Compatible and Incompatible Pollen-Styles Interaction in Pyrus communis L. Show Different Transglutaminase Features, Polyamine Pattern and Metabolomics Profiles

Abstract

Pollen-stigma interaction is a highly selective process, which leads to compatible or incompatible pollination, in the latter case, affecting quantitative and qualitative aspects of productivity in species of agronomic interest. While the genes and the corresponding protein partners involved in this highly specific pollen-stigma recognition have been studied, providing important insights into pollen-stigma recognition in self-incompatible (SI), many other factors involved in the SI response are not understood yet. This work concerns the study of transglutaminase (TGase), polyamines (PAs) pattern and metabolomic profiles following the pollination of Pyrus communis L. pistils with compatible and SI pollen in order to deepen their possible involvement in the reproduction of plants. Immunolocalization, abundance and activity of TGase as well as the content of free, soluble-conjugated and insoluble-bound PAs have been investigated. ¹H NMR-profiling coupled with multivariate data treatment (PCA and PLS-DA) allowed to compare, for the first time, the metabolic patterns of not-pollinated and pollinated styles. Results clearly indicate that during the SI response TGase activity increases, resulting in the accumulation of PAs conjugated to hydroxycinnamic acids and other small molecules. Metabolomic analysis showed a remarkable differences between pollinated and not-pollinated styles, where, except for glucose, all the other metabolites where less concentrated. Moreover, styles pollinated with compatible pollen showed the highest amount of sucrose than SI pollinated ones, which, in turn, contained highest amount of all the other metabolites, including aromatic compounds, such as flavonoids and a cynnamoil derivative.

Keywords: 1H NMR-metabolomics; Pyrus communis; polyamines; self-incompatibility; transglutaminase.

FIGURE 1

Transglutaminase activity was tested in NP, AxA and AxW styles of Pyrus communis on both N, N′-dimethylcasein (A) and fibronectin (B). Hydrated and in vitro germinated pollen were also analyzed for TGase activity (inserts). TGase activity is expressed as units (U) of specific activity (means ± SD) per mg of protein. Means ± SD of three experiments analyzed in triplicate are reported. Samples indicated with asterisks were significantly different each other. The data were analyzed by one-way ANOVA and Tukey’s post-test. A p value < 0.05 was regarded as statistically significant. ^∗∗p < 0.01. ^∗∗∗p < 0.001.

FIGURE 2

Transglutaminase relative quantification, expressed as absorbance units (Abs 450 nm) in NP, AxA and AxW styles of Pyrus communis. Means ± SD of three experiments analyzed in triplicate are reported. Samples indicated with different letters were significantly different each other. The data were analyzed by one-way ANOVA and Tukey’s post-test. A p value < 0.05 was regarded as statistically significant. Bars marked with the same letter are not significantly different.

FIGURE 3

Compatible (A,B) and incompatible (C–F) pollinated styles of pear analyzed with ID10 anti-TGase antibody. A,B: examples of AxW pollinated styles labeled with ID10 in indirect immunofluorescence. In addition to a diffuse background, no specific signal was detected. C,D: examples of AxA pollinated pear styles investigated in indirect immunofluorescence with ID10 antibody. Pollen tubes have a punctiform labeling along their surface (arrows). Bars in A–D: 10 μm. E,F: Immunogold labeling with ID10 antibody on AxA pollinated styles; the images show details of pollen tubes with cell wall-associated labeling in the form of distinct clusters (curly brackets in E, square bracket in F). Bars in E,F: 500 nm.

FIGURE 4

Free (A), soluble-conjugated (B) and insoluble-bound (C) polyamine pattern in NP, AxA, and AxW styles of Pyrus communis cv Abbé Fétel. For each polyamine, different letters (lowercase, italic, and bold) indicate significant differences at P < 0.05. Means ± SD of three experiments analyzed in triplicate are reported.

FIGURE 5

(A) From top to bottom: AxW; AxA and NP ¹H NMR full spectra of representative samples belonging to the three different classes (residual solvent signals have been removed; TMSP = standard). Extended spectral regions (B) form δ 0.5 to 3.3; (C) form δ 3.9 to 5.5; (D) δ 5.6 to 9.3. Numbers indicated diagnostic signals of the most variated metabolites: 1, fatty acids; 2, α-linolenic acids; 3, valine; 4, isoleucine; 5, alanine; 6, quinic acid; 7, GABA; 8, malic acid; 9, succininc acid;10, glutamine; 11, β-glucose; 12, sucrose; 13, α-glucose; 14, quercetin like flavonoids; 15, cynnamoyl derivative; 16, shikimic acid; 17, fumaric acid; 18, p-hydroxybenzoyc acid; 19, kaempherol like flavonoid; 20, trigonelline.

FIGURE 6

(A) 1H NMR based-PCA Score Scatter Plot, colored according to the three categories of samples. Each category included four replicates. Although the model is unsupervised, it individuated three different metabolic patterns among the given samples. (B) PLS-DA Score Scatter Plot, supervised model where the three classes (NP, AxW and AxA) were given as y variables. (C) B-Plot of the PLS-DA model, showing which NMR signals determined the grouping of the three classes indicated by the Score Scatter Plot. Black squares represent the three classes, dots are the 12 analyzed samples, inverted triangles are binned NMR signals and, among them, the red ones are the diagnostic signals of the most important metabolites. A general increase of all metabolites is observed in AxA, followed by AxW, while NP is less enriched in all metabolome, except glucose, which signals increase along component t[1]. Pollinated samples variate along the component t[2], indicating that AxW is more enriched in sucrose than AxA (this metabolite increases on positive component t[2]). 1, fatty acids; 2, α-linolenic acids; 3, valine; 4, isoleucine; 5, alanine; 6, quinic acid; 7, GABA; 8, malic acid; 9, succininc acid;10, glutamine; 11, β-glucose; 12, sucrose; 13, α-glucose; 14, quercetin like flavonoids; 15, cynnamoyl derivative; 16, shikimic acid; 17, fumaric acid; 18, p-hydroxybenzoyc acid; 19, kaempherol like flavonoid; 20, trigonelline.

FIGURE 7

Total flavonoid content of the three classes. Data are expressed in RU eq (rutin equivalent)/g of dry plant material. AxA samples showed a significant increase in total flavonoid content. Means ± SD of three experiments analyzed in triplicate are reported. The data were analyzed by one-way ANOVA and Tukey’s post-test. A p value < 0.05 was regarded as statistically significant as indicated by the different letters a, b.