 |
PDBsum entry 2bgf
|
|
|
|
References listed in PDB file
|
 |
|
Key reference
|
|
|
Title
|
|
Various strategies of using residual dipolar couplings in nmr-Driven protein docking: application to lys48-Linked di-Ubiquitin and validation against 15n-Relaxation data.
|
|
|
Authors
|
|
A.D.Van dijk,
D.Fushman,
A.M.Bonvin.
|
|
|
Ref.
|
|
Proteins, 2005,
60,
367-381.
[DOI no: ]
|
|
|
PubMed id
|
|
|
|
|
|
Abstract
|
|
|
When classical, Nuclear Overhauser Effect (NOE)-based approaches fail, it is
possible, given high-resolution structures of the free molecules, to model the
structure of a complex in solution based solely on chemical shift perturbation
(CSP) data in combination with orientational restraints from residual dipolar
couplings (RDCs) when available. RDCs can be incorporated into the docking
following various strategies: as direct restraints and/or as intermolecular
intervector projection angle restraints (Meiler et al., J Biomol NMR
2000;16:245-252). The advantage of the latter for docking is that they directly
define the relative orientation of the molecules. A combined protocol in which
RDCs are first introduced as intervector projection angle restraints and at a
later stage as direct restraints is shown here to give the best performance.
This approach, implemented in our information-driven docking approach HADDOCK
(Dominguez et al., J Am Chem Soc 2003;125:1731-1737), is used to determine the
solution structure of the Lys48-linked di-ubiquitin, for which chemical shift
mapping, RDCs, and (15)N-relaxation data have been previously obtained (Varadan
et al., J Mol Biol 2002;324:637-647). The resulting structures, derived from CSP
and RDC data, are cross-validated using (15)N-relaxation data. The solution
structure differs from the crystal structure by a 20 degrees rotation of the two
ubiquitin units relative to each other.
|
|
|
|
|
|
|
|
Figure 6.
Figure 6. Result of Dyndom[47] analysis, showing the rotation
of the proximal domain with respect to the distal domain when
comparing the representative solution structure (black) with the
crystal structure (gray). The structures are fitted on the
distal domain, and secondary structure elements are indicated.
Two orthogonal views are shown, corresponding to a 90°
rotation around a horizontal axis in the plane of the paper. The
rotation axis as determined by Dyndom is indicated in red.
|
|
Figure 7.
Figure 7. Detailed view of the interface of the Ub[2] solution
structure. Residues involved in hydrophobic non-bonded contacts
(ball-and-stick and transparent CPK representation) or in
inter-domain hydrogen bonds or salt-bridges (ball-and-stick
representation) are shown (see also Supporting Table 3S; note
that for a better visualization not all contacts are shown).
Dotted lines represent hydrogen bonds. The residues are labeled
with one-letter residue code and residue number, followed by D
or P to indicate the distal or proximal domain, respectively.
|
|
|
|
|
|
The above figures are
reprinted
by permission from John Wiley & Sons, Inc.:
Proteins
(2005,
60,
367-381)
copyright 2005.
|
|
|
Secondary reference #1
|
|
|
Title
|
|
Structural properties of polyubiquitin chains in solution.
|
|
|
Authors
|
|
R.Varadan,
O.Walker,
C.Pickart,
D.Fushman.
|
|
|
Ref.
|
|
J Mol Biol, 2002,
324,
637-647.
[DOI no: ]
|
|
|
PubMed id
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Figure 2.
Figure 2. pH-dependence of the amide chemical shift
difference between Ub[1] and Ub[2] for the proximal domain. As
can be seen from this plot, the chemical shift perturbations
(Ub[2] versus Ub[1]) saturate at higher pH, so that in most
amide groups there is practically no (or very small) difference
between the chemical shift perturbations observed at pH 6.8 and
7.5. These data suggest that the closed conformation of Ub[2] is
almost fully populated at pH 7.5. In the case of fast exchange,
the relative population of the closed or open conformations at
the intermediate pH values can then be estimated assuming that
the observed perturbation in the peak position is a weighted
average of the corresponding values for the closed (at pH 7.5)
and open (pH 4.5) conformations. The estimates of the population
of the open conformation obtained for the individual amide
groups at pH 6.8 ranged from less than 1% to 25%, with the mean
value of 15%.
|
|
Figure 3.
Figure 3. Comparison of the three-dimensional conformations
of Ub[2] in solution derived here at (a) acidic pH and ((b) and
(c)) at neutral conditions. For comparison, (d) shows the
crystal structure of Ub[2].[13.] Structures shown in (a) and (b)
were derived using 15N relaxation data (rotational diffusion
tensor) while that in (c) is on the basis of RDCs (alignment
tensor). The orientation of the principal axes of the rotational
diffusion or alignment tensors of the Ub[2] molecule, seen by
each individual Ub domain, is indicated by rods positioned at
the center of mass of the corresponding domain. The z-axes
(turquoise) are oriented in the horizontal direction, the y-axes
(pink) are vertical, and the x-axes are oriented toward the
reader. Structures for the individual domains are from
1aar.pdb;[13.] a similar orientation was obtained using other
protein coordinates (see the text). The domains are colored
green (proximal) and blue (distal). Cylindrical arrows (red)
indicate the orientation of a-helices. The location of L8, I44,
and V70 in both domains, as well as that of K48 in the proximal
domain are indicated in (b). Because this approach provides the
orientation (not the distance) between the domains, their
relative positions in (a)-(c) are somewhat arbitrary. Also, the
C terminus of the distal domain is unstructured/flexible and
should easily adopt a conformation accommodating closer contact
between the units. Its conformation shown here is from the
crystal structure and should be considered as an illustration
only. The picture was prepared using Molmol. [44.]
|
|
|
|
|
|
The above figures are
reproduced from the cited reference
with permission from Elsevier
|
|
|
|
|
|
|
