We describe a protocol using chamber slides and media to immobilize plant cotyledons for confocal imaging of the epidermis over several days of development, documenting stomatal differentiation. Fluorophore-tagged proteins can be tracked dynamically by expression and subcellular localization, increasing understanding of their possible roles during cell division and cell-type differentiation.
Imaging in vivo dynamics of cellular behavior throughout a developmental sequence can be a powerful technique for understanding the mechanics of tissue patterning. During animal development, key cell proliferation and patterning events occur very quickly. For instance, in Caenorhabditis elegans all cell divisions required for the larval body plan are completed within six hours after fertilization, with seven mitotic cycles1; the sixteen or more mitoses of Drosophila embryogenesis occur in less than 24 hr2. In contrast, cell divisions during plant development are slow, typically on the order of a day 3,4,5 . This imposes a unique challenge and a need for long-term live imaging for documenting dynamic behaviors of cell division and differentiation events during plant organogenesis. Arabidopsis epidermis is an excellent model system for investigating signaling, cell fate, and development in plants. In the cotyledon, this tissue consists of air- and water-resistant pavement cells interspersed with evenly distributed stomata, valves that open and close to control gas exchange and water loss. Proper spacing of these stomata is critical to their function, and their development follows a sequence of asymmetric division and cell differentiation steps to produce the organized epidermis (Fig. 1).
This protocol allows observation of cells and proteins in the epidermis over several days of development. This time frame enables precise documentation of stem-cell divisions and differentiation of epidermal cells, including stomata and epidermal pavement cells. Fluorescent proteins can be fused to proteins of interest to assess their dynamics during cell division and differentiation processes. This technique allows us to understand the localization of a novel protein, POLAR6, during the proliferation stage of stomatal-lineage cells in the Arabidopsis cotyledon epidermis, where it is expressed in cells preceding asymmetric division events and moves to a characteristic area of the cell cortex shortly before division occurs. Images can be registered and streamlined video easily produced using public domain software to visualize dynamic protein localization and cell types as they change over time.
1. Seed Sterilization
2. Preparation of Chamber Slide Media
3. Seed Dissection
4. Mounting Cotyledons in the Chamber Slide
5. Time Lapse Imaging
6. Video Editing
A set of informative time points collected with this method is shown in Figure 3. Cell membranes are labeled with RFP (pm-rb) and GFP is fused to POLAR protein under its native promoter (POLAR::POLAR-GFP)6 At the 30-min time scale, we see cell divisions along with the changes in protein localization preceding them. Asymmetric cell divisions in the stomatal lineage form stem-cell-like stomatal precursors called meristemoids, which retain the ability to undergo further asymmetric divisions, and their sister cells, called stomatal lineage ground cells (SLGCs). SLGCs do not assume a stomatal precursor fate; they frequently transdifferentiate into pavement cells, but may resume asymmetric division capability to produce another meristemoid spaced away from the first as the leaf expands. All of these fates are reflected in the expression and localization of POLAR::POLAR-GFP (Fig. 3, Video 1).
Figure 1. Schematic of stomatal development. Protodermal cells enter the stomatal lineage as meristemoid mother cells (MMC) in a step controlled by the basic helix-loop-helix transcription factors SPEECHLESS (SPCH), SCREAM (SCRM), and SCRM2. MMCs divide asymmetrically to produce a meristemoid (M) and stomatal lineage ground cell (SLGC); this step may occur multiple times and provides one mechanism by which stomata are spaced apart. The cell-state transition of the meristemoid to the guard mother cell (GMC) identity is directed specifically by the bHLH MUTE as well as SCRM/2. A final symmetric division of the GMC and its differentiation into two mature guard cells (GC) requires the bHLH FAMA with SCRM/2 10. Click here to view larger figure.
Figure 2. Diagram of seed dissection and mounting procedure. Seeds are sterilized and held at 4 °C, the seed coats and hypocotyls removed, and the cotyledons mounted under agar media in a chamber slide for imaging on an inverted confocal microscope.
Figure 3. Time-lapse imaging demonstrates POLAR::POLAR-GFP distal localization preceding asymmetric cell division. Series A: POLAR-GFP initially appears evenly distributed, then approximately two hours before asymmetric divisions (arrowheads) POLAR-GFP segregates away from the site of the incipient meristemoid. Series B: Following the initial asymmetric division (arrowhead), POLAR-GFP disappears in both cells, implying rapid transition to guard mother cell (GMC) state by the meristemoid. The GMC (asterisk) divides symmetrically and differentiates to form stomatal guard cells, while the sister cell regains POLAR-GFP expression, presumably presaging a later asymmetric division.
Video 1. Streamlined, registered video of POLAR::POLAR-GFP localization during amplification and spacing of one stomatal precursor cell. Initially, POLAR-GFP appears uniformly in the cell; by 39 hr after germination, it polarizes strongly, just before a division (40.0h) placing a smaller meristemoid at the opposite end of the cell. The larger cell at right also regains stomatal lineage identity, and by 45 hr POLAR-GFP localization moves adjacent to the stomatal precursor. The division at 47 hr produces a new meristemoid (right) oriented away from the existing meristemoid (left) from the first. The end result is two stomata separated by one cell. (Images missing from the video were not collected and do not represent significant changes.) Click here to view movie.
This time-lapse confocal technique allows longitudinal studies of fluorescently tagged protein expression and localization in individual cells of the Arabidopsis cotyledon epidermis, which in the case of POLAR and other dynamically changing proteins is critical to a proper understanding of their function. Previously, sustained time lapse imaging has been used to examine Arabidopsis root fungal infection11 and meristem growth5,12, but adding the cotyledon epidermis expands the versatility of this technique and allows its use for additional proteins, particularly assisting the expanding field of stomatal development.
The protocol’s main limitation is that it is currently restricted to cotyledons, which contain the nutrients they need for early development. Further, cotyledon development is not exactly identical to that undergone in air because the gel medium restricts gas exchange. Scanning laser exposure and room lights are adequate for photomorphogenesis, inducing cotyledon expansion and chloroplast maturation, but stomata may occasionally develop in adjacent pairs due to low CO2 concentration. This tendency can be reduced by using only a thin layer of media and minimal mounting water as directed in this protocol.
Fluorophore selection is also important for success with this technique. Both GFP and RFP work well, and allow double labeling to visualize the cell periphery or assess protein co-localization. Custom filters greatly reduce autofluorescence and allow sensitive detection of fluorescent protein tags even when chlorophyll is present. However, even filtered CFP autofluorescence in germinating cotyledons is strong enough to preclude visibility of almost all signal using the LSM700. For instruments with a spectral unmixing capability, a wider selection of fluorophores may be feasible.
The authors have nothing to disclose.
We thank Amanda Rychel for assistance in developing the time lapse protocol and Lynn Pillitteri for constructing POLAR::POLAR-eGFP. We are also grateful to ABRC for providing the pm-rb construct. This protocol was developed through a support from the PRESTO award from Japan Science Technology and Agency. Research on POLAR was also supported by the University of Washington Royalty Research Fund (RRF-4098) and the National Science Foundation (MCB-0855659). KMP is an NSF Graduate Research Fellow (DGE-0718124), and KUT is an HHMI-GBMF investigator.
Name of reagent | Company | Catalog Number | Comments |
Bacto Agar | BD | 214010 | |
One-chamber slide | Nunc (Thermo Scientific) | 155360 | Or two-chamber (155379) |
Laser scanning confocal microscope | Zeiss | LSM700 | Zen 2009 software |
20x objective lens | Zeiss | 420650-9901 | NA 0.8, Plan-APOCHROMAT |
Dissecting microscope | Benz (National) | 431TBL | Illuminates from below |
#5 forceps, biology tip | Roboz Surgical Instrument | RS-4978 | Very fine tips are critical |