Chrysanthemum transcription factor CmLBD1 direct lateral root formation in Arabidopsis thaliana

The plant-specific LATERAL ORGAN BOUNDARIES DOMAIN (LBD) genes are important regulators of growth and development. Here, a chrysanthemum class I LBD transcription factor gene, designated CmLBD1, was isolated and its function verified. CmLBD1 was transcribed in both the root and stem, but not in the leaf. The gene responded to auxin and was shown to participate in the process of adventitious root primordium formation. Its heterologous expression in Arabidopsis thaliana increased the number of lateral roots formed. When provided with exogenous auxin, lateral root emergence was promoted. CmLBD1 expression also favored callus formation from A. thaliana root explants in the absence of exogenously supplied phytohormones. In planta, CmLBD1 probably acts as a positive regulator of the response to auxin fluctuations and connects auxin signaling with lateral root formation.

polypeptide is a class I LBD member, harboring a LOB domain with both a C block and a GAS block. The former is a 22 residue sequence of the form CX 2 CX 6 CX 3 C, while the latter ends with a DPIYG motif (Fig. 1a).
CmLBD1 is likely involved in auxin-induced development of the adventitious root primordium. CmLBD1 transcripts were detected in the chrysanthemum stem and root, but not in the leaf. Transcript abundance was highest in the root (Fig. 1b); in the presence of auxin, the level peaked 2 h after the treatment, while in the presence of TIBA, the gene was down-regulated after 2 h, after which it slowly recovered (Fig. 2a). Once adventitious roots began to develop, the gene was markedly up-regulated, peaking after four days, and falling away once the adventitious root primordium had completely formed. In the presence of indole acetic acid (IAA), the transcription peak was brought forward by one day, while the effect of 2.3.5-triidobenzoid acid (TIBA) was to delay it by two days (Fig. 2b). Under normal conditions, early adventitious root primordium with apical meristems are visible under the microscope in five day old cuttings 22 ; in the presence of auxin, this developmental stage was brought forward to three days, while in the presence of TIBA, it was delayed to six days (Fig. 2c). Histological analyses further determined the time of early adventitious root primordium in the cutting bases (Fig. 2d).

Subcellular localization and transcriptional activation of CmLBD1. In onion epidermal cells tran-
siently expressing the construct 35 S::GFP-CmLBD1, GFP accumulated most markedly in the nucleus, while the product of the control GFP transgene was deposited in both the cytoplasm and the nucleus (Fig. 3a). The yeast strain harboring the complete GAL4 domain (pCL1) was able to grow on SD/-His-Ade medium; in contrast, the negative control strain harboring the pGBKT7 vector was able to grow on Trp-deficient medium, but not on SD/-His-Ade medium. Yeasts harboring CmLBD1 were able to activate the reporter genes His3, Ade2 and Mel1, allowing the cells to grow on SD/-His-Ade medium, as shown by the pigmentation induced when the (c) Adventitious root primodium formation in chrysanthemum 'Jinba' cuttings. Primordia developed after five days in the control treatment. Exposure to IAA and TIBA respectively accelerated (four days) and delayed (seven days) this developmental stage. Bar: 1 cm. The base of the cutting is shown to the right of each image, magnified seven fold. Adventitious roots were indicated using red arrows. (d) Histological properties of adventitious root primodium formation in the base of cuttings. Bar: 50 μ m. medium contained X-α -Gal (Fig. 3b). The result showed CmLBD1 had transcriptional activation activity in yeast cells in vitro. To further confirm the transactivation function of CmLBD1 in vivo, the 35 S::GAL4DB-CmLBD1 and a luciferase reporter 5 × GAL4-LUC were co-transfected into Arabidopsis protoplasts. In addition, we used 35 S::GAL4DB-AtARF5 as a positive control, and 35 S::GAL4DB as a negative control. Although the LUC activity of CmLBD1 was lower than positive control 35 S::GAL4DB-AtARF5, CmLBD1 resulted in strong LUC activity compared with negative control 35 S::GAL4DB in Arabidopsis protoplasts (Fig. 3c). These results suggested that CmLBD was an activator of transcription.
Heterologous expression of CmLBD1 increased lateral root growth and promoted callus growth from root explants. Two independent transgenic A. thaliana lines (35 S::CmLBD1-1 and 35 S::CmLBD1-2) were obtained. CmLBD1 was found in transgenic Arabidopsis 35 S::CmLBD1-1 and 35 S::CmLBD1-2, but not in wild type Col-0 (Fig. 4a). Lateral roots were more numerous in eight day old transgenic seedlings than in wild type ones (Fig. 4b). Lateral roots were seemed swollen in the upper primary roots. But the primary root length and the average of lateral root length are almost unchanged (Table 1). A qRT-PCR analysis of the transcription in the root of a set of genes involved in lateral root initiation (PLETHORAs, AtPLTs; PIN-FORMEDs, AtPINs and D-type cyclins, AtCYCDs) showed they were all up-regulated in roots of 5-day-old seedlings ( Fig. 4c) and 10-day-old seedlings (Fig. 4d). When root explants were cultured for ten days in the absence of exogenously supplied phytohormones, callus formation was strong in both transgenic line explants, but was suppressed in the wild type ones (Fig. 5).
CmLBD1 is inducible by auxin and affects lateral root formation. The effect of providing a source of exogenous auxin 1-Naphthaleneacetic acid (NAA) on the growth of lateral roots was contrasted between wild type A. thaliana and the two transgenic lines. The 6-day-old seedlings of wild type and two transgenic lines which growing in 1/2 MS medium without any exogenously phytohormones as control (Fig. 6a). In the presence of either 0.1 μ M or 1 μ M NAA, the transgenic seedlings developed at least three lateral roots within 72 h, while the wild type seedlings produced none (Fig. 6b,c).

Discussion
CmLBD1 could response to auxin signaling in Chrysanthemum. The sequence of CmLBD1 indicates that it encodes a class I LBD transcription factor, a family of proteins which includes a number of members involved in root initiation and auxin signaling, including AtLBD16, AtLBD17, AtLBD29, OsARL1 and ZmRTCS 11,13 (Fig. 1a). The gene was strongly transcribed in the root and not at all in the leaf (Fig. 1b), which implies that its function is associated with root processes. In particular, the indications are that CmLBD1 is a positive regulator of adventitious root initiation responsive to auxin signaling. In A. thaliana, lateral root and adventitious root initiation is responsive to auxin 23,24 , and the LBD genes AtLBD16, AtLBD17, AtLBD18 and AtLBD29 are all inducible when plants are grown in a medium containing the synthetic auxin 2,4-Dichlorophenoxyacetic acid (2,4-D) 25 . Similarly, tomato cotyledon explants tend to differentiate a higher number of adventitious roots when the medium contains NAA 26 . The inference is therefore that the product of CmLBD1 contributes to the process of adventitious root primordium formation from chrysanthemum cuttings (Fig. 2c).  The phenotype of eight day old WT and transgenic seedlings. Bar: 1 cm. qRT-PCR analysis of PLTs, PINs and CYCDs transcription in the roots of five day old (c) and ten day old seedlings (d), respectively. Values given as mean ± SE (n = 3).  proteins were localized predominantly in the nucleus. When basic amino acids, #113 Lys and #128 Arg changed into Thr and Ser in the coiled-coil motif, C-terminal fragment will be distributed in both the cytoplasm and the nucleus 29 . The maize rootless concerning crown and seminal roots (ZmRTCS), a LOB domin protein, localizes to the nucleus, while its paralog RTCS-LIKE (ZmRTCL) is deposited in both the nucleus and the cytoplasm 30 .
The transient expression experiment showed that CmLBD1 can be localized to the nucleus (Fig. 3a), consistent with the notion that its function is to regulate the transcription of other genes. Both ZmRTCS and ZmRTCL have demonstrated a self-activation capacity in yeast 30 , and AtLOB is also capable of transcriptional activation 28 .
Here, CmLBD1 was shown to be able to activate the three reporter genes His3, Ade2 and Mel1 in yeast, according to grow on SD/-His-Ade medium and show blue when the medium containing X-α -gal, further supporting the suggestion that CmLBD1 is a positive regulatory factor in yeast cells (Fig. 3b). Further detection of luminescence assay showed that it might act as a transcription activator in Arabidopsis proroplasts (Fig. 3c). Being similar to OsARL1, AtLOB and ZmRTCS, CmLBD1 can activate transcription in yeast and Arabidopsis proroplasts.
Lateral roots numbers was increased with or without auxin in two transgenic A. thaliana lines. Although adventitious roots are morphologically similar to lateral ones, the mechanistic basis of their initiation is less well understood 31 . The heterologous expression of CmLBD1 in A. thaliana resulted in a distinct phenotype, in which lateral root formation was promoted (Fig. 4b). Lateral roots were seemed swollen in upper primary root near hypocotyls. It might be a precursor of callus and form callus easier. The over-expression of AtEXP17 (a gene regulated by AtLBD18) enhances lateral root formation, a result of stimulation by auxin 3 . The polar nuclear migration is inhibited in transgenic plants expressing the fusion transgene AtLBD16-SRDX (SRDX is a 12 residue motif which converts transcription factors to dominant repressors). Only a small number of lateral roots are formed by Atlbd16 single and Atlbd16/lbd18/lbd33 triple mutants, while none are formed by the AtLBD16-SRDX transgenic 32 . AtARF17 negatively regulates adventitious root formation 20,21 . The over-expression of AtABCB19 generates a number of adventitious roots through its enhancement of auxin transport and accumulation 33 . Here, lateral root development was promoted in the CmLBD1 transgenic plants when they were provided with an exogenous source of auxin (Fig. 6), suggesting that CmLBD1 controls lateral root development in response to auxin. PLTs are key effectors of auxin synthesis [34][35][36] and calls formation 37 . PINs transport auxin from the center of the root (stele) to the new root tip, and then away again through the epidermis, which forms the basis of lateral root formation [38][39][40] . CYCDs controlled the G1-to-S phase of cell cycle transition and mediated pericycle responses to auxin signaling [41][42][43] . CmLBD1 also up-regulated the transcription of AtPLTs, AtPINs and AtCYCDs genes in 5-day-old seedlings and 10-day-old seedlings (Fig. 4c,d), which were implicated in controlling lateral root formation and callus formation 35,44,45 . With heterologous expression in A. thaliana, lateral roots could continue to grow within in a certain time. The results indicated that CmLBD1 might act as an important element to maintain the proliferative activity of pericycle cell.
CmLBD1 expression favored callus formation. Plant regeneration from in vitro grown callus is regulated by auxin and cytokinin. In A. thaliana, the genes AtLBD16, AtLBD17, AtLBD18 and AtLBD29 provide the necessary link between auxin signaling and regeneration 26 , while Micro160 (MiR160) and AtARF10 have been identified as important regulators of shoot regeneration from in vitro cultures 46 . Shoots were not found from callus in the absence of exogenously supplied phytohormones medium in our experiment. ARF10 is targeted by MiR160, but mechanisms are different in different tissues. The ability of CmLBD1 transgenic lines to regenerate callus was not dependent on the provision of any phytohormones, suggesting that the expression of the transgene drives callus formation in vitro (Fig. 5). Taken together, CmLBD1, a class I LBD transcription factor gene, played an important role in the process of adventitious root primordium formation of chrysanthemum. In A. thaliana, CmLBD1 positive regulated lateral root formation through response auxin signaling. This strongly suggests that CmLBD1 acts as a positive regulator and participates in root formation.

Materials and Methods
Plant materials and cultivation. Five to six leaf cuttings were taken from the Chrysanthemum cultivar 'Jinba' , which is maintained by the Chrysanthemum Germplasm Resource Preserving Centre (Nanjing Agricultural University, China). The cuttings were rooted in a 1:1 mixture of perlite and vermiculite. After two weeks under the 16 h photoperiod (80-100 μ mol/m 2 /s illumination) at 22 ± 1 °C conditions, the roots, stems and leaves were explored and analysed the tissue-specific transcription profiles of CmLBD1 gene. The transcription of CmLBD1 was monitored in cuttings held for 1 h in a liquid medium containing either 150 μ M auxin IAA or 150 μ M of the auxin inhibitor TIBA 47,48 , with a set of control cuttings placed in sterile water. Each cutting base with 8 mm in length was sampled before the transfer, then after 0.5, 1, 2, 4, 8, 12 and 24 hours, and subsequently once daily over the next six days. Samples were taken in triplicate.
Isolation and sequencing of CmLBD1 cDNA. RNA was isolated from leaves, stems and roots of 'Jinba' plants using the RNAiso reagent (TaKaRa, Tokyo, Japan) according the manufacturer's protocol, and a 1 μ g aliquot was converted into ss cDNA using M-MLV reverse transcriptase (TaKaRa). Internal fragment was identified based on other LBD genes sequences in GenBank. The full length CmLBD1 cDNA was obtained by applying 5′ -RACE and 3′ -RACE PCR 49 . For the 3′ reaction, the adaptor primer dT-AP was used for the reverse transcription step and the gene-specific primers GSP3′ -1/-2/-3 for the amplification step. For the 5′ reaction, the gene-specific primers GSP3′ -1/-2/-3 and 5′ RACE System kit (Invitrogen, Carlsbad, CA, USA). The resulting PCR product was purified and inserted into the plasmid pMD19-T (TaKaRa) for sequencing. Primer sequences are provided in Table S1. Subcellular localization of CmLBD1. The CmLBD1 coding sequence was amplified using a forward primer (LBD1-ENTR1A-F) and a reverse primer (LBD1-ENTR1A-R) (table S1 in file S1). The purified PCR product was restricted by Sal I and Not I and the resulting fragment inserted into the pENTR1A vector (Invitrogen) and thence into pMDC43 50 using LR Clonase TM II enzyme mix (Invitrogen). An empty vector (containing the N terminus of GFP) was used as a negative control. The recombinant plasmids were transiently expressed in onion epidermal cells, following He-driven particle bombardment (PDS-1000; Bio-Rad, Hercules, CA, USA) and a 16 h culture on Murashige-Skoog (MS) medium in the dark at 23 °C 51 . The expression of GFP was monitored by confocal laser scanning microscopy.

Transcriptional activity analysis of CmLBD1.
The CmLBD1 open reading frame was amplified using the primer pair LBD1-BD-F / LBD1-BD-R, which harbor, respectively, an Nde I and a BamH I recognition site (primer sequences given in Table S1) and then inserted into the yeast expression vector pGBKT7 (Clontech, Mountain View, CA, USA). The vector pCL1 (containing a full length copy of GAL4) was used as a positive control, and an empty pGBKT7 as the negative control. The constructs were introduced into yeast (Saccharomyces cereviseae) strain Y2HGold (Clontech) following the Yeastmaker Yeast Transformation System 2 protocol. The pCL1 transformants were incubated on a SD medium SD lacking leucine, while the pGBKT7 and pGBKT7-CmLBD1 ones were incubated on a SD medium lacking tryptophan. After culturing at 30 °C for 3 d, the transgenic cell lines were transferred onto a SD medium lacking both histidine and adenine (hemisulfate salt) either in the presence or absence of X-α -Gal 52 .
Then the luminescence assay of CmLBD1 was further examined for transactivation activity in Arabidopsis mesophyll protoplasts. The plasmid pENTR1A-CmLBD1 previously was subjected to vector 35S::GAL4DB using LR Clonase TM II enzyme mix. Arabidopsis mesophyll protoplast isolation and transformation were based on the protocol as described by Wu et al. 53 . 7.5 μ g 35S::GAL4DB-AtARF5, 35S::GAL4DB and 35S::GAL4DB-CmLBD1 were transfected, respectively. Additional 7.5 μ g GAL4-LUC as luciferase reporter plasmid were added 54 . Luciferase assay was as described by Fujikawa  Technology) replaced the ViviRen Live Cell Substrate. The protoplasts were incubated in 6-well plates for 16 h in light at 23 °C. LUC images were captured by a low-light cooled CCD imaging apparatus (DU934P Andor, UK) in 96-well plate. LUC activity was measured with 10 sec integration periods (Promega, Madison, Wisconsin, USA). Counts of luminescence were quantified with a 20/20n Luminometer (Turner BioSystems). Three independent experiments were performed for each assay.
A. thaliana transformation. The CmLBD1 sequence harbored by pENTR1A was transferred into pEarleyGate103 56 using LR Clonase TM II enzyme mix, then transformed into A. thaliana Col-0 using an Agrobacterium tumefaciens EHA105-mediated floral dip method 34 . The transgene was driven by the CaMV 35S promoter. Transgenic progeny were selected by including basta herbicide in the culture medium.

Auxin-induced lateral root formation in transgenic plants.
A. thaliana seedlings were grown on vertically oriented plates containing half strength MS medium, 3% w/v sucrose and 0.6% w/v agar under a 16 h photoperiod at 22 ± 1 °C. After three days, wild type and two independent transgenic A. thaliana seedlings were transferred to a half strength MS medium containing either 0.1 μ M or 1 μ M naphthalene acetic acid (NAA). The growth of lateral roots was observed after 72 h 32 .
Callus formation in vitro grown root explants. Root material was sampled from the maturation zone of both wild type and CmLBD1 transgenic seedlings cultured on half strength MS medium. The explants were transferred to B5 medium lacking phytohormones and cultured for 40 days to induce callus formation 26 .