Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Apr;176(4):2851-2870.
doi: 10.1104/pp.17.01590. Epub 2018 Feb 27.

Transcriptional Roadmap to Seasonal Variation in Wood Formation of Norway Spruce

Affiliations
Free PMC article

Transcriptional Roadmap to Seasonal Variation in Wood Formation of Norway Spruce

Soile Jokipii-Lukkari et al. Plant Physiol. .
Free PMC article

Abstract

Seasonal cues influence several aspects of the secondary growth of tree stems, including cambial activity, wood chemistry, and transition to latewood formation. We investigated seasonal changes in cambial activity, secondary cell wall formation, and tracheid cell death in woody tissues of Norway spruce (Picea abies) throughout one seasonal cycle. RNA sequencing was performed simultaneously in both the xylem and cambium/phloem tissues of the stem. Principal component analysis revealed gradual shifts in the transcriptomes that followed a chronological order throughout the season. A notable remodeling of the transcriptome was observed in the winter, with many genes having maximal expression during the coldest months of the year. A highly coexpressed set of monolignol biosynthesis genes showed high expression during the period of secondary cell wall formation as well as a second peak in midwinter. This midwinter peak in expression did not trigger lignin deposition, as determined by pyrolysis-gas chromatography/mass spectrometry. Coexpression consensus network analyses suggested the involvement of transcription factors belonging to the ASYMMETRIC LEAVES2/LATERAL ORGAN BOUNDARIES and MYELOBLASTOSIS-HOMEOBOX families in the seasonal control of secondary cell wall formation of tracheids. Interestingly, the lifetime of the latewood tracheids stretched beyond the winter dormancy period, correlating with a lack of cell death-related gene expression. Our transcriptomic analyses combined with phylogenetic and microscopic analyses also identified the cellulose and lignin biosynthetic genes and putative regulators for latewood formation and tracheid cell death in Norway spruce, providing a toolbox for further physiological and functional assays of these important phase transitions.

Figures

Figure 1.
Figure 1.
Seasonal variation in cambial activity and wood formation on the anatomical and RNA-Seq levels during one year in Hissjö, Sweden. A, Light microscopy sections from four different sampling dates representing different stages of cambial activity. Viability of the cells was assessed by nitroblue tetrazolium (NBT) staining. Images were taken (from left to right) on May 11, 2012; June 13, 2011; July 11, 2011; and September 6, 2011. c, Cambium; ew, earlywood; lw, latewood; p, phloem; plw, latewood of the previous growing season. Bars = 100 µm (first, third, and fourth images) and 200 µm (second image). B, Scheme of sample collection. Green rectangle, Summer months June, July, and August; brown rectangle, autumn months September, October, and November; blue rectangle, winter months December, January, and February; and yellow rectangle, spring months March, April, and May. C, PCA plot of the sequencing reads showing tissue type separation, the clustering of biological replicates, and the effect of time in a three-dimensional space. Different sampling points are represented by letters from A to W. Cambium/phloem and xylem samples are marked with green and magenta, respectively. A semantic summary (less than log10 P < −1.3) of GO term enrichment for the genes with highest loadings is given next to each component (Comp.).
Figure 2.
Figure 2.
Seasonal gene expression patterns in woody tissues of Norway spruce. Hierarchical clustering of genes is shown for all replicates across one season. Samples are ordered by sampling date. Expression values were library size corrected and variance stabilized (∼log2). The legibility of the heat map was improved by saturating the expression data between log2 values 3 and 10. The ribbon at the top of the heat map indicates the tissue type: magenta for xylem and green for cambium/phloem. Sampling months are marked by colors below the heat map: green, summer; brown, autumn; blue, winter; and yellow, spring.
Figure 3.
Figure 3.
Phase transitions in cambial growth and wood formation. Gene expression is shown for putative Norway spruce marker genes of cambial activity (CDKBs), cell expansion (EXPAs), phloem (APL) and xylem (CNA) specification, dormancy (FTL2), and SCW formation (CESAs). Values are means ± se.
Figure 4.
Figure 4.
Norway spruce monolignol genes showed expression peaks both in the summer and in the winter. A, Hierarchical clustering of all putative monolignol, laccase (LAC), and peroxidase (PRX) genes identified based on the phylogenetic analyses and expression in xylem samples. The candidate genes for lignin biosynthesis and polymerization were identified from clusters 1 and 2, respectively. B, Names of the genes in clusters 1 and 2. C and D, Expression profiles of the genes in clusters 1 and 2. E, Daily mean temperatures of sampling dates in Umeå. F, Proportions of lignin ± se in the part of the year ring that contained living tracheids. The periods of earlywood (ew) and latewood (lw) formation are indicated. Dashed lines show the date for the coldest sampling time point during the study period.
Figure 5.
Figure 5.
A subnet in the coexpression network connects several gene models related to cell wall formation and especially the monolignol pathway. A, Location of the subnet in the coexpression network. B, Gene cluster containing several cluster 1 lignin biosynthetic genes. The size of the octagons represents the betweenness centrality value, while the color denotes neighborhood connectivity (ascending from yellow to red). Double links indicate bidirectional edges. C, Expression profiles of the TF genes implicated in the seasonal control of lignification. Values are means ± se.
Figure 6.
Figure 6.
Onset of latewood formation in Norway spruce. A, Date of latewood onset and variation among the three replicate trees for each time point. B, Venn diagram showing the overlap between the results of two differential expression analyses (adjusted P < 0.01; 79 genes): the first time point when latewood cells were observed versus the last sampling before latewood detection (640 genes) and samples containing latewood versus those without over the course of latewood formation (467 genes; three samplings in July). C, GO term enrichment analysis of the common genes between comparisons.
Figure 7.
Figure 7.
The long lifespan of latewood tracheids. A, Average amount (%) of viable xylem tissue ± se, defined based on NBT viability staining, and average width (µm) of earlywood and latewood in sampled trees. B, Light microscopy image of a transverse section collected on March 27, 2012, and stained with NBT. The viability is shown by the presence of a pink/lilac precipitate that is formed by succinate dehydrogenase activity of living cells in the presence of NBT. Enzymatic activity was observed in several cell layers of the latewood. c, Cambium; ew, earlywood; lw, latewood; p, phloem; r, ray. C, A Toluidine Blue-stained section from March 27, 2012. D and E, Electron micrographs showing latewood tracheids with intact vacuoles in a sample collected May 2, 2012. Tonoplast is marked by the arrow. v, Vacuole. Bars = 100 µm (B and C), 5 µm (D), and 2 µm (E).
Figure 8.
Figure 8.
Norway spruce PaMC5, BFN, XCP, and CEP are strong candidate genes for executors of tracheid cell death. A, Hierarchical clustering of genes putatively related to PCD selected based on phylogenetic analyses. The genes that were differentially expressed (adjusted P < 0.01) during the appearance of dead earlywood cells (time point marked with the red arrow) are written in boldface. The cluster containing homologs for Arabidopsis MC9, BFN1, and XCP genes is marked with the blue rectangle next to the heat map. Sampling months are marked by colors below the heat map; green, summer; brown, autumn; blue, winter; and yellow, spring. B, Seven out of eight genes in the identified cluster were connected in the coexpression network of the spatial wood section series of the NorWood Web resource (threshold 3; http://NorWood.ConGenIE.org). PaMC5, BFN, XCP, and CEP showed the strongest correlation in their expression. The gray and red lines indicate positive and negative correlations, respectively. The width of the line is relative to the strength of the correlation. C, PaMC5, BFN, XCP, and CEP exhibited high expression in the cell death zone (red rectangle) of the tangential wood sections of NorWood.
Figure 9.
Figure 9.
Marker genes define the main phase transitions during the annual cycle of cambial growth in Norway spruce. The outermost and second outermost rings indicate the transcript abundances in xylem and cambium/phloem samples, respectively. The profiles represent the means of the specific gene groups.

Similar articles

See all similar articles

Cited by 4 articles

Publication types

MeSH terms

LinkOut - more resources

Feedback