Haplotype-phased genome and evolution of phytonutrient pathways of tetraploid blueberry

Abstract

Background: Highbush blueberry (Vaccinium corymbosum) has long been consumed for its unique flavor and composition of health-promoting phytonutrients. However, breeding efforts to improve fruit quality in blueberry have been greatly hampered by the lack of adequate genomic resources and a limited understanding of the underlying genetics encoding key traits. The genome of highbush blueberry has been particularly challenging to assemble due, in large part, to its polyploid nature and genome size.

Findings: Here, we present a chromosome-scale and haplotype-phased genome assembly of the cultivar "Draper," which has the highest antioxidant levels among a diversity panel of 71 cultivars and 13 wild Vaccinium species. We leveraged this genome, combined with gene expression and metabolite data measured across fruit development, to identify candidate genes involved in the biosynthesis of important phytonutrients among other metabolites associated with superior fruit quality. Genome-wide analyses revealed that both polyploidy and tandem gene duplications modified various pathways involved in the biosynthesis of key phytonutrients. Furthermore, gene expression analyses hint at the presence of a spatial-temporal specific dominantly expressed subgenome including during fruit development.

Conclusions: These findings and the reference genome will serve as a valuable resource to guide future genome-enabled breeding of important agronomic traits in highbush blueberry.

Keywords: Blueberry; Genome; Haplotype-phased; Phytonutrients; Polyploid; Subgenome Dominance; Tetraploid; Vaccinium.

Figure 1:

The haplotype-phased chromosome-scale highbush blueberry genome. (a) Collinearity among the homoeologous chromosomes. The gray lines represent conserved gene arrays between chromosomes. Chromosomes were drawn proportionally with respect to the number of genes on each chromosome. (b) Gene and transposable element (TE) density and LTR assembly index (LAI) in chromosomes 1–12 plotted in 300 Kb sliding window using Circos. The tracks from outside to inside are: 1 = chromosomes, 2 = gene density, 3 = TE density, and 4 = LAI score.

Figure 2:

Assessment of the origin of polyploid blueberry. (a) Gene content comparison of homoeologous chromosomes (1, 13, 25, and 37) plotted along 2,725 collinear syntenic regions. This analysis for all 48 chromosomes can be regenerated here: [65]. (b) Gene expression comparison (FPKM; fragments per kilobase per million) among the same four homoeologous chromosomes across different blueberry tissues (1 = flower bud; 2 = flower at anthesis; 3 = petal fall; 4 = green fruit; 5 = pink fruit; 6 = ripe fruit; 7, 8 = leaf collected at 12 p.m. and 12 a.m., respectively; 9, 10, 11 = methyl jasmonate-treated leaf collected after 1 hour, 8 hours, and 24 hours, respectively; 12 = shoot; 13 = root; 14 = salt-treated root).

Figure 3:

A schematic presentation of flavonoid biosynthesis in blueberry. (a) Predicted flavonoid biosynthetic pathway leading to production of anthocyanin. The proposed pathway is based on previously described flavonoid biosynthetic pathway in plants (Zifkin et al. [68]) and expression of predicted anthocyanin biosynthetic genes in blueberry. The core genes include phenylalanine ammonia-lyase (PAL), 4-hydroxycinnamoyl CoA ligase (4CL), trans-cinnamate 4-monooxygenase (C4H), cytochrome P450 98A3 (C3H), chalcone synthase (CHS), chalcone flavonone isomerase (CHI), flavanone-3β-hydroxylase (FHT), flavanone 3-hydroxylase (F3H), flavonoid 3′-hydroxylase (F3′H), flavonoid 3′,5′-hydroxylase (F3′5′H), dihydroflavonol reductase (DFR), leucoanthocyanidin reductase (LAR), anthocyanidin reductase (ANR), anthocyanidin synthase (ANS), UDP-glucose flavonoid 3-O-glucosyl transferase (UFGT), anthocyanin-O-methyltransferase (OMT), hydroxycinnamoyl-CoA shikimate/quinate hydroxycinnamoyltransferase (HCT), and hydroxycinnamoyl-CoA quinate hydroxycinnamoyltransferase (HQT). (b) Hypothetical regulatory pathway of anthocyanin biosynthetic genes based on the proposed model by Albert et al. [92]. (c) Developmental-specific expression pattern of key anthocyanin biosynthetic gene (green triangles = examples of genes upregulated during early fruit growth; red circles = examples of genes upregulated during late fruit development). (d) Chlorogenic acid biosynthetic genes (1 = petal fall, 2 = small green fruit, 3 = expanding green fruit, 4 = pink fruit, 5 = fruit color completely changed from pink to purple, 6 = unripe, 7 = ripe). (e) Expression profile of transcription factors predicted to regulate anthocyanin biosynthesis in blueberry. A high-resolution version of the heat maps is available on PURR (see Availability of Supporting Data section).