| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Biophys J, November 1999, p. 2331-2332, Vol. 77, No. 5
Groupe de Recherche en Transport Membranaire, Départements de Physique et Chimie, Université de Montréal, Montréal, Québec H3C 3J7 Canada
In this issue of the Biophysical Journal,
interesting experimental findings about proton fluxes through
gramicidin A (gA) channels and related analogs are reported. Phillips
et al. (1999) Phillips et al. (1999) The conduction of protons along a linear hydrogen bonded chain of water
molecules, a "proton wire," can be decomposed in two complementary
microscopic steps. The first step involves the rapid translocation of a
proton along the water chain by way of a Grotthuss mechanism, whereas
the second step involves the reorientation of the water chain. The
second step is necessary to complete the transport cycle of a proton
wire. After the translocation of the proton from entry to exit, the
water chain is left with all dipoles oriented in the opposite
direction. A preliminary reorientation of the water chain is thus
required to accept the next incoming proton. Using molecular dynamics
calculations based on atomic models, Pomes and Roux (1998) Transfer of H+ mediated by chains of water molecules is an
important mechanism in a wide range of proton transport phenomena involved in bioenergetics. For example, the functional importance of a
23-Å-long water chain observed in the photosynthetic reaction center
from Rhodobacter sphaeroides (Ermler et al., 1994 Understanding proton wires from a theoretical and computational point
of view is very challenging. In the long term, studies with analogs of
the gA channel such as reported by Phillips et al. (1999)
ARTICLE
compared the conduction of alkali metal ions and protons
for related analogs of the gA channel in which the dipole of the Trp
indole side chains were augmented (by fluorination) or decreased (by substitution with Phe) in two type of membranes with different headgroups. They observed that the rate-limiting factor governing H+ transport through the gA channel is modulated in an
opposite direction from that governing alkali metal cations upon
changes in the side chains or lipid headgroups. In particular, the
conductance of K+ through the gA channel in glyceryl
monoolein (GMO) membranes decreases when all four tryptophans
are substituted by phenylalanines, while the conductance of
H+ increases for the same substitution. When Trp-11 is
fluorinated, the K+ conductance increases, while the
conductance of H+ decreases. Interestingly, variations of
the lipid headgroups exhibit a similar trend: the conductance of
H+ in diphytanoyl phosphatidylcholine (DPhPC) membranes is
larger than in GMO membranes, contrary to the K+
conductance. These observations essentially demonstrate that transport
of H+ and alkali cations through the gA channel are
governed by qualitatively different mechanisms.
show that the observed variations in the
K+ conductance are correlated with the modulations in the
electrostatic field near the channel entrance. In particular, the
larger K+ conductance in GMO relative to DPhPC is
consistent with the interfacial dipole potential of GMO membranes,
which is larger than that of PC membranes (Pickard and Benz,
1978). Furthermore, the observed variations in the
K+ conductance are correlated with the modulations of the
indole dipole of the tryptophan side chains. Since they are oriented with the positive end of the indole dipole pointing toward the bulk
solution (Hu et al., 1993) the Trp side chains favor the passage of a
positive charge through the pore. Obviously, since proton also carry a
positive charge, the observed variations in the H+
conductance cannot be explained with such a mechanism. What could be
going on? The authors propose that part of the answer lies in a recent
theoretical study of a proton wire by Pomes and Roux (1998).
found that a
free energy barrier of a few kcal/mol opposes the reorientation of the
water chain, whereas the translocation of the H+ along the
chain is essentially barrierless. Furthermore, the simulations showed
that the rate-limiting water reorientation in a single file of hydrogen
bonded water molecules proceeds sequentially, i.e., it is initiated at
the end of one chain and the transition state occurs when the hydrogen
bond defect reaches the center. The observed variations in the
H+ conductance reported by Phillips et al. (1999) may be
the first evidence for the rate-limiting water reorientation step in
the transport cycle of a proton wire.
) was demonstrated by site-directed mutagenesis (Baciou and Michel, 1995).
Similarly, perturbation of the internal water chain in cytochrome
f shows a loss of the concerted reduction of cytochrome f and b6 (Ponamarev and Cramer,
1998). Nonetheless, the properties of proton fluxes in these complex
biological systems are not easy to characterize, and one is seeking
simpler model systems to investigate. The gA and its related analogs
provide good prototypical systems for investigating proton transport
along a hydrogen bonded water chain. Several aspects of proton
transport through the gA channel are fascinating and generate continued
interest (Cukierman, 1999; DeCoursey and Cherny, 1999). In another
article in this issue, Quigley et al. (1999) investigate the proton
fluxes through RR and SS dioxolane-linked gA analogs. Their
measurements show that dioxolane-linked channels exhibit stereospecific
modulation of H+ conductance. The proton conductance in the
SS-linked dimer is about 2 to 4 times larger than in the RR-linked
dimer. The authors propose that this reflects differences in
water-water and water-channel hydrogen bonding in the two stereoisomers
affecting the rate of the reorientation step. Although this explanation
is consistent with the analysis of Phillips et al. (1999), a definitive
conclusion will require a comparison of the transport properties of
H+ with those of alkali cations in the two stereoisomers.
and Quigley
et al. (1999) provide important information about H+
conduction along proton wires which will be essential for guiding the
construction of theoretical models. Nonetheless, the conclusions from
computer simulations of proton wires will remain speculative unless
further information is obtained at the atomic level from experiments. A
key question is, how can we learn more about proton wires at the
microscopic level? For example, x-ray scattering (Olah et al., 1991)
and nuclear magnetic resonance (Smith et al., 1990; Jing et al., 1995;
Tian et al., 1996) have provided a wealth of information about the
interaction of alkali cations with the gA. Perhaps those powerful
approaches could be used to investigate the gA proton wire. Lastly,
other experimental techniques such as Fourier transform infrared
spectroscopy offer promising avenues. Recently, Bartl et al. (1998)
detected large collective proton motions in the interior of the gA
channel using this technique. Meanwhile, the recent and interesting
results of Phillips et al. (1999) and Quigley et al. (1999) contribute
to the current reflection about proton transport. This is a story to follow.
| |
ACKNOWLEDGMENTS |
|---|
Discussions with David Busath, Sam Cukierman, and Régis Pomes are gratefully acknowledged.
| |
FOOTNOTES |
|---|
Received for publication 14 September 1999 and in final form 14 September 1999.
Address reprint requests to Benoit Roux, Dept. Chemistry, Rm. D603, C.P. 6128, Succursale Centre-Ville, Montreal, Quebec H3C 3J7, Canada. Tel.: 514-343-7105; Fax: 514-343-7586; E-mail: rouxb{at}plgcn.umontreal.ca.
| |
REFERENCES |
|---|
Biophys J, November 1999, p. 2331-2332, Vol. 77, No. 5
© 1999 by the Biophysical Society 0006-3495/99/11/2331/02 $2.00
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
