(Solanum lycopersicum) leaf mutants

710

Leaflet initiation is temporally and spatially separated in simple and complex tomato (Solanum lycopersicum) leaf mutants: a developmental analysis

Botany Downloaded from www.nrcresearchpress.com by Guangzhou Jinan University on 06/05/13 For personal use only.

Julie Kang and Neelima R. Sinha

Abstract: Formation of a compound leaf requires the involvement of multiple factors, including KNOX1 gene expression. To further characterize simple and complex tomato leaf mutants, we analyzed their morphology and development by assessing: leaf phenotypes, primary leaf morphogenesis, expression of the class I KNOX gene LeT6, and meristematic activity of the marginal blastozone. Mutants with alterations in lobing and (or) pinnation (decrease/increase) were analyzed. Primary leaflet initiation is delayed in mutants with decreased lobing. In contrast, leaflet initiation is advanced or similar to the wild type in mutants with deep lobes. Leaves with increased pinnation along the rachis require a protracted developmental program to form their final leaf morphology. Using a morphometric analysis, we show that leaf complexity can be quantified. The expression pattern of LeT6 correlates with histological analysis of meristematic activity of the marginal blastozone, suggesting that LeT6 may play a role, through some unknown mechanism, to regulate meristematic competence, not only in the marginal blastozone to regulate leaflet lobing, but along the entire length of the leaf to regulate pinnation in compound leaves. Key words: compound, KNOX, LeT6, leaf, leaflets, tomato. Re´sume´ : La formation d’une feuille composeé nećessite l’implication de multiples facteurs, incluant l’expression du ge`ne KNOX1. Afin de poursuivre la caracte´risation de mutants simples et complexes de la feuille de tomate, les auteurs ont analyse´ leur morphologie et leur de´veloppement en examinant : les pheńotypes foliaires, la morphogene`se de la feuille primaire, l’expression du ge`ne LeT6 de la classe 1 KNOX et l’activite´ me´riste´matique de la blastozone marginale. Ils ont analyse´ des mutants avec alte´rations dans la formation des lobes et (ou) la pinnation (augmentation/diminution). L’ini` l’oppose´, tiation des folioles primaires est retardeé chez les mutants avec une diminution de la formation des lobes. A l’initiation des folioles est devanceé ou semblable a` celle du type sauvage chez les mutants avec lobes profonds. Les feuilles avec pinnation accrue le long du rachi nećessitent un programme de´veloppemental plus long pour former leur morpho` l’aide d’une analyse morphome´trique, les auteurs montrent qu’il est possible de quantifier la logie foliaire finale. A complexite´ foliaire. Le patron d’expression du LeT6 montre une corre´lation avec l’analyse histologique de l’activite´ me´riste´matique de la blastozone marginale, ce qui sugge`re que le LeT6 peut jouer un roˆle, via un mećanisme inconnu, dans la re´gulation de la compe´tence me´riste´matique, non seulement dans la blastozone marginale pour re´guler la formation des lobes des folioles, mais tout au long de la feuille pour re´guler la pinnation chez les feuilles composeés. Mots-cle´s : composeé, KNOX, LeT6, feuille, folioles, tomate.

Introduction The main function of leaves, light capture and photosynthesis, remains the same despite vast morphological differences across the plant kingdom. Leaves are initiated on the periphery of the shoot apical meristem (SAM) and can be simple or compound. The primary difference between simple and compound leaves is the number of segments that make up the leaf. Simple leaves have only one blade unit that can have smooth to deeply lobed margins. Compound Received 19 January 2010. Accepted 27 April 2010. Published on the NRC Research Press Web site at botany.nrc.ca on 27 July 2010. J. Kang and N.R. Sinha.1 Section of Plant Biology, 1002 Life Sciences, University of California Davis, Davis, CA 95616, USA. 1Corresponding

author (e-mail: [email protected]).

Botany 88: 710–724 (2010)

leaves have multiple blade units, can vary in their morphology (peltate, palmate, pinnate), and can have multiple leaflet orders that are lobed or unlobed (Bharathan et al. 2002; Kim et al. 2003a). Leaves undergo three phases of development: (i) leaf initiation; (ii) primary morphogenesis; and (iii) secondary morphogenesis or leaf expansion (Dengler and Tsukaya 2001). During leaf initiation, leaves are produced at the SAM and establish adaxial and abaxial leaf polarity. Primary leaf morphogenesis is described as the phase where the basic leaf form is established. This phase also encompasses cellular processes such as cell division and differentiation of internal tissues, as well as leaf lamina and margin formation. Finally, during secondary morphogenesis or leaf expansion, leaves increase in surface area and may retain or alter their shape through differential patterns of expansion (Dengler and Tsukaya 2001). While we are beginning to understand the roles of each of these phases during compound leaf development, it has only been in recent years that genetic ad-

doi:10.1139/B10-051

Published by NRC Research Press


Kang and Sinha

vances have been made to answer the question of how the compound leaf pattern is established, particularly during primary leaf morphogenesis. The class1 Knotted-like homeobox (KNOX1) genes play a central role in regulating meristem maintenance and leaf morphology. In simple leafed species, KNOX1 genes are required for the acquisition of meristematic cell fate and SAM maintenance. One of the first indications of determinate primordium initiation at the indeterminate SAM is the downregulation of KNOX1 genes. For example, in simple leafed dicot species such as Arabidopsis thaliana and the monocot species Zea mays, loss of function in shoot meristemless (STM) and knotted1 (kn1), respectively, results in plants that are unable to maintain the SAM (Long et al. 1996b; Vollbrecht et al. 2000). Furthermore, KNOX1 genes in species with simple leaves are expressed in the SAM, and are downregulated in developing leaf primordia (Smith et al. 1992; Lincoln et al. 1994). In compound leafed species, however, KNOX1 genes are expressed in the SAM, downregulated in initiating leaf primordia, reappear in developing leaf primordia, and function during primary leaf morphogenesis (Hareven et al. 1996; Janssen et al. 1998b; Bharathan et al. 2002; Hay and Tsiantis 2006). Downregulation of the KNOX1 gene STM reduces leaf complexity in Cardamine hirsuta (Barkoulas et al. 2008). Overexpression of KNOXI genes in Arabidopsis (simple leafed species) increases marginal serrations (Lincoln et al. 1994; Chuck et al. 1996; Ori et al. 2000; Pautot et al. 2001), while overexpression of KNOX1 genes in tomato (compound leafed species) results in increased leaf orders via prolonged cell proliferation (Hareven et al. 1996; Chen et al. 1997; Parnis et al. 1997; Janssen et al. 1998a; Champagne et al. 2007; Barkoulas et al. 2008). Thus, expression of KNOX1 genes, together with cell proliferation, is required for compound leaf morphogenesis. One factor that plays an integral role in controlling the degree of leaf compounding is the meristematic activity of the marginal blastozone. The marginal blastozone (or marginal meristem) is described as a region along the lateral margins of leaf primordia where cells retain meristematic activity and organogenic potential (Hagemann and Gleissberg 1996; Kaplan 2001). Anatomically, the marginal blastozone derivatives include trichomes along the leaf margin, as well as specialized marginal cells and tissues (Kessler et al. 2001). In simple leafed dicot species, the meristematic activity of the marginal blastozone is short lived, ceasing only after a few plastochrons (Hagemann and Gleissberg 1996; Donnelly et al. 1999). The activity of the marginal zone does not extend entirely into the leaf base, defining a clear distinction between blade and rachis (Kang and Dengler 2005). In contrast, activity of the marginal blastozone is prolonged in compound leafed species, allowing for higher orders of branching to occur (Hagemann and Gleissberg 1996; Donnelly et al. 1999; Gunawardena and Dengler 2006). The reiteration of the branching process, called blastozone fractionation, occurs through enhancement and suppression of localized growth along the leaf margin (Hagemann and Gleissberg 1996). Currently, studies of genes that regulate both stem cell activity and (or) affect leaf morphology can be used to infer the role of the marginal blastozone during compound leaf development (Dolan

711

and Poethig 1998; Kessler et al. 2001; Kim et al. 2003b; Ori et al. 2007). There have been numerous studies on the anatomy, morphology, and development of tomato leaf mutants (Mathan and Jenkins 1962; Coleman and Greyson 1976; Dengler 1984; Chandra Sekhar and Sawhney 1990). These classical studies have provided insight into simple leafed mutants such as Lanceolate (La) (Mathan and Jenkins 1962; Stettler 1964; Dengler 1984) to more compound leaf mutants such as solanifolia (sf) (Chandra Sekhar and Sawhney 1990). More recently, regulatory factors that play a key role in compound leaf development in tomato have been identified (Avivi et al. 2000; Wang et al. 2005; Brand et al. 2007; Jasinski et al. 2007, 2008; Ori et al. 2007; Zhang et al. 2007; Barkoulas et al. 2008; Blein et al. 2008; Kimura et al. 2008; Berger et al. 2009; David-Schwartz et al. 2009). It is likely that the juxtaposition of both positive and negative factors along the rachis and in the lobes and sinuses of the leaves is a requirement for leaflet outgrowth and lobe formation in a compound leaf. While our knowledge of genetic factors that regulate compound leaf development is increasing, information regarding positional and temporal timing of leaflet formation along the main axis of the leaf is still lacking. Thus, a thorough study of positional and temporal regulation in wild type and mutant plants is necessary to fully understand compound leaf development. The vast number and variability of leaf shape mutants in tomato make this species an excellent model system to investigate mechanisms regulating leaf development. Since class I KNOX genes are known to play a critical role in leaf compounding (Jackson et al.1994; Hareven et al. 1996; Chen et al. 1997; Parnis et al. 1997; Janssen et al. 1998a; Kim et al. 2003b; Barkoulas et al. 2008), we analyzed the expression pattern of LeT6 (TKN2) in tomato leaf mutants. We also provide a detailed analysis of young and mature leaf morphology, leaf development during primary morphogenesis, and histological analysis of the marginal blastozone. The overall goal of this comparative study was to characterize the developmental morphology of tomato leaf mutants, focusing on leaflet initiation and leaf lobing, to gain more insight into compound leaf morphogenesis.

Materials and methods Plant material and growth conditions Wild type and mutant seed stocks were obtained from the Tomato Genetics Resource Center (TGRC) at the University of California, Davis, California, USA. All mutants used in this study are in the Ailsa Craig (AC) background except Lanceolate (La) (primitive cultivar) and trifoliate2 (tf2) (‘Condine red’). AC was used as the wild type in this study, since Kessler et al. (2001) found no differences in mature leaf phenotype when several wild-type cultivars were analyzed. We selected these particular mutants because each show combinations of leaflet reduction with or without lobes, or increased leaflet production with or without lobes, allowing us to examine more closely how these phenotypes (lobing/leaflets) develop in relation to each other. Seeds were sterilized in 10% bleach, rinsed 3 in distilled water for 15 min each, and planted directly into soil. Seedlings were grown in a Conviron E7 growth chamber (Controlled Published by NRC Research Press

712


Environments, Inc., Pembina, North Dakota, USA) under long day conditions [day: 18 h light (150 mmolm–2s–1) at 25 8C; night: 6 h dark at 22 8C] and then transferred to greenhouses to set seed. Days post germination (DPG) was used as a measure of staging to ensure that collected leaves were chronologically the same age. Leaf 5 was examined post-germination to avoid heterochronic variation between genotypes. The following stages were used in our study unless stated otherwise: ‘‘young leaves’’ (*45 d post-germination), ‘‘intermediate leaves’’ (*75 d post-germination), and ‘‘mature leaves’’ (fully expanded). Scanning electron microscopy Scanning electron microscopy (SEM) was conducted on young seedlings (*25 DPG) when leaf 5 was approximately at the P3 stage. Since we wanted to assess the overall development of leaflet initiation and lobe formation, P4 in the same shoot was also analyzed. Tissue was fixed in FAA (formalin – acetic acid) overnight, dehydrated through an ethanol series, and critical point dried with CO2 on a Tousimis Samdri-780A (Tousimi, Rockville, Maryland, USA). Samples were sputter coated with a Denton Vacuum Desk II (Denton Vacuum, Moorestown, New Jersey, USA) and viewed at 5 kV with a Hitachi S-3500N (Hitachi, Pleasanton, California, USA) scanning electron microscope. Images were captured digitally using Quartz PCI imaging (Hitachi, Pleasanton, Calif.). Approximately 20 individuals for each genotype were examined to confirm developmental timing of leaf primordia. Light microscopy Young leaves for each genotype (N = 15–20) were collected, cleared in 70% ethanol, and mounted in the same medium. For histology, shoots 25–30 DPG were collected, fixed in 70% FAA (formalin – acetic acid – 70% ethanol; 1:1:18, v/v/v), dehydrated through an ethanol series, and embedded in paraffin wax (Paraplast1, McCormick Scientific, St. Louis, Missouri, USA). Blocks were sectioned at 7 mmolL–1 on a Microme HM340E microtome (Heidelberg, Germany) and mounted on slides (Probe-On Plus, Thermo Fischer, Waltham, Massachusetts, USA). Slides were deparaffinized in Histoclear (National Diagnostics, Atlanta, Georgia, USA), hydrated through an ethanol series, stained with 0.5% aqueous toluidine blue O (TBO), and mounted in PermountTM (Fischer Scientific, Pittsburg, Pennsylvania, USA). Antibody production and immunolocalization The anti-LeT6 antibody was made by Research Genetics (Huntsville, Alabama, USA) using the MAP (Multiple Antigen Peptide) technique (Tam 1988) to the peptide sequence SFIDPQAEDRELK. Our 13 amino acid LeT6 peptide sequence has perfect sequence matches in two species within the genus Solanum (S. tuberosum L. and S. torvum Sw.). The peptide sequence shows two amino acid mismatches with the LeT6/STM ortholog in a species of Capsicum (C. annuum L.) (Capsicum is the sister group to the Solaneae, the group that contains species within the genus Solanum). However, our antibody fails to produce a signal on tissue from this species (N.R.S. Sinha, unpublished data). Further, the peptide sequence alignment to TKN1 (KNAT1) reveals 3 amino acid mismatches including a serine, phenyl-

Botany Vol. 88, 2010

alanine, and glutamine (STM) change to proline, glutamic acid, and arginine (TKN1), respectively. This would make the protein structures drastically different in this region. While the expression pattern of TKN1 is somewhat similar to LeT6 (Janssen et al. 1998b; Kim et al. 2003b), the antibody is presumed to not cross react with TKN1, or expression would be detected in the other species, like tobacco, which we have tested (N.R.S. Sinha, unpublished data). Combined, these data suggest the antibody is unique to detecting the STM-like genes from the genus Solanum. Each rabbit was immunized with 50 mg of peptide–MAP antigen in 500 mL of 10 mmolL–1 Tris (pH 7.5). This solution was injected 1:1 with Freunds incomplete adjuvant. Rabbits were subsequently boosted in an identical manner at 4 week intervals. Final bleeds were collected at *15 weeks. Peptide antigen was immobilized on an activated support, and antisera were affinity purified against the peptide and washed. Specific antibodies were eluted via a pH gradient, collected and stored in a borate buffer (0.125 molL–1 total borate). The concentration of the purified sera was determined (0.13 mgmL–1) and flow through titered at 1:32 (Research Genetics, Huntsville, Ala.). Immunolocalizations were performed as described (Kim et al. 2003b), pre-immune serum used as our control (dilutions of 1:500 and 1:1000) and done in triplicate for all genotypes. The polyclonal antibody was diluted 1:100 and specificity of expression compared to LeT6 insitu hybridization patterns (N.R.S. Sinha, unpublished data). Bright field images of sectioned material were taken on a Nikon Eclipse E600 (Nikon, Melville, New York, USA) and using the Spot RT camera (Diagnostic Instruments Inc., Sterling Heights, Michigan, USA). Bright field images of entire leaves were taken on a Zeiss Discovery V8, digital pictures taken on an AxioCam MRc, and images saved using AxioVision Ac (Carl Zeiss MicroImaging Inc, Thornwood, N.Y.). All images were processed in Adobe Photoshop CS31 (Adobe Systems Inc., San Jose´, Calif.). Reverse transcription analysis Total RNA (2 mg) was extracted from young leaves (*45 DPG; leaves *1 cm) using QIAGEN RNeasy Miniprep according to the manufacturer’s protocol (QIAGEN, Valencia, Calif.). Three biological replicates were collected for each genotype. SuperscriptIII (Invitrogen, Carlsbad, Calif.) with random hexamers was used according to the manufacturer’s recommendation. Solanum lycopersicum Actin primers (SlActin-1 5’-CGAACCGAGAAAATGACT-3’; SlActin-2 5’CTCAGCATCCTTACCGTTCT-3’) (GenBank accession SLU60478) were used to examine Actin mRNA levels. LeT6 primers are as follows: LeT6-1 5’-GCTCATCCTCACTACCATCG-3’; LeT6-2 5’-CACCACTACTACTACTGCTACG-3’ (GenBank accession AF000141). An annealing temperature of 58 8C was used for all primers with 40 cycles shown in Fig. 7. RT-PCR was replicated three times. Leaf shape factor Leaf shape factor (LSF) was calculated as a measure of leaf complexity. Area and perimeter of young, intermediate, and mature leaves were measured (N = 10). Circularity, which is defined as circularity = 4p(areaperimeter–2), was used as the parameter to measure leaf shape. A circularity value of 1.0 represents a perfect circle while a value apPublished by NRC Research Press


Kang and Sinha

proaching 0.0 represents an increasing elongate shape (Image J version 1.38; National Institute of Health). To determine the LSF, the –ln of the circularity value was calculated so that LSF ranges from 0 to ? for increasing leaf complexity. All measurements were conducted in ImageJ (National Institute of Health). Data for leaf area and LSF was assessed using a one-way ANOVA. Genotypes with the same letter are not significantly different, while those with different letters differ significantly by the Dunnett multiple comparison test (mutants) versus the control (wild type) (P < 0.001). Data are means of total samples (±SE). Analyses were performed using SigmaPlot1 and SigmaStat1 software (SPSS Science, Illinois, USA).

Results Wild-type tomato leaf development Development of wild-type ‘Ailsa Craig’ (AC) tomato leaves was analyzed to provide a baseline comparison for mutant leaves. The morphology of mature leaves was analyzed to determine the degree of compounding. Mature AC leaves are bipinnately compound with a terminal leaflet (TL) and three pairs of primary leaflets (PL) that are borne along the rachis (Figs. 1a–1b). The terminal and primary leaflets have moderately deep lobes. Secondary leaflets form along the petiolule of the primary leaflets while three to four pairs of intercalary leaflets (IL) develop along the rachis between the primary leaflets (Figs. 1a–1b). Secondary and intercalary leaflets have smoother margins with few or no serrations. To determine at what point during leaf development primary leaflet initiation and lobe formation occurs in wildtype leaves, SEM images and cleared young leaves were analyzed. The first pair of primary leaflets arise along the terminal leaflet margin in late leaf primordia 3 (P3)/early leaf primordia 4 (P4) (Fig. 2a). Subsequent pairs of primary leaflets form basipetally and progressively along the rachis (Fig. 2a, arrow) (Coleman and Greyson 1976; Dengler 1984). By late P4 (leaves are *1 mm in length), wild-type leaves typically have two prominent pairs of primary leaflets and 3 pairs of terminal leaflet lobes (Fig. 2b, asterisks). At this stage, the first pair of intercalary leaflets initiate between the first two primary leaflet pairs (Fig. 2b, arrow). Between mid-P5 and early-P6, the third pair of primary leaflets initiate (Fig. 2c, primary leaflets are outlined for clarity). At the P6/P7 stage of leaf morphogenesis (leaves are *1 cm in length), lobes of the terminal leaflet are well developed, all three primary leaflet pairs are present with the first pair forming leaflet lobes (Fig. 1c, arrowhead). Based on our wild-type leaf observations, two major developmental axes are present (Fig. 1b). The first axis (18) is along the rachis of the leaf (Fig. 1b, red box). The primary leaflets (PL) and intercalary leaflets (IL) form basipetally and sequentially (green arrow) along this axis. The second axis (28) is along the primary leaflet (Fig. 1b, yellow box) where secondary leaflets (SL) arise. Secondary leaflets develop along the petiolule of the primary leaflet. Lobes form acropetally along the leaflets (Fig. 1b, blue arrows); some of these results were also observed by Kessler et al. (2001). Since the mutants in our study have alterations along the ra-

713 Fig. 1. Leaf morphology in wild-type tomato. (a) Mature leaf 5 of wild-type cultivar ‘Ailsa Craig’ (AC). Wild-type tomato is bipinnately compound with three primary leaflets (PL) and secondary leaflets (SL). Intercalary leaflets (IL) are borne off the rachis between the primary leaflets. (b) Growth axes of wild type tomato leaves. Two axes of development are present in wild-type leaves: (1) The primary axis (red box, 18) along the rachis of the leaf. Primary leaflets and intercalary leaflets are formed basipetally and sequentially (green arrow) along this axis; and (2) The secondary axis (yellow box, 28) that occurs along the petiolule of the primary leaflets. Secondary leaflets (SL) form along this axis. Leaflet lobes initiate acropetally (blue arrows). (c) Young leaf. Two primary leaflet pairs are present and the third primary leaflet pair is initiating (PL arrow). The first intercalary leaflet (IL) pair is also initiating. Lobes of the primary leaflets are visible (arrowhead). (d) Intermediate leaf. All 3 primary leaflet pairs (PL) are well formed and intercalary leaflets (IL) arepresent. No secondary leaflets are formed at this stage. Scale bars = 5 cm (a); 0.5 cm (c); 1 cm (d). TL, terminal leaflet.

chis (18 axis), we focused our attention to the development of leaflets along this axis. Mature mutant leaf morphology Alterations in leaf complexity occur when there are changes (increase or decrease) in the depth of leaf lobes/serrations and (or) changes (increase or decrease) in leaf pinnation (leaflet orders) (Kessler et al. 2001). To determine whether pinnation and (or) lobing in the tomato leaf mutants differs from wild-type, mature leaf phenotypes were qualitatively assessed. Table 1 summarizes the number of primary Published by NRC Research Press

714



Fig. 2. Development of leaf compounding in wild-type tomato. Scanning electron microscopy (SEM) of young shoots. Leaf 5 is at the P3 stage of leaf development. (a) Primordium 1 (P1) through to early-primordium 4 (P4) is shown. The first primary leaflet pair (PL) and the second leaflet pair (arrow) is present by early P4. (b) SEM of an older shoot. A mid-primordium 3 (P3) and late-primordium 4 (P4) are shown. By late-P4, the first two primary leaflet pairs are formed, three lobes (asterisks) of the terminal leaflet are present, and the first intercalary leaflet is initiating (arrow). (c) SEM of a mid-P5 leaf. Leaf outline (white) is traced for clarity. The three primary leaflets (PL) are formed and first intercalary leaflet (IL) is visible. Scale bars = 100 mm.

Table 1. Summary of morphological features in tomato leaf mutants. Mature leaf

Primary morphogenesis

Pinnation Genotype AC La c sf tf2 bip clau

Primary leaflet 3 (0.14) – 2.6 (0.16) 7 (0.28) 2 (0.22) 3.8 (0.25) 2.8 (0.14)

Intercalary leaflet 3 (0.27) – – 5.6 (0.35) – 3.8 (0.31) 2.4 (0.17)

Lobes

P3

P4

Depth N/A – ; – : : :

Leaflet primordia + – – – + + +

Lobe primordia + – – – + + +

P5+ Leaflet primordia + – + – + + +

Leaflet primordia + – + + + + +

Note: +, present; –, absent; :, increase; ;, decrease. Numbers represent means where N = 20. Standard errors (±) are shown in parentheses. AC shown as (N/A, not applicable), since leaf lobing (: or ; in lobe depth) of the mutants are compared with wild-type leaves.

leaflets, intercalary leaflets, and lobe depth in mature leaves across all genotypes. The leaves of the Lanceolate (La) mutant are ovate in shape and have one single blade unit (lacks pinnation) with no marginal lobes/serrations (Fig. 3, La). The leaves of the potato leaf (c) mutant are unipinnately compound with a terminal leaflet, which has one or two serrations, and two pairs of primary leaflets with smooth margins (Fig. 3c). No intercalary leaflets are formed in c mutants. The leaves of solanifolia (sf) are unipinnately compound with a terminal leaflet and 5–7 pairs of primary leaflets (Fig. 3, sf). Pairs or single alternate intercalary leaflets form along the rachis and leaflet margins are completely smooth. The leaves of trifoliate2 (tf2) are unipinnately compound with a terminal leaflet and a single pair of primary leaflets (Fig. 3, tf2). The leaflets have a single pair of deep lobes. The leaves of bipinnata (bip) are bipinnately compound with a terminal leaflet, 3–4 pairs of primary leaflets, and secondary leaflets arising from the petiolule of the primary leaflets (Fig. 3, bip). Intercalary leaflets are also pinnatified with secondary leaflets. The leaflets of bip leaves are deeply lobed. The leaves of clausa (clau) are unipinnately compound with 2–3 pairs of primary leaflets and leaflets deeply lobed (Fig. 3, clau). Intercalary leaflets often lack a

petiolule and blade tissue is fused directly to the rachis of the leaf (Fig. 3, clau arrow). Primary leaf morphogenesis of mutant leaves In our study, we analyzed primary morphogenesis, since leaf shape is evident by this stage of development. To determine when leaf complexity (pinnation and lobing) is altered in the mutants during primary morphogenesis, SEM images of young shoots were analyzed. The leaves of the Lanceolate mutant do not form leaflets, and blade outgrowth (recognized externally by the presence of trichome differentiation) is evident by the P3/P4 stage of leaf development (Fig. 4, La). In smooth margined leaf mutants (c, sf), primary leaflet initiation is delayed along the rachis compared with wild-type leaves. Leaflet initiation in c and sf is first evident at the late-P4 (c) or late-P5/P6 stage (sf) of leaf development (Fig. 4; Table 1). Our observation of sf is similar to that observed by Chandra Sekhar and Sawhney (1990). In mutants with decreased pinnation and deep lobes (tf2, clau) or increased pinnation and deep lobes (bip), the timing of primary leaflet initiation is similar to wild type, in that the first pair of primary leaflets is initiated by P3 (Fig. 4; Published by NRC Research Press


Kang and Sinha Fig. 3. Morphology of mature leaves of mutant tomato species. The simple-leafed mutant Lanceolate (La). The leaf of potato leaf (c) is unipinnately compound with two primary leaflet pairs, no intercalary leaflets, and is smooth margined. The leaf of solanifolia (sf) is unipinnately compound with up to 5–7 primary leaflet pairs and intercalary leaflets. Leaf margins of sf are smooth. The leaves of trifoliate2 (tf2) mutants are unipinnately compound, with only one primary leaflet pair and deep lobes. The leaves of bipinnata (bip) mutants are bipinnately compound with three to four primary leaflet pairs, secondary leaflets (arrows), and are deeply lobed. The leaves of clausa (clau) mutants are unipinnately compound with two primary leaflet pairs. Intercalary leaflets are often fused along the rachis (arrow). IL, intercalary leaflet; PL, primary leaflet, SL, secondary leaflet; TL, terminal leaflet. Scale bars = 1 cm.

Table 1). By the P4 stage of leaf development, leaflet and lobe initiation was found to be either more advanced (tf2, bip) or similar (clau) to wild-type leaves (Fig. 4; Table 1). For instance, in tf2 mutant leaves, the first pair of primary leaflets is well formed, and lobes of the terminal and primary leaflets are initiating (Fig. 4, tf2, arrowheads). Similarly, bip mutant leaves have three primary leaflet pairs (two primary leaflet pairs in wild type at a similar stage), one intercalary leaflet pair, and a terminal leaflet lobe pair present (Fig. 4, bip, asterisks and arrowhead). In clau leaves, two primary leaflet pairs and one terminal leaflet lobe pair is evident by P4 (Fig. 4, clau, asterisks and arrowhead). Overall, we did not find evidence of earlier leaflet initiation in our mutants, but found that primary leaflet initiation along the rachis of the leaf is delayed in smooth margined leaf mutants (c, sf), while primary leaflet initiation is similar (clau) or more advanced (bip, tf2) in leaves with deep lobes (tf2, bip, clau), compared with wild-type leaves at comparable stages. This suggests that the timing of the morphogenetic program controlling primary leaflet initiation along the rachis is altered to increase or decrease pinnation. Primary leaflet initiation Since the timing of primary leaflet initiation along the rachis is altered in the leaf mutants, wild type and mutant leaf phenotypes were observed at an intermediate stage to assess when the full set of primary leaflet pairs (complementary to

715

mature leaf phenotypes) are present. First, young wild type and mutant leaf phenotypes were assessed. At our young leaf stage of development (*45 DPG), based on our growth conditions, all three primary leaflet pairs and one intercalary leaflet pair are completely formed or have initiated along the rachis in wild-type leaves (Fig. 1c, arrow). The simple leaf morphology of La remains unchanged, and undergoes blade expansion and rachis elongation (Fig. 5, La young). In c and sf leaves, primary leaflet initiation is still delayed along the rachis; potato leaf (c) mutants have 1 pair of primary leaflets (Fig. 5c, young) while sf leaves have 2–3 primary leaflet pairs (Fig. 5, sf young). Conversely, in tf2, bip, and clau leaves, the full set of primary leaflet pairs are initiated and leaflet lobes present at this stage of development (Fig. 5, young). Timing of intercalary leaflet initiation is similar (bip) or delayed (clau) compared with wild-type leaves (one pair present) (Fig. 5). Thus, in young leaves, the full set of primary leaflet pairs (complementary to mature leaf phenotypes) is apparent in tf2, bip, and clau but still delayed in c and sf leaves. Owing to the temporal delay in primary leaflet initiation in c and sf mutants, leaves were assessed at a later intermediate stage (*75 DPG) to determine when the full set of primary and intercalary leaflets form along the rachis. In all genotypes, with the exception of sf, the full set of primary leaflet and intercalary leaflet pairs (complementary to mature leaf phenotypes) are present when the leaves are *2– 3 cm in length (this stage not shown). In sf leaves, the full set of primary leaflet pairs (*5–7 pairs) is not evident until the leaves are *7–13 cm in length (4–5 primary leaflet pairs shown here in a leaf that is *9 cm), when leaves are almost half the length of a mature leaf (Fig. 5; Table 2). Primary leaflet pairs of sf leaves first develop in close proximity to each other along the rachis (Fig. 5, sf intermediate). Intercalary leaflets develop later when the primary leaflets have distanced from each other as the rachis elongates (data not shown). Secondary leaflets have not yet initiated in wild type plants and bip mutants (Figs. 1d and 5, bip intermediate). Based on our assessment of leaflet initiation, several conclusions can be made: (i) the signal and (or) mechanism controlling primary leaflet initiation along the rachis is altered in some of the leaf mutants (Fig. 8). For instance, the developmental program for primary leaflet initiation along the rachis is restricted or protracted in tf2 and sf mutants, respectively, causing decreased or increased primary leaflet formation; (ii) the developmental timing of primary leaflet initiation is not coordinated temporally with secondary leaflet initiation since secondary leaflets are not evident, even at a late stage in leaf development, in wild type plants and bip mutants (Figs. 1d and 5, bip intermediate); and (iii) leaflet lobes form early in leaf development, but that lobe depth can be further modified later in leaf development (Figs. 1d and 5, bip and clau intermediate). Together, it is likely that developmental genes and (or) signals are differentially controlled to regulate the temporal and spatial regulation of primary and secondary leaflet initiation and leaf lobing. Leaf-shape factor Since leaf complexity (pinnation and (or) lobing) is spatially and temporally altered in the leaf mutants, a morphometric approach was used to determine whether leaf Published by NRC Research Press


716


Fig. 4. Scanning electron microscopy of young shoots of mutant tomato species. Primordium 3 (P3) (Leaf 5) and primordium 4 (P4) stages are shown. Lanceolate (La) is a simple-leafed mutant with smooth margins. In the smooth margined leaf mutants, potato leaf (c) and solanifolia (sf), leaflet initiation is delayed compared with the wild type, regardless of whether there is a decrease (c) or increase (sf) in primary leaflet number. In c mutants, the first pair of primary leaflets initiate at P4 (asterisk), while leaflets are not evident in P4 leaves of sf mutants. Although sf leaf mutants have an increased number of primary leaflets (Fig. 3 sf), the first pair of primary leaflets is not evident until P6 (not shown). By late P6/P7, the first two pairs of primary leaflets are present (Fig. 5 sf young). The leaves of trifoliate (tf2) are well formed by the P4 stage of development. The lobe of the terminal leaflet is evident (black arrowhead) and the lobes of the primary leaflets initiating (white arrowhead). Leaves of bipinnata (bip) mutants are more developmentally advanced than wild-type leaves, at least by the mid-P4 stage of leaf development; three primary leaflet pairs are present (asterisks) (versus 2 pairs in wild-type at a similar stage) and the first intercalary leaflet pair initiating (arrow). Leaves of the leaf mutant clau (clau), has a similar developmental pattern to wild-type leaves. By the P4 stage of leaf development, the first two pairs of primary leaflets (asterisks) are present. Lobes of the terminal leaflet is also evident (arrowhead). Scale bars = 100 mm.

complexity could be assessed quantitatively (Bylesjo¨ et al. 2008). We used LSF as a measure of leaf complexity (see Materials and methods). LSF and leaf area were calculated for young, intermediate, and mature leaves (Fig. 6). In young leaves, leaf area is not significantly different across genotypes, although leaf shape is variable (Figs. 1c, 5 (young stage), and 6a). LSF is greater in leaves with more complex shape (AC, tf2, bip, clau) with a shape factor of *3 (Fig. 6b); this likely reflects terminal leaflet lobes as well as the presence of primary leaflets (Figs. 1c and 5, tf2, bip, clau young). LSF is lower in La, c, and sf, reflecting the prominent simple ovoid shape of the terminal leaflet and small (negligible) primary leaflets (Figs. 5 (La, sf, c young) and 6b). In intermediate leaves, the leaf area is significantly greater in wild-type leaves compared with the mutants (Fig. 6c). The large leaf area in wild-type leaves likely reflects the expanding terminal and primary leaflets (Fig. 1d). Among the mutants, leaf mutants with lobes (tf2, bip, clau) have larger leaf areas compared with smooth margined leaf mutants (La, c, sf) (Fig. 6c). While LSF does not change between the younger and intermediate stage in wild-type, bip, and clau leaves (LSF remains at *3), LSF increases in c and sf, reflecting the late initiation of primary leaflets

(Figs. 5c and 6d (sf intermediate)). There is no change in LSF (*1.5) for La, since the morphology (ovoid leaf shape) does not change (the rachis is negligible in overall LSF) (Fig. 6d). In mature leaves, overall leaf area varies considerably among the genotypes. Leaf area is greatest in wild-type, sf, and bip leaves (Fig. 6e) reflecting increased pinnation and (or) lobing, as well as overall greater leaf length (blade/rachis) (Figs. 1a and 3, sf and bip; Table 2). Leaf area is lower in c, tf2, and clau mutants (little change in La) reflecting their simpler morphology (decreased pinnation) and shorter leaf length (Fig. 3, La, c, tf2, clau; Table 2). Interestingly, LSF reflects final mature leaf morphology. For instance, there is little change in LSF for La and tf2 (*1.5 and *3, respectively) at any developmental time point (young, intermediate, mature) (Fig. 6f) since the mature morphological form is evident very early in leaf development (Fig. 4 La and tf2). LSF increases for wild-type, c, sf, and clau leaves reflecting the increase in primary, secondary and (or) intercalary leaflet number (AC, c and sf) or increase in lobe depth (clau) at later stages of development (Fig. 6f). LSF for bip leaves increases to *5 in mature leaves reflecting the increase in leaflets (primary, secondary, and intercalary) and lobe depth (Fig. 6f). While the overall graphical pattern Published by NRC Research Press

Kang and Sinha

717


Fig. 5. Development of primary leaflets in young [Primordia 6/7 (P6/P7); *45 DPG shown] and intermediate stage leaves (*75 DPG). Primary leaflet pairs for all the genotypes are present when leaves are *2–3 cm in length (not shown). Lanceolate (La) is a simple leafed mutant with no leaflets. All primary leaflet pairs are present in intermediate stage leaves but intercalary leaflets are not present in potato leaf (c) and solanifolia (sf) leaf mutants. Final leaf morphology (comparable with the mature leaf phenotype) of trifoliate (tf2) leaves is evident early in leaf development. Secondary leaflets of bipinnata (bip) leaves form later in development (not present in intermediate stage leaves). Lobes are well formed by the intermediate stage of leaf development in clausa (clau) leaf mutants. Abbreviations: IL, intercalary leaflets; PL, primary leaflet; TL, terminal leaflet. Scale bars = 0.5 cm (Young); 1 cm (Intermediate).

of leaf area and LSF in mature leaves is similar (Figs. 6e and 6f), it is clear, as indicated by the clau mutant (low leaf area, high LSF), that leaf area alone cannot be used as an indicator of leaf complexity. Thus, based on our results, LSF is a reliable method along with other parameters of leaf shape quantification (Bylesjo¨ et al. 2008; Weight et al. 2008), to quantify temporal changes in leaf complexity. Expression of LeT6 and the marginal blastozone Immunolocalization of LeT6 was performed on young wild type and mutant shoots to further characterize the role of LeT6 during primary leaf morphogenesis. RT-PCR was conducted on young leaves of all genotypes to determine the presence or absence of LeT6 transcript, and serial histological sections were analyzed to ascertain whether regions of meristematic activity (high density of dividing cells) along the marginal blastozone correlated with LeT6 expression. In wild-type leaves, LeT6 is expressed in the SAM, developing leaf primordia (arrowhead), initiating leaflets

(arrow), and adaxial domain of developing leaflets (asterisks) (Fig. 7, AC). Expression of LeT6 is lower in the developing rachis once leaflets are initiated (Fig. 7, AC, line). In histological sections, based on the dark staining density of nuclei, high meristematic activity is localized to regions of initiating and developing leaflets (Fig. 7, AC TBO stain, arrow, asterisk, and PL). In mutant leaves with no leaflets (La) or delayed primary leaflet initiation (c, sf), LeT6 is expressed in the SAM and adaxial domain of the terminal leaflet prior to leaflet initiation (in the case of c and sf) (Fig. 7, La, c, sf, arrow and arrowheads). This expression pattern is unlike simple leafed species such as Arabidopsis and maize that do not express KNOX genes in leaf primordia (Jackson et al. 1994; Long et al. 1996b). In tf2, bip and clau leaves, LeT6 expression pattern is similar to wild-type; LeT6 expression is detected in the SAM, initiating leaf primordia (arrowheads), but is reduced along the rachis (lines) once leaflets initiate (arrows) (Fig. 7, tf2, bip, clau). RT-PCR confirmed the presence of Published by NRC Research Press


718


Fig. 6. Morphometric analysis of leaf complexity in wild-type and mutant genotypes. Data was collected from young (a, b), intermediate (c, d), and mature leaves (e, f). (a, c, e) Leaf area (cm2) of young leaves (a), intermediate leaves (c), and mature leaves (e). (b, d, f). Leaf shape factor of young leaves (b), intermediate leaves (d), and mature leaves (f). Leaf complexity or leaf shape factor (LSF) was calculated as a measurement of circularity [circularity = 4p(areaperimeter–2)], where a circularity value of 1.0 represents a perfect circle, while a value approaching 0.0 represents an increasingly elongated shape (ImageJ, National Institute for Health). The circularity value was then calculated as –ln, so that a LSF value ranging from 1 (representing a perfect circle) to ? (representing increased complexity) could be obtained. Thus, the higher the –ln(LSF) value, the more complex the leaf shape. (b) Young LSF. LSF value is not significantly different between genotypes (compared with wild type) except La, c, and sf. (d) Intermediate LSF. LSF increases in c and sf leaves; Leaves of these genotypes show delayed primary leaflet initiation. (f) Increase in LSF is seen in AC, sf, and bip; Leaves of these genotypes have secondary leaflets (AC, bip) or increased primary leaflet pairs (sf). A total of 10 leaves (N = 10) were measured for each genotype. A one-way ANOVA followed by Dunnett’s tests (P = £ 0.001) was performed. Based on the Dunnett’s test, different letters indicate significant differences when compared with wild type.


Kang and Sinha

719 Table 2. Summary of leaf lengths in wild-type and mutant leaves.


Genotype AC La c sf tf2 bip clau

Mature leaf length 25 (0.51) 11 (0.58) 19 (0.81) 29 (0.81) 21 (0.93) 30 (0.89) 17 (0.6)

Leaf length* 7.5 (0.68) N/A 6.0 (0.52) 11.7 (0.54) 2.1 (0.26) 7.9 (0.38) 6.9 (0.42)

Note: Leaf length in centimetres; numbers represent means where N = 20; standard errors (±) are shown in parentheses; N/A, not applicable. *Average leaf length when full complementary set of primary and intercalary leaflet pairs (as seen in mature leaves) are present.

Fig. 7. Expression pattern of LeT6 and histological analysis of leaflets and lobes in wild type and mutant shoots. Immunolocalization of LeT6 protein in wild type and leaf-mutant shoots (*25 d post-germination). LeT6 is expressed in the SAM, young leaf primordia (arrowheads), and in developing leaflets and lobes (arrows and asterisks) . LeT6 expression is reduced in the developing rachis (black lines in wild type, tf2, and clau). RT-PCR of LeT6 expression reveals LeT6 RNA transcript in young leaves for all the tomato leaf mutants. Shown is RT-PCR at 40 cycles. Control for RT-PCR is the constitutively expressing gene Actin. Histological serial sections stained with TBO of wild type (AC) and mutant tomato (sf and tf2) P4 shoots. Low magnification showing a well developed primary leaflet (PL) and initiating primary leaflet (AC, asterisk). High meristematic activity is present in developing primary leaflets of a mid-P3 terminal leaflet (AC, arrow). Initiation of primary leaflets (PL) is extremely delayed in solanifolia (sf) compared with wild-type leaves. In P4 leaves, only two leaflets (PL) have initiated. Much like the c leaf (not shown), meristematic activity is high throughout the terminal leaflet (TL) and initiating leaflets. Leaflets of trifoliate2 (tf2) develop early (compared with wild type) and are well formed by the P4 stage of development. The terminal leaflet (TL) and primary leaflet (PL) show high meristematic activity while the rachis (black line) shows reduced meristematic activity. Scale bars = 10 mm.



720

LeT6 transcript in wild-type and all the mutants (Fig. 7). To determine whether LeT6 expression correlates with sites of high meristematic activity, paraffin sections were stained with TBO, a cationic dye that binds nucleic acids (Sharma and Sharma 1999). Thus, TBO was used as an indicator of small, dense cytoplasmic cells of high metabolic (meristematic) activity. Analysis of histological sections stained with TBO showed that in simple (La) or smooth margined leaves (c, sf) (sf shown here; Fig. 7, TBO), high meristematic activity (as indicated by small, high densely staining cytoplasmic cells) is present along the entire developing (terminal) leaflet, suggesting that prior to leaflet initiation, the marginal blastozone maintains morphogenetic competence. Once lobes and leaflets are initiated, meristematic activity of the marginal blastozone is restricted to those structures and down regulated along the rachis (tf2 shown here; Fig. 7, tf2, black lines). Overall, LeT6 expression is present in both simple and complex tomato leaf mutants. The fact that the observed histological patterns of meristematic activity of the marginal blastozone observed here correlates with the expression pattern domains of LeT6 (Fig. 7, immunolocalizations) lend support for the role of LeT6 (and possibly other class I KNOX genes) in maintaining morphogenetic and meristematic competence along the length of the leaf prior to leaflet initiation and maintains expression in developing leaflets in compound leaves.

Discussion This study was conducted to compare morphological and developmental differences between classic tomato mutants with simple and complex leaf phenotypes. We analyzed the degree of leaf complexity during primary morphogenesis, focusing on leaflet initiation along the rachis, to determine when during leaf development simple and complex leaf phenotypes diverge in tomato mutants. A morphometric analysis was also used to measure leaf complexity. Finally, the expression of the class1 KNOX gene LeT6 was analyzed to assess the role of LeT6 in leaf complexity. Together, the aim of this study is to provide a more in depth characterization of tomato leaf mutants to further understand compound leaf development. Leaf morphogenesis is defined by three phases: leaf initiation, primary morphogenesis (establishment of the basic leaf form), and secondary morphogenesis (cell expansion phase) (Dengler and Tsukaya 2001). In this study, we focused on the primary morphogenesis phase of leaf development, since formative events in leaf shape generation are usually evident at this stage in tomato. In simple (La) and simplified (tf2) (reduced pinnation) leaves, final leaf morphology is evident early in development (Fig. 4, La, tf2). It is known that elevated expression of LA leads to precocious differentiation along the blade margins causing a simple leaf phenotype (Ori et al. 2007). Although the tf2 gene has yet to be identified, it is possible that this gene may also play a similar role of regulating morphogenetic competence along the rachis during leaf development. In smooth margined leaf mutants such as c and sf, primary leaflet initiation is delayed (Fig. 4c, sf). In addition, secondary leaflet initiation is delayed in both wild-type and bip mutants relative to primary leaflet initiation (Figs. 1d and 5, bip intermediate). Thus, fi-


nal leaf morphology of these genotypes, particularly sf, is not evident until a very late stage of leaf development. The protraction of the developmental and spatial time course along the primary and secondary axis of the leaf suggests that this characteristic is a commonality found in compound leaves when leaf pinnation is altered. Based on our study of the developmental time course of primary leaflet formation, we suggest that the zone of morphogenetic competency along the rachis can be restricted or expanded to regulate the number of primary and intercalary leaflets formed along the rachis (Fig. 8). GOBLET, a gene recently identified as a NAC-domain transcription factor, may play such a role, since expression of GOB was found to be delimited to regions where leaflets initiate (Berger et al. 2009). Additionally, the MYB transcription factor PHANTASTICA may also play a role in regulating the spatial placement of leaflets (Kim et al. 2003a). Expression of the PHAN gene has been shown to set up the adaxial domain along the rachis and that the extent of PHAN expression correlates with final leaf morphology (Kim et al. 2003a). For instance, expression of PHAN along the entire length of the rachis correlates with a pinnately compound leaf (such as AC), while restriction of PHAN expression to the distal domain results in a peltately palmate compound leaf (reminiscent of a mutant such as tf2) (Kim et al. 2003a). Thus, PHAN may play a critical role in regulating both the spatial domain of primary leaflet position as well as regulating the timing of primary leaflet initiation along the rachis, as observed in our tomato leaf mutants (Fig. 8). In compound leafed species such as tomato, LeT6 is expressed in the SAM, downregulated in the site of leaf initiation, and then upregulated in leaf primordia (Hareven et al. 1996; Long et al. 1996b; Janssen et al. 1998a). In our study, we found that based on immunolocalizations, LeT6 is present in the SAM, young leaf primordia, and in developing leaflets in all the mutants, including the simple-leafed mutant La, in a pattern similar to wild-type leaves (Fig. 7). This is contrary to the data from species with simple-leafed species such as the dicot Arabidopsis and monocot maize, where KNOX genes are never present in incipient leaf primordia and developing leaves (Jackson et al. 1994; Lincoln et al. 1994; Long et al. 1996b). The presence of LeT6 in both simple and complex tomato leaves, and the fact that our histological observations (Fig. 7) overlap with our LeT6 expression patterns, suggests that LeT6 functions to maintain morphogenetic competence along the leaf margins in the tomato leaf as previously observed (Kim et al. 2003b; Ori et al. 2007; Barkoulas et al. 2008) (Fig. 8). For instance, when LeT6 is overexpressed, delayed differentiation in the leaf occurs such that the morphogenetic program is altered to produce higher order leaflets (Janssen et al. 1998a). Similarly, maturation of leaf primordia is also delayed in 35S::kn1 La2/- leaves (simple leaf phenotype) confirming the role of LeT6 (and La) in regulating the window of morphogenetic competence during leaf lobing and leaflet initiation (Ori et al. 2007). The fact that Lanceolate leaf mutants have precocious differentiation (Ori et al. 2007) suggests that this gene acts very early in leaf morphogenesis and that precocious differentiation would allow cessation of leaflet formation along the marginal region of the leaf leading to a simple leaf (Ori et al. 2007) (Fig. 8). The fact that we observed Published by NRC Research Press


Kang and Sinha

721

Fig. 8. Schematic diagram of genes that regulate leaf lobing and primary leaflet initiation along the rachis of a tomato leaf. A hypothetical zone of morphogenetic competency along the rachis where primary leaflets (open ovals) and intercalary leaflets (black circles) is present. During early stages of leaf development, both auxin and LeT6 are present along the leaf primordia. LeT6 is expressed along the entire length of the primordia (pink) regulating the morphogenetic competency along the leaf margin while auxin is present in punctuated gradients (green dots). These sites of auxin maxima are where leaflets will initiate. The action of the tomato gene ENTIRE functions in opposition to auxin to repress growth between leaflets and lobes. If the punctate pattern of auxin maxima is disrupted, smooth margined blades develop. The zone of morphogenetic competency can also be regulated through genes such as PHAN (orange) that is expressed specifically along the adaxial domain of the leaf. This domain (downward pointing time arrow) can be restricted as in the case of the tf2 leaf mutant, where final leaf morphology is evident early in leaf development and only one pair of leaflet primordia forms. Alternatively, if this zone is expanded (protracted temporally), as in the case of the sf leaf mutant, final leaf morphology is not evident until a very late stage of leaf development producing an increased number of leaflets. Primary leaflets of the sf leaf form early and develop in close proximity to each other. It is through rachis elongation (downward time arrow) that intercalary leaflets initiate later in leaf development as the primary leaflets separate in distance. It is known that the GOBLET (GOB) gene is required to maintain the boundary between primary leaflets (Berger et al. 2009). Thus, it is possible GOB activity may regulate leaflet initiation by acting on SF or TF2. In addition, Lanceolate (La) regulates temporal differentiation along the leaf primordia where precocious cell differentiation inhibits leaflet initiation leading to simple, smooth margined leaves (Ori et al. 2007). Thus, La is required both temporally and in a dosage-dependent manner to regulate leaflet (and thus lobing) initiation possibly causing the reduced lobing and smooth-margined lobes of tf2 and sf, respectively. Leaflet lobes are not shown and only one half of the primary leaflet pairs are shown for clarity.

LeT6 expression in Lanceolate and in all the other complex tomato mutant leaf primordia suggests that LeT6 not only plays an early role in regulating morphogenetic competency in tomato leaves, but that the action of genes such as Lanceolate may regulate blade outgrowth phenomena by regulating cell division activity after morphogenetic competency is established. Finally, it is possible that mutations that make simplified tomato leaves likely encode gene products that are downstream of LeT6 expression in the primordium during compound leaf development. KNOX1 genes have been studied in both simple and compound leaves and some of their functions are well known. In simple leaves such as maize and Arabidopsis, expression of KNOX1 genes is restricted to the SAM where they play a role in the production and maintenance of the stem cell niche and are downregulated in incipient leaf primordia (Sinha et al.1993; Jackson et al. 1994; Chuck et al. 1996; Long et al. 1996a). Overexpression of KNOX1 genes in compound leaves, however, leads to increased ramification of the leaf form (Chen et al. 1997; Parnis et al. 1997; Janssen et al. 1998a). The highly dissected leaf morphology of these plants suggests that expression of KNOX1 genes is involved in leaf complexity (Shani et al. 2009). It has been shown in simple leafed species that areas of the leaf margin, particularly leaf serrations, may be sites of meristematic activity since overexpression of KNAT6 (as well as KNAT1) revealed leaves with various degrees of lobing (Chuck et al. 1996; Dean et al. 2004). Genes such as TKN1 (Arabidopsis BP orthologue) play a role in leaf compounding and (or)

lobing through an unknown mechanism. In Arabidopsis, BP is known to promote cell divisions (among other functions) in various organs (Chuck et al. 1996; Douglas et al. 2002; Venglat et al. 2002), as well as promoting leaf serrations when overexpressed in saw1 saw2 double mutants (Kumar et al. 2007). More recently, it has been shown that the highly complex leaf phenotype of the bip mutant is a result of a loss of function of the BEL-like homeodomain gene, BIP, shown to be the tomato SAW ortholog (Kimura et al. 2008). Based on these reports, as well as our results of LeT6 expression in the tomato mutants, we suggest that other class I KNOX genes may also play a role in maintaining meristematic activity along the main rachis and (or) leaf margins. The existence of the leaf marginal blastozone and its role in organogenesis has been a subject of debate (Kaplan 1992; Hagemann and Gleissberg 1996). In simple leaves with smooth entire margins such as in Arabidopsis, the marginal blastozone is present for only a short period of time with cell divisions ceasing very early in leaf developing (Donnelly et al. 1999). The onset of early differentiation in this tissue suggests a limited period in which further dissection can occur (Donnelly et al. 1999). In lobed and compound leaves, however, suppression and enhancement in localized areas of the leaf results in a diversity of shapes in both dicots and monocots (Hagemann and Gleissberg 1996; Gunawardena and Dengler 2006). In compound and lobed leaves, it has been shown that prolonged activity of the marginal blastozone results in leaf dissection or leaf lobing (HagePublished by NRC Research Press


722

mann and Gleissberg 1996; Dolan and Poethig 1998; Kim et al. 2003a; Barth et al. 2009). Based on our histological observations (of dense nuclei staining), we showed that high meristematic (cell division) activity along the leaf margins mark the site of the marginal blastozone and is associated with sites of lobe and leaflet formation (Fig. 7). Furthermore, regions of high meristematic activity are localized to the lobe and (or) leaflets once they are initiated suggesting that these populations of cells are capable of forming new and (or) maintaining the integrity of existing lobes and developing leaflets. Our histological observations also corroborate with Histone H4 (a cell division marker) expression patterns in developing leaflets in the compound leafed species Cardamine hirsuta (Barkoulas et al. 2008). It is still largely unknown how leaflets are initiated and the leaf margin is elaborated during compound leaf blade outgrowth. It is likely that multiple factors, including competence of the marginal domain as well as the phytohormone auxin, play a critical role. The polar transport of PIN1, the auxin efflux carrier, has been shown to play numerous roles during plant development, including leaf initiation (Reinhardt et al. 2003; Scarpella et al. 2006; Bayer et al. 2009; Koenig et al. 2009) (Fig. 8). Additionally, marginal outgrowth is repressed by the action of the gene ENTIRE, a member of the AUX/IAA gene family (Wang et al. 2005; Zhang et al. 2007; Koenig et al. 2009). It has also been shown that PIN1 directed auxin activity is localized to sites of leaflet and lobe formation to regulate cell division activity (Barkoulas et al. 2008). It will be interesting to address the precise role of cell proliferation during compound leaf formation, since the tight spatial and temporal control of cell proliferation activity during leaf morphogenesis suggests that, at some cellular level, leaf compounding may be controlled through this mechanism. In conclusion, based on our study, we found that leaflet initiation is temporally and spatially altered along the rachis (primary axis) as well as along the petiolule (secondary axis) to regulate leaf complexity. Thus, our study points to further analysis of some specific mutants/genes and areas of future inquiry that will help us better understand compound leaf morphogenesis.

Acknowledgements The authors thank members of the Sinha Lab for critical reading of the manuscript and the Tomato Genetics Resource Center (TGRC) (University of California, Davis) for seed stocks. This work was supported by The Natural Sciences and Engineering Research Council of Canada (NSERC) and The Katherine Esau Fellowship (University of California, Davis) to J. Kang and National Science Foundation Grants (NSF) IOS 0641696 to N. Sinha.

References Avivi, Y., Lev-Yadun, S., Morozova, N., Libs, L., Williams, L., Zhao, J., Varghese, G., and Grafi, G. 2000. Clausa, a tomato mutant with a wide range of phenotypic perturbations, displays a cell type-dependent expression of the homeobox gene LeT6/ TKn2. Plant Physiol. 124(2): 541–552. doi:10.1104/pp.124.2. 541. PMID:11027705. Barkoulas, M., Hay, A., Kougioumoutzi, E., and Tsiantis, M. 2008. A developmental framework for dissected leaf formation in the

Botany Vol. 88, 2010 Arabidopsis relative Cardamine hirsuta. Nat. Genet. 40(9): 1136–1141. doi:10.1038/ng.189. PMID:19165928. Barth, S., Geier, T., Eimert, K., Watillon, B., Sangwan, R.S., and Gleissberg, S. 2009. KNOX overexpression in transgenic Kohleria (Gesneriaceae) prolongs the activity of proximal leaf blastozones and drastically alters segment fate. Planta, 230(6): 1081–1091. doi:10.1007/s00425-009-0997-0. PMID:19685246. Bayer, E.M., Smith, R.S., Mandel, T., Nakayama, N., Sauer, M., Prusinkiewicz, P., and Kuhlemeier, C. 2009. Integration of transportbased models for phyllotaxis and midvein formation. Genes Dev. 23(3): 373–384. doi:10.1101/gad.497009. PMID: 19204121. Berger, Y., Harpaz-Saad, S., Brand, A., Melnik, H., Sirding, N., Alvarez, J.P., Zinder, M., Samach, A., Eshed, Y., and Ori, N. 2009. The NAC-domain transcription factor GOBLET specifies leaflet boundaries in compound tomato leaves. Development, 136(5): 823–832. doi:10.1242/dev.031625. PMID:19176589. Bharathan, G., Goliber, T.E., Moore, C., Kessler, S., Pham, T., and Sinha, N.R. 2002. Homologies in leaf form inferred from KNOXI gene expression during development. Science, 296(5574): 1858–1860. doi:10.1126/science.1070343. PMID: 12052958. Blein, T., Pulido, A., Vialette-Guiraud, A., Nikovics, K., Morin, H., Hay, A., Johansen, I.E., Tsiantis, M., and Laufs, P. 2008. A conserved molecular framework for compound leaf development. Science, 322(5909): 1835–1839. doi:10.1126/science.1166168. PMID:19095941. Brand, A., Shirding, N., Shleizer, S., and Ori, N. 2007. Meristem maintenance and compound-leaf patterning utilize common genetic mechanisms in tomato. Planta, 226(4): 941–951. doi:10. 1007/s00425-007-0540-0. PMID:17520278. Bylesjo¨, M., Segura, V., Soolanayakanahally, R.Y., Rae, A.M., Trygg, J., Gustafsson, P., Jansson, S., and Street, N.R. 2008. LAMINA: a tool for rapid quantification of leaf size and shape parameters. BMC Plant Biol. 8(1): 82. doi:10.1186/1471-2229-882. PMID:18647399. Champagne, C.E., Goliber, T.E., Wojciechowski, M.F., Mei, R.W., Townsley, B.T., Wang, K., Paz, M.M., Geeta, R., and Sinha, N.R. 2007. Compound leaf development and evolution in the legumes. Plant Cell, 19(11): 3369–3378. doi:10.1105/tpc.107. 052886. PMID:17993625. Chandra Sekhar, K.N., and Sawhney, V.K. 1990. Leaf development in the normal and solanifolia mutant of tomato (Lycopersicon esculentum). Am. J. Bot. 77(1): 46–53. doi:10.2307/2444791. Chen, J.J., Janssen, B.J., Williams, A., and Sinha, N. 1997. A gene fusion at a homeobox locus: alterations in leaf shape and implications for morphological evolution. Plant Cell, 9(8): 1289– 1304. doi:10.1105/tpc.9.8.1289. PMID:9286107. Chuck, G., Lincoln, C., and Hake, S. 1996. KNAT1 induces lobed leaves with ectopic meristems when overexpressed in Arabidopsis. Plant Cell, 8(8): 1277–1289. doi:10.1105/tpc.8.8.1277. PMID:8776897. Coleman, W.K., and Greyson, R.I. 1976. The growth and development of the leaf in tomato (Lycopersicon esculentum). II. Leaf ontogeny. Can. J. Bot. 54(23): 2704–2717. doi:10.1139/b76-291. David-Schwartz, R., Koenig, D., and Sinha, N.R. 2009. LYRATE is a key regulator of leaflet initiation and lamina outgrowth in tomato. Plant Cell, 21(10): 3093–3104. doi:10.1105/tpc.109. 069948. PMID:19820188. Dean, G., Casson, S., and Lindsey, K. 2004. KNAT6 gene of Arabidopsis is expressed in roots and is required for correct lateral root formation. Plant Mol. Biol. 54(1): 71–84. doi:10.1023/ B:PLAN.0000028772.22892.2d. PMID:15159635. Dengler, N.G. 1984. Comparison of leaf development in normal (+/ Published by NRC Research Press


Kang and Sinha +), ENTIRE (e/e), and Lanceolate (La/+) plants of tomato, Lycopersicon esculentum ‘Ailsa Craig’. Bot. Gaz. 145(1): 66–77. doi:10.1086/337428. Dengler, N.G., and Tsukaya, H. 2001. Leaf morphogenesis in dicotyledons: current issues. Int. J. Plant Sci. 162(3): 459–464. doi:10.1086/320145. Dolan, L., and Poethig, R.S. 1998. Clonal analysis of leaf development in cotton. Am. J. Bot. 85(3): 315–321. doi:10.2307/ 2446322. Donnelly, P.M., Bonetta, D., Tsukaya, H., Dengler, R.E., and Dengler, N.G. 1999. Cell cycling and cell enlargement in developing leaves of Arabidopsis. Dev. Biol. 215(2): 407–419. doi:10.1006/ dbio.1999.9443. PMID:10545247. Douglas, S.J., Chuck, G., Dengler, R.E., Pelecanda, L., and Riggs, C.D. 2002. KNAT1 and ERECTA regulate inflorescence architecture in Arabidopsis. Plant Cell, 14(3): 547–558. doi:10.1105/tpc. 010391. PMID:11910003. Gunawardena, A.H.L.A.N., and Dengler, N.G. 2006. Alternative modes of leaf dissection in monocotyledons. Bot. J. Linn. Soc. 150(1): 24–44. doi:10.1111/j.1095-8339.2006.00487.x. Hagemann, W., and Gleissberg, S. 1996. Organogenetic capacity of leaves: the significance of marginal blastozones in angiosperms. Plant Syst. Evol. 199(3-4): 121–152. doi:10.1007/BF00984901. Hareven, D., Gutfinger, T., Parnis, A., Eshed, Y., and Lifschitz, E. 1996. The making of a compound leaf: genetic manipulation of leaf architecture in tomato. Cell, 84(5): 735–744. doi:10.1016/ S0092-8674(00)81051-X. PMID:8625411. Hay, A., and Tsiantis, M. 2006. The genetic basis for differences in leaf form between Arabidopsis thaliana and its wild relative Cardamine hirsuta. Nat. Genet. 38(8): 942–947. doi:10.1038/ ng1835. PMID:16823378. Jackson, D., Veit, B., and Hake, S. 1994. Expression of maize KNOTTED 1 related homeobox genes in the shoot apical meristem predicts patterns of morphogenesis in the vegetative shoot. Development, 120: 405–413. Janssen, B.J., Lund, L., and Sinha, N. 1998a. Overexpression of a homeobox gene, LeT6, reveals indeterminate features in the tomato compound leaf. Plant Physiol. 117(3): 771–786. doi:10. 1104/pp.117.3.771. PMID:9662520. Janssen, B.J., Williams, A., Chen, J.J., Mathern, J., Hake, S., and Sinha, N. 1998b. Isolation and characterization of two knottedlike homeobox genes from tomato. Plant Mol. Biol. 36(3): 417– 425. doi:10.1023/A:1005925508579. PMID:9484482. Jasinski, S., Kaur, H., Tattersall, A., and Tsiantis, M. 2007. Negative regulation of KNOX expression in tomato leaves. Planta, 226(5): 1255–1263. doi:10.1007/s00425-007-0572-5. PMID: 17628827. Jasinski, S., Tattersall, A., Piazza, P., Hay, A., Martinez-Garcia, J.F., Schmitz, G., Theres, K., McCormick, S., and Tsiantis, M. 2008. PROCERA encodes a DELLA protein that mediates control of dissected leaf form in tomato. Plant J. 56(4): 603–612. doi:10.1111/j.1365-313X.2008.03628.x. PMID:18643984. Kang, J., and Dengler, N. 2005. Leaf architecture — regulation of leaf position, shape, and internal structure. In Plant architecture and its manipulation. Edited by C. Turnbull. Blackwell Publishing, CRC Press, Oxford, UK. pp. 23–56. Kaplan, D.R. 1992. The relationship of cells to organisms in plants: problem and implications of an organismal perspective. Int. J. Plant Sci. 153(S3): S28–S37. doi:10.1086/297061. Kaplan, D. 2001. Fundamental concepts of leaf morphology and morphogenesis: a contribution to the interpretation of molecular genetic mutants. Int. J. Plant Sci. 162(3): 465–474. doi:10.1086/ 320135. Kessler, S., Kim, M., Pham, T., Weber, N., and Sinha, N. 2001.

723 Mutations altering leaf morphology in tomato. Int. J. Plant Sci. 162(3): 475–492. doi:10.1086/320136. Kim, M., McCormick, S., Timmermans, M., and Sinha, N. 2003a. The expression domain of PHANTASTICA determines leaflet placement in compound leaves. Nature, 424(6947): 438–443. doi:10.1038/nature01820. PMID:12879073. Kim, M., Pham, T., Hamidi, A., McCormick, S., Kuzoff, R.K., and Sinha, N. 2003b. Reduced leaf complexity in tomato wiry mutants suggests a role for PHAN and KNOX genes in generating compound leaves. Development, 130(18): 4405–4415. doi:10. 1242/dev.00655. PMID:12900456. Kimura, S., Koenig, D., Kang, J., Yoong, F.Y., and Sinha, N. 2008. Natural variation in leaf morphology results from mutation of a novel KNOX gene. Curr. Biol. 18(9): 672–677. doi:10.1016/j. cub.2008.04.008. PMID:18424140. Koenig, D., Bayer, E., Kang, J., Kuhlemeier, C., and Sinha, N. 2009. Auxin patterns Solanum lycopersicum leaf morphogenesis. Development, 136(17): 2997–3006. doi:10.1242/dev.033811. PMID:19666826. Kumar, R., Kushalappa, K., Godt, D., Pidkowich, M.S., Pastorelli, S., Hepworth, S.R., and Haughn, G.W. 2007. The Arabidopsis BEL1-LIKE HOMEODOMAIN proteins SAW1 and SAW2 act redundantly to regulate KNOX expression spatially in leaf margins. Plant Cell, 19(9): 2719–2735. doi:10.1105/tpc.106.048769. PMID:17873098. Lincoln, C., Long, J., Yamaguchi, J., Serikawa, K., and Hake, S. 1994. A knotted1-like homeobox gene in Arabidopsis is expressed in the vegetative meristem and dramatically alters leaf morphology when overexpressed in transgenic plants. Plant Cell, 6(12): 1859–1876. doi:10.1105/tpc.6.12.1859. PMID: 7866029. Long, J., McConnell, J., Fernandez, A., Grbic, V., and Barton, M. 1996a. Developmental genetics of shoot apical meristem formation in Arabidopsis. Plant Physiol. 111: 12. Long, J.A., Moan, E.I., Medford, J.I., and Barton, M.K. 1996b. A member of the KNOTTED class of homeodomain proteins encoded by the STM gene of Arabidopsis. Nature, 379(6560): 66– 69. doi:10.1038/379066a0. PMID:8538741. Mathan, D., and Jenkins, J. 1962. A morphogenetic study of Lanceolate, a leaf shape mutant in tomato. Am. J. Bot. 49(5): 504– 514. doi:10.2307/2439421. National Institute of Health. 2010. ImageJ: Image processing and analysis in java. Available from rsbweb.nih.gov/ij/index.html [accessed 29 October 2009]. Ori, N., Eshed, Y., Chuck, G., Bowman, J.L., and Hake, S. 2000. Mechanisms that control KNOX gene expression in the Arabidopsis shoot. Development, 127(24): 5523–5532. PMID: 11076771. Ori, N., Cohen, A.R., Etzioni, A., Brand, A., Yanai, O., Shleizer, S., Menda, N., Amsellem, Z., Efroni, I., Pekker, I., Alvarez, J.P., Blum, E., Zamir, D., and Eshed, Y. 2007. Regulation of LANCEOLATE by miR319 is required for compound-leaf development in tomato. Nat. Genet. 39(6): 787–791. doi:10.1038/ ng2036. PMID:17486095. Parnis, A., Cohen, O., Gutfinger, T., Hareven, D., Zamir, D., and Lifschitz, E. 1997. The dominant developmental mutants of tomato, Mouse-ear and Curl, are associated with distinct modes of abnormal transcriptional regulation of a Knotted gene. Plant Cell, 9(12): 2143–2158. doi:10.1105/tpc.9.12.2143. PMID: 9437860. Pautot, V., Dockx, J., Hamant, O., Kronenberger, J., Grandjean, O., Jublot, D., and Traas, J. 2001. KNAT2: evidence for a link between knotted-like genes and carpel development. Plant Cell, 13(8): 1719–1734. doi:10.1105/tpc.13.8.1719. PMID:11487688. Published by NRC Research Press


724 Reinhardt, D., Pesce, E.R., Stieger, P., Mandel, T., Baltensperger, K., Bennett, M., Traas, J., Friml, J., and Kuhlemeier, C. 2003. Regulation of phyllotaxis by polar auxin transport. Nature, 426(6964): 255–260. doi:10.1038/nature02081. PMID:14628043. Scarpella, E., Marcos, D., Friml, J., and Berleth, T. 2006. Control of leaf vascular patterning by polar auxin transport. Genes Dev. 20(8): 1015–1027. doi:10.1101/gad.1402406. PMID:16618807. Shani, E., Burko, Y., Ben-Yaakov, L., Berger, Y., Amsellem, Z., Goldshmidt, A., Sharon, E., and Ori, N. 2009. Stage-specific regulation of Solanum lycopersicum leaf maturation by class 1 KNOTTED1-LIKE HOMEOBOX proteins. Plant Cell, 21(10): 3078–3092. doi:10.1105/tpc.109.068148. PMID:19820191. Sharma, A.K., and Sharma, A. 1999. Plant chromosomes: analysis, manipulation and engineering. Harwood Academic Publishers, Chur, Switzerland. Sinha, N., Hake, S., and Freeling, M. 1993. Genetic and molecular analysis of leaf development. Curr. Top. Dev. Biol. 28: 47–80. doi:10.1016/S0070-2153(08)60209-4. PMID:8348842. Smith, L.G., Greene, B., Veit, B., and Hake, S. 1992. A dominant mutation in the maize homeobox gene, Knotted-1, causes its ectopic expression in leaf cells with altered fates. Development, 116(1): 21–30. PMID:1362381. Stettler, R.F. 1964. Dosage effects of the lanceolate gene in tomato. Am. J. Bot. 51(3): 253–264. doi:10.2307/2440294. Tam, J.P. 1988. Synthetic peptide vaccine design: synthesis and properties of a high-density multiple antigenic peptide system.

Botany Vol. 88, 2010 Proc. Natl. Acad. Sci. U.S.A. 85(15): 5409–5413. doi:10.1073/ pnas.85.15.5409. PMID:3399498. Venglat, S.P., Dumonceaux, T., Rozwadowski, K., Parnell, L., Babic, V., Keller, W., Martienssen, R., Selvaraj, G., and Datla, R. 2002. The homeobox gene BREVIPEDICELLUS is a key regulator of inflorescence architecture in Arabidopsis. Proc. Natl. Acad. Sci. U.S.A. 99(7): 4730–4735. doi:10.1073/pnas. 072626099. PMID:11917137. Vollbrecht, E., Reiser, L., and Hake, S. 2000. Shoot meristem size is dependent on inbred background and presence of the maize homeobox gene, knotted1. Development, 127(14): 3161–3172. PMID:10862752. Wang, H., Jones, B., Li, Z., Frasse, P., Delalande, C., Regad, F., Chaabouni, S., Latche´, A., Pech, J.C., and Bouzayen, M. 2005. The tomato Aux/IAA transcription factor IAA9 is involved in fruit development and leaf morphogenesis. Plant Cell, 17(10): 2676–2692. doi:10.1105/tpc.105.033415. PMID:16126837. Weight, C., Parnham, D., and Waites, R. 2008. LeafAnalyser: a computational method for rapid and large-scale analyses of leaf shape variation. Plant J. 53(3): 578–586. doi:10.1111/j.1365313X.2007.03330.x. PMID:18028263. Zhang, J., Chen, R., Xiao, J., Qian, C., Wang, T., Li, H., Ouyang, B., and Ye, Z. 2007. A single-base deletion mutation in SlIAA9 gene causes tomato (Solanum lycopersicum) entire mutant. J. Plant Res. 120(6): 671–678. doi:10.1007/s10265-007-0109-9. PMID:17955175.
