|
Detailed structure-function characterization of IDPs
|
|
|
The classical structure-function paradigm suggests that solving structures at high resolution holds the key to understanding protein function at atomic resolution. Whereas in many instances we have come close to this ambitious goal, the recognition of protein disorder has posed a significant challenge to this simplifying view. Now we work under the premise that the extension of the structure-function paradigm offers a similar atomistic understanding of function through the description of the ensemble and dynamics of IDP structures. Whereas these goals are clear and many techniques can contribute to unraveling the structural puzzle of IDPs, NMR clearly stands out because it can provide atomistic description of the disordered ensemble. In our laboratory, we are studying two inhibitor-enzyme pairs in great detail.
Calpain is an intracellular Ca2+-activated cysteine protease, which is common in eukaryotic cells and fulfills basic cellular functions such as regulation of cell division, differentiation and cell motility. The activity of the enzyme is tightly controlled by the level of free Ca2+ and its endogenous ID inhibitor, calpastatin. Vertebrates contain multiple isoforms of the enzyme: some, like mu- and m-calpain, are ubiquitously expressed while others, such as p94 or p82, are more tissue specific. Calpastatin is also ubiquitously expressed, usually at levels exceeding that of the enzyme(s). The effective control of calpain activity is of particular importance for its proper function, as witnessed by a variety of pathological conditions that result from the loss of control of the calpain-calpastatin system.
Calpastatin is composed of four equivalent domains, each of which is
capable of inhibiting a separate calpain molecule. Within each domain,
there are three short, conserved segments termed subdomains, which are
primarily responsible for the observed inhibitory effect. Calpastatin
as a whole is fully disordered (random-coil like), whereas its
subdomains A and C, which anchor the inhibitor to the enzyme in a
Ca2+-dependent manner and subdomain B, which binds in the active-site
region and directly inhibits the enzyme, have been thought to
transiently sample the conformation the attain in the bound state. We
have tested this inference by full NMR assignment of domain I of human
calpastatin and by the detailed structural characterization of
calpastatin in both the solution and bound states. Several NMR
observables, such as secondary chemical shift values (chemical shift
index, CSI) and R1, R2 and hetero NOE relaxation data showed a the
non-random conformational state of calpastatin, with reduced
flexibility within the subdomains, and an increased flexibility in the
regions linking them. The interpretation of these observations is that
these important regions probably show a preference for the local
structure they attain in he bound state, partial helical backbone
conformations in subdomains A and C, and beta structure in subdomain B.
NMR relaxation data of the bound state showed the corresponding
tripartite binding mode of the inhibitor, in which calpastatin wraps
around and contacts the enzyme at three points (subdomains A, B and C),
facilitated by linkers which remain disordered in the bound state. This
unprecedented binding mode probably permits a unique combination of
specificity, speed and binding strength in regulation.
|
 |
1H-15N HSQC spectrum of calpastatin domain 1. From Kiss et al. (2008) Biochemistry 47: 6936-45.
|
|
NMR parameters of calpastatin domain 1 showing non-random behavior
in subdomains A through D.
From Kiss et al. (2008) Biochemistry 47: 6936-45.
|
 |
Human securin (also known as pituitary tumor transforming gene product,
PTTG) plays an important role in regulating the cell cycle. Securin is
the inhibitor of separase, the protease that initiates chromosome
separation at the onset of anaphase in mitosis. Malfunction of the
separase-securin system causes inappropriate chromosome segregation,
i.e. aneuploidy, a leading cause of malignant transformations in
humans. Human separase is a large protein of 2120 amino acids, with an
armadillo-repeat region and two caspase-like folds, where securin
probably binds and exerts its inhibitory function. Human securin of 202
amino acids in length is fully disordered (random coil-like). We could
achieve full assignment of NMR backbone amide resonances by a
combination of proton-detected and 13C-detected protonless NMR
experiments. CSI, 15N relaxation rates (R1, R2, 1H-15N NOEs), 1H
exchange rates with the solvent (CLEANEX-PM) and 1H-15N residual
dipolar couplings determined along the entire length of the protein
showed that securin is not entirely disordered, but segregates into a
largely disordered N-terminal half, and a C-terminal half with
transient segmental order. A short segment of about 10 residues within
the C-terminal half has a significant helical tendency, which suggests
that this site is probably directly involved in separase recognition
and inhibition. The importance of this finding also resides in that
understanding the structural background of separase inhibition by
securin may provide novel target(s) in drug development.
|
| Scheme of cell-cycle: securin regulates the transition from metaphase to anaphase. |
|