Evaluation of the Major Seed Storage Proteins, the Conglutins, Across Genetically Diverse Narrow-Leafed Lupin Varieties

Abstract

Lupin seeds have an excellent nutritional profile, including a high proportion of protein and dietary fiber. These qualities make lupin seeds an ideal candidate to help meet the growing global demand for complementary sources of protein. Of consequence to this application, there are nutritional and antinutritional properties assigned to the major lupin seed storage proteins-referred to as α-, β-, δ- and γ-conglutins The variation in the abundance of these protein families can impact the nutritional and bioactive properties of different lupin varieties. Hence, exploring the conglutin protein profiles across a diverse range of lupin varieties will yield knowledge that can facilitate the selection of superior genotypes for food applications or lupin crop improvement. To support this knowledge generation, discovery proteomics was applied for the identification of the 16 known conglutin subfamilies from 46 domestic and wild narrow-leafed lupin (NLL) genotypes. Consequently, the diversity of abundance of these proteins was evaluated using liquid chromatography-multiple reaction monitoring-mass spectrometry (LC-MRM-MS). This comparative study revealed a larger variability for the β- and δ-conglutin content across the lines under study. The absence/lower abundance of the β2- to β6-conglutin subfamilies in a subset of the domesticated cultivars led to substantially lower overall levels of the allergenic β-conglutin content in these NLLs, for which the elevation of the other conglutin families were observed. The diversity of the conglutin profiles revealed through this study-and the identification of potential hypoallergenic genotypes-will have great significance for lupin allergic consumers, food manufactures as well as grain breeders through the future development of lupin varieties with higher levels of desirable bioactive proteins and lower allergen content.

Keywords: LC-MRM-MS; Lupinus angustifolius; conglutin; legume; narrow-leafed lupin; plant-based protein; proteomics.

FIGURE 1

Schematic representation of the main structural findings and conserved regions in (A) α-, (B) β-, (C) δ-, and (D) γ-conglutin sequences, as well as the marker peptides used in LC-MRM-MS measurement of these proteins. The Pfam domains are highlighted in blue (Cupin_1), yellow (Tryp_alpha_amyl), purple (TAXi_N), and pink (TAXi_C). The red lines show the position of cysteine residues, and the green blocks are representative of MRM marker peptides (quantitative peptides). The conglutin peptides identified by IDA (qualitative peptides) are depicted by gray blocks, wherein the darker color indicates sequence regions commonly identified across the analyzed lines (white blocks, 1–5 cultivars; black blocks, 40–46 cultivars).

FIGURE 2

Multivariate analysis of the peptide level abundance data. (A) The PCA scores plot, PC1 vs. PC2, showing the separation of 46 NLL genotypes based on the conglutin-derived peptide data (log10). Each color represents an NLL genotype where the wild (triangles) and domesticated (circles) varieties are indicated. (B) The heatmap displaying the monitored marker peptide peak areas (log10) across the analyzed lines. The dendrogram from the unsupervised HCA illustrating the similarity of the NLL varieties in terms of conglutin abundance. Each column corresponds to a genotype and every row represents a peptide. The highlighted row at the top illustrates the counties of origin for the lines under study. The country names are abbreviated as follows: DZA (Algeria), AU (Australia), BLR (Belarus), CYP (Cyprus), FRA (France), GRC (Greece), ISR (Israel), ITA (Italy), MAR (Morocco), POL (Poland), PRT (Portugal), ESP (Spain), Syria (SYR), and TUR (Turkey).

FIGURE 3

Relative quantitation of the marker peptides monitored for (A) α-, (B) β-, (C) δ-, and (D) γ-conglutin families. The line graphs in each panel compare the relative abundance of the peptides measured for each conglutin subfamily. For ease of comparison the MRM peak area of each peptide was converted to percentage relative to the average peak area for all accessions. One-way ANOVA with Dunnett test was conducted for the multiple comparisons of the experimental groups and determining the significant differences (*p < 0.05).