Opposite physiological effects upon jasmonic acid and brassinosteroid treatment on laticifer proliferation and co-occurrence of differential expression of genes involved in vascular development in rubber tree

Abstract

During growth of woody plant-trunk, the secondary meristem functions in giving rise the xylem and phloem. Rubber tree (Hevea brasiliensis Muell. Arg.), in addition, contains laticifers (latex producing vessels) in the vicinity of phloem. Insights into regulatory mechanisms of gene networks underlying laticifer proliferation in rubber tree has remained very limited. The candidate vascular development-related genes were selected to investigate for expression profile in phloem and xylem tissues of high latex yield- and high wood yield-clones of rubber tree. The differential gene expression between the mature branch-xylem and -phloem tissues was clearly observed. The cis-regulatory motif analysis revealed the existent of putative jasmonic acid (JA)- and brassinosteroid (BR)-responsive regulatory motifs in promoter regions of these genes, and consequently the effect of exogenous application of JA, BR or their respective signaling inhibitors, on the formation of laticifers in rubber tree was demonstrated. Interestingly, the laticifer numbers were significantly increased in JA-treatment, correlated with up-regulation of phloem development-related genes in both rubber tree clones. On the contrary, the laticifers were decreased in BR-treatment accompanying by up-regulation of xylem development-related genes, especially in high wood yield-rubber tree clone. BR-inhibitor treatment also enhanced laticifer numbers, while JA-inhibitor suppressed laticifer differentiation. Taken together, this study unveils the molecular interplay between JA/BR on vascular development in rubber tree and how this impacts the appearance of laticifers in this plant. This process is vital for a better understanding on laticifer differentiation and its impact in the manipulation of wood and latex yield in rubber tree improvement program.

Electronic supplementary material

The online version of this article (10.1007/s12298-019-00686-0) contains supplementary material, which is available to authorized users.

Introduction

A higher plant continues to grow by forming new cells and organs in both apical and lateral directions throughout the life cycle. Vascular tissue in a woody plant composes of two basic units to support length-wise growth by providing both physical support and long-distance fluid transport (xylem) and nutrient transport (phloem). During secondary growth of stem, vascular cambium that is a meristematic tissue differentiates to secondary xylem to the inside of stem, commonly is called wood, and secondary phloem to the outside of stem, as soft bark. Xylem is a complex tissue, consisting of tracheary elements (water-conducting tissues), xylem fibers (lignified secondary walls to provide mechanical strength), and xylem parenchyma cells (living and store organic nutrients). The xylem is responsible for the movement of water and solutes from the root to the shoot, and the large woody secondary cell wall of xylem is responsible for mechanical strength of plant, represents the majority of plant biomass (Ye 2002). Phloem consists of sieve elements (photoassimilate-conducting tissue), companion cells (tightly connected with sieve elements to support and loading/unloading nutrients), phloem fibers and phloem parenchyma cells (differentiate from cells adjacent to sieve elements with extensive wall in growths). The phloem is responsible for movement of products from the photosynthetic leaf to other part and the root system (Offler et al. 2003).

Among the latex-producing plants, Para rubber tree (Hevea brasiliensis Muell. Arg.) is the sole commercial source of natural rubber. Although, the natural rubber is a key raw material in everything from elastomers to materials blended with synthetic rubber, yet, production is still tied to century-old agricultural practices. The global natural rubber market has been continued in short supply because of the steady growth in rubber demand and underproduction. The rubber tree is primarily grown for the production of natural latex, however rubber plantation is also an eco-friendly sustainable source of timber. Beside the huge information on rubber biosynthesis study, the vascular development including phloem, articulated laticifers (latex producing vessels) and xylem, has not much progress. In rubber tree, laticifers differentiate from cambium and locate vicinity of phloem and laticifer loading takes up sucrose from phloem surrounding laticifer and then converts sucrose into latex (Hagel et al. 2008; Tang et al. 2010). Wood yield is depend on xylem differentiation, but the latex yield is influenced by the number of laticifers as well as soft bark (phloem tissue) development (Gomez 1982; Elo et al. 2009). The regulatory network for laticifer differentiation in rubber tree has not been well documented. The latest discovery on the laticifer organization and its biological implications was on laticifer-deficient mutants (poor in latex, pil) in a latex bearing non-articulated laticifer plant, Euphorbia lathyris, in which the distinct pil loci revealed regulation of laticifer differentiation (Castelblanque et al. 2016).

Our knowledge regarding the vascular development in plants has rapidly expanded with the advancement in genomics and other molecular tools. Various molecular and genetic components, including phytohormones, components of signaling mechanisms, and transcription factors, required to initiate, pattern and expand the vascular tissue have been identified and characterized in the model plants (Xu et al. 2013; Ohashi-Ito and Fukuda 2014; Etchells et al. 2016). Although significant progress has been made in studies on the role of plant hormones and interconnections among some of the major signaling networks that regulate rubber tree vascular development, yet nothing was known about laticifer differentiation. Number of laticifers in rubber tree can be increased by Jasmonic acid (JA) and its precursor-treatment (Hao and Wu 2000; Zhang et al. 2015; Tian et al. 2015). Using Subtractive Hybridization Suppression (SSH), the result suggested that inhibition of cell proliferation may result in induction of laticifer differentiation (Zhang and Tian 2016). The correlation of primary and/or secondary laticifer vessel proliferation in phloem tissue and differential gene expression upon jasmonic acid (JA) treatment was reported (Tan et al. 2014; Laosombut et al. 2016; Loh et al. 2016; Wu et al. 2016; Tan et al. 2017). Brassinosteroid (BR) is one of phytohormone that is well-known to promote xylem differentiation from cambium meristem that determines wood yield (Ohashi-Ito et al. 2002; Ohashi-Ito and Fukuda 2003). Xylem and phloem differentiate antagonistically (Bonke et al. 2003; Caño-Delgado et al. 2004), therefore, laticifers that are coordinated with phloem may differentiate antagonistic to xylem.

The key genes related to vascular development were selected for this investigation. The candidate genes were classified into two groups depending on their previously reported function during developmental process in Arabidopsis and poplar. The phloem development-related genes are Lateral organ boundaries (LOB) domain-containing protein 1-like (LBD1), KANADI1 (KAN1) and HIGH CAMBIAL ACTIVITY (HCA2). Overexpression of PtaLBD1 significantly increased secondary phloem width (Yordanov et al. 2010). At a very early stage of inflorescence stem development, HCA2 promotes interfascicular cambium formation (Guo et al. 2009). KAN1 functions antagonistically to HB-Zip classIII genes, which are auxin responsive gene and promote xylem differentiation (Huang et al. 2014). The xylem development-related genes are HOMEOBOX8 (HB8), VASCULAR-RELATED NAC-DOMAIN7 (VND7) and BRASSINOSTEROID INSENSITIVE 1-like3 (BRL3). Ectopic expression of ATHB8 in Arabidopsis increases the production of xylem tissue consistent with overexpression of ATHB8 promote vascular differentiation (the procambial cells differentiated precociously into xylem) (Baima et al. 1995). VND7 encodes a NAC transcription factor, that plays an essential role in regulation of protoxylem development and regulates genes related to secondary wall formation during xylem development (Yamaguchi et al. 2011). BR is a phytohormone, and one of its well-known functions is to promote tracheary elements differentiation (Iwasaki and Shibaoka 1991). BRI1 encodes leucine-rich repeat receptor-like kinase, BRASSINOSTEROID INSENSITIVE 1 (BRI1), functions in BRs signaling (Li and Chory 1997). The bri1 mutant does not respond to BRs, shows multiple deficiencies in growth and development pathways consistent with the inhibition of secondary xylem development (Nagata et al. 2001). BRL3 (BRI1-like3) was reported to express mainly in vasculature and the expression pattern of BRL3 was BR dose-dependent (Caño-Delgado et al. 2004).

In this study, the molecular interplay between JA/BR on vascular development in rubber tree and how this impacts the appearance of laticifers in this plant were focused. The expression profiles of candidate genes related to vascular development in phloem and xylem tissue of the high latex yield- and the high wood yield-rubber tree clones were demonstrated. To elucidate their response to plant hormones and hormone inhibitors in relation with laticifer differentiation in rubber tree, methyl-jasmonate (MeJA) and brassinolide (BL) were used to determine the activity of JA and BR, while Ibuprofen (IBU) and propiconazole (PCZ), which are the potent inhibitors of JA and BR, respectively, were used to confirm the effect of endogenous phytohormone synthesis and the exogenous plant hormone treatments. The histochemical staining of laticifers upon plant hormone and inhibitor treatments in the two rubber tree clones were investigated and discussed in relation with the expression pattern of candidate genes related to vascular development.

Materials and methods

Rubber tree clones and tissue collection

The two rubber tree clones were selected from clone list recommended by Rubber Research Institute of Thailand (RRIT). RRIT251 (selected from illegitimate clonal seed of RRIT, recommended clone for high latex yield with 2140–2980 kg/ha/year) and RRIT402 (Chachoengsao50, selected from illegitimate clonal seed of RRIC110, recommended clone for high wood yield with 113 m³/ha of 9-year-old tree) (Supplementary Figure S1). The budded-seedlings of each clone were grown in pots until 3 month-old at the experimental station at Rubber Research Center at Chachoengsao Province (RRCC), Thailand, before using for hormonal treatments.

For the mature xylem and phloem tissues, the 5 cm–diameter of rubber tree branches were collected from the 15-year-old of RRIT251 and RRIT402 rubber tree plantation at RRCC. The mature phloem and xylem tissues were separated apart. The periderm layer (epidermis and cork), which is the outer-most layer of stem and overlays the vascular tissues, was peeled out. Then, the soft bark (phloem tissue, including laticifers) was separated from wood (xylem tissue) by peeling off and collected (Supplementary Figure S1). The plant tissues were quickly frozen using liquid nitrogen and collected at − 80 °C until use.

Phytohormone and inhibitor treatments

The 3-month-old rubber trees, with first extension unit (EU1), were treated for 1 month and the plants were allowed to grow without treatment for another 2 months, until the new extension unit (EU2) had generated in this 6-month-old plants (Fig. 1). The treatments were performed at 0, 12, 24 h, and then once a week for 1 month. For JA treatment, using cotton pad soaked 0.1 M Methyl-jasmonate (MeJA) + 0.1% tween 20 to wrap the plant stem at different times of treatment. For BR, JA-inhibitor or BR-inhibitor treatments, using foliar spray to the whole plant with 10 ppm Brassinolide (BL), 1.2 mM Ibuprofen (IBU) or 5 μm Propiconazole (PCZ) at different times of treatment, respectively.

Fig. 1.

Rubber tree was treated with plant hormones. The 3-month-old of rubber trees, with one extension unit (EU1), were treated by plant hormone at 0, 12, 24 h and then every week, for 1 month. The plants were further grown without treatment for another 2 months, until the second extended unit (EU2) was generated (6-month-old). The stems of plant hormones- and inhibitors-treated rubber trees were collected for histological staining and gene expression analysis

Histochemical study of the laticifer structure

The third node of stem, about 5-mm diameter, was cut by cross section and soaked into FAA (FORMALIN-ACETO-ALCOHOL) solution for 48 h. The samples were washed with glacial acetic acid for 1 h, and treated by Iodine and Bromine in glacial acetic acid followed by performing paraffin method (Laosombut et al. 2016). The embedded tissues in paraffin were sliced into 18-µm thick section with sliding microtome (LEICA, Germany) and observed under light microscope. The number of laticifer vessels was counted in 30 randomly selected visual fields under the microscope. The average number of laticifer vessels per area of stem section was calculated for standard error. To compare the effect of the treatments for each rubber tree clone, the t test was used. For analysis of the different treatments, the one-way ANOVA was used and the results were considered statistically significant when the P value was < 0.05.

Gene information, primer design and cis-regulatory elements prediction

Nucleotide sequences of each gene were searched against NCBI database, focusing on information of vascular development from rubber tree and other related plants (Supplementary Table S1). The specific primers for each gene were designed by PRIMER3 program (http://frodo.wi.mit.edu) for further gene expression analysis (Supplementary Table S2). The 3000 nucleotide sequences located upstream of the ATG sequence of each target gene were retrieved, and the cis-regulatory elements were predicted using the Plant Promoter Analysis Navigator (PlantPAN; http://PlantPAN2.itps.ncku.edu.tw) (Chow et al. 2016).

RNA preparation

The plant samples were ground to the fine powder in liquid nitrogen and using modified protocol to prepare for high quality total RNA. Cold Concert™ Plant RNA Reagent kit (Invitrogen, USA) 0.5 ml was added to 0.1 g of frozen ground tissue. The mixture was mixed thoroughly and incubated for 5 min at room temperature before centrifugation at 10,000g for 5 min to get supernatant. The 0.1 ml of 5 M NaCl was added to the supernatant following by extraction with 0.3 ml of chloroform-isoamylalcohol (24:1). After centifugation, the supernatant was transferred to a new RNase–free tube, and RNA was precipitated with equal volume of isopropyl alcohol for 10 min, and subsequently centrifuged at 4 °C for 15 min at 10,000g. The pellet was washed with cold 75% ethanol. Finally, 20 µl of RNase–free water was added to dissolve the RNA pellet. The samples were stored at − 80 °C until use.

First-strand cDNA synthesis

To ensure that the obtained RT-PCR products were originated only from the RNA template, the contaminated DNA was removed from RNA preparation using DNA-free™ Kit (Ambion, USA), according to the manufacturer’s instruction. Then, the first-strand cDNA was synthesized using the SuperScript™ III First-Strand Synthesis System for RT-PCR (Invitrogen, USA), according to the manufacturer’s instructions. The RNA template was removed using RNase H, following the previous protocol (Laosombut et al. 2016), and the first-strand cDNA synthesis reactions were either stored at − 20 °C or used immediately for PCR.

Gene expression analysis by quantitative real time-PCR

PCR amplification using the gene-specific primers was performed and validated. The PCR sequences were blasted to check for the correct target genes. Amplification efficiency of primers was performed and shown in Supplementary Table S15.

The quantitative RT-PCR (qRT-PCR) analysis was performed with iTaq™ Universal SYBR^® Green Supermix using CFX96 Touch™ Real-Time PCR Detection System (Bio-rad). A 20 μl PCR reaction mixture was prepared according to the manufacturer’s instruction. The amplification was monitored as fluorescence signals in real time during instrument run. For the control reaction, no cDNA template was added. The actin was selected as housekeeping gene. Expression level of the endogenous gene actin was used to normalize the detected level of gene expression (Laosombut et al. 2016; Guo et al. 2017). The qRT-PCR experiments were run by 3 individual experiments and performed in triplicate. The C_T values were calculated and the fold change in gene expression, after normalization of the C_T of non-treated and treated plants, to the reference gene (actin) was calculated. The relative expression of genes was manually calculated from the comparative C_T method, 2^−ΔΔCT. Significance of different testing of qRT-PCR within groups of sample (control and plant hormone treatments) was determined using ΔCT (ΔCT_(sample) = CT_(target) − CT_(ref)) (http://blog.mcbryan.co.uk/2013/06/qpcr-normalisation.html) and using one-way ANOVA with honestly significant difference (HSD) post hoc test for statistical analysis. Independent sample t test was performed to compare the difference between control and treatment at each time point. The results were considered statistically significant when the P value < 0.05. All statistical analysis was carried out using SPSS (release 18.0.0) software.

Result

Orthologous genes involved in vascular cambium, phloem and xylem differentiation in rubber tree were identified

The vascular development-related genes that have been previously reported in various plants, especially Arabidopsis and Populus, were selected. The six candidate genes were classified into two groups based on phloem and xylem differentiation. The nucleotide sequences of candidate genes in Arabidopsis thaliana were obtained, and blasted to seek for the orthologs in Hevea brasiliensis genome (Supplementary Table S1). There are 3, 10 and 2 gene accession numbers of phloem-related genes, LBD1, KAN1 and HCA2 in rubber tree genome database, respectively. Each xylem-related gene, HB8, VND7 and BRL3 has 2 gene accession numbers in rubber tree genome database. It should be noted that BRI1-like3 (BRL3) is a member of BRI1. There was no VND7 annotated in rubber tree genome database, however, there were 2 gene accession numbers of NAC domain-containing protein 30-like protein of rubber tree (HbNAC30) showed high similarity to Populus euphratica XM_011039829.1 NAC domian-contianing protein 7 (VND7). Therefore, HbNAC30 could be VND7 ortholog.

The PCR reactions were performed using cDNA templates generated from RNA of xylem and phloem tissues of rubber tree, and the PCR products were checked for expected size by agarose gel electrophoresis (Supplementary Figure S2). The expected PCR products were sent for DNA sequencing and blasted against NCBI database. Based on the blasted results, these rubber tree PCR-amplified products showed highest similarity to LBD1, KAN1, HCA2, VND7, HB8 and BRL3 of Hevea brasiliensis (Supplementary Table S1, S3-S8). The primers of each gene could be used to amplify cDNA of every gene members in each group.

Putative regulatory motifs in promoter region of candidate genes were predicted from available rubber tree genome database

The coding sequences of every members of LBD1, KAN1, HCA2, HB8, VND7 and BRL3 of Hevea brasiliensis were blasted in NCBI database to find the genome sequences containing each corresponding gene (Supplementary Table S9). The blasted results indicated that some gene accession numbers were actually the same member of one gene, thus located and driven by the same promoter region. This information suggested that there are 1 member of LBD1, and 2 members of each KAN1, HCA2, HB8, VND7 and BRL3. The upstream sequence of each candidate gene was analyzed for the putative cis-regulatory elements and shown in Supplementary Table S10. All candidate gene members with coding sequences and upstream sequences from genome sequence database of rubber tree clones RRIM600 were shown in Supplementary Table S11, and the hormonal responsive elements with typical motifs and functions were shown in Supplementary Table S12.

Up to now, genome scaffolds data from only two rubber tree clones is available, RRIM600 (medium latex yield clone, BioProject: PRJDB4387) and BMP24 (high latex yield clone, BioProject: PRJDB4987). In order to compare the types and numbers of hormonal responsive elements on the promoter region of each gene between these two rubber tree clones, the Hevea brasiliensis coding sequences of LBD1, KAN1, HCA2, HB8, VND7 and BRL3 were blasted in NCBI database to find the genome sequences containing the corresponding genes in each rubber tree clone (Supplementary Table S13). The upstream sequence of each candidate gene was analyzed and shown in Supplementary Table S14. Candidate genes with coding sequences and upstream sequences from genome sequence database of rubber tree clones RRIM600 and BMP24 are  shown in Supplementary Table S11. The number of putative hormonal-responsive regulatory motifs in the promoter region of all candidate gene members from genome sequences of rubber tree clones RRIM600 is shown in Supplementary Table S10 and Supplementary Figure S3. The hormonal-responsive regulatory motifs were focused, including jasmonic acid-responsive elements, auxin-responsive elements, cytokinin-responsive elements, ethylene-responsive elements and brassinosteroid-responsive elements. The putative auxin-responsive elements are the most available motifs, while jasmonic acid-responsive elements and cytokinin-responsive elements are rare in every gene in both rubber tree clones. Different gene members contain different number of hormonal-responsive regulatory motifs in their promoter region. The number of putative hormonal-responsive regulatory motifs in the promoter region of candidate genes from genome sequences of rubber tree clones RRIM600 and BMP24 is shown in Supplementary Table S14 and Fig. 2. There was no JA-responsive element in promoter region of KAN1-2 in RRIM600. The promoter region of HB8-1 in BMP24 lacked of both JA- and cytokinin-responsive elements. The number of JA-responsive elements in promoter of HCA2-2 more than HCA2-1, VND7-1 and more than VND7-2. The number of BR-responsive elements in promoter of VND7-2 was more than VND7-1. The number of JA-responsive elements in promoter of LBD1, HCA2, HB8 and BRL3 in RRIM600 was more than in BMP24. The number of BR-responsive elements in promoter of LBD1 in RRIM600 was more than in BMP24, while there were in promoter of KAN1, HCA2, HB8, VND7 and BRL3 in BMP24 more than in RRIM600.

Fig. 2.

Comparison of putative hormonal-responsive regulatory motif number in the promoter region. The hormonal-responsive regulatory motifs in the promoter region of six candidate vascular development-related genes from genome sequences of RRIM600 and BMP24 rubber tree clones were compared

Differential expression of vascular development-related genes in mature xylem and phloem tissues of the high latex-yield RRIT251 and high wood-yield RRIT402 rubber tree clones

Expression of LBD1 was 1.6-fold higher in phloem tissue compared to xylem tissue of RRIT402, while phloem and xylem tissues of RRIT251 exhibited no significantly different expression level. In addition, the expression level of LBD1 in RRIT402 phloem was about 1.9-fold higher than in RRIT251 phloem (Fig. 3a). KAN1 showed 2.1-fold higher expression in phloem tissue than in RRIT251 xylem, while phloem and xylem tissues of RRIT402 exhibited not significantly different expression level. In addition, the expression level of KAN1 in RRIT251 phloem was about 2.1-fold higher than in RRIT402 phloem (Fig. 3b). Expression of HCA2 was predominantly 4.3-fold higher in xylem compared to phloem of RRIT251, while phloem and xylem of RRIT402 exhibited no significantly different expression level. In addition, the expression level of HCA2 in xylem of RRIT251 was about 4.3-fold higher than in RRIT402 (Fig. 3c).

Fig. 3.

qRT-PCR analysis of LBD1 (a), KAN1 (b), HCA2 (c), HB8 (d), VND7 (e) and BRL3 (f) expression in RRIT251 and RRIT402 rubber tree clones. 251X, 251P, 402X and 402P are xylem and phloem tissues of RRIT251 and RRIT402, respectively. The relative expression of target genes was determined by the comparative C_T method (2^−ΔΔCT). The housekeeping reference gene, actin, was used as an internal control for normalization of the target gene detection levels. Statistical analysis was performed using Tukey’s test, one-way ANOVA. Significant differences at P value < 0.05

HB8 is predominantly expressed in xylem of both rubber tree clones, whereas apparently almost no expression was found in phloem tissue. Interestingly, HB8 expression in RRIT402 xylem was about 1.7-fold higher than in RRIT251 xylem (Fig. 3d). Expression of VND7 was significantly higher in xylem tissue than the almost no expression in phloem tissue of both rubber tree clones. Moreover, VND7 was expressed predominantly in RRIT402 xylem which was about sixfold higher as compared to RRIT251 xylem (Fig. 3e). Expression of BRL3 was predominantly fourfold in xylem than in phloem of RRIT251, while phloem and xylem of RRIT402 exhibited no significantly different expression level. In addition, the expression level of BRL3 in RRIT251 xylem was about 2.4-fold higher than in RRIT402 (Fig. 3f). It should be noted that the relative expression corresponds to the ratio of the transcript abundance of target gene/actin was highest in HB8 (0.075 in RRIT402 xylem) as compared to LBD1 (0.015 in RRIT402 phloem), HCA2 (0.013 in RRIT251 xylem), BRL3 (0.006 in RRIT251 xylem), VND7 (0.0018 in RRIT402 xylem) and KAN (0.00017 in RRIT251 phloem), respectively.

Effects of JA, BR and inhibitor treatments on the high latex-yield and high wood-yield rubber tree clones

As expected, RRIT251, a recommended high-latex yield clone, showed higher average number of laticifers than RRIT402, a recommended high wood-yield clone (Fig. 4). Average number of laticifers per one area of the stem-treated by MeJA was 1.3-fold increased in both rubber tree clones compared to untreated trees. Interestingly, BL treatment showed 0.4 to 0.6-fold decreasing of the number of laticifers in both rubber tree clones compared to untreated trees. The MeJA-treated rubber trees showed significantly higher number of laticifers in comparison to BL-treated rubber trees. Interestingly, the PCZ-treated rubber trees showed 1.2-fold increasing number of laticifers in comparison to untreated trees, in both rubber tree clones (Fig. 5). In contrast, IBU-treated rubber trees showed opposite effect to PCZ treatment. The rubber tree treated by IBU showed 0.7 to 0.8-fold decreasing number of laticifers in both clones compared to untreated trees. The PCZ-treated rubber trees showed significantly higher number of laticifers in comparison to IBU-treated rubber trees.

Fig. 4.

Histological study of laticifer in plant hormones-treated rubber tree. Cross-section of rubber tree stem at the 3rd node of EU2, showing the effect of MeJA and BL treatment on laticifer development in RRIT251 (left) and RRIT402 (right). The treatments had done with the 3-month-old rubber trees and let the plants further grew for another 3 months to generate EU2. The rubber trees were treated by water (control) (a–b), 10 ppm Brassinolide (BL treatment) (c–d) and 0.1 M Methyl-jasmonate (MeJA treatment) (e–f). (Scale bars 200 μm) (upper panel). Relative laticifer number per one unit area of rubber tree stem treated with BL/MeJA (lower panel). The treatments were done with the 3-month-old rubber trees and let the plants further grew for another 3 months to generate EU2. The rubber trees were treated by water (Control), 10 ppm Brassinolide (BL), and 0.1 M Methyl-jasmonate (MeJA). Data were collected from four randomly chosen areas for each sample. Using t test, the results were considered statistically significant when the P value < 0.05. The letter indicated significant differences

Fig. 5.

Histological study of laticifer in hormone inhibitor-treated rubber tree. Cross-section of rubber tree stem at the 3rd node of EU2, showing PCZ and IBU has an effect on laticifer development in RRIT251 (left) and RRIT402 (right). The treatments were done with the 3-month-old rubber trees and let the plants further grew for another 3 months to generate EU2. The rubber trees were treated by water (Control) (a–b), 5 µM PCZ (PCZ treatment) (c–d) and 1.2 mM IBU (IBU treatment) (e–f). (Scale bars 200 μm) (upper panel). Relative number of laticifers per one unit area of rubber tree stem treated with PCZ/IBU (lower panel). The treatments had done with the 3-month-old rubber trees and let the plants further grew for another 3 months to generate EU2. The rubber trees were treated by water (control), 5 µM PCZ and 1.2 mM IBU. Data were collected from four randomly chosen areas for each sample. Using t test, the results were considered statistically significant when the P value < 0.05. The letter indicated significant differences

Differential expression of vascular development-related genes upon JA, BR and inhibitor treatments

LBD1 was 1.7-fold up-regulated at 12 h after MeJA treatment of both rubber tree clones, while the expression was maintained to 24 h in RRIT251 but was decreased after 12 h in RRIT402. For the IBU treatment, the fourfold induction was observed at 12 h in RRIT251, in contrast to the down-regulation in RRIT402. Interestingly, the down-regulation of this gene in BL treatment was observed until 24 h in both clones. The expression in PCZ treatment was up-regulated in both clones, in which 3.5-fold sharply increased in RRIT251 (Fig. 6a, b). These evidences suggested that applying exogenous JA could rapidly induce and on the other hand, applying exogenous BR could rapidly suppress LBD1 expression in both rubber tree clones. Expression of LBD1 in RRIT251 and RRIT402 responded differently to JA-inhibitor treatment suggested that endogenous JA level or inhibitor uptake or metabolism rates may be different between these two rubber tree clones. KAN1 was not differently expressed in both rubber tree clones upon MeJA and BL treatments in comparison with un-treated plants. However, treatment with both JA- and BR-inhibitors revealed sharply 25-fold up-regulation with IBU and 13-fold up-regulation with PCZ in RRIT251 at 12 h, while only gradual increase was observed in RRIT402 (Fig. 6c, d). HCA2 expression was not induced at 12 h to 24 h upon MeJA treatment but up-regulated at 3 months in both rubber tree clones. In RRIT251, the 1.5-fold up-regulation was observed at 12 h after treatment with both JA and BR-inhibitors. In contrast, the down-regulation was observed at 12 h after treatment with both JA and BR-inhibitors in RRIT402 (Fig. 6e, f). These evidences suggested that exogenous applying JA could induce the expression of HCA2 after 3 months of treatment in both rubber tree clones which correlate to increased laticifer number.

Fig. 6.

The expression profile of LBD1, KAN1 and HCA2 in stem of RRIT251 and RRIT402 rubber tree clones upon MeJA, BL and inhibitor treatments. The expression analysis of LBD1 (a, b), KAN1 (c, d) and HCA2 (e, f) in stem of RRIT251 (left) and RRIT402 (right) upon treatments was shown. The treatments were done with the 3-month-old rubber trees and plants were allowed to grow further for another 3 months to generate EU2. The rubber trees were treated by water as control, 10 ppm BL, 0.1 M MeJA, 5 µM PCZ and 1.2 mM IBU. The relative expression of target genes was determined by the comparative C_T method (2^−ΔΔCT). The housekeeping reference gene, actin, was used as an internal control for the normalization of the target gene detection levels. Error bars correspond to standard error. Statistical analysis was performed using one-way ANOVA with honestly significant difference (HSD) post hoc test. Significant differences at P value < 0.05. The letter indicated significant differences

MeJA and BL treatment induced HB8 expression at 12 h after treatment in both rubber tree clones, in which BL could up-regulate HB8 expression higher than MeJA treatment. At 24 h after treatment, the HB8 expression was down-regulated or maintained but still higher than in control. Significantly, MeJA treatment up-regulated HB8 expression at 3 months in both rubber tree clones. These evidences suggested that applying exogenous JA and BR could rapidly induce the expression of HB8 in both rubber tree clones, in which JA showed prolong effect. IBU treatment could also induce HB8 expression, which was significantly up-regulated to 11.5-fold in RRIT251 at 12 h then gradually increased to 24 h. The up-regulation to twofold at 12 h then sharply induced to 12-fold at 24 h was observed in RRIT402. PCZ induced HB8 expression to tenfold in RRIT251 and sixfold in RRIT402 at 24 h after treatment (Fig. 7a, b). Expression of VND7 in RRIT251-treated MeJA and BL was down-regulated, in contrast to the up-regulation in the first 12 h after treatment in RRIT402. The expression of VND7 was sharply increased to 4.5-fold in PCZ treatment of RRIT251 at 12 h and to 3.5-fold in RRIT402 at 24 h. IBU induced VND7 expression in RRIT402 to 4.5-fold at 24 h, but suppressed the expression in RRIT251 (Fig. 7c, d). These evidences suggested that applying exogenous JA and BR could rapidly induce the expression of VND7 in RRIT402. BRL3 expression upon MeJA and BL treatment was up-regulated at 12 h after treatment and then down-regulated at 24 h. At 12 h after treatment, BRL3 was higher expressed up to 3.5-fold by MeJA treatment in RRIT402 whereas the effect of BL treatment was not different in both rubber tree clones (Fig. 7e, f). These evidences suggested that applying exogenous JA and BR could induce the expression of BRL3 especially at 12 h after treatment.

Fig. 7.

The expression profile of HB8, VND7 and BRL3 in stem of RRIT251 and RRIT402 rubber tree clones upon MeJA, BL and inhibitor treatments. The expression analysis of HB8 (a, b), VND7 (c, d) and BRL3 (e, f) in stem of RRIT251 (left) and RRIT402 (right) upon treatments was shown. The treatments were done with the 3-month-old rubber trees and plants further allowed to grow for another 3 months to generate EU2. The rubber trees were treated by water as control, 10 ppm BL, 0.1 M MeJA, 5 µM PCZ and 1.2 mM BU. The relative expression of target genes was determined by the comparative C_T method (2^−ΔΔCT). The housekeeping reference gene, actin, was used as an internal control for the normalization of the target gene detection levels. Error bars correspond to standard error. Statistical analysis was performed using one-way ANOVA with honestly significant difference (HSD) post hoc test. Significant differences at P value < 0.05. The letter indicated significant differences

Discussion

Different types and numbers of putative regulatory motifs existed on promoter regions of the six target vascular development-related genes in rubber tree

Vascular differentiation in plants is controlled by both plant hormones and transcription factors (Elo et al. 2009). As xylem differentiation and phloem differentiation are antagonistic (Bonke et al. 2003; Caño-Delgado et al. 2004), therefore, laticifer, that coordinates with phloem, may differentiate antagonistic to xylem. As indicated above, number of laticifers in rubber tree can be increased by JA-treatment, but so far, no specific gene function in laticifer differentiation was unraveled, thus this research project focused on the six vascular development-related genes, based on the previously reported function in phloem- and xylem-differentiation. The target vascular development-related genes were the phloem development-related genes; LBD1, KAN1, HCA2; and the xylem development-related genes; HB8, VND7, BRL3. The bioinformatics result suggested that there are 1 member of LBD1, and 2 members of KAN1, HCA2, HB8, VND7 and BRL3 on rubber tree genome.

As the rubber tree clones RRIM251 and RRIM402 used in this research still have no genome database, the available putative hormonal-responsive regulatory motifs in the promoter region of candidate genes from the genome sequences of rubber tree clone RRIM600 and BMP24 were considered. The number of putative hormonal-responsive regulatory motifs in the promoter region of all candidate gene members from genome sequences of both rubber tree clones revealed that the putative auxin-responsive elements are the most available motif, while JA-responsive elements and cytokinin-responsive elements are pretty rare in every genes in both rubber tree varieties. Different gene members contain different number of hormonal-responsive regulatory motifs in their promoter regions. All candidate gene member promoters contain less number of JA-responsive elements than BR-responsive elements. The putative auxin-responsive elements are mostly available, while JA-responsive elements and cytokinin-responsive elements are pretty rare in every genes, in both RRIM600 and BMP24. Auxin is a major regulator of plant growth and development. It has been reported that a local auxin-signaling maximum is required for transcription factors to specify a stem-cell organizer (Smetana et al. 2019). Auxin can induce phloem differentiation, while brassinosteroid can induce xylem differentiation in plants (Iwasaki and Shibaoka 1991). Auxin/indole-3-acetic acid (AUX/IAA) proteins are key players in auxin signaling pathway which involved in BR responses. The relationship of BR and auxin has been well documented in plant growth and developmental processes as well as involved in abiotic stress tolerance. A comprehensive microarray analysis revealed that most of the known auxin-inducible genes and various homologues of such genes are also induced by BR application in tobacco, suggesting that a marked overlap exists between the BR and auxin signaling pathways. Since auxin is a major growth factor that is required for the proliferative growth of plant cells, the relationship between the effects of BR and auxin on cell proliferation is an intriguing question. The mechanism that supports the short-term promotion of cell division by BR is unique and probably distinct from that regulated by the balance of auxin and cytokinin levels. BR quickly promotes cell division mediated by the enhanced accumulation of cell cycle-related genes (Miyazawa et al. 2003). It has long been recognized that JA/auxin crosstalk plays an important role in lateral root development under normal growth conditions. However, the involvement of auxin in the interplay between JA and BR in vascular development has not been reported (Saini et al. 2015; Peres et al. 2019). As the 6 candidate genes in this research were selected from the reported function that may relate to vascular development, thus the dynamic, differential distribution of auxin within plant tissues may control an impressive variety of developmental processes, which tailor plant growth and morphology to environmental conditions. Various environmental and endogenous signals can be integrated into changes in auxin distribution through their effects on local auxin biosynthesis and intercellular auxin transport. Different rubber tree clones contain different number of hormonal-responsive regulatory motifs in their promoter regions. It should be noted that all candidate gene promoters contain both JA- and BR-responsive elements, although less number of JA-responsive elements than BR-responsive elements was observed, thus JA and BR have an effect on  their gene expression profiles.

Correlation of vascular development-related genes in mature xylem and phloem tissues of high latex-yield and high wood-yield rubber tree clones suggested tissue specific expression

LBD1 has been reported to be highly expressed in phloem and cambium zone and regulated secondary development in poplar tree (Yordanov et al. 2010). Both rubber tree clones in this study also showed that  LBD1 was slightly higher expressed in phloem than in xylem tissue. LBD1 is preferentially highly expressed in mature phloem tissue of the high wood yield rubber tree clone, RRIT402, thus it may play role to control phloem differentiation in this clone. Interestingly, LBD1 was up-regulated in response to MeJA treatment in both rubber tree clones which was consistent with  the increasing laticifer number in both clones, hence, supported the possible role of LBD1 in laticifer differentiation. The co-occurance of gene induction by MeJA until 3 months in RRIT251 strongly supported to its involvement in laticifer formation. KAN1 expression was significantly high in phloem tissue of high latex yield clone, RRIT251. This finding was supported by the previous report which revealed the expression of KAN1 in early development and later became restricted to phloem (Nieminen et al. 2015). KAN1, KAN2, KAN3 and KAN4 belong to GARP family transcription factors, and act redundantly in the early development to promote abaxial cells that are destined to be future phloem (Emery et al. 2003). KAN1 acts antagonistic to HD-Zip classIII by inhibiting PIN1 expression, and subsequently negatively regulate auxin movement. In this context, KAN1 may play important role to control phloem differentiation as well as laticifer development in high latex yield rubber tree clone, RRIT251. However, upon exogenous JA treatment, KAN1 expression did not show any significant response after treatment as compared with control, thus KAN1 might be involved in laticifer development at certain growth stage,and highest expression was observed in phloem tissue from mature branches of RRIT251. HCA2 promotes interfascicular cambium at the very first stage of stem development thereby provide adjacent zones of dividing and more-quiescent cells, which forms a foundation for further radial growth (Guo et al. 2009). Preferentially high expressed HCA2 was found in mature xylem tissue of high latex-yield rubber tree clone, RRIT251, thus HCA2 may also play role to control xylem differentiation in this clone. Moreover, HCA2 expression was turned to up-regulation at 3 months after MeJA treatment, therefore JA-signaling might trigger the HCA2 expression and consequently promote interfascicular cambium and laticifer differentiation in phloem tissue in younger stem.

HB8 is a member of HD-Zip class III transcription factor family which act downstream of auxin to promote vascular meristem (Donner et al. 2009). High expression of HB8 promoted differentiation of vascular cambium to be xylem and in addition, homolog of HB8 in Zinnia, ZeHB10, was found to be induced by brassinosteroids (Ohashi-Ito et al. 2002). These evidences supported our present study in which HB8 was preferentially expressed in mature xylem tissue of both rubber tree clones, whereas slightly low expression level was found in phloem tissue. HB8 may be used as a marker to indicate mature xylem tissue of rubber tree. HB8 was highly expressed in xylem tissue of the high wood yield rubber tree clone, RRIT402, thus HB8 may play role to promote xylem differentiation in this clone. VND7 is the master regulator of xylem differentiation. During xylem development, VND7 triggers the cells to differentiate into xylem and then regulates many genes downstream for completely xylem formation (Yamaguchi et al. 2011). Both rubber tree clones in this study also showed consistent finding in which VND7 was higher expressed in xylem than in phloem tissue. Interestingly, the high wood yield rubber tree clone, RRIT402, exhibited preferential highest expression of VND7 in xylem tissue. VND7 may play important role to promote xylem differentiation in high wood yield rubber tree clone, RRIT402. BRL3 was reported to express mainly in vasculature and the expression pattern of BRL3 was BR dose-dependent (Caño-Delgado et al. 2004). In Arabidopsis root, at concentration of endogenous BR, the expression of BRL3 was increased and extended toward the transition zone between undifferentiated and differentiated protophloem (Salazar-Henao et al. 2016). In this present study, the expression of BRL3 can be observed in both mature xylem and phloem tissue of both rubber tree clones. Taken together, these evidences suggested that the expression of BRL3 can be observed in both xylem and phloem tissue depend on endogenous BR in each clone. Interestingly, the high latex yield rubber tree clone, RRIT251, exhibited preferential highest expression of BRL3 in xylem tissue. BRL3 may play important role to control xylem differentiation in RRIT251.

JA, BR and inhibitor treatments revealed opposite effects on number of laticifers in high latex-yield and high wood-yield rubber tree clones

The average number of laticifers was significantly higher in MeJA treatment as compared with BL treatment and control. The laticifer numbers in BL treatment was lower as compared with control. This observation suggested the effect of exogenous BR supplied to the rubber trees. BRs interact with various phytohormones in regulating wide range of physiological and developmental processes in plants. The interaction of BR and JA plays crucial roles in plant development, in both biotic and abiotic responses. BR antagonizes JA-signaling pathway primarily reported in root growth and development in rice. Exogenous foliar spraying with JA leads to strong down-regulation of BR biosynthesis and signaling genes indicating antagonistic interaction between BR and JA in the rice roots (Saini et al. 2015). Recently BR-JA antagonistic interactions in regulating rice innate immunity during infection with the root-knot nematode and the defense against herbivory has been demonstrated (Nahar et al. 2013; Peres et al. 2019). JA and BR were reported to play opposite function, therefore, the decreasing of laticifer numbers in BL treatment might be the effect of exogenous BR that suppressed JA biosynthesis, and consequently decreased the laticifer development. Correspondingly to the effect of BL treatment, the IBU treatment, that can inhibit the JA biosynthesis, also showed decrease of the laticifer numbers. This finding was supported by PCZ treatment, which can inhibit BR biosynthesis, also induced the number of laticifers. These results suggested the homeostasis of endogenous JA/BR biosynthesis which control balance of laticifer differentiation in rubber tree. Taken together, the results suggested that the application of exogenous JA can promote laticifer differentiation while exogenous BR interfere differentiation of these vessels. Thus, suppression of BR biosynthesis can lead to more laticifer differentiation, while inhibition of JA biosynthesis can interfere laticifer differentiation. However, the mechanism of cross-talk between JA and BR on this vascular tissue development has been poorly elucidated, hence needed to further investigate more in order to better understanding.

Differential expression of vascular development-related genes were exhibited upon JA, BR and inhibitor treatments

The effect of JA, BR and inhibitor treatment was investigated in the high latex-yield clone compared with the high wood-yield clone. Recent studies provide evidence that conjugate jasmonoyl isoleucine (JA-Ile) initiate signal transduction upon formation of co-receptor complex of JA signaling perception, COI1-JA-JAZ ternary complex. In the presence of JA, JAZ proteins interact with COI1 and are ubiquitinated and subsequently degraded by the 26S proteasome. The removal of JAZ proteins releases MYC2, enabling the transcriptional activator to regulate the expression of early jasmonate-responsive genes (Thines et al. 2007). Trichostatin A (TSA), an inhibitor of histone deacetylation that caused inhibition of cambium proliferation, was reported to induce secondary laticifers in rubber tree (Zhang and Tian 2016). Histone deacetylase 6 (HDA6) was found to interact with JAZ protein that is known to be a repressor of JA signaling (Zhu et al. 2011). A poplar mutant over-expressing PtaLBD1 gene showed enhancing of secondary phloem growth in stem and up-regulating APL (Altered Phloem Development) gene, that is a master regulator of phloem differentiation, but down-regulated ARK1 and ARK2, transcription factors that are essential for cambium proliferation (Yordanov et al. 2010). In this study, LBD1 was up-regulated in response to MeJA treatment in both rubber tree clones and the number of laticifers was also increased. These evidences suggested that the up-regulated LBD1 upon MeJA treatment might negatively regulate genes involved in cambium proliferation that can be interacted with JAZ to turn on the JA signaling and consequently resulted in promoting laticifer differentiation. This finding was supported by the up-regulated XM_002315243.1 gene (LOB domain protein) upon JA- and linolenic acid-treatment of rubber tree clone RRIM600 study by Loh et al. (2016). IBU treatment showed down-regulation of LBD1 expression in RRIT402, but sharply up-regulation in RRIT251 clone. This finding suggested that endogenous JA level may be clone specific which is different between these two rubber tree clones. IBU suppressed endogenous JA-biosynthesis so that endogenous BR might be increased. BR and JA treatment had been reported to be opposite in rice roots, which means supplying BR to the rice had an effect on suppressing JA-biosynthesis (Nahar et al. 2013). In this present study, LBD1 expression was down-regulated upon BL treatment but up-regulated upon suppression of BR-biosynthesis by PCZ treatment, in both rubber tree clones. This finding was consistent with the higher number of laticifers in PCZ treatment compared to BL treatment. The results suggested that LBD1 might be positively related to laticifer development through the early JA-signaling that inhibit cell proliferation (Zhang et al. 2015), resulting in induction of laticifer differentiation. RRIT251 might have mechanism to maintain the differentiation of laticifers, therefore it is the high latex-yield clone.

Ectopic KAN1 induction in Arabidopsis revealed that KAN1 is a transcription repressor of genes related to auxin biosynthesis, abscisic acid and BR. Genes involved in BR that were repressed by KAN1 are DWARF4 (DWF4) encoding a P450 enzyme involved in BR synthesis, PHYB ACTIVATION TAGGED SUPPRESSOR1 (BAS1) encoding a P450 enzyme involved in BR catabolism, BRASSICOSTEROID-ENHANCED EXPRESSION1 (BEE1) encoding a basic helix-loop-helix transcription factor, and BRH1 encoding a ring-finger protein (Xie et al. 2015). Therefore, KAN1 expression might not respond to BL treatment as a result of that KAN1 acts upstream of BR-related genes. Besides, the relation between JA and KAN1 expression has not been elucidated yet. KAN1 had been reported to have opposing roles to HD-ZIPIII genes, and function in vascular patterning (Ilegems et al. 2010). KAN1 is expressed in phloem and HD-ZIPIII is expressed at cambium and developing xylem; nevertheless, KAN1 loss-of-function mutant develop phloem cells, indicating that KAN1 is not required for phloem identity (Eshed et al. 2001). In this present study, higher KAN1 expression in mature phloem tissue of RRIT251 compared to RRIT402, as well as KAN1 expression did not respond to exogenous JA, supported that KAN1 might involve in laticifer development at certain growth stage. JA treatment induces interfascicular cambium, which is meristematic tissues undergoing secondary growth, through JA-signaling (Sehr et al. 2010). HCA2, Dof 5.6 transcription factor, had been reported to promote interfascicular cambium in stem of Arabidopsis, besides, gain-of-function hca2 mutant influenced phloem formation as a result of up-regulating phloem-specific genes and down-regulating xylem-specific genes (Guo et al. 2009). In this present study, HCA2 expression was down-regulated to 24 h and turned to up-regulation at 3 months after MeJA treatment, whereas it was slightly down-regulated upon BL treatment. Taken together, JA-signaling might trigger the HCA2 expression and consequently promote interfascicular cambium and laticifer differentiation in phloem tissue.

AtHB8 was a positive regulator of cambium proliferation and ectopic expression of AtHB8 was found in cells that differentiated into tracheary elements in wounded tobacco transgenic plants and increased xylem tissue in Arabidopsis (Baima et al. 2001). The expression of ZeHB10, which is a homolog of ATHB8, in Zinnia elegans can be induced by BR (Ohashi-Ito et al. 2002). Upon BL treatment, HbHB8 was up-regulated at the first 12 h-after treatment in both rubber tree clones. HbHB8 responded to BL treatment by maintaining the cambium meristem, and promoting the differentiation into secondary xylem located in the adaxial side, therefore, the number of laticifers in phloem tissue of the abaxial side were decreased. JA had been reported to positively regulate cambium cells division in Arabidopsis stem, and influences secondary growth (Sehr et al. 2010). Thus, HbHB8 gene that specifically expressed in the cambium might respond to JA treatment by maintaining the cambium meristem, and might not relate to laticifer differentiation. This finding is consistent with the higher HB8 expression in mature xylem tissue of RRIT402 as compared to RRIT251. The rubber trees treated with IBU and PCZ also showed up-regulated HB8 expression in both two clones. Since, HB8 was reported to be directly controlled by auxin responsive transcription factor, MONOPTEROS (MP) (Donner et al. 2009); therefore, the expression of HB8 upon inhibitor treatments might involve in several mechanisms.

VND7 gene is a key regulator of xylem differentiation (Endo et al. 2015; Kubo et al. 2005). The effect of three plant hormones (auxin, cytokinin and BR) that regulate xylem development on VND7 expression in Arabidopsis were reported by Kubo et al. (2005). Therefore, the VND7 expression in this present study might not involve in the effect of BR and JA on laticifer development. However, RRIT402, which is a high wood yield clone, showed slightly higher expression of VND7 relative to RRIT251 at 12 h after treatments of BL and MeJA, and then it was down-regulated at 24 h after treatments in both clones. This finding is consistent with the higher VND7 expression in mature xylem tissue of RRIT402 compared to RRIT251, thus VND7 may involve in xylem formation in RRIT402. Since VND7 response to various plant hormones, auxin, cytokinin and BR, the expression of VND7 upon inhibitor treatments have to be further elucidated. BRL3 (BRI1-like3) is a key receptor for BR-signaling (Li and Chory 1997). Applying exogenous JA and BR could induce the expression of BRL3 especially at 12 h after treatment in both rubber tree clones. BRL3 was higher expressed by MeJA treatment in RRIT402. BRL3 was reported to express mainly in vasculature (Caño-Delgado et al. 2004) and the expression pattern of BRL3 was BR dose-dependent. Low level exogenous BR (< 0.04 nM) increased expression of BRL3 in protophloem but high level of BR repressed BRL3 expression in protophloem and spatially shifted toward stele (Salazar-Henao et al. 2016), and repressed phloem differentiation (González-García et al. 2011). In this present study, 10 ppm BL was used to treat rubber trees and the expression of BRL3 were up-regulated to 12 h, then continued down-regulation to 3 months in both clones. This observation is consistent with previous report that the expression pattern of BRL3 is BR dose-dependent, high level of exogenous BR could repress BRL3 expression since 12 h after BL treatment, and the suppression could prolong to 3 months. Applying exogenous JA could induce the expression of BRL3 better than BR treatment.

In conclusion, the results suggested the tissue specific expression of the six vascular development-related genes in mature xylem and phloem tissues of high latex-yield and high wood-yield rubber tree clones. The different transcript abundance of each gene in xylem and phloem tissues obtained from mature branches of RRIT251 and RRIT402 suggested the effect of clone specific as well as homeostasis of endogenous JA/BR in each tissue of each rubber tree clone at different growth stages. In mature vascular tissues, LBD1 may play role  in controlling phloem differentiation in the high wood-yield rubber tree clone, and play possible role in laticifer differentiation, while KAN1 play important role in controlling phloem differentiation in high latex-yield rubber tree clone, and laticifer development at certain growth stage. HCA2 may also control xylem differentiation in high latex-yield rubber tree clone, and laticifer differentiation in phloem tissue. Exogenous application of JA can promote more laticifer numbers, while BR showed opposite effect on reducing laticifer numbers. LBD1, KAN1, and HCA2 positively relate to laticifer differentiation through the early JA signaling in high-latex yield rubber tree clone. HB8 and VND7 promote xylem differentiation in high wood yield rubber tree clone, while BRL3 control xylem differentiation in high latex-yield rubber tree clone. This is the first report on the opposite effect between BR and JA on laticifer differentiation in rubber tree. To understand more complexity underlying this laticifer differentiation system, gene function study experiment should be a better tool for direct exploring this mechanism in rubber tree. Moreover, there are a number of uncharacterized transcription factors especially in woody plants, hence, further characterization of those genes is necessary to unravel their role in laticifer differentiation in rubber tree. The opposite effects of BR and JA on vascular system pave the way to manipulate rubber tree plantation for high latex-yield and high wood-yield.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

BL

Brassinolide

BR

Brassinosteroid

BRL3

BRASSINOSTEROID INSENSITIVE 1-like3

HB8

HOMEOBOX8

HCA2

HIGH CAMBIAL ACTIVITY

IBU

Ibuprofen

JA

Jasmonic acid

KAN1

KANADI1

LBD1

Lateral organ boundaries domain-containing protein 1-like

LOB

Lateral organ boundaries

MeJA

Methyl-jasmonate

PCZ

Propiconazole

VND7

VASCULAR-RELATED NAC-DOMAIN7

Author contributions

JN conceived and designed of the research. PA conducted main parts of the research and PS-L performed some bioinformatics. KN contributed rubber tree from the Rubber Research Institute of Thailand. JN, PK and UV provided technical support, analyzed and discussed the results. JN and PA wrote the manuscript. All authors read and approved the manuscript.

Conflict of interest

The authors declare that they have no conflict of interest.