The WRKY family is one of the largest transcription factor (TF) families in plants and plays central roles in modulating plant stress responses and developmental processes, as well as secondary metabolic regulations. Lotus (Nelumbo nucifera) is an aquatic crop that has significant food, ornamental and pharmacological values. Here, we performed an overview analysis of WRKY TF family members in lotus, and studied their functions in environmental adaptation and regulation of lotus benzylisoquinoline alkaloid (BIA) biosynthesis. A total of 65 WRKY genes were identified in the lotus genome and they were well clustered in a similar pattern with their Arabidopsis homologs in seven groups (designated I, IIa-IIe, and III), although no lotus WRKY was clustered in the group IIIa. Most lotus WRKYs were functionally paired, which was attributed to the recently occurred whole genome duplication in lotus. In addition, lotus WRKYs were regulated dramatically by salicilic acid (SA), jasmonic acid (JA), and submergence treatments, and two lotus WRKYs, NnWRKY40a and NnWRKY40b, were significantly induced by JA and promoted lotus BIA biosynthesis through activating BIA biosynthetic genes. The investigation of WRKY TFs for this basal eudicot reveals new insights into the evolution of the WRKY family, and provides fundamental information for their functional studies and lotus breeding.

SUMMARY: Nelumbo lutea (American lotus), which differs from Nelumbo nucifera (Asian lotus) morphologically, is one of the two remaining species in the basal eudicot family Nelumbonaceae. Here, we assembled the 843‐Mb genome of American lotus into eight pseudochromosomes containing 31 382 protein‐coding genes. Comparative analyses revealed conserved synteny without large chromosomal rearrangements between the genomes of American and Asian lotus and identified 29 533 structural variants (SVs). Carotenoid and anthocyanin pigments determine the yellow and red petal colors of American and Asian lotus, respectively. The structural genes encoding enzymes of the carotenoid and anthocyanin biosynthesis pathways were conserved between two species but differed in expression. We detected SVs caused by repetitive sequence expansion or contraction among the anthocyanin biosynthesis regulatory MYB genes. Further transient overexpression of candidate NnMYB5 induced anthocyanin accumulation in lotus petals. Alternative oxidase (AOX), uncoupling proteins (UCPs), and sugar metabolism and transportation contributed to carpel thermogenesis. Carpels produce heat with sugars transported from leaves as the main substrates, because there was weak tonoplast sugar transporter (TST) activity, and with SWEETs were highly expressed during thermogenesis. Cell proliferation‐related activities were particularly enhanced in the warmer carpels compared with stamens during the cold night before blooming, which suggested that thermogenesis plays an important role in flower protogyny. Population genomic analyses revealed deep divergence between American and Asian lotus, and independent domestication affecting seed, rhizome, and flower traits. Our findings provide a high‐quality reference genome of American lotus for exploring the genetic divergence and variation between two species and revealed possible genomic bases for petal color, carpel thermogenesis and domestication in lotus.

Background: The genus Nelumbo Adans. comprises two living species, N. nucifera Gaertan. (Asian lotus) and N. lutea Pers. (American lotus). A genetic linkage map is an essential resource for plant genetic studies and crop improvement but has not been generated for Nelumbo. We aimed to develop genomic simple sequence repeat (SSR) markers from the genome sequence and construct two genetic maps for Nelumbo to assist genome assembly and integration of a genetic map with the genome sequence. Results: A total of 86,089 SSR motifs were identified from the genome sequences. Di- and tri-nucleotide repeat motifs were the most abundant, and accounted for 60.73% and 31.66% of all SSRs, respectively. AG/GA repeats constituted 51.17% of dinucleotide repeat motifs, followed by AT/TA (44.29%). Of 500 SSR primers tested, 386 (77.20%) produced scorable alleles with an average of 2.59 per primer, and 185 (37.00%) showed polymorphism among two parental genotypes, N. nucifera ‘Chinese Antique’ and N. lutea ‘AL1’, and six progenies of their F₁ population. The normally segregating markers, which comprised 268 newly developed SSRs, 37 previously published SSRs and 53 sequence-related amplified polymorphism markers, were used for genetic map construction. The map for Asian lotus was 365.67 cM with 47 markers distributed in seven linkage groups. The map for American lotus was 524.51 cM, and contained 177 markers distributed in 11 genetic linkage groups. The number of markers per linkage group ranged from three to 34 with an average genetic distance of 3.97 cM between adjacent markers. Moreover, 171 SSR markers contained in linkage groups were anchored to 97 genomic DNA sequence contigs of ‘Chinese Antique’. The 97 contigs were merged into 60 scaffolds. Conclusion: Genetic mapping of SSR markers derived from sequenced contigs in Nelumbo enabled the associated contigs to be anchored in the linkage map and facilitated assembly of the genome sequences of ‘Chinese Antique’. The present study reports the first construction of genetic linkage maps for Nelumbo, which can serve as reference linkage maps to accelerate characterization germplasm, genetic mapping for traits of economic interest, and molecular breeding with marker-assisted selection.

Background: Starch in the lotus seed contains a high proportion of amylose, which endows lotus seed a promising property in the development of hypoglycemic and low-glycemic index functional food. Currently, improving starch content is one of the major goals for seed-lotus breeding. ADP-glucose pyrophosphorylase (AGPase) plays an essential role in regulating starch biosynthesis in plants, but little is known about its characterization in lotus. Results: We describe the nutritional compositions of lotus seed among 30 varieties with starch as a major component. Comparative transcriptome analysis showed that AGPase genes were differentially expressed in two varieties (CA and JX) with significant different starch content. Seven putative AGPase genes were identified in the lotus genome (Nelumbo nucifera Gaertn.), which could be grouped into two subfamilies. Selective pressure analysis indicated that purifying selection acted as a vital force in the evolution of AGPase genes. Expression analysis revealed that lotus AGPase genes have varying expression patterns, with NnAGPL2a and NnAGPS1a as the most predominantly expressed, especially in seed and rhizome. NnAGPL2a and NnAGPS1a were co-expressed with a number of starch and sucrose metabolism pathway related genes, and their expressions were accompanied by increased AGPase activity and starch content in lotus seed. Conclusions: Seven AGPase genes were characterized in lotus, with NnAGPL2a and NnAGPS1a, as the key genes involved in starch biosynthesis in lotus seed. These results considerably extend our understanding on lotus AGPase genes and provide theoretical basis for breeding new lotus varieties with high-starch content.