|
Figure 1.
Figure 1. Stereo MOLSCRIPT[23.] Figure of the
PhoE-phosphate complex. Residues contacting the bound phosphate
(larger spheres; shaded bonds) are drawn and labeled, along with
the C-terminal residue Val208 whose carboxyl group interacts
with Arg9. Residue His10 lies beneath the bound phosphate and is
not labeled for reasons of clarity. Potential hydrogen bonds
involving Arg9 are drawn as dotted lines. A blue C^a trace
replaces part of the a-helix, which would otherwise obscure
Gln22. The C-terminal tails of superimposed rat F26BPase
(magenta) and phosphorylated E. coli dPGM (orange) are compared
to the PhoE tail. The N terminus of the PhoE-phosphate structure
is labeled in black and the C termini of the other structures
labeled in their respective colors. The B. stearothermophilus
PhoE protein was produced and purified as described[11.] and
stored at -80 °C with preservation of activity in 10.0 mM
Tris-HCl (pH 7.4), 1.0 mM DTT, 2.0 mM EDTA, 150.0 mM NaCl at a
concentration of 25 mg/ml. Crystallization of the B.
stearothermophilus PhoE-phosphate complex was accomplished using
the vapor diffusion, hanging drop method[24.] in 24 well VDX
culture plates (Hampton Research Inc.). Equal volumes of the
protein sample and reservoir solution (1 µl each), and 0.5
µl of AMP substrate[11.] (25 mM) were used and the mixture
was equilibrated against 1 ml of reservoir solution at 22
°C. A sparse matrix screen (Hampton Research Inc.) was used
to obtain initial conditions. Diffraction quality irregularly
shaped crystals were obtained after the refinement of initial
conditions using 15.0% (v/v) ethylene glycol (EG), and 85.0 mM
sodium cacodylate buffer (pH 4.5), as the reservoir solution, in
one to two weeks (0.5 mm×0.4 mm×0.3 mm). The
PhoE-vanadate complex was obtained by soaking the crystals
described above in 50.0 mM of ammonium metavanadate
(Sigma-Aldrich Co.), 30.0% EG, 20.0 mM sodium acetate buffer (pH
5.0) for three days in a depression dish. The PhoE-phosphate
complex crystals were soaked for 20 seconds in a cryo-protecting
solution (30.0% EG, 25.0 mM AMP, and 55.0 mM sodium cacodylate
buffer (pH 4.5)) and flash frozen at -170 °C in liquid
nitrogen immediately before data collection. The PhoE-vanadate
complex crystals were also flash frozen in liquid nitrogen
directly from their soaking solution. Both sets of crystals were
diffracted using a synchrotron X-ray radiation source, beamline
5.0.1 at the Advanced Light Source, Lawrence Berkeley National
Laboratory, with a 1.0 Å wavelength. The oscillation
diffraction images were recorded using an ADSC Quantum 4u CCD
detector. The diffraction data were processed with the HKL2000
package.[25.] A Free R-factor,[26.] calculated from 5% of
reflections set aside at the outset, was used to monitor the
progress of refinement. The free R set from the earlier
crystal[17.] was transferred to the new data for the phosphate
complex and completed randomly in the higher resolution range.
This free R assignment was then transferred to the trivanadate
complex data. The new complex structures were solved using rigid
body refinement with a maximum likelihood target using CNS[27.]
and atoms of the refined model subjected to a random positional
shift of up to 0.3 Å in each of the x, y and z dimensions
before further refinement in order to eliminate phase bias. All
data were used throughout, without the application of sigma or
amplitude-based cutoffs. Further rounds of positional and
restrained individual B-refinement were carried out with a
variety of stereochemical[28.] and other analyses [29. and 30.]
periodically performed in order to locate possible model errors.
Alternate conformations were introduced into the model when
electron density clearly justified their inclusion. Ligands were
modeled and refined with the aid of the HIC-UP database. [31.]
The phosphate complex contains a phosphate ion and an ethylene
glycol molecule in the catalytic site, and a second ethylene
glycol molecule at the protein surface. The trivanadate complex
contains trivanadate in the catalytic site and a partially
occupied phosphate group at the protein surface. Water molecules
were added into 3s peaks of the s-A weighted |2F[o] -F[c]| map
within hydrogen bonding distance of suitable model atoms.
Programs of the CCP4 package[32.] were used for manipulations
and O [29.] for visualization and manual rebuilding. Structural
superpositions were made with LSQMAN. [33.] The structures and
structure factors have been deposited in the RCSB PDB with codes
1h2e and 1h2esf (phosphate complex) and 1h2f and 1h2fsf
(trivanadate complex).
|
