Cold acclimation is accompanied by complex responses of glycosylphosphatidylinositol (GPI)-anchored proteins in Arabidopsis

Abstract

Cold acclimation results in changes of the plasma membrane (PM) composition. The PM is considered to contain specific lipid/protein-enriched microdomains which can be extracted as detergent-resistant plasma membrane (DRM). Previous studies in animal cells have demonstrated that glycosylphosphatidylinositol-anchored proteins (GPI-APs) can be targeted to microdomains and/or the apoplast. However, the functional significance of GPI-APs during cold acclimation in plants is not yet fully understood. In this study, we aimed to investigate the responsiveness of GPI-APs to cold acclimation treatment in Arabidopsis We isolated the PM, DRM, and apoplast fractions separately and, in addition, GPI-AP-enriched fractions were prepared from the PM preparation. Label-free quantitative shotgun proteomics identified a number of GPI-APs (163 proteins). Among them, some GPI-APs such as fasciclin-like arabinogalactan proteins and glycerophosphoryldiester phosphodiesterase-like proteins predominantly increased in PM- and GPI-AP-enriched fractions while the changes of GPI-APs in the DRM and apoplast fractions during cold acclimation were considerably different from those of other fractions. These proteins are thought to be associated with cell wall structure and properties. Therefore, this study demonstrated that each GPI-AP responded to cold acclimation in a different manner, suggesting that these changes during cold acclimation are involved in rearrangement of the extracellular matrix including the cell wall towards acquisition of freezing tolerance.

Keywords: Apoplast; Arabidopsis; GPI-anchored protein; cold acclimation; freezing tolerance; microdomain; plasma membrane; proteomics..

Fig. 1.

One-dimensional SDS–PAGE profiles of GPI-AP fractions with or without PI-PLC treatment. Each GPI-AP fraction (1 μg of protein equivalent) was loaded on a polyacrylamide gel, separated, and visualized by silver staining (Kawamura and Uemura, 2003). Arrowheads indicate major protein bands visualized in the presence or absence of exogenous PI-PLC during the preparation process. Non-acclimated (NA) and cold-acclimated (CA) samples were used.

Fig. 2.

Venn diagrams and bar graph of GPI-APs identified in each GPI-AP predictor and sample fraction. (A) The relationship among five different GPI-AP prediction programs in GPI-AP lists identified in the present study. (B) Similarities and dissimilarities of GPI-APs identified among each fraction. Venn diagrams were generated based on the AGI code using the online Venn diagram drawing program (http://bioinformatics.psb.ugent.be/webtools/Venn/). (C) The number of GPI-APs identified in each sample fraction in the following five different GPI-AP prediction programs: Borner et al. (2003), big-PI Plant Predictor (Eisenhaber et al., 2003), PredGPI (Pierleoni et al., 2008), GPI-SOM (Fankhauser and Maser, 2005), and fragAnchor (Poisson et al., 2007).

Fig. 3.

Functional categorization of total proteins and GPI-APs identified in non-acclimated samples. All GPI-APs identified were classified into 11 functional categories based on the definition proposed by Bevan et al. (1998). Proportions of each functional category in bar charts were calculated by the proportions of each functional category based on the normalized abundance of each GPI-AP identified. The upper and lower part of the graph indicate the functional distribution of total proteins and only GPI-APs identified in non-acclimated samples.

Fig. 4.

Cold acclimation responsiveness of GPI-APs. Non-acclimated (NA) and cold-acclimated (CA) PM samples (equivalent of 2mg of protein) were incubated with or without PI-PLC. After the treatment, the protein content of each fraction was determined by Bradford assay (n=4; A). Alternatively, the proportions of identified GPI-APs in the total proteins in each sample were calculated based on the abundance of each protein obtained from the quantitative proteomic approach (n=3–7; B).

Fig. 5.

Global proteomic changes of GPI-APs in each functional category during cold acclimation. GPI-APs identified in the four sample fractions were categorized into 11 functions as described in Fig. 3. The mean normalized abundance of GPI-APs in each functional category was summed and differences in this value between non-acclimation (NA) and cold acclimation (CA) were calculated.

Fig. 6.

Distribution of cold acclimation changes of GPI-APs from GPI-AP/PI-PLC (+), PM, DRM, and apoplast fractions. Changes in abundance of each GPI-AP during cold acclimation were transformed to log₂ values and are shown in a histogram.

Fig. 7.

Schematic presentation of potential roles of representative GPI-APs identified in the present study in cell wall changes during cold acclimation. In the present study, 163 GPI-APs were identified in total. Among them, lipid transfer proteins (LTPs), fasciclin-like arabinogaractan proteins (FLAs), glycerophosphoryldiester phosphodiesterase-like proteins (GPDLs), blue-copper-binding protein (BCB), and O-glycosyl hydrolase family 17 proteins (GH17) predominantly increased during cold acclimation (CA). The proteins included in the four protein families have potential roles in wax deposition, cell wall organization, cellulose and pectin modification, and lignin synthesis and callose degradation, respectively. Although detailed mechanisms are not yet fully understood, physical and biochemical properties of the extracellular matrix including the cell wall are considered to be modified during cold acclimation in order to withstand extracellular ice propagation and freeze-induced dehydration. This study proposed the possibility that GPI-APs could be involved in remodeling of cell wall and plasmodesmal communication during cold acclimation, and PM microdomains may regulate cold-acclimation-associated activities of GPI-APs. On the other hand, changes of GPI-APs in the apoplast fraction were vastly different from those in the PM and GPI-AP/PI-PLC (+) fractions, suggesting that the activity and/or cellular localization is partly regulated by GPI-PLC.