synthesis
David Birch,a Roger R. Hill,*b G. E. Jeffsb and Michael Northa
a Department of Chemistry, University of Wales, Bangor, Gwynedd, UK LL57 2UW
b Department of Chemistry, The Open University, Milton Keynes, UK MK7 6AA. E-mail: r.r.hill@open.ac.uk
Received (in Cambridge, UK) 4th March 1999, Accepted 13th April 1999
Competitive activated couplings of N-phthaloyl amino acids
with amino acid dimethylamides show little selectivity
among substrates in respect of their sidechains, but a
consistent and significant preference for heterochiral out-
comes.
amino acid derivatives of opposite configuration. Further
confirmation of a substantial preference for heterochiral
coupling was found in the outcome of reactions between
equimolar racemic reagents. Thus, the percentages of hetero-
chiral diastereoisomer in reactions between racemic Pth-alanine
and racemic dimethylamides of alanine, valine and phenyl-
alanine were 75.6, 89.3 and 87.9, respectively. More complex
mixtures also behaved consistently. For example, reaction
between 1 equiv. of Pth-l-phenylalanine and a mixture of 5
equiv. each of the dimethylamides of l-alanine, l-phenyl-
Spontaneous abiotic formation of peptides from amino acid
derivatives is a topic of considerable interest within prebiotic
chemistry,1 peptide synthesis2 and combinatorial chemistry.3
Whilst non-random thermal condensations in amino acid
mixtures are seen as routes to incipient informational bio-
polymers,4 evidence for selectivity at lower temperatures is
equivocal.2,5 Moreover, we are unaware of any study that
addresses the question of stereoselectivity in competitive
reactions where epimerizing equilibration is not a factor. We
have found that, in DCC/HOBt-mediated coupling of N-
phthaloyl amino acids and amino acid dimethylamides, stereo-
selectivity not only is the dominant controlling factor but also
favours reaction between substrates of opposite configuration.
An explanation of this phenomenon implies that such selectivity
could be more widely applicable, with potentially significant
consequences.
N-Phthaloyl (Pth) and N,N-dimethylamido (DMA) deriva-
tives were selected for convenient chromophorically-based
detection and as an inert C-terminus, respectively. Competitive
couplings were carried out mostly pair-wise, where one type of
substrate had the opportunity to react with two of the other type
of substrate, the latter usually present in considerable molar
excess. In an illustrative procedure, a solution of Pth-l-
phenylalanine 1-hydroxybenzotriazole (HOBt) ester (0.1
mmol) in CH2Cl2 (10 cm3), prepared in situ from Pth-l-
phenylalanine, HOBt (0.12 mmol) and DCC (0.12 mmol), was
added to racemic phenylalanine dimethylamide hydrobromide
(1.0 mmol) dissolved in CH2Cl2 (25 cm3) and Et3N (1.2
mmol).6 The mixture was stirred for two days, the CH2Cl2
removed by evaporation, the residue taken up in the initial
HPLC eluent (30% aq. MeCN) and a homogeneous sample
analysed by the area ratio of peaks identified by standards7 as
the l,l- and d,l-diastereoisomers of the corresponding deriva-
tized dipeptides.
The outcomes of 28 such competition experiments are
summarised in Tables 1 and 2, results being averaged in the
cases where experiments were repeated. Table 1 shows the
consistent and significant preference for heterochiral couplings,
irrespective of whether the excess reagent is the Pth component
or the DMA component. In order to confirm that products
resulted from kinetic rather than thermodynamic control, we
allowed one experiment to proceed with 1 equiv. of the racemic
DMA and found the ratio of diastereomers to be close to 1:1 at
completion. Table 2 demonstrates that when a choice of amino
acids is presented, selectivity additional to that of ster-
eochemistry is only significant in respect of the preferred
coupling with glycyl residues.
The presence of one reactant in significant excess allows the
approximation that product ratios in competing reactions
correspond with the ratio of rate constants. Thus, the rate of
peptide bond formation in these reactions is largely independent
of b-substitution but significantly and uniformly faster between
Table 1 Competition outcomes in reactions between Pth-l-X and racemic
Y-DMA using DCC/HOBt coupling
ld-dipeptide
Initial [Y-DMA]/
[Pth-l-X]
derivativea
(%)
X
Y
Ala
Ala
Ala
Ala
Ala
Ala
Ala
Phe
Phe
Phe
Phe
Phe
Val
Val
Val
Ala
Ala
Ala
Phe
Phe
Phe
Val
Phe
Phe
Phe
Ala
Val
Val
Ala
Phe
9.9
86
55
85
92
90
85
94
89
87
75
90
94
96
95
83
1.0
0.08b
7.0
18.0
45.0
8.0
10.0
6.3
0.1b
7.5
10.0
10.0
10.0
8.3
a
Determined by integration of HPLC traces, typically at 220 nm using a
250 3 4.6 mm (ID) Phenomenex Luna C18 (2) column at 35 °C and a linear
mobile phase gradient from 30 to 95% MeCN in water over 20 min. Error
< ± 5%. b Racemic Pth-X and l-Y-DMA.
Table 2 Competition outcomes in reactions between Pth-X and equimolar
mixtures of Y-DMA and Z-DMA using DCC/HOBt coupling.
% LD-
dipeptide
Initial
Final
derivativea
X
Y
Z
{[Y] + [Z]}/[X]b [XY]/[XZ]c (%)
l-Ala l-Ala l-Phe
l-Ala l-Ala l-Phe
6.0
0.17
1.0
0.8
0.3
3.5
1.4
0.2
9.0
2.7
7.3
4.2
0.07
1.06
7.3
—
—
78
78
—
85
90
—
—
—
93
—
88
l-Ala l-Ala d-Phe 13.0
l-Ala d-Ala l-Phe 10.2
l-Ala l-Ala l-Val 10.0
l-Ala l-Ala d-Val 11.0
l-Ala d-Ala l-Val 10.0
Gly
Gly
Gly
Gly
Gly
Gly
l-Ala
l-Phe
l-Va
9.3
8.1
10.0
l-Phe l-Phe d-Val 10.0
l-Phe l-Phe l-Val 10.0
l-Phe d-Phe l-Val 10.0
a
b
c
See Table 1.
[PthXZDMA].
{[Y-DMA] + [Z-DMA]}/[Pth-X].
[PthXYDMA]/
Chem. Commun., 1999, 941–942
941

alanine and racemic valine gave a product mixture with a ratio
15:6:4(ll):75(ld), respectively.
The dominance of chirality over sidechain as an influence in
these competitive reactions may be explained by considering
the probable structure of the activated complex 1, where it can
as the latter interact positively. (Similar independence has been
noted in respect of amino acid selectivity in the reactions of
different amino acids with the phosphoanhydrides of alanine
and various nucleoside monophosphates; specificity was unaf-
fected by alternative nucleosides.8) The results support Dose’s
contention that rate differences in condensations in amino acid
mixtures would generally be too small to direct sidechain
sequences,5 but introduce the prospect that, in the absence of
epimerization equilibria, such sequences in extendable systems
would entail alternating configurations at Ca until interrupted
by a glycyl residue. We are presently exploring the generality of
this effect with other systems.
O
δ−
R1
O
N
O
N
N
δ+
NH2
N
O
R2
We thank Mrs J Burrage and Mr Eric Williams for technical
assistance.
O
1
NMe2
be seen that the two a-carbon atoms are separated by only two
other atoms, whereas b-carbon atoms are more remote from
potentially interacting centres. The consistent preference for a
heterocyclic outcome is less readily anticipated, but would
follow if 1 gained secondary stabilization by interactions
between the terminal groups in a six-membered chair conforma-
tion, where the more favoured diequatorial disposition of
sidechains would lead to the observed results (2b).
Notes and references
1 S. A. Kaufman, The Origins of Order, OUP, Oxford, 1993; B. M. Rode,
J. Phys. Chem., 1992, 96, 4170; A. Brack, Origins Life, 1987, 17, 367; K.
Dose, Naturwissenschaften, 1983, 70, 378.
2 M. Mutter, Int. J. Peptide Protein Res., 1979, 13, 274.
3 N. K. Terrett, Combinatorial Chemistry, OUP, Oxford, 1998; S. R.
Wilson and A. W. Czarnik, Combinatorial Chemistry, Wiley, New York,
1997.
4 S. W. Fox and K. Dose, Molecular Evolution and the Origin of Life,
Dekker, New York, 1974.
5 K. Dose, Interdiscip. Sci. Rev., 1988, 13, 348.
6 M. Bodansky and A. Bodansky, The Practice of Peptide Synthesis,
Springer-Verlag, Berlin, 1984.
O
O
R1
δ +
H2N
δ +
H2N
R2
R2
NMe2
NPth
NMe2
NPth
BtO
BtO
R1
7 Authentic standards for each coupling experiment were prepared by
reaction between a single N-phthaloyl amino acid and a single amino acid
dimethyl amide, and gave satisfactory spectroscopic and analytical
data.
δ–
δ–
O
O
L, L
L, D
2a
2b
8 S. Tyagi and C. Pomamperuma, J. Mol. Evol., 1990, 30, 391.
If valid, this interpretation would suggest that the selectivity
is largely independent of activating and terminal groups as long
Communication 9/01730E
942
Chem. Commun., 1999, 941–942