SEARCH

SEARCH BY CITATION

Keywords:

  • Arabidopsis thaliana;
  • leaf development;
  • vascular patterning;
  • vein formation;
  • procambium;
  • auxin transport;
  • ATHB8;
  • SHR;
  • SCL29;
  • SCL32

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

The processes underlying the formation of leaf vascular networks have long captured the attention of developmental biologists, especially because files of elongated vascular-precursor procambial cells seem to differentiate from apparently equivalent, isodiametric ground cells. In Arabidopsis leaves, ground cells that have been specified to vascular fate engage expression of ARABIDOPSIS THALIANA HOMEOBOX8 (ATHB8). While definition of the transcriptional state of ATHB8-expressing ground cells would be particularly informative, no other genes have been identified whose expression is initiated at this stage. Here we show that expression of SHORT-ROOT (SHR) is activated simultaneously with that of ATHB8 in leaf development. Congruence between SHR and ATHB8 expression domains persists under conditions of manipulated vein patterning, suggesting that inception of expression of SHR and ATHB8 identifies transition to a preprocambial cell state that presages vein formation. Our observations further characterize the molecular identity of cells at anatomically inconspicuous stages of leaf vein development. Developmental Dynamics, 2011. © 2010 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

The vascular system of plants is a network of veins that extends throughout all organs (Esau, 1965). Veins transport water and nutrients, and are source of signals that act locally, to assign identity to surrounding cells, and systemically, to coordinate initiation of new shoot organs with that of new roots (Berleth and Sachs, 2001). Sites of vein formation are foreshadowed by appearance of files of elongated procambial cells, which in leaf development seem to emerge de novo from within a homogeneous population of isodiametric ground cells (Louis, 1935; Esau, 1943; Foster, 1952).

The molecular events that lead to acquisition of procambial cell identity during leaf development are not entirely clear, but available evidence supports a decisive role for transport and transduction of the plant signaling molecule auxin in specifying paths of leaf vein formation. Auxin application to leaf primordia induces formation of new veins (Sachs, 1975, 1989; Scarpella et al., 2006), and chemical inhibition of auxin transport during leaf development severely disturbs vein patterning (Mattsson et al., 1999; Sieburth, 1999). Consistent with these observations, mutants impaired in auxin biosynthesis, response, or transport display diagnostic alterations in leaf vein patterns (Przemeck et al., 1996; Mattsson et al., 1999; Alonso-Peral et al., 2006; Cheng et al., 2006). During leaf development, ground cells are directed toward procambial fate through induction of wide domains of expression of the PIN-FORMED1 (PIN1) auxin exporter and of the auxin response transcription factor MONOPTEROS (MP; Hardtke and Berleth, 1998; Scarpella et al., 2006; Sawchuk et al., 2007; Wenzel et al., 2007; Donner et al., 2009). Cessation of PIN1 and MP expression occurs in some of the cells, as fields of PIN1 and MP expression become restricted to individual lines of elongating procambial cells (Scarpella et al., 2006; Wenzel et al., 2007; Donner et al., 2009).

While initiation of PIN1 and MP expression identifies a reversible state in leaf vein formation, files of PIN1- and MP-expressing ground cells that are stabilized toward procambial fate activate expression of the class III HOMEODOMAIN-LEUCINE ZIPPER (HD-ZIP III) gene ARABIDOPSIS THALIANA HOMEOBOX8 (ATHB8; Baima et al., 1995; Kang and Dengler, 2004; Scarpella et al., 2004; Sawchuk et al., 2007; Donner et al., 2009). Onset of ATHB8 expression is directly controlled by MP (Donner et al., 2009), and identifies transition to a typically irreversible “preprocambial” cell state that accurately predicts sites of leaf vein formation (e.g., Koizumi et al., 2000; Carland and Nelson, 2004; Kang and Dengler, 2004; Scarpella et al., 2004, 2006; Pineau et al., 2005; Alonso-Peral et al., 2006; Cnops et al., 2006; Petricka and Nelson, 2007; Sawchuk et al., 2007, 2008; Donner et al., 2009; Carland et al., 2010). Therefore, characterization of the transcriptional profile of ground cells that have switched to preprocambial state would be particularly desirable as it may provide insight into the molecular pathways controlling vein formation. However, as of yet, no genes have been identified whose expression in vein development is initiated simultaneously with that of ATHB8.

In this study, we searched for gene expression patterns associated with early stages of vein development in Arabidopsis leaves. We found that onset of expression of SHORT-ROOT (SHR), which encodes a transcription factor of the GRAS family (after GIBBERELLIC ACID INSENSITIVE, REPRESSOR OF gibberellic acid1–3 and SCARECROW; Di Laurenzio et al., 1996; Peng et al., 1997; Silverstone et al., 1998; Pysh et al., 1999; Helariutta et al., 2000), coincides with that of ATHB8 during undisturbed leaf development. Parallel initiation of expression of SHR and ATHB8 persisted under conditions of experimentally manipulated leaf vascular patterning, suggesting that synchronous activation of expression of SHR and ATHB8 operationally defines a reproducible cell state that presages vein appearance. While the ATHB8 protein remained confined to leaf vascular cells, however, the SHR protein additionally localized to adjacent, periveinal positions, suggesting functions of preprocambial cells beyond vein formation. Our observations assist in the molecular characterization of cell state at morphologically indistinguishable, preprocambial stages of leaf vein formation.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

In Arabidopsis leaves, veins are arranged in a ramified pattern that largely reflects the shape of the leaf (Nelson and Dengler, 1997; Candela et al., 1999; Dengler and Kang, 2001; Fig. 1A). Lateral veins depart from either side of a conspicuous central vein (midvein), extend along the leaf margin and connect to distal veins to form prominent closed loops. A series of higher-order veins branch from midvein and loops, and can either terminate in the lamina or join two veins. Veins of succeeding orders become recognizable progressively later in the same area of the developing leaf primordium, and veins of the same order appear in a tip-to-base sequence during leaf development (Telfer and Poethig, 1994; Kinsman and Pyke, 1998; Candela et al., 1999; Mattsson et al., 1999; Sieburth, 1999; Kang and Dengler, 2002, 2004; Steynen and Schultz, 2003; Scarpella et al., 2004; Fig. 1B–D). The illustrations in Figure 1 (Fig. 1A–D) schematically depict the temporal sequence of vein formation events in Arabidopsis leaf development, and define stages and terminology to which we refer throughout this study (for additional details on nomenclature, see the Experimental Procedures section).

thumbnail image

Figure 1. Vein development in the Arabidopsis first leaf. A,C,D: Abaxial (i.e., ventral) view. B: Lateral view (abaxial side to the left). A–D: Illustrations depicting the vein pattern of the mature first leaf (A) and the spatiotemporal course of vein formation in first leaf development (B–D) as inferred from published works (see text for references), and definition of terms used in this study; see also the Experimental Procedures section. B: Two days after germination (DAG). C: Three DAG. D: Four DAG. Green, mature veins; indigo, procambial stages; lavender, preprocambial stages; hv, higher-order vein; l1, l2, and l3, first, second and third loop, respectively; mv, midvein.

Download figure to PowerPoint

Leaf Expression of Root Vascular Markers

All the genes whose expression has previously been assigned to early stages of leaf vein development have also been reported to be expressed in the root procambium (e.g., Baima et al., 1995; Hardtke and Berleth, 1998; Steinmann et al., 1999; Kang and Dengler, 2004; Scarpella et al., 2004, 2006; Alonso-Peral et al., 2006; Konishi and Yanagisawa, 2007; Wenzel et al., 2007; Carland and Nelson, 2009; Gardiner et al., 2010), and identification of leaf vascular gene expression profiles based on root procambial expression has proved to be an effective strategy (Gardiner et al., 2010). Reporter gene expression in the J2501 and Q0990::mGFP5er enhancer-trap lines and in transcriptional fusions to SHR or to WOODENLEG/CYTOKININ RESPONSE1/ARABIDOPSIS HISTIDINE KINASE4 (Mahonen et al., 2000; Inoue et al., 2001; Suzuki et al., 2001; WOL hereafter) has consistently been used as reliable marker of root procambial cells (e.g., Benkova et al., 2003; Birnbaum et al., 2003; Wang et al., 2005; Zhang et al., 2005; Dello Ioio et al., 2007; Hirota et al., 2007; Mustroph et al., 2009; Petersson et al., 2009; Fig. 2A–D). Activation of Q0990::mGFP5er expression in the leaf coincides with acquisition of procambial cell identity (Sawchuk et al., 2007; Fig. 2E), further supporting the value of root procambial expression filtering for discovery of leaf vascular expression patterns. Therefore, to identify new preprocambial expression profiles, we asked whether reporter gene expression in J2501:: mGFP5er and in transcriptional fusions to SHR or WOL retained, like Q0990::mGFP5er, vascular specificity in the leaf. To address this question, we visualized fluorescence protein activity in J2501::mGFP5er and in transcriptional fusions of SHR or WOL to nuclear localized YFP or GFP (HTA6:EYFP or HTA6:EGFP; Zhang et al., 2005), and compared it with that of Q0990::mGFP5er, in first leaves of seedlings 4 days after germination (DAG) as their venation is predominantly preprocambial and procambial (Sawchuk et al., 2007; Donner et al., 2009; Fig. 1D).

thumbnail image

Figure 2. Marker expression in seedling organs. A–H: Overlay of confocal laser scanning and differential interference contrast microscopy, subepidermal focal plane. A look-up-table (LUT; displayed in A), in which black was used to encode background, and cyan, green, yellow, orange, and red to encode increasing signal intensities (Sawchuk et al., 2008), was applied to eight-bit gray scaled images to generate color-coded images. Top right, marker identity. Bottom left, fraction of samples showing the displayed features. A–D: Root tips 4 days after germination (DAG). E–H: Four-DAG first leaves, abaxial view. F: See Supp. Fig. S1 for additional expression patterns and their frequencies. Scale bars = 50 μm.

Download figure to PowerPoint

While, in agreement with previous observations (Sawchuk et al., 2007; Donner et al., 2009), Q0990:: mGFP5er signals in 4-DAG leaves were restricted to procambial midvein and first loops (Fig. 2E), expression of J2501::mGFP5er was not detected (Fig. 2F), and weak WOL::HTA6: EGFP fluorescence was observed in nearly all cells (Fig. 2G). However, territories of SHR::HTA6:EYFP activity were associated with sites of formation of midvein, first and second loops, and higher-order veins (Fig. 2H). Because neither expression of J2501::mGFP5er nor that of WOL::HTA6:EGFP displayed leaf vascular bias, successive characterization focused on SHR::HTA6:EYFP.

Expression of SHR During Leaf Development

Expression of SHR in second loops of 4-DAG leaves (Fig. 2H; compare with Fig. 1D and Fig. 2E), suggests that, like ATHB8 (Kang and Dengler, 2004; Scarpella et al., 2004), SHR is expressed in ground cells that have shifted to preprocambial state. However, patterns of initiation, progression and termination of SHR expression could be dramatically different from those of ATHB8, even if the two genes are expressed similarly at a single stage of leaf development. Therefore, to visualize dynamics of SHR expression in leaf vein formation, we monitored activity of SHR::HTA6: EYFP and of the reference preprocambial marker ATHB8::HTA6:EYFP (Sawchuk et al., 2007; Donner et al., 2009) in first leaf primordia at 2, 3, 4, and 5 DAG.

At 2 DAG, SHR::HTA6:EYFP and ATHB8::HTA6:EYFP signals were confined to a single cell file along the midline of the leaf primordium (Fig. 3A,E). At 3 DAG, SHR and ATHB8 transcriptional fusions were expressed in narrow domains at sites of midvein and first loop formation (Fig. 3B,F). At 4 DAG, slender zones of SHR::HTA6:EYFP and ATHB8::HTA6:EYFP activity marked appearance of midvein, first and second loops, and higher-order veins (Fig. 3C,G). Finally, at 5 DAG, SHR and ATHB8 promoters directed expression in developing midvein, first, second, and third loops, and higher-order veins (Fig. 3D,H). However, while ATHB8::HTA6:EYFP expression had subsided from the apical portion of midvein and first loops (Fig. 3D), the SHR transcriptional fusion was evenly active throughout the leaf vasculature (Fig. 3H).

thumbnail image

Figure 3. ATHB8 and SHR expression in first leaf development. A–H: Overlay of confocal laser scanning and differential interference contrast microscopy, subepidermal focal plane. Top right, leaf primordium age and gene identity. Bottom left, fraction of samples showing the displayed features. A,E: Lateral view (abaxial side to the left). B–D,F–H: Abaxial view. A–D: Green, ATHB8::HTA6:EYFP expression. E–H: Green, SHR::HTA6:EYFP expression. Scale bars = 10 μm in A,E, 20 μm in B,F, 50 μm in C,G, 75 μm in D,H.

Download figure to PowerPoint

In summary, expression of SHR seemed to be tightly associated with regions of ATHB8-labeled vein formation throughout leaf development.

Stage-Specific SHR Expression in Leaf Vein Formation

Comparison between SHR and ATHB8 expression profiles during leaf development (Fig. 3) suggests that expression of SHR is initiated as early as that of ATHB8, and that therefore SHR expression could be assigned to ground cells that have switched to preprocambial state. We adopted two criteria to test such a hypothesis: (1) visualization of shape of cells expressing SHR; (2) detection of SHR and ATHB8 expression within the same sample.

Simultaneous imaging of activity of SHR transcriptional fusions and plasma-membrane-localized GFP (Sawchuk et al., 2008) in basal regions of 4-DAG first leaves showed that, like ATHB8 (Kang and Dengler, 2004; Scarpella et al., 2004; Fig. 4A), SHR is expressed in isodiametric cells (Fig. 4B,C), suggesting that SHR expression is initiated in ground cells.

thumbnail image

Figure 4. Stage-specific SHR expression in leaf vein development. A–O: Details of basal regions (A–C) or second loops (D–O) of first leaves 4 days after germination (DAG), abaxial view. Confocal laser scanning microscopy, subepidermal focal plane. Top right, marker identity. Bottom left, fraction of samples showing the displayed features. A– C: White, UBQ10::EGFP:LTI6B expression. A: Green, ATHB8::HTA6:EYFP expression. B: Green, SHR::HTA6:EYFP expression. C: Green, SHR::mCherry-Nuc expression. D,F: Magenta, ATHB8::HTA6:EYFP expression. E,F,H,I: Cyan, ATHB8::ECFP-Nuc expression. G,I: Magenta, SHR::HTA6:EYFP expression. J,L: Cyan, SHR::HTA6:EYFP expression. K,L,N,O: Magenta, SHR::mCherry-Nuc expression. M,O: Cyan, ATHB8::HTA6:EYFP expression. F,I,L,O: Merge of images in D and E, G and H, J and K, and M and N, respectively. Images are color-coded with a dual-channel LUT from cyan to magenta through green, yellow and red (Demandolx and Davoust, 1997). Fluorescence in each detection channel was displayed in either magenta or cyan. Single-fluorophore images were then merged using a differential operator. As a result, preponderance of cyan signal over colocalized magenta signal is encoded in green, opposite in red, and colocalized cyan and magenta signals of equal intensity in yellow. Scale bars = 5 μm in A–C, 10 μm in D–O.

Download figure to PowerPoint

Covisualization of signals of SHR:: HTA6:EYFP and ATHB::ECFP-Nuc (Sawchuk et al., 2007) in second loops of 4-DAG first leaves showed matching expression of fluorescent reporters (Fig. 4G–I), suggesting that expression of SHR is initiated simultaneously with that ATHB8. To test for possible artifacts induced by fluorophore intrinsic properties (e.g., different maturation times and stabilities of HTA6:EYFP and ECFP-Nuc) or detection parameters (e.g., suboptimal excitation wavelengths and emission intervals), we visualized extent of coexpression between SHR::mCherry-Nuc and ATHB8::HTA6:EYFP signals. The reproducible coincidence of fluorescence in reciprocal permutations of SHR and ATHB8 regulatory regions with YFP and CFP, or mCherry (compare Fig. 4M–O to Fig. 4G–I), suggests that our covisualization data are fluorophore independent, further supporting that expression of SHR and ATHB8 is simultaneously activated in ground cells that have transitioned to preprocambial state.

SHR Expression in Auxin Transport-Inhibited Leaves

Domains of SHR expression may be rigidly specified in leaf development and only incidentally matching with zones of vein appearance. Therefore, we asked whether fields of SHR expression remained associated with areas of leaf vein formation upon experimental interference with vascular patterning. Auxin transport has been shown to define sites of vein appearance in developing leaf primordia (Mattsson et al., 1999; Sieburth, 1999). Therefore, we grew seedlings harboring the SHR and ATHB8 transcriptional fusions in the presence of the auxin transport inhibitor 1-N-naphthylphthalamic acid (NPA) and imaged fluorescent protein expression in first leaves at 3, 4, and 5 DAG.

Leaves of plants germinated and grown in the presence of auxin transport inhibitors are characterized by several reproducible, distinct abnormalities in vein network configuration; most conspicuously, great numbers of broad vein loops that fuse along the entire edge of the leaf, to give rise to a wide marginal zone of vascular differentiation, and that extend parallel to one another at the centre of the leaf, to give rise to a laterally expanded midvein (Mattsson et al., 1999; Sieburth, 1999). As shown in Figure 5, domains of SHR::mCherry-Nuc and ATHB8::HTA6:EYFP expression retained their tight relation to sites of vein formation throughout development of auxin transport-inhibited leaves (Fig. 5A–F). Furthermore, strict congruence between regions of SHR and ATHB8 promoter activity was preserved under conditions of reduced auxin transport (Fig. 5G–I). However, as observed in undisturbed leaf development, SHR::mCherry-Nuc signals persisted at later stages of vein differentiation, while expression of the ATHB8 transcriptional fusion had declined (Fig. 5H,I).

thumbnail image

Figure 5. SHR and ATHB8 expression in auxin transport-inhibited leaves. A–I: First leaves, abaxial view, developing in the presence of 2.5 μM 1-N-naphthylphthalamic acid (NPA). Confocal laser scanning microscopy, subepidermal focal plane. Top right, leaf primordium age and gene identity. Bottom left, fraction of samples showing the displayed features. A–C: Cyan, ATHB8::HTA6:EYFP expression. D–F: Magenta, SHR::mCherry-Nuc expression. G–I: Merge of images in A and D, B and E, and C and F, respectively. Images color-coded with a dual-channel LUT as described for Figure 4. Scale bars = 50 μm.

Download figure to PowerPoint

In conclusion, association between SHR expression domains with areas of ATHB8-marked vein formation observed under undisturbed conditions persisted in auxin transport-inhibited leaves, suggesting noncircumstantial correlation between SHR expression and leaf vein emergence.

SHR Expression in Leaf Vein Development

In the root, SHR transcription is restricted to the procambium, but SHR protein is additionally localized to the cell layer surrounding the root vasculature (Helariutta et al., 2000; Nakajima et al., 2001). We therefore asked whether SHR displayed similar behavior in the leaf. To address this question, we visualized expression of a translational fusion of SHR to YFP in 4-DAG first leaves, and compared it to expression of the nonmobile ATHB8::ATHB8:mCherry translational fusion (Donner et al., 2009).

In agreement with previous observations (Donner et al., 2009), expression of the fluorescently tagged ATHB8 protein mimicked ATHB8 promoter activity in leaf vascular cells (Fig. 6A–C). In contrast, SHR::SHR: EYFP signals were further detected in cells adjacent the preprocambial and procambial domains of expression of the SHR transcriptional fusion (Fig. 6D–I). However, while SHR::SHR: EYFP fluorescence was distributed in both nucleus and cytoplasm of cells within the vascular expression territory, fusion protein localization in the periveinal cell layer was markedly nuclear (Fig. 6D–I).

thumbnail image

Figure 6. SHR expression in first leaves. A–I: First leaves 4 days after germination (DAG), abaxial view. Confocal laser scanning microscopy, subepidermal focal plane. Top right, marker identity. Bottom left, fraction of samples showing the displayed features. A–C,G–I: Details of second loops. D–F: Details of first loops. A,C: Cyan, ATHB8::HTA6:EYFP expression. B,C: Magenta, ATHB8::ATHB8:mCherry expression. D,F,G,I: Magenta, SHR::mCherry-Nuc expression. E,F,H,I: Cyan, SHR::SHR:EYFP expression. C,F,I: Merge of images in A and B, D and E, and G and H, respectively. Images color-coded with a dual-channel LUT as described for Figure 4. Scale bars = 10 μm in A–C, 50 μm in D–F, 5 μm in G–L.

Download figure to PowerPoint

Leaf Expression of SHR-Related Genes

SHR belongs to a small clade of GRAS genes that includes SCARECROW-LIKE29 (SCL29) and SCL32 (Bolle, 2004; Lee et al., 2008). Therefore, we asked whether SCL29 and SCL32 were expressed in the leaf in a pattern similar to that of SHR. To address this question, we visualized expression of transcriptional and translational fusions of SCL29 or SCL32 to YFP in 4-DAG first leaves.

While expression of SCL29 fusions was confined to epidermal cells (Fig. 7A,D), activity of SCL32 fusions was detected at nearly all subepidermal positions (Fig. 7B,E). We therefore asked whether the expression domain of SCL32 in the leaf comprised vascular cells. To address this question, we imaged degree of signal overlap in leaves simultaneously expressing SHR::mCherry-Nuc and transcriptional or translational fusions of SCL32. We observed separate activity of SCL32 fusions and of SHR:: mCherry-Nuc (Fig. 7C,F), suggesting nonvascular expression of SCL32 in the leaf.

thumbnail image

Figure 7. SCL29 and SCL32 expression in first leaves. A–F: First leaves 4 days after germination (DAG), abaxial view. Top right, marker identity. Bottom left, fraction of samples showing the displayed features. A,D: Overlay of confocal laser scanning and differential interference contrast microscopy. B,C,E,F: Confocal laser scanning microscopy. A–C,E,F: Subepidermal focal plane. D: Epidermal focal plane. A: Green, SCL29::HTA6:EYFP expression. B,C: Cyan, SCL32::HTA6:EYFP expression. C,F: Magenta, SHR::mCherry-Nuc expression. Images color-coded with a dual-channel LUT as described for Figure 4. D: Green, SCL29::SCL29:EYFP expression. E,F: Cyan, SCL32::SCL32:EYFP expression. Scale bars = 50 μm.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

While the molecular events that control recruitment of ground cells toward procambium formation in leaf development are largely unknown, available evidence suggests that the selection process culminates with initiation of expression of the HD-ZIP III gene ATHB8 (Koizumi et al., 2000; Carland and Nelson, 2004; Kang and Dengler, 2004; Scarpella et al., 2004, 2006; Pineau et al., 2005; Alonso-Peral et al., 2006; Cnops et al., 2006; Petricka and Nelson, 2007; Sawchuk et al., 2007, 2008; Donner et al., 2009; Carland et al., 2010). Activation of ATHB8 expression defines transition to a morphologically inconspicuous preprocambial cell state that preludes to procambium appearance. Therefore, characterization of the molecular identity of ground cells that have switched to preprocambial state would be particularly informative as it may provide insight into the molecular circuits controlling vein formation.

In this study, we searched for gene expression profiles associated with preprocambial stages of vein development in Arabidopsis leaves. We found that expression of SHR, which encodes a member of the GRAS family of plant-specific transcription factors (Di Laurenzio et al., 1996; Pysh et al., 1999; Helariutta et al., 2000), emerges in synchrony with that of ATHB8 in leaf development, suggesting that parallel activation of expression of SHR and ATHB8 identifies a preprocambial cell state that announces vein formation. However, while ATHB8 protein expression remained confined to developing veins, the SHR protein expression domain further included a contiguous, perivascular cell layer, suggestive of activities of procambium-precursor cells beyond vein formation.

Transition to Preprocambial Cell State

During leaf development, SHR and ATHB8 were expressed in seemingly overlapping subepidermal domains and with amazingly comparable dynamics. Expression of both SHR and ATHB8 was initiated in narrow domains that became associated with sites of vein formation. Further, vein-associated expression fields of SHR and ATHB8 emerged in the same temporal sequence: midvein, first loops, second loops and higher-order veins, third loops. Finally, vein order-specific expression domains of SHR and ATHB8 became apparent at the same stage of leaf development. However, expression of SHR was sustained at all stages of vein formation, while that of ATHB8 became dissipated at later stages of vascular differentiation, in agreement with previous reports (Kang and Dengler, 2002, 2004; Scarpella et al., 2004). Expression of both SHR and ATHB8 was initiated in files of polygonal, isodiametric ground cells, and positions of activation of SHR expression overlapped with sites of initiation of ATHB8 expression, suggesting that SHR is expressed at preprocambial stages of vein development. Moreover, that SHR and ATHB8 preprocambial expression domains reproducibly coincided with one another suggests that expression of SHR is initiated concurrently with transition to ATHB8 preprocambial cell state.

If coincidence between expression of SHR and ATHB8 were merely circumstantial, one would not expect such association to endure under conditions of manipulated ATHB8 expression. Behavior of SHR expression in leaves developing under conditions of reduced auxin transport, which dramatically changes the architecture of ATHB8 expression domains and of vein networks (Mattsson et al., 1999; Sieburth, 1999; Gardiner et al., 2010), was comparable to that observed under undisturbed vein patterning. All aspects of SHR expression, including relation to ATHB8 expression and association with positions of vein formation, proved to be highly reproducible under all experimental conditions. We therefore suggest that, together with ATHB8, activation of expression of SHR defines switch to a morphologically inconspicuous transcriptional state that foreshadows procambial development.

Unlike ATHB8, the SHR protein is additionally localized to a layer of nonvascular cells that surrounds leaf veins. This observation is consistent with events occurring in root development, where SHR movement from vascular to neighboring cells is required for the formation of the cell sheath that envelops the single vein (Helariutta et al., 2000; Nakajima et al., 2001; Gallagher et al., 2004). Leaf veins have long been suspected to provide positional cues that control differentiation of adjoining photosynthetic cell types (Langdale and Nelson, 1991), and the pattern of SHR expression in the leaf suggests that such organizing influence may arise simultaneously with transition to preprocambial cell state.

Correct initiation of ATHB8 expression at preprocambial stages of leaf vein development strictly depends on the presence of a TGTCTG regulatory element in the ATHB8 promoter (Donner et al., 2009). The SHR promoter does not contain any TGTCTG element, suggesting an independent mechanism controlling onset of SHR expression. It will be interesting to understand the molecular basis of SHR preprocambial expression; nevertheless, our findings already contribute to molecularly define cells at incipient stages of leaf vascular development.

Complementary Leaf Expression Profiles of SHR-Related Genes

Members of gene families frequently display overlapping expression profiles (e.g., Mason et al., 2004; Tsuchisaka and Theologis, 2004; Sawchuk et al., 2008). In contrast, the expression of the related SHR, SCL29, and SCL32 genes defines complementary territories of cells in the leaf.

Epidermal domains of SCL29 promoter activity become further compartmentalized by presence of the intronless SCL29 coding sequence. Reports of regulatory elements within the coding region are not unprecedented (e.g., Ito et al., 2003), and various post-transcriptional control mechanisms have been described that could account for the differential behavior of SCL29 transcriptional and translational fusions, including regulated nuclear export (Bailey-Serres et al., 2009), mRNA decay (Belostotsky and Sieburth, 2009), and intercellular mRNA trafficking (Ueki and Citovsky, 2000).

Subepidermal cells that express either type of SCL32 fusion lack expression of the preprocambial marker gene SHR, and mutual exclusivity of SCL32 and SHR expression domains is consistent with the view that photosynthetic and vascular cell identity acquisition represent antagonistic pathways in leaf subepidermal tissue ontogeny (Scarpella et al., 2004; Kang et al., 2007; Sawchuk et al., 2008).

Tissue-specific expression data are available for 21 of the 32 GRAS genes in Arabidopsis, but function is only known for 10 of them (Di Laurenzio et al., 1996; Peng et al., 1997; Silverstone et al., 1998; Pysh et al., 1999; Bolle et al., 2000; Dill et al., 2001; Dill and Sun, 2001; Lee et al., 2002; Wen and Chang, 2002; Greb et al., 2003; Fu et al., 2004; Tyler et al., 2004; Torres-Galea et al., 2006; Fode et al., 2008; Lee et al., 2008). While it will be interesting to learn whether the nonoverlapping expression patterns of SHR, SCL29, and SCL32 are associated with equally distinct functions, our results already assist in the characterization of a family of plant-specific transcription factors in leaf development.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

Terminology and Notation

We apply the generic term “subepidermal” to all positions of the leaf beneath the epidermis. We refer to “ground cells” as polygonal, isodiametric, subepidermal cells of the leaf. We use the terms “procambial” and “procambium” to indicate morphologically identifiable vascular cell precursors. We designate as “preprocambial” all stages of vein development before procambium formation. We adopt the “::” and “:” symbols to denote transcriptional and translational gene fusions, respectively.

Vector Construction

All amplifications were performed on Arabidopsis (Arabidopsis thaliana) ecotype Col-0 genomic DNA using Finnzymes Phusion high-fidelity DNA polymerase (New England Biolabs Inc., Ipswich, MA) and gene-specific primers (Supp. Table S1, which is available online). To generate the SHR::HTA6:EYFP construct, the 2487-bp region from −2503 to −16 of the SHR gene (AT4G37650) was recombined into the pFYTAG vector (Zhang et al., 2005). To generate the SHR::mCherry-Nuc construct, the 2494-bp region of the SHR gene from −2504 to −10 was cloned upstream of a translational fusion of the mCherry coding sequence (Shaner et al., 2004) to the 3xSV40 nuclear localization signal from pEYFP-Nuc (Clontech Laboratories, Mountain View, CA). To generate the SHR::SHR:EYFP construct, the 4107-bp region of the SHR gene from −2514 to +1593 was cloned upstream of the EYFP coding sequence (Clontech) using an Asp-Pro-Gly linker, as described in Gallagher et al., 2004. To generate the SCL29::HTA6:EYFP construct, the 1679-bp region from −1686 to −7 of the SCL29 gene (AT3G13840) was recombined into the pFYTAG vector (Zhang et al., 2005). To generate the SCL29::SCL29:EYFP construct, the 3227-bp region of the SCL29 gene from −1697 to +1530 was cloned upstream of the EYFP coding sequence (Clontech) using a Pro-Asp-Pro-Gly linker. To generate the SCL32::HTA6:EYFP construct, the 2886-bp region from −2888 to −2 of the SCL32 gene (AT3G49950) was recombined into the pFYTAG vector (Zhang et al., 2005). To generate the SCL32::SCL32:EYFP construct, the 4169-bp of the SCL32 gene from −2940 to +1229 was cloned upstream of the EYFP coding sequence (Clontech) using an Asp-Pro-Gly linker.

Plant Material and Growth Conditions

The J2501 and Q0990::mGFP5er enhancer-trap lines of the Haseloff collection (Haseloff, 1999) were obtained from the Arabidopsis Biological Resource Center. The WOL::HTA6:EGFP line was a generous gift of David Galbraith's. The origins of the ATHB8::HTA6:EYFP, UBQ10::EGFP:LTI6B, ATHB8::ECFP-Nuc and ATHB8:: ATHB8:mCherry lines have been described (Sawchuk et al., 2007, 2008; Donner et al., 2009). Seeds were sterilized and germinated, and seedlings and plants were grown, transformed, and selected as described (Sawchuk et al., 2007, 2008). For SHR::HTA6:EYFP, SHR::mCherry-Nuc, SCL29::HTA6:EYFP, SCL32::HTA6:EYFP, SHR::SHR: EYFP, SCL29::SCL29:EYFP and SCL32::SCL32:EYFP, the progeny of 10 to 26 independent transgenic lines were inspected to identify the most representative expression pattern. Successive expression analysis was performed on the progeny of at least three lines per construct, which were selected because of strong fluorescent protein expression that was emblematic of the expression profile observed across the entire series of transgenic lines and that resulted from single insertion of the transgene. In genetic crosses, the progeny of at least two independent lines per construct were examined. For auxin transport inhibition, seeds were germinated on growth medium supplemented with 2.5 μM NPA (Chem Service Inc., West Chester, PA). We define “days after germination” (DAG) as days following exposure of imbibed seeds to light.

Microscopy and Image Processing

Dissected seedling organs were mounted and imaged as described (Sawchuk et al., 2007, 2008; Donner et al., 2009). Brightness and contrast were adjusted through linear stretching of the histogram in ImageJ (National Institutes of Health, http://rsb.info.nih.gov/ij). Signal levels and colocalization were visualized as described (Sawchuk et al., 2008).

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

We thank the Arabidopsis Biological Resource Center, Robert Campbell, David Galbraith, and Roger Tsien for kindly providing seeds and plasmids; and Megan Sawchuk and Osama Odat for critically reading the manuscript. This work was supported by a Discovery Grant of the Natural Sciences and Engineering Research Council of Canada (NSERC). E.S. was supported by the Canada Research Chairs Program. T.J.D. was supported by an NSERC CGS-M Scholarship, an NSERC CGS-D Scholarship and an Alberta Ingenuity Student Scholarship.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
DVDY_22516_sm_SuppInfoFigS1.tif813KSupp. Fig. S1. Additional expression patterns of J2501 in leaves. A–C: The 4 days after germination (DAG) first leaves, abaxial view. Overlay of confocal laser scanning and differential interference contrast microscopy. Green, J2501::mGFP5er expression. Bottom left, fraction of samples showing the displayed features. A: Epidermal focal plane. B,C: Subepidermal focal plane. Note erratic expression in epidermal cells (A, magenta arrow), in subepidermal cells (B, yellow arrow) or in both positions (C). Scale bars = 50 μm.
DVDY_22516_sm_SuppInfoTableS1.xls22KSupplementary Table S1. Sequences of primers used in this study

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.