Our understanding of the signalling mechanisms involved in the process of stomatal closure is reviewed. Work has concentrated on the mechanisms by which abscisic acid (ABA) induces changes in specific ion channels at both the plasmalemma and the tonoplast leading to efflux of both K
+
and anions at both membranes, requiring four essential changes. For each we need to identify the specific channels concerned, and the detailed signalling chains by which each is linked through signalling intermediates to ABA. There are two global changes that are identified following ABA treatment, an increase in cytoplasmic pH and an increase in cytoplasmic Ca
2+
, although stomata can close without any measurable global increase in cytoplasmic Ca
2+
. There is also evidence for the importance of several protein phosphatases and protein kinases in the regulation of channel activity.
At the plasmalemma, loss of K
+
requires depolarization of the membrane potential into the range at which the outward K
+
channel is open. ABA–induced activation of a non–specific cation channel, permeable to Ca
2+
, may contribute to the necessary depolarization, together with ABA–induced activation of S–type anion channels in the plasmalemma, which are then responsible for the necessary anion efflux. The anion channels are activated by Ca
2+
and by phosphorylation, but the precise mechanism of their activation by ABA is not yet clear. ABA also up–regulates the outward K
+
current at any given membrane potential; this activation is Ca
2+
–independent and is attributed to the increase in cytoplasmic pH, perhaps through the marked pH–sensitivity of protein phosphatase type 2C.
Our understanding of mechanisms at the tonoplast is much less complete. A total of two channels, both Ca
2+
–activated, have been identified which are capable of K
+
efflux; these are the voltage–independent VK channel specific to K
+
, and the slow vacuolar (SV) channel which opens only at non–physiological tonoplast potentials (cytoplasm positive). The SV channel is permeable to K
+
and Ca
2+
, and although it has been argued that it could be responsible for Ca
2+
–induced Ca
2+
release, it now seems likely that it opens only under conditions where Ca
2+
will flow from cytoplasm to vacuole. Although tracer measurements show unequivocally that ABA does activate efflux of Cl
–
from vacuole to cytoplasm, no vacuolar anion channel has yet been identified.
There is clear evidence that ABA activates release of Ca
2+
from internal stores, but the source and trigger for ABA–induced increase in cytoplasmic Ca
2+
are uncertain. The tonoplast and another membrane, probably ER, have IP
3
–sensitive Ca
2+
release channels, and the tonoplast has also cADPR–activated Ca
2+
channels. Their relative contributions to ABA–induced release of Ca
2+
from internal stores remain to be established. There is some evidence for activation of phospholipase C by ABA, by an unknown mechanism; plant phospholipase C may be activated by Ca
2+
rather than by the G–proteins used in many animal cell signalling systems.
A further ABA–induced channel modulation is the inhibition of the inward K
+
channel, which is not essential for closing but will prevent opening. It is suggested that this is mediated through the Ca
2+
–activated protein phosphatase, calcineurin.
The question of Ca
2+
–independent stomatal closure remains controversial. At the plasmalemma the stimulation of K
+
efflux is Ca
2+
–independent and, at least in
Arabidopsis
, activation of anion efflux by ABA may also be Ca
2+
–independent. But there are no indications of Ca
2+
–independent mechanisms for K
+
efflux at the tonoplast, and the appropriate anion channel at the tonoplast is still to be found.
There is also evidence that ABA interferes with a control system in the guard cell, resetting its set–point to lower contents, suggesting that stretch–activated channels also feature in the regulation of guard cell ion channels, perhaps through interactions with cytoskeletal proteins. There is evidence for involvement of actin in the control of guard cell ion channels, although possible mechanisms are still to be identified.
Stomatal closure involves net loss of vacuolar sugars as well as potassium salts, and there is an urgent need to address the question of the nature of the signalling chains linking transport and metabolism of sugars to the closing signal.
Stomatal immunity against fungal invasion comprises not only chitin-induced stomatal closure but also chitosan-induced guard cell death
