Plants have diversified their leaf morphologies to adapt to diverse ecological niches. The molecular components responsible for regulating leaf morphology, however, have not been fully elucidated. By screening Arabidopsis activation-tagging lines, we identified a dominant mutant, which we designated longifolia1-1D (lng1-1D). lng1-1D plants were characterized by long petioles, narrow but extremely long leaf blades with serrated margins, elongated floral organs, and elongated siliques. The elongated leaves of the mutant were due to increased polar cell elongation rather than increased cell proliferation. Molecular characterization revealed that this phenotype was caused by overexpression of the novel gene LNG1, which was found to have a homolog, LNG2,in Arabidopsis. To further examine the role of the LNG genes, we characterized lng1 and lng2 loss-of-function mutant lines. In contrast to the elongated leaves of lng1-1D plants,the lng1 and lng2 mutants showed slightly decreased leaf length. Furthermore, the lng1-3 lng2-1 double mutant showed further decreased leaf length associated with less longitudinal polar cell elongation. The leaf widths in lng1-3 lng2-1 mutant plants were similar to those in wild type, implying that the role of LNG1 and LNG2 on polar cell elongation is similar to that of ROTUNDIFOLIA3(ROT3). However, analysis of a lng1-3 lng2-1 rot3-1 triple mutant and of a lng1-1D rot3-1 double mutant indicated that LNG1 and LNG2 promote longitudinal cell elongation independently of ROT3. Taken together, these findings indicate that LNG1 and LNG2 are new components that regulate leaf morphology by positively promoting longitudinal polar cell elongation independently of ROT3 in Arabidopsis.

As sessile organisms, plants have had to adapt their growth and development to diverse habitats and ecological niches, in part by diversifying the shape,size, form and arrangement of their leaves. As leaf morphology is closely associated with the habitats and ecological niches in which a plant is able to grow, the potential benefits from genetically engineering leaf morphology are immense. In addition, the diversity in leaf morphology between closely related species or within a single species provides an excellent opportunity to investigate how a living form is shaped by the interplay of various genes, and how this interplay is modified during evolution.

Leaf morphogenesis is conceptually divided into three processes: leaf initiation, establishment of polarity and leaf expansion. During initiation,the leaf primordia are formed at the periphery of the shoot apical meristem(SAM). Polarity is established with regards to the three main axes: the proximodistal axis (also called the apicobasal axis, and herein referred to as the leaf-length direction); the dorsoventral axis (also called the adaxial-abaxial axis); and the left/right axis (also called the mediolateral axis, and herein referred to as the leaf-width direction). Subsequently, the leaves expand to their final shapes and sizes(Sinha, 1999; Bowman et al., 2002; Kim and Cho, 2006).

The final shape and size of an adult leaf is modulated by leaf expansion,which is in turn dependent upon cell division and expansion. The importance of cell proliferation on leaf morphology can be observed in Arabidopsistransgenic lines expressing cyclin-dependent kinase inhibitor 1(ICK1) or Kip-related protein 2 (KRP2), which inhibit leaf cell proliferation through interactions with CDKA1/cyclin complexes (Wang et al., 2000; Verkest et al., 2005). Overexpression of these genes has been associated with the development of small serrated leaves that have reduced cell numbers. However, although this suggests that general cell proliferation is crucial for the determination of leaf shape and size, changes in cell proliferation do not always result in altered leaf shapes. For example, overexpression of ABP1 results in decreased cell proliferation but increased cell volume without gross changes in leaf morphology (Jones et al.,1998). Furthermore, studies have shown that increased cell numbers could be compensated for by decreases in cell size, or vice versa (Hemerley et al., 1995; De Veylder et al.,2001; Ullah et al.,2001; Tsukaya,2003). In addition, other factors regulate leaf morphogenesis by affecting polarity-dependent cell proliferation. Recently, ROTUNDIFOLIA4 (ROT4), which encodes a membrane-bound small peptide, was shown to inhibit cell proliferation in only the leaf-length direction in Arabidopsis. Conversely, it was also shown that ectopic expression of ROT4 caused short leaves(Narita et al., 2004). Thus,both general and specific cell proliferation can affect leaf morphology.

Cell expansion also plays an important role in the leaf expansion required for the proper formation of a mature leaf. Genetic analyses have indicated that cell expansion is regulated independently in the leaf-length and leaf-width directions. Mutation of the ANGUSTIFOLIA (AN)gene, which encodes a C-terminal binding protein (CtBP) that is responsible for regulating the arrangement of microtubules in leaf cells, caused a defect in cell expansion in the leaf-width direction, resulting in narrower leaves(Kim et al., 2002). By contrast, the mutation of ROT3, which encodes a P450 involved in brassinosteroid (BR) biosynthesis, decreased cell expansion in the leaf-length direction, resulting in shorter leaves(Kim et al., 1998). This finding of an association between ROT3 and leaf expansion provided evidence that BR promotes cell expansion in the leaf-length direction(Kim et al., 1998; Kim et al., 1999; Kim et al., 2005). Similar findings were observed in other BR biosynthetic mutants, including de-etiolated 2 (det2) and dwarf4 (dwf4) -both of which show shortened leaves(Altmann, 1998; Fujioka et al., 1997; Li and Chory, 1997; Azpiroz et al., 1998; Choe et al., 1998). Other factors that affect cell expansion, including auxins and light, have also been found to regulate leaf morphology (Timpte et al., 1992; Qin et al.,1997; Franklin et al.,2003; Kozuka et al.,2005)

The different leaf morphologies observed in the various Arabidopsis ecotypes are likely to be determined by multiple factors. A genetic screen of leaf-shape mutants from 5770 M1 EMS-mutagenized Arabidopsis plants generated 94 complementation groups that showed altered leaf morphology (Berna et al.,1999). A quantitative trait loci (QTL) analysis using recombinant inbred Ler and Col-4 lines identified 21 loci that affect six different characteristics of the adult leaf(Perez-Perez et al., 2002). These previous studies suggest that leaf morphogenesis is a complex process requiring the interplay of many different genetic components. However,although many factors involved in leaf morphogenesis have been identified,other genetic components remain to be examined before the determination of leaf shape is fully understood. Here, we have screened Arabidopsisleaf-shape mutants in an effort to discover new genetic components and to report the identification of two new genetic components, LONGIFOLIA1(LNG1) and LONGIFOLIA2 (LNG2), that are responsible for regulating leaf morphology by promoting cell expansion in the leaf-length direction in a ROT3-independent fashion.

Plant materials and growth conditions

Arabidopsis thaliana plants (Col-0 ecotype background) were maintained in a greenhouse at 22-24°C with a 16 hour/8 hour light/dark cycle for general growth. Activation-tagged T-DNA insertion lines were generated by a dipping procedure (Clough and Bent, 1998), using Agrobacterium harboring the activation-tagging vector pSKI1015 (Weigel et al., 2000). The lng1-1D mutant plant was selected from herbicide (ammonium glufosinate, known as Basta)-resistant transgenic lines. Loss-of-function mutant alleles (lng1-2, lng1-3, lng2-1 and lng2-2) were obtained from the SALK Institute Genomic Analysis Laboratory (salk_107002, salk_135585, salk_067658, salk_034645 and salk_130006, respectively). For generation of the lng1 lng2 double homozygous mutant, homozygous lng1 plants were crossed with homozygous lng2 plants. Homozygous lng1 lng2 double mutants were identified by PCR analysis from the F2 population using LNG1-and LNG2-specific primers (LBa1, 5′-TGG TTC ACG TAG TGG GCC ATC G-3′ and LBb1: 5′-GCG TGG ACC GCT TGC TGC AAC T-3′).

Expression analysis

For northern analysis, total RNA was extracted from 3-week-old seedlings using the RNeasy plant mini kit (Qiagen) and 20 μg total RNA was resolved from each sample by 1.2% agarose gel electrophoresis. The separated RNAs were transferred to a nylon membrane (Hybond-N; Amersham), and the filters were prehybridized for 1-2 hours at 42°C in 50% formamide, 5×SSPE,5×Denhardt's solution, 0.1% SDS and 100 μg/ml denatured salmon sperm DNA. For hybridization, a 32P-dCTP-labeled probe (described below)was added to the prehybridization buffer and the filter was hybridized overnight. Washes were performed twice at room temperature for 10 minutes in 2×SSC and 0.1% SDS, once at 65°C for 15 minutes in 1×SSC and 0.1% SDS, and twice at 65°C for 15 minutes in 0.1×SSC and 0.1% SDS. Results were analyzed with a scanning densitometer (Pharmacia), with adjustments being made for small differences in RNA loading. The probes were generated by RT-PCR from total RNA (described below) using specific primers(At5g15580 F, 5′-TTG GTT GCC CAC TGA GCA TC-3′ and R, 5′-ACA AGT ACA TCT CAG AGA TA-3′; At5g15600 F, 5′-ATG GGG AAA GCT AGA GGA GTT-3′ and R, 5′-TCA TTT GTC AGA TCC ACA AGG-3′; At5g15610 F, 5′-CCT CTT CAA TCT ATA TAA TCT-3′ and R, 5′-CTG AAA TAT ACT GGG AGC CTT-3′). The amplified fragments were cloned into the pGEM-T easy vector.

For RT-PCR analysis, Arabidopsis plants were grown on 1/2 MS medium and total RNA was extracted from 23-day-old plants using the total RNA Isolation System (Qiagen). RT-PCR was performed with the Access RT-PCR System(Invitrogen), according to the manufacturer's protocol. Briefly, 5 μg tRNA was reverse-transcribed to cDNA, diluted tenfold and amplified with the following specific primers: LNG1 F, 5′-ATG TCG GCG AAG CTT TTG TAT A-3′ and R, 5′-GTC TCT TTC AAC TTG GCC CCT G-3′; LNG2 F, 5′-CGA CTT GAG GAG TCT AAG GTC-3′ and R,5′-GGA TCC CTG ATA ACC TTA AAA AAA TTA G-3′; actin (control) F,5′-ATG ACT CAA ATC ATG TTT GAG ACC TTC-3′ and R, 5′-ACC TTA ATC TTC ATG CTG CTT GGA GC-3′; AN F, 5′-TGA GAC GGT GCC GTG GTA TGG-3′ and R, 5′-GTT GCC TAC TGG TGG ATT CC-3′; ROT3F, 5′-TGA GGC CTC GGT TGT TCT CA-3′ and ROT3R, 5′-TCT CTA CGA TCT TTC CGC TG-3′; ATHB13-550R, 5′-TTC TGT TAC TGC AAG ATC CTT CAG T-3′; ATHB13-1F, 5′-ATG TCT TGT AAT AAT GGA ATG TC-3′; CYP90D1-F, 5′-GCA CAA GTT TTT GTC GGA ATC C-3′; CYP90D1-R, 5′-GTC GAT CAT ATT GTT AGC AAT C-3′.

Anatomical analysis

Samples for the anatomical analysis were examined as described by Tsukaya et al. (Tsukaya et al., 1993). All plants were grown under identical conditions. The fifth leaves were fixed overnight in FAA solution (5% acetic acid, 45% ethanol and 5% formaldehyde),dried under a vacuum for 20 minutes and dehydrated in a graded ethanol series(50, 60, 70, 80, 90, 95, 99 and 100%) at room temperature. The dehydrated samples were incubated at room temperature in 50% Tchnovit 7100 resin (Kulzer and Company) and 50% ethanol for 2 hours, and then in 100% resin overnight. The resin was hardened at room temperature and the resin-impregnated leaves were sectioned at 4 μm through the center of the broad blade for longitudinal sections and through the widest part of the blade for transverse sections. The slices were fixed to glass slides, dried at room temperature,stained with 0.1% Toluidine Blue in 0.1 M phosphate buffer (pH 7.0), and observed and photographed under a microscope.

The lng1-1D mutant has longer lateral organs in its aerial part

To identify new genetic components involved in leaf morphogenesis, we screened Arabidopsis activation-tagging lines and isolated a dominant mutant, which we named lng1-1D. Fig. 1A shows the overall morphology of a soil-grown lng1-1D mutant plant. A leaf-by-leaf comparison with wild-type plants showed that all leaves of lng1-1Dwere longer and slightly narrower than those of the wild type(Fig. 1B), mainly owing to elongated leaf blades and leaf petioles. In addition, the leaves of lng1-1D were more deeply serrated than those of the wild type; this exacerbated serrated phenotype was due to the protrusion of existing hydathodes rather than to an increase in the number of hydathodes(Fig. 1B, right panel). The elongated phenotype of lng1-1D was not restricted to the rosette leaves. As shown in Fig. 1C-F,other aerial parts showed elongated phenotypes, including the floral tissues,siliques, seeds and cotyledons. The hypocotyl length of lng1-1Dplants was slightly increased compared with the wild type, but the lengths of stems and primary roots were unchanged (see Table S1 in the supplementary material).

For quantitative comparison, we measured the lengths of various organs and structures, including primary roots, hypocotyls, petioles, leaf blades,internodes, siliques and seeds. The fully expanded leaf blades of lng1-1D mutants were 43% greater in length, but 29% narrower in width, compared with the wild type, suggesting that the mutation caused increased longitudinal leaf expansion with a concomitant decrease in transverse expansion (see Table S1 in the supplementary material). Consistent with the increased leaf length, the leaf petioles of lng1-1D mutants were 19% longer than those of the wild type, and the mutant seeds were longer(19%) and slightly narrower (6%) than those of wild type. Although there was no significant difference in stem length, the stem diameter was noticeably smaller (∼25%) than that of wild-type plants. Taken together, the results indicated that the dominant lng1-1D mutation affects the longitudinal expansion of various parts of the plant, including the leaf blade, leaf petiole, flower, cotyledon and seed.

In Arabidopsis, the fifth leaf is regarded as a good representative of the rosette leaves(Tsuge et al., 1996). To investigate the kinetics of leaf expansion in the lng1-1D mutant, we monitored changes over time in the lengths and widths of fifth rosette leaf blades from at least 15 plants grown in soil under identical environmental conditions. As shown in Fig. 2A, the expansion rate of leaf blade width was initially comparable between lng1-1D and wild-type plants, but decreased faster in the lng1-1D mutant, leading to narrower mature leaves in the mutant. By contrast, the expansion rate of leaf blade length was higher in the lng1-1D mutant from the beginning, resulting in a greater final length of lng1-1D leaf blades compared with wild type(Fig. 2B). Expansion of the fifth leaves stopped around 39 days after sowing, both in wild type and lng1-1D mutants, suggesting that the mutation altered the expansion rate but not the timing of expansion during leaf development.

lng1-1D plants have elongated cells in their leaves, petals and siliques

Leaf expansion is determined by cell proliferation and expansion. To investigate whether cell expansion was affected in the lng1-1Dmutant, we subjected the fifth rosette leaves of mutant and wild-type plants to scanning electron microscopy (SEM). Magnification of the leaf midvein area revealed that the adaxial epidermal cells in the midveins and the epidermal cells of nearby midveins were longitudinally elongated in lng1-1Dleaves compared with wild type (Fig. 3A,B,D,E). On the adaxial side of the leaf blade, the epidermal cells of the lng1-1D mutant plants were longitudinally elongated compared with the wild type (Fig. 3C,F). To semi-quantify differences in cell size, we measured the lengths and widths of more than 150 adaxial epidermal cells in mutant and wild-type plants. The wild-type and lng1-1D cells had lengths of 88.0±11.6 μm (six plants) and 132.6±19.1 μm (five plants),respectively, and widths of 83.0±7.4 μm (six plants) and 59.8±9.4 μm (five plants), respectively. The lng1-1Depidermal cells were 51% longer and 28% narrower than their wild-type counterparts, suggesting that the longer, narrower cells of the mutant largely accounted for the differences in leaf phenotype. Similarly, the adaxial epidermal cells in the both proximal and distal mutant petals were longitudinally elongated and transversely narrowed in the lng1-1Dmutant compared with wild type (Fig. 3G,H,K,L), as were the cells of the siliques(Fig. 3I,J,M,N).

LNG1 and LNG2 encode novel proteins that activate longitudinal plant organ expansion

We performed a plasmid rescue experiment to identify the gene responsible for the lng1-1D mutant phenotypes and identified a single T-DNA insertion at 5477 bp upstream of At5g15880(Fig. 4A). Since the T-DNA used included multiple copies of the 35S enhancer element, we tested whether the insertion caused the increased expression of nearby genes. Northern blot analysis indicated that the expression of one gene (At5g15880) was higher in mutant plants compared with wild type (Fig. 4B). This gene encodes a 927 amino acid protein containing a putative nuclear localization signal. BLAST searching identified no homologous proteins with known function, but did identify one closely related Arabidopsis homolog (At3g02170, 63% identical) of unknown function and a few related proteins from Arabidopsis and rice (∼22-27%identical; Fig. 4C). Based on the sequence similarity and our functional analysis, we designated At5g15580 as LNG1 and At3g02170 as LNG2. Amino acid sequence alignment indicated that several regions are well conserved among LNG1, LNG2and the related rice proteins (see Fig. S1 in the supplementary material). One of the conserved regions is serine rich, but none shows any similarity to known motifs.

To determine whether the elongated phenotypes of lng1-1D mutant plants were caused by the increased expression of LNG1, we generated transgenic Arabidopsis plants expressing this gene under the control of the CAMV 35S promoter. Similar to the phenotypes seen in the lng1-1D mutant, seven out of 15 independent transgenic lines displayed longer leaf blades, longer petioles and serrated leaf margins(Fig. 4F), indicating that overexpression of LNG1 is responsible for the elongated phenotypes of the lng1-1D mutant. As LNG2 is closely related to LNG1, we further tested whether overexpression of LNG2 could cause elongated phenotypes. As shown in Fig. 4G, five out of 18 independent transgenic plants overexpressing LNG2 also had longer petioles and slightly longer leaf blades. Interestingly, none of the LNG2 overexpression lines had serrated leaves, suggesting that LNG1 and LNG2 have distinct, but overlapping, functions. Collectively, these results indicate that LNG1 and LNG2 are involved in directing leaf morphology by activating longitudinal cell expansion.

As LNG proteins have not been characterized in previous studies, we experimentally determined the subcellular localization of LNG1 in transgenic Arabidopsis plants expressing a GFP-LNG1 fusion protein and in onion epidermis that had been bombarded with GFP-LNG1 fusion constructs. Bombardment with vectors encoding GFP alone caused green fluorescence in both the cytosol and nucleus (data not shown), with the nuclear localization likely to be due to the low molecular weight of GFP alone(von Arnim et al., 1998). The GFP-LNG1 fusion protein showed similar localizations in both the cytosol and nucleus (Fig. 5A-C). However,since the molecular weight of the fusion protein is 131 kDa, the nuclear localization of the fusion protein is likely to reflect the actual nuclear localization of LNG1 protein. To confirm this finding, we generated transgenic plants expressing GFP-LNG1 fusion proteins under the control of the CAMV 35S promoter. The transgenic plants showed phenotypes similar to those of the lng1-1D mutants, including very elongated and serrated leaves (data not shown), suggesting that the GFP-LNG1 fusion protein is functional. The GFP signal, however, was relatively weak and was detected only faintly in the root tip cells of transgenic plants (Fig. 5D-G). Consistent with the subcellular localization in bombarded onion epidermal cells, GFP signals were detected in both the cytosol and nucleus. It is, however, possible that the cytosolic GFP signals were the result of partially degraded GFP-LNG1 proteins.

To further investigate the physiological functions of LNG1 and LNG2, we determined the expression patterns of these genes in promoter:GUS transgenic plants. Promoter sequences (∼2 kb) from the LNG1 and LNG2 genes were used to drive expression of GUS. The ProLNG1:GUS and ProLNG2:GUS transgenes were expressed in various parts of adult transgenic plants, including the petioles, leaf blades, floral organs and roots (Fig. 6). In roots,GUS staining was mainly detected in the lateral roots. GUS was strongly expressed in and at the base of the petioles(Fig. 6B,F). In leaves, GUS staining was observed in all parts, with stronger staining in the veins(Fig. 6C,G). Consistent with these observations, microarray analysis data compiled in Genevestigator( indicate that LNG1 and LNG2 are expressed widely in various organs. These results collectively suggest that LNG1 and LNG2 are expressed in various Arabidopsis tissues, where they appear to regulate cell elongation.

We then characterized T-DNA-inserted putative loss-of-function LNG1 (lng1-2 and lng1-3) and LNG2(lng2-1 and lng2-2) mutants, which were obtained from the ABRC (Arabidopsis Biological Resource Center; Fig. 7A). To determine whether the T-DNA insertion lines were true loss-of-function mutants, we tested their expression levels of LNG1 and LNG. We observed weak expression of LNG1 in lng1-2 plants, but no such expression in lng1-3, indicating that lng1-2 was a weak mutant allele and that lng1-3 was a strong mutant allele (data not shown). No detectable expression of LNG2 was observed in the lng2-1 and lng2-2 mutants, indicating that both lng2-1 and lng2-2 were strong mutant alleles. To determine whether LNG1and LNG2 regulate each other, we examined the reciprocal expression levels in each line. As shown in Fig. 7B, loss-of-function of LNG1 or LNG2 did not affect the expression of the other gene. In addition, overexpression of LNG1 (observed in the lng1-1D mutant) did not alter the expression of LNG2.

Next, we examined the phenotypes of the lng1-3, lng2-1 and lng1-3 lng2-1 loss-of-function mutants to determine the effects of LNG1 and LNG2 on various organs and structures. The lengths of cotyledons,rosette leaves, siliques and flowers were negligibly or only mildly affected in the lng1-3 and lng2-1 single mutant plants, but the lng1-2 lng2-1 double mutant showed substantial decreases in the lengths of these organs (Fig. 7C). To quantify these differences, we measured various dimensions of these organs and structures in wild-type plants, as well as in lng1-3,lng2-1 and lng1-3 lng2-1 mutants (see Table S1 in the supplementary material). The widths of the leaf blades did not differ significantly among wild-type, lng1-3, lng2-1 and lng1-3 lng2-1 plants. By contrast, the leaf blade length was ∼15% shorter in both single mutants, and almost 30% shorter in the double mutant, indicating that LNG1 and LNG2 additively regulate the leaf-length expansion. The additivity of LNG1 and LNG2 in leaf-length expansion was also observed in other organs. Petiole length did not significantly differ between the lng1-3 and lng2-1 single mutants and wild-type plants, but was 37% shorter in the double mutant compared with wild type. Similarly, the silique length of the double mutant was about 40% shorter (see Table S1 in the supplementary material; Fig. 7C). In addition to the decreased leaf-length expansion of these organs, we observed that the stem diameters of the double mutant plants were increased by almost 60% compared with the wild type. Taken together, these data suggest that LNG1 and LNG2 additively promote leaf-length expansion in various organs,while inhibiting the expansion of stem thickness.

LNG1 and LNG2 additively promote longitudinal polar cell elongation in the leaf

SEM analysis of lng1-1D mutant plants suggested that LNG1promotes longitudinal polar cell elongation. To further investigate the role of LNG1 and LNG2 in polar cell elongation, we determined cell sizes in transverse and longitudinal leaf sections(Table 1; see Fig. S2 in the supplementary material). In longitudinal sections, the palisade cells were 46%longer in the lng1-1D mutant than wild type, but 24% shorter in the lng1-3 lng2-1 double mutant (Table 1; see Fig. S2 in the supplementary material). In transverse sections (representing width), however, we did not observe significant differences among the different lines(Table 1). Interestingly, cell proliferation did not appear to be significantly altered in the mutants. Both transverse and longitudinal sections showed similar numbers of palisade and mesophyll cells in wild-type, lng1-1D, lng1-3, lng2-1 and lng1-3 lng2-1 plants (Table 1). These results indicate that the altered leaf blade sizes in these plants are not due to altered cell proliferation, and that LNG1 and LNG2 regulate leaf expansion by promoting longitudinal polar cell elongation, not by regulating cell proliferation.

LNG1 and LNG2 regulate longitudinal polar cell elongation independent of ROT3

Previous studies have shown that ROT3 regulates leaf expansion by promoting longitudinal cell expansion, not proliferation(Kim et al., 1998; Kim et al., 1999; Kim et al., 2002), whereas AN and ATHB13 regulate leaf expansion by promoting transverse cell expansion(Tsuge et al., 1996; Hanson et al., 2001; Kim et al., 2002). As our data indicate that cell expansion is also regulated by LNG1 and LNG2, we tested for functional relationships among these factors. We first tested for interrelation of their expression levels. Our results revealed that the expression levels of ROT3, CYP90D1 (a ROT3 homolog), AN and ATHB13 were not altered in lng1-1D or lng1-3 lng2-1 double mutant plants(Fig. 8A), and that LNG1 and LNG2 expression levels were unchanged in the rot3-1 mutant (Fig. 8B), compared to wild type. These results indicate that LNG1 and LNG2 do not regulate expression of ROT3,CYP90D1, AN or ATHB13. Similarly, ROT3 does not appear to regulate expression of LNG1 or LNG2.

The functional relationship among these genes was further examined by analysis of a lng1-3 lng2-1 rot3-1 triple mutant and of a lng1-1D rot3-1 double mutant. The petiole length was shorter in the triple mutant compared with the lng1-3 lng2-1 double mutant and rot3-1single mutant, both of which had shorter petioles than wild type(Fig. 8D). This difference in petiole length was mainly due to longitudinal shortening of the cells (see Table S2 in the supplementary material). By contrast, the leaf blade size of the triple mutant was similar to that of the lng1-3 lng2-1 double mutant (Fig. 8C), and the blade cells were similar in size among the mutants(Fig. 8G; see Table S3 in the supplementary material). The leaf blades were already short in the lng1-3 lng2-1 and rot3-1 mutants, perhaps masking an additive shortening effect. Kozuka et al. suggested that the mechanism of petiole elongation differs from that of leaf blade elongation(Kozuka et al., 2005). Therefore, it is also possible that the LNG genes and ROT3 may regulate petiole length additively, while regulating blade length in a collaborative fashion. To further clarify the relationships between LNG1,LNG2 and ROT3, we measured the leaf blade length of an lng1-1D rot3-1 double mutant. The blade lengths of lng1-1D rot3-1 double mutant plants were intermediate between those of the lng1-1D overexpression line and the rot3-1 loss-of-function mutant plants (Fig. 8C). The intermediate blade length of the lng1-1D rot3-1 double mutant was due to cells of intermediate length, not to a change in cell number(Fig. 8G; see Table S3 in the supplementary material). Independent regulation of longitudinal elongation by LNG1 and ROT3 can also be seen in silique length (see Table S4 in the supplementary material). The lng1-1D had longer siliques than wild type, whereas the lng1-3 lng2-1 double mutant and the rot3-1single mutant had shorter siliques. In support of independent regulation of longitudinal elongation by LNG1 and ROT3, the lng1-3 lng2-1 rot3-1triple mutant had even shorter siliques than the lng1-3 lng2-1 double mutant or rot3-1 single mutant, whereas the lng1-1D rot3-1double mutant had intermediate siliques compared with lng1-1D and rot3-1 mutants. As LNG1 does not regulate the expression of ROT3 or of its close homolog CYP90D1(Fig. 8A), these results collectively suggest that LNG1 and ROT3 independently regulate longitudinal organ size (Fig. 8C-E).

The relationship between LNG and ROT3 is less clear in terms of regulating leaf-width expansion. The lng1-1D mutant had a slightly narrower leaf blade compared with wild type, whereas the rot3-1 mutant and the lng1-1D rot3-1 double mutant showed similar phenotypes to each other, with a marginally wider leaf blade. The latter observation may suggest that rot3-1 is epistatic to lng1-1D for leaf-width expansion. However, because the leaf blades of the lng1-1D rot3-1 double mutants were shorter than those of the lng1-1D mutants, it is unclear whether the marginally wider leaf blade of the double mutant is caused by rot3-1, or is instead associated with the shortening of the leaf length.

Previous studies have shown that AN regulates transverse leaf expansion. To investigate whether LNG1 and AN themselves regulate longitudinal and transverse leaf expansion independently, we generated a lng1-1D andouble mutant and examined leaf expansion. The lng1-1D an double mutants had longer petioles and longer leaf blades than wild-type plants(Fig. 8C,E). These results support the notion that longitudinal leaf expansion is regulated by LNG1 in an AN-independent fashion. For width direction, the relationship is not clear. The lng1-1D an double mutant had narrower leaf blades than wild type,but its widths were similar to those of the an single mutant. This suggests that AN is epistatic to LNG1 for the width elongation of leaf blades. Alternatively, because the leaf blade of the an mutant is already narrow, it is possible that further narrowing in lng1-1D an double mutants was masked. To investigate whether LNG1 functions at a specific leaf developmental stage, we determined the cell numbers and cell sizes throughout the various leaf developmental stages. The cell numbers in the fifth leaf increased until 21 days after sowing, and no difference was observed between wild-type and mutant plants(Fig. 8F; see Table S3 in the supplementary material), indicating that cell numbers are not regulated by LNG1, LNG2 or ROT3. The cell sizes increased longitudinally up to 39 days after sowing, both in wild-type and mutant plants(Fig. 8G), but the cell elongation rates of lng1-1D and lng1-3 lng2-1 cells were higher and lower, respectively, than that of wild-type cells at all tested leaf developmental stages. Taken together, these results suggest that LNG1, LNG2 and ROT3 additively promote longitudinal cell expansion in petioles and leaf blades throughout leaf development, rather than increasing cell number or acting at a specific leaf developmental stage.

We report that the novel Arabidopsis genes LNG1 and LNG2 are responsible for regulating the expansion of various aerial parts, including leaves, flowers and siliques, by promoting longitudinal cell expansion at the expense of transverse cell expansion. Our genetic analysis reveals that LNG1 and LNG2 regulate longitudinal polar cell elongation independently of ROT3, suggesting that Arabidopsis contains at least two independent pathways for promoting longitudinal cell expansion.

LNG1 and LNG2 promote longitudinal polar cell elongation in a ROT3-independent fashion

Characterization of a dominant Arabidopsis mutant with unusually long petioles, leaves, petals and siliques allowed us to identify a gene involved in longitudinal expansion of aerial plant organs. We designated this gene LNG1 and identified a homolog in Arabidopsis, which we designated LNG2. LNG1 and LNG2 play positive roles in longitudinal organ expansion in length direction, as evidenced by the elongated organs of the lng1-1D mutant and transgenic lines overexpressing LNG1 or LNG2(Fig. 4F,G), or the shortened organs of plants harboring loss-of-function mutations in LNG1 and LNG2. Consistent with their amino acid sequence similarity (63%),analysis of the lng1 lng2 double loss-of-function mutant revealed that LNG1 and LNG2 are functionally overlapping genes.

Paradermal and section image analysis of leaf blade cells showed that the cell numbers in the leaf blades were almost identical between wild-type and mutant/transgenic lines, suggesting that LNG1 and LNG2 do not affect cell proliferation (Table 1, see Fig. S2 and Table S3 in the supplementary material). However, the same analysis indicated that cells were lengthened in the lng1-1D plants, and additively shortened in lng1-3 lng2-1double mutants in the leaf-length direction. The ratio of wild-type to mutant cell length was roughly equivalent to the ratio of wild-type to mutant leaf length. Taken together, these results indicate that LNG1 and LNG2 regulate organ expansion by regulating polar cell elongation,not cell proliferation.

The roles of LNG1 and LNG2 in longitudinal cell expansion appear similar to that of the previously characterized ROT3. For example, ROT3-overexpression lines have longer leaf cells compared with the wild type (Kim et al.,1999), whereas rot3-1 loss-of-function mutants have shorter organs comprising shorter cells(Tsuge et al., 1996). Although these similar phenotypes could indicate that these genes are members of the same pathway, several lines of evidence suggest that LNG1 and LNG2 regulate longitudinal cell expansion independently of ROT3. First, our expression analysis revealed that altering the expression levels of LNG1 and LNG2 did not alter the expression levels of ROT3 or its homolog, CYP90D1(Fig. 8A). Second, the lng1-3 lng2-1 rot3-1 triple loss-of-function mutant showed additive shortening of cell length, whereas the lng1-1D rot3-1 double mutant showed an intermediate cell length, particularly in the case of petiole cells(see Table S2 in the supplementary material). Third, the lng1-1D rot3-1 double mutant showed longer leaf blades and petioles compared with the rot3-1 single mutant, but shorter leaf blades and petioles compared with the lng1-1D single mutant. These data indicate that the two classes of genes do not have an epistatic relationship with regard to longitudinal cell expansion, suggesting that longitudinal cell expansion is regulated by at least two independent pathways in Arabidopsis: one involving ROT3, and the other involving LNG1 and LNG2. As ROT3 encodes an enzyme required for BR biosynthesis(Kim et al., 2005), these data further suggest that Arabidopsis contains a BR-independent pathway that regulates longitudinal polar cell elongation.

The novel proteins LNG1 and LNG2 have homologs in other plant species

Analysis of the predicted protein sequences of LNG1 and LNG2 indicated that they are novel proteins of unknown function. However, BLAST analysis identified homologous protein sequences in Arabidopsis and rice. The Arabidopsis genome contains two additional LNG-like proteins showing overall sequence similarity and a few proteins that show partial sequence similarity. In rice, we identified three proteins showing ∼22-27% overall sequence similarity to LNG1 (Fig. 4C; see Fig. S1 in the supplementary material). Multiple sequence alignment of the four Arabidopsis proteins and the three rice proteins allowed us to identify several regions that are well conserved among the family members. One conserved region is highly enriched with serine residues, but overall these regions overall lack known motifs. Taken together,these results suggest that LNG family members are novel plant-specific proteins that promote longitudinal polar cell elongation through unknown mechanisms.

The presence of LNG-like genes in other plant species raises the possibility that these genes might play important roles in shaping various plant forms. In Arabidopsis, different ecotypes display different leaf shapes, ranging from rot3-like round leaves to an-like narrow leaves. Previous QTL analysis using recombinant inbred lines identified 21 QTL responsible for the natural variations found in adult leaves of the Ler and Col-4 ecotypes (Perez-Perez et al.,2002). Notably, one of the reported QTL loci (ad-LSI2) overlaps with the chromosome position of LNG1. According to the QTL analysis,ad-LSI2 was responsible for about 5% of the variance found in the recombinant inbred lines. Further experiments are needed to prove that LNG1corresponds to ad-LSI2.

This work was supported, in part, by grants from KOSEF(R21-2003-000-1002-0), the Agricultural Plant Stress Research Center(R11-2001-092-01002-0) and the Plant Metabolism Research Center(R11-2000-081), and by a MOST/KOSEF grant from the Environmental Biotechnology National Core Research Center (R15-2003-002-01002-0). We also thank the Korea Basic Science Institute (KBSI) for assistance with the Scanning Electron Microscope (SEM) analysis.

Altmann, T. (
). Recent advances in brassinosteroid molecular genetics.
Curr. Opin. Plant Biol.
Azpiroz, R., Wu, Y., LoCascio, J. C. and Feldmann, K. A.(
). An Arabidopsis brassinosteroid-dependent mutant is blocked in cell elongation.
Plant Cell
Berna, G., Robles, P. and Micol, J. L. (
). A mutational analysis of leaf morphogenesis in Arabidopsis thaliana.
Bowman, J. L., Eshed, Y. and Baum, S. F.(
). Establishment of the polarity in angiosperm lateral organs.
Trends Genet.
Choe, S., Dilkes, B. P., Fujioka, S., Takatsuto, S., Sakurai, A. and Feldmann, K. A. (
). The DWF4 gene of Arabidopsis encodes a cytochrome P450 that mediates multiple 22alpha-hydroxylation steps in brassinosteroid biosynthesis.
Plant Cell
Clough, S. J. and Bent, A. F. (
). Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana.
Plant J.
De Veylder, L., Beeckman, T., Beemster, G. T., Krols, L.,Terras, F., Landrieu, I., van der Schueren, E., Maes, S., Naudts, M. and Inze,D. (
). Functional analysis of cyclin-dependent kinase inhibitors of Arabidopsis.
Plant Cell
Franklin, K. A., Davis, S. J., Stoddart, W. M., Vierstra, R. D. and Whitelam, G. C. (
). Mutant analyses define multiple roles for phytochrome C in Arabidopsis photomorphogenesis.
Plant Cell
Fujioka, S., Li, J., Choi, Y. H., Seto, H., Takatsuto, S.,Noguchi, T., Watanabe, T., Kuriyama, H., Yokota, T., Chory, J. et al.(
). The Arabidopsis deetiolated2 mutant is blocked early in brassinosteroid biosynthesis.
Plant Cell
Hanson, J., Johannesson, H. and Engstrom, P.(
). Sugar-dependent alterations in cotyledon and leaf development in transgenic plants expressing the HDZhdip gene ATHB13.
Plant Mol. Biol.
Hemerly, A., Engler Jde, A., Bergounioux, C., Van Montagu, M.,Engler, G., Inze, D. and Ferreira, P. (
). Dominant negative mutants of the Cdc2 kinase uncouple cell division from iterative plant development.
Jones, A. M., Im, K. H., Savka, M. A., Wu, M. J., DeWitt, N. G.,Shillito, R. and Binns, A. N. (
). Auxin-dependent cell expansion mediated by overexpressed auxin-binding protein 1.
Kim, G. T. and Cho, K. H. (
). Recent advances in the genetic regulation of the shape of simple leaves.
Physiol. Plantarum
Kim, G. T., Tsukaya, H. and Uchimiya, H.(
). The ROTUNDIFOLIA3 gene of Arabidopsis thaliana encodes a new member of the cytochrome P-450 family that is required for the regulated polar elongation of leaf cells.
Genes Dev.
Kim, G. T., Tsukaya, H., Saito, Y. and Uchimiya, H.(
). Changes in the shapes of leaves and flowers upon overexpression of cytochrome P450 in Arabidopsis.
Proc. Natl. Acad. Sci. USA
Kim, G. T., Shoda, K., Tsuge, T., Cho, K. H., Uchimiya, H.,Yokoyama, R., Nishitani, K. and Tsukaya, H. (
). The ANGUSTIFOLIA gene of Arabidopsis, a plant CtBP gene, regulates leaf-cell expansion, the arrangement of cortical microtubules in leaf cells and expression of a gene involved in cell-wall formation.
Kim, G. T., Fujioka, S., Kozuka, T., Tax, F. E., Takatsuto, S.,Yoshida, S. and Tsukaya, H. (
). CYP90C1 and CYP90D1 are involved in different steps in the brassinosteroid biosynthesis pathway in Arabidopsis thaliana.
Plant J.
Kozuka, T., Horiguchi, G., Kim, G. T., Ohgishi, M., Sakai, T. and Tsukaya, H. (
). The different growth responses of the Arabidopsis thaliana leaf blade and the petiole during shade avoidance are regulated by photoreceptors and sugar.
Plant Cell Physiol.
Li, J. and Chory, J. (
). A putative leucine-rich repeat receptor kinase involved in brassinosteroid signal transduction.
Narita, N. N., Moore, S., Horiguchi, G., Kubo, M., Demura, T.,Fukuda, H., Goodrich, J. and Tsukaya, H. (
). Overexpression of a novel small peptide ROTUNDIFOLIA4 decreases cell proliferation and alters leaf shape in Arabidopsis thaliana.
Plant J
Perez-Perez, J. M., Serrano-Cartagena, J. and Micol, J. L.(
). Genetic analysis of natural variations in the architecture of Arabidopsis thaliana vegetative leaves.
Qin, M., Kuhn, R., Moran, S. and Quail, P. H.(
). Overexpressed phytochrome C has similar photosensory specificity to phytochrome B but a distinctive capacity to enhance primary leaf expansion.
Plant J.
Sinha, N. (
). Leaf development in angiosperms.
Annu. Rev. Plant Physiol. Plant Mol. Biol.
Timpte, C. C., Wilson, A. K. and Estelle, M.(
). Effects of the axr2 mutation of Arabidopsis on cell shape in hypocotyl and inflorescence.
Tsuge, T., Tsukaya, H. and Uchimiya, H. (
). Two independent and polarized processes of cell elongation regulate leaf blade expansion in Arabidopsis thaliana (L.) Heynh.
Tsukaya, H. (
). Organ shape and size: a lesson from studies of leaf morphogenesis.
Curr. Opin. Plant Biol.
Tsukaya, H., Naito, S., Rédei, G. P. and Komeda, Y.(
). A new class of mutations in Arabidopsis thaliana,acaulis1, affecting the development of both inflorescences and leaves.
Ullah, H., Chen, J. G., Young, J. C., Im, K. H., Sussman, M. R. and Jones, A. M. (
). Modulation of cell proliferation by heterotrimeric G protein in Arabidopsis.
Verkest, A., Manes, C. L., Vercruysse, S., Maes, S., Van Der Schueren, E., Beeckman, T., Genschik, P., Kuiper, M., Inze, D. and De Veylder,L. (
). The cyclin-dependent kinase inhibitor KRP2 controls the onset of the endoreduplication cycle during Arabidopsis leaf development through inhibition of mitotic CDKA;1 kinase complexes.
Plant Cell
von Arnim, A. G., Deng, X. W. and Stacey, M. G.(
). Cloning vectors for the expression of green fluorescent protein fusion proteins in transgenic plants.
Wang, H., Zhou, Y., Gilmer, S., Whitwill, S. and Fowke, L. C. (
). Expression of the plant cyclin-dependent kinase inhibitor ICK1 affects cell division, plant growth and morphology.
Plant J.
Weigel, D., Ahn, J. H., Blazquez, M. A., Borevitz, J. O.,Christensen, S. K., Fankhauser, C., Ferrandiz, C., Kardailsky, I.,Malancharuvil, E. J., Neff, M. M. et al. (
). Activation tagging in Arabidopsis.
Plant Physiol.