Trafficking, lateral mobility and segregation of the plant K channel KAT1.

Functioning of ion channels depends not only on the control of their activity but also on their density and lateral organisation in the membrane. The two latter aspects have recently attracted increasing attention. Here, we summarize studies on trafficking and plasma membrane distribution of the plant K(+) channel KAT1 from Arabidopsis thaliana. In guard cells, KAT1 was found to be subject to constitutive and pressure-driven turnover and ABA-stimulated endocytosis. These results point to a role of exo- and endocytosis in regulating KAT1 density and thus ion transport during guard cell functioning. Recent studies indicate that KAT1 density can also be adjusted at the site of ER export. Efficient ER export of KAT1 was shown to depend on an acidic motif that interacts with Sec24, a component of ER-derived vesicles. Surface expression of ER export mutants of KAT1 can be rescued through heterotetrameric assembly with wild-type KAT1, implying that not all subunits of the channel tetramer need to carry an ER export motif. Analysis of the distribution of KAT1 in the plasma membrane revealed segregation of the channel into microdomains, and low lateral mobility in both plant and mammalian cells. In plant cells, SNAREs have been shown to be involved in anchoring KAT1 in the plasma membrane. Studies on guard cells imply a role for the cell wall in organisation of KAT1 microdomains. Together, these findings underline the importance of investigating mechanisms of KAT1 trafficking and lateral organisation in order to fully understand channel functioning.

Intracellular localisation of PPI1 (proton pump interactor, isoform 1), a regulatory protein of the plasma membrane H(+)-ATPase of Arabidopsis thaliana.

PPI1 (proton pump interactor isoform 1) is a novel protein able to interact with the C-terminal autoinhibitory domain of the Arabidopsis thaliana plasma membrane (PM) H(+)-ATPase. In vitro, PPI1 binds the PM H(+)-ATPase in a site different from the known 14-3-3 binding site and stimulates its activity. In this study, we analysed the intracellular localisation of PPI1. The intracellular distribution was monitored in A. thaliana cultured cells by immunolocalisation using an antiserum against the PPI1 N-terminus and in Vicia faba guard cells and epidermal cells by transient expression of a GFP::PPI1 fusion. The results indicate that the bulk of PPI1 is localised at the endoplasmic reticulum, from which it might be recruited to the PM for interaction with the H(+)-ATPase in response to as yet unidentified signals.

Guard cells undergo constitutive and pressure-driven membrane turnover.

During stomatal movement, guard cells undergo large and reversible changes in cell volume and consequently surface area. These alterations in surface area require addition and removal of plasma membrane material. How this is achieved is largely unknown. Here we summarize recent studies of membrane turnover in guard cells using electrophysiology and fluorescent imaging techniques. The results implicate that membrane turnover in guard cells and most likely in plant cells in general is sensitive to changes in membrane tension. We suggest that this provides a mechanism for the adaptation of surface area of guard cells to osmotically driven changes in cell volume. In addition, guard cells also exhibit constitutive membrane turnover. Constitutive and pressure-driven membrane turnover were found to be associated with addition and removal of K+ channels. This implies that some of the exo- and endocytic vesicles carry K+ channels. Together the results demonstrate that exo- and endocytosis is an essential process in guard cell functioning.

Cytochalasin D attenuates the desensitisation of pressure-stimulated vesicle fusion in guard cell protoplasts.

Fusion of vesicular membranes with the plasma membrane during pressure-driven swelling of guard cell protoplasts was studied using patch clamp capacitance measurements. Hydrostatic pressure pulses were applied via the patch pipette and resulted in an immediate and linear increase in membrane capacitance, a parameter proportional to the surface area. In any given protoplast, pressure-stimulated increases in membrane capacitance could be provoked repetitively. However, the rate of rise in capacitance upon the same strength of stimulation decreased exponentially with time (tau = 4 min) for subsequent pressure stimuli. This process was the result of a desensitisation of the plasma membrane to mechanical forces. Incubation of guard cell protoplasts in cytochalasin D, which depolymerises actin filaments, nearly abolished this desensitisation process. These results suggest that membrane stretch initiates a reactive process that may fortify or stabilise the plasma membrane of guard cell protoplasts.

Cell surface area regulation and membrane tension.

The beautifully orchestrated regulation of cell shape and volume are central themes in cell biology and physiology. Though it is less well recognized, cell surface area regulation also constitutes a distinct task for cells. Maintaining an appropriate surface area is no automatic side effect of volume regulation or shape change. The issue of surface area regulation (SAR) would be moot if all cells resembled mammalian erythrocytes in being constrained to change shape and volume using existing surface membrane. But these enucleate cells are anomalies, possessing no endomembrane. Most cells use endomembrane to continually rework their plasma membrane, even while maintaining a given size or shape. This membrane traffic is intensively studied, generally with the emphasis on targeting and turnover of proteins and delivery of vesicle contents. But surface area (SA) homeostasis, including the controlled increase or decrease of SA, is another of the outcomes of trafficking. Our principal aims, then, are to highlight SAR as a discrete cellular task and to survey evidence for the idea that membrane tension is central to the task. Cells cannot directly "measure" their volume or SA, yet must regulate both. We posit that a homeostatic relationship exists between plasma membrane tension and plasma membrane area, which implies that cells detect and respond to deviations around a membrane tension set point. Maintenance of membrane strength during membrane turnover, a seldom-addressed aspect of SA dynamics, we examine in the context of SAR. SAR occurs in both animal and plant cells. The review shows the latter to be a continuing source of groundbreaking work on tension-sensitive SAR, but is principally slanted to animal cells.

Ca(2)+-stimulated exocytosis in maize coleoptile cells.

Changes in membrane capacitance (C(m)) after photolysis of the caged Ca(2)+ compound dimethoxynitrophenamine were studied in protoplasts from maize coleoptiles. Changes in C(m) values resulting from increased concentrations of free Ca(2)+ in the cytoplasm ([Ca(2)+](cyt)) were interpreted as representing changes in [Ca(2)+](cyt)-sensitive exocytosis and endocytosis. A continuous increase in [Ca(2)+](cyt) resulted in a sigmoidal increase in C(m) values with a half-maximal concentration at approximately 1 microM. The steep increase in C(m) values was followed by a variable slow phase in changing C(m) values. When [Ca(2)+](cyt) increased at a rate of 0.6 micromol L(-)(1) sec(-)(1), the initial steep increase in C(m) values lasted approximately 5 to 10 sec. During this time, protoplasts increased in surface area by approximately 2.5%. The biphasic dynamics of [Ca(2)+](cyt)-stimulated increases in C(m) values can be described by a kinetic model containing two pools of vesicles with two [Ca(2)+](cyt)-sensitive steps in the exocytotic pathway.

Osmotically evoked shrinking of guard-cell protoplasts causes vesicular retrieval of plasma membrane into the cytoplasm.

The dye FM1-43 was used alone or in combination with measurements of the membrane capacitance (Cm) to monitor membrane changes in protoplasts from Vicia faba L. guard cells. Confocal images of protoplasts incubated with FM1-43 (10 microM) at constant ambient osmotic pressure (pi omicron) revealed in confocal images a slow internalisation of FM1-43-labelled membrane into the cytoplasm. As a result of this process the relative fluorescence intensity of the cell interior (fFM,i) increased with reference to the total fluorescence (fFM,t) by 7.4 x 10(-4) min(-1). This steady internalisation of dye suggests the occurrence of constitutive endocytosis under constant osmotic pressure. Steady internalisation of FM1-43 labelled membrane caused a prominent staining of a ring-like structure located beneath the plasma membrane. Abrupt elevation of pi omicron by 200 mosmol kg(-1) caused, over the first minutes of incubation, a rapid internalisation of FM1-43 fluorescence into the cytoplasm concomitant with a decrease in cell perimeter. Within the first 5 min the cell perimeter decreased by 7.9%. Over the same time fFM,i/fFM,t increased by 0.13, reflecting internalisation of fluorescent label into the cytoplasm. Combined measurements of Cm and total fluorescence of a protoplast (fFM,p) showed that an increase in pi omicron evoked a decrease in Cm but no change in fFM,p. This means that surface contraction of the protoplast is due to retrieval of excess membrane from the plasma membrane and internalisation into the cytoplasm. Further inspection of confocal images revealed that protoplast shrinking was only occasionally associated with internalisation of giant vesicles (median diameter 2.7 microm) with FM 1-43-labelled membrane. But, in all cases, osmotic contraction was correlated with a diffuse distribution of FM1-43 label throughout the cytoplasm. From this, we conclude that endocytosis of small vesicles into the cytoplasm is the obligatory process by which cells accommodate an osmotically driven decrease in membrane surface area.

Transient and permanent fusion of vesicles in Zea mays coleoptile protoplasts measured in the cell-attached configuration.

Exocytosis in protoplasts from Zea mays L. coleoptiles was studied using patch-clamp techniques. Fusion of individual vesicles with the plasma membrane was monitored as a step increase of the membrane capacitance (Cm). Vesicle fusion was observed as (i) An irreversible step increase in Cm. (ii) Occasionally, irreversible Cm steps were preceded by transient changes in Cm, suggesting that the electrical connection between the vesicle with the plasma membrane opens and closes reversibly before full connection is achieved. (iii) Most frequently, however, stepwise transient changes in Cm did not lead to an irreversible Cm step. Within one patch of membrane capacitance steps due to transient and irreversible fusions were of similar amplitude. This suggests that the exocytosis events do not result from the fusion of vesicles with different sizes but are due to kinetically different states in a fusion process of the same vesicle type. The dwell time histogram of the transient fusion events peaked at about 100 msec. Fusion can be described with a circular three-state model for the fusion process of two fused states and one nonfused state. It predicts that energy input is required to drive the system into a prevailing direction.

Ca2+-sensitive and Ca2+-insensitive exocytosis in maize coleoptile protoplasts.

Ca2+ and osmotic driven extension of the surface area of maize coleoptile protoplasts was investigated using capacitance measurements and photolysis of the caged compound DM-nitrophen. Protoplasts responded to an elevation of cytoplasmic Ca2+ (Ca(i)) with a rapid burst in capacitance reaching a maximal increase of 1.3+/-1.1% over the resting cell capacitance. Subsequent lowering of the osmotic potential in the external medium by 210 mosmol caused a further increase in Cm by 26+/-6%. These data indicate two independent pathways for insertion of membrane into the plasma membrane. One is driven by Ca(i) and recruits membrane from a small pool. The osmotic evoked rise in surface area draws membrane from a much larger reservoir and may be driven by membrane tension.

Unitary exocytotic and endocytotic events in guard-cell protoplasts during osmotically driven volume changes.

Osmotically driven swelling and shrinking of guard-cell protoplasts (GCPs) requires adjustment of surface area which is achieved by addition and removal of plasma membrane material. To investigate the mechanism for adaptation of surface area we have used patch-clamp capacitance measurements. The recorded membrane capacitance (C(m)) trace of swelling and shrinking GCPs occasionally revealed discrete upward and downward deflecting capacitance steps, respectively, with a median value of about 2 fF. The observed capacitance steps resulted from the fusion and fission of single vesicles with a diameter of around 300 nm. We conclude that exo- and endocytosis of these vesicles accommodate for osmotically driven surface area changes in GCPs.

Ca2+-independent and Ca2+/GTP-binding protein-controlled exocytosis in a plant cell.

Exocytosis allows the release of secretory products and the delivery of new membrane material to the plasma membrane. So far, little is known about the underlying molecular mechanism and its control in plant cells. We have used the whole-cell patch-clamp technique to monitor changes in membrane capacitance to study exocytosis in barley aleurone protoplasts. To investigate the involvement of Ca2+ and GTP-binding proteins in exocytosis, protoplasts were dialyzed with very low (<2 nM) and high (1 microM) free Ca2+ and nonhydrolyzable guanine nucleotides guanosine 5'-gamma-thio]triphosphate (GTP[gammaS]) or guanosine 5'-[beta-thio]diphosphate (GDP[betaS]). With less than 2 nM cytoplasmic free Ca2+, the membrane capacitance increased significantly over 20 min. This increase was not altered by GTP[gammaS] or GDP[betaS]. In contrast, dialyzing protoplasts with 1 microM free Ca2+ resulted in a large increase in membrane capacitance that was slightly reduced by GTP[gammaS] and strongly inhibited by GDP[betaS]. We conclude that two exocytotic pathways exist in barley aleurone protoplasts: one that is Ca2+-independent and whose regulation is currently not known and another that is stimulated by Ca2+ and modulated by GTP-binding proteins. We suggest that Ca2+-independent exocytosis may be involved in cell expansion in developing protoplasts. Ca2+-stimulated exocytosis may play a role in gibberellic acid-stimulated alpha-amylase secretion in barley aleurone and, more generally, may be involved in membrane resealing in response to cell damage.

Ion channel activity during the action potential in Chara: new insights with new techniques.

The dynamics of macroscopic currents underlying the electrically triggered action potential (AP) in the giant alga Chara corallina were directly recorded with an action potential clamp method. In this technique an AP is recorded and repetitively replayed as the command voltage to the same cell under voltage control. Upon adding the channel blockers niflumic acid and/or Ba(2+) to the bath, the excitation current, i.e. the current crossing the membrane during an AP, can be dissected into a transient, fast-appearing Cl(-) inward current and a transient delayed K(+) outward current. The delayed onset of the K(+) outward current demands the postulation of an additional outward current in order to balance the excess Cl(-) inward current at the onset of the AP. The capacitive current that alters the charge on the membrane during excitation is several orders of magnitude too small to be relevant for charge balance. Measurements of single channel activity in the plasma membrane of C. corallina by the patch clamp method shows two types of Cl(-) channel (15 and 38 pS with 100 mM Cl(-) in the pipette) and one type of K(+) channel (about 40 pS with 100 mM K(+) in the pipette) which become transiently active during an AP. Typically, variable numbers of CI(-) channels activate in a random fashion for short periods of time when favoured by positive voltages in combination with high concentrations of extracellular Ca(2+) (Ca(2+)(o)) or during an AP of the whole cell. The peak values of these Cl(-) channel currents measured in a patch are such that they can account quantitatively for the peak of the whole cell Cl(-) excitation current studied under comparable ionic conditions. Furthermore, the short dura- tion of channel activity, as well as the fast rising and somewhat slower trailing kinetics is similar in duration and dynamics to AP-associated changes in membrane permeability of the whole Chara cell to Cl(-) (P(Cl(-))). Taken together, the data stress that the characteristic, transient activation of random numbers of Cl(-) channels seen in membrane patches is the elementary unit of the Cl(-) excitation current. However, due to the random nature of this transient activity, gating of Cl(-) channels can not be explained on the basis of previous models for excitation: gating can neither be due to intrinsic voltage sensitivity of the Cl(-) channels, nor to a voltage-dependent influx of Ca(2+) and subsequent activation of Ca(2+)-sensitive Cl(-) channels. To account for the short life-time and for the randomness of Cl(-) channel activity, the putative gating factors Ca(2+) and voltage must be uncoupled in time. This could be explained by a random release of Ca(2+) from stores, the latter being filled in a voltage-sensitive manner via non-specific cation channels from the outside. A 4 pS non-selective cation channel in the plasma membrane may serve this purpose. The 40 pS K(+) channel, which becomes transiently active in C. corallina during a cell AP, is an outward rectifier. At negative resting voltages the channel has a low open probability (< <1%). At voltages reached during an AP the open probability rises significantly reaching half-maximal open probability at -25 mV. The elevated activity of the 40 pS channel associated with membrane excitation relaxes at the end of an AP with a time constant of about 2.5 s. A comparable time constant of 2 s can be obtained for the decay of the transiently elevated permeability of the membrane to K(+) (P(K(+))), stressing that the kinetic properties of the 40 pS K(+) channel are responsible for the course of whole cell P(K(+)) changes. Voltage sensitivity of the K(+) channels suggests that they are activated during an AP by the drop in membrane voltage in order to aid repolarization. However, the rise and decay of P(K(+)) during an AP also shares similarity with the time-course of transient changes in cytoplasmic concentration of free Ca(2+), [Ca(2+)](cyt), the latter being measured in parallel experiments with the Ca(2+)-sensitive fluorescent dye, Fura-2, in excited C. corallina cells. This similarity could suggest that gating of the 40 pS K(+) channel is also sensitive to [Ca(2+)](cyt) and that the latter sensitivity is rate-limiting for activity during an AP.

Cl- and K+ channel currents during the action potential in Chara. Simultaneous recording of membrane voltage and patch currents.

Patch currents in the cell attached-mode and action potentials (AP) have been recorded simultaneously in internodal cells of Chara corallina. The action potentials are closely correlated with transient patch currents. With pipettes containing either 50 mM CaCl2 or 100 mM KCl plus 1 or 5 mM CaCl2, these transients measured up to 100 to 200 pA per patch at zero mV. Transients had a mean duration (time during which the current was > or = half maximum peak amplitude) of about 1 sec, a maximum slope for current rising of about 400 pA sec-1 and a maximum rate of about 100 pA sec-1 for current decay, with no obvious effect of external Ca2+ on either of these parameters. In well-resolved recordings of current transients triggered by an action potential (AP), activities of two types of Cl--conducting channels (15 and 38 pS) have been identified. Since activity of these channels was only observed during action potentials but not upon positive voltage steps, these channels are not directly voltage gated but point to a cytoplasmic gating factor which accumulates during excitation and propagates from excited areas to the patch. A K(+)-conducting channel (40 pS) could be identified as well during an AP, when 100 mM KCl was in the pipette solution. The activity of this channel relaxed at the end of the APs with a time constant of about 3 sec. Stimulated activity of this channel is understood to cause the repolarization overshoot during the final phase of the action potential, whereas the transient activation of the Cl- channels determines the fast voltage changes of the action potential.

Microscopic elements of electrical excitation in Chara: transient activity of Cl- channels in the plasma membrane.

The plasma membrane of Chara corallina was made accessible for patch pipettes by cutting a small window through the cell wall of plasmolyzed internodal cells. With pipettes containing Cl- as Ca2+ or Ba2+ (50 or 100 mM), but not as Mg2+ or K+ salt, it was possible to record in the cell-attached mode for long periods with little channel activity, randomly interspersed with intervals of transient activation of two Cl- channel types (cord conductance at +50 mV: 52 and 16 pS, respectively). During these periods of transient channel activity, variable numbers (up to some 10) of the two Cl- channel types activated and again inactivated over several 100 msec in a coordinated fashion. Transient Cl- channel activity was favored by voltages positive of the free running membrane voltage (> -45 mV); but positive voltage alone was neither a sufficient nor a necessary condition for activation of these channels. Neither type of Cl- channel was markedly voltage dependent. A third, nonselective 4 pS channel is a candidate for Ca2+ translocation. The activity of this channel does not correlate in time with the transient activity of the Cl- channels. The entire set of results is consistent with the following microscopic mechanism of action potentials in Chara, concerning the role of Ca2+ and Cl- for triggering and time course: Ca2+ uptake does not activate Cl- channels directly but first supplies a membrane-associated population of Ca2+ storage sites. Depolarization enhances discharge of Ca2+ from these elements (none or few under the patch pipette) resulting in a local and transient increase of free Ca2+ concentration ([Ca2+]cyt) at the inner side of the membrane before being scavenged by the cytoplasmic Ca2+ buffer system. In turn, the transient rise in [Ca2+]cyt causes the transient activity of those Cl- channels, which are more likely to open at an elevated Ca2+ concentration.