Effect of the suppression of BpAP1 on the expression of lignin related genes in birch

Haijiao Huang · Shuo Wang · Huiyu Li · Jing Jiang

Abstract Lignin is an integral part of secondary cell walls in plants and plays important roles in maintaining the strength of stems, enhancing transport ability of stems, and providing resistance to multiple stresses.Lignin biosynthesis has become one of the hotspots in molecular forest biology research.The AP1 transcription factor plays important roles in plant flower development.However, in this study, suppression of BpAP1 altered the transcription profiles of white birch and RNA-seq was used to find that suppression of BpAP1 changed the expression of lignin pathway-related genes; C4H/CYP73A, POD were down-regulated and HCT, CCoAOMT, REF1 and CAD were up-regulated.Cell walls of the suppressed transgenic birch were significantly thinner than the wild type of birch, and BpAP1- repressed birch contained less lignin.In addition to regulation of floral development, BpAP1 might play a role in regulating the expression of genes in lignin biosynthesis of birch.This study could provide a new insight into the function of AP1 genes in woody species.

Keywords BpAP1 · Lignin biosynthesis · RNA-seq · Wood fiber cell walls

Introduction

Transcription factor (TF), also known as trans-acting factors, directly or indirectly interact with cis-acting elements in the promoter regions of downstream genes, a group of proteins that regulate the initiation of gene transcription.Transcription factors have been described figuratively as “switches that control several lights” at once.Their expression activates or inhibits the expression of a series of genes.The transcription translation product ofAPETALA1(AP1) gene is a transcription factor in plants.Several studies have reported that over expression of theAP1gene can accelerate flowering time and significantly shorten the juvenile phase (Irish and Sussex 1990; Mandel et al.1992; Elo et al.2001; Pea et al.2001; Cseke et al.2003; Fernando and Zhang 2005; Kotoda et al.2005; Jaya et al.2009; Chi et al.2011; Huang et al.2014).

Published research ofAP1mainly focuses on its canonical function in regulating flower development, and a previous study by Huang et al.(2014) found that theBpAP1inBetula platyphyllaSuk.is involved in regulation of flowering time as well; overexpression ofBpAP1resulted in early flowering.However, during the study, we found an interesting phenomenon; the suppression ofBpAP1could make cell walls thinner and lignin contents decreased significantly.

Lignin, a complex organic polymer, accumulates in secondary cell walls of the xylem.It is the second most abundant biopolymer after cellulose and accounts for approximately 25% of the total biomass of woody plants.In addition, lignin is a major form of carbon sequestration because of its resistance to biodegradation (Boerjan et al.2003; Boudet et al.2003).Lignin plays a significant role in the structural integrity of plant cell walls which has the function of connecting cells.It is also crucial for stem hardness and strength, enabling plants to grow upward (Chabannes et al.2001).It is important for plants to strengthen cell wall surfaces, and to resist bacteria and other invasive biological materials (Zabala et al.2006; Ithal et al.2007).Lignin is synthesized from three precursors of p-hydroxycinammyl alcohol, and the three hydroxycinnamyl alcohols (p-coumaryl, coniferyl and sinapyl) generate p-hydroxyphenyl (H), guaiacyl (G) and syringyl (S) units, respectively.It can be divided into three types due to the different monomers: syringyl lignin, guaiacyl lignin, and para-hydroxy phenyl lignin.

The content and composition of lignin varies considerably within and between plant species, between tissues and organs in the same plant (Plomion et al.2001), during different developmental stages, and in response to the environment (Bhuiyan et al.2009).Lignin of gymnosperms is mainly the G-type (guaicyl).Dicotyledonous lignin is G-and S-types (guaiacyl and syringyl), monocotyledons contain all three types.Betulais an important economic genus in many boreal regions and are dicots.In this study, RNA-seq, a widely used method (Chen et al.2019; Ritonga et al.2020) was used to determine ifBpAP1alters lignin and if the lignin biosynthesis pathway was changed in theBpAP1transgenic birch.

Materials and methods

Plant materials

The materials used were xylem tissues of 5-year-old transgenic birch lines in whichBpAP1was repressed (BpAP1RE, FA1-FA5) and the wild type (WT) (Huang et al.2015).Xylem from each line was collected from two individual trees and immersed in liquid nitrogen for further study.

Basal stem segments of theBpAP1RE transgenic line and WT were used (Fig.1c), including investigation of lignin contents and microstructures.

Fig.1 Plant material of BpAP1 RE transgenic line and WT.a The whole plant of BpAP1RE transgenic line and WT, from left to right the wild type and the two BpAP1RE transgenic lines; b and c basal stem segments of BpAP1 RE transgenic line and WT, from top to bottom the wild type and the two BpAP1RE transgenic lines

Investigation of stem straightness

We defined a standard for the degree of straightness according to the appearance of the transgenic birch: there is no bend for degree I; there is an obvious bend for degree II; there are two or three obvious bends for degree III; and, there are more than three obvious bends for degree IV.

qRT-PCR

Total RNA was extracted from transgenic and WT lines using the CTAB method (Chang et al.1993).First-strand cDNA was synthesized using a ReverTreAce_qPCR RT Kit (Toyobo Co., Ltd., Osaka, Japan) and qRT-PCR was performed using quantitative SYBR green PCR Master Mix (Toyobo Co., Ltd., Osaka, Japan) on an ABI 7500 real-time PCR system (45 cycles of 95 °C for 30 s, 95 °C for 15 s, and 60 °C for 1 min).A 18SrRNA(Accession number: GU476453.1) was used as housekeeping genes to analyze the expression of the selected unigenes.All experiments were replicated three times per sample and the results analyzed using the 2-△△Ctmethod (Livak and Schmittgen 2001).The primer sequences were shown in Table S1.Statistical analysis was performed using SPSS version 16.0.0 (SPSS Inc., Chicago, IL, USA) software to evaluate the expression of transgenic lines and WT.ANOVA was employed to test expression differences between the transgenic lines and WT.

Measurement of lignin content

The lignin content of 2-cm basal stem segments was determined by the Klason procedure and lignin acid solution (Whiting et al.1981; Dence 1992).The lignin-derived monomers were identified by gas chromatography-mass spectrometry of their trimethylsilylated derivatives (Lapierre et al.1983).All lignin analyses were duplicated.SPSS 16.0 was used to perform variance analysis (the same with qRT-PCR).

Observation of stem cross sections

A scanning electron microscope (SEM) (S-4800; Hitachi, Ibaraki, Japan) was used to observe the microstructures of the secondary xylem and the static digital images saved in BMP format.The cell wall thickness was determined using the binary morphology technique developed by Yu et al.(2008).

RNA-seq library construction and illumina sequencing

Total RNA of each sample was isolated using the CTAB method (Chang et al.1993).The contaminating genomic DNA was removed by DNaseI digestion (Promega, Madison, WI, USA).The quality and concentration of RNA was analyzed using Agilent 2100 Biological Analyzer (Agilent Technologies, Palo Alto, CA, USA).A 20 μg amount of total RNA was used for RNA sequencing library construction for each sample and were sequenced on Illumina Hiseq 4000 platform (Illumina, San Diego, CA, USA) in biomarker (Biomarker Technologies Co., Ltd, Beijing, China).

Data collection, functional annotation and differential expression analysis

Reads, including ambiguous bases such as N or low-quality bases and adapter contamination, were first filtered out.The resulting clean reads were aligned to theB.pendulagenome using Top Hataligner (ver.2.0.9) (Trapnell et al.2009).Gene expression levels were normalized using the FPKM method (fragments per kilobase of transcript per million mapped reads) by Cufflinks (Trapnell et al.2010).The differentially expressed genes (DEGs) were identified with thresholds of false discovery rate (FDR) ＜ 0.05 and fold change (FC) ≥ 2 (Robinson et al.2010).The DEGs were annotated using the databases of NCBI non-redundant (Nr), Swiss-Prot and Kyoto Encyclopedia of Genes and Genomes (KEGG).Blast2GO (Conesa et al.2005; Gotz et al.2008) was used to obtain gene ontology (GO) annotation of the genes based on the Nr annotation.Our annotation method was based on sequence homology alignment and the annotations that accompanied them.The aim of this method was to capture the richest and most complete annotation information possible.In addition, putative metabolic pathways were assigned to each gene by performing Blastx searches against the KEGG pathway database with an E value threshold of 1 × 10-5.

Results

Phenotypic changes of BpAP1 transgenic birch

In previous studies,BpAP1transgenic birch was obtained, includingBpAP1over expressing (OE) and repressing (RE) lines (Huang et al.2014, 2015).Over expression ofBpAP1significantly accelerated birch flowering time.In addition to regulation of flowering time,BpAP1may be involved in lignin biosynthesis.Stem straightness of 78.9%BpAP1RE lines were degree III (Table 1, Fig.1).The expression ofBpAP1was analyzed in the transgenic birch and the results showed that it decreased to varying degrees in theBpAP1RE transgenic lines (Fig.2).Since lignin composition could determine the mechanical strength of wood, lignin contents ofBpAP1RE transgenic lines and WT were measured.The results show that lignin contents ofBpAP1RE transgenic lines were significantly lower than WT (P＜ 0.01), and 13.7% lower than WT (Fig.3a).As noted previously, lignin contains three monomer types, guaiacyl lignin (G-lignin), syringyl lignin (S-lignin) and hydroxy-phenyl lignin (H-lignin).The lignin monomers were measured and found that G-lignin and S-lignin ofBpAP1RE transgenic lines significantly decreased, 12.3% and 14.0% lower than those in WT, respectively (Fig.3b and c).

Fig.2 Expression of BpAP1 in different transgenic lines and the wild type.Data with the same letter indicates no significant difference; different letters indicate a significant difference (P ＜ 0.05; Duncan’s multiple range analysis).The error bars represent standard deviation (SD)

Fig.3 Lignin content and ultra-microstructure analysis of transgenic birch and WT.a Total lignin content; b guajacyl lignin; c syringyl lignin; d-f SEM observation in cell wall thickness of transgenic lines and WT; d WT; e FA1; f FA4, magnification 5000×; g cell wall thickness comparison of transgenic lines and WT.The error bars represent SD; * represents a significant difference (P ＜ 0.05; Duncan’s multiple range analysis)

Table 1 Investigation of the trunk straightness

According to the lignin contents of the transgenic lines, transgenic birch of FA1 and FA4 were selected for xylem microstructure analysis.The secondary xylem microstructures of the stem bases were photographed using the S-4800 SEM (Fig.3d to f).Cell wall thickness was measured by the binary morphology methodology and it was found that the average xylem cell wall thickness in FA1 and FA4 was 0.98 μm, significantly thinner and 27.3% lower than that in non-transgenic plants (P＜ 0.01) (Fig.3g).

Generation of transcriptome sequencing data for BpAP1RE transgenic lines and WT

To analyze transcriptional alterations ofBpAP1RE transgenic lines, 12 cDNA libraries were constructed using RNA extracted from the xylem ofBpAP1RE (FA1-FA5) transgenic lines and WT.The sequencing reads were filtered to remove low-quality ones containing ambiguous nucleotides and adaptor sequences, resulting in approximately 40.0 Gb clean data for the 12 libraries.The Q30 percentages were over 91.0%.The GC content of the clean reads ranged from 45.2% to 46.2% in different libraries (Table 2).The clean reads libraries were aligned to theB.pendulagenome.Of these, 84.5% were mapped in each sample (Table 2).

Table 2 Summary of RNA-Seq

Suppression of BpAP1 altered transcriptional profiles of birch

Two biological replicates of RNA-seq data for WT and eachBpAP1RE transgenic lines were generated (Fig.4a).K-means cluster analysis using the gene expression data showed that all the replicates in the same line were clustered into one group, and the five transgenic lines further clustered into one group (Fig.4b), showing that transcriptome data of birch were reliable and suppression ofBpAP1changed the transcriptional profiles.Compared with WT, 2953, 3005, 2760, 2330, and 2049 DEGs were identified in FA1 to FA5, respectively (Fig.4c).The diversity of DEGs in the different transgenic lines may be due to multiple factors, such as different T-DNA insertion sites or different levels of repressing levels ofBpAP1.Therefore, it may be that the common DEGs in all five transgenic lines are affected byBpAP1.Among the DEGs, 886 genes were differentially expressed in all five transgenic lines (Fig.4d).

Fig.4 Transcriptome analysis overview of BpAP1 transgenic birch.a Two biological repeats of transcriptome sequencing, WT represents the wild birch, FA1-FA5 different BpAP1 transgenic lines; b K-means cluster analysis of the gene expression data; c DEG numbers of different transgenic lines (FA1-FA5); d Venn diagram of DEGs from FA1-FA5

GO enrichment is a widely used method in transcriptome analysis to uncover the molecular mechanism behind transcriptional alterations (Gang et al.2019; Wang et al.2019, 2020).To explore the potential functions of genes with significant transcriptional changes between theBpAP1RE transgenic lines and WT, GO categories were assigned to the 21,310 genes and 875 DEGs.The categorizations according to the biological process, cellular component, and molecular function ontology are shown in Fig.5.For the cellular component, the analysis revealed a high percentage of cell parts (723 DEGs, 82.6%), cell (701 DEGs, 80.1%), organelle (558 DEGs, 63.8%) and membrane (425 DEGs, 48.6%).For molecular function, the genes were classified into 17 categories.The two most over represented GO terms were catalytic activity (510 DEGs, 58.3%) and binding (518 DEGs, 59.2%).DEGs were related to 20 biological processes, including cellular process (720, 82.3%), single-organism process (698, 79.8%), metabolic process (676, 77.3%), response to stimulus (604, 69.0%), biological regulation (461, 52.7%), and the developmental process (Fig.5).

Fig.5 Gene ontology classification of assembled genes and DEGs.The genes were categorized based on gene ontology (GO) annotation, and the proportion of each category is displayed based on the ontologies biological process, cellular component and molecular function

Significant DEGs were also subjected to KEGG pathway analysis and all were mapped to 116 KEGG pathways; only two were enriched which were the flavonoid biosynthesis pathway and alpha-linolenic acid metabolism.However, there were other pathways which had obvious changed, and included phenylpropanoid biosynthesis, phenylalanine metabolism, and amino sugar and nucleotide sugar metabolism (Fig.S1).

Expression of lignin-related genes affected by BpAP1

BpAP1RE transgenic lines were obviously bent, lignin contents of both were lower, and xylem cell wall thickness thinner.Based on these changes, the lignin biosynthesis pathwayof transgenic lines and the wild type were analyzed to provide information on molecular mechanisms of the changes.Lignin biosynthesis is a part of the phenylpropanoid biosynthesis pathway which is enriched inBpAP1RE transgenic lines.Fourteen DEGs were identified that are involved in the lignin biosynthesis pathway, including genes encoding trans-cinnamate 4-monooxygenase (C4H/CYP73A), shikimate O-hydroxy cinnamoyl transferase (HCT), caffeoyl-CoA O-methyl transferase (CCoAOMT), coniferyl-aldehyde dehydrogenase (REF1), cinnamyl-alcohol dehydrogenase (CAD), and peroxidase (POD).The expression ofHCT,CCoAOMT,REF1andCADshowed significant up-regulation and the expression ofCYP73A,β-glucosidase,UGT72Esignificant down-regulation (Fig.6).The results of the expression of these genes were consistent with those of lignin monomer content.

Fig.6 Lignin biosynthesis pathway.PAL: Phenylalanine ammonia-lyase; C4H/CYP73A: trans-cinnamate 4-monooxygenase; CYP98A8_9: cytochrome P450 family 98 subfamily A polypeptide 8/9; COMT: caffeic acid 3-O-methyltransferase; F5H: flavanone 5-hydroxylase; CYP98A: 5-O-(4-coumaroyl)-D-quinate 3′-monooxygenase; 4CL: 4-coumarate-CoA ligase; CCR: cinnamoyl-CoA reductase; CCoAOMT: caffeoyl-CoA O-methyltransferase; CAD: cinnamyl-alcohol dehydrogenase; POD: Peroxidase; UGT72E: coniferylalcohol glucosyltransferase; REF1: coniferyl-aldehyde dehydrogenase.Blue represents down-regulation, red represents up-regulation

Discussion

Wood cells go through mitogenetic enlargement and wall thickening in the process of growth and development.The major wood cells are sclerenchyma, cellulose, hemicellulose and lignin constitute the cell wall, and lignin improves the mechanical strength.Lignin biosynthesis is divided into three stages: the photo contract product forms phenylalanine which is metabolized into lignin monomer, and the monomers polymerize into different types of lignin.There are numerous key enzymes in the lignin biosynthesis pathway.Several studies have regulated the biosynthesis of lignin through genetic engineering, for example,COMTandCCoAOMT, encoding methylase at different substrate levels, and the transformation ofCOMTandCCoAOMTantisense expression vectors into tabacoo can decrease the lignin content of transgenic plants (Zhao et al.2002).Reducing the expression ofCOMTin alfalfa can also decrease the lignin content (Guo et al.2001).Moreover, the decrease inCCoAOMTactivity in flax results in the thinning of xylem cells (Day et al.2009).A key enzyme in the early phenylpropane metabolic pathways of lignin biosynthesis isC4H(Boerjan et al.2003).Silencing theC4Hgene in poplar can reduce lignin content in transgenic plants (Bjurhager et al.2010), and down-regulated expression ofC4HinArabidopsis thaliana, tobacco and alfalfa can also decrease lignin content (Sewalt et al.1997; Blee et al.2001; Reddy et al.2005; Schilmiller et al.2009).These results indicate thatC4H,COMTandCCoAOMTare involved in the regulation of lignin biosynthesis.

This study found that the performance ofBpAP1RE transgenic lines had large amount of growth, 45.4% significantly higher than the non-transgenic line.However, silencing ofBpAP1causes a curved trunk (Huang et al.2015).BpAP1is not only involved in the flowering of birch but also participates in the formation of cellulose, hemicellulose, and lignin.The results show that G-lignin and S-lignin ofBpAP1RE transgenic lines were significantly lower than in the wild type, and cell walls were also significantly thinner.A study by Li et al.(2016) showed that another MADS-box gene,BpMADS12, promotes lignin synthesis through regulation of key enzymes in response to brassinosteroid signaling.

Moreover, the expression ofC4H/CYP73AandPODinvolved in lignin biosynthesis ofBpAP1RE transgenic lines was significantly down-regulated, while the expression ofHCT,CCoAOMT,REF1, andCADshowed significant upregulation (Fig.6).These results indicate that the change ofBpAP1expression caused the change ofC4H/CYP73A,CCoAOMT,HCTand other gene expressions.Thus, the lignin biosynthesis was affected.These results suggest that this MADS-box transcript factor is crucial to plant structure and provides a foundation for studies aiming to elucidate the developmental mechanisms underlying the formation of wood.

Conclusion

BpAP1repressing transgenic line and the wild type were used as plant materials with RNA sequencing technology to identify differentially expressed genes in the xylem.The results show that suppression ofBpAP1affected the lignin biosynthesis pathway-related genes, including the down-regulated genesC4H/CYP73A, andPOD, andHCT,CCoAOMT,REF1andCADwere up-regulated.Cell walls of the suppression transgenic birch were significantly thinner than in the wild type, andBpAP1suppressed birch contained less lignin than the wild type.We speculate thatBpAP1may play a role in regulating the expression of lignin pathway genes of birch.

AcknowledgementsWe sincerely thank Yongzhi Cui for technical assistance with scanning electron microscopy.