Methylation of aquaporins in plant plasma membrane

Abstract

A thorough analysis, using MS, of aquaporins expressed in plant root PM (plasma membrane) was performed, with the objective of revealing novel post-translational regulations. Here we show that the N-terminal tail of PIP (PM intrinsic protein) aquaporins can exhibit multiple modifications and is differentially processed between members of the PIP1 and PIP2 subclasses. Thus the initiating methionine was acetylated or cleaved in native PIP1 and PIP2 isoforms respectively. In addition, several residues were detected to be methylated in PIP2 aquaporins. Lys3 and Glu6 of PIP2;1, one of the most abundant aquaporins in the PM, occurred as di- and mono-methylated residues respectively. Ectopic expression in Arabidopsis suspension cells of PIP2;1, either wild-type or with altered methylation sites, revealed an interplay between methylation at the two sites. Measurements of water transport in PM vesicles purified from these cells suggested that PIP2;1 methylation does not interfere with the aquaporin intrinsic water permeability. In conclusion, the present study identifies methylation as a novel post-translational modification of aquaporins, and even plant membrane proteins, and may represent a critical advance towards the identification of new regulatory mechanisms of membrane transport.

Figure 1. Representative sequence analysis of an N-terminal PIP1 tryptic peptide from root PM

A Met¹–Arg⁹ peptide of PIP1 was analysed by ESI–MS/MS and both N-terminal (b-type) and C-terminal (y-type) fragments were used to interpret the spectra. Note that PIP1;1, PIP1;2, PIP1;3 and PIP1;4 isoforms share a perfectly conserved Met¹–Arg⁹ sequence. Analysis was performed on the [M+2H]²⁺ species of 567.76 Da that corresponds to the [M+H]⁺ species of 1134.52 Da, as described in Table 2. The whole sequence can be deduced from the spectrum. _AcM, acetylated methionine.

Figure 2. Representative sequence analysis of N-terminal PIP2;1 tryptic peptides from root PM

Ala²–Arg¹⁶ peptides of PIP2;1 were analysed by ESI–MS/MS, and N-terminal (b-type), C-terminal (y-type) and internal fragments were used to interpret the spectra. Fragmented peptides are indicated according to either b and y series, or by their presumed sequence in the case of internal fragments. (A) Analysis was performed on the [M+2H]²⁺ species of 816.40 Da that corresponds to the [M+H]⁺ species of 1631.80 Da, as described in Table 3. The whole sequence can be deduced from the spectrum. The initiating methionine is cleaved. Lys³ is dimethylated (_2meK). Z=K+28 Da. (B) Analysis was performed on the [M+3H]³⁺ species of 549.27 Da that corresponds to the [M+H]⁺ species of 1645.80 Da, as described in Table 3. The whole sequence can be deduced from the spectrum. The initiating methionine is cleaved. Lys³ is dimethylated (_2meK), whereas Glu⁶ is monomethylated (_1meE). Z=K+28 Da; J=E+14 Da.

Figure 3. Ectopic expression of PIP2;1 in suspension cells

(A) Western-blot analysis, using an anti-PIP2;1 peptide antibody, of total protein extracts from independent cell lines, either untransformed (Untr.) or transformed with an empty vector (PG) or with a CaMV70S::PIP2;1 construct (lines 1–5). Equal amounts of proteins (5 μg) were loaded on to each lane. (B) MALDI–TOF analysis of the trypsin-cleaved 28 kDa band from PM of suspension cells overexpressing PIP2;1. Peptides derived from PIP2;1 are described by their monoisotopic mass, sequence and position inside the protein sequence. Three peptides were assigned to endogenous PIP aquaporins: (●), [M+H]⁺=1017.55 Da, attributed to sequence VGANKFPER of PIP1 isoforms; (●●), [M+H]⁺=1134.52 Da, attributed to sequence _acMEGKEEDVR of PIP1;1/PIP1;2/PIP1;3/PIP1;4 isoforms; (●●●), [M+H]⁺=1312.65 Da, attributed to sequence SFGAAVIYNNEK of PIP2;4/PIP2;7/PIP2;8 isoforms. The trypsin autolysis peptides are also indicated (*). _1meE, monomethylated glutamic acid; _2meK: dimethylated lysine.

Figure 4. Relationship between water transport activity and PIP2 abundance in purified PM vesicles

(A) Representative time course of stopped-flow scattered light intensity following imposition of an inwardly directed osmotic gradient on PM vesicles purified from PG (bottom trace) and WT-PIP2;1 (top trace) cells. The data were fitted with a single exponential function with a rate constant of k_exp=5.90 and 9.39 s⁻¹ for PG and WT-PIP2;1 cells respectively. (B) PM vesicles were prepared from at least two representative cell lines of each genotype, i.e. from cells containing an empty vector (PG) or overexpressing the WT, E6A, K3A or K3R form of PIP2;1. PM vesicles were probed for their water transport activity by stopped-flow measurements (k_exp) and for PIP2 abundance by an ELISA. For each cell line, water transport activity was calculated as the ratio of k_exp to PIP2. The specific activity of overexpressed PIP2;1 was deduced by substitution of the mean water transport activity measured in PG cells and was expressed as percentage of the specific activity of WT form. Each measurement was performed in triplicate from the indicated number (n) of independent cell lines.