Structure, Dynamics and Function of the Proton Pump Bacteriorhodopsin
235
At a pH above 9 another group on the cytoplasmic side might be deprotonated
and compensate at least partly the added positive charge of the arginine. This
would allow the structural changes associated with the transition to the M2 state
to take place.
Reconsidering the earlier experiments of Glaeser et al. (1986) it seems now
very probable that the MJ intermediate was accumulated and therefore no
changes in the tertiary structure were observed.
3
The Crystal Structure
3.1
Crystallization of bR
bR was one of the first integral membrane proteins that was crystallized in three
dimensions (Michel and Oesterhelt, 1980). Considerable progress was made by
Schertler et al. (1993) using the surfaces of freshly formed benzamidine crystals
as nucleation sites. These plate-shaped crystals gave diffraction to 3.6 A in a and
b direction and 4.2 A in the c direction.
A novel concept for crystallization of membrane proteins in a lipidic cubic
phase was reported by Landau et al. (1996). They crystallized bR in mono oleinwater cubic phase and obtained microcrystals of space group P63 with a unit cell
of a = b = 61.76 A; c = 104.16 A, a = ~ = 90° and y = l20°. Although the exact
mechanism of crystallization is still unknown, it seems probable that bR, once
inserted into the continuous 3D curved lipid bilayer of the cubic phase, diffuses
within this bilayer to form a nucleation particle from which well-ordered crystals
can grow. They started with very small microcrystals of up to 30 !lm in diameter
and 5 !lm in thickness. The crystallization conditions were continuously
improved so that crystals of 150 !lm in diameter are available now. The cell constant in a,b-direction of 61.76 A together with the hexagonal space group gave an
indicates an arrangement of bR molecules similar to the purple membrane. In
accordance with the screw axis along the c direction these purple membrane like
2D lattices are stacked on each other, where two adjacent lattices are rotated by 60
degrees (Pebay-Peyroula et al., 1997). Using a highly focused beam with a diameter of about the same size as the crystal, the signal to noise ratio was strongly
increased, so that diffraction patterns of up to 2 A resolution could be obtained.
3.2
Vibrational Spectroscopy to Assess the Functionality of bR in Microcrystals
After solving the problem of crystallizing a membrane protein the proper functionality of the protein within the crystal lattice has to be proven. Common tests
on functionality of membrane proteins usually call for compartmentation. This
strategy is apparently impossible with crystals. Thus, we employed vibrational
spectroscopy as a non-invasive tool to assess the functionality of bR microcrystals (Helerle et al., 1998).
235
At a pH above 9 another group on the cytoplasmic side might be deprotonated
and compensate at least partly the added positive charge of the arginine. This
would allow the structural changes associated with the transition to the M2 state
to take place.
Reconsidering the earlier experiments of Glaeser et al. (1986) it seems now
very probable that the MJ intermediate was accumulated and therefore no
changes in the tertiary structure were observed.
3
The Crystal Structure
3.1
Crystallization of bR
bR was one of the first integral membrane proteins that was crystallized in three
dimensions (Michel and Oesterhelt, 1980). Considerable progress was made by
Schertler et al. (1993) using the surfaces of freshly formed benzamidine crystals
as nucleation sites. These plate-shaped crystals gave diffraction to 3.6 A in a and
b direction and 4.2 A in the c direction.
A novel concept for crystallization of membrane proteins in a lipidic cubic
phase was reported by Landau et al. (1996). They crystallized bR in mono oleinwater cubic phase and obtained microcrystals of space group P63 with a unit cell
of a = b = 61.76 A; c = 104.16 A, a = ~ = 90° and y = l20°. Although the exact
mechanism of crystallization is still unknown, it seems probable that bR, once
inserted into the continuous 3D curved lipid bilayer of the cubic phase, diffuses
within this bilayer to form a nucleation particle from which well-ordered crystals
can grow. They started with very small microcrystals of up to 30 !lm in diameter
and 5 !lm in thickness. The crystallization conditions were continuously
improved so that crystals of 150 !lm in diameter are available now. The cell constant in a,b-direction of 61.76 A together with the hexagonal space group gave an
indicates an arrangement of bR molecules similar to the purple membrane. In
accordance with the screw axis along the c direction these purple membrane like
2D lattices are stacked on each other, where two adjacent lattices are rotated by 60
degrees (Pebay-Peyroula et al., 1997). Using a highly focused beam with a diameter of about the same size as the crystal, the signal to noise ratio was strongly
increased, so that diffraction patterns of up to 2 A resolution could be obtained.
3.2
Vibrational Spectroscopy to Assess the Functionality of bR in Microcrystals
After solving the problem of crystallizing a membrane protein the proper functionality of the protein within the crystal lattice has to be proven. Common tests
on functionality of membrane proteins usually call for compartmentation. This
strategy is apparently impossible with crystals. Thus, we employed vibrational
spectroscopy as a non-invasive tool to assess the functionality of bR microcrystals (Helerle et al., 1998).