Full text provided by www.sciencedirect.com
^jfj^ '%* ScienceDirect
ELSEVIER
Stomatal development: from neighborly to global communication
Dominique Bergmann
Stomata are epidermal structures that are responsible for modulating the exchange of gases between the plant and the environment. Stomata are formed and patterned by asymmetric cell divisions. The number and orientation of these asymmetric divisions is informed by plant intrinsic signals acting locally (among epidermal cells) or at a distance (from mature to young leaves) and by plant extrinsic factors such as the quantity of light, water and CO2 in the atmosphere. Recent studies have implicated a set of conserved cell surface receptors and intracellular signaling molecules in the perception and response to developmental cues. Complementary studies have probed the nature of environmental signals and how these signals are transduced from the site of perception to the cells in the stomatal lineage.
Addresses
Department of Biological Sciences, 371 Serra Mall, Stanford University, Stanford, California 94305, USA
Corresponding author: Bergmann, Dominique (dbergmann@stanford.edu)
Current Opinion in Plant Biology 2006, 9:478-483
This review comes from a themed issue on Cell signalling and gene recognition Edited by Joseph Kieber and Takashi Araki
Available online 4th August 2006
1369-5266/$ - see front matter
© 2006 Elsevier Ltd. All rights reserved.
DOI 10.1016/j.pbi.2006.07.001
Introduction
Stomatal development is a paradigm for studying the integration of genetic and environmental inputs in developmental decisions. Stomata are structures in the epidermis of aerial organs that function as conduits for the exchange of carbon dioxide (CO2) and water vapor between the plant and the atmosphere. A stoma consists of a pore, flanked by two epidermal cells, that lies above an airspace in the mesophyll (Figure 1a). Stomata are separated from one another by intervening epidermal cells, a pattern thought to be a functional requirement for efficient gas exchange [1]. In Arabidopsis, this pattern is referred to as the one-cell spacing rule [1]. Stomatal density (stomata mm~2) and stomatal index (number of stomata per total number of epidermal cells) vary with growth conditions, whereas the spacing pattern does not. This points to possible differences in the source of regulatory information for each parameter. For example,
stomatal density is highly regulated by the environment, but spacing might be under the sole jurisdiction of developmental genes.
Stomatal development is characterized by an orchestrated series of asymmetric cell divisions followed by a single symmetric division (Figure 1b). 'Entry divisions' initiate the lineage by creating a small cell called a meristemoid and a larger sister cell. The sister cell may become an epidermal (pavement) cell or undergo an asymmetric division later. 'Amplifying divisions' occur in meriste-moids. These divisions not only regenerate the meriste-moid but also increase the number oflarger daughter cells, and thus the total number of epidermal cells. 'Spacing divisions' resemble entry divisions in their outcome but have additional constraints because they occur in cells next to a stoma or precursor, and because their asymmetric division must be oriented so that the larger daughter cell intercalates between the stoma and the new meristemoid. The differentiation of stomatal guard cells requires that the meristemoid transits through its amplifying divisions and converts into a new cell type, the guard mother cell (GMC). The GMC divides only once, and its symmetric division creates the two guard cells of the stoma.
Three transcription factors have been identified that control the GMC to guard cell transition [2*,3*]. No genes that are required exclusively for entry divisions have yet been identified, but a set of genes that encode cell surface receptors is critical for oriented divisions and plays a role in modulating all of the asymmetric divisions in the stomatal lineage.
Receptors mediate cell-cell communication in epidermal pattern
ERECTA (ER) encodes a receptor-like protein that contains leucine-rich repeat (LRR) extracellular domains, a single transmembrane domain and a cytoplasmic kinase (a receptor-like kinase [RLK]). This protein has far-reaching effects on plant growth, developmental pattern and responses to biotic and abiotic stresses [4,5]. Many of the processes that are affected by loss of ER function are united in their requirement for cell proliferation. The expression of an ER protein without the kinase domain (ER-Akinase) results in plant growth phenotypes that are more severe than those found in er null plants, suggesting that the truncated ER protein acts as a dominant negative, probably by forming unproductive heterodimers with other LRR-RLKs [6]. Good candidates to be partners in these heterodimers are ER's closest homologues ERECTA-LIKE1 (ERL1) and ERL2. Both ERL1 and
Current Opinion in Plant Biology 2006, 9:478-483
www.sciencedirect.com
Improving allogeneic stem cell transplantation Barrett 479
Figure 1
Entry Amplifying Spacing
-»Proliferation
Current Opinion in Plant Biology
Stomatal structure and developmental pathway. (a) Scanning electron micrograph of closed Arabidopsis stoma (arrowhead) surrounded by sepal epidermal cells. (b) Diagram of divisions in the stomatal pathway. Young epidermal cells make entry divisions to create meristemoids (M). The meristemoid might continue dividing (amplifying divisions) or undergo differentiation to form a guard mother cell (GMC) and, via a symmetric division, a pair of guard cells (GCs). Spacing divisions can occur next to GCs, GMCs, or older Ms as indicated.
ERL2 can functionally substitute for ER in overall plant growth when expressed under the control of ER regulatory sequences [4]. erll and erl2 single mutants and erl1;erl2 double mutants are indistinguishable from wild-type at the whole-plant level, but erl1;erl2;er triple mutants are dwarfed and sterile [4,7**].
A closer examination of the er,erl1;erl2 mutants revealed a striking stomatal overproliferation and spacing phenotype in all organs that normally make stomata [7**]. The phenotypes of the er erll erl2 mutant suggest that the three LRR-RLKs are redundantly required for stomatal development. However, careful studies of the pheno-types of all single and double combinations of mutated ER family genes revealed subtle differences in each gene's function during asymmetric stomatal lineage divisions. Only the triple mutant exhibits defects in patterning, suggesting that activity of any of the ER family members is sufficient to orient the spacing divisions [7**].
The excitement surrounding the ER findings derives not only from the dramatic effect of these genes on stomatal development, but also because previous studies with the TOO MANY MOUTHS (TMM) gene hinted at the involvement of an LRR-RLK in stomatal development. TMM encodes an LRR receptor-like protein (RLP) that lacks an intracellular kinase domain (LRR-RLP), suggesting that TMM participates in signal transduction through interactions with an LRR-RLK partner [8].
In contrast to er phenotypes, the phenotypes of tmm mutants are exclusively related to stomatal development.
The primary mechanism for the suite of tmml phenotypes appears to be a failure of neighbor cells to respond to positional cues: tmm mutants fail to orient spacing divisions and fail to inhibit entry divisions in cells adjacent to multiple stomata or stomatal precursors. tmm mutant cells also undergo fewer amplification divisions and produce a higher percentage of stomata relative to pavement cells [9].
It is clear from the phenotypes of mutations in TMMand ER-family members that all four of the proteins encoded by these genes play roles in stomatal development. It is not clear whether these proteins are part of a single signaling complex or whether some can signal independently. The phenotypes of loss-of-function mutations in tmm and the er-family are similar but not identical. ER-family proteins primarily modulate entry divisions and amplifying divisions whereas TMM's preeminent role is in spacing divisions.
Combinations of tmml mutants with single and multiple mutants in the ER-family result in complex stomatal development phenotypes. Many of the published genetic interactions between tmm and the er-family mutants are consistent with the affected proteins being part of one physical complex, but the phenotypes in several organs can also be explained by TMM acting independently of the ER family ([7**,8]; Table 1). One of the issues confounding the interpretation of the function of these genes is that different methods and different stages of tissue development were examined by the groups studying mutants in which TMM and the ER-family were affected.
www.sciencedirect.com
Current Opinion in Plant Biology 2006, 9:478^83
480 Transplantation
Table 1
Summary of phenotypes in stomatal signaling mutants					
tmm er	erl1	erl2	er erll erl2	yoda	Other
Cotyledon          Increased SI,     Decreased SI	Wildtype	Wildtype	Greatly increased	Greatly increased	
^^^^^^^^ clusters			SI, large clusters	SI, large clusters	
Stem                No stomata Wildtype	Slight increase	Slight increase	Small clusters	Small clusters	tmm epistatic to
	in SI	in SI			single er family
/					mutants only
u Silique              Decreased SD,   Some arrested	Slight increase	Wildtype	Greatly increased	ND	er tmm double
^ clusters	in SI		SI, large clusters		mutants show no
					differentiation of
					stomata,
\					neomorphism
\ Pedicle             Gradient           Decreased SI	Slight increase	Wildtype	Greatly increased	ND	
in SI
SI
TMM, YODA and the ER family have different roles in different tissues. Examples of some of these roles are summarized in the table. Each row represents a tissue, each column a mutant phenotype. ND, no data; SD, stomatal density; SI, stomatal index.
If ER/ERL1/ERL2 and TMM do function together, what is the physical nature of their interactions? Animal receptor tyrosine kinases dimerize, and ligand binding induces phosphorylation of the receptors and the activation of downstream signaling pathways. Genetic and protein interaction studies suggest that plant LRR-RLKs probably share some of these activities [10]. TMM and the ER-family proteins contain paired cysteine residues in their extracellular domains that might be responsible for dimerization [10]. Mechanistic models that are consistent with the genetic interaction data and the coincident expression patterns of TMM, ERL1 and ERL2 in stomatal lineage cells would suggest that TMM and ER-family members interact with each other in both competitive and cooperative modes [7**]. Many interaction models can be tested by careful genetic-dosage studies and by looking at in planta physical interactions among the proteins. Discovering physical connections among these four proteins will have a significant impact not only in the field of stomatal development but in the field of plant receptor signaling. Previous experiments with the LRR-RLK-LRR-RLP pair encoded by the CLA-VATA1 and CLAVATA2 proteins suggested that hetero-dimerization activated or stabilized the complex [10]. If TMM inhibits members of the ER-family by forming heterodimers, this will demonstrate a novel mode of
plant receptor interaction, and also provides a mechanism by which specificity can be granted to broadly used
LRR-RLKs.
Regardless of whether the ER family and TMM form a single signaling complex or several, loss of these proteins results in the disruption of multiple downstream cellular behaviors, including transcription, cytoskeletal organization and targeted secretion. Intracellular signaling is required to transduce signals from the cell periphery to these numerous targets. A mitogen-activated protein (MAP) kinase signaling pathway was implicated in stomatal development by the discovery of mutations in the MAP kinase kinase kinase (MAPKKK) gene YODA that profoundly alter stomatal density and spacing [3*]. The phenotype of the yoda mutant is similar to that of the er,erl1;erl2 triple mutant; it includes a severe reduction in overall plant height and internode length and excess production of guard cells [3*,7**]. Studies of developing cotyledons suggest that the asymmetric divisions that are compromised in tmm and er.,erl1;erl2 arealsoaffectedin yoda. Cells in the epidermis of yoda cotyledons exhibit excessive entry divisions, fail to orient spacing divisions and fail to prevent the division of neighbor cells that contact two cells of the stomatal lineage.
Current Opinion in Plant Biology 2006, 9:478-483
www.sciencedirect.com
Improving allogeneic stem cell transplantation Barrett 481
Constitutive activation of YODA eliminates stomata [3*]. This effect is dosage dependant and heterozygous CA-YODA/+ plants do not have dramatically affected stomata. However, CA-YODA can dominantly suppress the stomatal proliferation and patterning defects of tmm1 [3*]. Suppression of tmm-1 by CA-YODA is consistent with YODA's acting downstream in a common signaling pathway, but does not rule out the possibility that YODA participates in an independent and parallel signaling pathway in young epidermal cells. A physical or biochemical connection between the YODA MAPKK kinase and the stomatal regulators at the membrane will be needed to rule out this second possibility. Obtaining definitive in vivo interaction data might be challenging. Even in the intensively studied plant defense response LRR-RLK/ MAPK pathways, it is not clear whether the LRR-RLK directly phosphorylates and activates the MAPKKK or whether this activation requires an intermediary protein or protein complex [10].
Environmental influences: direct and mediated through signals from old to new
The experimental focus on the 'developmental' genes that act within the epidermis to control cell identity and division behavior will identify many of the signaling pathways and cell autonomous factors that are required for stomatal development and pattern. However, signals from underlying tissues and from the environment clearly influence the stomatal density and index. Long-range
signals could act by modulating the activity of genes such as TMM, YODA and members of the ER family or they could impinge directly on the cell-cycle machinery and other downstream targets (Figure 2).
Ecophysiologists have long noticed a correlation between stomatal density and environmental characteristics, and have used this correlation to estimate past global CO2 concentrations [11]. These studies tracked changes over hundreds of years. Because plants continuously generate organs, they might also be capable of responding to environmental change within one lifetime. Dicotyledonous plants such as Arabidopsis generate new leaves in a tightly packed arrangement around the meristem. Entry of protodermal cells into stomatal lineages precedes the emergence of the leaf into the open atmosphere. Therefore, leaves that are capable of altering their epidermal development are not capable of accurately measuring the surrounding environment. A cuvette system designed to manipulate the environment around single leaves of a plant revealed that it was the CO2 concentration around mature leaves that determined the stomatal density of emerging leaves [12].
If mature leaves perceive the environment, there must be a long-range signal that transmits this information to young leaves. Recently, several labs have used modified cuvette systems to identify or test candidates for the mobile signal and to describe the cellular basis of altered
Figure 2
Integration of environmental and developmental signals. Cuvette studies indicate where the environment must be perceived and developmental studies indicate which cells are capable of responding. The site of action of developmental genes is noted in the stomatal lineage to suggest where crosstalk between signaling pathways might occur.
www.sciencedirect.com
Current Opinion in Plant Biology 2006, 9:478^83
482 Transplantation
stomatal density in response to environmental change. Studies of maize [13*], Arabidopsis [14*], and poplar [15*] suggest that environmental perception by old leaves and response by new leaves are universal, but the details of the response can vary among species. In general, elevated CO2 levels near mature leaves increased the size of new leaves but decreased the stomatal density of these leaves. Shading mature leaves decreased new leaf size but also decreased stomatal density. Shading and high CO2 concentrations produce the same stomatal density effect in new leaves despite having opposite effects on carbohydrate production, effectively ruling out most models that invoke photosynthetic rates or sugar production as a signal that controls stomatal density. In poplar, the parameter that directly correlates with stomatal density response of new leaves is the rate at which water and CO2 pass through stomata (i.e. stomatal conductance) [15*]of mature leaves. Stomatal conductance is itself a measure of multiple inputs, including light, humidity and CO2 concentration. It is not obvious how this information would travel long distance, but it does make sense to integrate multiple inputs in the mature leaf to transmit a summed picture of the atmosphere to the new leaf.
The cell size and morphology of both guard and pavement cells can be affected by environmental treatments, suggesting several cellular mechanisms by which stomatal density decreases. Possible scenarios include altered expansion of pavement cells, altered number of entry divisions or rounds of asymmetric proliferative divisions in the stomatal lineage, and arrest (or transdifferentiation) of meristemoids or GMCs. The present studies focus on the final stomatal density, stomatal index and arrangement of stomata in responding leaves. Time-course studies would allow hypotheses about which developmental steps or genes might be targets of environmental regulation. Some evidence already points to the expansion of pavement cells and the division of stomatal lineage cells being under independent control [13*,15*].
Crosstalk in the system: roles of ERECTA in environmental signaling
The stomatal development and ecophysiology fields were recently united by the discovery of ER as a major quantitative trait locus (QTL) underlying differences in transpiration efficiency among Arabidopsis ecotypes [16**]. Transpiration efficiency is the ratio of carbon fixation though photosynthesis to water loss through transpiration — parameters that are tightly linked to the behavior of stomata. Given the role of ER in stomatal development and the implication that ER might have different partners for different developmental processes, what is the mechanism by which ER controls both development and transpiration efficiency? It is possible that the ER kinase acts with one set of partners for its stomatal development roles and a unique set of partners in a signaling pathway that coordinates transpiration and
photosynthesis. It is also possible that changes in the transpiration efficiency of er mutants are an indirect consequence of ER's developmental roles (i.e. altered plant morphology, leaf thickness and stomatal density). Genetic or protein interaction screens to find ER's partners or to find downstream targets that are unique to each process might distinguish the direct and indirect models of ER action.
Conclusions
Signaling is required at several levels to ensure the correct specification, pattern and function of stomata. Proteins that are elements of canonical signal-response pathways, ER, TMM and YODA, are essential for many aspects of stomatal pattern, but many questions about their activities and interactions remain. Do ER, TMM and YODA work in a concerted signaling pathway? If so, to what do they respond and what are the ultimate downstream targets of their activity? Do any of these proteins serve as nodes for the integration of developmental and environmental signals?
Whole-genome and proteome approaches might hold the greatest promise for filling gaps in the pathway of stomatal development. For example, many more genes that are expressed in and required for the stomatal lineage might be identified through transcriptional profiling of cells that are isolated at each stage of stomatal development [17], by profiles of mutants that affect stomatal development [3*] or through a combination of both approaches. Many resources are available that can be used to identify the MAP kinase components downstream of YODA [18,19], and a recently developed protein array has already proven to be useful in a high-throughput screen for targets of plant MAP kinases [20*].
Acknowledgements
I wish to thank members of my laboratory for discussions and comments on the manuscript, and Fred Sack and Keiko Torii for our ongoing conversations about stomatal development. Research in the lab is supported by the US National Science Foundation (NSF-IOB-0544895), a Terman fellowship and Stanford University funds.
References and recommended reading
Papers of particular interest, published within the annual period of review, have been highlighted as:
* of special interest
** of outstanding interest
1. Nadeau JA, Sack FD: Stomatal development in Arabidopsis In
The Arabidopsis Book. Edited by Somerville CR, Meyerowitz EM. Rockville, MD: American Society of Plant Biologists; 2002:1-28.
2. Lai LB, Nadeau JA, Lucas J, Lee EK, Nakagawa T, Zhao L,
*     Geisler M, Sack FD: The Arabidopsis R2R3 MYB proteins FOUR LIPS and MYB88 restrict divisions late in the stomatal cell lineage. Plant Cell 2005, 17:2754-2767. After cells are committed to enter the stomatal lineage through the action of signaling genes, a cell differentiation process that requires a change in division potential ensues. Two MYB-type transcription factors, MYB88 and FOUR LIPS (FLP), are required to stop excessive divisions of the guard mother cell and to ensure the creation of a normal pair of guard cells.
Current Opinion in Plant Biology 2006, 9:478-483
www.sciencedirect.com
Improving allogeneic stem cell transplantation Barrett 483
3. Bergmann DC, Lukowitz W, Somerville CR: Stomatal
• development and pattern controlled by a MAPKK kinase.
Science 2004, 304:1494-1497. Characterization of loss-of-function and constitutively active forms of the MAPKK kinase YODA implicates MAP kinase signaling as part of a switch between guard cell and pavement cell development. The authors also describe a transcriptional profiling approach to identify new stomatal regulators by comparing gene expression between plants that lack stomata and those with excess stomata.
4. Shpak ED, Berthiaume CT, Hill EJ, Torii KU: Synergistic interaction of three ERECTA-family receptor-like kinases controls Arabidopsis organ growth and flower development by promoting cell proliferation. Development 2004, 131:1491-1501.
5. Lease KA, Lau NY, Schuster RA, Torii KU, Walker JC: Receptor serine/threonine protein kinases in signalling: analysis of the ERECTA receptor-like kinase of Arabidopsis thaliana.
NewPhytol 2001, 1:133-143.
6. Shpak ED, Lakeman MB, Torii KU: Dominant-negative receptor uncovers redundancy in the Arabidopsis ERECTA leucine-rich repeat receptor-like kinase signaling pathway that regulates organ shape. Plant Cell 2003, 15:1095-1110.
7. Shpak ED, McAbee JM, Pillitteri LJ, Torii KU: Stomatal patterning ••   and differentiation by synergistic interactions of receptor
kinases. Science 2005, 309:290-293. Shpak et al. perform detailed genetic and phenotypic characterization of single to quadruple mutant combinations of tMm and the ER-family genes. This paper is important in pointing out that closely related genes might have both redundant and unique functions, and it suggests several novel (and testable) mechanisms by which LRR-RLKs could interact.
8. Nadeau JA, Sack FD: Control of stomatal distribution on the Arabidopsis leaf surface. Science 2002, 296:1697-1700.
9. Geisler M, Nadeau J, Sack FD: Oriented asymmetric divisions that generate the stomatal spacing pattern in Arabidopsis are disrupted by the too many mouths mutation. Plant Cell 2000,
12:2075-2086.
10. Torii KU: Leucine-rich repeat receptor kinases in plants: structure, function, and signal transduction pathways.
Int Rev Cytol 2004, 234:1-46.
11. Hetherington AM, Woodward FI: The role of stomata in sensing and driving environmental change. Nature 2003, 424:901-908.
12. Lake JA, Woodward FI, Quick WP: Long-distance CO2 signalling in plants. J Exp Bot 2002, 53:183-193.
13. Driscoll SP, Prins A, Olmos E, Kunert KJ, Foyer CH: Specification
• of adaxial and abaxial stomata, epidermal structure and photosynthesis to CO2 enrichment in maize leaves. J Exp Bot
2006, 57:381-390.
Driscoll et al.[13^], Coupe et al.[14^] and Miyazawa et al.[15^] all use variations on the cuvette system of Lake et al. [12] to examine the mature to new leaf signals in response to environmental change. Each group found that mature leaves can sense environmental change and produce a response in new leaves, but that the amplitude and timing of that response varies among species.
14. Coupe SA, Palmer BG, Lake JA, Overy SA, Oxborough K,
• Woodward FI, Gray JE, Quick WP: Systemic signalling of environmental cues in Arabidopsis leaves. J Exp Bot 2006,
57:329-341.
See annotation for [13^].
15. Miyazawa S, Livingston NJ, Turpin DH: Stomatal development in
• new leaves is related to the stomatal conductance of mature leaves in poplar (Populus trichocarpax P. deltoides). J Exp Bot
2006, 57:373-380.
See annotation for [13^].
16. Masle J, GilmoreSR, FarquharGD: The ERECTA gene regulates plant transpiration efficiency in Arabidopsis. Nature 2005,
43613:866-870.
Using 813C discrimination as a method for estimating transpiration efficiency (TE), Masle et al. identified QTLs for this trait in Arabidopsis accessions and mapped a major QTL to the ER region. They showed that ER itself is responsible for altered TE. Some of the developmental defects, including changes in stomatal density, that are associated with mutation of ER and that could lead to altered TE were explored. However, no single morphological character is solely responsible for the effect.
17. Birnbaum K, Jung JW, Wang JY, Lambert GM, Hirst JA, Galbraith DW, Benfey PN: Cell type-specific expression profiling in plants via cell sorting of protoplastsfrom fluorescent reporter lines. Nat Methods 2005, 2:615-619.
18. Asai T, Tena G, Plotnikova J, Willmann MR, Chiu WL, Gomez-Gomez L, Boller T, Ausubel FM, Sheen J: MAP kinase signalling cascade in Arabidopsis innate immunity. Nature 2002, 415:977-983.
19. Liu Y, Zhang S: Phosphorylation of 1-aminocyclopropane-1-carboxylic acid synthase by MPK6, a stress-responsive mitogen-activated protein kinase, induces ethylene biosynthesis in Arabidopsis. Plant Cell 2004, 16:3386-3399.
20. Feilner T, Hultschig C, Lee J, Meyer S, Immink RGH, Koenig A,
• Possling A, Seitz H, Beveridge A, Scheel D etal.: High throughput identification of potential Arabidopsis mitogen-activated protein kinases substrates. Mol Cell Proteomics 2005, 4:1558-1568.
This paper describes the creation of a protein microarray that contains about 1400 Arabidopsis proteins. This microarray was used to identify kinase substrates of MPK3 and MPK6. The validation of targets for these kinases though conventional kinase assays suggests that this high-throughput approach will have great utility in identifying possible targets of many plant signaling pathways.
www.sciencedirect.com
Current Opinion in Plant Biology 2006, 9:478^83