THE THREE DIMENSIONAL STRUCTURE OF A NINE-HAEM CYTOCHROME C FROM D.DESULFURICANS ATCC 27774 FROM FE MAD DATA REVEALS AN UNUSUAL REPLAY OF A FAMILIAR THEME.
M.A. Carrondo1, P. M. Matias1, R.Coelho1, A.V.Coelho1,2 and A.W. Thompson3
1 ITQB, Universidade Nova de Lisboa, P.O.
Box 127, 2780 Oeiras, Portugal;
2 Chemistry Department, Universidade de
Évora, 7000 &EACUTEvora, Portugal; 3 EMBL
Grenoble Outstation, c/o ILL 20, B.P. 156, F-38042 Grenoble
Cedex, France
Keywords: MAD phasing method, cytochromes, electron
transport
Nine-heme cytochrome c has been purified
from Desulfovibrio desulfuricans ATCC 27774 cells grown
under both nitrate and sulfate-respiring conditions. Therefore,
it is likely to play a role in the electron transfer system of
both respiratory chains. Monoclinic crystals of this protein were
obtained in space-group P21 by vapour diffusion from a
solution containing 2% PEG 6000 and 0.25-0.75M acetate buffer
pH=5.5, with cell parameters a=61.00 A, b=106.19 A,
c=82.05 A, b=103.61o. Density measurements suggested that
there are two independent molecules in the asymmetric unit and
self-rotation function calculations indicated the presence of a
non-crystallographic axis perpendicular to the crystallographic
twofold axis. Diffraction data were collected at ESRF BM-14 to
2.9 A resolution from a small frozen crystal, at 3 wavelengths
suitably chosen near the Fe K-absorption edge by means of an
X-ray fluorescence scan. A fourth data set was collected to 1.8 A
resolution near l=0.9 A for wavelength scaling purposes and also
to provide additional phasing information, as well as the best
possible electron density map for side-chain assignment, since
only the first 39 aminoacids in the polypeptide chain are known.
The Fe atomic positions were used to derive phase information and
the electron density maps obtained clearly showed a
solvent-protein boundary. Improvement by density modification and
phase extension to 2.4 A, allowed a clear and complete tracing
of the 292 residues in the polypeptide chain, missing only one
residue at the N-terminal and another at the C-terminal. Most of
the side-chains of the residues beyond 39 were assigned by
inspection and refinement was begun at 2.4 A resolution, using
non-crystallographic symmetry restraints between the two
independent molecules in the asymmetric unit. The initial
R-factor was 37.6%, and the free R-factor 38.7%, based on a
random sample of 5 % of the reflections in the 1.8 A dataset,
using only data to 2.4 A. The current refinement results using
1.8 A data with no NCS restraints and the inclusion of 412
solvent molecules and six acetate ions in the model gave an
R-factor of 20.6% and a free R-factor of 23.4 %, with more than
90% of the side-chains correctly assigned on the basis of the fit
to the 2Fo-Fc electron density map (apart from ambiguities
impossible to resolve at this resolution, such as GLU vs. GLN and
ASP vs. ASN residues). The three-dimensional structure of this
cytochrome is an unusual replay of a familiar theme since the
nine heme groups are arranged into two tetra-heme units, located
at both ends of the molecule, with geometrical parameters very
similar to those encountered in tetraheme cytochromes c3,
and the ninth heme is located asymmetrically between the two
tetra-heme units. This presentation will briefly describe the
structure determination and refinement procedure, and will focus
on a description of the current model as well as present some
comparisons with the known three-dimensional structure of the
tetra-heme cytochrome c3 from the same
organism.