The Arabidopsis LRR-RLK,PXC1, is a regulator of secondary wall formation correlated with the TDIF-PXY/TDR-WOX4 signaling pathway

Background

Although a number of leucine-rich repeat receptor-like kinase-encoding genes (LRR-RLKs) have been identified in plants, a functional role has been determined for only a few. Recent studies have demonstrated that an LRR-RLK, PXY/TDR, is important for the process of secondary vascular development. Other studies have indicated that PXY/TDR is unlikely to be the sole LRR-RLK involved in this complex process.

Results

In this study,in silicoanalyses led to the identification of threeArabidopsis LRR-RLKgenes (PXY-correlated; PXC1, 2, 3) with transcript accumulation profiles that correlated strongly with several key regulators of vascular development, includingPXY/TDR,HB-8,REV, andCLE41. Expression profiling using qPCR and promoter:reporter lines indicated that all threePXCgenes are associated with the vasculature. One in particular,PXC1(At2g36570),had a strong correlation withPXY/TDR. Shiftingpxc1mutants from long-days to short-days showed that loss of the gene led to a dramatic reduction in secondary wall formation in xylem fibers. Transcript analysis of mutants for a variety of secondary cell wall-associated genes, includingPXY/TDRindicated that the pathways mediated by PXC1 connect with those mediated by the TDIF-PXY/TDR-WOX4 system.

Conclusions

数据表明,LRR-RLK PXC1 involved in secondary cell wall formation in xylem fibers. Whereas further study is needed to identify the ligands and mode of action of the PXC1 protein, it is clear from this work that similarly to the shoot apical meristem (SAM), secondary vascular development requires contributions from a number of LRR-RLKs.

LRR-RLKs (leucine-rich repeat receptor-like kinases) comprise the largest group within the RLK (receptor-like kinase) superfamily in plants. Among the more than 400RLKgenes identified in theArabidopsisgenome, over half areLRR-RLKs[1]。TheLRR-RLKscan be grouped into 13 subfamilies (I to XIII) [2]。Although the functions of most LRR-RLKs remain undiscovered, it has been suggested that plant LRR-RLKs can be divided into two broad functional categories [2]。That is, some appear to function in plant growth and developmental processes such as morphogenesis, organogenesis and hormone signaling, while others appear to be primarily involved in mediating responses to biotic or abiotic stresses and therefore can be said to be defense-related. Some LRR-RLKs have been demonstrated to possess dual functions, either through signaling pathway cross-talk or due to their ability to recognize multiple ligands [3]。Well known examples of LRR-RLKs involved in the regulation of plant growth and development are the CLV3 (CLAVATA3)-CLV1 (CLAVATA1)-WUS (WUSCHEL) signaling system in the SAM (Shoot Apical Meristem) and the similar, CLE40 (CLAVATA3/ESR40)-CLV1-ACR4 (CRINKLY4)-WOX5 (WUSCHEL RELATED HOMEOBOX 5) signaling system in the RAM (Root Apical Meristem) [4–9]。In the SAM, the CLV1 LRR-RLK is essential for maintaining a balance between stem cell division and differentiation, and therefore, growth control in the shoot [6,10,11]。Loss of CLV1 leads to the accumulation of undifferentiated cells in the SAM. Three other LRR-RLKs that operate in the SAM, the BAM (BARELY ANY MERISTEM) proteins, form a monophyletic group with CLV1, are also involved in maintaining meristem function. Although CLV1 and the BAMs both operate in maintaining SAM function, their expression profiles and functions differ. In contrast to CLV1, the loss of BAM function leads to a reduction in the number of undifferentiated cells [12]。Clearly, LRR-RLKs fulfill multiple roles in complex processes such as meristem function.

LRR-RLKs have now been shown to be involved in all three major plant meristems, the SAM and RAM, and the vascular cambium, that produces cells for secondary vascular development. InArabidopsis,信号系统组成的一个小CLE peptide, the TDIF (TRACHEARY ELEMENT DIFFERENTIATION INHIBITORY FACTOR), and its receptor PXY/TDR (TDIF RECEPTOR/ PHLOEM INTERCALATED WITH XYLEM) regulates the behavior of vascular stem cells [13]。Genetic analyses showed that at least two pathways diverge early in TDIF-PXY/TDR signaling and the WOX4 (WUSCHEL RELATED HOMEOBOX 4), which belongs to the WUS subclade in the WOX family [14］, is required for promoting the proliferation of procambial/cambial stem cells but not for repressing their commitment to xylem differentiation in response to the TDIF signal [15]。Correct spatial separation of the expression of the genes encoding PXY/TDR and TDIF, is essential for generating the spatial cues necessary for ordered secondary vascular development[16]。

As the only well described pathway so far, the TDIF-PXY/TDR signal transduction pathway has been suggested to be required both very early in vascular development to orientate the polarity of the vascular bundle, and continuously throughout development to regulate the process [16]。Ectopic expression of TDIF-related genes results in pleiotropic phenotypes including a bushy appearance with small leaves [17,18]。Recently, TDIF and CLE42 peptide were found to have anin vivoactivity to enhance axillary bud formation and there are indications that PXY/TDR is involved in this process [19]。Together, these results indicate that the TDIF ligand-PXY/TDR signal transduction pathway is an important regulator of multiple developmental processes [19]。

Given the size of the LRR-RLK gene family and the evidence of multiple active LRR-RLKs in the SAM, it is likely that other multiple LRR-RLKs are involved in the complex process of secondary vascular development. In this work, we aimed to identify LRR-RLKs other than PXY/TDR that contribute to secondary vascular development through interactions, direct or indirect, with the TDIF-PXY/TDR pathway. Given the importance of the localization of gene expression for PXY/TDR function, we initially performed anin silicoco-expression and functional clustering analyses. ThreeLRR-RLKs(At2g36570, At5g01890 and At2g41820) were identified that had similar transcript profiles toPXY/TDR. We named themPXCforPXY/TDR-correlatedgenes. Evidence from loss-of-function and gain-of-function analyses showed thatPXC1in particular plays a TDIF-PXY/TDR associated role in the process of secondary cell formation in fiber cells.

Co-expression profiling and functional clustering analyses identified threeAtLRR-RLKs与PXY/TDR

In order to develop our understanding of the functions of AtLRR-RLKs in vascular development, a hierarchical cluster analyses using the microarray data in the Genevestigator database was performed for allAtLRR-RLKs[20]。The dendrogram in Additional file1was used to assess transcript profile similarities between the genes. Six out of the 7 genes that clustered withPXY/TDRexhibited preferential expression in the vasculature, with the highest transcript levels occuring in the stem, apex and floral organs (Figure1A,1B). Of the 7,MOL1(At5g51350, MORE LATERAL GROWTH1),RUL1(At5g05160, REDUCED IN LATERAL GROWTH1) andVH1(At2g01950, VASCULAR HIGHWAY1) have already been reported for their vascular functions [21,22]。The other three,At2g41820,At5g01890andAt2g36570are largely uncharacterised. We named the genes asPXC PXY/TDR-correlatedgenes,PXC1(At2g36570),PXC2(At5g01890), andPXC3(At2g41820). TAIR database information of these three genes indicated that they may be broadly functional. For example,PXC1displayed a decreased expression level in theArabidopsisleaves treated by salt [23］andPXC2appear to be down-regulated inArabidopsisseedlings under anoxia [24]。ForPXC3, its transcription was dramatically repressed inArabidopsiscell suspensions upon salicylic acid treatment [25]。In an investigation into the roles of LRR-RLKs inArabidopsisroot development, germination of a line mutated inPXC1, N634974耐盐(200米M NaCl) and osmotic stress. Apxc2mutant allele was found to be sensitive to darkness and resistant to osmotic stress treatment (400 mM mannitol) [1]。在一起,结果显示的作用Arabidopsis PXCgenes in defense and other stress-related responses. A putative soybean ortholog of PXC1,GmLRK1gene has been studied in some detail [26]。MutatingGmLRK1led to reduced lignification in leaf cells and defective leaf cell elongation. The authors hypothesized that GmLRK1is is involved in the regulation of cell expansion by influencing the development of cell wall architecture [26]。

In order to further explore the possibility of a role for the threePXCgenes in vasculature development, we developed a putative gene co-regulation network using the ATTED-II suite of programs (based on publicly available microarray data of 58 experiments, 1388 slides collected by AtGenExpress) [27,28］(Additional file2). The analyses led to the identification of a putative relationship betweenPXY/TDRandPXC1(Figure1C). The homeobox transcription factor,AtHB8(HOMEOBOX GENE 8), that is regarded as a procambium and protoxylem cell identity marker, appeared to be similarly associated withPXY/TDRandPXC1[29]。PXC1also correlated strongly with genes encoding enzymes involved in the cell expansion process (Figure1C). ThePXC2 gene correlated withPXC3 and withREVOLUTA(REVAt5g60690).AtHB8andREVare Class III HD-ZIP transcriptional regulators and both play important roles in vascular differentiation [30]。CLE41, that encodes the peptide ligand for the PXY/TDR receptor [31］, was located betweenPXC1andPXC2in the coexpression network (Figure1C).The bioinformatic data, therefore, indicated connections between PXY/TDR, and the three PXC proteins in vascular development.

Expression patterns ofPXYand threePXCgenes in vascular tissues

In order to further explore the associations betweenPXC1/2/3andPXY/TDR, we examined their respective expression patterns using a native promoter-driven GUS (β-glucuronidase) reporter system.PXY/TDRhas been reported to be expressed in the vasculature of a variety of organs including leaves, roots and the stem [16]。Its expression has been shown to be confined to the procambial cells in the developing vascular bundles [16]。The pattern of GUS activity observed in thepPXY::GUSwas similar to the pattern oberseved in thepPXC1::GUS, pPXC2::GUS,andpPXC3::GUSlines. Specifically, GUS was observed primarily in the vascular strands in cotyledons, the shoot apex, hypocotyls, roots and leaves (Figure2A-2D,2F). In the inflorescence stems of thepPXY::GUSline, GUS was observed in the protoxylem, procambial cells and in the interfascicular cambial regions (Figure2E). In thepPXC1::GUSline,GUS staining was observed primarily in the developing xylem of both inflorescence stem and hypocotyl (Figure2E, Figure3B).PXC1expression overlapped withPXYexpression in the basal stem, except that the staining was very faint in protoxylem (Figure2E). InpPXC2::GUSplants, GUS staining in the inflorescence stem was primarily observed in the differentiating vessel cells (Figure2E, Figure3A,3C). The GUS staining pattern in the secondary vasculature of thepPXC3::GUSplants was similar to that ofpPXY::GUS, except that inPXC3::GUS, no GUS was not expressed in the interfascicular region (Figure2E). The GUS staining of the four lines diverged in the roots. GUS was observed in the root tip in thepPXC2::GUSline, close to the quiescent center in thepPXC1::GUSandpPXC3::GUS行和elongation zone in thepPXY::GUSline (Figure2D) [13,15,16]。In contrast to the similarities in the GUS staining patterns in the vascular tissues, there was little similarity between the GUS outside these tissues (i.e. in floral tissues) (Additional file3).

Expression levels ofPXC1in vascular tissues

Because the strongest predicted links were betweenPXC1andPXY/TDR, we concentrated our efforts on this gene. To further investigate the function ofPXC1in vascular development, transcript levels ofPXC1were determined in parallel with seven known regulators of vascular development in different plant tissues including the leaf lamina (as a control), petiole, young inflorescence stem, old inflorescence stem and hypocotyl. Our data showed that the transcripts forPXYwere most abundant in the xylem fraction of hypocotyls, with levels increasing in older inflorescence stems (Additional file4C). Expression patterns forHB8,HB15(HOMEOBOX GENE 15) andNST3(NAC TRANSCRIPTION FACTOR3) were similar to those observed forPXY(Additional file4A, B and D).HB8andHB15are both recognized as molecular markers of procambial cells [32,33］andNST3encodes a known regulator of secondary cell wall formation in xylem fibers [34,35]。Interestingly, the qRT-PCR data also indicated a similar expression pattern betweenPXC1andWOX4, with the highest level of transcript observed in young stem (Additional file4G and H).CLE41andCLE44both encode the B-type CLE peptides that act as the ligands for the TDIF-PXY/TDR signal transduction pathway [31]。As previously reported, theCLE41transcript was found to be mainly associated with the phloem fraction (Additional file4E) [13]。This fraction may, however, include some cambium cells as a result of the sample collection (peeling) method used [13]。CLE44transcript appeared more evenly distributed among the tissue fractions, with a significantly lower level of transcripts in the xylem tissue (Additional file4F), which is consistent with the recent report of phloem-specific expression ofCLE44::GUS[36]。To summarize the expression patterns described above, a heat map was generated using the qPCR data. Strikingly, the heat map highlighted the similarities betweenPXC1andWOX4in vascular tissues (Figure4).

Loss ofPXC1function suppresses secondary cell wall formation in xylem fibers

ThePXC1gene encodes a predicted protein containing a putative kinase domain and an extracellular domain with 21 leucine-rich repeats (eLRR). The predicted PXC1 belongs to the LRR-RLK subfamily III [2,37］along with CLV1 (27% amino acid identity) and PXY/TDR (28% amino acid identity). Protein motif search in the Pfam and SMART databases predicted that the kinase domain of PXC1 resembles that of the animal receptor tyrosine kinase domain, which is an unusual characteristic for a plant RLK [26]。Phosphorylation of the soybean homologue of PXC1 (GmLRK1) has been demonstrated to be induced by plant protein extracts, which suggested that some plant proteins may interact with GmLRK1 and phosphorylate itin vivo[26]。Based on the high degree of sequence divergence between PXC1 and other subfamilies of LRR-RLKs, no firm prediction can be made as to its functional role.

Three T-DNA insertion lines were identified forPXC1in the NASC (the European Arabidopsis Stock Center) mutant collection, including two SALK lines and one WiscDsLox T-DNA line. These three alleles were designated aspxc1-1 (SALK_134974),pxc1-2 (SALK_134975), andpxc1-3 (WiscDsLox470G6), respectively. Bothpxc1-1andpxc1-3contain insertions in the coding sequence, whilepxc1-2has a T-DNA inserted in the 3'-UTR (Figure5A). Results of qPCR analyses using a gene specific primer set located in the 3’-UTR indicated that the level ofPXC1mRNA in three mutants waspxc1-1>pxc1-2>pxc1-3(Figure5B).pxc1-3displayed only background level ofPXC1expression, suggesting that this allele might represent a null mutant (Figure5B). Several aspects of plant morphology were affected by the mutation ofPXC1. The inflorescence stems of thepxc1mutants were taller than those of the wild-type, with the phenotype most pronounced in thepxc1-3line (Additional file5). By contrast, the inflorescence stem ofpxymutant plants were shorter than the wild-type [16]。When grown under long-day conditions,pxc1mutants did not show significant difference from the wild-type in terms of cellular morphology as seen from transverse sections of the inflorescence stem (data not shown). Secondary vascular development is enhanced in the inflorescence stems ofArabidopsiswhen plants are grown under long day conditions (16/8 day/night) and transferred to short-day conditions (8/16 day/night) immediately after bolting. In both wild-type andpxc1-2plants, the inflorescence stem grew vertically from this point. In contrast, the stems ofpxc1-1andpxc1-3plants were unable to support the weight of the continued upright growth of the stem (Figure5C). Ligin staining in cross sections taken from the base ofpxc1-1,pxc1-3and wild-type inflorescent stems indicated that the two mutant lines were defective in vascular lignification (Figure5D-L), which then provided an explanation to the inability of the inflorescence stem to support an upright growth. Tissue polarity in thepxc1mutants appeared to be retained, as opposed to the polarity phenotype in thepxymutant (Figure5G-J). Secondary cell wall thicking in fiber cells was also considerably reduced inpxc1-1plants and absent frompxc1-3plants (Figure5I, 5L), indicating a reduced capacity for secondary cell wall synthesis and lignifications in thepxc1mutants (Figure5H, 5K). The vascular bundles of the long day grown plants and those shifted from long to short days wer compared by close-ups in Figure6. Interestingly,PXC1has been identified in the repertoire of genes regulated bySND2, which is an indirect target of a principal regulator of fiber secondary cell wall formation, SND1 [38,39]。Overexpression ofSND2produced a fiber cell-specific increase in secondary cell wall thickness inArabidopsisstems andPXC1was slightly up-regulated in this transgenic line [38]。Thus, the reduced cell wall thickness in the interfascicular fiber cell ofpxc1-3andpxc1-1mutants indicated that PXC1 is likely playing a role in secondary cell wall formation of fiber cells.

The interactions betweenPXC1andTDIF-PXY/TDRsignaling pathway

qPCR analysis ofPXY,PXC1,CLE41,CLE44andWOX4expression was analyzed in thepxy/tdr,pxc1,cle41,cle44andwox4mutants. Transcript abundances ofPXYandCLE41were not dramatically affected in thepxy/tdr,pxc1andwox4knockout lines (Figure7A). The transcript level ofPXC1, WOX4andCLE44were significantly increased inpxy,pxc1andwox4mutants compared to the wild-type (Figure7A). The elevated expression level ofWOX4in thepxybackground was unexpected since thatWOX4is a key downstream target of the TDIF-PXY/TDR signaling pathway [13]。This result indicated a possibility that PXY might not be the only receptor acting upstream of WOX4. Meanwhile, instead of young seedlings [13］, 5-week-old hypocotyls, which contain more secondary growth, were used in this study and the implication of PXY/TDR in xylem development has not been investigated in details. The dramatic increase inPXC1transcripts occurred in thepxyandwox4knockout mutants indicated that either the PXY-WOX4 pathway negatively regulate the expression ofPXC1, or thatPXC1is upregulated in these mutants as a compensatory measure (Figure7A). These data also suggested that the signal transduction pathways mediated by PXY and PXC1 are at least partly overlapping. However, it is hard to specify the position of PXC1 relative to the TDIF/PXY/WOX4 pathway because we could not judge which alterations in gene expression resulting from forward responses or feedback responses.

As opposed to the expression profiles of plants grown under long-day conditions, the short-day shift experiment induced a strong decrease inPXY/TDRandWOX4transcript levels in the inflorescence stems of thepxc1mutants, particularly in thepxc1-3line (Figure7B). The reduced capacity to produce an interfascicular cambium in thepxc1mutants may explain the reduction inPXY/TDRandWOX4transcript levels, or vice versa, the reduced expression ofPXY/TDRandWOX4may be the underlying cause of the lack of interfascicular cambium. Future research will attempt to answer this question. It has been reported that the exogenous application of the TDIF/CLE41/44 peptide ligand resulted in an increase in the transcript levels ofHB8,HB15,WOX4andPXYgenes [40]。We similarly over-expressedCLE44inArabidopsisand analyzed its effects onPXC1transcript levels. ThreeCLE44oxlines exhibiting from 65 to 85-fold increases inCLE44transcript levels compared to wild-type level were identified and analyzed (Figure7C). These lines exhibited similar phenotypes to theCLE41oxandCLE42oxlines reported previously [13,31]。Transcripts forPXC1,PXY/TDR,WOX4,HB-8andREVwere elevated compared to wild-type plants in all three of ourCLE44oxlines (Figure7D). Once again, the explanation may be that the changes in gene expression are the result of developmental abnormalities. That is to say that the affected genes might not be directly regulated byCLE44over-expression, but the result of changes in cellular composition, such as an increase in the abundance of dividing/undifferentiated cells in the transgenic line (Figure7D). However, the similarities betweenPXC1and elements in theTDIF-PXY/TDRpathway indicate that PXC1 functions synergistically with the TDIF-PXY/TDR signaling pathway.

LRR-RLK receptors have been shown to mediate multiple signal transduction pathways. It is clear from work in the SAM that the combined actions of multiple LRR-RLKs, each with defined functions, are required to maintain the balance between stem cell division and differentiation in meristems. In vascular tissue, TDIF-PXY/TDR signal transduction pathway plays multiple roles in xylem development including the promotion of cambial cell division and repression of xylogenesis [31]。In this work, fromin silicoanalyses, a new LRR-RLK component involved in the regulation of plant vasculature development, PXC1, was introduced with its expression patterns correlated to that ofPXYgene. PXC1 probably plays its roles in a regulatory network which also incorporates the PXY/TDR-WOX4 signaling pathway and regulates the maturation of interfascicular fiber cells. The co-regulation network suggested that the loss ofPXC1function might retard the initiation of secondary cell wall deposition by prolonging the course of cell wall remodeling and reorganization during the procedure of cell expansion.

Expression profiling, co-expression analyses and gene functional clustering

Microarray expression data sets were explored for the predictedAtLRR-RLKgenes using the Arabidopsis Affymetrix GeneChip® average data available on the GENEVESTIGATOR analysis tool site (http://www.genevestigator.ethz.ch) [20]。The Gene Co-expression Analysis (GeneCAT) Toolbox athttp://genecat.mpg.de/cgi-bin/Ainitiator.py[41］was used to generate the Expression Tree which clustered genes by the similarity of their expression profiles and visualized those similarities using a dendogram. To minimize the effects of experimental artifacts, data were renormalized, and Pearson’s correlation coefficient between genes was weighted in ATTED-II [28]。

Identification of loss-of-function mutants and construction of transgenic Arabidopsis

Seeds for segregating T3 plants harboring thepxc1alleles in the Col-0 ecotype background were obtained from the NASC (Nottingham Arabidopsis Stock Centre. The location of the T-DNA insertion was determined by sequencing and homozygous lines were identified by PCR. For the production of the35Spro::CLE44construct, the coding sequence of Arabidopsis CLE44 was amplified from genomic DNA, cloned into pDONR201 vector (Invitrogen), and subcloned into the pK2GW7 vector using the Gateway cloning system (Invitrogen). Nativepromoter::GUS-GFPfusion constructs were made forPXYandPXC1/2/3by cloning the amplified promoter regions into the binary vector pKGWFS7™ [42］viapDONR201 (Invitrogen). Vectors were then transformed intoAgrobacterium tumefaciensstrain GV3101 (pMP90). Arabidopsis plants were transformed using the floral dip method [44]。Positive transgenic Arabidopsis plants were selected based on kanamycin resistance conferred by the T-DNA. T2 seeds from at least 24 independent positive lines for each construct were harvested for expression analyses.

Histochemistry

Plant tissues at various developmental stages were vacuum-infiltrated for 2 min in GUS solution including 1 mM X-gluc, 50 mM sodium phosphate (pH 7.0), 0.1% Triton X-100, 1 mM potassium ferricyanide and 1 mM potassium ferrocyanide, and incubated at 37°C overnight. Destaining of the samples were performed by incubation in 0.24 M HCl and 20% MeOH solution at 55°C for 15 min, then in 7% NaOH and 60% EtOH solution at room temperature for 15 min. The samples were dehydrated through ethanol series (40%, 20%, and 10% in water). For sectioning, samples were embedded in 4% (w/v) agar and sectioned at 50μm with a vibratome (Leica vt1000s). Samples were then mounted in glycerol and analyzed by bright field transmitted microscopy using an Axioplan 2 microscope (Carl Zeiss Inc. Thornwood, NY, USA). Images were captured by AxioCam HRc and Axiovision software (AxionVs40 V4.5.0.0).

Safranin/alcian-blue staining

Plants were grown under long day conditions and then were moved into short day conditions right after bolting. Long day conditions were 16h light/8h dark, 75% humidity, 150μE of irradiance and short day conditions were 8 h light/16 h dark, 75% humidity, 150μE of irradiance. For sectioning, samples were embedded in 4% (w/v) agar, and sectioned at 50μm with a vibratome (Leica vt1000s). Sections were then stained in one part safranin (1% w/v safranin in 50% ethanol) and two parts alcian-blue (1% v/w alcian-blue, 1% v/v formalin, 36% formaldehyde) in 0.05% glacial acetic acid. The sections were subsequently rinsed in ddH₂O and mounted in 50% glycerol. Slide-mounted sections were viewed using a Zeiss Axioplan 2 compound microscope with a Zeiss AxioCamHRc digital camera (Carl Zeiss, Inc., Thornwood, NY, USA).

Gene expression analyses by quantitative PCR

All samples came from five-week-old wild type plants under long-day conditions when the inflorescence stems reached a height of 25 cm, except that stem1 was from younger plants whose inflorescence was 10 cm high. Fully expanded leaves were harvested without the midrib, and petioles were harvested from the fully expanded leaves. Stem1 (younger stem) and stem2 (older stem) were both from the basal part of the inflorescence stems at 1–5 cm above the rosette. All nodes were removed from the stem samples. The phloem and xylem samples were obtained simply by separating the bark/phloem from the xylem core. Total RNA (0.2μg) was used for cDNA synthesis using the Thermoscript RT-PCR kit (Invitrogen Life Technologies, USA). Real-time qPCR was performed with the LightCycler instrument (Roche Diagnostics, Germany). Each mRNA value was corrected by the measurements obtained in the same sample for 18S mRNA and elongation factor 1 (EF1α) using Delta Delta method [44]。The primer sequences utilized in this study were listed in Additional file6. Each amplification included three technical replicates and their results were averaged to give the value for a single biological replicate. Three biological replicates were prepared for each treatment using material harvested from 10–15 plants in each case. The means of relative expression for each sample were examined using one-way analysis of variance (ANOVA) method (significance atP< 0.05).

JW, MK, LZ, PC and DD performed the experimental work. BZ, ON and GS designed and coordinated the project. JW, BZ and BJ wrote the paper. All authors read and approved the final manuscript.

这项研究是由国家项目上客y Basic Research Project (2012CB114500), National High Technology Research and Development Program (2011AA100200) and National Natural Science Foundation (31270644, 31170564 and 31070600) of China. It was also supported by Swedish Foundation for Strategic Research.

Affiliations

1. School of Environmental Science and Engineering, Tianjin University, Tianjin, 300072, China

   Jiehua Wang

2. Umeå Plant Science Centre, Department of Forest Genetics and Plant Physiology, Swedish University of Agricultural Sciences, SE-901 83, Umeå, Sweden

   Melis Kucukoglu, Linbin Zhang & Ove Nilsson

3. College of Plant Science and Technology, Huazhong Agricultural University, Wuhan, 430070, China

   Peng Chen

4. Umeå Plant Science Centre, Department of Plant Physiology, Umeå University, Umeå, SE-901 87, Sweden

   Daniel Decker & Göran Sandberg

5. Faculty of Agriculture and Environment, Department of Plant and Food Sciences, University of Sydney, Sydney, Australia

   Brian Jones

6. College of Horticulture and Forestry Sciences, Huazhong Agricultural University, Wuhan, 430070, China

   Bo Zheng

Corresponding author

Correspondence toBo Zheng.

Competing interests

The authors declare that they have no competing interests.

Open AccessThis article is published under license to BioMed Central Ltd. This is an Open Access article is distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Reprints and Permissions