Identification and Investigation of the Genetic Variations and Candidate Genes Responsible for Seed Weight via GWAS in Paper Mulberry

Abstract

Seeds directly determine the survival and population size of woody plants, but the genetic basis of seed weight in woody plants remain poorly explored. To identify genetic variations and candidate genes responsible for seed weight in natural woody populations, we investigated the hundred-seed weight of 198 paper mulberry individuals from different areas. Our results showed that the hundred-seed weight of paper mulberry was significantly associated with the bioclimatic variables of sampling sites, which increased from south to north along the latitudinal-temperature gradient. Using 2,414,978 high-quality SNPs from re-sequencing data, the genome-wide association analysis of the hundred-seed weight was performed under three models, which identified 148, 19 and 12 associated genes, respectively. Among them, 25 candidate genes were directly hit by the significant SNPs, including the WRKY transcription factor, fatty acid desaturase, F-box protein, etc. Most importantly, we identified three crucial genetic variations in the coding regions of candidate genes (Bp02g2123, Bp01g3291 and Bp10g1642), and significant differences in the hundred-seed weight were detected among the individuals carrying different genotypes. Further analysis revealed that Bp02g2123 encoding a fatty acid desaturase (FAD) might be a key factor affecting the seed weight and local climate adaptation of woody plants. Furthermore, the genome-wide investigation and expression analysis of FAD genes were performed, and the results suggested that BpFADs widely expressed in various tissues and responded to multiple phytohormone and stress treatments. Overall, our study identifies valuable genetic variations and candidate genes, and provides a better understanding of the genetic basis of seed weight in woody plants.

Keywords: candidate gene; fatty acid desaturase; genetic variation; genome-wide association study; seed weight; woody plant.

Figure 1

The seed size and distribution of the hundred-seed weight of paper mulberry. (A) One hundred seeds of paper mulberry individuals from different areas. The seeds on the left were collected from Chengde, Hebei, China (41° N), and the seeds on the right were collected from Dongfang, Hainan, China (19° N). (B) The seed sizes of paper mulberry individuals from different areas. The collection areas are the same as subfigure A. (C) The frequency distribution of the hundred-seed weight of paper mulberry. (D) The geographic distribution of the hundred-seed weight. The graded color scale from blue to red was used to display the hundred-seed weight from small to large.

Figure 2

The linear fitting of the hundred-seed weight and bioclimatic variables of sampling sites. (A) Latitude. (B) Longitude. (C) Mean temperature. (D) Active accumulated temperature ≥10 °C. (E) Extreme low temperature. (F) Frost-free period. (G) Annual precipitation. (H) Sunshine hours.

Figure 3

Genome-wide association study on the hundred-seed weight using three association analysis models. (A–C) The Manhattan plots of the GWAS on the hundred-seed weight using the MLM+K model, MLM+Q+K model, and FarmCPU model, respectively. The black solid line and dashed line indicate the significance threshold of 0.01 and 0.05, respectively. The spectrum column was used to represent the SNP density along 13 chromosomes of paper mulberry. (D–F) The QQ-plots of the GWAS on the hundred-seed weight using the MLM+K model, MLM+Q+K model, and FarmCPU model, respectively.

Figure 4

The GO annotation of the associated genes identified by the GWAS of hundred-seed weight through multiple models. (A) The Gene Ontology (GO) classification of the associated genes. (B) The top 20 GO terms in biological processes.

Figure 5

The analysis of the SNPs 2:33502257, 1:44171110 and 10:18312485. (A) The linkage disequilibrium heat map surrounding the SNP 2:33502257. (B) The gene structure of Bp02g2123 and the DNA polymorphism of the nonsynonymous SNP 2:33502257. (C) Comparison of the hundred-seed weight of the individuals carrying different genotypes of 2:33502257. (D) The linkage disequilibrium heat map surrounding the SNP 1:44171110. (E) The gene structure of Bp01g3291 and the DNA polymorphism of the nonsynonymous SNP 1:44171110. (F) Comparison of the hundred-seed weight of the individuals carrying different genotypes of 1:44171110. (G) The linkage disequilibrium heat map surrounding the SNP 10:18312485. (H) The gene structure of Bp10g1642 and the DNA polymorphism of the nonsynonymous SNP 10:18312485. (I) Comparison of the hundred-seed weight of the individuals carrying different genotypes of 10:18312485.

Figure 6

The analysis of the phylogenetic relationship, conserved motifs and gene structures of FAD family genes in paper mulberry. (A) The phylogenetic relationship of FAD proteins from paper mulberry (red filled circle) and Arabidopsis (green filled square). The neighbor-joining tree was constructed through the MEGA X program with 1000 bootstraps. (B) The analysis of the motifs and gene structures of the FAD family genes in paper mulberry. The conserved motifs were analyzed using the online analysis tool MEME.

Figure 7

The analysis of the cis-acting regulatory elements of FAD family genes in paper mulberry. (A) The location of the cis-acting regulatory elements in the upstream 2000-bp of the BpFADs, which were identified through the online analysis tool PlantCARE. (B) The number of the cis-acting regulatory elements in the promoter regions of BpFAD genes. The color scale represents the number of each cis element in every BpFAD gene.

Figure 8

The expression profiles of BpFAD family genes. (A) The expression patterns of BpFAD genes in different tissues based on the FPKM values. The transcript levels of BpFAD genes were shown through the color gradient; green to red represents the transcript levels from low to high. Leaf-A represents the young leaf; Leaf-B represents the developing leaf; Leaf-C represents the climax leaf; Root-A represents the root tip; Root-B represents the taproot; Stem-A represents the apical bud; Stem-B represents the immature stem; Stem-C represents the partially lignified stem; Stem-D represents the mature stem. (B) The expression profiles of BpFAD genes under 4 °C treated with different times based on the FPKM values; blue to red represents the transcript levels from low to high. (C) The expression patterns of BpFAD genes under abiotic stresses and hormone treatments. A quantitative RT-PCR was used to explore the expression levels of BpFAD genes. BpGAPDH was selected as an internal control. Salt: leaves treated with 250 mM NaCl. Drought: leaves treated with 20% PEG6000. ABA: leaves treated with 100 µM ABA. SA: leaves treated with 100 µM SA. Me-JA: leaves treated with 100 µM Me-JA. * represents p < 0.05, ** represents p < 0.01, *** represents p < 0.001.