This review is focused on the chloroplast (i.e. thylakoid) ATP synthase-hyd
rolase (ATPase) but also refers to the bacterial and mitochondrial cases wh
en relevant. (1) A proton-translocating F-ATPase (CF-ATPase in chloroplasts
) comprises an intramembrane proton channel, F-0, and an extramembrane cata
lytic head, F-1, the latter forming a 'ball' of 3 alternative couples of al
pha and beta subunits plus lone copy of each gamma, delta, epsilon subunit;
alpha and beta subunits bear regulatory and catalytic nucleotide binding s
ites, respectively F-0 acts as a motor, with a stator made of subunits IV (
a in prokaryotes), I and II (b, b'), and a rotor arranged in a crown of 9-1
2 subunits III (c). Subunit gamma, anchored to the rotor, thus turns with t
he latter while its other extremity, entering into F-1, induces periodical
conformational changes that allow binding of substrates on one beta and rel
ease of products from another The energy is supplied by the proton gradient
built by the redox chain. Thus, one H+ from the high potential compartment
(thylakoid lumen) protonates, via part of subunit a, a neighboring subunit
c while another c nearby, facing the other part of a, is deprotonated and
the resulting H+ escapes into the low potential compartment (chloroplast st
roma). (2) The mechanistic stoichiometry H+/ATP of protons transported per
ATP formed was initially found to equal 2-3 but is now considered to equal
4. Such a shift may be due to the diversity of material (chloroplasts, mito
chondria, bacteria, liposomes), techniques (steady-state vs. pH or salt jum
ps) and approaches (kinetic: 'flow-flow', or thermodynamic: 'force-force' m
ethods). In mitochondria, after subtraction of the H+ due to P-i translocat
ion, H+/ATP is close to 3, perhaps due to a smaller number of c subunits (H
+/ATP = c/beta). (3) The osmotic (Delta pH) and electric (Delta Psi) compon
ents of total Delta<(mu)over tilde>(H+) are thermodynamically interconverti
ble and, despite some conflicting results, also seem kinetically equivalent
. This is thanks to a 'proton well', a scheme of which is presented. A rela
ted problem is whether phosphorylation depends on pH(in) and pH(out) or sol
ely on their difference (= Delta pH). No consensus exists on how the enzyme
affinity for ADP, expressed by its K-m, may depend on Delta<(mu)over tilde
>(H+). Two models are discussed. One is the simultaneous binding of an ADP
and release of an ATP, driven by the simultaneous translocation of 4 proton
s while the other introduces an intermediary step. (4) Besides being an ene
rgetic source for ATP synthesis, Delta<(mu)over tilde>(H+) switches ATPase
from inactive to active state and, specifically for CF1, exposes a disulfid
e bridge of gamma to thiol-reducing agents, thioredoxin in vivo (reduced by
Photosystem I), DTT in vitro. This reduction lowers the activation for thr
eshold Delta<(mu)over tilde>(H+) and at the same time prolongs the life-tim
e of the active enzyme from milliseconds to minutes after dissipation of De
lta<(mu)over tilde>(H+). By using ATPase inhibitors before or after activat
ion, we have suggested that activating and catalytic protons are different
(actually, activation does not necessarily require a proton flow). Activati
on is also accompanied - or governed - by conformational changes. In mitoch
ondria, there is an unbinding of a special subunit IF1 from F-1; its equiva
lent for CF1 could be a conformational change of epsilon. A final remark co
ncerns the physiological role of (de)activation, if any.
It was proposed that inactive state of CF could prevent futile hydrolysis o
f ATP in dark, but in fact, thanks to the low Delta<(mu)over tilde>(H+) thr
eshold required for activation of reduced CF1 and to low membrane permeabil
ity to protons in vivo, a minute consumption of ATP is sufficient to mainta
in a high enough Delta<(mu)over tilde>(H+) to oppose a counter-protonmotive
force limiting sustained ATP hydrolysis.