Wood Transcriptome Profiling Identifies Critical Pathway Genes of Secondary Wall Biosynthesis and Novel Regulators for Vascular Cambium Development in Populus

Abstract

Wood, the most abundant biomass on Earth, is composed of secondary xylem differentiated from vascular cambium. However, the underlying molecular mechanisms of wood formation remain largely unclear. To gain insight into wood formation, we performed a series of wood-forming tissue-specific transcriptome analyses from a hybrid poplar (Populus alba × P. glandulosa, clone BH) using RNA-seq. Together with shoot apex and leaf tissue, cambium and xylem tissues were isolated from vertical stem segments representing a gradient of secondary growth developmental stages (i.e., immature, intermediate, and mature stem). In a comparative transcriptome analysis of the 'developing xylem' and 'leaf' tissue, we could identify critical players catalyzing each biosynthetic step of secondary wall components (e.g., cellulose, xylan, and lignin). Several candidate genes involved in the initiation of vascular cambium formation were found via a co-expression network analysis using abundantly expressed genes in the 'intermediate stem-derived cambium' tissue. We found that transgenic Arabidopsis plants overexpressing the PtrHAM4-1, a GRAS family transcription factor, resulted in a significant increase of vascular cambium development. This phenotype was successfully reproduced in the transgenic poplars overexpressing the PtrHAM4-1. Taken together, our results may serve as a springboard for further research to unravel the molecular mechanism of wood formation, one of the most important biological processes on this planet.

Keywords: Ptrham4-1; poplar; tissue-specific transcriptome; vascular cambium; wood formation.

Figure 1

Tissue-specific transcriptome analysis of a hybrid poplar. RNA-seq analysis of various poplar tissues for tissue-specific study. (a) Strategy for poplar tissue sampling used in this study. Schematic diagram of tissue samples collected. SL (shoot apical meristem with leaf primordia), IS (immature stem), IC (intermediate cambium), IDX (intermediate developing xylem), MC (mature cambium), MDX (mature developing xylem), and ML (mature leaf without major veins). (b,c) Sample correlation by heatmap (b) and correlation matrix (c), which were produced by Trinity (analyze_diff_expr.pl) with log2 fold change value (p-value, 0.005).

Figure 2

Confirmation of RNA-seq results by independent qRT-PCR analysis. Expression of each indicated gene is plotted. In each plot, blue bars (right Y-axis) indicate FPKM (Fragment Per Kilobase of transcript per Million mapped reads) values from RNA-seq data while red bars (left Y-axis) indicate qRT-PCR results. A gene model name for P. trichocarpa (v3.0) is shown in parenthesis as: PtrCesA4 (Potri.002G257900.1); PtrCesA7 (Potri.018G103900.1); PtrCesA8 (Potri.011G069600.1); PtrIRX9 (Potri.016G086400.1); PtrPAL1 (Potri.006G126800.1); PtrC4H1 (Potri.013G157900.1); Ptr4CL1 (Potri.001G036900.1); PtrCHI1 (Potri.010G213000.1); PtrANS1 (Potri.003G119100.1); PtrMYB46 (Potri.009G053900.1); PtrPXY (Potri.003G107600.1); and PtrWOX4 (Potri.014G025300.1). PtrActin2 (Potri.019G010400.1) was used as a reference gene. Error bars indicate S.E. gene-specific primer sequences used in this analysis are shown in Table S2.

Figure 3

Identification of major players within cellulose and xylan biosynthetic pathways for secondary wall formation. Metabolites in each step of the biosynthetic pathway are shown in the box and all related genes are shown to the side. Fold change (X/L) was calculated in Log2. X is a FPKM value of mature developing xylem (MDX) and L (Leaf without major veins) represents the leaf FPKM value. Color gradient according to the fold change values was visualized.

Figure 4

Identification of major players of the monolignol biosynthetic pathway in secondary wall formation. Metabolites in each step of the biosynthetic pathway are shown in the box and all the related genes are shown to the side. Fold changes (X/L) were calculated in Log2. X is a FPKM value of mature developing xylem (MDX) and L is the leaf FPKM value. Color gradient according to fold change values was visualized.

Figure 5

Wood-forming tissue-specific expression of secondary wall biosynthetic genes in poplar. Transcript abundance of all secondary wall biosynthetic genes in each tissue was plotted. (a) Genes responsible for biosynthesis of cellulose and hemicellulose: CesA4 (Potri.002G257900.1); CesA7-A (Potri.006G181900.1); CesA8-A (Potri.011G069600.1); CesA8-B (Potri.004G059600.2); CslA1 (Potri.008G026400.1); CslA2 (Potri.010G234100.1); FRA8 (Potri.009G006500.1); GUX1a (Potri.007G107200.2); GUX1b (Potri.005G061600.4); GUX2 (Potri.014G029900.1); IRX8-1 (Potri.011G132600.1); IRX8-2 (Potri.001G416800.1); IRX9-1 (Potri.006G131000.1); IRX9-2 (Potri.016G086400.1); IRX10-1 (Potri.001G068100.3); IRX10-2 (Potri.003G162000.1); IRX14-1 (Potri.005G141500.1); IRX14-2 (Potri.007G047500.1); PARVUS-1 (Potri.002G132900.1); PARVUS-2 (Potri.014G040300.1); PARVUS-L-2 (Potri.007G031700.1). (b) Genes responsible for monolignol biosynthesis: PAL1 (Potri.006G126800.1); PAL2 (Potri.008G038200.2); PAL3 (Potri.016G091100.1); PAL4 (Potri.010G224100.1); C4H1 (Potri.013G157900.1); C4H2 (Potri.019G130700.1); 4CL3 (Potri.001G036900.1); 4CL5 (Potri.003G188500.1); HCT1 (Potri.003G183900.2); HCT6 (Potri.001G042900.2); CSE1 (Potri.003G059200.1); CSE2 (Potri.001G175000.1); C3H3 (Potri.006G033300.3); CCoAOMT1 (Potri.009G099800.1); CCoAOMT2 (Potri.001G304800.1); CCoAOMT3 (Potri.008G136600.1); CCR2 (Potri.003G181400.1); CAld5H2 (Potri.007G016400.1); COMT2 (Potri.012G006400.1); CAD1 (Potri.009G095800.1). Transcript abundance on the Y-axis represents FPKM values.

Figure 6

Identification of PtrHAM4-1, preferentially expressed in cambium tissue based on in Silico analysis. (a) Co-expression network of the intermediate cambium (IC) tissue preferentially expressed transcriptional regulators. The co-expression network was obtained from the AspWood website (http://aspwood.popgenie.org/aspwood-v3.0/) by querying a total of 64 poplar genes (Potri. ID) in the Table 2. PtrHAM4-1 (Potri.005G125800) was relocated to emphasize. (b) Tissue-specific expression of both Potri.005G125800.1/PtrHAM4-1 and its closest homolog, Potri.007G029200.1/PtrHAM4-2. This diagram was reconstructed from our RNAseq data. (c) PtrHAM4-1 is highly expressed in the phloem and cambial tissues. To obtain a gene expression profile by exploiting the high spatial-resolution wood formation data [37] the list of genes from Table 2 was queried to the AspWood website. The resulting heatmap showed that PtrHAM4-1 (indicated by red letters) is highly expressed in the phloem and cambial tissues.

Figure 7

Overexpression of PtrHAM4-1 in transgenic Arabidopsis increased cambium development. (a) Overall growth phenotypes of transgenic Arabidopsis plants. Five independent T3 homozygote transgenic lines (e.g., 2–7, 4–1, 6–7, 7–6, and 16–6) are shown. (b) Expression of the PtrHAM4-1 gene in the independent transgenic Arabidopsis lines compared with Col-0. First-strand cDNA was synthesized from the total RNA extracted from stem tissues and used as a template in semi-quantitative RT-PCR experiments. (c) Observation of cambium development in stem cross sections from 60-day-old transgenic Arabidopsis and Col-0. Yellow braces indicate the ICD (interfascicular cambium-derived tissue) extension and red arrowheads point to secondary xylem vessels within the interfascicular region. Scale bars represent 100 µm. (d) Quantification of ICD extension in 60-day-old transgenic Arabidopsis plants compared to Col-0. ICD extensions were measured in all interfascicular regions of rosette level stem sections from five independent transgenic lines. Error bars indicate S.E. (n = 5).

Figure 8

Transgenic poplar overexpressing PtrHAM4-1 resulted in an increased cambium development. (a) Observation of cambium development in stem cross sections of 3-month-old transgenic poplars and wild type BH clone (BH) grown in test tube. Red braces under the red star indicate the cambial layers. Scale bars represent 50 µm. (b) Quantification of cambial layers of transgenic poplars compared to BH. The length of cambial layers was measured in stem sections from five ramets of each transgenic lines described in (a). Error bars indicate S.E. (n = 5). (c) Expression of the PtrHAM4-1 gene in the independent transgenic poplar lines compared to BH. First-strand cDNA was synthesized from the total RNA extracted from stem tissues and used as a template in the qRT-PCR experiments.