Water loss from leaves is regulated by the state of stomatal pores, whose aperture is controlled by the level of potassium salt accumulation in guard cells. In water stress conditions abscisic acid (ABA), produced or imported into leaves, and acting on the outside of the guard cell induces net loss of potassium salts, and hence stomatal closure. The mechanism of ABA-induced closure and the role of calcium in the process are discussed. There are two questions at issue, whether Ca
2+
-regulated fluxes of specific ions are an obligatory part of the signal cascade, and if this is the case, w hether the necessary ABA-induced increase in cytoplasmic Ca
2+
arises from Ca
2+
influx at the plasmalemma, or by Ca
2+
release from internal stores, or both. Tracer flux studies establish that ABA-induced closure involves transient stimulation of both anion and cation fluxes at the plasmalemma, and stimulation of the transfer of both anions and cations from vacuole to cytoplasm. ABA-induced efflux transients can occur in very low external Ca
2+
, but their reduction in the presence of La
3+
suggests that Ca
2+
influx is required for the response. The flux work can only be interpreted in terms of defined ion channels identified by electrical work, either whole-cell voltage clamping or patch clamp studies, and of the responses of these channels to Ca
2+
and to ABA. Electrical work identifies a number of ion channels in the plasmalemma; these include an inward K
+
channel open at negative membrane potentials, and inhibited by increase in cytoplasmic Ca
2+
, an outward K
+
channel open at more positive membrane potentials, which is insensitive to Ca
2+
but is more active at higher pH, a voltage-sensitive, Ca
2+
-dependent anion channel, active only over a restricted range of potentials (about — 100 mV to —50 mV), and some ill-defined conductances lumped together as the ‘leak’ or background conductance, which may include channels (selective or nonselective) allowing Ca
2+
influx. The leak conductance is increased by increase in cytoplasmic Ca
2+
. Guard cells are capable of responding to inositol 1,4,5-trisphosphate released in the cytoplasm, by increasing cytoplasmic Ca
2+
, by inhibition of the inward K
+
channel and by stimulation of the leak conductance (but w ithout effect on the outw ard K
+
channel), and by stomatal closure. Recent work suggests that there is considerable turnover in the phosphoinositide cycle in guard cells, within 30 s of treatment with ABA. Measurements by fluorescence techniques of cytoplasmic Ca
2+
in guard cells following treatment with ABA give conflicting results. Some work shows increase in cytoplasmic Ca
2+
in response to ABA, other studies show variable behaviour, with most cells closing in response to ABA, but without detectable changes in cytoplasmic Ca
2+
. Nevertheless it seems likely that increases in cytoplasmic Ca
2+
, at least locally, are a universal feature of the ABA response, but that they may be difficult to detect with present techniques. Fluorescence studies also show alkalinization of guard cell cytoplasm in response to ABA. Whole cell electrical studies identify a number of ABA-induced changes. They show (i) depolarization of cells with very negative membrane potentials to potentials which are positive to
E
K
, and thus out of the activation range for the inward K
+
channel, and within the range for the outward K
+
channel, (ii) activation of an inward current, a voltage-insensitive component of the leak conductance, which is responsible for the depolarization, (iii) deactivation of the inward K
+
channel, (iv) activation of a voltage-sensitive channel carrying inward current, probably the Ca
2+
-sensitive anion channel, and (v) the slower activation of the outward K
+
channel. The activation of the inward leak current seems to be the primary response, but its nature is not clearly established; a non-selective cation channel, which may allow Ca
2+
influx, is perhaps most likely. Thus the early events in the ABA-response include stimulation of tracer efflux, activation of an ill-defined component of the leak conductance, producing an inward current, and turnover in the phosphoinositide cycle. These occur within the first minute, but their time sequence and causal relationships are not yet clear. A plausible scheme for ABA-induced closure can be devised, involving Ca
2+
influx through a non-selective cation channel as the first event, producing depolarization and increase (possibly local) in cytoplasmic Ca
2+
. This may then be supplemented by release of Ca
2+
from internal stores, triggered by inositol 1,4,5-trisphosphate produced by activation of phospholipase C. Increase in cytoplasmic Ca
2+
will give deactivation of the inward K
+
channel, and activation of the Ca
2+
-dependent anion channel, but some other trigger is required to explain the activation of the outward K
+
channel; increase in cytoplasmic pH (observed, but of mechanism unknown) is the most likely candidate. This is one scheme, but others can also be devised, and with the gaps still existing in our description of the events involved, and their time sequence, a definitive hypothesis is not yet available.
Mitochondria accumulate Ca2+ following intense glutamate stimulation of cultured rat forebrain neurones.