During the development of multicellular organisms many cell types are produced. Depending on their position, each cell perceives different signals, responds through intracellular signalling pathways and, eventually, adopts a specific cell fate. Subsequent cell differentiation usually involves complex changes. For example, cells exit the mitotic cycle or enter cycles of ENDOREDUPLICATION, the cellular architecture alters to meet the functional requirements of the respective cell type, and the metabolism of the cell changes according to its function. Compared with animals, plant development faces additional constraints because the rigid cell walls prevent any cell movement. In plants, a few single-celled Arabidopsis thaliana model systems — in particular root hairs and trichomes — have greatly improved our understanding of the development of single cells.
The goal of this review is to summarize how the study of A. thaliana trichomes facilitates the understanding of development at the single-cell level. A large number of mutants have been characterized that enabled the identification of subsequent developmental processes. These include the selection of trichomes in a field of epidermal cells, cell-fate determination, changes in the cell-cycle mode and cell-shape control. The genetic, molecular and cell-biological analysis of trichome development has revealed only a few trichome-specific processes, as most developmental steps involve the regulation of general cellular machineries. Therefore, studying the trichome system has provided unique insights into the function of transcription factors, the microtubule and actin cytoskeleton, the cell cycle and cell-death control. The study of all the developmental stages of a single cell is a first step towards an understanding of how general cellular processes are integrated during development.
Steps in trichome development
Shoot epidermal hairs are known as trichomes, a term that is derived from the Greek word for hairs, trichos. Trichomes are found in most plants and can comprise either single or several cells and can be secretory glandular or nonglandular1, 2. The functions that are ascribed to trichomes range from protecting the plant against insect herbivores and UV light, to reducing transpiration and increasing tolerance to freezing3, 4.
Trichomes are an excellent model system because they are of epidermal origin and are therefore easily accessible. In addition, A. thaliana trichomes are not essential for the plant under laboratory conditions, which facilitates the isolation of trichome-specific mutants5, 6. So far, most studies have been carried out on leaf trichomes (Fig. 1). At the base of young leaves, single cells that are spaced out at regular distances in an area of apparently equivalent PROTODERMAL CELLS develop into trichomes7, 8. Incipient trichomes stop mitotic cell divisions and initiate endoreduplication cycles. As a result, the trichome cell increases in size and changes its direction of growth such that it grows perpendicular to the leaf surface. Further growth is characterized by a total of about four endoreduplication cycles that result in a DNA content of 32C (1C is the DNA content of the unreplicated haploid genome), which is accompanied by rapid cell enlargement7, 9. The growing cell undergoes two consecutive branching events, the orientations of which are co-aligned with respect to the basal–distal leaf axis10.
Figure 1 | Trichome development.
A schematic presentation of trichome development is shown at the top. Below, representative examples of mutant or overexpression phenotypes, which enabled the identification of the developmental processes, are shown. First, a trichome cell is selected from protodermal cells. The try cpc (triptychon caprice) double mutant shows a defective trichome pattern. A cell that has adopted the trichome cell fate switches from mitotic cycles to endoreduplication cycles. A situation in which this switch is suppressed is exemplified by a trichome in which a specific B-type cyclin, CYCB1;2, is overexpressed under the control of the trichome-specific GL2 promoter (GL2:CYCB1;2). The glabra2 (gl2) mutant phenotype indicates that cell-fate determination requires specific gene functions. The angustifolia (an) mutant illustrates the requirement of genes to undergo proper branching. The directionality of expansion growth requires a group of several so-called DISTORTED genes, of which the wurm (wrm) mutant is shown. The kaktus (kak) mutant has twice the DNA content of wild-type plants and represents the class of mutants that regulates the number of endoreduplication cycles. Cell-death control can also be studied using trichomes. Certain mutants — and the overexpression of the cell-cycle kinase inhibitor/interactor of cyclin-dependent kinases (ICK) under the control of the trichome-specific GL2 promoter (GL2:ICK), which is shown here — lead to unscheduled cell death. Finally, several mutants have a fragile and glassy appearance and are therefore thought to be involved in the maturation of the trichome cell (not shown). The drawings of developmental stages are adapted from Ref. 6 and the gl2 image is reproduced from Ref. 98. Other images are reproduced with permission as follows: try cpc from Ref. 19 © (2002) Macmillan Magazines Ltd; GL2:CYCB1;2 from Ref. 35 © (2002) Elsevier; an from Ref. 10 © (1997) The Company of Biologists Ltd; wrm from Ref. 76 © (2003) Springer–Verlag GmbH; kak from Ref. 7 © (1994) Elsevier; GL2:ICK from Ref. 91 © (2003) The American Society of Plant Biologists.
A large number of mutations that affect trichome development were identified in several mutagenesis screens5, 6. The mutations helped to define regulatory processes in trichome development according to their specific developmental defects (Fig. 1). The selection of trichome cells and the initiation of the trichome cell fate are under the control of a small group of so-called patterning genes. One gene seems to specifically translate the patterning cues into cell-fate differentiation. The switch from mitotic cycles to endoreduplication cycles and the number of endoreduplication cycles are controlled by the endoreduplication genes. A large number of genes are known to be important for branching. The directionality of expansion growth is affected in the so-called distorted mutants. One mutant is known to cause unscheduled cell death, and several other mutants seem to affect the maturation of the trichome.
Now that the genetic interactions are well understood and most of the genes have been cloned, the emerging picture is that only very few genes are, in fact, trichome specific. Most genes are relevant for many cell types and are involved in general cellular processes (Fig. 1). It seems that mutations in these genes have little effect in most cell types but are crucial during trichome development — possibly because trichomes, with their rapid growth and enormous size, are more demanding.
Trichome patterning and initiation
Wild-type trichomes are initiated with an average distance of about three cells between developing trichomes, and almost never form directly next to each other — as would be expected if they were randomly distributed — which indicates that there must be an underlying patterning mechanism7, 8. A mechanism that would explain trichome patterning by a standardized cell-division pattern that segregates trichome cell fates was excluded by clonal analysis8, 11. It is therefore hypothesized that trichome selection is based on a mutual-inhibition mechanism12, 13, 14, 15 (Fig. 2): cells that are initially equivalent produce a trichome-promoting factor (or factors) that activates a factor (or factors) that suppresses trichome development in the neighbouring cells. This way, cells begin to compete and, due to stochastic fluctuations, individual cells will gain higher levels of the promoting factor, produce more of the suppressing factor and, in turn, inhibit the neighbouring cells more strongly. Eventually, these cells would become committed to the trichome cell fate. For this mechanism to work a number of criteria have to be met (for theoretical considerations, see Box 1). First, the positive and the negative regulators need to be involved in a feedback loop with the activator activating the inhibitor and the inhibitor inhibiting the activator. A second requirement is that the inhibitor can move.
Figure 2 | Redundancy of trichome-patterning genes.
The top row of phenotypes shows the redundancy of the negative regulators of trichome patterning. From left to right, wild-type leaves, the patterning defects of triptychon (try) single, try caprice (cpc) double and try cpc enhancer caprice triptychon (etc1) triple mutants are shown. Compared with wild-type plants, try mutants have only a few trichome clusters that contain 2–3 trichomes. The additional removal of CPC results in an increased cluster size (up to 40 trichomes) and mutations in ETC1 lead to the formation of trichome clusters with several hundred trichomes. It is noteworthy that cpc single mutants (not shown) show a subtle increase of the trichome density and that etc1 mutants show no phenotype. The lower row of phenotypes shows that GLABRA3 (GL3) and ENHANCER OF GL3 (EGL3) are redundant. Both single mutants show a slight reduction in trichome number compared with wild-type plants and the double mutant lacks trichomes. The images are reproduced with permission as follows: top row from Ref. 21 © (2004) Elsevier; lower row from Ref. 18 © (2003) The Company of Biologists Ltd.
The genetic and molecular analysis of the trichome-patterning genes is consistent with this model, although little proof is available that could directly demonstrate the patterning mechanism. Both the positive and the negative regulator are represented by a group of several factors (Fig. 2). Four of the trichome-patterning genes function as positive regulators of trichome development. Mutations in the GLABRA1 (GL1) and TRANSPARENT TESTA GLABRA1 (TTG1) genes each result in the complete absence of trichomes16, 17, whereas GLABRA3 (GL3) and ENHANCER OF GL3 (EGL3) function in a redundant manner — gl3 mutants exhibit fewer trichomes compared with wild-type plants, whereas gl3 egl3 double mutants are devoid of trichomes18. The trichome-suppressing genes are represented by three redundantly acting genes (see below). Mutations in the TRIPTYCHON (TRY) gene result in trichome clusters, mutations in the CAPRICE (CPC) gene cause an increased number of trichomes19, 20 and a single mutant of ENHANCER CAPRICE TRIPTYCHON1 (ETC1), which is an enhancer of cpc and try mutants, is indistinguishable from wild-type plants21.
Genetic analysis has established the functional relationships between the four positive factors. The findings that co-overexpression of GL3 and EGL3, as well as GL3 together with GL1, can rescue the ttg1-mutant phenotype indicates that TTG1 functions upstream of these genes and that the other three factors function together at the same point in the pathway18, 22. These data have been confirmed at the molecular level. All trichome-promoting genes, except for TTG1, encode putative transcription factors. GL1 encodes a MYB-RELATED TRANSCRIPTION FACTOR23, GL3 a BASIC HELIX–LOOP–HELIX (BHLH) FACTOR22, EGL3 is a close homologue of GL3 (Ref. 18), whereas TTG1 encodes a WD40 PROTEIN whose molecular function is unknown24. Yeast two-hybrid data indicate that the four positive factors form a transcriptional-activation complex in which GL3 forms a homodimer that binds to GL1 (Ref. 18,22). The GL3 protein also binds to the TTG1 protein, but through a different domain. No direct interaction was found between GL1 and TTG1 (Ref. 22). It is likely that GL1 and GL3 mediate the transcriptional activation, as both proteins contain transcriptional-activation domains. This complex is expected to be active in trichome precursor cells and be inactivated in all other protodermis cells by one or more known negative regulators of trichome initiation.
The negative regulators TRY, CPC and ETC1 all belong to a small family of single-repeat MYB proteins with no obvious transcriptional-activation domain19, 20, 21. Overexpression of any of these proteins abolishes trichome formation. They seem to function in a highly redundant manner: try mutants have small trichome clusters consisting of two or three trichomes, try cpc double mutants have large clusters of up to 40 trichomes and in try cpc etc1 triple mutants, fields of several hundred trichomes are observed (Fig. 2)7, 19, 21. These negative factors seem to interfere with the function of the transcriptional-activation complex by a competition mechanism25. Three-hybrid analysis has shown that the interaction between GL1 and GL3 is counteracted by TRY, thereby disturbing the formation of the proposed functional transcriptional-activation complex26.
How cell–cell communication and a regulatory feedback loop are achieved is, at present, unknown. However, evidence is available for root-hair patterning in A. thaliana, which requires a set of identical and closely related genes (Box 2). Here it was shown — using a green fluorescent protein (GFP)–CPC fusion protein — that the negative regulator CPC moves between cells, probably via PLASMODESMATA, which indicates that travelling transcription factors can mediate cell–cell communication27. It was also shown that the positive root-hair-patterning genes WEREWOLF (WER; a GL1 homologue) and GLABRA2 (GL2) are involved in a negative-feedback loop with CPC28. It is conceivable that a similar mechanism operates during trichome patterning (Box 2).
Trichome differentiation
The homeodomain leucine-zipper protein that is encoded by the GL2 gene29, 30 is thought to translate the cues that are provided by the patterning genes into cell-specific differentiation of several epidermal cell types including the seed coat, root hairs and trichomes17, 29, 30 (Box 2). This is documented by the finding that co-overexpression of GL1 and the maize R gene product (a homologue of GL3) results in an increased and ectopic expression of GL2 (Ref. 31). Although some evidence indicates that GL2 might also have a role in trichome patterning, most of the available data indicate that GL2 triggers downstream differentiation events32. The gl2-mutant-trichome phenotype is characterized by undifferentiated trichomes that resemble the combined phenotypes of various trichome-morphogenesis genes, which indicates that GL2 activates trichome-specific differentiation genes7, 29. Supporting evidence comes from the root-hair system, where it was shown that GL2 regulates the gene that encodes phospholipase D
1, which, in turn, promotes root-hair differentiation33.
Cell-cycle control during trichome development
Trichome cell-cycle-regulation mutants affect either the switch from mitotic cycles to endoreduplication cycles, or the number of endoreduplication rounds, and thereby the ploidy level (Fig. 3).
Figure 3 | Regulation of the cell cycle in trichomes.
The mitotic cell cycle proceeds through four phases: M (mitosis), G1 (gap 1), S (synthesis) and G2. Endoreduplication cycles cut the cell cycle short by skipping the G2 and M phases. Processes that affect the number of endoreduplication cycles (blue) were identified by mutant analysis. The factors that are involved in the switch from mitosis to endoreduplication are indicated in red. Overexpression studies have shown that the mitotic cyclin CYCB1;2 is sufficient to trigger complete cell divisions. The SIAMESE (SIM) gene represses CYCB1;2 as sim mutants express CYCB1;2 in trichomes. Also CYCD3;1 overexpression causes a switch from endoreduplication to mitosis. In contrast to the situation with CYCB1;2, however, the total number of mitotic/endoreduplication cycles is drastically increased during CYCD3;1 overexpression, which indicates an additional role for this cyclin at G1.
The SIAMESE (SIM) gene suppresses the switch from mitotic divisions to endoreduplication cycles34. In sim mutants, trichomes are multicellular and contain between 2 and 15 cells. If the first cycle is already mitotic, two trichomes are formed instead of one; if the switch to endoreduplication from mitotic divisions is late, multicellular trichomes that are morphologically normal are formed. As SIM has not been cloned yet, little is known about the molecular mechanism that is involved. However, some information is available from a different line of experiments in which the role of known cell-cycle genes in the control of endoreduplication was tested. The expression of cell-cycle genes that are normally not active in trichomes was used to test the effects on trichome endoreduplication. To avoid organism-wide defects that could cause sickness or even lethality, trichome-specific gene promoters were used. B-TYPE AND D-TYPE CYCLINS could trigger the formation of multicellular trichomes, and B-type cyclins are involved in the transition from G2 phase to mitosis. The overexpression of a specific B-type cyclin, CYCB1;2, is sufficient to switch from endoreduplication cycles to mitotic cycles, which indicates that B-type cyclins are important for this regulatory step35. Surprisingly, the overexpression of the D-type cyclin CYCD3;1 can also lead to the switch36. D-type cyclins are thought to control the transition from the G1 to the S phase of the cell cycle in animals; but these results indicate that, in plants, they have an additional function in regulating the entry into mitosis. In sim mutants, CYCB1;2 is expressed in trichomes, which indicates that SIM inhibits the expression of mitotic cyclins. However, this is not its only function, as overexpression of CYCB1;2 in a sim mutant background shows a much stronger phenotype than the single mutants. By contrast, the initiation of mitotic cycles in sim mutants is independent of D-type cyclins36.
The number of trichome endoreduplication cycles is affected in at least ten mutants that display either lower or higher ploidy levels than normal. The picture that is emerging from their genetic and molecular analysis is that several different molecular pathways are involved in the regulation of trichome endoreduplication cycles.
Regulation by patterning genes. Two of the patterning genes that are described above, GL3 and TRY, also function as positive and negative regulators of endoreduplication cycles; try trichomes have a DNA content of 64C and different gl3 alleles exist that have either a reduced or an increased DNA content7, 26. This dual function raises the fascinating possibility that trichome cell-fate choice is functionally linked with cell-cycle regulation.
DNA-catenation-dependent endoreduplication. ROOT HAIRLESS2 (RHL2) and HYPOCOTYL6 (HYP6) are positive regulators of endoreduplication cycles in trichomes and in other cell types37. Both are plant homologues of the archaeal DNA TOPOISOMERASE VI complex. These topoisomerases can DECATENATE DNA and promote ATP-dependent separation of entangled DNA37. It is unclear whether the observed defect in the progression of trichome endoreduplication is due to a physical block of further DNA replication or the activation of a checkpoint that controls progression of the endoreduplication cycle.
Regulation by plant hormones. The plant hormone gibberellin promotes endoreduplication cycles. In spindly (spy) mutants, which exhibit a constitutive gibberellin response, trichomes have a DNA content of 64C (Ref. 38). Conversely, in a mutant that is incapable of gibberellin synthesis, ga1-3, no trichomes are formed39, 40.
Regulation by protein degradation. A class of four trichome mutants, kaktus (kak), rastafari (rfi), polychome (poc) and hirsute (hir), all show a very similar phenotype: they all have a ploidy level of 64C. The cloning of the KAK gene revealed that it encodes a protein with sequence similarity to a UBIQUITIN E3 LIGASE41, 42. It is therefore assumed that ubiquitin-regulated protein degradation controls the progression of endoreduplication.
Regulation by a cell-death pathway. Two lines of evidence indicate that a pathway exists that controls both the progression of endoreduplication cycles (and mitotic cycles) as well as programmed cell death. First, the CONSTITUTIVE PATHOGEN RESPONSE5 (CPR5) gene is involved in both processes. Second, overexpression of an inhibitor of the cell-cycle kinase INHIBITOR/INTERACTOR OF CYCLIN-DEPENDENT KINASES/KIP-RELATED PROTEINS (ICK/KRP) leads to reduced ploidy and early trichome cell death (see below).
Trichome branching
The typical three-dimensional branching pattern of trichomes is a unique model system for studying how several axes of polarity and cell morphogenesis are established. Except for the sim mutant, all mutants described so far affect the number of branches but not their orientation with respect to each other. Genetic and molecular data indicate that several independent molecular pathways participate in trichome branching7, 10, 43, 44, 45, 46 (Fig. 4).
Figure 4 | Regulation of trichome branching.
Trichome branching is controlled by at least four different molecular processes. The molecular analysis of the ANGUSTIFOLIA (AN) gene indicates that transcriptional regulation and/or Golgi-related processes are important for branching. The corresponding an mutant is underbranched. Several mutants that affect microtubule function or organization have reduced branching; for example, the tubulin folding cofactor c (tfcc) mutant has two short branches. The branch number also correlates with the ploidy levels. Higher ploidy levels lead to more branches, and trichomes with reduced ploidy levels have fewer branches. The triptychon (try) mutant shown here has twice the DNA content of wild-type cells and has five branches. A key regulator of branching is the STICHEL (STI) gene, as the corresponding sti mutant trichomes do not initiate branches. However, the biochemical mechanisms through which STICHEL regulates branching are, at present, unknown. The images are reproduced with permission as follows: an, try and sti from Ref. 10 © (1997) The Company of Biologists Ltd; tfcc from Ref. 51 © (2002) Elsevier; WT from Ref. 76 © (2003) Springer–Verlag GmbH.
Regulation by endoreduplication levels. The number of branches on a trichome depends on the ploidy level of the cell. Tetraploid plants, in which the DNA content of all cells is doubled, have trichomes with supernumerary branches47. Similarly, mutants that have trichomes with increased DNA levels — such as kak, poc, rfi, try and spy — have trichomes with up to eight branches7, 47. The reduction of the ploidy level results in a reduced branch number — as found in cpr5, gl3 rhl2 and hyp6 (Refs 7,37,48). It is likely that this observed control of trichome branching by the ploidy level is indirect, and that the cell size or the time of actual cell growth provides the frame for branch initiation.
Regulation by microtubules. Microtubules have an important role in trichome branching, as shown in experiments with microtubule antagonists. If the microtubule cytoskeleton is defective during trichome growth, the cell expands almost equally in all directions (this is known as isotropic growth) and does not initiate branches49. Several branching genes encode components that are involved in the biogenesis of 
/
-tubulin dimers, the formation and stability of microtubules or microtubule-based transport processes. Two weak mutants of TUBULIN FOLDING COFACTOR (TFC)A and TFCC — which are involved in the correct folding of tubulins and therefore the formation of assembly-competent / tubulin dimers — exhibit a 'bloated' and underbranched trichome phenotype50, 51, 52. The analysis of the microtubule cytoskeleton in these mutants sheds some light on how the microtubules are reoriented during branch initiation. As the microtubule density and orientation is normal, it is likely that the failure of branch formation is due to problems in de novo synthesis rather than in the reorientation of pre-existing microtubules. If branching requires the synthesis of new microtubules, it is conceivable that it also requires the fragmentation of pre-existing ones to allow a reorientation of growth. This view is supported by the reduced-branching phenotype of mutants of the KATANIN-P60 gene53, 54, 55. Katanins are known to cut pre-existing microtubules into smaller fragments. Microtubule reorientation during branch formation is therefore assumed to be controlled by severing pre-existing microtubules combined with de novo synthesis.
The spatial control of the orientation of microtubules is regulated by at least two branching genes. The FASS/TONNEAU2 gene regulates microtubules in the context of cell divisions as can be inferred from the observation that in fass/tonneau2 mutants cell-division orientation is randomized56, 57. It encodes a novel protein, phosphatase-2A regulatory subunit, which indicates that it regulates microtubules by protein phosphorylation58. The second regulator of microtubule organisation is SPIKE59. This protein shows sequence similarity to CDM-family adaptor proteins (Caenorhabditis elegans CED-5; Homo sapiens DOCK180; Drosophila melanogaster MYOBLAST CITY). These proteins function as guanine nucleotide-exchange factors (GEFs)60 and are thought to modulate the cytoskeleton through small RHO-like GTPases (known as ROPs in plants)61.
In addition, specific microtubule-based transport processes seem to be important for branch formation. The branching gene ZWICHEL (ZWI) encodes a calmodulin-binding kinesin motor protein that binds microtubules in a calmodulin-dependent manner62, 63, 64, 65, 66, 67. The activity of ZWI is modulated by the KIC protein, which binds to ZWI in a Ca2+-dependent manner68. This indicates that ZWI-dependent transport processes might ultimately be controlled by the intracellular second messenger Ca2+.
Regulation by transcription or Golgi-related processes. The ANGUSTIFOLIA (AN) gene regulates branching by two possible pathways, by Golgi-related transport processes or by transcriptional co-activation. It encodes a protein with sequence similarity to carboxy-terminal binding protein (CtBP) and brefeldin-A-ribosylated substrate (BARS)69, 70. In D. melanogaster, CtBP binds to the zinc-finger transcription factors and functions as a transcriptional co-repressor71. In the rat, BARS proteins were identified as proteins that are ADP-ribosylated after treatment with the fungal toxin brefeldin A. Brefeldin-A treatments result in the transformation of Golgi stacks into a tubular-reticular network and it is therefore thought that BARS is involved in Golgi functions72, 73. Biochemical data are not available for the plant CtBP/BARS protein; however, the findings that an mutants have microtubule defects and that AN physically interacts with ZWI in a yeast two-hybrid screen indicates that AN regulates microtubule organization69.
Regulation by the STICHEL gene. The STICHEL (STI) gene regulates trichome branching in a dosage-dependent manner; branch reduction is subtle in weak sti alleles, becomes more pronounced in stronger alleles and trichomes are unbranched in null-alleles. Conversely, overexpression of STI leads to extra branch formation74. This genetic behaviour indicates a key regulatory role for STI, although its molecular function is still elusive. STI encodes a protein that contains a domain with sequence similarity to eubacterial DNA-polymerase-III subunits. However, it is unlikely that STI functions as a DNA polymerase subunit, as no replication effects were found to be associated with the branching phenotype.
An underlying scheme of how branch formation is controlled is not evident from the current analysis of the branching genes. One model, however, accommodates all the available data by assuming that branching is evolutionarily derived from multicellular trichomes, in which branching is the result of a certain division pattern (Box 3).
Directionality of trichome cell expansion
Like most plant cells, trichomes enlarge several-fold during the later stages of differentiation and expand in a polarized manner. This expansion occurs, unlike in the growing tip of root hairs or pollen tubes, along the whole cell surface75, 76. The directionality of expansion growth is affected in mutants of eight genes, which are collectively known as the DISTORTED genes. All distorted mutants show a very similar phenotype: trichomes show turns and twists, some regions of the cell become bulged and others are underdeveloped. Following the movement of small beads that had been placed on the trichome surface, it was shown that this phenotype is caused by the regionally unbalanced expansion of the cell76.
Findings from different experimental approaches indicate that the directionality of trichome cell expansion depends on the actin cytoskeleton. First, the application of drugs that interfere with actin function perfectly phenocopies the distorted mutant phenotype75, 77. Second, the actin cytoskeleton is organized aberrantly in distorted mutants (Fig. 5b)75, 76, 77. Third, all DISTORTED genes that have been cloned so far encode components of the ARP2/3 COMPLEX78, 79, 80, 81, which promotes actin polymerization by enhancing F-actin nucleation and side-binding activities that result in the initiation of fine actin filaments from pre-existing F-actin82, 83.
Figure 5 | Control of expansion polarity.
Trichome cell expansion is controlled by the actin cytoskeleton. a | In wild-type cells, actin is organized in long filaments (as indicated by the arrow). b | Mutants of the genes that are collectively known as the DISTORTED genes exhibit fragmented actin (as indicated by the arrow). All distorted-gene mutants, except for one, show the same phenotype, which is a result of the fact that trichome expansion is no longer coordinated. c | The phenotype of one distorted-gene mutant, the wurm mutant, is shown as a scanning electron microscope picture. d | Inside a cell of one of the distorted mutants, the crooked mutant, both growing regions (left panel) and non-growing regions (right panel) show actin and Golgi vesicles. The difference is that growing regions have fine actin, whereas non-growing regions have bundled actin. This indicates that fusion of vesicles with the membrane is only promoted in regions with fine actin. Online supplementary information S2 (movie) and S3 (movie) show actin-based movement in peroxisomes, which, despite the actin-organization defects, is not generally affected in the mutants (see online supplementary information S3 (movie)) compared with the wild-type cells (see online supplementary information S2 (movie)). The images are reproduced with permission as follows: parts a and b from Ref. 79 © (2003) The American Society of Plant Biologists; part c from Ref. 76 © (2003) Springer–Verlag GmbH; the image in part d from Ref. 78 © (2003) The Company of Biologists Ltd.
The analysis of the distorted mutants demonstrated that actin has a role in expansion growth that goes beyond its mere requirement for general growth. The observation that in distorted mutants actin-based movement of organelles, such as peroxisomes or the Golgi, is not generally impaired indicates that F-actin is still functional (see supplementary information S2 (movie) and S3 (movie))78, 79. Defects were found locally in those parts of the cell that were not growing (Fig. 5d). Non-growth regions contain heavily bundled actin, whereas regions in the distorted mutants that exhibit growth comprise a fine network of actin known as 'fine actin'. It is conceivable that the creation of a local fine-actin network promotes the transport of membrane and cell wall material for the actual growth. It is speculated that the actin cytoskeleton is also involved in the fusion of membranes, as the fusion of small vacuoles, which normally leads to the formation of the large central vacuole, does not take place in distorted mutants79.
It is unknown how the ARP2/3-complex-dependent formation of fine actin is spatially controlled in trichomes. Some of the canonical pathways such as the Rho and Rac/Cdc42 signal-transduction pathways that are known in animals and yeast are, in principle, present in plants, although they are strongly modified. In agreement with this, ROPs were shown to control the local actin configuration in epidermal cells and downstream components, such as the HSPC300 (haematopoetic stem/progenitor-cell clone-300) complex, are known to be involved in the control of actin organization84, 85, 86.
Cell-death control in trichomes
The analysis of trichome development has revealed two pathways that suppress cell death and also regulate endoreduplication (see above). One pathway is represented by ICK/KRP, which shows homology to the animal cell-cycle inhibitor p27Kip1 (Ref. 87). In animals, p27Kip1 can induce apoptosis in the absence of growth factors in some specific cell types88. When ubiquitously expressed in the whole plant, ICK/KRP causes severe growth reduction87, 89, 90, and when expressed under the control of a trichome-specific promoter, trichome cells stop endoreduplication cycles after two cycles and begin to die with symptoms that are characteristic of programmed cell death, such as the degeneration of CHROMOCENTRES and nucleoli91.
A second pathway is linked to the response of plants to plant pathogens. A number of mutants mimic the plant pathogen response. Many of these mutants show a cell-death phenotype combined with growth defects92. One of these mutants, cpr5, shows a trichome phenotype that is similar to that of ICK/KRP-overexpressing lines; trichomes have a ploidy level of about 8C and undergo unscheduled cell death48. It seems that in both cases cell-cycle or endoreduplication-cycle progression and the control of cell death are somehow linked; however, the mechanistic basis of this link remains to be determined.
Control of maturation
Trichome maturation is affected in a group of diverse mutants in which adult trichomes appear transparent or underdeveloped. Three poorly characterized mutants, chablis, chardonnay and retsina, have transparent trichomes and the underdeveloped trichome mutant has no papilla on the trichome surface93. The trichome birefringence mutant is defective in the production of cellulose94.
Conclusions and perspectives
Almost all trichome genes are involved not only in trichome development, but also in the development of other cell types and represent important components of generally important regulatory pathways. The analysis of trichome initiation has uncovered an evolutionarily conserved gene cassette of transcription factors that are involved in patterning processes and anthocyanin-synthesis control. Their evolution and functional diversification will be very interesting to study. Also, the theoretical model that explains pattern formation (Box 1) is far from being proven; for example, at present, there are no target promoters known that could be used to test the genetic predictions. Therefore, it will be challenging to show not only that the inhibitor proteins can move, but also how this is relevant for patterning.
Several pathways seem to have a role in how the switch from mitosis to endoreduplication and the cycle number are controlled. The isolation of further genes, in combination with trichome-specific overexpression approaches, should be a valuable addition to the cell-cycle field. The analysis of branching genes has led to the identification of proteins that are involved in processes as different as intracellular transport, cell-size control, transcriptional control and Golgi-dependent processes, as well as still unknown processes such as those controlled by STI. Each group of genes has opened new research areas in the plant sciences and it will be interesting to see whether the common branching phenotype will tie these processes together. With the discovery that cell-expansion genes encode components of the ARP2/3 complex, key components that regulate actin-based growth have been identified and will allow the study of the up- and downstream regulatory processes in plants. Further analysis of trichomes as a single-cell model system offers the chance to connect the above-mentioned, seemingly unrelated, processes in the future.