Plant Development: YODA theStomatal Switch

Dispatch

Julie E. Gray1 and Alistair M. Hetherington2

The appearance of stomatal pores during plantevolution is believed to have been a crucial step inland colonisation. A recent screen for genes involvedin stomatal development has identified for the firsttime a mutant plant with no stomata; the results impli-cate a MAP kinase cascade in stomatal development.

Stomata are small pores on the surfaces of leaves thatcontrol the exchange of gasses between the interior ofthe plant and the atmosphere [1]. They make a signifi-cant contribution to global water and carbon cycles, asan estimated 440 x 1015 g CO2 and 32 x 1015 kg ofwater vapour pass through stomatal pores annually [2].Stomatal pore opening and closure are regulated byplant hormones and environmental signals. The pres-ence of these control systems allows the plant to con-serve water during periods of drought and to maximiseCO2 uptake during the day. It has been convincinglyargued that the acquisition of stomata is a key elementin the evolution of advanced terrestrial plants [3], asthey allow the plant to inhabit different environmentswhile maintaining control over their water content.Stomatal development and patterning are also influ-enced by environmental factors, and the identity ofsome of the components involved in controlling theseprocesses is beginning to emerge [4–6].

In this context, recent work from Bergmann, Lukowitzand Somerville [7] is of great interest, as these authorshave shown that a newly characterised mitogen acti-vated protein (MAP) kinase kinase kinase known asYODA is an important negative regulator of stomataldevelopment. Of equal significance, however, is theirobservation that plants engineered to express a consti-tutively active YODA have no stomata and die beforethey reach reproductive maturity. Although further workis required before it is possible to reach definitive con-clusions, this result is potentially very importantbecause it suggests that the possession of stomata isvital to plant survival and as such lends support toRaven’s suggestion [3] that the acquisition of thesestructures was a key element in the evolution ofadvanced terrestrial plants.

In the Arabidopsis epidermis, stomatal developmentstarts with the initiation of a meristemoid mother cell.When this cell divides asymmetrically the smallerproduct is known as the meristemoid and this, afterseveral asymmetric divisions, is eventually convertedinto a guard mother cell, which divides symmetrically toform the pair of guard cells that gate the stomatal pore[4–6]. At least three rules govern stomatal development

and patterning. First, their development follows theseries of asymmetric divisions described above;second, a patterning rule ensures that stomata are sep-arated from each other by at least one non-stomatalepidermal cell; and third, the number of stomata thatform, relative to the non-stomatal epidermal cells, issubject to control by environmental factors. The isola-tion of Arabidopsis mutants that carry lesions in stom-atal development and patterning has providedsignificant insights into the signal transduction path-ways that underlie these processes [4,5]. Arabidopsismutant screening also formed the basis of the workwhich allowed Bergmann et al. [7] to uncover the role ofa putative MAP kinase kinase kinase encoded by theYODA gene (YDA) in stomatal development.

Bergmann et al. [7] carried out a visual screen foraberrant stomatal patterning and out of the 19 mutantsrecovered, six were alleles of YDA [8]. This gene hadrecently been identified as a site of mutations that affectearly embryo development. After fertilization, the Ara-bidopsis zygote undergoes an asymmetric division toform a small apical cell that goes on to form the embryoand a larger lower cell that develops into an extraembryonic structure known as the suspensor. In ydamutants, the suspensor does not form and the cells ofthe lower cell lineage are incorporated into the embryo.

On the basis of these observations Lukowitz et al. [8]concluded that the putative MAP kinase kinase kinaseencoded by the YDA gene acts as a switch that pro-motes extra-embryonic fate. From what we know ofMAP kinase cascades, it seems reasonable to concludethat a MAP kinase signalling cassette is involved in thisexample of the control of post asymmetric division cellfate. The six yda alleles identified by Bergmann et al. [7]also cause defects in post-embryonic development;additionally, they result in plants developing withextreme overproduction and clustering of stomata inthe epidermis of the cotyledons and hypocotyls. Mostyda seedlings fail to survive on transfer to soil, butthose that do were found to be dwarfed and character-ized by compact leaves filled with clusters of stomata.

When the yda phenotype was investigated in detail,it was found that, compared to wild type, more yda epi-dermal cells had entered the stomatal fate pathway.This indicates that just as in embryos, YDA plays a rolein cell fate choice in the leaf epidermis. To investigatethe role of YDA further, Bergmann et al. [7] made trans-genic plants expressing a gene (∆∆N-YDA) encoding aform of the YDA MAP kinase kinase kinase designed tobe constitutively active. Seedlings homozygous for ∆∆N-YDA totally lacked stomata, supporting the idea thatYDA acts as a switch that negatively regulates stomatalfate. The authors also used genetic means to ‘place’YDA in the stomatal development pathway relative toother gene products that affect stomatal development.The TOO MANY MOUTH (TMM) and STOMATALDENSITY AND DISTRIBUTION (SDD1) loci encode aleucine-rich repeat receptor-like protein and a putative

Current Biology, Vol. 14, R488–R490, June 22, 2004, ©2004 Elsevier Ltd. All rights reserved. DOI 10.1016/j.cub.2004.06.019

1Department of Molecular Biology and Biotechnology,University of Sheffield, Sheffield S10 2TN, UK. 2Department ofBiology, LEC, Lancaster University, Lancaster LA1 4YQ, UK.

subtilisin-like serine protease, respectively [9,10]. Muta-tions in these genes result in plants that exhibit anincrease in stomatal number, and current models toaccount for these phenotypes are based on the ideathat the SDD1 protein generates a peptide signal that isrecognized by TMM receptor [5]. As indicated in Figure1, genetic analysis places YDA downstream of boththese gene products.

Finally, Bergmann et al. [7] used whole-genome tran-script analysis to identify further genes involved in thecontrol of stomatal patterning and development. Theyreasoned that transcripts up-regulated in yda plants(mostly stomata) and down-regulated in ∆∆N-YDAplants (no stomata), compared to wild-type, would begood candidate regulators of stomatal patterning anddevelopment. Indeed, expression of three genesknown to be involved in stomatal development, TMM,SDD1 and HIC (involved in the control of stomataldevelopment by atmospheric CO2 concentration [11]),followed this expression pattern [7]. Analysis of genesthat clustered with these three known stomatal genesrevealed 109 that are strong candidates for involve-ment in the control of stomatal patterning and devel-opment. This was confirmed by selecting one of thesegenes known as FAMA, which encodes a putative tran-scription factor; when plants homozygous for a T-DNAinsertion in FAMA were analysed it was found that theytoo lack stomata [7].

The impact of the work from Bergmann et al. [7] islikely to be profound. They have identified over 100candidate genes involved in guard cell developmentand patterning which are set to be a rich source foruncovering the intricacies of these developmental path-ways for a considerable time. In addition, they haveprovided a tantalizing glimpse into the possible signifi-cance of the acquisition of stomata for the evolution ofadvanced terrestrial plants [9]. Of course, because YDAis implicated in multiple developmental responses, it isnot possible to conclude that ∆∆N-YDA plants fail toreach reproductive maturity because of their lack ofstomata. But the fact that fama plants are small andsterile certainly adds weight to this possibility.

The issue should be resolved in the near future asBergmann et al. [7] are currently constructing plants inwhich ∆∆N-YDA is expressed only in the epidermis (D. Bergmann, personal communication). Phenotypicanalysis of these plants might result in the productionof key data supporting Raven’s suggestion [3] that theacquisition of stomata is key to the success of theadvanced terrestrial plants. Even though a tobaccoMAP kinase kinase kinase had previously been impli-cated in the control of guard cell development [12], theidentification of YDA MAP kinase kinase kinase and thefate switch it controls is a major advance. Indeed, it isan advance that prompts many new questions. Weneed to know whether YDA functions in the same (asin Figure 1) or a separate pathway to TMM and SDD. Ifthey are in the same pathway how does TMM feed intothe MAP kinase signalling cassette? Might FAMA be aphosphorylation target for the putative MAP kinase sig-nalling cassette? And how do components involved inthe environmental control of stomatal development, suchas HIC, interact with the YDA stomatal development

pathway? Finally, we urgently need to identify the othercomponents, such as the receptor-like kinase that oneassumes interacts with TMM to inactivate YDA andactivate stomatal development, and the MAP kinasekinase and MAP kinase scaffold proteins that will pre-sumably be responsible for controlling the specificityof YDA interactions in the cells of the zygote and stom-atal precursors [13].

References1. Hetherington, A.M. and Woodward, F.I. (2003). The role of stomata

in sensing and driving environmental change. Nature 424, 901-908.2. Ciais, P., Denning, A.S., Tans, P.P., Berry, J.A., Randall, D.A. Collatz,

C.J., Sellers, P.J., White, J.W.C., Trolier, M., Meijer, H.A.J., et al.(1997). A three-dimensional synthesis study of dO18 in atmosphericCO2. I. Surface fluxes. J. Geophy. Res. — Atmosphere 102, 5857-5872.

3. Raven, J. (2002). Selection pressures on stomatal evolution. NewPhytol. 153, 371-386.

4. Bergmann D.C. (2004). Integrating signals in stomatal development.Curr. Opin. Plant Biol. 7, 26-32.

5. Nadeau, J.A. and Sack, F.D. (2003). Stomatal development: crosstalk puts mouths in place. Trends Plant Sci. 8, 294-299.

6. Bird, S.M. and Gray, J.E. (2003). Signals from the cuticle affect epi-dermal cell differentiation. New Phytol. 157, 9-23.

7. Bergmann, D.C., Lukowitz, W. and Somerville, C.R. (2004). Stomataldevelopment and pattern controlled by a MAPKK kinase. Science304, 1494-1497.

8. Lukowitz W., Roeder, A., Parmenter, D. and Somerville C. (2004). AMAPKK kinase gene regulates extra-embryonic cell fate in Ara-bidopsis. Cell 116, 109-119.

Current BiologyR489

Figure 1. Hypothetical scheme linking components known tobe involved in the control of stomatal development.

Genetic analysis places the MAPKKK YDA downstream ofTMM, a leucine-rich repeat receptor-like protein, and SDD1, asubtilisin-like protease. Coloured components represent geneproducts known to affect stomatal development. Black com-ponents are predicted to be present on the basis of knowledgegained from other better characterised signalling pathways.Although this model for signalling during stomatal developmentassumes that all components shown act in the same pathway,it is quite possible, on the basis of our current knowledge, thatsome components act in separate signalling pathways.

P

P

P

YDA

Receptor-likeprotein kinase

MAPK

MAPKK

Mature peptidesignal

SDD1

Stomatal development

TMM

–ve

P

Current Biology

DispatchR490

9. Nadeau, J.A. and Sack, F.D. (2000). Control of stomatal distributionon the Arabidopsis leaf surface. Science 296, 1697-1700.

10. Berger, D. and Altmann, T. (2000). A subtilisin-like serine proteaseinvolved in the regulation of stomatal density and distribution inArabidopsis thaliana. Genes Dev. 14, 1119-1131.

11. Gray, J.E., Holroyd, G.H., van der Lee, F., Bahrami, A.R., Sijmons,P.C., Woodward, F.I., Schuch, W., and Hetherington A.M. (2000).The HIC signalling pathway links CO2 perception to stomatal devel-opment. Nature 408, 713-716.

12. Nishihama, R., Ishikawa, M., Araki, S., Soyano, T., Asada, T. andMachida, Y. (2001). The NPK1 mitogen-activated protein kinasekinase kinase is a regulator of cell-plate formation in plant cytoki-nesis. Genes Dev. 15, 352-363.

13. Ptashne M. and Gann, A. (2003). Imposing specificity on kinases.Science 299, 1025-1027.