藜麦是什么样子？

In the map, red-covered area shows the natural distribution of quinoa, and the green parts show the cultivated distribution. Some evidences show that quinoa originated in the South America Andean region of Peru, Bolivia, Colombia and Chile, and was domesticated 3,000 to 4,000 years ago. It has been adapted to the harsh climates of Andean area, which has a wide range of altitudes, temperature and annual precipitation. It is regarded as a halophyte, shows strong resistance to drought, low temperature, soil salinity and frost et al.

Maybe you have seen the sale of quinoa in marketing as a grain. Quinoa has gained international attention because of the nutritional value of its seeds, which are gluten-free, have a low glycemic index, and contain an excellent balance of essential amino acids, fibre, lipids, vitamins, and minerals. FAO declared 2013 to be the "International Year of Quinoa".

So based on above knowledge, we may wonder two questions. First, Why does it have so strong environmental adaptibility compared to other crops? what genes have been selected by environments to tolerate the tough climate and living conditions?(Environmental Adaptibility) Genomic and transcriptome data may shed light on its mechanism. Second, Now that there are so many exceptional nurtritions beneficial to human health and strong environmental adaptibility, why is the cultivating area of quinoa still so limited and small ?(Seed Quality)

With the two questions, we collected 3 papers published recently focusing on quinoa genome and its specific characters. Here I just summarize and list out main contents in each paper, we'd like to answer questions from the following 4 parts.

Genome assembly.The genome size estimate is from 1.45 to 1.5 Gb, and because of its allotetraploidy maybe derived from the genome fusion of its two diploid parental ancestors, we can see secondary peak at 200 which is 2-fold value of the peak.

Here I list out some key pipelines and statistics of 3 versions of genome according the 3 papers.

The first version was published on DNA research last year, they firstly used WGS data to assemble genome and used Pacbio long reads to do gap-closing and scaffolding. Finally they got a genome of 1.1 Gb with scaffold N50 86Kb and contig N50 14Kb.

The second version was published on Nature this year, they used pure Pacbio 3GS data to assemble the genome, also they integrated BioNano, Chicago and linkage map to assist the genome assembly, finally they got a nice result. The total size is ahout 1.4 Gb, bigger than 1st version. Scaffold N50 and contig N50 had improved significantly.

The 3rd version was assembled by Zhu Jiankang group and published on Cell Research 2 months ago. In this work, they assembled with WGS data and Pacbio data independently, and pacbio pipeline contained 2 assemblies using software Facon and Canu independently. Then they merged these 3 versions of assemblies together with HABOT2.

Now we have got the genome, generally, we would talk something about evolution.

In time scale.

Just as what I said before, quinoa is derived from 2 diploids. To understand evolution in quinoa further, A-genome diploid C. pallidicaule and the B-genome diploid C. suecicum was sequenced and assembled by Jarvis group. Here is some statistics about the 2 diploid genomes.

By examining single-copy gene families from eight sequenced plant genomes, they found that both quinoa and spinach, belonging to a subfamily Chenopodiaceae, diverged 16 million years ago, and shared a common ancestor with A. hypochondriacus, diverging about 25 million years ago.

Fourth degenerate sites (4DTv) also identified the critical time points in quinoa genome evolution. Age distribution was calculated using 9 890 paralogous gene pairs after excluding local duplications. A sharp peak could be observed  around 0.028, indicating a recent whole-genome duplication (WGD) event that likely reflects the genome fusion between two parents. The time of fusion is about 4.3 million years ago, which is much later than the divergence between quinoa and spinach or between quinoa and sugar beet.

In space scale.

Multiple inter-fertile tetraploid species have arisen from the ancestral tetraploid following hybridization, including C. berlandieri and C. hircinum, although the evolutionary relationships among quinoa and its diploid and tetraploid relatives remain unclear. To begin to resolve these issues, 15 additional quinoa samples were re-sequenced, representing the two major recognized groups of quinoa: highland and coastal. five accessions of C. berlandieri and one accession each of C. hircinum from the Pacific and Atlantic Andean watersheds were also involved.

Phylogenetic analysis indicates that North American C. berlandieri is the basal member of the species complex. Quinoa was thought to have been domesticated from C. hircinum in a single event, from which coastal quinoa was later derived; however, sequencing data place a C. hircinum sample basal to coastal ecotypes, suggesting maybe quinoa was domesticated independently in highland and coastal environments.

Come back to the question about environmental adaptibility.Phytohormone abscisic acid (ABA) can regulate abiotic stress tolerance.Phylogenetic analysis revealed that gene families involved in ABA signalling were substantially expanded in the quinoa genome compared with other Amaranthaceae plant species. For example, the PYR/PYL/RCAR gene family, which encodes an ABA receptor, plays key roles in ABA signalling, quinoa contains ~2 gene copies due to genome fusion.

Here is shown the biosynthetic pathway of ABA. ZEP /VDE/NSY are all enzymes that catalyse each step. The step catalyzed by NCEDs , and produce the C15 intermediate xanthoxin, is considered to be the rate-limiting step for ABA biosynthesis. Quinoa contains 11 NCEDs, while other plant species contain 4-5. They also identified the expansion of other genes regulating the biosynthesis of ABA and the numbers of these genes are roughly two-fold those in other diploid plants,suggesting that the duplicated ABA de novo synthesis genes were retained after genome fusion in quinoa.

Moreover, quinoa contains 22 PYL genes, while the others contain at most 10. Quinoa contains a higher number of ABCGs, the group of ABC transporters that is utilized for ABA transportation.

The expression of ABA-responsive genes in different quinoa tissues was examined. Among the 108 genes that are assigned the GO term “response to ABA stimulus”, 42% of them have a RPKM value of over 20, indicating that quinoa has a high basal level of ABA response.They also analyzed transcriptome of epidermal bladder cells which are observed in ~50% of halophytic plants. So expansion of gene families involved in stress response, ABA signaling and ion transport and a constitutive stress response at the transcript level were found to be correlated with the remarkable stress tolerance in quinoa.

The second question is about seed quality.

Quinoa seeds contain a mixture of saponins. Although saponins may be beneficial for plant growth, they must be removed before human consumption as they produce a bitter flavour. Because this process is costly, reduce the nutrition of the seeds, the development of saponin-free lines is a major breeding objective in quinoa.

Luckily in nature sweet quinoa that contain very low levels of saponins are present. To identify these genes regulating the presence of saponins, They performed linkage mapping and bulk segregant analysis (BSA) using two populations segregating for the presence of saponins in the seeds: Kurmi (sweet)×0654 (bitter), and Atlas (sweet)×Carina Red (bitter). Segregation ratios of 3:1 in 2 populations indicate that the presence of saponins is controlled by a single gene, with the presence of saponins being dominant.

Linkage mapping and BSA in each population identified the same region on CqB16 that distinguishes the bitter and sweet lines. Frequencies of the sweet allele in both populations reached 100% for markers located in scaffold 3489. Of the 54 annotated genes in this region, two are similar to genes previously shown to play a role in saponin biosynthesis. AUR62017204 and AUR62017206 are neighbouring genes annotated as basic helix-loop-helix (bHLH) transcription factors that are known to regulate triterpenoid biosynthesis.

In Medicago truncatula, overexpression of the triterpene saponin biosynthesis activating regulator 1 (TSAR1) and TSAR2 bHLH transcription factors was recently shown to increase the expression of genes in the triterpenoid biosynthetic pathway, resulting in increased accumulation of triterpene saponins. TSAR1 and TSAR2 were also found to bind to the DNA motif 5'-CACGHG-3'(where H can be A, C, or T).

In quinoa, they found AUR62017206 (TSARL2) was expressed in root tissue but not in flowers or immature seeds, whereas AUR62017204 (TSARL1) was almost exclusively expressed in seeds, with significantly lower expression levels in sweet lines.

They identified the DNA motif within 2 kb upstream of the start codon in several saponin biosynthetic pathway genes in quinoa. Expression levels of these genes in the saponin biosynthetic pathway were significantly downregulated in sweet lines. Together, these results suggest that TSARL1 might be a functional TSAR orthologue.

The TSARL1 transcript was alternatively spliced in the sweet progeny of Kurmi and 0654. A SNP in the last position of exon 3 (G2078C) co-segregates with the presence of saponins in the Kurmi×0654 progeny. The SNP alters the canonical intron/exon splice boundary, probably leading to the alternative splicing at an upstream cryptic splice site in the sweet lines. This alternative splicing of TSARL1 results in a premature stop codon and a truncated protein that modelling predicts to be compromised in its ability to form homodimers and to bind DNA. All bitter strains in re-sequencing pool share the same allele (G) found in the bitter progeny of Kurmi and 0654.

However, none of the sweet progeny in the Atlas×Carina Red population were found to have the G2078C allele. Additional sequencing of individual plants of the Atlas variety revealed a low level of heterogeneity within the variety for the TSARL1 gene, with some plants containing the G2078C allele and others containing sequence insertions. Strong evidence of insertions in and around the TSARL1 gene in all the sweet progeny of the Atlas×Carina Red population was found. In particular, two exonic insertions in TSARL1 in the sweet progeny probably inactivate the gene and result in a sweet phenotype.

Identification of multiple, independent mutations in TSARL1 that co-segregate with the sweet phenotype strongly suggests that this gene regulates the presence and absence of saponins in quinoa seeds.