Chromosome-level genome assembly of bunching onion illuminates genome evolution and flavor formation in Allium crops

Liao, Nanqiao; Hu, Zhongyuan; Miao, Jinshan; Hu, Xiaodi; Lyu, Xiaolong; Fang, Haitian; Zhou, Yi-Mei; Mahmoud, Ahmed; Deng, Guancong; Meng, Yi-Qing; Zhang, Kejia; Ma, Yu-Yuan; Xia, Yuelin; Zhao, Meng; Yang, Haiyang; Zhao, Yong; Kang, Ling; Wang, Yiming; Yang, Jing-Hua; Zhou, Yan-Hong; Zhang, Ming-Fang; Yu, Jing-Quan

doi:10.1038/s41467-022-34491-3

Download PDF

Article
Open access
Published: 05 November 2022

Chromosome-level genome assembly of bunching onion illuminates genome evolution and flavor formation in Allium crops

Nature Communications volume 13, Article number: 6690 (2022) Cite this article

13k Accesses
25 Citations
47 Altmetric
Metrics details

Subjects

Abstract

The Allium genus is cultivated globally as vegetables, condiments, or medicinal plants and is characterized by large genomes and strong pungency. However, the genome evolution and genomic basis underlying their unique flavor formation remain poorly understood. Herein, we report an 11.27-Gb chromosome-scale genome assembly for bunching onion (A. fistulosum). The uneven bursts of long-terminal repeats contribute to diversity in genome constituents, and dispersed duplication events largely account for gene expansion in Allium genomes. The extensive duplication and differentiation of alliinase and lachrymatory factor synthase manifest as important evolutionary events during flavor formation in Allium crops. Furthermore, differential selective preference for flavor-related genes likely lead to the variations in isoalliin content in bunching onions. Moreover, we reveal that China is the origin and domestication center for bunching onions. Our findings provide insights into Allium genome evolution, flavor formation and domestication history and enable future genome-assisted breeding of important traits in these crops.

Chromosome-level genomes of three key Allium crops and their trait evolution

Article 06 November 2023

Chromosome-scale genome assembly provides insights into the evolution and flavor synthesis of passion fruit (Passiflora edulis Sims)

Article Open access 08 January 2021

Combined genomic, transcriptomic, and metabolomic analyses provide insights into chayote (Sechium edule) evolution and fruit development

Article Open access 31 January 2021

Introduction

The Allium genus is naturally distributed geologically in the northern hemisphere with a major center of diversity ranging from the Mediterranean Basin to Central Asia^{1, 2}. It possesses strong adaptability to seasonal drought and harsh conditions. Awareness of the potential of Allium vegetables to enhance human health is driving the increasing demand for these functional crops. In addition, the Allium genus is notable for a unique set of sulfoxides derived from cysteine^3,4,5, which contribute to their characteristic flavor, medicinal properties, disease resistance, and pest repellency^5,6,7. Compared to quality-related components such as sugar, organic-acid, alkaloid-, and terpene-derived metabolites, less is known about the biosynthesis and regulation of Allium-specific alk(en)yl cysteine sulfoxides (ACSOs)^8,9,10, such as alliin, isoalliin, and methiin. Among the major ACSOs in the Allium genus, isoalliin can be hydrolyzed to propanthial S-oxide (commonly known as lachrymatory factor) with the aid of alliinase (ALL) and lachrymatory factor synthase (LFS)³, whereas alliin can be spontaneously catalyzed into allicin upon tissue damage^{3, 4, 11}. However, the genomic basis of this unique flavor formation in Allium crops remains unclear.

The Allium genus possesses some major cash crops including bunching onions, Welsh onions or storey onions (A. fistulosum), bulb onions (A. cepa), garlic (A. sativum), and shallots (A. ascalonicum). Owing to their morphological resemblance, weak reproductive isolation, and similar chromosome karyotypes, the genetic relationships among Allium species are not well defined^{2, 12}. Traditional Allium taxonomy is mostly based on plant morphology^{9, 13}. For instance, A. fistulosum is generally classified based on its tillering characteristics and pseudo-stem morphology. Shallots (A. ascalonicum) are often mistaken for bunching onions because of their similar tillering characteristics. Chinese red onions (A. cepa var. proliferum) are often confused with storey onions (A. fistulosum L. var. viviparum), because of their similar aerial bulbs. These ambiguities in Allium taxonomy often leads to misunderstanding of the evolution and domestication in this genus, which needs to be clarified at the genomic level.

Natural or artificial interspecific hybridizations are ubiquitous among the Allium genus, which have been empirically and randomly utilized to introduce disease resistance and quality traits^14,15,16,17. Genome-informed interspecific hybridization could be an effective strategy to expand Allium genetic diversity for desirable agronomic traits. However, the Allium genera are notorious for their large genomes with a high proportion of repetitive sequences, which consequently lead to lack of high-quality genome assemblies in this genus^{18, 19}. As a result, genome-assisted breeding of these important crops lags that of crops with relatively small genomes. Therefore, genomic information is highly demanded and desirable for Allium research and breeding communities.

Here, we report a chromosome-scale reference genome assembly of A. fistulosum obtained by combining PacBio, Bionano, HiC, and Illumina sequencing technology toolkits. Comparative genomics and genome collinearity analyses have robustly illuminated the genome evolution and relatedness in Allium crops. In addition, we investigate the evolution of genes involved in the biosynthesis and hydrolysis of ACSOs to elucidate the genomic mechanisms underlying the special flavor formations in Allium crops. Furthermore, 135 Allium accessions are re-sequenced to clarify the phylogenomic evolution and migration routes of the Allium crops. Additionally, transcriptomic and metabolic analyses are performed to explore the formation of sulfur-containing flavors in Allium crops. The present study provides insights into genome evolution and expansion of the Allium species and will enable genomic-aided breeding.

Results

Chromosome-scale mega-genome assembly and gene annotations

We assembled a chromosome-level genome of the bunching onion (A. fistulosum) using de novo genome sequencing of the bunching onion landrace. First, the k-mer analysis (k = 17) revealed an estimated genome size of 11.97 Giga base-pairs (Gb), a heterozygosity rate of 0.64%, and the repetitive sequences accounted for 89.89% of the whole genome (Supplementary Fig. 1 and Supplementary Table 2). PacBio sequences (1,649.82 Gb, 138.1 × genome coverage) were generated to assemble the genome into contigs, yielding a draft assembly of 27,972 contigs with a total length of 15.36 Gb and contig N50 of 4.72 mega base-pairs (Mb, Supplementary Table 1). After polishing and quality improvement, an improved assembly (scaffold N50 of 8.98 Mb) was obtained (Table 1 and Supplementary Table 3). Finally, Hi-C interaction datasets (1,136.4 Gb, 95.1 × genome coverage) were performed to construct the genome into super-scaffolds, and 98.71% of the assembled contigs were anchored into eight chromosomes (Fig. 1b and Supplementary Table 1, 3–5). The final genome assembly was 11.27 Gb with a contig N50 of 7.34 Mb and a super scaffold N50 of 1.34 Gb (Fig. 1, Table 1, and Supplementary Tables 3-4), which represents the best contiguity among the currently released Allium genome assemblies.

Table 1 The main genome assembly features of three major Allium crops

Full size table

**Fig. 1: High-quality genome assembly of the *A. fistulosum*.**

Multiple genome assessments supported the high quality of A. fistulosum genome assembly. First, Benchmarking Universal Single-Copy Orthologue (BUSCO, 91.0% of the 1614 core eukaryotic genes) and Core Eukaryotic Gene Mapping Approach (CEGMA, 94.35% of conserved genes) analyses indicated the high completeness of the assembled genome (Supplementary Tables 6, 7). Meanwhile, the consensus quality value (QV) and the completeness of the A. fistulosum genome were 37.06% and 93.61%, respectively, indicating the high accuracy of our assembly²⁰. By mapping short Illumina reads to the genome assembly, we obtained a mapping rate of 99.65%, a genome coverage of 99.34%, and an SNP rate of 0.0063%, demonstrating the high quality of the assembled genome (Supplementary Table 8). In addition, we observed good collinear relationships between the previous genetic map²¹ and the assembled chromosomes (Supplementary Fig. 2), which further supported our high-quality genome assembly. Regardless of the large genome size, the high long-terminal repeat retrotransposon assembly index (LAI, 24.81) suggested the gold continuity of the A. fistulosum genome assembly²².

A total of 7885 Mb (69.81% of the total genome) of repetitive sequences was annotated using RepeatMasker, TRF, and RepeatProteinMask (Supplementary Table 9). Most of the repetitive sequences were transposable elements (TEs), whereas long-terminal repeat (LTR) retrotransposons were the most abundant TEs in the A. fistulosum genome (62.18% of the total genome; Supplementary Fig. 3a, Supplementary Table 10). Complementary methods using de novo gene prediction, protein-based homology searches, and transcriptome-based predictions have been used to annotate protein coding sequences. In total, 62,255 genes were predicted, with an average gene length of 5000.13 base pairs (bp), an average gene coding sequence length of 820.11 bp, an average of 3.93 exons per gene, an average exon length of 208.47 bp, and a gene annotation rate of 98.30% (Supplementary Tables 11–13). Moreover, approximately a quarter (14,862 of 62,259) of the predicted genes were inserted by or overlapped with LTRs (Supplementary Fig. 3b). It has been reported that LTR in/around genes provide potential for their transduction, duplication and recombination, as well as alternative splicing and epigenetic control^{23, 24}. Thus, we hypothesized that gene duplication or movement in A. fistulosum might occur along with the proliferation and transposition of LTRs. We also identified 1361 miRNA, 4365 tRNA, 2832 rRNA, and 3444 snRNAs in the A. fistulosum genome using the tRNAscan-SE pipeline (Supplementary Table 14).

Allium genome evolution and chromosome homology

To explore the genomic evolution of Allium, we selected 13 species from Asparagale, Arecales, Poales, Amborellaceae, and Scitamineae for further comparative genomic analysis. Among these plants, we discovered that the three Allium crops had the highest number of total and unique gene families (Fig. 2a). In total, 492 single-copy gene families were selected to reconstruct a maximum-likelihood (ML) tree, which illustrated that A. fistulosum and A. cepa diverged about 7.4 million years ago (MYA), and the common ancestor of A. fistulosum and A. cepa diverged from A. sativum about 16.7 MYA^{11, 25} (Fig. 2a). Among the selected genomes, the Allium crops exhibited numerous expanded and contracted genes, and 308 gene families were commonly expanded in all three Allium crops. Importantly, we found these gene families predominately enriched in gene ontology (GO) pathways like “Catalytic activity,” “Carbon-sulfur lyase activity,” and “Cysteine-type peptidase activity,” implying that the high sulfur-metabolism activities of Allium crops are responsible for the biosynthesis of sulfur compound related to flavor formation (Fig. 2a, b).

**Fig. 2: Genome evolution and relatedness of *Allium* crops.**

The distribution of Ks values of homologous pairs in the syntenic regions of A. fistulosum and A. sativum was similar to those of A. sativum and Asparagus officinalis¹¹ (Supplementary Fig. 4), indicating that, similar to garlic, A. fistulosum also underwent three WGD events. In addition, most of the genes of the three major Allium crops corresponded to a single block in the collinearity analysis. Therefore, no unique WGD event occurred among these three genomes after divergence (Supplementary Fig. 5). Previously, a high level of macro-synteny between A. fistulosum and A. cepa has been reported using 103 anchor markers²¹. Here, we observed an almost one-to-one syntenic relationship at the chromosome level, except for Chr4, which underwent a clear inversion between the two Allium crops (Fig. 2c, Supplementary Fig. 6). In addition, four A. fistulosum chromosomes (Chr3, Chr6, Chr7, and Chr8) exhibited nearly one-to-one syntenic relationships with those of A. sativum (Chr2, Chr8, Chr3, and Chr1; Fig. 2c, Supplementary Fig. 6). The remaining four chromosomes exhibited several rearrangements. In details, a clear inversion was observed between Chr1 of A. fistulosum and Chr7 of A. sativum, Chr2 of A. fistulosum was partly syntenic to Chr5 and Chr6 of A. sativum, Chr4 of A. fistulosum showed partially syntenic relationships with Chr4 and Chr5 of A. sativum, and Chr5 of A. fistulosum was partially rearranged from Chr4 and Chr6 of A. sativum (Fig. 2c, Supplementary Fig. 6). Even though there were significant gaps among their genome sizes, high chromosome relatedness was observed among the three Allium genomes, which indicated that they were evolutionarily close to each other.

Insights into genome expansions and gene duplications

Allium plants have long been known for their large genomes, which are among the largest in vegetable crops²⁶. Therefore, we analyzed the key factors potentially associated with Allium genome expansion. We identified LTRs as the leading genomic constituents in the A. fistulosum (Supplementary Fig. 3a and Supplementary Table 10) and A. sativum genomes¹⁰. Gypsy was the predominant LTR and burst mostly within the last 2 million years (Fig. 3a and Supplementary Fig. 7). Within Gypsy, the Tekey and Tat clades were the most abundant LTRs in both A. fistulosum and A. sativum genomes. The CRM clade contributed significantly (10.38%) to the LTR components only in A. fistulosum (Fig. 3b). Although Copia LTRs comprised a small proportion of repeat sequence in both crops, their components significantly differed between the A. fistulosum and A. sativum genomes (Fig. 3a, Supplementary Figs. 7 and 8). Therefore, the accumulation of LTRs, especially the Gypsy-type, is responsible for genome enlargement in Allium crops, while uneven expansions of different LTR clades lead to diversity in Allium genome constituents.

**Fig. 3: Extensive genome expansions and gene duplications in *Allium* crops.**

The three Allium crops showed the highest number of total and unique gene families among the 13 selected genomes (Fig. 2a). To this end, we investigated the global landscape of gene duplications in 16 species with genome sizes ranging from 0.2 to 28 Gb. The large genomes had a relatively high frequency of dispersed duplication (DSD) events, demonstrating that DSD events play an important role in the gene expansion of large genomes (Fig. 3c, Supplementary Fig. 9). After classifying the duplicated paralogs into ancestral and new DSD genes based on synteny²⁷, we identified 13,676, 17,817, and 8424 gene pairs (ancestral/new) in A. fistulosum, A. sativum, and A. cepa, respectively. The ancestral DSD genes exhibited significantly higher expression levels than the new genes (Fig. 3d). The expression bias between ancestral and new DSD genes might result from their different environmental responses or functional redundancies, which also implies the possible pseudogenization, sub-functionalization, or neo-functionalization after gene duplications. We propose that ubiquitous DSD events are responsible for gene expansion and divergence in the Allium genomes.

Expansions and differentiations of flavor-related genes

To explore the evolutionary events of the putative genes involved in Allium flavor formation, we identified all ACSO-related genes in the 14 plant genomes (Fig. 4a). Seven γ-glutamylcysteine synthetase (GSH1) orthologs, one glutathione synthetase (GSH2) ortholog, one phytochelatin synthase (PCS) ortholog, three γ-glutamyl transpeptidase (GGT) orthologs, and three flavin-containing monooxygenase (FMO) orthologs were identified in the A. fistulosum genome (Fig. 4a). Among these genes, AfGSH1b, AfGSH1g, AfGSH2, AfPSC1, AfGGT1, AfGGT2, AfFMO1, and AfFMO2 were constitutively expressed in the roots, pseudo-stems, leaves, floral stalks, and flowers, indicating that these genes might play roles in ACSOs biosynthesis in Allium fistulosum (Fig. 4b). The ALL gene encoding a key enzyme for ACSOs hydrolysis was found to be extensively expanded in bunching onion (56 AfALLs), bulb onion (65 AcALLs), and garlic (48 AsALLs) genomes³ compared with other selected genomes (Fig. 4a, Supplementary Fig. 10). Importantly, 43, 27, and 29 LFS orthologs were identified in the genomes of A. fistulosum, A. cepa, and A. sativum, respectively, whereas no LFS genes were found in species other than Allium among the selected genomes (Fig. 4a, d).

**Fig. 4: Massive duplication and differentiation of sulfur-related genes in *Allium*.**

According to our ACSOs contents analysis and relevant report⁴, isoalliin was the major sulfur-containing bioactive substance in both A. fistulosum and A. cepa, whereas alliin was not detected in these two crops. However, alliin was the dominant sulfoxide in A. sativum and a small amount of isoalliin was detected (Fig. 4c). LFS is the key enzyme that catabolizes hydrolyzed isoalliin into the characteristic propanthial S-oxide (lachrymatory factor) in Allium crops. We classified the LFS gene family into four groups, LFS-a, LFS-b, LFS-c, and LFS-d, according to their phylogenetic structure annotated in three major Allium crops (Fig. 4d). In addition, BAC21275.1 was verified as a functional LFS in bulb onions²⁸ and was clustered into the LFS-c family (Fig. 4d). AfLFS01, the best hit with BAC21275.1 in bunching onions, exhibited the highest transcriptional levels in most tissues, except roots, among all the AfLFS genes examined (Fig. 4d). Importantly, eight AfLFS, seven AcLFS and three AsLFS genes in this group presented similar expression patterns (Fig. 4d). Based on the close phylogenetic relationships with BAC21275.1 and their similar expression patterns, we suppose that the LFS-c group might play synchronized roles in ACSOs hydrolysis in Allium crops. However, as the phylogenetically closest duplications, AfLFS15 and AfLFS16 showed different expression patterns against AfLFS01 (Fig. 4d), suggesting the potential for functional differentiation of these homologous genes. We also observed different expression biases of the LFS groups in these crops. A. fistulosum and A. cepa preferred to extensively express LFS-c group members, whereas A. sativum predominately expressed LFS-a group members (Fig. 4d). Similarly, three proximally duplicated genes (AfALL47, AfALL48, and AfALL49) in Chr4 of bunching onion, four AcALL genes (AcALL05, AcALL08, AcALL09, and AcALL47) and four tandemly duplicated AsALL genes (AsALL31, AsALL32, AsALL33, and AsALL34) belong to the same cluster as known ALL genes (AAD26853.1, AAA32639.1, AAB32477.1, and ACN78838.1; Supplementary Fig. 10), all of which exhibited high expression levels in most tissues. These findings suggested that this gene cluster may be functional ALL genes in these three crops.

The LFS-b group consisted of nine AfLFS genes that exhibited extremely low gene expression in all tissues (Fig. 4d). In contrast, 12 AsLFS genes clustered in an independent clade in the LFS-d group and were mainly expressed in the roots (Fig. 4d). Furthermore, the micro-collinearity analysis indicated that several LFS and ALL genes in bunching onion and garlic exhibited reciprocal one-to-many relationships (Fig. 5a, b). Thus, we inferred that the independent expansion and differentiation of LFS and ALL gene families after the divergence of bunching onion and garlic from their common ancestor might relate to their flavor diversity (Figs. 4d, 5a, b and Supplementary Fig. 10). Notably, the dispersed duplication, tandem duplication, and proximal duplication events of the LFS and ALL gene families were attributed to gene duplications (Fig. 5c, d, Supplementary Data 1). In particular, 33 AfLFSs, among 43 duplicates, were clustered in the form of tandem repeats within a 10 Mb region on Chr5 and the intertwined coexistence of LTRs and LFS genes in this region suggested that LTRs likely contribute to LFS gene family proliferation (Fig. 5c, Supplementary Fig. 11). We concluded that the independent expansion and differentiation of sulfur-metabolism-related genes prevail in Allium genomes, which might relate to the characteristic pungent flavor evolution in Allium crops.

**Fig. 5: Independent expansions of *LFS* and *ALL* gene families between bunching onion and garlic.**

Population structure and migration routes of A. fistulosum

We re-sequenced 135 diverse Allium accessions to elucidate the phylogenetic relationships of A. fistulosum crops and their close relatives (Supplementary Figs. 12–13 and Supplementary Data 4). We generated 16,453.99 Gb Illumina short-reads with an average depth of 9.49 × genome coverage. In total, 48,218,339 SNPs and 18,182,723 insertions and deletions (InDels) with an average of 4.08 SNPs and 1.54 InDels per kb were identified. The maximum-likelihood tree of 135 Allium accessions revealed four main clusters centered on Chinese red onions (A. cepa var. proliferum), shallots (A. ascalonicum), A. fistulosum Group 1, and A. fistulosum Group 2 (Fig. 6a). All nine Chinese red onions and the A. altaicum accession clustered tightly into one group. Twenty-two shallots clustered together and displayed two distinct subgroups, whereas the remaining two shallots were close to the Chinese red onion group (Fig. 6a). Chinese red onions and shallots clustered apart from A. fistulosum, fitting well with their relatively low genome mapping rate (79–89%) to our reference genome. Therefore, we concluded that Chinese red onions and shallots are close relatives or sibling species of A. fistulosum. Ninety-six A. fistulosum accessions were clustered into two groups; the remaining five A. fistulosum accessions were clustered with shallots and Chinese red onions in the cladogram, which was indicative of ambiguity or mis-assignment in the Allium taxonomy (Fig. 6a and Supplementary Data 4).

**Fig. 6: Population structure and the plausible migration routes of *A. fistulosum*.**

To clarify the evolution and migration of A. fistulosum, we performed phylogenetic relationships and population structure analyses with 96 affirmed A. fistulosum accessions, using A. altaicum as an outgroup. The population structure of A. fistulosum was investigated and the cross validation (CV) error was minimized when K = 5 (Supplementary Fig. 14). The 96 A. fistulosum accessions could be categorized into five subgroups and fell into two groups (Fig. 6a, b). Group 1 (G1) could be divided into three subgroups with distinctive geographical distributions based on the population structure. G1-1 accessions were mainly distributed in western China, G1-2 accessions were distributed in Middle Asia and the Russian Federation, and G1-3 accessions were distributed in the northern and northeastern areas of China. Group 2 (G2) also displayed two differentiable geographic subgroups, with accessions from southeastern China clustered in subgroup G2-1. Accessions distributed in Japan, America, and Europe showed high genomic similarities and were clustered as G2-2 (Fig. 6b, e).

Furthermore, the principal components analysis (PCA) revealed that G1-2 and G2-2 clustered individually, whereas the three China-originated groups (G1-1, G1-3, and G2-1) showed a closer relationship (Fig. 6c). G1-3 possessed the lowest linkage disequilibrium (LD) decay, indicating that north and northeast China might be a diversity center of A. fistulosum (Supplementary Fig. 15). In contrast, G1-2 showed a unique structural pattern (when K = 3, 4, 5, and 6) with relatively low nucleotide diversity (π) and high LD decay (Fig. 6b, d and Supplementary Fig. 15). Furthermore, the fixation indexes (F_ST) of G1-2/G1-1 (0.0783), G1-2/G1-3 (0.0891), G1-2/G2-1 (0.1095), and G1-2/G2-2 (0.1491) were higher than those of other subgroup pairs (Fig. 6d). The long leading branch of G1-2 (Fig. 6b) suggested that there might be an early divergence and independent domestication of G1-2 group, which also partly explains the relatively independent PCA clustering of G1-2 (Fig. 6b, c). Moreover, four accessions from western China (G1-1) clustered with Middle Asian accessions (G1-2, Fig. 6a), and there was a clear gene flow from the G1-1 group to the G1-2 group (Supplementary Fig. 16), indicating that Middle Asian accessions might have migrated from western China. Besides, the genetic structure of G2 exhibited a gradually changing trend from G2-1 to G2-2. When K = 2 and 3, G2-1 possessed both G1-3/G1-1 and G2-2 genetic structures, which illustrated that the phylogenetic relationship between G2-1 and G1 was closer than that between G2-2 and G1(Fig. 6b). In addition, F_ST values between G1 and G2-1 were lower than those between G1 and G2-2 (Fig. 6d). Together with their geographical locations, we speculated that the Japanese, American, and European accessions (G2-2) migrated or were introduced from southeast China (G2-1, Fig. 6e).

Genomic signals of isoalliin variations in A. fistulosum

We quantified the isoalliin contents of 91 A. fistulosum accessions to determine the genomic basis of the isoalliin variations. The isoalliin content of G1-2 accessions were observed to be significantly higher than those of G1-3 accessions (Fig. 7a). By comparing the selective sweeps between these two groups, we discovered 2718 and 2133 genes located in the selective regions of G1-2 and G1-3, respectively. Several sulfur-flavor-formation-related genes, such as AfGSH2, AfFMO3, AfGGT3, AfALL48, AfALL49, AfALL55, and AfLFS11 were under selection in G1-2. AfPCS1 and AfALL42 were identified to be under selection in G1-3 (Fig.7b). Moreover, a positive correlation between the endogenous isoalliin amount and expression levels of several ACSO-biosynthesis or -hydrolysis genes was observed in the leaves of six accessions with significant difference in isoalliin content, among which AfGSH2 and AfPCS1 exhibited the highest correlation (Fig. 7c). Among the 25 flavor-formation-related genes that were highly expressed in low-isoalliin accessions (A12, C155, and C16), 19 ACSOs hydrolysis-related genes (ALL and LFS) were identified, implying that both weak biosynthesis and strong hydrolysis may contribute to their lower isoalliin accumulation. Among these 19 ACSOs hydrolysis-related genes, differential LFS or ALL gene expression patterns in six accessions were observed (Fig. 7c), implying that the extensive expansion of ACSOs hydrolysis-related genes might offer more evolutionary opportunities for the variation of isoalliin contents in bunching onion.

**Fig. 7: Insights into the variation of isoalliin accumulations in bunching onion accessions.**

Discussion

In the present study, we generated a chromosome-level mega-genome of A. fistulosum, with a genome size of 11.27 Gb and contig N50 of 7.34 Mb (Table 1). Our reference genome exhibited superior quality based on both major assembly indicators and several assessments (BUSCO, CEGMA, LAI, and Merqury; Supplementary Tables 6 and 7). Undoubtedly, the achievement of this genome assembly will accelerate our understanding of the evolution and relatedness of major Allium crops. The genome enlargements of both bunching onion and garlic crops arise from bursts of common Gypsy-type LTRs, but they exhibit uneven expansions of the LTR subclades. Meanwhile, DSD events dominated the contributions to gene expansion in major Allium crops (Fig. 3). As highly conserved genome synteny has been observed in bunching onion and bulb onion with an SSR-markers based genetic map²¹, high genome collinearity between bulb onion, bunching onion, and garlic was shown at the chromosome level. The chromosomes of the bunching onion (A. fistulosum) and bulb onion (A. cepa) showed almost one-to-one collinearity (Fig. 2c, Supplementary Fig. 6), indicating high consanguinity between these two Allium crops. Therefore, the estimated sequence of each chromosome of the bunching onion can be used as a template for onions with high commonality in the chromosome structure. Meanwhile, this genomic evidence well explains the weak reproductive isolation and ubiquitous existence of interspecific hybridization among Allium crops^14,15,16,17, which provides valuable information for directing future genome-aided breeding of Allium crops.

The Allium genus is characterized by unique flavors and medicinal properties owing to its specific cysteine-derived sulfur compounds^3,4,5. We observed that sulfur-metabolism-related genes in GO terms of “Carbon-sulfur lyase activity” and “Cysteine-type peptidase activity” were specifically expanded in the three Allium crops examined (Fig. 2b). Importantly, extensive expansion of ALL and LFS gene families is a striking feature of Allium crops. In intact tissues, ACSOs are stored in the cytosol of mesophyll cells. Upon tissue damage, the vacuolar ALL enzyme contacts and hydrolyzes ACSOs to produce sulfenic acids, which are further converted into various sulfur-containing bioactive compounds via spontaneous reactions⁵.Thus, we speculated that the extensive expansion of ACSOs hydrolysis-related genes might lead to the over-accumulation of related transcripts and resultant enzymes, which might be helpful for quick responses to external stimuli. In particular, the LFS gene family was detected only in Allium crops among the 14 selected genomes (Fig. 4a). These results strongly support that the Allium crops evolved large and special ALL and LFS gene families possibly to resist diseases and insects, by utilizing its volatile and pungent end-products^5,6,7.

The species-specific clustering of ALL and LFS gene families indicated their independent expansions in the three Allium crops (Fig. 4, Supplementary Fig. 10), implying that their pungent flavor might have resulted from convergent evolution. It is known that the activity of LFS is reduced to 1/100 or less even by replacing only one amino acid²⁹. Therefore, we speculate that the extensive duplication and differentiation of the LFS gene family may lead to the difference of protein activities or functional diversities. LFS genes with high activity and high expression levels are likely involved in the production of tearing properties and characteristic flavors in Allium crops. Those members with no or negligible LFS activity, might act in other secondary metabolic pathways instead. It has been reported that different terpene synthase (TPS) subgroups participate in different bioactive terpene biosynthesis³⁰. Similarly, A. fistulosum and A. cepa, preferred to extensively express LFS-c group members, which may function in isoalliin hydrolysis. However, A. sativum predominately expresses both LFS-a and LFS-c members (Fig. 4c, d), which might act on alliin and isoalliin decompositions. ACSOs can be decomposed into various sulfur constituents including diallyl and methyl allyl³¹, thus, we conjecture that different clades of the LFS and ALL gene families might play roles in formulating other sulfur-containing bioactive compounds, which need to be further verified in future genetic and biochemical studies.

Moreover, the expansion of ALL and LFS gene families were possibly synchronized with the proliferation of LTRs (Supplementary Fig. 11)^{23, 24}. In addition, FMO genes, involved in the biosynthesis of both ACSOs and glucosinolates, underwent extensive expansions only in Arabidopsis among the 14 selected genomes (Fig. 4a), which might be related to the accumulation of sulfur-containing glucosinolates in Cruciferous crops^{32, 33}. However, the absence of several glucosinolate-related genes in Allium crops suggested that they might select different evolutionary directions toward the biosynthesis of sulfur-containing bioactive compounds, compared with Cruciferous plants (Fig. 4a and Supplementary Fig. 17).

Traditional Allium categories cause confusion in evolutionary studies and exploitation of agricultural usage, mainly because of the lack of genome information. Previously, A. altaicum was commonly considered the ancestor of A. fistulosum based on its phenotypic resemblance and noncoding chloroplast DNA similarity^{12, 34} (Supplementary Fig. 12). Here, we found that it did not cluster with A. fistulosum in either SNP-based phylogeny or PCA analyses, but was phylogenetically closer to the Chinese red onion (Fig. 6a and Supplementary Data 4). Thus, we postulated that A. altaicum is closely related to A. fistulosum, rather than its proposed ancestor. In contrast to earlier morphological classifications, three storey onions (A. fistulosum var. viviparum) accessions and the Chinese red onions (A. cepa var. proliferum), propagated by aerial bulbs, were clustered into two different groups (Fig. 6a, Supplementary Fig. S18 and Supplementary Data 4). Their characterized aerial bulbs, which could be used for asexual propagation, may have resulted from convergent evolution.

The geographic positions of G1-1 are close to the major diversity center (ranging from the Mediterranean Basin to Central Asia) for Allium genus², and the highest nucleotide diversity was observed in this subgroup (Fig. 6d). These findings are consistent with the historical records that northwestern China was the origin center of A. fistulosum^{35, 36}. In addition, the literature on the cultivation of A. fistulosum in China dates back to the third century BC. The Japanese A. fistulosum accessions were considered to be introduced from China at approximately 720 AC. Thereafter, A. fistulosum was spread to Western Europe during or at the end of the Middle Ages (1000-1400 AC) from Asia^35,36,37. These records are supportive of our phylogenetic results that the G2-2 (accessions from Japan and America and European countries) might have migrated from southern China.

Furthermore, among A. fistulosum, diverse accessions exhibited distinct geographic clustering rather than morphological classification in phylogenomic analysis (Fig. 6b). Middle Asian accessions (G1-2), which are distributed in Siberia and its adjacent areas, might migrate from western China (G1-1), whereas Japanese, American, and European accessions (G2-2) migrated or were introduced from southeast China G2-1 (Fig. 6e). Collectively, we validate that Northwestern China (G1-1) is the single origin center of A. fistulosum.

In summary, the high-quality chromosome-level genome of A. fistulosum greatly advances our understanding of genome diversity and expansion of Allium crops. The highest diversification of China-originated accessions and the existence of close relatives indicate that China is the primary origin and domestication center for A. fistulosum crops. The expansion and differentiation of the alliinase and lachrymatory factor synthase gene families might relate to Allium-specific flavor formation. Our current study will enable further functional genomic studies and genomic selection of important agronomic traits in Allium crops.

Methods

Plant materials and sequencing libraries preparations

The bunching onion accession “SXSJC” (grown in Hangzhou, Zhejiang Province, China) was sequenced for genome assembly. Detailed information on the 135 Allium accessions used for re-sequencing is provided in Supplementary Data 4. A total of 101 A. fistulosum accessions were distributed worldwide, including in China, Japan, and European, American, and Central Asian countries. Shallots and Chinese red onions were collected from China, and the feral A. altaicum, which was considered the closest relative and potential ancestor for cultivated A. fistulosum^{13, 38}, was collected from the Altai Mountains (89.44E, 47.05 N; Xinjiang, China).

For genome assembly, genomic DNA was extracted from the young leaves of bunching onions using a DNAsecure Plant Kit (Tiangen Biotech, Beijing, China), and DNA quality was detected by agarose gel electrophoresis. Single molecular real-time (SMRT) PacBio sequencing libraries were built following the standard protocols of Pacific Biosciences. Briefly, high-quality genomic DNA was sheared to ~20 kb, the damaged ends were repaired and ligated with a blunt-end adaptor, and the resulting libraries were sequenced using the PacBio Sequel platform.

For Bionano sequencing, DNA was extracted using the suggested kits (Bionano Prep Plant DNA Isolation Kit), and the specific sequences across the entire genome were labeled. The labeled DNA was transferred into a cartridge for scanning, and the images were converted into molecules.

Hi-C libraries were constructed according to the suggested procedure. Briefly, leaf samples were fixed with formaldehyde solution before chromatin extraction, and chromatin was digested with 400U of DPNII restriction enzyme at 37 °C. DNA ends were labeled with biotin, and DNA ligation was performed using T4 DNA ligase (NEB). After ligation, proteinase K was added for reverse crosslinking. The DNA fragments were then purified and dissolved. The purified DNA was fragmented to 300–500 bp, and the DNA ends were repaired. Biotin-labeled DNA fragments were separated using Dynabeads® M-280 Streptavidin (Life Technologies). Hi-C libraries were controlled for quality and sequenced on an Illumina HiSeq X Ten sequencer.

For Illumina sequencing, DNA libraries were constructed according to a previously described procedure (Illumina). Briefly, DNA was broken randomly into segments using a Covaris ultrasonic crusher, DNA ends were repaired, and poly (A) adaptors were added. PCR was performed using purified DNA, and the final libraries were sequenced on an Illumina platform after assessment.

Genome assembly and quality assessment

The genome size, heterozygosity ratio, and repeat sequence ratio were evaluated by k-mer³⁹ distribution analysis (k = 17) using Illumina short-reads. PacBio reads were used to assemble the contig-level genome using CANU⁴⁰ (v1.9, parameters: genomeSize = 12 G, corOutCoverge = 40 G). The contigs were polished with PacBio reads using NextPolish (v1.2.4; rerun = 3, https://github.com/Nextomics/NextPolish) and Illumina short-reads using Pilon (v1.2.2)⁴¹. Subsequently, redundant sequences were removed using Purge_dups (v1.2.3)⁴². Bionano data were used for auxiliary assembly using Bionano Solve (v3.5.1, DeNovo Assembly: -i 5 -F 1 -W 1 -c 1, Hybrid Scaffold: -B 2 -N 2) to improve the accuracy of the assembly⁴³. Hi-C data were used to assemble the chromosome-level genome using ALLHiC (v0.9.8, https://github.com/tangerzhang/ALLHiC) software.

Complementary methods were employed to evaluate the quality of the genome assembly. First, genome completeness was assessed based on conserved plant genes in the BUSCO (v3.0.2) and CEGMA (v2.5) databases. Second, Illumina short-reads were mapped to the assembled genome using BWA (v0.7.8)⁴⁴ to assess coverage rate and average depth. The LTR_retriever (v1.0.7)²² package was used to evaluate genome quality. Mequery (v1.3) was used to assess the consensus quality (QV) value and completeness of the genome assembly²⁰. The markers of genetic map²¹ were mapped to our genome using Chromonomer (v1.07, https://github.com/jleluyer/chromonomer_workflow), and collinear relationships were displayed using Python scripts.

Genome annotations

RepeatMasker and RepeatProteinMask (v4.07)⁴⁵ were used to identify TEs by alignment to the repeat library (Repbase v15.02), whereas de novo prediction of TEs was performed using RepeatModeler (v1.05, http://www.repeatmasker.org/RepeatModeler/, version 1.0.5), RepeatScout, and LTR_FINDER. Tandem Repeats Finder (TRF, v4.09) analysis was performed to identify the tandem repeats in the genome.

De novo prediction was performed using five ab initio gene prediction programs: Augustus (v3.2.3)⁴⁶, Geneid (v1.4, https://genome.crg.cat/software/geneid/index.html), Genescan (v1.0, http://genes.mit.edu/GENSCAN.html), GlimmerHMM (v3.0.2, https://ccb.jhu.edu/software/glimmerhmm/), and SNAP (v2013.11.29, https://snap.stanford.edu/). The protein sequences of 12 species (A. sativum¹¹, A. cepa²⁵, Asparagus setaceus⁴⁷, Asparagus officinalis⁴⁸, Dendrobium officinale⁴⁹, Apostasia shenzhenica⁵⁰, Phalaenopsis equestris⁵¹, Phoenix dactylifera⁵², Oryza sativa⁵³, Ananas comosus⁵⁴, Amborella trichopoda⁵⁵, and Musa schizocarpa⁵⁶) from NCBI or ENSEMBLE were aligned against the A. fistulosum genome using TBLASTN (v2.2.26, E-value <10e-5), which was used to predict gene models. Tophat (v2.0.13, http://ccb.jhu.edu/software/tophat/index.shtml) and Cufflinks (v2.1.1, http://cufflinks.cbcb.umd.edu/) pipelines were used to map the RNA-seq data to the A. fistulosum genome for gene prediction. Trinity (v2.1.1) and PASA (v2.2.0)⁵⁷ were used to analyze the gene structures. All gene models predicted using these approaches were merged using the weighted and non-redundant gene set 1.1.1 (EVM, v1.1.1, http://evidencemodeler.github.io/).

Gene annotation was performed by blasting the SwissProt, TrEMBL, and NCBI non-redundant protein databases, and the best hits were selected for annotations. Motifs and domains were annotated using InterProScan (v4.7) to search the InterPro (v29.0) databases. The GO and KEGG pathways for each gene were obtained using the best-match classification. Noncoding RNAs were predicted using the tRNAscan-SE (v1.4)⁵⁸ and INFERNAL (v1.1.2)⁵⁹ software.

Analysis of WGD events

FASTKs (https://github.com/mrmckain/FASTKs) was used to calculate the synonymous substitution (Ks) values of the paralogous blocks in A. fistulosum. The density estimation of the Ks value distributions was based on finite normal mixture modeling using MCLUST (https://sites.stat.washington.edu/mclust/), and the distribution of Ks values was used to determine WGD events.

Comparative genomics and genome evolution analysis

Briefly, orthologous gene families (single-copy and multi-copy families) of A. fistulosum and other genomes (A. sativum¹¹, A. cepa²⁵, Asparagus setaceus⁴⁷, Asparagus officinalis⁴⁸, Dendrobium officinale⁴⁹, Apostasia shenzhenica⁵⁰, Phalaenopsis equestris⁵¹, Phoenix dactylifera⁵², Oryza sativa⁵³, Ananas comosus⁵⁴, Amborella trichopoda⁵⁵ and Musa schizocarpa⁵⁶) were obtained using OrthoMCL (v1.4, http://orthomcl.org/orthomcl/). A super-alignment matrix was obtained by aligning single-copy gene families with MUSCLE (v3.8.31, http://www.drive5.com/muscle/), which was subsequently used for ML phylogenetic tree construction using RaxML (ML tree, Model GTRGAMMA, v8.2.12, http://sco.h-its.org/exelixis/web/software/raxml/index.html). Mcmctree (burn-in = 10,000, sample number = 100,000, sample frequency = 2; http://abacus.gene.ucl.ac.uk/software/paml.html) was used to calculate the divergence time based on the TimeTree database (http://www.timetree.org/) and a previous report⁵⁵. Gene family expansion and contraction were detected using CAFÉ (v4.2)⁶⁰ software. Genomic collinearity between garlic and bunching onions was analyzed using the MCScanx pipeline⁶¹.

Insertion time and phylogenetic analysis of LTRs

The ends of the LTR retrotransposons were aligned using MUSCLE (v3.8.31, (http://www.drive5.com/muscle/), and the insertion time (T) of the LTRs was calculated using the formula T = K/2r. K is the genetic distance calculated using the formula K = −0.75ln (1-4λ/3). The nucleotide divergence rate (λ) between the two LTRs less than 0.75 was retained for further analysis. The nucleotide substitution rate “r” was set to 1.3e-8 substitutions per site per year. Uncorrected pairwise distances were used to construct a neighbor-joining unrooted phylogenetic tree using TreeBest (v1.9.2, http://treesoft.sourceforge.net/treebest.shtml) with the suggested parameters. We classified LTRs into Copia, Gypsy, and other superfamilies using LTRdigest (v1.07)⁶², while the secondary LTRs were determined according to their phylogenetic properties.

Identification of ancestral and new DSD genes

Gene duplication events were classified as dispersed, proximal, WGD, tandem, and transposed duplications using the DupGen_finder (https://github.com/DXXDR/DupGen_finder) pipeline. We investigated the global landscape of gene duplications in 13 selected genomes and several large genomes, including three large genomes of Ginkgo biloba⁶³, rye⁶⁴, and Chinese pine⁶⁵. Chromosome collinearity was analyzed to determine whether a gene was located in the collinearity region based on JCVI⁶¹. Both A. sativum and A. cepa used A. fistulosum as the reference genome, whereas A. sativum was selected as the reference genome for A. fistulosum. Genes in collinear regions were considered ancestral DSD genes, whereas those outside the collinear regions were considered new DSD genes. We selected dispersed gene pairs in which one was an ancestral gene and the corresponding gene was a new gene to analyze the expression difference.

Identification and evolution analysis of flavor-related genes

To identify genes involved in the sulfur-metabolism biosynthesis pathways, the key parameters of BLASTP (v 2.2.26) were E-value <1e-5 and the identity >50%. The protein domains of the candidate genes were identified using HMMER (v3.1b1) software, and genes with different domains were filtered. The protein sequence of bulb onion LFS gene (GenBank accession no. AB089203) and garlic ALL genes (GenBank accession no. AAB32477.1, ACN78838.1, AAD26853.1, and AAA32639.1) from the NCBI database were used to identify the candidate orthologs. The query sequences of glucosinolate-related genes were obtained from a review³³. Full-length amino acid sequences of LFS and ALL genes were aligned using MUSCLE (v3.8.31), and a phylogenetic tree was constructed using Treebest (v1.9.2, http://treesoft.sourceforge.net/treebest.shtml) and ITOLs (https://itol.embl.de/).

Spatial expressions of the sulfoxide-related genes

Illumina RNA-seq was performed on the roots, pseudo-stems, leaves, floral stalks, and flowers of bunching onions, and the roots, bulbs, leaves, and pseudo-stems of bulb onions with four biological replicates. Gene expression datasets were obtained using the Tophat (http://ccb.jhu.edu/software/tophat/index.shtml) and Cufflinks (http://cufflinks.cbcb.umd.edu/) pipelines. The expression patterns of the target genes were visualized using a heatmap library (https://CRAN.R-project.org/package=pheatmap) in the R software with log₂ (FPKM + 1) values.

Re-sequencing and population genomics analysis

A total of 350 bp pair-end reads from 135 accessions were generated using the Illumina HIseq2500 platform. The mapping rates of A. ascalonicum and A. cepa var. proliferum were less than 90, and 4 × genomic coverage was less than 60%. We used 96 A. fistulosum accessions to filter SNPs. The SNPs were subjected to a quality control procedure following these steps: (1) removing the Illumina library construction adapters, (2) removing the reads containing more than 10% unknown bases (N bases), and (3) removing the low-quality reads containing more than 50% of low-quality bases (sequencing quality value ≤ 5). BWA (v0.7.8)⁴⁴ was used to align the clean reads to the reference genome, and Samtools (https://github.com/samtools/samtools) was used to call SNPs/InDels with the following parameters (-q 1 -C 50 -m 2 -F 0.002 -d 1000). The raw SNP/InDel sets with a mapping quality <20, or a depth of the variant position <3, or a missing rate >0.05, or an SNP frequency <5% were filtered. After filtering, we selected high-quality SNPs from the 135 Allium accessions for subsequent analyses. ANNOVAR (v20191024, https://annovar.openbioinformatics.org/en/latest/) was used for the functional annotation of variants. A phylogenetic tree was constructed using SNPhylo (v20180901)⁶⁶. The population structure of the accessions was investigated using Admixture (v1.23). In addition, PCA was performed using VCFtools (v0.1.12b, https://vcftools.github.io/index.html). Linkage disequilibrium analysis (LD decay) was calculated using PopLDdecay (v03.41)⁶⁷. The XP-CLR scores were calculated using XP-CLR (v1.0) with a sliding window size of 40 kb and a step size of 20 kb⁶⁸.The top 5% region was selected as the candidate selection region.

Determination of ACSOs in Allium crops

To prevent ACSO degradation, O-(carboxymethyl) hydroxylamine hemihydrochloride (OCMHA) was used to inhibit the activity of alliinase⁶⁹. The extraction procedure was as follows: 0.15–0.20 g flesh weight samples were transferred to 2 mL centrifuge tube containing a 5 mm steel ball, and 990 μL OCMHA solution (1.1 g/L methanol/deionized water 4:1, v/v) and 10 μL L-norleucin solution (1 mg/mL water) were added. After rubbing with a tissue grinding machine, the samples were homogenized for 1 min. The resulting slurry was centrifuged (10 min, room temperature, 12,000 rpm) and the supernatant was then diluted 10,000 × g. with a mixture of acetonitrile and deionized water (4:1, v/v). The final extracts were stored in vials at −20 °C until analysis.

ACSOs were identified by comparison with the authentic standards (alliin, Beijing SoIarbio Life Science Company, Beijing, China; isoalliin, from Chengdu Caoyuankang Biotechnology Company, Chengdu, China; and methiin, from Shanghai Yuanye Biotechnology Company, Shanghai, China). Mixed calibration curves were used at concentrations from 0.01 to 500 ng/mL for ACSOs quantification. The ACSO content was determined by Agilent 1290 ultra-high-performance liquid chromatographer coupled to a 6470 triple quadrupole spectrometer (Agilent Technologies, Waldbronn, Germany). The chromatographic separation was performed on a Waters BEH amide column (100 × 2.1 mm, 1.7 μm) maintained at 60 °C with a flow rate of 0.6 mL/min. The mobile phase consisted of (A) water with 0.5% formic acid and (B) acetonitrile with 0.5% formic acid. A gradient elution was used: 0–4 min, 10–15% A; 4–8 min, 15–60% A; 8–15 min, 10% A; injection volume 1 μL. Ionization was conducted in positive ion electrospray mode, and the operating parameters were as follows: capillary voltage, 3000 V; drying gas temperature, 325 °C; drying gas flow, 5 L/min; nebulizer pressure, 45 psi; sheath gas temperature, 350 °C; and sheath gas flow, 11 L/min. In addition, three biological replicates were used for all the accessions.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

All raw sequencing data generated in this study have been deposited to CNSA (https://db.cngb.org/cnsa/) under accession CNP0002276. The expression data of A. sativum were downloaded from NCBI GEO database under accession GSE145455. Source data are provided with this paper.

References

Friesen, N., Fritsch, R. M. & Blattner, F. R. Phylogeny and new intrageneric classification of Allium (Alliaceae) based on nuclear ribosomal DNA its sequences. Aliso 22, 372–395 (2006).
Article Google Scholar
Li, Q. Q. et al. Phylogeny and biogeography of Allium (Amaryllidaceae: Allieae) based on nuclear ribosomal internal transcribed spacer and chloroplast rps16 sequences, focusing on the inclusion of species endemic to China. Ann. Bot. 106, 709–733 (2010).
Article CAS PubMed PubMed Central Google Scholar
Eady, C. C. et al. Silencing onion lachrymatory factor synthase causes a significant change in the sulfur secondary metabolite profile. Plant Physiol. 147, 2096–2106 (2008).
Article CAS PubMed PubMed Central Google Scholar
Fritsch, R. & Keusgen, M. Occurrence and taxonomic significance of cysteine sulphoxides in the genus Allium L. (Alliaceae). Phytochemistry 67, 1127–1135 (2006).
Article CAS PubMed Google Scholar
Yoshimoto, N. & Saito, K. S-Alk(en)ylcysteine sulfoxides in the genus Allium: proposed biosynthesis, chemical conversion, and bioactivities. J. Exp. Bot. 70, 4123–4137 (2019).
Article PubMed Google Scholar
Fu, X., Li, C., Zhou, X., Liu, S. & Wu, F. Physiological response and sulfur metabolism of the V. dahliae-infected tomato plants in tomato/potato onion companion cropping. Sci. Rep. 6, 36445 (2016).
Article ADS CAS PubMed PubMed Central Google Scholar
Yan, Z. et al. Repellency of forty-one aromatic plant species to the Asian citrus psyllid, vector of the bacterium associated with huanglongbing. Ecol. Evol. 10, 12940–12948 (2020).
Article PubMed PubMed Central Google Scholar
Kang, M. et al. A chromosome-level Camptotheca acuminata genome assembly provides insights into the evolutionary origin of camptothecin biosynthesis. Nat. Commun. 12, 3531 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
Xia, E. et al. The reference genome of tea plant and resequencing of 81 diverse accessions provide insights into its genome evolution and adaptation. Mol. Plant 13, 1013–1026 (2020).
Article CAS PubMed Google Scholar
Guo, S. et al. Resequencing of 414 cultivated and wild watermelon accessions identifies selection for fruit quality traits. Nat. Genet. 51, 1616–1623 (2019).
Article CAS PubMed Google Scholar
Sun, X. et al. A Chromosome-Level Genome Assembly of Garlic (Allium sativum) Provides Insights into Genome Evolution and Allicin Biosynthesis. Mol. Plant 13, 1328–1339 (2020).
Article CAS PubMed Google Scholar
Yusupov, Z. et al. Phylogenomics of Allium section Cepa (Amaryllidaceae) provides new insights on domestication of onion. Plant Divers 43, 102–110 (2021).
Article PubMed Google Scholar
Peffley, E. B. in High-Tech and Micropropagation III (ed. Bajaj, Y.P.S.) 244–261 (Springer Berlin Heidelberg, 1992).
Budylin, M. V., Kan, L. Y., Romanov, V. S. & Khrustaleva, L. I. GISH study of advanced generation of the interspecific hybrids between Allium cepa L. and Allium fistulosum L. with relative resistance to downy mildew. Russ. J. Plant Physiol. 50, 387–394 (2014).
CAS Google Scholar
Benke, A. P., Mahajan, V., Manjunathagowda, D. C. & Mokat, D. N. Interspecific hybridization in Allium crops: status and prospectus. Genet. Resour. Crop Evol. 69, 1–9 (2022).
Umehara, M. et al. Interspecific hybrids between Allium fistulosum and Allium schoenoprasum reveal carotene-rich phenotype. Euphytica 148, 295 (2006).
Article CAS Google Scholar
Stice, S. P. et al. Thiosulfinate tolerance is a virulence strategy of an atypical bacterial pathogen of onion. Curr. Biol. 30, 3130–3140.e6 (2020).
Article CAS PubMed Google Scholar
Han, T. et al. Polyploidy promotes species diversification of Allium through ecological shifts. N. Phytol. 225, 571–583 (2020).
Article CAS Google Scholar
Ricroch, A., Yockteng, R., Brown, S. C. & Nadot, S. Evolution of genome size across some cultivated Allium species. Genome 48, 511–520 (2005).
Article CAS PubMed Google Scholar
Rhie, A., Walenz, B. P., Koren, S. & Phillippy, A. M. Merqury: reference-free quality, completeness, and phasing assessment for genome assemblies. Genome Biol. 21, 245 (2020).
Article CAS PubMed PubMed Central Google Scholar
Tsukazaki, H. et al. Construction of SSR-based chromosome map in bunching onion (Allium fistulosum). Theor. Appl. Genet. 117, 1213–1223 (2008).
Article CAS PubMed Google Scholar
Ou, S., Chen, J. & Jiang, N. Assessing genome assembly quality using the LTR Assembly Index (LAI). Nucleic Acids Res. 46, e126 (2018).
PubMed PubMed Central Google Scholar
Galindo-González, L., Mhiri, C., Deyholos, M. K. & Grandbastien, M. A. LTR-retrotransposons in plants: Engines of evolution. Gene 626, 14–25 (2017).
Article PubMed Google Scholar
Xiao, H., Jiang, N., Schaffner, E., Stockinger, E. J. & van der Knaap, E. A Retrotransposon-Mediated Gene Duplication Underlies Morphological Variation of Tomato Fruit. Sience 319, 1527–1530 (2008).
Article ADS CAS Google Scholar
Finkers, R. et al. Insights from the first genome assembly of Onion (Allium cepa). G3 (Bethesda) 11, jkab243 (2021).
Shukla, S. et al. The Onion Genomic Resource: a genomics and bioinformatics driven resource for onion breeding. Plant Gene 8, 9–15 (2016).
Article CAS Google Scholar
Lichman, B. R., Godden, G. T. & Buell, C. R. Gene and genome duplications in the evolution of chemodiversity: perspectives from studies of Lamiaceae. Curr. Opin. Plant Biol. 55, 74–83 (2020).
Article CAS PubMed Google Scholar
Imai, S. et al. An onion enzyme that makes the eyes water. Nature 419, 685–685 (2002).
Article ADS CAS PubMed Google Scholar
Arakawa, T. et al. Dissecting the stereocontrolled conversion of short-lived sulfenic acid by lachrymatory factor synthase. ACS Catal. 10, 9–19 (2019).
Article Google Scholar
Shen, Y. et al. Chromosome-level and haplotype-resolved genome provides insight into the tetraploid hybrid origin of patchouli. Nat. Commun. 13, 3511 (2022).
Article ADS CAS PubMed PubMed Central Google Scholar
Rose, P., Whiteman, M., Moore, P. K. & Zhu, Y. Z. Bioactive S-alk(en)yl cysteine sulfoxide metabolites in the genus Allium: the chemistry of potential therapeutic agents. Nat. Prod. Rep. 22, 351–368 (2005).
Article CAS PubMed Google Scholar
Ishida, M., Hara, M., Fukino, N., Kakizaki, T. & Morimitsu, Y. Glucosinolate metabolism, functionality and breeding for the improvement of Brassicaceae vegetables. Breed. Sci. 64, 48–59 (2014).
Article CAS PubMed PubMed Central Google Scholar
Sønderby, I. E., Geu-Flores, F. & Halkier, B. A. Biosynthesis of glucosinolates gene discovery and beyond. Trends Plant Sci. 15, 283–290 (2010).
Article PubMed Google Scholar
Friesen, N., Pollner, S., Bachmann, K. & Blattner, F. R. RAPDs and noncoding chloroplast DNA reveal a single origin of the cultivated Allium fistulosum from A. altaicum (Alliaceae). Am. J. Bot. 86, 554–562 (1999).
Article CAS PubMed Google Scholar
Singh, B. & Ramakrishna, Y. Welsh onion (Allium fistulosum L.): a promising spicing-culinary herb of Mizoram. Indian J. Hill. Farming 30, 201–208 (2017).
Google Scholar
Ford-Lloyd, B.V. & Armstrong, S.J. in Genetic Improvement of Vegetable Crops (eds. Kalloo, G. & Bergh, B.O.) 51–58 (Pergamon, Amsterdam, 1993).
Kayat, F., Mohammed, A. & Ibrahim, A. M. in Advances in Plant Breeding Strategies: Vegetable Crops: Volume 10: Leaves, Flowerheads, Green Pods, Mushrooms and Truffles (eds. Al-Khayri, J.M., Jain, S.M. & Johnson, D.V.) 135–182 (Springer International Publishing, Cham, 2021).
Ford-Lloyd, B. V. & Armstrong, S. J. in Genetic Improvement of Vegetable Crops (eds. Kalloo, G. & Bergh, B.O.) 51–58 (Pergamon, Amsterdam, 1993).
Marcais, G. & Kingsford, C. A fast, lock-free approach for efficient parallel counting of occurrences of k-mers. Bioinformatics 27, 764–770 (2011).
Article CAS PubMed PubMed Central Google Scholar
Koren, S. et al. Canu: scalable and accurate long-read assembly via adaptive k-mer weighting and repeat separation. Genome Res. 27, 722–736 (2017).
Article CAS PubMed PubMed Central Google Scholar
Walker, B. J. et al. Pilon: an integrated tool for comprehensive microbial variant detection and genome assembly improvement. PLoS ONE 9, e112963 (2014).
Article ADS PubMed PubMed Central Google Scholar
Guan, D. et al. Identifying and removing haplotypic duplication in primary genome assemblies. Bioinformatics 36, 2896–2898 (2020).
Article CAS PubMed PubMed Central Google Scholar
Lam, E. T. et al. Genome mapping on nanochannel arrays for structural variation analysis and sequence assembly. Nat. Biotechnol. 30, 771–776 (2012).
Article CAS PubMed Google Scholar
Li, H. & Durbin, R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics 25, 1754–1760 (2009).
Article CAS PubMed PubMed Central Google Scholar
Tarailo‐Graovac, M. & Chen, N. Using RepeatMasker to identify repetitive elements in genomic sequences. Curr. Protoc. Bioinforma. 25, 4.10.1-4.10.14 (2009).
Article Google Scholar
Stanke, M. et al. AUGUSTUS: ab initio prediction of alternative transcripts. Nucleic Acids Res. 34, W435–W439 (2006).
Article CAS PubMed PubMed Central Google Scholar
Li, S. et al. Chromosome-level genome assembly, annotation and evolutionary analysis of the ornamental plant Asparagus setaceus. Hort. Res. 7, 48 (2020).
Article CAS Google Scholar
Harkess, A. et al. The asparagus genome sheds light on the origin and evolution of a young Y chromosome. Nat. Commun. 8, 1279 (2017).
Article ADS PubMed PubMed Central Google Scholar
Yan, L. et al. The genome of Dendrobium officinale illuminates the biology of the important traditional Chinese orchid herb. Mol. Plant 8, 922–934 (2015).
Article CAS PubMed Google Scholar
Zhang, G. Q. et al. The Apostasia genome and the evolution of orchids. Nature 549, 379–383 (2017).
Article ADS CAS PubMed PubMed Central Google Scholar
Cai, J. et al. The genome sequence of the orchid Phalaenopsis equestris. Nat. Genet. 47, 65–72 (2015).
Article CAS PubMed Google Scholar
Al-Mssallem, I. S. et al. Genome sequence of the date palm Phoenix dactylifera L. Nat. Commun. 4, 2274 (2013).
Article ADS PubMed Google Scholar
Gao, Z.-Y. et al. Dissecting yield-associated loci in super hybrid rice by resequencing recombinant inbred lines and improving parental genome sequences. Proc. Natl Acad. Sci. USA 110, 14492–14497 (2013).
Article ADS CAS PubMed PubMed Central Google Scholar
Ming, R. et al. The pineapple genome and the evolution of CAM photosynthesis. Nat. Genet. 47, 1435–1442 (2015).
Article CAS PubMed PubMed Central Google Scholar
Albert, V. A. et al. The Amborella genome and the evolution of flowering plants. Science 342, 1467 (2013).
Google Scholar
Belser, C. et al. Chromosome-scale assemblies of plant genomes using nanopore long reads and optical maps. Nat. Plants 4, 879–887 (2018).
Article CAS PubMed Google Scholar
Haas, B. J. Improving the Arabidopsis genome annotation using maximal transcript alignment assemblies. Nucleic Acids Res. 31, 5654–5666 (2003).
Article CAS PubMed PubMed Central Google Scholar
Lowe, T. M. & Eddy, S. R. tRNAscan-SE: a program for improved detection of transfer RNA genes in genomic sequence. Nucleic Acids Res. 25, 955–964 (1997).
Article CAS PubMed PubMed Central Google Scholar
Nawrocki, E. P., Kolbe, D. L. & Eddy, S. R. Infernal 1.0: inference of RNA alignments. Bioinformatics 25, 1335–1337 (2009).
Article CAS PubMed PubMed Central Google Scholar
De Bie, T., Cristianini, N., Demuth, J. P. & Hahn, M. W. CAFE: a computational tool for the study of gene family evolution. Bioinformatics 22, 1269–1271 (2006).
Article PubMed Google Scholar
Tang, H. et al. Synteny and collinearity in plant genomes. Science 320, 486–488 (2008).
Article ADS CAS PubMed Google Scholar
Steinbiss, S., Willhoeft, U., Gremme, G. & Kurtz, S. Fine-grained annotation and classification of de novo predicted LTR retrotransposons. Nucleic Acids Res. 37, 7002–7013 (2009).
Article CAS PubMed PubMed Central Google Scholar
Liu, H. et al. The nearly complete genome of Ginkgo biloba illuminates gymnosperm evolution. Nat. Plants 7, 748–756 (2021).
Article CAS PubMed Google Scholar
Li, G. et al. A high-quality genome assembly highlights rye genomic characteristics and agronomically important genes. Nat. Genet. 53, 574–584 (2021).
Article CAS PubMed PubMed Central Google Scholar
Niu, S. et al. The Chinese pine genome and methylome unveil key features of conifer evolution. Cell 185, 204–217.e14 (2022).
Article CAS PubMed Google Scholar
Lee, T., Guo, H., Wang, X., Kim, C. & Paterson, A. H. SNPhylo: a pipeline to construct a phylogenetic tree from huge SNP data. BMC Genomics 15, 162 (2014).
Article PubMed PubMed Central Google Scholar
Zhang, C., Dong, S., Xu, J., He, W. & Yang, T. PopLDdecay: a fast and effective tool for linkage disequilibrium decay analysis based on variant call format files. Bioinformatics 35, 1786–1788 (2018).
Article Google Scholar
Chen, H., Patterson, N. & Reich, D. Population differentiation as a test for selective sweeps. Genome Res. 20, 393–402 (2010).
Article CAS PubMed PubMed Central Google Scholar
Rektorisova, M., Hrbek, V., Jiru, M., Ovesna, J. & Hajslova, J. Variability in S-Alk(en)yl-L-cysteine sulfoxides in garlic within a seven-month period determined by a liquid chromatography - tandem mass spectrometry method. Plant Foods Hum. Nutr. 75, 376–382 (2020).
Article CAS PubMed Google Scholar

Download references

Acknowledgements

We thank the U.S. National Plant Germplasm System for providing the A. fistulosum accessions. We thank Mrs. Xiaodan Wu and Mei Fang from Analysis Center of Agrobiology and Environmental Science, Zhejiang University for their technical assistance in ACSOs quantifications, as well as Mrs. Quanfang Zhang from Agricultural Technology Extension Center of Fuyang District for providing the A. fistulosum var. viviparum accession “HZC”. This work was supported by the China Agriculture Research System of MOF and MARA (CARS-24-B-01, Y-H.Z.) and the Fundamental Research Funds for the Central Universities (+226−2022-00100, M-F.Z.).

Author information

These authors contributed equally: Nanqiao Liao, Zhongyuan Hu, Jinshan Miao, Xiaodi Hu.

Authors and Affiliations

Institute of Vegetable Science, Zhejiang University, 310058, Hangzhou, Zhejiang, P. R. China
Nanqiao Liao, Zhongyuan Hu, Xiaolong Lyu, Yi-Mei Zhou, Ahmed Mahmoud, Guancong Deng, Yi-Qing Meng, Kejia Zhang, Yu-Yuan Ma, Yuelin Xia, Meng Zhao, Haiyang Yang, Jing-Hua Yang, Yan-Hong Zhou, Ming-Fang Zhang & Jing-Quan Yu
Horticultural Institute of Science and Technology, Weifang University of Science and Technology, 262700, Weifang, Shandong, P. R. China
Jinshan Miao
Novogene Bioinformatics Institute, 100083, Beijing, P. R. China
Xiaodi Hu, Yong Zhao, Ling Kang & Yiming Wang
Key Laboratory of Horticultural Plant Growth and Development, Ministry of Agriculture and Rural Affairs, 310058, Hangzhou, Zhejiang, P. R. China
Xiaolong Lyu, Jing-Hua Yang, Yan-Hong Zhou, Ming-Fang Zhang & Jing-Quan Yu
Hainan Institute of Zhejiang University, Yazhou Bay Science and Technology City, 572025, Sanya, Hainan, P. R. China
Xiaolong Lyu, Jing-Hua Yang & Ming-Fang Zhang
Ningxia Key Laboratory for Food Microbial-applications Technology and Safety Control, School of Food & Wine, Ningxia University, 750021, Yinchuan, Ningxia, P. R. China
Haitian Fang

Authors

Nanqiao Liao
View author publications
You can also search for this author in PubMed Google Scholar
Zhongyuan Hu
View author publications
You can also search for this author in PubMed Google Scholar
Jinshan Miao
View author publications
You can also search for this author in PubMed Google Scholar
Xiaodi Hu
View author publications
You can also search for this author in PubMed Google Scholar
Xiaolong Lyu
View author publications
You can also search for this author in PubMed Google Scholar
Haitian Fang
View author publications
You can also search for this author in PubMed Google Scholar
Yi-Mei Zhou
View author publications
You can also search for this author in PubMed Google Scholar
Ahmed Mahmoud
View author publications
You can also search for this author in PubMed Google Scholar
Guancong Deng
View author publications
You can also search for this author in PubMed Google Scholar
Yi-Qing Meng
View author publications
You can also search for this author in PubMed Google Scholar
Kejia Zhang
View author publications
You can also search for this author in PubMed Google Scholar
Yu-Yuan Ma
View author publications
You can also search for this author in PubMed Google Scholar
Yuelin Xia
View author publications
You can also search for this author in PubMed Google Scholar
Meng Zhao
View author publications
You can also search for this author in PubMed Google Scholar
Haiyang Yang
View author publications
You can also search for this author in PubMed Google Scholar
Yong Zhao
View author publications
You can also search for this author in PubMed Google Scholar
Ling Kang
View author publications
You can also search for this author in PubMed Google Scholar
Yiming Wang
View author publications
You can also search for this author in PubMed Google Scholar
Jing-Hua Yang
View author publications
You can also search for this author in PubMed Google Scholar
Yan-Hong Zhou
View author publications
You can also search for this author in PubMed Google Scholar
Ming-Fang Zhang
View author publications
You can also search for this author in PubMed Google Scholar
Jing-Quan Yu
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

J-Q.Y., M-F.Z., and Y-H.Z. conceived of and designed the experiments. N.L., Z.H., X.H., Y.Z., L.K., and Y.W. performed genome assembly and assessment, comparative genome analysis, and other bioinformatic analyses; J-H.Y., X.L., Y-M.Z., G.D., Y-M.M., K.Z., Y-Y.M., M.A., and Y.X. conducted transcriptome assays and re-sequencing; J.M. and H.F. provided most of the A. fistulosum accessions; J.M., H.Y., and M.Z. performed field cultivation and phenotype investigation; N.L. and Z.H. wrote the draft manuscript; J-Q.Y., M-F.Z., and Y-H.Z. revised the manuscript.

Corresponding authors

Correspondence to Yan-Hong Zhou, Ming-Fang Zhang or Jing-Quan Yu.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Description of Additional Supplementary Files

Supplementary Data 1

Supplementary Data 2

Supplementary Data 3

Supplementary Data 4

Reporting Summary

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Liao, N., Hu, Z., Miao, J. et al. Chromosome-level genome assembly of bunching onion illuminates genome evolution and flavor formation in Allium crops. Nat Commun 13, 6690 (2022). https://doi.org/10.1038/s41467-022-34491-3

Download citation

Received: 10 July 2021
Accepted: 19 October 2022
Published: 05 November 2022
DOI: https://doi.org/10.1038/s41467-022-34491-3

This article is cited by

Identification of a candidate gene for the I locus determining the dominant white bulb color in onion (Allium cepa L.)
- Geonjoong Kim
- Heejung Cho
- Sunggil Kim
Theoretical and Applied Genetics (2024)
Chromosome-level genome assembly of Microplitis manilae Ashmead, 1904 (Hymenoptera: Braconidae)
- Xiaohan Shu
- Ruizhong Yuan
- Xuexin Chen
Scientific Data (2023)
Evaluation of homozygosity levels of onion (Allium cepa L.) inbred lines using molecular markers evenly distributed in a high-resolution linkage map
- Jin Seong Moon
- Bongju Kim
- Sunggil Kim
Euphytica (2023)
Chromosome-level genomes of three key Allium crops and their trait evolution
- Fei Hao
- Xue Liu
- Jing Cai
Nature Genetics (2023)

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.