Biochemistry (Moscow), Vol. 66, No. 11, 2001, pp. 1192-1196. Translated from Biokhimiya, Vol. 66, No. 11, 2001, pp. 1477-1482.
Original Russian Text Copyright © 2001 by Lanyi.
REVIEW
X-Ray Crystallography of Bacteriorhodopsin
and Its Photointermediates: Insights into the Mechanism
of Proton Transport
J. K. Lanyi
Department of Physiology and Biophysics, University of California, Irvine, CA 92697, USA; E-mail: jlanyi@orion.oac.uci.edu
Received March 20, 2001
Revision received May 9, 2001
Abstract—In the last few years, detailed structural information from high-resolution x-ray diffraction has been added to the
already large body of spectroscopic and mutational data on the bacteriorhodopsin proton transport cycle. Although there are
still many gaps, it is now possible to reconstruct the main events in the translocation of the proton and how they are coupled
to the photoisomerization of the retinal chromophore. Future structural work will concentrate on describing the details of the
individual proton transfer steps during the photocycle.
Key words: bacteriorhodopsin, x-ray diffraction, proton transport cycle, retinal
There is a huge amount of spectroscopic information, collected over many years, to describe the photoreaction cycle of wild type bacteriorhodopsin, as well as site
specific mutants modified at critical resides [1-6]. It has
resulted in a fairly complete assessment of the pathway of
the transported proton across the membrane, and gave
rise to many hypotheses for the conversion of the free
energy gain in the excited state of the retinal into the
movement of a proton and ultimately into a transmembrane electrochemical gradient. However, as with
enzymes in general, many of the questions concerning
the transformations of the protein in the reaction cycle
are structural in nature, and for that reason the answers
need to be framed in structural terms also. This short
review will discuss insights gained into the transport
mechanism from x-ray crystallography of the unilluminated bacteriorhodopsin and some of its photointermediates.
The acidic groups which play proton transfer roles in
the transport are located in the interhelical cleft formed
by the seven transmembrane helices, and are well known.
The photoisomerization of the retinal from all-trans to
13-cis,15-anti drives these transfers during the cyclic
reaction called the “photocycle”. Roughly speaking, the
photocycle can be described as a linear sequence of the K,
L, M, N, and O states [7-9]. Because they are defined by
their spectra in the visible, infrared, etc. they are intermediates distinguished by the different states of the retinal and its immediate environment. Some are produced
as several substates, the chromophore being the same but
with additional changes elsewhere in the protein. The K
to L transition reflects partial relaxation of the isomerized
but twisted retinal, and many small changes in the protein
and bound water that sets the stage for the L to M reaction, in which the retinal Schiff base becomes deprotonated and Asp-85 protonated. The M intermediate
exhibits complex rise and decay kinetics, which must
originate from several M substates that reflect the fact that
numerous important events must occur in the protein
before the Schiff base is reprotonated. During the lifetime
of M, there will have to be a change in the accessibility of
the Schiff base to a proton donor/acceptor from one side
to the other. Related to this [10], there will be release of a
proton to the extracellular surface, from a so-far undefined site, perhaps a hydrogen-bonded network of water
molecules [11], upon protonation of Asp-85. The Schiff
base had become deprotonated with a lowered pKa, but in
the next step it will be reprotonated, and one would
expect that its pKa will rise. The pKa of the proton donor
from the cytoplasmic direction, Asp-96, on the other
hand, will decrease from its initial value >11. Finally, a
pathway for transferring the proton from Asp-96 to the
Schiff base must be built.
After the Schiff base is reprotonated during the M to
N reaction, Asp-96 becomes reprotonated from the cytoplasmic surface producing a late N substate. This requires
a new protein conformation in which Asp-96 is accessible
to the surface but no longer to the Schiff base [5, 12]. The
0006-2979/01/6611-1192$25.00 ©2001 MAIK “Nauka / Interperiodica”
CRYSTALLOGRAPHY OF BACTERIORHODOPSIN
retinal may now thermally reisomerize to all-trans, producing the O state. The decay of O, that regenerates the
initial state, depends on deprotonation of Asp-85 and
reprotonation of the extracellular proton release site [1315].
1193
A
B
T46
2.96
502
2.87
A215
3.00
W182
2.86
STRUCTURE OF THE BR STATE
501
The latest and highest resolution structures, at 1.55
[16] and 1.9 Å [17], are from x-ray diffraction of crystals
grown by the cubic phase method [18, 19]. The P63 space
group arises by multiple stacking of the naturally occurring purple membrane sheets with P3 symmetry. This
means that the environment of the protein in the crystals
is very similar to that in the original membrane. Indeed,
the entire complement of the archaeal lipids native to the
purple membrane appear to be present in the crystals [16,
17], to the exclusion of detergent and the lipids used for
the cubic lipid matrix. The photocycle of bacteriorhodopsin in the crystals is similar to that in membrane
sheets [20].
The seven transmembrane helices span the membrane at various angles not far from the membrane normal, as known already from cryo-electron microscopy
[21-23]. The cytoplasmic half of the protein projects out
of the lipid bilayer by about 5 Å farther than the extracellular half. This, and the greater mobility of the cytoplasmic region, as reflected in the higher temperature factors
[16], is consistent with the fact that it is near the cytoplasmic surface where large-scale conformational
changes occur in the photocycle [24-27].
The active site comprises the protonated retinal
Schiff base, Asp-85, Asp-212, and their hydrogen bonds
to water 402, which separates the retinylidene cation and
the two aspartate anions [28]. The site is connected
through chains of interaction to both extracellular and
cytoplasmic regions. It is through these chains that longrange effects occur in the protein during the photocycle.
In the extracellular region, an extended 3-dimensional
hydrogen-bonded chain, that involves Arg-82, Glu-194,
Glu-204, and at least seven water molecules, leads to the
protein surface [16, 17]. Mutation of any of these three
residues abolishes or strongly inhibits the release of a proton upon protonation of Asp-85. Hydrogen bonds with
Thr-89, Tyr-185, Tyr-57, and Ser-193 lend additional
stability to the network.
The cytoplasmic region contains no such polar network, but it does contain a movable structural element.
The α-helical repeat of helix G is interrupted by a π-bulge
at Ala-215 and Lys-216 [16]. It is caused by hydrogen
bonds of the two main-chain C=O groups to water molecules 501 and 502, as shown in Fig. 1. This feature offers
the possibility of conformational change near the location
on helix G where the retinal is bound, through shuttling
between a more π-helical and a more α-helical arrangeBIOCHEMISTRY (Moscow) Vol. 66 No. 11
2001
K216
2.81
3.09
501
A215
W182
502
2.95
K216
2.90
T46
C
D
B
C
E
A
G
F
Fig. 1. Irregularities in the α-helices of bacteriorhodopsin. A
and B represent views, along the c-axis and the a-axis, respectively, of the π-bulge in helix G, and the two water molecules,
501 and 502, that stabilize it. The chain that extends from the
vicinity of the retinal to Thr-46 near the cytoplasmic surface is
evident. C is a view down the c-axis, to show the three prolineassociated kinks, on helix B, C, and F, as well as the π-bulge on
helix G. In B and C, helix G is shown with an additional helix
that represents a perfect α-helix aligned with the extracellular
segment. Coordinates used, 1CW3. Reprinted with permission
from [16].
ment. Water 501 is further hydrogen bonded to the indole
nitrogen of Trp-182, a residue that contacts the retinal
polyene chain. Water 502, in turn, is hydrogen bonded to
the C=O of Thr-46, whose side-chain is hydrogen bonded to Asp-96. This forms a chain of covalent and hydrogen bonds, with Trp-182, water 501, Ala-215, Lys-216,
water 502, Thr-46, and Asp-96 as participants. The chain
connects helices C, F, and G together. Breaking and
remaking it would offer the possibility of communication
between the isomeric state of the retinal and the protonation state of the cytoplasmic proton donor.
STRUCTURAL CHANGES AT THE RETINAL
In solution, photoisomerization of the retinal
from all-trans to 13-cis,15-anti drastically changes the
shape of the polyene chain and rotates the Schiff base
1194
LANYI
N–H bond. To a degree that is not yet quite clear, the
retinal binding pocket will oppose these changes in
geometry. The resolution of the structures for the K
[29] and L [30] states determined so far does not reveal
the kind of distortion this conflict will produce at the
retinal. The low occupancy of these states is also a serious problem. On the other hand, comparison of the
structures of the two M states, the “early” M from the
E204Q mutant [31] where proton release is blocked,
and the “late” M from the D96N mutant [32] where
reprotonation of the Schiff base is blocked, indicates
progressive relaxation of an initially distorted retinal.
In the “early” M the distortion is mainly of an increase
of the bond angle at C13 because the polyene chain is
still straight like all-trans, and less bent at C13 than in
the “late” M. The relaxation therefore consists of C13,
and therefore the 13-methyl group, buckling upward.
In both M states the angle of the Schiff base nitrogen
has inverted so that the N: points toward the cytoplasmic direction. If the small difference observed [31] in
this angle in the two M states is significant (and this is
not yet clear), the direction of the rotation is clockwise
(sweeping past Asp-85).
Because the geometry of the retinal in the L intermediate is not yet known, the mechanism for the deprotonation of the Schiff base is unclear. If the N–H bond in
L continues to point toward Asp-85, there will be direct
proton transfer to the aspartate. If it points toward the
cytoplasmic direction, direct transfer is less likely, and the
participation of bound water near this location might
need to be invoked. One possibility [33] would be the dissociation of a water molecule, with the H+ moving to
Asp-85 and the OH– moving to the Schiff base and
receiving its proton. This would be tantamount to
hydroxyl ion transport in the opposite direction, analogous to chloride transport by halorhodopsin [2, 34, 35]
and the D85T and D85S mutants of bacteriorhodopsin
[36, 37].
unprotonated). The results for M produced by illumination of the D96N mutant, as shown [32], as well as
for M of the wild type and the E204Q mutant [31, 44],
strongly suggest that the means for this coupling is the
shuttling of the positively charged side-chain of Arg-82
between the “up” position where it is connected to Asp85 through water 406, and the “down” position where it
is connected to the Glu-194/Glu-204 pair through
water 405.
The link between the protonation of Asp-85 and the
release of a proton to the surface will influence the
reversibility of the deprotonation of the Schiff base. Free
energy is dissipated at the release of the proton, proportional to the difference between the pH and the pKa for
proton release. This will raise the pKa of Asp-85 and
thereby block the return of the Schiff base proton.
Indeed, double-perturbation experiments indicated [45]
that with increasing pH, and with an apparent pKa of 6
(i.e., the pKa for proton release), the protonation equilibrium of the Schiff base in the M state is drawn away
from Asp-85. From this observation it appears that shifts
of proton affinity during the photocycle play a greater
part in the protonation switch than changes in the
geometry of the groups involved in the proton transfer.
This is in accord with the “local-access” mechanism for
the transport switch [46].
B
A
all-trans
retinal
T89
T89
Y185
402
D212
D85
D212
D85
401
Y57
Protonation of Asp-85 induces proton release from
a poorly defined site near the extracellular surface,
which has been suggested to be Glu-204 [38], the Glu194/Glu-204 pair [39], Tyr-185 [40], or the water molecules in the hydrogen-bonded network [11]. Lack of a
negative C=O stretch band [11], as well as UV-Raman
[41] and NMR [42] evidence argue against the first
three alternatives. The pKa of Asp-85 and a second proton binding site, that must correspond to the release
site, are coupled in the BR state already [10, 43]. Figure
2 compares the structure of the BR state (Asp-85 anionic, proton release site protonated) with the structure of
the M state (Asp-85 protonated, proton release site
Y185
W86
401
406
RELEASE OF A PROTON
TO THE EXTRACELLULAR SURFACE
13-cis
retinal
R82
405
403
E194
404
407
R82
Y57
407
Y83
E204
405
E204
E194
404
Fig. 2. Structural changes in the extracellular domain upon
formation of the M intermediate. A and B are views of selected features of the BR and M states of the D96N mutant,
respectively. The protonation of Asp-85 results in downward
movement of the side-chain of Arg-82, and the breaking
hydrogen bonds that connect the Schiff base region to the
extracellular surface. Coordinates used 1C8R and 1C8S.
Reprinted with permission from [32].
BIOCHEMISTRY (Moscow) Vol. 66 No. 11 2001
CRYSTALLOGRAPHY OF BACTERIORHODOPSIN
REPROTONATION
OF THE RETINAL SCHIFF BASE
In the M intermediate produced at 230 K, that
should correspond to a state immediately after loss of the
Schiff base proton, the O–H stretch band of Thr-89 indicates that its hydrogen bond to Asp-85 is retained [47]. In
the M state produced at room temperature, however, this
bond is broken [31]. The gradual changes at the retinal
cause upward displacement of the 13-methyl group of the
retinal after deprotonation of the Schiff base pushes the
indole ring of Trp-182 in the cytoplasmic direction [31].
Water 501 loses its hydrogen bond to the C=O of Ala-215
on helix G and forms a new one to Thr-178 on helix F.
This breaks the connection of Trp-182 on helix F to helix
G through water 501. Partial recovery of an α-helical
structure at the π-bulge, displacement of the side-chain
of Lys-216, and repacking of the side-chains between
helices F and G cause Asp-96 and Thr-46 to move apart
[31]. In the M state already, this allows the entry of water
into the cytoplasmic region, either from the bulk or from
other locations in the protein. Thus, a hydrogen-bonded
network of water molecules is assembled that leads from a
water that now bridges Asp-96 and Thr-46 in the direction of the Schiff base. Presumably, completion of this
network is the rate-limiting step in the reprotonation of
the Schiff base in the M to N reaction.
In the D96N mutant, where it could be easily determined [48], the pKa of the Schiff base at this stage of the
photocycle is 8, and proton transfer will not occur unless
the pKa of Asp-96 is lowered correspondingly. The water
molecule intercalated between Asp-96 and Thr-46, and
the generally less hydrophobic character of the now wellhydrated cytoplasmic region should suitably decrease the
initially very high pKa (>11) [49] of Asp-96.
1195
Asp-96 [24]. Nevertheless, it is clear that Asp-96 cannot
be in protonation equilibrium with the Schiff base and the
cytoplasmic surface at the same time [5], because reprotonation of the Schiff base is largely independent of pH.
Because the two protonation reactions proceed with not
greatly different time-constants, an additional conformational change in the cytoplasmic region is needed, which
is distinct from the one that allows deprotonation of Asp96. The nature of this conformation change is still uncertain.
Deprotonation of Asp-85 will be the consequence of
the reestablishment of the initial geometry at the Schiff
base. Loss of the Asp-85 proton limits the rate of the final
photocycle step [14], and its irreversibility must reflect
recovery of the very low pKa (2.5) of Asp-85.
PERSPECTIVES
High-resolution x-ray crystallography will continue
to contribute unique and essential information to uncover the transport mechanism in bacteriorhodopsin. The
main unsolved questions are about the exact path of the
proton transfer reactions. However, the limitations of the
crystallographic method are also becoming evident.
States or substates of the photocycle, which arise in rate
limiting steps, might be so transient that they can never be
stabilized. Visualizing the most interesting changes at the
retinal, which occur in the K and L states, will need crystals that diffract much better than the current resolution.
Describing the large-scale conformational changes in the
second half of the photocycle will require new approaches which avoid disorder in the tightly packed crystals.
Successful solution of the still outstanding problems will
no doubt utilize many experimental strategies and much
ingenuity.
THE LAST STEPS OF THE PHOTOCYCLE
REFERENCES
Return to the initial state requires several steps after
the N state: reprotonation of Asp-96 from the cytoplasmic surface, reisomerization of the retinal to all-trans,
deprotonation of Asp-85, and conformational recovery
[5, 6]. There is no direct structural evidence for how these
occur, but there are clues. The properties of a mutant bacteriorhodopsin indicate that in the dark the thermal equilibrium between the 13-cis,15-anti and all-trans isomeric
states is shifted toward 13-cis,15-anti when Asp-96 is
deprotonated [50]. This implies a second functional coupling in the protein during the photocycle, between the
retinal and the cytoplasmic region, perhaps through
reversal of the events that had mobilized the proton of
Asp-96 for the reprotonation of the Schiff base.
It was recently suggested that a single protein conformation can account for proton transfer between Asp96 and the Schiff base and the cytoplasmic surface and
BIOCHEMISTRY (Moscow) Vol. 66 No. 11
2001
1. Lanyi, J. K. (1993) Biochim. Biophys. Acta, 1183, 241-261.
2. Haupts, U., Tittor, J., and Oesterhelt, D. (1999) Annu. Rev.
Biophys. Biomol. Struct., 28, 367-399.
3. Lanyi, J. K. (1999) Int. Rev. Cytol., 187, 161-202.
4. Lanyi, J. K. (2000) Biochim. Biophys. Acta, 1459, 339-345.
5. Balashov, S. P. (2000) Biochim. Biophys. Acta, 1460, 75-94.
6. Kaulen, A. D. (2000) Biochim. Biophys. Acta, 1460, 204219.
7. Lozier, R. H., Xie, A., Hofrichter, J., and Clore, G. M.
(1992) Proc. Natl. Acad. Sci. USA, 89, 3610-3614.
8. Mathies, R. A., Lin, S. W., Ames, J. B., and Pollard, W. T.
(1991) Annu. Rev. Biophys. Biophys. Chem., 20, 491-518.
9. Lanyi, J. K., and Váró, G. (1995) Israel J. Chem., 35, 365386.
10. Balashov, S. P., Imasheva, E. S., Govindjee, R., and Ebrey,
T. G. (1996) Biophys. J., 70, 473-481.
11. Rammelsberg, R., Huhn, G., Lubben, M., and Gerwert, K.
(1998) Biochemistry, 37, 5001-5009.
1196
LANYI
12. Radionov, A. N., and Kaulen, A. D. (1999) FEBS Lett.,
451, 147-151.
13. Kandori, H., Yamazaki, Y., Hatanaka, M., Needleman, R.,
Brown, L. S., Richter, H. T., Lanyi, J. K., and Maeda, A.
(1997) Biochemistry, 36, 5134-5141.
14. Richter, H. T., Needleman, R., Kandori, H., Maeda, A.,
and Lanyi, J. K. (1996) Biochemistry, 35, 15461-15466.
15. Balashov, S. P., Lu, M., Imasheva, E. S., Govindjee, R.,
Ebrey, T. G., Othersen, B., Chen, Y., Crouch, R. K., and
Menick, D. R. (1999) Biochemistry, 38, 2026-2039.
16. Luecke, H., Schobert, B., Richter, H. T., Cartailler, J. P.,
and Lanyi, J. K. (1999) J. Mol. Biol., 291, 899-911.
17. Belrhali, H., Nollert, P., Royant, A., Menzel, C.,
Rosenbusch, J. P., Landau, E. M., and Pebay-Peyroula, E.
(1999) Structure, 7, 909-917.
18. Landau, E. M., and Rosenbusch, J. P. (1996) Proc. Natl.
Acad. Sci. USA, 93, 14532-14535.
19. Rummel, G., Hardmeyer, A., Widmer, C., Chiu, M. L.,
Nollert, P., Locher, K. P., Pedruzzi, I., Landau, E. M., and
Rosenbusch, J. P. (1998) J. Struct. Biol., 121, 82-91.
20. Heberle, J., Buldt, G., Koglin, E., Rosenbusch, J. P., and
Landau, E. M. (1998) J. Mol. Biol., 281, 587-592.
21. Grigorieff, N., Ceska, T. A., Downing, K. H., Baldwin, J.
M., and Henderson, R. (1996) J. Mol. Biol., 259, 393-421.
22. Kimura, Y., Vassylyev, D. G., Miyazawa, A., Kidera, A.,
Matsushima, M., Mitsuoka, K., Murata, K., Hirai, T., and
Fujiyoshi, Y. (1997) Nature, 389, 206-211.
23. Mitsuoka, K., Hirai, T., Murata, K., Miyazawa, A., Kidera,
A., Kimura, Y., and Fujiyoshi, Y. (1999) J. Mol. Biol., 286,
861-882.
24. Kamikubo, H., Oka, T., Imamoto, Y., Tokunaga, F., Lanyi,
J. K., and Kataoka, M. (1997) Biochemistry, 36, 1228212287.
25. Subramaniam, S., Lindahl, M., Bullough, P., Faruqi, A. R.,
Tittor, J., Oesterhelt, D., Brown, L., Lanyi, J., and
Henderson, R. (1999) J. Mol. Biol., 287, 145-161.
26. Subramaniam, S., and Henderson, R. (2000) Nature, 406,
653-657.
27. Oka, T., Yagi, N., Fujisawa, T., Kamikubo, H., Tokunaga,
F., and Kataoka, M. (2000) Proc. Natl. Acad. Sci. USA, 97,
14278-14282.
28. Luecke, H., Richter, H. T., and Lanyi, J. K. (1998) Science,
280, 1934-1937.
29. Edman, K., Nollert, P., Royant, A., Belrhali, H., PebayPeyroula, E., Hajdu, J., Neutze, R., and Landau, E. M.
(1999) Nature, 401, 822-826.
30. Royant, A., Edman, K., Ursby, T., Pebay-Peyroula, E.,
Landau, E. M., and Neutze, R. (2000) Nature, 406, 645648.
31. Luecke, H., Schobert, B., Richter, H. T., Cartailler, J.-P.,
Rosengarth, A., Needleman, R., and Lanyi, J. K. (2000) J.
Mol. Biol., 300, 1237-1255.
32. Luecke, H., Schobert, B., Richter, H. T., Cartailler, J. P.,
and Lanyi, J. K. (1999) Science, 286, 255-261.
33. Luecke, H. (2000) Biochim. Biophys. Acta, 1460, 133-156.
34. Lanyi, J. K. (1990) Physiol. Rev., 70, 319-330.
35. Oesterhelt, D. (1995) Isr. J. Chem., 35, 475-494.
36. Sasaki, J., Brown, L. S., Chon, Y., Kandori, H., Maeda, A.,
Needleman, R., and Lanyi, J. K. (1995) Science, 269, 7375.
37. Brown, L. S., Needleman, R., and Lanyi, J. K. (1996)
Biochemistry, 35, 16048-16054.
38. Brown, L. S., Sasaki, J., Kandori, H., Maeda, A.,
Needleman, R., and Lanyi, J. K. (1995) J. Biol. Chem.,
270, 27122-27126.
39. Essen, L., Siegert, R., Lehmann, W. D., and Oesterhelt, D.
(1998) Proc. Natl. Acad. Sci. USA, 95, 11673-11678.
40. Wang, J., and El Sayed, M. A. (2001) Biophys. J., 80, 961971.
41. Ames, J. B., Bolton, S. R., Netto, M. M., and Mathies, R.
A. (1990) J. Am. Chem. Soc., 112, 9007-9009.
42. McDermott, A. E., Thompson, L. K., Winkel, C., Farrar,
M. R., Pelletier, S., Lugtenburg, J., Herzfeld, J., and
Griffin, R. G. (1991) Biochemistry, 30, 8366-8371.
43. Richter, H. T., Brown, L. S., Needleman, R., and Lanyi, J.
K. (1996) Biochemistry, 35, 4054-4062.
44. Sass, H. J., Buldt, G., Gessenich, R., Hehn, D., Neff, D.,
Schlesinger, R., Berendzen, J., and Ormos, P. (2000)
Nature, 406, 649-653.
45. Brown, L. S., Dioumaev, A. K., Needleman, R., and Lanyi,
J. K. (1998) Biophys. J., 75, 1455-1465.
46. Brown, L. S., Dioumaev, A. K., Needleman, R., and Lanyi,
J. K. (1998) Biochemistry, 37, 3982-3993.
47. Kandori, H., Yamazaki, Y., Shichida, Y., Raap, J.,
Lugtenburg, J., Belenky, M., and Herzfeld, J. (2001) Proc.
Natl. Acad. Sci. USA, 98, 1571-1576.
48. Brown, L. S., and Lanyi, J. K. (1996) Proc. Natl. Acad. Sci.
USA, 93, 1731-1734.
49. Száraz, S., Oesterhelt, D., and Ormos, P. (1994) Biophys.
J., 67, 1706-1712.
50. Dioumaev, A. K., Brown, L. S., Needleman, R., and Lanyi,
J. K. (1998) Biochemistry, 37, 9889-9893.
BIOCHEMISTRY (Moscow) Vol. 66 No. 11 2001