what-when-how
In Depth Tutorials and Information
slower folding to complete triple-helix formation of the
entire molecule.
Folding of normal type I collagen is a complicated
process, involving C-terminal trimerization, many post-
translational modifications, and multiple cis-trans isom-
erization events in imino acids to reach the final folded
molecule with its marginal stability (see Chapter 7 in
this volume).
28
This process may be easily perturbed
when a Gly missense mutation is present within the
triple helix, and the aberrant folding of OI collagens is
likely to involve defective interactions with chaperones,
retention in the ER, and processes of ER stress and deg-
radation (see Chapters 6 and 7 in this volume).
NMR and other biophysical studies on collagen
model peptides containing Gly substitutions have pro-
vided “snapshots” of which residues are triple heli-
cal and successfully folded, and which are disordered
and incapable of folding.
48
For example, a peptide
containing six triplets from the α1(I) chain attached to
a C-terminal (Gly-Pro-Hyp)
4
cap folded fully into a
triple helix, and the introduction of a Gly→Ser substitu-
tion, resulted in triple-helix formation only C-terminal
to the substitution site.
49
The Gly substitution muta-
tion appeared to trap a stage along the directional
folding pathway where the triple helix is unable to
continue, suggesting a re-nucleation event is neces-
sary. Incorporation of a natural imino-rich sequence
N-terminal to the mutation site in the peptide allowed
the formation of the complete triple helix, N-terminal
to as well as C-terminal to the Gly substitution site.
50
These results suggest the mutation causes a pause in
triple-helix propagation which requires an internal re-
nucleation sequence to restart the folding process on the
other side of the mutation. Peptide studies also showed
that the sequence environment of the mutation affects
the folding as well as stability.
40
The triple-helix folding in model peptides dif-
fers from full length collagens in their short triple-
helix length and the lack of both a defined initiation
site and linear propagation, but a recently developed
recombinant bacterial system based on the
S. pyo-
genes
Scl2 collagen-like protein seems like a good fold-
ing model. This bacterial collagen-like protein contains
an N-terminal globular trimerization domain which
is essential for the folding of the adjacent collagen
domain (Gly-Xaa-Yaa)
79
,
51-54
and has a
T
m
~36-37°C,
stability similar to that of human collagens, despite the
absence of hydroxyproline.
51,55
Gly missense mutations
have been introduced at specific sites within the triple
helix, and folding of such homotrimers investigated.
39
A Gly→Arg mutation near the center of the triple helix
led to a significant folding delay, going from a half time
of 10 min for the control to 55 min for the mutated mol-
ecule. When the Gly→Arg mutation was placed at a site
three triplets away from the N-terminus of the bacterial
triple helix, very close to the N-terminal trimerization
domain, a dramatic decrease was observed in the fold-
ing rate (
t
1/2
>1000 min) and the extent of refolding. The
extremely slow folding suggested this mutation dis-
rupted the triple-helix nucleation process. A mutation
introduced near an interruption sequence found within
type IV collagen also led to significant delay in folding.
56
OVER
VIEW AND FUTURE DIREC
TION
Missense collagen mutations lead to local distortions
in the triple-helix conformation, decreased local stabil-
ity and retardation of triple-helix folding. The sequence
of events that follows from these alterations to pathol-
ogy is in the process of being defined through model
peptides, computational approaches,
in vitro
studies
on OI collagens expressed in fibroblasts and OI mouse
models. Changes in conformation could impact the
fibril structure
57
or lead to abnormal interactions with
cell receptors or other matrix proteins.
10
The delayed
folding could lead to ER stress and cell pathology and
abnormal secretion. Studies on cells and animal mod-
els (see Chapter 21 in this volume) hold promise of
connecting the initial molecular level perturbations to
physiological consequences.
The identity of the amino acid replacing Gly and the
location of the mutation are critical features in deter-
mining the OI clinical outcome. Other important fac-
tors may include proximity of the mutation to binding
sites for cell receptors, matrix molecules, chaperones
or to residues essential for collagen self-association,
as well as the existence of folding domains. The grow-
ing OI clinical database represents a very important
resource for clarifying the molecular mechanisms of
OI which can lead to the development of rational treat-
ment of this serious human disorder. Formulation of
an algorithm for predicting the clinical outcome of a
given missense mutation in type I collagen still remains
a challenging task and will require input from experi-
mental, computational, genetic and clinical results.
References
