The amino acid residues comprising the interface between strands of the coi
led-coil motif are usually hydrophobic and make a major contribution to coi
led-coil folding and stability. However, in some cases the presence of exce
llent hydrophobic residues at the coiled-coil interface is insufficient for
folding. It has been proposed that a "consensus trigger sequence" exists t
hat is necessary within the coiled-coil domains of various proteins to trig
ger folding. Therefore, in this study we designed a 31-residue hybrid seque
nce based on sequences from the two-stranded parallel coiled-coil domains o
f the yeast transcriptional activator GCN4 and the actin-bundling protein D
ictyostelium discoideum cortexillin I. The hybrid and its analogs were stud
ied by CD spectroscopy and analytical ultracentrifugation. The hybrid had s
table residues in the core "a" and "d" positions in the 3-4 hydrophobic rep
eat, denoted (abcdefg)(n), but did not have a consensus trigger sequence an
d did not possess appreciable secondary structure as determined by CD spect
roscopy. The substitutions in the parent peptide were introduced at positio
ns other than "a" and "d", altering a variety of interactions including alp
ha -helical propensity, interchain and intrachain electrostatics, and hydro
phobicity. Although the substitutions did not bring the overall sequence in
closer agreement to the consensus trigger sequence, they increased coiled-
coil folding and stability. Therefore, our results suggest that the combina
tion of stabilizing effects along a protein sequence is a more general indi
cator of protein folding in coiled-coils than the identification of a speci
fic trigger sequence. We propose that surpassing a critical threshold stabi
lity value using any type or combination of stabilizing effects will allow
coiled-coils to fold, in the absence of a specific trigger sequence per se.
(C) 2001 Academic Press.