IMPROVED FOLDING OF APO-RETINOL-BINDING-PROTEIN IN THE PERIPLASM OF ESCHERICHIA-COLI - POSITIVE INFLUENCES OF DSBC COEXPRESSION AND OF AN AMINO-ACID EXCHANGE IN THE VITAMIN-A BINDING-SITE
Am. Schmidt et al., IMPROVED FOLDING OF APO-RETINOL-BINDING-PROTEIN IN THE PERIPLASM OF ESCHERICHIA-COLI - POSITIVE INFLUENCES OF DSBC COEXPRESSION AND OF AN AMINO-ACID EXCHANGE IN THE VITAMIN-A BINDING-SITE, Protein engineering (Print), 11(7), 1998, pp. 601-607
The in vivo folding of the serum retinol-binding protein (RBP), a repr
esentative of the lipocalin structural family, is known to be complex.
In order to gain insight into the essential steps along its folding p
athway the heterologous production of the functional protein in Escher
ichia coli was investigated. Simultaneous overexpression of the bacter
ial dsbC gene, which codes for a periplasmic thiol-disulphide oxidored
uctase, prevented the formation of soluble REP variants with non-nativ
e disulphide bonds that were otherwise observed. Although the coexpres
sion of dsbC had furthermore a stabilizing effect on the cell viabilit
y, the relative yield of the solubly produced REP was not much better.
In an attempt to enhance its folding efficiency, a favourable point m
utation in the inner part of the retinol-binding pocket was predicted.
Replacement of the polar Gln117 with an Ile side chain seemed not onl
y to relieve the unfavourable energetics of the carboxamide group in t
he environment of predominantly non-polar residues but also to fill an
adjacent cavity in the hydrophobic core. Indeed, this single substitu
tion reproducibly resulted in a more than threefold increase in the am
ount of functional recombinant REP. Ligand binding experiments showed
that the affinity of this mutant for retinol was slightly enhanced. Ki
netic measurements revealed that this was due to a higher association
rate whereas the dissociation of the complex with retinol was essentia
lly unaffected. Although the question remained why nature did not sele
ct this obviously beneficial mutation, our results demonstrate that th
e folding pathway of a lipocalin can be optimized by protein engineeri
ng.