Vitamin B6 Enzyme Models - Artificial Enzymes - page 48

the transamination between L- phenylalanine and pyruvic acid to form D- alanine with
an ee as high as 90%. Since the presence of L- phenylalanine proved indispensable for
the enantioselective transamination, a five-coordinated Cu( II ) complex was proposed
as a key intermediate in the stereoselective protonation (Figure 2.2), in which the L-
phenylalanine molecule present in excess over 37 acts as a chiral bidentate ligand to be
bound to the quinoid-Cu( II ) complex [27]. The ammonium group of the lysine residue
of 34 acts as a proton source to protonate the quinoid-Cu( II ) complex preferentially
from the less hindered si face of the imino carbon to afford the aldimine chelate.

2.2.4
Vitamin B6-Polypeptide Systems

All the above vitamin B6 enzyme models were built with small molecules. Natural
enzymes, however, are macromolecules. Their macromolecular structures offer ideal
frames for the construction of versatile, robust catalytic sites. Strong and selective
binding of the substrate is attained through a combination of the hydrophobic effect
and specific substrate-enzyme interactions such as hydrogen bonding. The macromo-
lecular structure can also create regions in which the catalyzed reactions occur in a less
than fully aqueous medium. It is interesting to learn how to construct macromolecular
enzyme mimics. A straightforward idea would be to employ polypeptides.
Imperiali and Roy were the first to introduce a pyridoxal cofactor to a polypeptide
[28-30]. Their idea was to incorporate a pyridoxal-based amino acid residue in the S-
peptide of the RNase-S complexes. These RNase-S complexes are semisynthetic con-
structs consisting residues 21-124 of native ribonuclease A (S-protein) and an oligo-
peptide based on residues 1-20 (S-peptide) (Figure 2.3). Analysis of the crystal struc-
ture of RNase-S revealed that a Phe 8 pyridoxal (Pal 8 ) substitution in the S-peptide

would position the residue in proximity to the general acid-base pair (His 12 , His 119 )
utilized by the native enzyme. Thus, a few S-peptides containing a Pal 8 residue were
designed, synthesized by standard Fmoc solid phase procedures. These Pal 8 -contain-
ing S-peptides bound with the S-protein with affinities comparable to that of native S-
peptide. The S-peptide-S-protein complexes were then evaluated for their ability to
convert L- alanine into pyruvate under single turnover conditions in the presence or
absence of copper( II ). In one case, the binding to S-protein increased the transamina-
tion rate of the Pal 8 -containing S-peptide by 16.7-fold.
In a similar approach, pyridoxamine was introduced into an S-peptide at position 8
to maintain the interactions with His 12 and His 119 [30]. Upon formation of the RNase
complex, the rate was enhanced 7-fold compared with uncomplexed peptides under
single turnover conditions. However, replacing the His residue at position 12 with Ser
afforded only a 3-fold rate increase for the S-peptide-S-protein complex. Under cata-
lytic conditions with pyruvate and L- phenylalanine as the substrates, uncomplexed pep-
tides did not show catalytic turnover, suggesting that a hydrophobic microenviron-
ment in the peptide-protein complex is critical for catalysis. However, in the presence
of the S-protein, catalysis ensued. Up to 1.5 turnovers were observed in 160 h from the
S-peptide-S-protein complex.

!