An understanding of allelic diversity and population structure is important in developing association studies and constructing core collections for tree breeding. We examined population genetic differentiation in the native Populus tomentosa by genotyping 460 unrelated individuals using 20 species-specific microsatellite markers. We identified 99 alleles with a mean of 4.95 observed alleles per locus, indicating a moderate level of polymorphism across all individuals. A model-based population structure analysis divided P. tomentosa into 11 subpopulations (K = 11). The pattern of individual assignments into the subsets (K = 3) provided reasonable evidence for treating climatic zones as genetic regions for population genetics. The highest level of genetic variation was found in the southern region (i.e., N = 93, N (P) = 11, H (E) = 0.445, F = -0.102), followed by the northeastern and northwestern regions. Thus, the southern region is probably the center of the current species distribution. No correlation was found between population genetic distance and geographic distance (r = 0.0855, P = 0.3140), indicating that geographical distance was not the principal factor influencing genetic differentiation in P. tomentosa. These data provide a starting point for conserving valuable natural resources and optimizing breeding programs.

SummaryDeciphering the genetic architecture underlying polygenic traits in perennial species can inform molecular marker-assisted breeding. Recent advances in high-throughput sequencing have enabled strategies that integrate linkage-linkage disequilibrium (LD) mapping in Populus.We used an integrated method of quantitative trait locus (QTL) dissection with a high-resolution linkage map and multi-gene association mapping to decipher the nature of genetic architecture (additive, dominant, and epistatic effects) of potential QTLs for growth traits in a Populus linkage population (1200 progeny) and a natural population (435 individuals).Seventeen QTLs for tree height, diameter at breast height, and stem volume mapped to 11 linkage groups (logarithm of odds (LOD) ≥ 2.5), and explained 2.7-18.5% of the phenotypic variance. After comparative mapping and transcriptome analysis, 187 expressed genes (10 046 common single nucleotide polymorphisms (SNPs)) were selected from the segmental homology regions (SHRs) of 13 QTLs. Using multi-gene association models, we observed 202 significant SNPs in 63 promising genes from 10 QTLs (P ≤ 0.0001; FDR ≤ 0.10) that exhibited reproducible associations with additive/dominant effects, and further determined 11 topranked genes tightly linked to the QTLs. Epistasis analysis uncovered a uniquely interconnected gene-gene network for each trait.This study opens up opportunities to uncover the causal networks of interacting genes in plants using an integrated linkage-LD mapping approach.

Long non-coding RNAs (lncRNAs) participate in a wide range of biological processes, but lncRNAs in plants remain largely unknown; in particular, we lack a systematic identification of plant lncRNAs involved in hormone responses. Moreover, allelic variation in lncRNAs remains poorly characterized at a large scale. Here, we conducted high-throughput RNA-sequencing of leaves from control and gibberellin (GA)-treated Populus tomentosa and identified 7655 reliably expressed lncRNAs. Among the 7655 lncRNAs, the levels of 410 lncRNAs changed in response to GA. Seven GA-responsive lncRNAs were predicted to be putative targets of 18 miRNAs, and one GA-responsive lncRNA (TCONS_00264314) was predicted to be a target mimic of ptc-miR6459b. Computational analysis predicted 939 potential cis-regulated target genes and 965 potential trans-regulated target genes for GA-responsive lncRNAs. Functional annotation of these potential target genes showed that they participate in many different biological processes, including auxin signal transduction and synthesis of cellulose and pectin, indicating that GA-responsive lncRNAs may influence growth and wood properties. Finally, single nucleotide polymorphism (SNP)-based association analysis showed that 112 SNPs from 52 GA-responsive lncRNAs and 1014 SNPs from 296 potential target genes were significantly associated with growth and wood properties. Epistasis analysis also provided evidence for interactions between lncRNAs and their potential target genes. Our study provides a comprehensive view of P. tomentosa lncRNAs and offers insights into the potential functions and regulatory interactions of GA-responsive lncRNAs, thus forming the foundation for future functional analysis of GA-responsive lncRNAs in P. tomentosa.

SummaryChinese white poplar (Populus tomentosa), an important commercial tree species for timber and pulp production in northern China, has been used to examine the individual genes and allelic diversity responsible for complex traits controlling growth and lignocellulosic biosynthesis. Taking advantage of the low degree of linkage disequilibrium (LD) within P. tomentosa association populations, we examined associations between 15 cellulose synthase (PtoCesA) genes and traits including growth and wood properties.Thirty-six novel simple sequence repeat (SSR) markers within PtoCesA genes were detected by re-sequencing and genotyped in an association population (460 individuals). Single-marker and haplotype-based LD approaches were used to identify significant marker-trait associations. Family-based linkage studies and real-time PCR testing were conducted to validate the functional significance of SSR variation.Fifteen single-marker associations from seven PtoCesA genes and nine haplotype-based associations within six genes were identified in the association population (false discovery rate Q < 0.05). Next, five SSR marker-trait associations (Q < 0.05) from four PtoCesA genes were successfully validated in a linkage mapping population (1200 individuals).The results imply a functional role for these genes in mediating wood properties, demonstrating the potential of combining single-marker and haplotype-based LD approaches to detect functional allelic variation underlying quantitative traits in a low-LD population.

Despite the significance of actin in plant growth and development, little is known of the structure, expression and evolution of the actin gene family in woody plants. In this study, we systematically examined the diversification of the actin gene family in Populus by integrating genomic organization, expression, and phylogeny data. Genome-wide analysis of the Populus genome indicated that actin is a multigene family consisting of eight members, all predicted to encode 377-amino acid polypeptides that share high sequence homology ranging from 94.2 to 100% identity. Microarray and real-time PCR expression analysis showed that the PtrACT family members are differentially expressed in different tissues, exhibiting overlapping and unique expression patterns. Of particular interest, all PtrACT genes have been found to be preferentially expressed in the stem phloem and xylem, suggesting that poplar PtrACTs are involved in the wood formation. Gene structural and phylogenetic analyses revealed that the PtrACT family is composed of two main subgroups that share an ancient common ancestor. Extremely high intraspecies synonymous nucleotide diversity of pi(syn) = 0.01205 was detected, and the pi(non-syn)/pi(syn) ratio was significantly less than 1; therefore, the PtACT1 appears to be evolving in Populus, primarily under purifying selection. We demonstrated that the actin gene family in Populus is divided into two distinct subgroups, suggesting functional divergence. The results reported here will be useful in conducting future functional genomics studies to understand the detailed function of actin genes in tree growth and development.

Economically important traits in many species generally show polygenic, quantitative inheritance. The components of genetic variation (additive, dominant and epistatic effects) of these traits conferred by multiple genes in shared biological pathways remain to be defined. Here, we investigated 11 full-length genes in cellulose biosynthesis, on 10 growth and wood-property traits, within a population of 460 unrelated Populus tomentosa individuals, via multi-gene association. To validate positive associations, we conducted single-marker analysis in a linkage population of 1,200 individuals. We identified 118, 121, and 43 associations (P< 0.01) corresponding to additive, dominant, and epistatic effects, respectively, with low to moderate proportions of phenotypic variance (R2). Epistatic interaction models uncovered a combination of three non-synonymous sites from three unique genes, representing a significant epistasis for diameter at breast height and stem volume. Single-marker analysis validated 61 associations (false discovery rate, Q ≤ 0.10), representing 38 SNPs from nine genes, and its average effect (R2 = 3.8%) nearly 2-fold higher than that identified with multi-gene association, suggesting that multi-gene association can capture smaller individual variants. Moreover, a structural gene–gene network based on tissue-specific transcript abundances provides a better understanding of the multi-gene pathway affecting tree growth and lignocellulose biosynthesis. Our study highlights the importance of pathway-based multiple gene associations to uncover the nature of genetic variance for quantitative traits and may drive novel progress in molecular breeding.

Lignin provides structural support in perennial woody plants and is a complex phenolic polymer derived from phenylpropanoid pathway. Lignin biosynthesis is regulated by coordinated networks involving transcription factors (TFs), microRNAs (miRNAs) and long noncoding RNAs (lncRNAs). However, the genetic networks underlying the lignin biosynthesis pathway for tree growth and wood properties remain unknown. Here, we used association genetics (additive, dominant and epistasis) and expression quantitative trait nucleotide (eQTN) mapping to decipher the genetic networks for tree growth and wood properties in 435 unrelated individuals of Populus tomentosa. We detected 124 significant associations (P ≤ 6.89E-05) for 10 growth and wood property traits using 30 265 single nucleotide polymorphisms from 203 lignin biosynthetic genes, 81 TF genes, 36 miRNA genes and 71 lncRNA loci, implying their common roles in wood formation. Epistasis analysis uncovered 745 significant pairwise interactions, which helped to construct proposed genetic networks of lignin biosynthesis pathway and found that these regulators might affect phenotypes by linking two lignin biosynthetic genes. eQTNs were used to interpret how causal genes contributed to phenotypes. Lastly, we investigated the possible functions of the genes encoding 4-coumarate: CoA ligase and cinnamate-4-hydroxylase in wood traits using epistasis, eQTN mapping and enzymatic activity assays. Our study provides new insights into the lignin biosynthesis pathway in poplar and enables the novel genetic factors as biomarkers for facilitating genetic improvement of trees.

In perennial woody plants, the coordinated increase of stem height and diameter during juvenile growth improves competitiveness (i.e. access to light); however, the factors underlying variation in stem growth remain unknown in trees. Here, we used linkage-linkage disequilibrium (linkage-LD) mapping to decipher the genetic architecture underlying three growth traits during juvenile stem growth. We used two Populus populations: a linkage mapping population comprising a full-sib family of 1,200 progeny and an association mapping panel comprising 435 unrelated individuals from nearly the entire natural range of Populus tomentosa. We mapped 311 quantitative trait loci (QTL) for three growth traits at 12 timepoints to 42 regions in 17 linkage groups. Of these, 28 regions encompassing 233 QTL were annotated as 27 segmental homology regions (SHRs). Using SNPs identified by whole-genome re-sequencing of the 435-member association mapping panel, we identified significant SNPs (P ≤ 9.4 × 10 ) within 27 SHRs that affect stem growth at nine timepoints with diverse additive and dominance patterns, and these SNPs exhibited complex allelic epistasis over the juvenile growth period. Nineteen genes linked to potential causative alleles that have time-specific or pleiotropic effects, and mostly overlapped with significant signatures of selection within SHRs between climatic regions represented by the association mapping panel. Five genes with potential time-specific effects showed species-specific temporal expression profiles during the juvenile stages of stem growth in five representative Populus species. Our observations revealed the importance of considering temporal genetic basis of complex traits, which will facilitate the molecular design of tree ideotypes.