Tissue patterning ofArabidopsiscotyledons - Wiley Online Library

NPH_342.fm Page 461 Thursday, January 31, 2002 8:22 PM

Research

Tissue patterning of Arabidopsis cotyledons Blackwell Science Ltd

Gregory J. Bean1, M. David Marks2, Martin Hülskamp3, Murray Clayton4 and Judith L. Croxdale1 1

Department of Botany, University of Wisconsin, Madison, WI 53706, USA; 2Department of Plant Biology, University of Minnesota, St. Paul, MN 55108,

USA; 3Botanik III, Universität Köln, Köln, DE D50931, Germany; 4Department of Statistics, University of Wisconsin, Madison, WI 53706, USA

Summary Author for correspondence: Judith L. Croxdale Tel: +1 608 262 2743 Fax: +1 608 262 7509 Email: [email protected] Received: 12 September 2001 Accepted: 22 November 2001

• Trichome and stomatal patterning are not independent events because trichomes form before stomata. We thought trichome genes might provide spatial referents to ensure proper distribution of stomata for gas exchange, and therefore studied mutants of GL1 and TRY using stomatal pattern of the entire cotyledon surface as the indicator. • Mature cotyledons were imaged by SEM, stomatal maps were generated, and data were spatially analysed. Expression of GL1 and TRY was determined in wild type and mutant samples by reverse transcriptase-PCR (RT-PCR) analysis. • At the tissue level, findings showed wild type cotyledons had a random stomatal pattern, whereas gl1–1 and try240 cotyledons had ordered and clustered stomatal patterns, respectively. Regardless of overall pattern type, c. 10% of the stomatal population – those closest to one another – were always ordered, the result of genes regulating cellular differentiation. • These results indicate epidermal cells respond to GL1 and TRY signals that affect distribution of both stomata and trichomes in postembryogenic events. The GL1 and TRY genes play dual roles in the epidermis, one role regulating epidermal tissue patterning and a second role connected with trichome development. Key words: Arabidopsis, stomata, trichomes, epidermal patterning. © New Phytologist (2002) 153: 461–467

Introduction Patterning of stomata and trichomes in the epidermis are not independent events, but represent the interaction of spatial organization in a single field of cells. Within the field, cells commit to one of two differentiation pathways on discrete temporal and spatial bases. How this organization occurs and what molecular agents are involved is not known, although there are hints from published work. Leaves are ideal for the study of tissue pattern because they present two-dimensional patterns, their surface is accessible, and two specialized cells – trichomes and stomata – can be used to monitor spatial organization of the epidermis. Studies of tissue spatial organization require examination of the entire cellular field and of entire populations of stomata and trichomes, with each serving as an indicator of tissue spatial organization. This approach differs significantly from the molecular genetic studies of trichomes and stomata, which identify defects in cellular patterning (Oppenheimer et al., 1991; Larkin et al., 1994; Larkin et al., 1996; Larkin et al.,

© New Phytologist (2002) 153: 461– 467 www.newphytologist.com

1997; Marks, 1997; Szymanski et al., 1998a,b; Perazza et al., 1999; Berger & Altmann, 2000; Szymanski et al., 2000). Some defects are strictly concerned with differentiation (Glover, 2000; Martin & Glover, 1998; Walker et al., 1999), but others, such as clustered trichomes and stomata, reveal disruptions in cell fate decisions (Hülskamp et al., 1994; Yang & Sack, 1995; Schnittger et al., 1999; Geisler et al., 2000). Misplaced fate decisions disrupt local pattern but do not reflect tissue pattern. Since trichomes arise earlier than stomata, which need to be appropriately placed for effective gas exchange, we felt genes that influence the siting of trichomes might also influence the siting of stomata. Here we examined that possibility using genes with known effects in the Arabidopsis trichome pathway. Precedent for this possibility is known from studies of roots (Scheres, 2000), hypocotyls (Berger et al., 1998), and the leaf epidermis (Benfey, 1999). Stomatal development in the hypocotyl is regulated by genes active in the root epidermis (Masucci et al., 1996; Berger et al., 1998; Hung et al., 1998). Given that hypocotyl stomata and trichoblasts in roots have the same positional relationship with respect to underlying

461


462 Research

cells, a common mechanism exists for epidermal patterning (Masucci et al., 1996; Berger et al., 1998; Hung et al., 1998). Several genes (TTG1, GL2, CPC ) are implicated in regulating the position and fate of particular cell types in both root and shoot epidermal patterning, although regulation is negative in roots and positive in leaves (Benfey, 1999). In the leaf epidermis, a similar situation exists with GL1, GL2, TTG1, and TRY, their mutants, and overexpression lines, which have complex interactions in promoting and inhibiting trichome cell fate (Scheres, 2000). These findings point to common genetic mechanisms that regulate cellular patterning (Dolan & Scheres, 1998). Indeed, parsimony dictates that patterning of all epidermal cells would use the same or similar spatial referents. The evidence cited above lends substantial support to such a model. Since trichome and stomatal patterning are not independent events, we studied cotyledons, leaves that bear no trichomes, to simplify interpretation of pattern data. We compared the overall distribution of stomata on cotyledons of gl1–1 (a glabrous mutant) and try240 (a mutant with clustered trichomes) to those of wild type (WT) Arabidopsis. Cotyledons of all three genotypes have similar, uniform epidermal fields, where all cells are available for commitment as a guard mother cell because they cannot develop into a trichome. Our results show qualitative differences in stomatal patterns. Wild type stomatal pattern on cotyledons is random, whereas pattern is ordered in mutant lines homozygous for gl1–1 and clustered in mutant lines homozygous for try240. Therefore, epidermal cells that become stomata read the same underlying spatial signals as leaf epidermal cells that become trichomes. GL1 and TRY play dual roles in epidermal patterning – one role in connection with patterning epidermal space and a second role in connection with trichome differentiation. These genes have a broad, regulatory effect on the spatial organization of the epidermis and yield signals that affect the distribution of both trichomes and stomata

Materials and Methods Plant materials and growth conditions Plant lines of Arabidopsis thaliana (L.) Heynh selected for this study included wild type Nossen, gl1–1, and try240. The Nossen and gl1–1 (backcrossed three times into the Nossen background) seeds were provided by Fred Hemphel and Pat Zambryski, University of California-Berkeley. The try240 seeds were in the Columbia background (backcrossed three times). The gl1–1 (Koornneef et al., 1982) and try240 (Hülskamp et al., 1994) mutations were originally found in the Landsberg erecta ecotype. All plants were grown in Ferti-Lome potting mix (VPG, Bonham TX USA). Seeds were scattered onto moistened soil mixture in 2.5-inch square pots, which were then covered with plastic film and placed into a 4°C refrigerator for 3 d. The plants were then

transferred to a growth chamber, and the plastic film was left over the pots for 2 d. The WT and gl1–1 plants were grown at 150 µmol m−2 s−1 on a 8-h light: 16 h dark cycle, while the try240 plants were grown at 130 µmol m−2 s−1 on a 16-h light: 8 h dark cycle. All plants were grown at 22°C and 75% humidity and were watered from below with tap water. Sample preparation, microscopic imaging, and stomatal index sampling Developing and mature cotyledons were collected and their abaxial surfaces were imaged by SEM. Nossen and gl1–1 plants were prepared and viewed according to Ahlstrand (1996) with the following modification: cotyledons were excised and attached to 1/4 inch SEM stubs, and the entire stub was immersed in liquid nitrogen for 2 min before being placed onto the cold stage for low temperature (−160°C)-low voltage (1.5 kV) scanning electron microscopy (LTLVSEM). Cotyledons of try240 were fixed in FAA for at least 2 h. The samples were then dehydrated in a graded ethanol series, critical point dried, mounted on SEM stubs with double-stick carbon disks, and sputter coated with gold. Samples were viewed at 10 kV on an S-570 SEM (Hitachi, Tokyo, Japan). For each line, samples were collected from plants beginning at day 2 and ending with full expansion of the cotyledon. Twenty-seven Nossen, 25 gl1–1, and 16 try240 samples were imaged; a minimum of three samples was used for analysis of mature pattern. Each sample was captured in a series of digital images, which were pieced together in Photoshop 5.5 (Adobe, San Jose, CA, USA) to reconstruct the image of the entire cotyledon (Fig. 1). The area of each cotyledon was measured with NIH Image 1.62, and maps of stomatal distribution were made using Photoshop. Stomatal maps were transformed using DataThief II into X-Y coordinate data for subsequent statistical analysis. Stomatal index data (stomata as a percentage of the epidermal cell population) were collected from each sample. Their relationship to area was analyzed by using the regression (REG) procedure of SAS to fit a regression model of the square root of index as a linear function of the square root of area. Square root transformations were used to account for nonlinearities in the original data. The general linear model (GLM) procedure of SAS was used to test for possible differences in slopes or intercepts of the lines fit to each genotype by modeling square root of index as a function of square root of area, plus a main effect for genotype and an interaction between genotype and square root of index. Statistical analysis of stomatal pattern Data were analyzed using a ‘nearest neighbour’ approach (Upton & Fingleton, 1985) with the spatial module of S-PLUS (MathSoft, Cambridge, MA, USA). Specifically, for each stoma the distance to the nearest stoma was calculated. These observed distances were used to calculate the estimated

www.newphytologist.com © New Phytologist (2002) 153: 461– 467


Research

Fig. 1 Composite scanning electron microscopy (SEM) image of the abaxial surface of a WT cotyledon. Bar, 500 µm.

distribution function of these distances, G(w). Envelopes corresponding to a pointwise significance of P = 0.05 were generated by Monte Carlo simulation. In other words, for any given value w, the value of G(w) represents the estimated probability that the distance from a given stoma to its nearest neighbour will be less than or equal to w. Moreover, for a given value of w, if G(w) lies below the simulation envelope, then that represents evidence of spatial regularity at the scale of that w; similarly, G(w) above the simulation envelope provides evidence of spatial clustering at that scale. The statistical analysis graphs are organized to display the distance in µm between stomata on the abscissa (X axis) and the entire stomatal population by percentage on the ordinate (Y axis). The more or less parallel lines that lie at a 45 degree angle indicate the region of random distribution with a 95% confidence level. Data points that lie between the lines reflect a random distribution, whereas data points that lie above the uppermost line reflect a clustered distribution and data points that lie below the lowermost line reflect an ordered distribution. RNA extraction and RT-PCR For Nossen samples, the cotyledons (c. 4-d-old), rosette leaves one and two (c. 10-d-old), or immature seeds were removed


and frozen in liquid nitrogen. For the mutant samples, entire aerial portions of 10-d-old-plants were frozen in liquid nitrogen. Tissues were stored at –80°C until RNA was extracted by the procedure in Shirzadegan et al. (1991). DNA was removed from the extract with RQ1 DNase (Promega, Madison, WI, USA), and the samples were then phenol-chloroform extracted and ethanol precipitated. RNA concentrations were determined spectrophotometrically and approximately equal amounts of RNA template were used in reverse transcription. M-MLV reverse transcriptase and oligo(dT)15 primers (Promega) were used for reverse transcription as per the manufacturer’s instructions. One microliter of this cDNA was used as the template in 25 µl PCR reactions. Ex Taq DNA polymerase (Takara, Kyoto, Japan) was used in the PCR reactions following the manufacturer’s instructions. GL1 primer sequences were obtained from Oppenheimer et al. (1991). Primers amplifying a 500-bp region of Arabidopsis APRT served as a positive control in all reactions as in Payne et al. (2000). Reactions were carried out in a PTC-200 thermal cycler (MJ Research, Waltham, MA, USA) set to run the following program: 96°C for 5 min, then 40 cycles of 96°C for 30 s, 61°C for 30 s, and 72°C for 60 s, followed by a final extension of 72°C for 7 min for the GL1 primers. When the TRY primers were used, the annealing temperature was set to 57.5°C. PCR products

463


464 Research

were separated on 1% agarose gels stained with ethidium bromide.

Results Spatial analysis of stomatal distribution on WT cotyledons indicated a random pattern (Fig. 2a), whereas similar analysis of gl1–1 and try240 cotyledons showed ordered and clustered patterns, respectively (Fig. 2b,c). Try240 was also backcrossed three times to Nossen and they also display the clustered pattern (data not shown). Although the overall pattern in each genotype can easily be assigned, note that stomata in close proximity are always ordered, regardless of genotype. These closely spaced stomata represent approximately 10% of the stomatal population. On an index basis, stomata account for 15.4% of the epidermal cells in WT (SE ± 0.8, N = 23) and 15.9% in gl1 (SE ± 0.7, N = 20), but comprise 20.7% of the epidermal cells in try (SE ± 1.2, N = 16). The relationships of these indices to cotyledon area are not significantly different across genotypes when evaluated by the REG and GLM procedures. Restoration of WT Stomatal Pattern. To ensure that the stomatal patterns were the result of the mutation, both gl1–1 and try240 plants were backcrossed to their respective WT and examined for stomatal pattern type. In both instances, the WT stomatal pattern type was restored in the first backcross (data not shown) indicating the mutations of the GL1 and TRY genes are in fact responsible for the observed changes in stomatal pattern. Determination of GL1 and TRY expression by RT-PCR. Given that mutants of GL1 and TRY qualitatively affect stomatal pattern type, we determined expression of these genes in immature WT seeds, cotyledons, and first leaf pair by RT-PCR (Fig. 3). GL1 and TRY both appeared in cotyledons and the first leaf pair, whereas only TRY was present in immature WT seeds containing developing cotyledons (Fig. 3). Gene expression in gl1–1 and try240 plants showed the alternate gene is expressed (Fig. 3). Although no GL1 is detected in

Fig. 2 (a) Dot map of stomata on a wild type (WT) cotyledon (left) with its statistical analysis (right). (b) Dot map of stomata on a gl1–1 cotyledon (left) with its statistical analysis (right). (c) Dot map of stomata on a try240 cotyledon (left) with its statistical analysis (right). The statistical graphs show the percentage of the entire population of stomata on the y-axis as a function of the proximity of nearest neighboring stomata. Based on stomatal number and area over which they occur, the region between the thin lines indicates the positions of randomly distributed stomata at the 95% confidence level. Data points which lie above this envelope indicate a clustered pattern and those which lie below the envelope indicate an ordered pattern.

Fig. 3 GL1 and TRY gene expression in wild type (WT) (Nossen) leaves, cotyledons, and immature seeds (left), and in gl1–1 and try240 mutant plants (right) as determined by RT-PCR. The labels at the top of the figure indicate which sample was used in each reaction, while the labels at the bottom of the figure indicate which primer was used. In WT, GL1 transcript is found in leaves and cotyledons, whereas TRY transcript is found in leaves, cotyledons, and immature seeds. In gl1–1 mutants, TRY transcript is present, and in try240 mutants, both GL1 and TRY transcript is present. A 500-bp segment of Arabidopsis APRT (adenine phosphoribosyl transferase), which is expressed at low levels in all tissues, serves as a positive control in each reaction.



Research

gl1–1 plant samples because the mutation encompasses the entire coding region and promoter sequences, TRY is found in try240 because the mutation is a single base pair change resulting in a misplaced stop codon, which can only be detected by protein analysis.

Discussion The ordered pattern of specialized cells on leaf surfaces is much studied. It was Bünning, however, who first presented a comprehensive approach to understanding patterning (Bünning, 1956). He interpreted the pattern of stomata in dicots as a direct result of a simple rule: developing stomata generate inhibitory fields, which prevent new stomata from arising nearby. This idea of an inhibitory field is attractive and sufficient to explain order. Unfortunately, the size of such inhibitory fields has never been established for any species, nor have any molecules been isolated and shown to prevent the formation of stomata. Nevertheless, Bünning’s idea of lateral inhibition continues to hold great sway. This view of stomatal patterning has regulation based on dynamic interactions between individual cells and thus, order is the product of a cellular property. Sachs combined the study of the distribution of stomata relative to one another with studies of the cell divisions that generate a mature stomatal complex (Sachs, 1979; Sachs et al., 1993). Like Bünning, he argued that stomata were not prepatterned, and explained order in the pattern as a result of an obligate series of divisions, a minimal clone size to generate one stoma. This view of the origins of pattern finds answers at the cellular level rather than at the tissue level. His concept that the leaf surface was a composite of mature and variously arrested stomata-generating cell lineages does connect events at the cellular level to the tissue level. Recent studies of stomata center on events of cellular patterning or differentiation (Berger & Altmann, 2000; Yang & Sack, 1995; Glover et al., 1998; Serna & Fenoll, 2000). Charlton hypothesized that ability to alter developmental fate of cells was a function of stage in the cell cycle (Charlton, 1990), and our studies of both cellular and tissue pattern using the monocot Tradescantia do support such a link to the cell cycle (Croxdale, 2000). In this view, cell differentiation is regulated by decisions at the cellular level, but tissue patterning directs where in a field of cells this cell differentiation will occur. Our studies with Arabidopsis, reported briefly in this paper and more extensively elsewhere, extend this organ-level viewpoint to dicots. Stomatal pattern on cotyledons is qualitatively different in three genotypes of Arabidopsis. Using the Nossen ecotype whose flat leaves facilitate imaging, we analyzed images of the entire leaf surface, since sampling portions as large as half a cotyledon do not accurately represent stomatal pattern. Stomata on cotyledons of WT Nossen were randomly distributed, at odds with much published literature (citations in


Croxdale, 2000) and in agreement with published assertions (Kagan et al., 1992). By contrast, stomatal pattern on try240 and gl1–1 cotyledons is clustered and ordered, respectively. These genes, TRY and GL1, are known to regulate the differentiation of trichomes. The data in this paper show that they also regulate epidermal tissue pattern. Our results cannot be the outcome of cell-cell interference between differentiating trichomes and cells intended to become stomata because our studies were carried out with cotyledons, a leaf-type with no trichomes. We also showed that the pattern of stomata in heterozygotes, WT/mutant, is the characteristic random pattern of WT homozygotes, confirming that the effect on tissue patterning conferred by these two mutant alleles is recessive, as are their effects on trichome differentiation. These two mutations, gl1–1 and try240, identify two loci essential for establishing the correct overall pattern of differentiated epidermal features. Presumably, the WT alleles of these genes participate in the formation or propagation of some spatial referent or they perceive a spatial referent. Epidermal cells perceive or respond to the spatial referent by producing the final distribution of stomata. The molecular nature of this spatial referent is unknown. Current information from other plant systems indicates an assemblage of molecules is more likely than one acting alone. Such examples include the transcription factors which specify floral organ primordia in time and space (Bisseling & Weigel, 2001) or receptor kinasemediated signaling in defined regions of the shoot apical meristem (Clark et al., 1997) or receptor–ligand interactions (Fletcher et al., 1999). It seems likely to us that the distribution of these signals is also tied to mitotic activity and to the stage in the cell cycle of epidermal cells. Indications of a link to the cell cycle come from trichome and stomatal findings. Endoreduplication is a hallmark of trichome differentiation, yielding mature cells with a DNA level as large as 64C (Melaragno et al., 1993; Marks, 1997; Walker et al., 2000). Mutants of particular genes (SIM, GL3, TRY, GL1, COT1, KAK) are implicated in altering endoreduplication levels of trichomes (Walker et al., 2000; Hülskamp et al., 1994). The formation and differentiation of Tradescantia stomata in series (linear groups or strings) correlates to cell cycle position and synchrony of sister cells in division (Croxdale, 2000). The sequence of string lengths is nonrandom, indicating a cell cycle influence on tissue pattern. The separation of leaf epidermal cells into a diploid population (stomata) and an endopolyploid population (trichomes and epidermal cells) indicates a fundamental link to the cell cycle. Interaction between cyclin genes and transcription factors like GL3 (Meijer & Murray, 2000) including plant MYB genes such as GL1 and TRY (Ito, 2000) is known and indicates distribution of these signals can be tied to mitotic activity and the cell cycle in the epidermis. The stomatal patterns in the three Arabidopsis genotypes differ from one another, but each overall pattern includes a section of ordered stomata. The ordered stomata – about 10%

465


466 Research

of the population – are always those stomata closest to one another. We believe their close spacing reflects the action of known stomatal genes, for example, TMM, FLP, and SDD, which are involved in cell fate decisions, ensuring that stomata do not directly abut one another. These genes of stomatal differentiation regulate the close spacing of stomata, but not their overall pattern. The expression of GL1 and TRY determined in leaves, cotyledons, and immature seeds with developing cotyledons indicates patterning of stomata takes place in expanding cotyledons. GL1 transcript is found only in cotyledons and leaves whereas TRY is found in all three sample types. In WT, the GL1 product expressed in expanding cotyledons is read by epidermal cells and results in a random pattern of stomata. When the GL1 product is absent from cotyledons, as in gl1–1 mutants, the epidermal cells perceive or respond to the spatial referents and form an ordered pattern of stomata. Based on these expression data, we conclude that patterning of stomata occurs when cotyledons expand during germination, not when they form in the seed. Since TRY expression occurs in all three sample types, we are unable to confirm a similar role for this gene in altering spatial referents, although stomatal pattern in try240 mutants differed from WT, so its absence is registered by epidermal cells. The epidermis of the root system is continuous with that of the shoot system, and like leaves, roots have specialized epidermal cells arranged in ordered ways. The organization of the root epidermis, linear files of cells, is simpler than the complex organization of dicot leaves, but the two systems share several developmental and molecular parallels. The pattern of stomata on the hypocotyl is altered by mutations that alter patterning of the root epidermis (Masucci et al., 1996; Berger et al., 1998; Hung et al., 1998). Additionally, patterning of trichoblasts and atrichoblast cells of the root epidermis involves some genes with known activity in the trichome pathway (GL2, TTG1, CPC; Wada et al., 1997; Dolan & Scheres, 1998; Schneider et al., 1998; Lee & Schiefelbein, 1999; Scheres, 2000) as well as unique genes such as WER (Lee & Schiefelbein, 1999) and TRN (Cnops et al., 2000). The classic work of Murry & Christianson (1987) and Murry et al. (1987) illustrate other similarities: root hairs are typically endoreduplicate, like trichomes, although amplification disappears under salt stress. The regularity of cell files, cell shape, and division in roots and hypocotyl may permit patterning to occur entirely by cellular fate decisions. The origin and fate of cells from the root apical meristem can be predicted with high accuracy (van den Berg et al., 1998), but spatial referents are a key determinant of their cell fate based on evidence from clonal analysis and arrays of positive and negative genetic regulators (Dolan & Okada, 1999; Kidner et al., 2000). Determinants operate in both radial and circumferential patterning of the root based on the temporal and spatial dynamics of gene expression (Lim et al., 2000). Interestingly, the same mechanism of radial patterning

is postulated for both roots and shoots based on a cassette of genes operating in the shoot, hypocotyl, and root (Benfey, 1999; Wysocka-Diller et al., 2000). The common genes, the need for spatial referents, and the common regulation of shoot and root epidermal cells by the same or similar sets of genes indicate that the entire epidermis may be patterned by a cassette of genes which operate by similar mechanisms. Our findings indicate that tissue patterning of stomata is a separate and distinct process from cellular patterning (differentiation) of stomata, although both are under genetic control and may well use some of the same regulatory elements. This conclusion is based on the demonstration that GL1 and TRY mutations, known previously to affect the differentiation of trichomes, affect the pattern of stomata on cotyledons. These genes are leaf epidermal patterning genes. Expression of these patterning genes indicates that cotyledon pattern is specified by postembryogenic events. While cell lineage and other celllevel factors control the presence and local number of stomata, organ-level pattern is controlled by larger scale spatial referents.

Acknowledgements We thank Huyen Nguyen, Maria Spletter, and Patricia Brack for assistance with reconstruction of digital images and data collection. Gib Ahlstrand and Phil Oshel provided technical assistance with SEM. We also thank Pat Zambryski and Fred Hemphel for Arabidopsis seed. This research was funded by a Vilas Associate Fellows grant.

References Ahlstrand GG. 1996. Low-temperature low-voltage scanning electron microscopy. (LTLVSEM) of uncoated frozen biological samples: a simple alternative. In: Bailey G, Corbett J, Dimlich R, Michael J, Zaluzec N, eds. Proceedings of microscopy and microanalysis. San Francisco, CA, USA: San Francisco Press, 918–919. Benfey PN. 1999. Is the shoot a root with a view? Current Opinion in Plant Biology 2: 39–43. van den Berg C, Weisbeek P, Scheres B. 1998. Cell fate and cell differentiation status in the Arabidopsis root. Planta 205: 483 – 491. Berger D, Altmann T. 2000. A subtilisin-like serine protease involved in the regulation of stomatal density and distribution in Arabidopsis thaliana. Genes and Development 14: 1119–1131. Berger F, Linstead P, Dolan L, Haseloff J. 1998. Stomata patterning on the hypocotyl of Arabidopsis thaliana is controlled by genes involved in the control of root epidermis patterning. Developmental Biology 194: 226–234. Bisseling T, Weigel D. 2001. Plant development: From cell fate to organ formation. Plant Cell 13: 221–227. Bünning E. 1956. General processes of differentiation. In: Milthorpe F, ed. The growth of leaves. London, UK: Butterworths, 18 –30. Charlton WA. 1990. Differentiation in the leaf epidermis of Chlorophytum comosum Baker. Annals of Botany 66: 567–578. Clark SE, Williams RW, Meyerowitz EM. 1997. The CLAVATA1 gene encodes a putative receptor kinase that controls shoot and floral meristem size in Arabidopsis. Cell 89: 575–585. Cnops G, Wang X, Linstead P, Van Montagu M, Van Lijsebettens M, Dolan L. 2000. TORNADO1 and TORNADO2 are required for the specification of radial and circumferential pattern in the Arabidopsis root. Development 127: 3385–3394.



Research Croxdale JL. 2000. Stomatal patterning in angiosperms. American Journal of Botany 87: 1069 –1080. Dolan L, Okada K. 1999. Signaling in cell type specification. Seminars in Cell & Developmental Biology 10: 149 –156. Dolan L, Scheres B. 1998. Root pattern: shooting in the dark? Seminars in Cell & Developmental Biology 9: 201–206. Fletcher JC, Brand U, Running MP, Simon R, Meyerowitz EM. 1999. Signaling of cell fate decisions by CLAVATA3. Arabidopsis shoot meristems. Science 283: 1911–1914. Geisler M, Nadeau J, Sack FD. 2000. Oriented asymmetric divisions that generate the stomatal spacing pattern in Arabidopsis are disrupted by the too many mouths mutation. Plant Cell 12: 2075–2086. Glover B. 2000. Differentiation in plant epidermal cells. Journal of Experimental Botany. 51: 497–505. Glover B, Perez-Rodriques M, Martin C. 1998. Development of several epidermal cell types can be specified by the same MYB-related transcription factor. Development 125: 3497–3508. Hülskamp M, Miséra S, Jürgens G. 1994. Genetic dissection of trichome cell development in Arabidopsis. Cell 76: 555–566. Hung C-Y, Lin Y, Zhang M, Pollock S, Marks MD, Schiefelbein J. 1998. A common position-dependent mechanism controls cell-type patterning and GLABRA2 regulation in the root and hypocotyl epidermis of Arabidopsis. Plant Physiology 117: 73 – 84. Ito M. 2000. Factors controlling cyclin B expression. Plant Molecular Biology 43: 677– 690. Kagan ML, Novoplansky N, Sachs T. 1992. Variable cell lineages form the functional pea epidermis. Annals of Botany 69: 303–312. Kidner C, Sundaresan V, Roberts K, Dolan L. 2000. Clonal analysis of the Arabidopsis root confirms that position, not lineage, determines cell fate. Planta 211: 191–199. Koornneef M, Dellaert WM, van der Veen JH. 1982. EMS- and radiationinduced mutation frequencies at individual loci in Arabidopsis thaliana (L.) Hyenh. Mutation Research 93: 109 –123. Larkin JC, Marks MD, Nadeau J, Sack F. 1997. Epidermal cell fate and patterning in leaves. Plant Cell 9: 1109 –1120. Larkin JC, Oppenheimer DG, Lloyd AM, Paparozzi ET, Marks MD. 1994. Roles of the GLABROUS1 and TRANSPARENT TESTA GLABRA genes in Arabidopsis trichome development. Plant Cell 6: 1065–1076. Larkin JC, Young N, Prigge M, Marks MD. 1996. The control of trichome spacing and number in Arabidopsis. Development 122: 997–1005. Lee MM, Schiefelbein J. 1999. WEREWOLF, a MYB-related protein in Arabidopsis, is a position-dependent regulator of epidermal cell patterning. Cell 99: 473 – 483. Lim J, Helariutta Y, Specht CD, Jung J, Sims L, Bruce WB, Diehn S, Benfey PN. 2000. Molecular analysis of the SCARECROW gene in maize reveals a common basis for radial patterning in diverse meristems. Plant Cell 12: 1307–1318. Marks MD. 1997. Molecular Genetic Analysis of Trichome Development in Arabidopsis. Annual Review of Plant Physiology and Plant Molecular Biology 48: 137–163. Martin C, Glover BJ. 1998. Cellular differentiation in the shoot epidermis. Current Opinion in Plant Biology 1: 511–519. Masucci JD, Rerie WG, Foreman DR, Zhang M, Galway ME, Marks MD, Schiefelbein JW. 1996. The homeobox gene GLABRA2 is required for position-dependent cell differentiation in the root epidermis of Arabidopsis thaliana. Development 122: 1253–1260. Meijer M, Murray JAH. 2000. The role and regulation of d-type cyclins in the plant cell cycle. Plant Molecular Biology 43: 621–633. Melaragno JE, Mehrotra B, Coleman AW. 1993. Relationship between endopolyploidy and cell size in epidermal tissue of Arabidopsis. Plant Cell 5: 1661–1668.


Murry LE, Christianson ML. 1987. Phylogenetic comparison of large nuclear DNA contents of differentiated cells in the roots of Equisetum, Tradescantia, and Hordeum. American Journal of Botany 74: 1772–1778. Murry LE, Christianson ML, Alfinito SH, Garger SJ. 1987. Characterization of the nuclear DNA of Hordeum vulgare root hairs: Amplification disappears under salt stress. American Journal of Botany 74: 1779–1786. Oppenheimer DG, Herman PL, Sivakumaran S, Esch J, Marks MD. 1991. A myb gene required for leaf trichome differentiation in Arabidopsis is expressed in stipules. Cell 67: 483–493. Payne CT, Zhang F, Lloyd AM. 2000. GL3 encodes a bHLH protein that regulates trichome development in Arabidopsis through interaction with GL1 and TTG1. Genetics 156: 1349–1362. Perazza D, Herzog M, Hülskamp M, Brown S, Dorne AM, Bonneville J-M. 1999. Trichome cell growth in Arabidopsis thaliana can be derepressed by mutations in at least five genes. Genetics 152: 461– 476. Sachs T. 1979. Cellular interactions in the development of stomatal pattern in Vinca major L. Annals of Botany 43: 693 –700. Sachs T, Novoplansky N, Kagan ML. 1993. Variable development and cellular patterning in the epidermis of Ruscus hypoglossum. Annals of Botany 71: 237–243. Scheres B. 2000. Non-linear signaling for pattern formation? Current Opinion in Plant. Biology 3: 412–417. Schneider K, Mathur J, Boudonck K, Wells B, Dolan L, Roberts K. 1998. The ROOT HAIRLESS1 gene encodes a nuclear protein required for root hair initiation in Arabidopsis. Genes and Development 12: 2013 –2021. Schnittger A, Folkers U, Schwab B, Jürgens G, Hülskamp M. 1999. Generation of a spacing pattern: the role of TRIPTYCHON in trichome patterning in Arabidopsis. Plant Cell 11: 1105 –1116. Serna L, Fenoll C. 2000. Stomatal development in Arabidopsis : how to make a functional pattern. Trends in Plant Science 5: 458 – 460. Shirzadegan M, Christie P, Seemann JR. 1991. An efficient method for isolation of RNA from tissue cultured plant cells. Nucleic Acids Research 19: 6055. Szymanski DB, Jilk RA, Pollock SM, Marks MD. 1998a. Control of GL2 expression in Arabidopsis leaves and trichomes. Development 125: 1161– 1171. Szymanski DB, Klis DA, Larkin JC, Marks MD. 1998b. cot1: a regulator of Arabidopsis trichome initiation. Genetics 149: 565 –577. Szymanski D, Lloyd AM, Marks MD. 2000. Progress in the molecular genetic analysis of trichome initiation and morphogenesis in Arabidopsis. Trends in Plant Science 5: 214–219. Upton GJG, Fingleton B. 1985. Spatial data analysis by example. New York, USA: Wiley, 79–87. Wada T, Tachibana T, Shimura Y, Okada K. 1997. Epidermal cell differentiation in Arabidopsis determined by a Myb homolog, CPC. Science 277: 1113–1116. Walker AR, Davison PA, Bolognesi-Winfield AC, James CM, Srinivasan N, Blundell TL, Esch JJ, Marks MD, Gray JC. 1999. The TRANSPARENT TESTA GLABRA1 locus, which regulates trichome differentiation and anthocyanin biosynthesis in Arabidopsis, encodes a WD40 repeat protein. Plant Cell 11: 1337–1349. Walker JD, Oppenheimer DG, Concienne J, Larkin JC. 2000. SIAMESE, a gene controlling the endoreduplication cycle in Arabidopsis thaliana trichomes. Development 127: 3931–3940. Wysocka-Diller JW, Helariutta Y, Fukaki H, Malamy JE, Benfey PN. 2000. Molecular analysis of SCARECROW function reveals a radial patterning mechanism common to root and shoot. Development 127: 595–603. Yang M, Sack FD. 1995. The too many mouths and four lips mutations affect stomatal production in Arabidopsis. Plant Cell 7: 2227–2239.

467