SEARCH

SEARCH BY CITATION

Keywords:

  • MIXTA;
  • transcriptome;
  • ROP;
  • trichomes;
  • conical cell;
  • Arabidopsis

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Transcriptome analysis using the Affymetrix ATH1 platform has been completed on purified trichomes from the gl3-sst mutant. These trichomes display immature features, such as glassy cell walls and blunted branches. The gl3-sst trichome transcriptome was greatly enriched for genes involved in lipid biosynthesis, including those mediating the synthesis of fatty acids and wax. In addition, gl3-sst trichomes displayed reduced expression of the R3 MYBs TRY and CPC, which normally function to limit trichome development. The expression of the MIXTA-like MYB gene NOK was elevated. Members of the MIXTA-like family promote conical cell outgrowth, and in some cases, trichome initiation in diverse plant species. In contrast, NOK limits trichome outgrowth in wild-type Arabidopsis plants. Similar to other MIXTA-like genes, NOK was required for the expansion of gl3-sst trichomes, as the gl3-sst nok double mutant trichomes were greatly reduced in size. Expression of NOK in nok mutants reduced branch formation, whereas in gl3-sst nok, NOK expression promoted trichome cell outgrowth, illustrating duel roles for NOK in both promoting and limiting trichome development. MIXTA-like genes from phylogenetically diverse plant species could substitute for NOK in both nok and gl3-sst nok backgrounds. These findings suggest that certain aspects of NOK and MIXTA-like gene function have been conserved.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Studies using Arabidopsis thaliana trichomes have addressed basic biological questions concerning cell fate specification and differentiation (Marks et al., 1991; Szymanski et al., 2000; Schellmann et al., 2007; Larkin, 2009). In Arabidopsis, leaf trichomes are unicellular and typically consist of a stalk and three or four branches (Figures 1a and S1a,e). Arabidopsis trichomes initiate in a non-random, non-cell lineage dependent fashion that is influenced by the cell-to-cell movement of inhibitors of an activation complex that induces trichome cell fate (Larkin et al., 1996; Digiuni et al., 2008). The activation complex contains the proteins TRANSPARENT TESTA GLABRA1 (TTG1; a WD40 repeat protein), GLABRA1 (GL1; an R2R3 MYB), and either GLABRA3 (GL3) or ENHANCHER OF GL3 (EGL3; both containing bHLH domains) (Zhao et al., 2008). This complex controls the transcription of other regulatory genes such as TRANSPARENT TESTA GLABRA2 (TTG2) and GLABRA2 (GL2), which are required for trichome cell outgrowth (Wang and Chen, 2008; Zhao et al., 2008; Morohashi and Grotewold, 2009). Proteins encoded by TTG1 and genes related to GL1 and GL3/EGL3 form a different complex that regulates anthocyanin pigment production (Gonzalez et al., 2008). These complexes are negatively regulated by a family of single repeat R3 MYBs including TRIPTYCHON (TRY) for trichome formation and CAPRICE (CPC) for both trichome and pigment formation (Wada et al., 1997; Schellmann et al., 2002; Wang et al., 2007; Zhu et al., 2009). A model for the interaction between R3 MYBs and the complex was posited in Szymanski et al., 2000; in which the R3 MYB CPC was predicted to compete with GL1 for a binding site on GL3. This model has been supported by experimental data (Payne et al., 2000; Bernhardt et al., 2003; Esch et al., 2003, 2004) and more recently it has been shown that all related R3 MYBs may function in a similar fashion (Wester et al., 2009).

image

Figure 1.  Comparison of mature wild-type and gl3-sst trichomes. (a, b) Col wild-type and gl3-sst trichomes, respectively. Images (c, d) show the same wild-type trichomes spaced by 8.3 h, while (e, f) show the same gl3-sst trichome spaced by 29.2 h. (g–i) gl3-sst trichomes showing different degrees of maturity relative to mature wild-type trichomes. Arrows in (c, d), and in (e, f) highlight the same trichomes. Bars in (a, b, g–i) = 100 μm, and (c–f) = 50 μm.

Download figure to PowerPoint

Arabidopsis trichome development can be divided into six stages. These stages include: (i) termination of cell division followed by radial expansion; (ii) expansion out of the epidermal plane; (iii) branch initiation; (iv) tip-like growth of branches; (v) diffuse expansion of the entire nascent trichome; (vi) cession of expansion and thickening of the cell wall (Hülskamp et al., 1994; Szymanski et al., 1998, 1999; see Figure 2b for images of representative stages). In gl3-shapeshifter (gl3-sst) mutants, trichomes rarely reach stage 6 (Esch et al., 2003; see Figures 1b,g–i and S1c,g online for representative images). The mutation in the gl3-sst results in a weakened interaction between gl3-sst and GL1. When gl3-sst is coupled with siamese1 (sim), the double mutant trichomes cells retain mitotic competency (Walker et al., 2000; Churchman et al., 2006; Marks et al., 2007). Cells in the interior of the resulting trichome clusters undergo little endoreduplication or expansion, and resemble early stage developing trichomes. These trichomes express high levels of GL1 and GASA4, which are markers for early trichome development (Larkin et al., 1993; Kryvych et al., 2008; Marks et al., 2009).

image

Figure 2.  Comparison of developing nok and wild-type trichomes, and suppression of nok phenotype. (a) Mature nok trichome. Developing trichomes in (b) wild-type and (c) nok. (d) Fluorescent image of a stage 2 nok trichome expressing nuclear localized NOK–GFP. (e) Merger of light and fluorescent images of nuclear localized NOK–GFP in rescued nok mature trichomes. (f) SEM of suppressed nok trichome phenotype by TRY:NOK. White arrows in (b) and (c) illustrate trichomes in early to later stages of development. Black arrows in (c) highlight ectopic trichome branch initiation in nok. Bars in (a, e, f) = 100 μm; (b, c) = 50 μm; (d) = 10 μm.

Download figure to PowerPoint

Over 50 loss-of-function mutations in Arabidopsis alter trichome cell fate or development (see Marks et al., 2009 and Morohashi and Grotewold, 2009) for recent lists of mutants. gl3-sst along with noeck (nok), which has larger extra-branched trichomes, are the main subjects of this report. NOK encodes a MIXTA-like R2R3 MYB, and functions to negatively regulate trichome branch formation (Figures 2a and S1b,f) (Folkers et al., 1997; Jakoby et al., 2008). MIXTA and other MIXTA-like genes were first discovered in the asterid Antirrhinum majus (Am) where they promote the outgrowth of conical cells in the petals and can induce the formation of ectopic trichomes in tobacco (Noda et al., 1994; Glover et al., 1998; Perez-Rodriguez et al., 2005; Baumann et al., 2007; Jaffe et al., 2007). Compared to other MIXTA-like genes, the opposite function of NOK could indicate that NOK acquired a different function during evolutionary events leading to trichome formation in Arabidopsis.

We previously preformed a comparative transcriptome analysis of purified wild-type mature and gl3-sst sim trichomes (Marks et al., 2009). Here we have extended the analysis to gl3-sst and gl3-sst nok trichomes. Among the findings was the discovery that NOK was preferentially expressed in gl3-sst trichomes. In contrast to its role as a negative regulator of trichome outgrowth in wild-type, NOK was required for the outgrowth of gl3-sst trichomes. Further, it was found that MIXTA-like genes from divergent plant species could substitute for NOK by limiting trichome branch formation in wild-type and stimulating outgrowth in gl3-sst. Comparative analyses of gl3-sst nok and gl3-sst sim trichomes showed that their transcriptomes are closely related, and that they share many features associated with early stage developing trichomes. These studies shed light on some of the biological activities required for early trichome development and have led to new insights on the relationships between NOK and MIXTA-like genes.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

gl3-sst as a resource for studying the transcriptome of intermediate staged trichomes

Mutant gl3-sst trichomes are under-developed compared with those of wild-type (compare Figure 1a,b; note the glassy cell wall surface of gl3-sst trichomes). A comparison of time-lapse images is shown in Figure 1(c–f). The time between images of the wild-type trichomes in Figure 1(c,d) was 8 h, whereas the time frame between gl3-sst images in Figure 1(e,f) was nearly 30 h. Even though these trichomes initiated at approximately the same time, the wild-type trichomes have expanding branches, while the gl3-sst trichomes have yet to branch. It is only after several days of slow steady expansion that gl3-sst trichomes begin to form branches. The resulting trichomes display varying degrees of stunted development (Figures 1g–i and S1c,g).

RNA was extracted from trichomes isolated from three biological replicates of gl3-sst plants. The RNA was used to generate probes for hybridization to the Affymetrix ATH1 GeneChip® as previously described [(Marks et al., 2008, 2009); see Table S1 for the normalized means and standard deviation for all genes declared detected (i.e. Affymetrix MAS5 P < 0.04) with all values normalized to 1000 per Chip]. The replicates were grown under different conditions and the samples were processed at two locations. These environmental and experimental differences account for some of the variance in the expression values of individual probe sets. Those probe sets showing reduced variances likely represent genes expressed at a constant level in trichomes under most conditions. Those with higher variances likely represent genes with expression patterns more heavily influenced by the environment. For validation, the expression values of 12 transcription factors and four other genes more highly expressed in trichomes relative to intact shoots were considered [see (Marks et al., 2009)]. TTG1 serves as a control as it is expressed throughout the shoot. Eleven of the 12 transcription factors and the four other genes in the list were more highly expressed in gl3-sst trichomes than in shoots, whereas TTG1 and TRY were expressed at similar levels (Table 1).

Table 1.   Comparison of isolated trichome expression values for genes known to be up in trichomes relative to shoots as previously shown in Marks et al. (2009)
AGIGenegl3-sstsst nokPro Lfa
  1. aData from processed leaves –Marks et al. (2009).

  2. bMean ± SD – all values were normalized to 1000.

  3. cnd, not detected.

AT1G79840GL25798 ± 1828b7797 ± 822ndc
AT5G40330MYB238232 ± 227015 960 ± 1497nd
AT2G37260TTG22164 ± 4013640 ± 784nd
AT1G05230HDG21731 ± 6041909 ± 166nd
AT1G73360HDG11492 ± 80225 ± 27nd
AT3G01140NOK1005 ± 80177 ± 27nd
AT3G27920GL1455 ± 4825067 ± 429nd
AT2G46410CPC404 ± 1601580 ± 450nd
AT1G01380ETC15919 ± 23748361 ± 2383nd
AT1G17920HDG1299 ± 63169 ± 47nd
AT1G63650EGL3136 ± 49ndnd
AT5G53200TRY260 ± 140351 ± 94200 ± 101
AT5G24520TTG1842 ± 65572 ± 38863 ± 76
AT5G04470SIM5209 ± 13606032 ± 968275 ± 137
AT1G56580SVB15 650 ± 302439 030 ± 12 5001846 ± 958
AT1G64690BLT133 ± 821588 ± 131nd
AT5G57800YRE4199 ± 13002959 ± 511674 ± 324

To identify predominant cellular activities in gl3-sst trichomes, a Gene Set Enrichment Analysis (GSEA) was performed on highly expressed genes using GeneTrail (http://genetrail.bioinf.uni-sb.de/index.php). The complete output of the analysis of 3245 genes is shown in Table S2. The largest enriched set of genes (427 of 3245) contained those involved in stress responses (GO:0006950) and 35 of these genes were found among the 100 most highly expressed. One of the more prevalent subsets of stress genes corresponded to those involved in dealing with osmotic stress (149; GO:0006970). Other sets of more highly expressed genes were involved in translation (229; GO:0006412), cell walls (195; GO:0005618), vacuoles (262; GO:0005773), plasma membrane (719; GO:0005886), and oxidative phosphorylation (32; GO:0006119).

A previous study indicated that TRY expression was reduced or absent in gl3-sst trichomes (Esch et al., 2004). In this study, the comparison between gl3-sst and wild-type showed that TRY expression was reduced by a factor of 2.52 (< 0.05) (see Table S3). This reduction in TRY expression supports the idea that the gl3-sst trichome phenotype is caused in part by the reduced expression of genes required to limit trichome outgrowth such as TRY. Also in support of this hypothesis, CPC which functions redundantly with TRY, was expressed 6.64-fold lower (< 0.05) (Schellmann et al., 2002). GL1 expression was 3.4-fold higher (< 0.05) in gl3-sst trichomes. This finding was important because previous analyses showed that GL1 expression is highest in early stage trichomes (Larkin et al., 1993). Compared with gl3-sst sim trichomes, GL1 expression in gl3-sst was 7.8-fold lower (Marks et al., 2009). In using GL1 expression levels as a guide for estimating the degree of immaturity, these results along with trichome morphology indicate that gl3-sst trichomes are stunted at an intermediate stage of development.

To identify other differences between gl3-sst and wild-type, several types of analyses were performed. First, genes declared detected in gl3-sst trichomes but not in wild-type trichomes were identified. For those genes detected in both sets, the ratios between the expression values were determined and subjected to a t-test (see Table S3 online for complete results). A list containing genes only detected in gl3-sst (and not in wild-type trichomes) along with those expressed at least 2.5-fold higher in gl3-sst than wild-type was used in an overrepresentation analysis (ORA) using GeneTool. It was found that genes involved in lipid metabolic processes (GO:0006629) were specifically enriched in gl3-sst (P-value (fdr) < 8.56 × 10−6) (Table S4). These genes included those involved in fatty acid and wax biosynthesis. Genes involved in trichome differentiation (GO:0010026) also were over represented [P-value (fdr) < 0.011].

A reverse genetics screen of genes up-regulated in gl3-sst highlights the role of NOK in controlling trichome branching and outgrowth

A search was conducted to identify transcription factors more highly expressed in gl3-sst trichomes than in wild-type mature trichomes. This search led to the detailed analysis of AT3G01140 (MYB106), which was expressed six-fold higher. T-DNA insertion lines for this gene, SALK_110059 and SALK_025449, were obtained from the Arabidopsis Biological Resource Center (ABRC, http://www.arabidopsis.org; (Alonso et al., 2003)). Mutants with extra-branched trichomes were associated with these lines (Figure 2a and S1b,f) and linkage analysis showed that the phenotype required homozygosity of the T-DNA inserts. SEM analysis showed that the extra-branched phenotype was due to a prolonged period of branch formation (compare Figure 2b,c). As these studies were being conducted, Jakoby et al., 2008 reported that AT3G01140 corresponded to the previously characterized NOK locus (Folkers et al., 1997; Jakoby et al., 2008).

A gl3-sst nok double mutant was created to determine if the enhanced NOK expression in gl3-sst limited trichome expansion or branch formation as in wild-type. Unexpectedly, expansion and branch formation were greatly curtailed (Figures 3a and S1d,h). The frequency of trichome initiation and clustering was not altered. Crosses with a second nok allele, yielded a double mutant with a similar phenotype (not shown).

image

Figure 3.  Analysis of the genetic interaction between gl3-sst and nok, and the rescue of double mutant. (a) gl3-sst nok double mutant trichome. (b) Suppression of double mutant phenotype by expression of wild-type NOK fused to GFP coding sequence. (c) Fluorescent image of developing double mutant trichome expressing nuclear localized NOK–GFP. (d) Isolated gl3-sst nok trichome. Arrow in (c) highlights the fluorescent nucleus. Bars in (a, b, d) = 100 μm and in (c)  = 10 μm.

Download figure to PowerPoint

A very early event during normal trichome differentiation is the onset of endoreduplication (Hülskamp et al., 1994). As shown in Figure 4, gl3-sst nok and gl3-sst trichomes, exhibited enhanced endoreduplication compared to mature wild-type trichomes. These analyses also showed that the nok single mutant undergoes extra rounds of endoreduplication, exhibiting levels similar to those seen with try trichomes.

image

Figure 4.  Relative nuclear DNA content in wild-type and mutant trichomes as determined by DAPI staining. X-axis shows relative DNA concentrations using DNA content in guard cells as 2C, and the Y-axis indicates the number of nuclei for each C value.

Download figure to PowerPoint

The gl3-sst nok double mutant transcriptome shares many features with gl3-sst sim double mutant trichomes

A transcriptome analysis was pre-formed on purified gl3-sst nok mutant trichomes (see Figure 3d). Trichome RNA was prepared from three independently grown populations of plants and used for ATH1 Affymetrix analysis as previously described (Marks et al., 2008; see Table S1). The trichome specific gene list in Table 1 was used to validate the gl3-sst nok data set. Eleven of the 12 transcription factors and the four other genes showed enhanced expression in the double mutant trichomes relative to the levels in shoots. As expected, NOK expression only was detected at a very low level.

The transcriptome of gl3-sst nok resembled that of gl3-sst sim (gl3-sst sim data available from ArrayExpress and Marks et al., 2009). In both, the genes encoding proteins involved in translation or ribosome structure were the most highly expressed (represented by approximately 50% of the 150 most highly expressed genes in both datasets). MYB30 and associated lipid metabolism related genes also were highly expressed (Raffaele et al., 2008). GL1 expression was higher in both gl3-sst nok and gl3-sst sim than in gl3-sst trichomes (11-fold and 7.8-fold higher, respectively). The close relationship between the double mutants was further demonstrated by a hierarchical analysis showing that datasets from gl3-sst nok and gl3-sst sim were more closely related to each other than those from processed shoots or gl3-sst (Figure 5). This comparison also revealed the reproducibility of these analyses.

image

Figure 5.  Hierarchical relationships between datasets derived from individual ATH1 GeneChips for processed shoots (Pro Sh), and trichomes isolated from gl3-sst, gl3-sst nok (sst nok), and gl3-sst sim (sst sim).

Download figure to PowerPoint

Morohashi and Grotewold (2009) identified 18 direct gene targets of the GL1/GL3 activation complex. These included known regulatory genes expressed during early trichome development as well as other newly characterized genes. Table S5 shows that of the 18 gene targets, the expression of 17 in gl3-sst sim, 16 in gl3-sst nok, and 16 in gl3-sst was detected in trichomes. Furthermore, 12 in gl3-sst sim, 10 in gl3-sst nok, and seven in gl3-sst were expressed more highly in trichomes than in the processed wild type shoots. GL1 was found to possess increased expression in trichomes versus processed wild type shoots of the same three genotypes. Overall, gl3-sst showed an intermediate level of expression of these genes when compared with gl3-sst sim and gl3-sst nok. These results support the notion that the double and single mutants represent early and intermediately staged trichomes, respectively.

There were some significant differences in the gl3-sst sim and gl3-sst nok profiles concerning the expression of genes involved in the cell cycle. For example, CDKB2s 1 and 2, which promote cell division, were more highly expressed in gl3-sst sim [(295 ± 128 and 322 ± 147 versus 108 ± 27 and not detected); (Boudolf et al., 2004; Andersen et al., 2008)]. CDKA;1, which is required for both endoreduplication and cell division, was expressed at higher levels in gl3-sst nok (5833 ± 584 versus 2084 ± 523; Dissmeyer et al., 2007, 2009; Iwakawa et al., 2006). Similarly, CDC6 expression, which is required for the initiation of DNA replication and endoreduplication, was expressed at a higher level in gl3-sst nok trichomes (308 ± 37 versus 106 ± 32; Castellano et al., 2001). An interesting contrast and potentially inconsistent finding was the elevated expression in gl3-sst nok (1584 ± 358) versus gl3-sst sim (206 ± 62) of LOG7, which encodes cytokinin riboside 5′-monophosphate phosphoribohydrolase that promotes cell division (Kuroha et al., 2009). The key difference between the double mutants was SIM. SIM was the most highly expressed cell cycle related gene in gl3-sst nok.

The triple gl3-sst nok sim mutant was generated. The gl3-sst sim trichomes, which express high levels of NOK, typically have an outer layer of cells exhibiting a more differentiated phenotype than the inner layer of cells (Figure 6a). The triple mutant resembled the gl3-sst sim double with the exception that the outer layer of cells remained relatively undifferentiated (Figure 6b).

image

Figure 6.  SEM analysis of mature double and triple mutant trichomes. Trichome clusters on (a) gl3-sst sim double and (b) gl3-sst sim nok triple mutants. Bars = 100 μm.

Download figure to PowerPoint

Potential basis for morphological differences between gl3-sst and gl3-sst nok trichomes

To begin to identify the molecular mechanisms responsible for the morphological differences between the trichomes on the gl3-sst and gl3-sst nok mutants, a survey was made of genes involved in controlling the cytoskeleton. During trichome development, dynamic arrays of actin microfilaments (AFs) and microtubules (MTs) mediate the changes in directional cell expansion that are associated with the different stages of trichome development (Szymanski et al., 1999). The RAC/ROP regulatory GTPases (referred to as ROPs in this report) likely play an important role in controlling these changes (Fu, 2010). An intriguing difference between gl3-sst and gl3-sst nok trichomes is that ROP2 was expressed five-fold higher in gl3-sst (592 ± 185 versus 115 ± 8; see Table S1). Although ROP2 was expressed at similar levels in mature wild-type (583 ± 121) and gl3-sst trichomes, RIC4, a target of ROP2, was only detected in gl3-sst (428 ± 79). Activation of RIC4 has been found to stimulate the formation of AFs (Fu et al., 2005; Gu et al., 2005). Thus, the transition from gl3-sst nok-like to gl3-sst-like trichomes likely requires the formation of distinct AFs that could be generated through ROP2 signaling. The transition also may involve the repression of ROP11 expression. This ROP was expressed three-fold higher in gl3-sst nok (2755 ± 463) than in gl3-sst (824 ± 218). Expression of a constitutively active ROP11 has been associated with a stabilization of AFs, a loss of endocytosis, and reduced cell expansion (Bloch et al., 2005). These phenotypes fit with the arrested development of gl3-sst nok trichomes. While these ROPs are differentially expressed, NOK is not absolutely required for these changes in expression. ROP11 was highly expressed in gl3-sst sim trichomes, which also express high levels of NOK, and ROP2 was expressed in mature nok trichomes at similar levels as in gl3-sst. However, RIC4 remains a possible direct target as its expression was not detected in either gl3-sst nok or nok trichomes.

Rescue of nok and gl3-sst nok phenotypes by NOK expression

Mutant nok plants were transformed with a construct containing the NOK coding sequence fused to that of GFP under the transcriptional control of the TRY promoter. GFP–NOK protein was specifically detected in trichome nuclei during both early (Figure 2d) and late stages (Figure 2e) of development. This pattern qualitatively mirrors that of NOK expression under the control of its own promoter (Jakoby et al., 2008). Expression of the fused protein suppressed the extra-branch nok phenotype and in many cases resulted in less-branched trichomes [Figures 2f and S1i,j; two branched trichomes and spikes typically are not seen in Col wild-type as shown in Figures 1(a) and S1(a,e)]. The over-correction was correlated with enhanced expression of NOK as shown by qPCR analysis. Three independently isolated under-branched transformants showed higher NOK expression (Table 2 and Figure S3). The GFP–NOK transgene also was moved into the gl3-sst nok double mutant, and as shown in Figure 3(c) the fused gene was appropriately expressed and restored the over-expanded gl3-sst trichome phenotype and induced extra-branch formation (Figures 3b and S1k,l).

Table 2.   Relative expression levels of NOK and HSC70 in wild-type, and nok transformed with TRY:NOK
Plant tissueNOKHSC70
  1. aValues normalized to wild-type.

Wild-type1.0 ± 0.01.0 ± 0.0
TRY:NOK Sa7.0 ± 0.21.5 ± 0.0
TRY:NOK Ua22.8 ± 0.42.1 ± 0.1
TRY:NOK Va5.3 ± 0.11.0 ± 0.0

Function of MIXTA-like genes from diverse species in Arabidopsis

Representative MIXTA-like genes were isolated from Medicago truncatula (Mt), the asterid Antirrhinum majus (Am), and the monocot Dendrobium crumenatum (Dc) also known as pigeon orchid (see Figure S2a online). The relationships between the polypeptide sequences encoded by these MIXTA-like genes are shown in Figures 7 and S4 online. In Medicago, which contains trichomes with long unicellular spikes and smaller multi-cellular glandular trichomes (Damerval, 1983; Pang et al., 2009), three MIXTA-like (MtMYBML) genes were found. MtMYBML1 and 3 were more closely related to NOK (Figure 7). According to the Medicago Gene Expression Atlas [http://bioinfo.noble.org/gene-atlas/v2/ (Benedito et al., 2008)], these genes were highly expressed in young vegetative buds that contain developing trichomes. Based on this information, MtMYBML3 was chosen for this study. Between MIXTA and the three AmMYBML genes in Antirrhinum, AmMYBML2 and 3 were more closely related to NOK (Figure 7). The phylogeny indicates that MIXTA and AmMYBML2 and 3 genes diverged before the monocot/dicot split. Previous RT-PCR analyses showed that AmMYBML3 was expressed throughout the Antirrhinum shoot, and that an AmMYBML3 promoter:GUS reporter gene drove expression of GUS in tobacco trichomes (Jaffe et al., 2007). Accordingly, it was chosen for study. DcMYBML1 was derived from mRNA expressed in developing flowers of Dc. While the flowers of this species do not contain trichomes or conical cells, the labellum [the modified ventral petal of orchid flowers that acts as a main determinate of insect attraction – see Figure S2(a)] contains cells that display an irregular pattern of outgrowth compared to the more regularly shaped cells on the other petals; compare Figure S2(b) (petal surface) with S2(c) (labellum). The position of DcMYBML1 in the phylogeny indicates that it is more similar to genes in the lineage leading to NOK than to MIXTA.

image

Figure 7.  Relationships among MIXTA and MIXTA-like proteins. NOK, Q9LE63; MIXTA, CAA55725; AmMYBML1, CAB43399; AmMYBML2, AAV70655; AmMYBML3, AY661654; MtMYBML1, translated from AC122724_9; MtMYBML2 translated from AC123899_49; MtMYBML3 translated from AC140544_6; AtMYB16, NP_001078589; GL1, AAL01215; DcMYBML1 (in progress). The tree was generated using sequences aligned by ClustalW (see Figure S4 for aligned sequences) for manipulation by MEGA4 software using the Neighbor-Joining method (Tamura et al., 2007; with 1000 replicates). The tree is drawn to scale with a sum branch length of 3.084 (Bar = 0.1).

Download figure to PowerPoint

The chosen MIXTA-like MYBs were expressed in nok trichomes and rescued the extra-branched nok phenotype (Figures 8a,c,e and S5). In addition, the three genes promoted trichome outgrowth and branch formation of gl3-sst nok trichomes (Figures 8b,d,f and S5). The best rescue of the double mutant was induced by the monocot DcMYBML1 (compare Figures 8f with 8b,d). The rescue was less dramatic with either MtMYBML3 or AmMYBML3, but enhanced branch formation and expansion were evident (compare Figure 8b,d with 3a), and some trichomes on these plants resembled those normally found on gl3-sst plants (compare Figure 8 b,d with 1g–i). t-test analyses of trichome size measurements from the non-transformed double mutant and from either of the AmMYBML3 or MtMYBML3 transformants showed that there were significant differences. Trichomes on both AmMYBML3 and MtMYBL3 transformants were 1.75-times larger than those on gl3-sst nok (< 0.0001).

image

Figure 8.  Suppression of nok phenotype by expression of MIXTA-like MYBs. (a, c, e) Reduction in trichome branch number of nok trichomes expressing the MtMYBML3, AmMYBML3, and DcMYBML1, respectively. (b, d, f) Enhanced outgrowth and trichome branch formation on gl3-sst nok trichomes expressing the MtMYBML3, AmMYBML3, and DcMYBML1, respectively. Representative bar for all images in (f) = 100 μm.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Use of mutants to study early stages of trichome development

We have used purified trichomes from mutants as surrogates to study early phases of Arabidopsis trichome development. Affymetrix ATH1 transcriptome studies by others on developing trichomes have been reported (Morohashi and Grotewold, 2009). However, the experimental procedures used were not sufficiently robust to regularly detect many of regulatory genes that were used to validate the data in this report (Table 1). The lack of sensitivity was likely due to the fact that intact plant tissues were used, where trichomes make up only a small fraction of the total number of cells. Morohashi and Grotewold (2009) did use intact shoot tissue to identify new trichome genes using techniques such as qPCR, ChIP-PCR, and ChIP-chip. Many of the genes identified using those techniques were preferentially expressed in the mutant trichomes in this study.

Previously, we characterized the trichome transcriptomes of wild-type and gl3-sst sim plants (Marks et al., 2009). The trichomes on the double mutant are composed of clusters of cells that contain an outer layer of more differentiated cells and a core of cells resembling trichomes at the earliest stages of development (Marks et al., 2007, 2009). The cells in these clusters express high levels of GL1 and other transcription factors highly expressed in young trichomes. The gl3-sst single mutant was studied in this report. Trichomes on this mutant exhibit a prolonged period of slow expansion. The resulting trichomes maintain the morphological features of intermediately staged trichomes, such as a lack of papillae and blunt branch tips. The ability to isolate large quantities of these trichomes has allowed us to study the transcriptome associated with the gl3-sst phenotype.

The amino acid change in gl3-sst protein reduces its interaction between with GL1 (Esch et al., 2003). This change in affinity likely plays a major role in altering the gl3-sst trichome transcriptome. The gl3-sst profile shared some properties with mature wild-type trichomes such as the heightened expression of genes involved in stress responses, and some properties with the gl3-sst sim immature-like trichomes such as the increased expression of genes involved in translation and lipid biosynthesis. The transcriptome profile of gl3-sst trichomes also indicates that the weakened interaction between the gl3-sst and GL1 proteins resulted in a reduction in the expression of negative regulators such as TRY.Morohashi et al. (2007) reported that GL3 can singly bind to the TRY promoter (a complex containing both GL3 and GL1 was required for binding to other trichome gene promoters), but GL1 was still needed for the activation of TRY transcription. It is possible that either the mutant gl3-sst protein has less affinity for the TRY promoter or that the mutation alters its ability to recruit GL1 to the TRY promoter. Either of these would result in reduced TRY expression as seen in gl3-sst.

The analysis of gl3-sst the transcriptome can be justified in several ways. First, as indicated above, gl3-sst trichomes have many features associated with intermediate staged trichomes. Second, similar mutations may have occurred during the evolution of trichome development in other plants. The results presented here shed light on mechanisms that can drastically alter trichome development that may have evolutionary significance. Finally, the analysis of gl3-sst trichomes suggests strategies that can be used to manipulate trichome development in ways that may enhance their ability to defend against herbivorous insects. For example, we have found that the thinned-walled gl3-sst trichomes are preferentially attacked by thrips (M. David Marks, unpublished data). Thus, it should be possible to alter the metabolism of gl3-sst trichomes in ways that can deter thrips and perhaps other insects. Further manipulation may lead to the formation of trichomes that readily rupture and release their chemical contents onto the leaf surface. This action would provide a means to deliver such substances as the defensive proteins encoded by phylloplannin genes (Shepherd et al., 2005).

NOK acts as both a positive and negative regulator of cellular outgrowth

NOK expression was greatly elevated in gl3-sst trichomes. In a wild-type GL3 background, mutations in NOK result in a prolonged period of trichome branch formation (compare Figure 2b,c). Extra expression of NOK in a wild-type or nok background reduced the formation of branches (Figure 2f). These results indicate that NOK functions as a negative regulator of branch formation. NOK is a member of the MIXTA-like MYB gene family, which is probably conserved in all plants (Serna and Martin, 2006). The founding member, MIXTA from the asterid Antirrhinum, functions to promote the outgrowth of conical epidermal cell on the adaxial surface of petals (Noda et al., 1994). Ectopic expression of MIXTA and other related genes also can result in the initiation of ectopic trichomes in the asterid tobacco (Glover et al., 1998). In Physcomitrella patens, MIXTA-like MYBs are required for viability and cell expansion (Leech et al., 1993). The contrasting function of NOK as a negative regulator is unique for this family. For this reason, we expected to observe increased outgrowth of the gl3-sst nok double mutant trichomes. Instead, we found that both trichome expansion and branch formation were greatly reduced in the double mutant. Expression of NOK in the double mutant was able to promote both trichome expansion and branch formation. This shows that NOK can function as both a negative regulator, and similar to other MIXTA-like genes, as a positive regulator of Arabidopsis trichome outgrowth.

Similarities between gl3-sst nok and gl3-sst sim trichomes

Additional characterization of the gl3-sst nok trichomes suggests that they represent an earlier stage in trichome development than gl3-sst trichomes. They expressed nearly the complete suite of transcription factors and other genes that are known targets of the GL1/GL3 activation complex. They also exhibit prolonged endoreduplication, whose onset in wild-type plants marks one of the first visible signs of entry into the trichome pathway (Hülskamp et al., 1994).

The gl3-sst nok transcriptome most closely resembled that of gl3-sst sim trichomes. The trichomes on gl3-sst sim mutants continue to divide and contain an inner cluster of cells resembling pre-stage one to stage two trichomes (Marks et al., 2007). The highest expressed genes in both double mutants were associated with ribosome function and translation, as well as those involved in lipid/wax biosynthesis.

The double mutants did display differential expression of cell cycle genes. CDKB2;1 and 2 were more highly expressed in gl3-sst sim. These CDKs are normally expressed during the G2[RIGHTWARDS ARROW]M phases to promote mitosis (Boudolf et al., 2004; Andersen et al., 2008). In contrast, CDKA;1 expression was higher in gl3-sst nok. CDKA;1 is required for cell division and complete loss of function mutations are lethal; however weak alleles result in reduced endoreduplication and reduced trichome branch formation (Iwakawa et al., 2006; Dissmeyer et al., 2007, 2009). A potential inconsistency in gl3-sst nok was the heightened expression of LOG7. This enzyme activates cytokinin nucleotides and overexpression is associated with enhanced cell division (Kuroha et al., 2009). However, Kuroha et al. (2009) showed that this gene was expressed in immature wild-type trichomes that do not divide. This factor could reflect on the natural ability of cytokinins to delay the cellular differentiation of early stage trichomes, while still allowing or promoting endoreduplication. The key difference was SIM. SIM, which encodes a cyclin D inhibitor, was the highest expressed cell cycle-related gene in gl3-sst nok (Churchman et al., 2006).

Evolution of trichome development

Members of the MIXTA family could substitute for NOK in Arabidopsis. The use of the TRY promoter may have been crucial in this analysis. Previous expression of MIXTA-like genes in Arabidopsis using the CaMV 35SRNA (35S) promoter failed to identify a trichome phenotype (Glover et al., 1998; Payne et al., 1999). In our studies, the expression of NOK with the 35S promoter in nok never resulted in the rescue of the mutant phenotype, but instead often induced a more severe nok-like phenotype (results not shown). This was likely due to co-suppression, as fluorescence from the GFP fused to NOK was not detected. Using the TRY promoter, MIXTA-like genes from diverse plant lineages functioned as both positive and negative regulators of Arabidopsis trichome development. All of the constructs contained GFP fusions, and rescue was always associated with nuclear localized GFP fluorescence.

Given that the Antirrhinum Mixta-like gene used is this study is naturally expressed in Antirrhinum trichomes, the results in this study suggest the presence of a conserved developmental program controlled by MIXTA-like genes. This implies that some of the underpinnings of trichome development in Arabidopsis may have been conserved and divergent evolution has led to the changes unique to Arabidopsis trichomes. This is further supported by the recent finding that a MIXTA-like gene from cotton (Gossypium hirsutum) is a positive regulator of trichome development (Machado et al., 2009). Cotton and Arabidopsis both belong to eurosid II clade.

The ability of the diverged MIXTA-like genes to suppress the nok phenotype in either wild-type or gl3-sst backgrounds may represent conservation in transcription factor binding sites rather than function. This action would be similar to the ability of the R gene from maize to suppress the trichome defects in Arabidopsis GL3 mutants, as R is not required for trichome formation in maize (Lloyd et al., 1992; Zhang et al., 2003). Instead R and other R-like genes in Arabidopsis and other species control anthocyanin biosynthesis. However, as is the case for the widely conserved regulatory mechanisms that control anthocyanin biosynthesis, it is possible that a developmental program regulated by MIXTA-like genes is conserved. Mutations in the single MIXTA-like gene in Petunia hybrida only affect conical cell development (Baumann et al., 2007). This situation implicates this event as the one requiring the potentially conserved development program. Indeed, Baumann et al., 2007 reported subtle changes in the shape of Arabidopsis petal cells in plants overexpressing AtMYB16.

A possible common function for MIXTA-like and NOK genes is highlighted by a potential similarity between conical cell outgrowth and the inhibition of trichome branch formation. During conical cell development, the radial expansion at the base of the cells is restricted resulting in greater aerial expansion. In Phmyb1 and mixta mutants, radial expansion is less restricted resulting in flatter petal cells (Noda et al., 1994; Baumann et al., 2007). Mechanistically, the cell machinery regulated by the MIXTA-like genes could affect conical cell formation by either actively inhibiting lateral expansion or by promoting aerial expansion (Figure 9a). The extra branches on nok mutants appear to be due to the ectopic formation of branch initiation sites (compare Figure 2b,c). NOK regulated genes may encode proteins that suppress these foci using mechanisms similar to those involved in conical cell development. They could either promote branch elongation in a way that suppresses the formation of ectopic branch foci (Figure 9b) or actively inhibit the anticlinal expansion through these nascent foci (Figure 9c). The inappropriate timing of either function could prolong the slow expansion of gl3-sst trichomes by either delaying branch formation or by actively driving cell expansion. The loss of either of these functions in the context of the activities ongoing in gl3-sst trichomes could result in a breakdown of cell expansion as seen in the double mutant. One intriguing difference between gl3-sst and gl3-sst nok trichomes is that ROP2 is expressed five-fold higher in gl3-sst. ROP2 plays an important role in regulating the reorientation of AFs and MTs, and studies have shown that the expression of a constitutive active ROP2 leads to the formation of trichomes that share some characteristics with those on gl3-sst mutants (Fu et al., 2002). This loss of ROP2 signaling could be responsible for the spherical morphology of gl3-sst nok trichomes. The direct or indirect regulation of similar genes by NOK or MIXTA could positively or negatively control cellular outgrowth. Clearly, additional information is needed to understand the mechanisms controlling directional cell wall expansion. The analysis of mutants such as Phmyb1, mixta, gl3-sst, and nok should yield important information concerning this process.

image

Figure 9.  Models for the function of genes regulated by MIXTA-like genes. (a) Conical cell development requires the modulation of cell expansion by either promoting anticlinal (top) or inhibiting periclinal (bottom) expansion. Ectopic branch formation during Arabidopsis trichome development (white arrow) may be prevented by (b) active cell wall expansion in a plane that prevents the formation of the foci or (c) through the inhibition of expansion through these foci.

Download figure to PowerPoint

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Plant material

Columbia (Col-0) Arabidopsis plants were used as wild-type. Seeds for isolating nok-025449 and nok-110059 were derived from SALK_025449 and SALK_110059 lines available from ABRC (http://www.arabidopsis.org; (Alonso et al., 2003)). SALK_110059 was used for all of the analyses described in this report. The gl3-sst nok double mutant was generated by manually crossing kanamycin-resistant nok-110059 pollen onto gl3-sst flowers. Kanamycin-resistant F2 plants were selected on 0.8% agar containing 50 mg L−1 kanamycin. Plants with the reduced trichome phenotype were selected and propagated through single seed descent to the F4 generation. Backcrossing double mutants to wild-type yielded plants with gl3-sst and nok phenotypes. The triple mutant was generated by crossing gl3-sst nok and gl3-sst sim-1 plants. F2 plants exhibiting the clustered mounds of trichomes associated with gl3-sst sim were selected. Seed from individual selected plants were tested for kanamycin resistance, and lines showing 100% resistance were considered to be triple mutants.

Image analysis

SEM analyses were performed as previously described using a Cyro-preparation system (Ahlstrand, 1996; Esch et al., 2004). Time-lapse images were taken at 5-min intervals with a Canon G5 camera (Canon USA Inc., http://www.usa.canon.com) attached to a Nikon SMZ1500 stereomicroscope (Nikon Instruments Inc., http://www.nikoninstruments.com/). Stereoscopic images were captured with the same camera and scope setup. Fluorescence microscopy was performed with a Nikon Diaphot 200 using a QImaging Retiga Exi camera (QImaging, http://www.qimaging.com) as previously described Esch et al. (2004).

Trichome isolation and Affymetrix GeneChip analyses

Three replicates of gl3-sst nok plants were grown under 24 h light during different times of the year in a growth room controlled for temperature (23°C) at the University of Minnesota. Two replicates of gl3-sst plants were similarly grown at the University of Minnesota, and a third was grown at the Samuel Roberts Noble Foundation (Ardmore, OK, USA). gl3-sst trichomes were isolated as described previously in (Marks et al., 2008). gl3-sst nok trichomes were isolated with two modifications. Instead of capturing the dislodged trichomes on a 100 μm mesh, the filtrate obtained after vortexing the seedlings to dislodge the trichomes was strained through the 100-μm mesh. The trichomes were then sedimented by centrifugation at 80 g for 2 min. The pellet was washed in PBS and centrifuged again. RNA was isolated and used to generate probes for hybridization to the Affymetrix ATH1 GeneChip (Affymetrix, http://www.affymetrix.com) as previously described (Marks et al., 2008). The Expressionist software package (Genedata, http://www.genedata.com) was used for the analysis. Data were subsequently transferred to Excel (Microsoft, http://office.microsoft.com) spread sheets for sorting and other manipulations.

Gene constructs

All gene constructs were made using the modified pEGAD vector containing a TRY promoter, EGFP coding sequence, and the gateway RFA recombination sequence (Invitrogen; http://www.invitrogen.com), which was described in detail in (Marks et al., 2009). Total RNA from Antirrhinum and Dendrobium flower buds, and from a mixture of flower buds and shoot tissue from Medicago was isolated using the RNeasy Plant Mini kit (Qiagen; http://www1.qiagen.com/); this RNA was converted into cDNA and used in PCR reactions. The primers used for the Medicago, Antirrhinum and Dendrobium MYBs contained attB recombination sites. The primers used to amplify Dc MYB1 were derived from the first and last 21 bases of the coding sequence of the MYB1 gene from the closely related DendrobiumWoo Leng’ orchid (Wu et al., 2003). The primers for cloning the Medicago and Antirrhinum genes were derived from the native sequences found in GenBank. The template used for cloning the Arabidopsis NOK gene was the plasmid pYAT3G01140 (Gong et al., 2004) obtained from the Arabidopsis Stock Center (http://www.arabidopsis.org). Primers derived from the first and last 20 nucleotides of the NOK coding sequence were used to generate a cDNA that was cloned into pCR8 (Invitrogen). The LR-Clonase II (Invitrogen) recombination kit was used to move all constructs into the pEGAD vector. The resulting constructs were verified by DNA sequencing, and then were moved into the Agrobacterium tumefaciens strain GV3101 via electroporation. The resulting Agrobacterium were used to transform plants as described by (Clough and Bent, 1998). Plants derived from at least three independent transformations for each construct were used, and the genotypes were confirmed by PCR.

qPCR

RNA was isolated from the apexes of wild-type and three independently transformed plant lines containing the TRYproGFP–NOK transgene. This RNA was converted to cDNA and used in the Roche LightCycler for qPCR analysis as previously described (Marks et al., 2008). The primers used for this analysis were: NOK for GAC TAA ACC AAA CCA AGG AAA CGG and NOK rev CCA AGT TGA GAA TGC TAT TCC AG. The HSC70-1 (At5g02500) gene was used a test for RNA integrity, as previously described (Marks et al., 2008). The actual tracings obtained from the LightCycler are presented in Figure S3.

DNA content

The relative fluorescence of DAPI stained trichome and guard cell nuclei was used to measure nuclear DNA content as described in detail in (Marks et al., 2007).

ArrayExpress ID numbers for arrays used in this study

The following raw Cell file data are available from ArrayExpress (http://www.ebi.ac.uk/microarray-as/ae/): E-MEXP-2008 Mature Columbia leaf trichomes; E-MEXP-2009 gl3-sst trichomes; E-MEXP-2021 gl3-sst nok trichomes; E-MEXP-2013 gl3-sst sim trichomes; E-MEXP-2014 processed Columbia shoot tissue used for trichome isolation.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

We thank Dr Jonathan P. Wenger for technical assistance and advice, Edward Stronge for translating (Damerval, 1983), and Dr George Weiblen for comments. We also thank Dr Yu Hao at the University of Singapore for hosting EKG while working with Dc, and Dr. Rick Dixon at the Samuel Roberts Noble Foundation for hosting MDM on sabbatical leave. NSF awards 0343982 and 0605033 to MDM and 0812633 to EKG funded this work.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Ahlstrand, G. (1996) Low-temperature low-voltage scanning microscopy (LTLVSEM) of uncoated frozen biological materials: a simple alternative. In Proceedings of Microscopy Microanalysis (Bailey, G., Corbett, J., Dimlich, R., Michael, J. and Zaluzec, N., eds). San Francisco: San Francisco Press, pp. 918.
  • Alonso, J.M., Stepanova, A.N., Leisse, T.J. et al. (2003) Genome-wide insertional mutagenesis of Arabidopsis thaliana. Science, 301, 653657.
  • Andersen, S.U., Buechel, S., Zhao, Z., Ljung, K., Novak, O., Busch, W., Schuster, C. and Lohmann, J.U. (2008) Requirement of B2-type cyclin-dependent kinases for meristem integrity in Arabidopsis thaliana. Plant Cell, 20, 88100.
  • Baumann, K., Perez-Rodriguez, M., Bradley, D., Venail, J., Bailey, P., Jin, H.L., Koes, R., Roberts, K. and Martin, C. (2007) Control of cell and petal morphogenesis by R2R3 MYB transcription factors. Development, 134, 16911701.
  • Benedito, V.A., Torres-Jerez, I., Murray, J.D. et al. (2008) A gene expression atlas of the model legume Medicago truncatula. Plant J. 55, 504513.
  • Bernhardt, C., Lee, M.M., Gonzalez, A., Zhang, F., Lloyd, A. and Schiefelbein, J. (2003) The bHLH genes GLABRA3 (GL3) and ENHANCER OF GLABRA3 (EGL3) specify epidermal cell fate in the Arabidopsis root. Development, 130, 64316439.
  • Bloch, D., Lavy, M., Efrat, Y., Efroni, I., Bracha-Drori, K., Abu-Abied, M., Sadot, E. and Yalovsky, S. (2005) Ectopic expression of an activated RAC in Arabidopsis disrupts membrane cycling. Mol. Biol. Cell, 16, 19131927.
  • Boudolf, V., Barroco, R., De Engler, J.A., Verkest, A., Beeckman, T., Naudts, M., Inze, D. and De Veylder, L. (2004) B1-type cyclin-dependent kinases are essential for the formation of stomatal complexes in Arabidopsis thaliana. Plant Cell, 16, 945955.
  • Castellano, M.M., Del Pozo, J.C., Ramirez-Parra, E., Brown, S. and Gutierrez, C. (2001) Expression and stability of Arabidopsis CDC6 are associated with endoreplication. Plant Cell, 13, 26712686.
  • Churchman, M.L., Brown, M.L., Kato, N. et al. (2006) SIAMESE, a plant-specific cell cycle regulator, controls endoreplication onset in Arabidopsis thaliana. Plant Cell, 18, 31453157.
  • Clough, S.J. and Bent, A.F. (1998) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J. 16, 735743.
  • Damerval, C. (1983) Micromorphologie des épidermes foliaires chez quelques espèces de Medicago. Can. J. Bot. 61, 34613470.
  • Digiuni, S., Schellmann, S., Geier, F. et al. (2008) A competitive complex formation mechanism underlies trichome patterning on Arabidopsis leaves. Mol. Syst. Biol. 4, 217.
  • Dissmeyer, N., Nowack, M.K., Pusch, S., Stals, H., Inze, D., Grini, P.E. and Schnittger, A. (2007) T-loop phosphorylation of Arabidopsis CDKA;1 is required for its function and can be partially substituted by an aspartate residue. Plant Cell, 19, 972985.
  • Dissmeyer, N., Weimer, A.K., Pusch, S. et al. (2009) Control of cell proliferation, organ growth and DNA damage response operate independently of dephosphorylation of the Arabidopsis Cdk1 homolog CDKA;1. Plant Cell, 21, 36413654.
  • Esch, J.J., Chen, M., Sanders, M., Hillestad, M., Ndkium, S., Idelkope, B., Neizer, J. and Marks, M.D. (2003) A contradictory GLABRA3 allele helps define gene interactions controlling trichome development in Arabidopsis. Development, 130, 58855894.
  • Esch, J.J., Chen, M.A., Hillestad, M. and Marks, M.D. (2004) Comparison of TRY and the closely related At1g01380 gene in controlling Arabidopsis trichome patterning. Plant J. 40, 860869.
  • Folkers, U., Berger, J. and Hülskamp, M. (1997) Cell morphogenesis of trichomes in Arabidopsis: differential control of primary and secondary branching by branch initiation regulators and cell growth. Development, 124, 37793786.
  • Fu, Y.. (2010) ROP GTPases and the cytoskeleton. In Integrated G Proteins Signaling in Plants (Yalovsky, S., Baluska, F. and Jones, A., eds). Berlin Heidelberg: Springer-Verlag, pp. 91104.
  • Fu, Y., Li, H. and Yang, Z. (2002) The ROP2 GTPase controls the formation of cortical fine F-actin and the early phase of directional cell expansion during Arabidopsis organogenesis. Plant Cell, 14, 777794.
  • Fu, Y., Gu, Y., Zheng, Z., Wasteneys, G. and Yang, Z. (2005) Arabidopsis interdigitating cell growth requires two antagonistic pathways with opposing action on cell morphogenesis. Cell, 120, 687700.
  • Glover, B.J., Perez-Rodriguez, M. and Martin, C. (1998) Development of several epidermal cell types can be specified by the same MYB-related plant transcription factor. Development, 125, 34973508.
  • Gong, W., Shen, Y.P., Ma, L.G. et al. (2004) Genome-wide ORFeome cloning and analysis of Arabidopsis transcription factor genes. Plant Physiol. 135, 773782.
  • Gonzalez, A., Zhao, M., Leavitt, J.M. and Lloyd, A.M. (2008) Regulation of the anthocyanin biosynthetic pathway by the TTG1/bHLH/Myb transcriptional complex in Arabidopsis seedlings. Plant J. 53, 814827.
  • Gu, Y., Fu, Y., Dowd, P., Li, S., Vernoud, V., Gilroy, S. and Yang, Z. (2005) A Rho family GTPase controls actin dynamics and tip growth via two counteracting downstream pathways in pollen tubes. J. Cell Biol. 169, 127138.
  • Hülskamp, M., Misra, S. and Jurgens, G. (1994) Genetic dissection of trichome cell development in Arabidopsis. Cell, 76, 555566.
  • Iwakawa, H., Shinmyo, A. and Sekine, M. (2006) Arabidopsis CDKA;1, a cdc2 homologue, controls proliferation of generative cells in male gametogenesis. Plant J. 45, 819831.
  • Jaffe, F.W., Tattersall, A. and Glover, B.J. (2007) A truncated MYB transcription factor from Antirrhinum majus regulates epidermal cell outgrowth. J. Exp. Bot. 58, 15151524.
  • Jakoby, M.J., Falkenhan, D., Mader, M.T., Brininstool, G., Wischnitzki, E., Platz, N., Hudson, A., Hülskamp, M., Larkin, J. and Schnittger, A. (2008) Transcriptional profiling of mature Arabidopsis trichomes reveals that NOECK encodes the MIXTA-like transcriptional regulator MYB106. Plant Physiol. 148, 15831602.
  • Kryvych, S., Nikiforova, V., Herzog, M., Perazza, D. and Fisahn, J. (2008) Gene expression profiling of the different stages of Arabidopsis thaliana trichome development on the single cell level. Plant Physiol. Biochem. 46, 160173.
  • Kuroha, T., Tokunaga, H., Kojima, M., Ueda, N., Ishida, T., Nagawa, S., Fukuda, H., Sugimoto, K. and Sakakibara, H. (2009) Functional analyses of LONELY GUY cytokinin-activating enzymes reveal the importance of the direct activation pathway in Arabidopsis. Plant Cell, 21, 31523169.
  • Larkin, J.C. (2009) Morphological evolution: by any means necessary? Curr. Biol. 19, R953R954.
  • Larkin, J.C., Oppenheimer, D.G., Pollock, S. and Marks, M.D. (1993) Arabidopsis GLABROUS1 gene requires downstream sequences for function. Plant Cell, 5, 17391748.
  • Larkin, J.C., Young, N., Prigge, M. and Marks, M.D. (1996) The control of trichome spacing and number in Arabidopsis. Development, 122, 9971005.
  • Leech, M.J., Kammerer, W., Cove, D.J., Martin, C. and Wang, T.L. (1993) Expression of myb-related genes in the moss, Physcomitrella patens. Plant J. 3, 5161.
  • Lloyd, A.M., Walbot, V. and Davis, R.W. (1992) Arabidopsis and Nicotiana anthocyanin production activated by maize regulators R and C1. Science, 258, 17731775.
  • Machado, A., Wu, Y., Yang, Y., Llewellyn, D.J. and Dennis, E.S. (2009) The MYB transcription factor GhMYB25 regulates early fibre and trichome development. Plant J. 59, 5262.
  • Marks, M.D., Esch, J., Herman, P., Sivakumaran, S. and Oppenheimer, D. (1991) A model for cell-type determination and differentiation in plants. Symp. Soc. Exp. Biol. 45, 7787.
  • Marks, M.D., Gilding, E. and Wenger, J.P. (2007) Genetic interaction between glabra3-shapeshifter and siamese in Arabidopsis thaliana converts trichome precursors into cells with meristematic activity. Plant J. 52, 352361.
  • Marks, M.D., Betancur, L., Gilding, E., Chen, F., Bauer, S., Wenger, J.P., Dixon, R.A. and Haigler, C.H. (2008) A new method for isolating large quantities of Arabidopsis trichomes for transcriptome, cell wall and other types of analyses. Plant J. 56, 483492.
  • Marks, M.D., Wenger, J.P., Gilding, E., Jilk, R. and Dixon, R.A. (2009) Transcriptome analysis of Arabidopsis wild-type and gl3-sst sim trichomes identifies four additional genes required for trichome development. Mol. Plant, 2, 803822.
  • Morohashi, K. and Grotewold, E. (2009) A systems approach reveals regulatory circuitry for Arabidopsis trichome initiation by the GL3 and GL1 selectors. PloS Genet. 5, e1000396.
  • Morohashi, K., Zhao, M., Yang, M., Read, B., Lloyd, A., Lamb, R. and Grotewold, E. (2007) Participation of the Arabidopsis bHLH factor GL3 in trichome initiation regulatory events. Plant Physiol. 145, 736746.
  • Noda, K., Glover, B.J., Linstead, P. and Martin, C. (1994) Flower color intensity depends on specialized cell-shape controlled by a myb-related transcription factor. Nature, 369, 661664.
  • Pang, Y., Wenger, J.P., Saathoff, K. et al. (2009) A WD40 repeat protein from Medicago truncatula is necessary for tissue-specific anthocyanin and proanthocyanidin biosynthesis but not for trichome development. Plant Physiol. 151, 11141129.
  • Payne, T., Clement, J., Arnold, D. and Lloyd, A. (1999) Heterologous myb genes distinct from GL1 enhance trichome production when overexpressed in Nicotiana tabacum. Development, 126, 671682.
  • Payne, C.T., Zhang, F. and Lloyd, A.M. (2000) GL3 encodes a bHLH protein that regulates trichome development in Arabidopsis through interaction with GL1 and TTG1. Genetics, 156, 13491362.
  • Perez-Rodriguez, M., Jaffe, F.W., Butelli, E., Glover, B.J. and Martin, C. (2005) Development of three different cell types is associated with the activity of a specific MYB transcription factor in the ventral petal of Antirrhinum majus flowers. Development, 132, 359370.
  • Raffaele, S., Vailleau, F., Leger, A., Joubes, J., Miersch, O., Huard, C., Blee, E., Mongrand, S., Domergue, F. and Roby, D. (2008) A MYB transcription factor regulates very-long-chain fatty acid biosynthesis for activation of the hypersensitive cell death response in Arabidopsis. Plant Cell, 20, 752767.
  • Schellmann, S., Schnittger, A., Kirik, V., Wada, T., Okada, K., Beermann, A., Thumfahrt, J., Jurgens, G. and Hülskamp, M. (2002) TRIPTYCHON and CAPRICE mediate lateral inhibition during trichome and root hair patterning in Arabidopsis. EMBO J. 21, 50365046.
  • Schellmann, S., Hülskamp, M. and Uhrig, J. (2007) Epidermal pattern formation in the root and shoot of Arabidopsis. Biochem. Soc. Trans. 35, 146148.
  • Serna, L. and Martin, C. (2006) Trichomes: different regulatory networks lead to convergent structures. Trends Plant Sci. 11, 274280.
  • Shepherd, R.W., Bass, W.T., Houtz, R.L. and Wagner, G.J. (2005) Phylloplanins of tobacco are defensive proteins deployed on aerial surfaces by short glandular trichomes. Plant Cell, 17, 18511861.
  • Szymanski, D.B., Jilk, R.A., Pollock, S.M. and Marks, M.D. (1998) Control of GL2 expression in Arabidopsis leaves and trichomes. Development, 125, 11611171.
  • Szymanski, D.B., Marks, M.D. and Wick, S.M. (1999) Organized F-actin is essential for normal trichome morphogenesis in Arabidopsis. Plant Cell, 11, 23312347.
  • Szymanski, D.B., Lloyd, A.M. and Marks, M.D. (2000) Progress in the molecular genetic analysis of trichome initiation and morphogenesis in Arabidopsis. Trends Plant Sci. 5, 214219.
  • Tamura, K., Dudley, J., Nei, M. and Kumar, S. (2007) MEGA4: molecular evolutionary genetics analysis (MEGA) software version 4.0. Mol. Biol. Evol. 24, 15961599.
  • Wada, T., Tachibana, T., Shimura, Y. and Okada, K. (1997) Epidermal cell differentiation in Arabidopsis determined by a Myb homolog, CPC. Science, 277, 11131116.
  • Walker, J.D., Oppenheimer, D.G., Concienne, J. and Larkin, J.C. (2000) SIAMESE, a gene controlling the endoreduplication cell cycle in Arabidopsis thaliana trichomes. Development, 127, 39313940.
  • Wang, S.C. and Chen, J.G. (2008) Arabidopsis transient expression analysis reveals that activation of GLABRA2 may require concurrent binding of GLABRA1 and GLABRA3 to the promoter of GLABRA2. Plant Cell Physiol. 49, 17921804.
  • Wang, S.C., Kwak, S.H., Zeng, Q.N., Ellis, B.E., Chen, X.Y., Schiefelbein, J. and Chen, J.G. (2007) TRICHOMELESS1 regulates trichome patterning by suppressing GLABRA1 in Arabidopsis. Development, 134, 38733882.
  • Wester, K., Digiuni, S., Geier, F., Timmer, J., Fleck, C. and Hülskamp, M. (2009) Functional diversity of R3 single-repeat genes in trichome development. Development, 136, 14871496.
  • Wu, X.M., Lim, S.H. and Yang, W.C. (2003) Characterization, expression and phylogenetic study of R2R3-MYB genes in orchid. Plant Mol. Biol. 51, 959972.
  • Zhang, F., Gonzalez, A., Zhao, M.Z., Payne, C.T. and Lloyd, A. (2003) A network of redundant bHLH proteins functions in all TTG1-dependent pathways of Arabidopsis. Development, 130, 48594869.
  • Zhao, M., Morohashi, K., Hatlestad, G., Grotewold, E. and Lloyd, A. (2008) The TTG1–bHLH–MYB complex controls trichome cell fate and patterning through direct targeting of regulatory loci. Development, 135, 19911999.
  • Zhu, H.F., Fitzsimmons, K., Khandelwal, A. and Kranz, R.G. (2009) CPC, a single-repeat R3 MYB, is a negative regulator of anthocyanin biosynthesis in Arabidopsis. Mol. Plant, 2, 790802.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Figure S1. Additional images of wholes leaves and trichomes of Col wild type, nok and rescued nok plants. (a–d) and (e–h) show ×10 and ×40 stereoscopic images of representative leaves from Col, nok, gl3-sst, and gl3-sst nok, respectively. (i, j) and (k, i) are ×10 and ×40 stereoscopic images of rescued nok, and gl3-sst nok, respectively. Bars in (a–d, i, k) and (e–h, j, l) equal 1 mm and 200 μm, respectively.

Figure S2. Flower on the Pigeon orchid, Dendrobium crumenatum. (a) The organs labeled with LS, DS, LP, and LB are the lateral sepals, the dorsal sepal, the lateral petals, and the labellum or ventral petal, respectively. Black arrows highlight the ridges on the labellum. Adaxial epidermal cells on (b) petal and (c) labellum. Bars equal 200 μm.

Figure S3. qPCR analysis of NOK expression in wild-type and TRY:NOK transformed shoot apexes. RNA was isolated from the young shoot apexes (prebolting) of wild-type and three independent transformants (S, U, V), and was used as templates for cDNA synthesis. Replicate aliquots of cDNA were subjected to qPCR analysis using the primers described in the text and the Roche LightCycler. Hundred-fold dilutions of the target fragments (previously generated by PCR and gel purified) were used for standards. These standards were only used to assess relative relationships between the samples and not for quantification. LightCycler3 software was used to analyze the data. The top panel shows results obtained using NOK primers. The bottom panel used the HSC70 gene as a control for RNA integrity, as we previously found that this gene is fairly uniformly expressed in shoot tissue and trichomes (Marks et  al., 2008). (Note: in the top panel TRY:NOK S and the 10−4 diluted standard co-ran.).

Figure S4. Alignment of amino acid sequences of MIXTA-like MYBs with GL1 as an outgroup. Names and associated accession numbers are shown on the left. Alignment was performed using the ClustalW program in the Lasergene 8 software package.

Figure S5. Additional images of wholes leaves and trichomes of nok and gl3-sst nok rescued with MIXTA-like genes. (a, b, e, f, i, j) are stereoscopic images of nok plants rescued by expression of DcMYBML1, MtMYBML3, and AmMYBML3, respectively. (c, d, g, h, k, l) are stereoscopic images of gl3-sst nok plants rescued by expression of DcMYBML1, MtMYBML3, and AmMYBML3, respectively. In the images of whole leaves, bars = 1 mm; in those showing close-ups of trichomes, bars = 200 μm.

Table S1. Means and standard deviations of genes detected in gl3-sst and gl3-sst nok trichomes. All values were normalized to 1000 on a chip-by-chip basis.

Table S2. GSEA analysis of gl3-sst trichome dataset to identify gene categories with expression signals significantly above the median.

Table S3. Genes detected in wild-type and not in gl3-sst trichomes; detected in gl3-sst and not in wild-type trichomes; t-test of expression levels for genes detected in both gl3-sst and wild-type trichomes.

Table S4. Overrepresentation analysis (ORA) of genes preferentially expressed in gl3-sst trichomes compared with wild-type.

Table S5. Expression levels of direct gene targets of GL3/GL1 as defined by Morohashi and Grotewold (2009).

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

FilenameFormatSizeDescription
TPJ_4329_sm_Table-S1.xls3185KSupporting info item
TPJ_4329_sm_Table-S2.xls44KSupporting info item
TPJ_4329_sm_Table-S3.xls4294KSupporting info item
TPJ_4329_sm_Table-S4.xls40KSupporting info item
TPJ_4329_sm_Table-S5.xls21KSupporting info item
TPJ_4329_sm_Figure-S1.doc1321KSupporting info item
TPJ_4329_sm_Figure-S2.doc1858KSupporting info item
TPJ_4329_sm_Figure-S3.doc371KSupporting info item
TPJ_4329_sm_Figure-S4.doc50KSupporting info item
TPJ_4329_sm_Figure-S5.doc11454KSupporting info item
TPJ_4329_sm_Legends.doc36KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.