Specificity and Plasticity of the Functional Ionome of Brassica napus and Triticum aestivum Exposed to Micronutrient or Beneficial Nutrient Deprivation and Predictive Sensitivity of the Ionomic Signatures

Abstract

The specific variation in the functional ionome was studied in Brassica napus and Triticum aestivum plants subjected to micronutrient or beneficial mineral nutrient deprivation. Effects of these deprivations were compared to those of macronutrient deprivation. In order to identify early events, plants were harvested after 22 days, i.e., before any significant reduction in growth relative to control plants. Root uptake, tissue concentrations and relative root nutrient contents were analyzed revealing numerous interactions with respect to the 20 elements quantified. The assessment of the functional ionome under individual mineral nutrient deficiency allows the identification of a large number of interactions between elements, although it is not totally exhaustive, and gives access to specific ionomic signatures that discriminate among deficiencies in N, P, S, K, Ca, Mn, Fe, Zn, Na, Si, and Se in both species, plus Mg, Cl, Cu, and Mo in wheat. Ionome modifications and components of ionomic signatures are discussed in relation to well-known mechanisms that may explain crosstalks between mineral nutrients, such as between Na and K, V, Se, Mo and S or Fe, Zn and Cu. More surprisingly, when deprived of beneficial nutrients such as Na, Si, Co, or Se, the plant ionome was strongly modified while these beneficial nutrients contributed greatly to the leaf ionomic signature of most mineral deficiencies.

Keywords: ionome; ionomic signatures; nutrient deficiency; nutrient interactions; rapeseed; wheat.

FIGURE 1

Flow diagram of the ionomic data analysis procedure using the PLS-DA model. The dataset containing the elemental concentrations (i.e., the ionome) (X) and deficiency classes (Y) was randomly split ten times into a training set and a test set (80/20 percent, respectively). Each time a five-fold and 50 repeat cross validation on the PLS-DA was used to measure model performance and find the best parameters (i.e., latent components to be retained) for class prediction. Thereafter a confusion matrix was used to assess prediction quality and the results were averaged after running the analysis ten times. Four methods and associated datasets were used (A) all ionomic data, (B) all ionomic data except the deprived nutrient, (C) macronutrient ionomic data, and (D) the deprived nutrient data.

FIGURE 2

(A) Aboveground, (B) root, and (C) whole plant dry weight of B. napus and T. aestivum plants after 22 days of micronutrient or beneficial nutrient deprivation under hydroponic conditions. Data are given as the mean ± SE (n = 5) and significant differences between control and nutrient-deprived plants are indicated as follows: *, p < 0.05.

FIGURE 3

Relative root nutrient contents (RRNC) of B. napus and T. aestivum plants after 22 days of micronutrient or beneficial nutrient deprivation (A). The RRNC was calculated as the ratio of nutrient content in roots/entire plant. A color gradient for large increases (red) or decreases (blue) in the RRNC are given only if they were significantly different from control plants for p < 0.01. The ratios of root biomass/whole plant biomass (B) are also given as the mean ± SE. for comparisons with RRNC, and only the values indicated in bold differ significantly from control plant biomass.

FIGURE 4

Heatmap of relative net uptake by (A) B. napus and (B) T. aestivum plants after 22 days of micronutrient or beneficial nutrient deprivation under hydroponic conditions. Relative net nutrient uptake was calculated (see section “Materials and Methods”) as the ratio of nutrients taken up by deprived plants/nutrients taken up by control plants. Only values significant for p < 0.05 are given, with the color gradient indicating values relative to control plants between 0.2 (blue = low) and 5 (orange = high). Blank cells in heatmaps corresponded to non-significant variations in net uptake compared to control plants.

FIGURE 5

BnaIRT1 gene expression relative to control plants in roots of B. napus subjected to micronutrient or beneficial nutrient deprivation for 22 days. Values are given as the mean ± SE (n = 5) and significant differences between control and nutrient-deprived plants are indicated as follows: ^∗p < 0.05; ^∗∗p < 0.01.

FIGURE 6

Heatmap of relative mineral nutrient concentrations in (A) B. napus and (B) T. aestivum plants after 22 days of micronutrient or beneficial nutrient deprivation under hydroponic conditions. Relative mineral nutrient concentration was calculated (see section “Materials and Methods”) as the ratio of the nutrient concentration in deprived plants/nutrient concentration in control plants. Tissues developed before or after the beginning of deprivation (D₀) are indicated as “young leaf blades” (YLBs) and “old leaf blades” (OLBs), respectively. Only values significant for p < 0.05 are given, with the color gradient indicating values relative to control plants between 0.2 (blue = low) and 5 (orange = high).

FIGURE 7

Principal component analysis (PCA) score plots of the complete elemental content data. Projection of the data onto the subspace spanned by components 1 (PC 1) and 3 (PC3), which are colored according to species (A) or tissue (B), with each individual labeled with its treatment class (control or deprived plant). The contribution plots (C,D) depict the importance of each element in component 1 and 3, respectively, the bar length representing regression coefficients with either positive or negative signs. Variables are ranked by decreasing importance starting from the bottom.

FIGURE 8

Heatmap of the elements with the highest VIP scores for each deficiency applied to B. napus (A) and T. aestivum (B) which were retrieved from the PLS-DA model on YLBs samples. Elements with VIP scores ≥1 are significant (≥1, light gray; ≥1.5, mid gray; ≥2, dark gray) and indicate the importance of each element (X variable) to predict deficiencies (Y). The specific combination of elements and their concentration variation (high or low in tissue) discriminate samples and generate signature.