Membrane trafficking in guard cells during stomatal movement

Abstract

Pairs of guard cells form small pores called stoma in the epidermis, and the reversible swelling and shrinking of these guard cells regulate the stomatal apertures. The well-documented changes in guard cell volume have been associated with their vacuolar structures. To investigate the contribution of the guard cell vacuoles to stomatal movement, the dynamics of these vacuolar structures were recently monitored during stomatal movement in vacuolar-membrane visualized Arabidopsis plants. Calculation of the vacuolar volume and surface area after reconstruction of three-dimensional images revealed a decrease in the vacuolar volume but an increase in the vacuolar surface area upon stomatal closure. These results implied the possible acceleration of membrane trafficking to the vacuole upon stomatal closure and membrane recycling from the vacuole to the plasma membrane upon stomatal opening. To clarify and quantify membrane trafficking during stomatal movement, we describe in this addendum our development of an improved image processing system.

Stomatal Vacuolar Dymamics During Stomatal Closure

Ever since a 100 years ago, when Francis Darwin discovered the stomatal responses to light,¹ the mechanism of stomatal movement upon perception of environmental stimuli has been a topic of intense investigation^2–4 and the volume change of their surrounding guard cells has been demonstrated to regulate the pore size.

In general, plant cells expand by increasing the volume of their vacuoles. Guard cells also use this large organelle to affect changes in their volumes, with parallel increases and decreases in vacuolar volumes during stomatal opening and closure, respectively.⁵ To investigate the contribution of guard cell vacuoles to stomatal movement, we recently examined guard cell vacuolar dynamics using a transgenic Arabidopsis line expressing GFP-AtVAM3 that visualized the vacuolar membrane (VM).⁶ The VM structures in guard cells were simple when the stomata were open, but became complicated and contained luminal vacuolar structures when they were closed.⁶ A reconstruction of the three-dimensional (3-D) images by our originally-developed software from successive images taken along the z-axis showed a large, topologically-connected vacuole. In addition, this software allowed us to quantify the structural volumes and their surface areas.⁷ Our calculations demonstrated a decrease of about 20% in the volumes of the guard cells and their vacuoles and in the surface area of the guard cells during stomatal closure,⁶ consistent with previous observations (refs. ^{5, 8–10}). In contrast, we found a 20% increase in the vacuolar surface area during stomatal closure and an accumulation of the membrane dye, FM4–64, in the VM.⁶

Membrane Trafficking During Stomatal Movement

The above increases and decreases in surface area could not be explained by simple membrane stretching and shrinkage since biomembranes have been shown to possess no more than about 3% elasticity.¹¹ In this context, the decrease in guard cell surface area was found to result in internalization of the plasma membrane by endocytosis, as monitored using membrane staining dyes or a marker GFP-tagged plasma membrane K⁺ channel protein, KAT1.^9,12,13 During stomatal movement, the endosytosis was triggered by ABA application and ABA washout recycled the internalized membrane material to the plasma membrane.¹⁴

Both the increase in the VM and accumulation of the dye in the VM suggested an acceleration of membrane trafficking towards vacuoles during stomatal closure. Such increases in the VM have been suggested to act as membrane reservoirs for membrane reversion.¹⁵ Potential membrane sources are thought to include the ER and endosomes,6 since the ER supplies many vacuolar proteins¹⁶ and the endosomes act as a membrane store of the plasma membrane.¹⁷ Although the mechanism that accelerates membrane trafficking to the vacuole has not yet been clarified, it is known that endosomal proteins are either recycled to the plasma membrane or targeted to the vacuole for degradation.^18,19 To facilitate fusion of endosomes with the vacuoles, some endosomal proteases have also been implicated.²⁰

Image Processing to Investigate Membrane Trafficking

Can the increased vacuolar membranes be recycled to the plasma membrane upon stomatal opening? Recycling these lipids appears to be an economical strategy that allows rapid stomatal movement without de novo lipid synthesis. However, the mechanism of lipid trafficking and the mode of plasma membrane maturation from the vacuolar membranous lipids are still open to debate. To clarify the process of endomembrane trafficking to vacuoles, we measured the fluorescence intensities of FM4-64 in the plasma and vacuolar membranes and endosomes.⁶ While fluorescence intensities were found to correlate with FM4-64 densities,²¹ we found that the obtained intensities were biased by the angle of the objects and were therefore unsuitable for quantitative comparisons. In our measurements, even though we employed the maximum intensities in every membrane component to reduce the bias, we could still only approximate the general tendency of endocytotic activity. The bias has been shown to arise from the optical character of fluorescence microscopy, as well as of confocal laser scanning microscopy.²² For example, when we measured the fluorescence distribution of a small globular bead along the x, y, and z-axes by capturing its 3-D image (point-spread function, PSF), the distribution was found to be elongated along the z-axis in the resulting image (Fig. 1A). Therefore, because of the bias introduced by this PSF anisotropy,^23,24 by which objects along the z-axis appeared brighter and horizontal ones appeared darker (Fig. 1B), it was difficult to estimate the density of fluorophores at the membranous structures. We are now in the process of developing an image processing algorithm to compensate for the bias from PSF anisotropy of membranous objects. In our preliminary trials, measurements of FM4–64 fluorescence intensities from tobacco BY-2 cell confocal images (Fig. 1C) demonstrated reduced errors by incorporation of the compensation (Fig. 1D).

Figure 1.

Measurement of fluorescence intensities in membranous structures. (A) A XZ-section of point-spread function (PSF) obtained from a 0.22 µm fluorescent bead. Note that the fluorescence distribution is elongated along the z-axis compared to the x-axis. The white circle represents the real dimensions of the fluorescent bead. Image acquisition was achieved by spinning-disc confocal laser scanning microscopy (olympus IX70 with Yokogawa Electric CSU 10 scanning unit) using a 60× oil immersion objective lens (olympus PlanApo; NA 1.40). Bar: 1 µm. (B) Intensity bias observed in a virtual membranous structure caused by PSF anisotropy. Relative fluorescence intensities of the objects were attained every 10 degrees to the z-axis. n = 10. Error bar: standard deviation. Inset: Images of the virtual membranous structure after reconstruction from successive images along the z-axis at 0.5 mm intervals. Note that the intensity of objects parallel to the z-axis appeared bright whereas the more horizontal objects appeared darker. (C) Confocal image of a tobacco BY-2 cell 4 h after pulse-labeling with the FM4-64 membrane dye. The arrow indicates the vacuolar membrane. Bar: 10 µm. (D) Fluorescence intensities of FM4–64 in the vacuolar membrane. ◆, Raw intensities in confocal images. , Intensities after compensation to reduce bias. n = 10. Note that the errors of the intensities were reduced after compensation.

Image processing techniques have allowed us to reconstruct the 3-D images of guard cell vacuoles and to calculate their structural volumes and surface area.⁶ Our next mission is to clarify and quantify lipid dynamics in the guard cells during stomatal movement by improving the image processing system.

Abbreviations

PSF

point-spread function

VM

vacuolar membrane

Addendum to: Tanaka Y, Kutsuna N, Kanazawa Y, Kondo N, Hasezawa S, Sano T. Intra-vacuolar reserves of membranes during stomatal closure: The possible role of guard cell vacuoles estimated by 3-D reconstruction. Plant Cell Physiol. 2007;48:1159–1169. doi: 10.1093/pcp/pcm085.