Molecular identification and functional characterization of the first Nα-acetyltransferase in plastids by global acetylome profiling

Abstract

Protein N(α) -terminal acetylation represents one of the most abundant protein modifications of higher eukaryotes. In humans, six N(α) -acetyltransferases (Nats) are responsible for the acetylation of approximately 80% of the cytosolic proteins. N-terminal protein acetylation has not been evidenced in organelles of metazoans, but in higher plants is a widespread modification not only in the cytosol but also in the chloroplast. In this study, we identify and characterize the first organellar-localized Nat in eukaryotes. A primary sequence-based search in Arabidopsis thaliana revealed seven putatively plastid-localized Nats of which AT2G39000 (AtNAA70) showed the highest conservation of the acetyl-CoA binding pocket. The chloroplastic localization of AtNAA70 was demonstrated by transient expression of AtNAA70:YFP in Arabidopsis mesophyll protoplasts. Homology modeling uncovered a significant conservation of tertiary structural elements between human HsNAA50 and AtNAA70. The in vivo acetylation activity of AtNAA70 was demonstrated on a number of distinct protein N(α) -termini with a newly established global acetylome profiling test after expression of AtNAA70 in E. coli. AtNAA70 predominately acetylated proteins starting with M, A, S and T, providing an explanation for most protein N-termini acetylation events found in chloroplasts. Like HsNAA50, AtNAA70 displays N(ε) -acetyltransferase activity on three internal lysine residues. All MS data have been deposited in the ProteomeXchange with identifier PXD001947 (http://proteomecentral.proteomexchange.org/dataset/PXD001947).

Keywords: Arabidopsis thaliana; AtNAA70; Chloroplast; Nα-acetyltransferase; Plant proteomics.

Figure 1

Structure modeling of chloroplastic AT2G39000 from Arabidopsis thaliana. (A) Superposition of modeled N-domain (brown) and C-domain (blue) from AT2G39000 with the template 2PSW (light gray) is shown. The enzyme structures are in cartoon representation, whereas CoA molecule is displayed in sticks representation. The modeled domains are labeled by the first and the last modeled residues. (B) Enlargement of modeled AT2G39000 C-domain harboring a CoA molecule (with its molecular surface). The CoA molecule and selected residues that probably interact with CoA are shown in stick presentation. (C) Schematic representation of AT2G39000 domain organization. TP, transit peptide. (D) Structural alignment of N- and C-domains with 2PSW. The orange box highlights the linker of AT2G39000 that is missing in 2PSW. The violet box designates the acetyl-CoA binding motif (RxxGxG/A). Conserved residues are marked by horizontal bars above the sequences. Green areas represent α-helices, yellow areas indicate β-strands.

Figure 2

AtNAA70 is localized in chloroplasts. Full length AtNAA70 (AT2G39000) in fusion with EFYP was ectopically expressed under control of the 35S-promotor in mesophyll protoplasts extracted from six-week-old Arabidopsis leaves. The YFP signal (green) of AtNAA70-EYP (A) perfectly overlap (C) with chlorophyll auto-fluorescence (red signal, B) determined in the same protoplasts. Protoplasts transiently transformed with the empty pFF19-EYFP vector (D–F) display chlorophyll auto-fluorescence (E) but no specific EFYP signal (D) using the same settings.

Figure 3

A newly established MS/MS-based in vivo Nat activity test demonstrates enzymatic acetylation activity of AtNAA70. (A) Number of identified acetylated N-terminal peptides after digestion of E. coli soluble protein lysates (black bar) with trypsin and enrichment of N-terminal peptides by SCX. Wild type E. coli and E. coli cells expressing either the auxiliary subunit (AtNAA15, negative control) or the catalytically active subunit of NatA (AtNAA10, positive control) were analyzed to verify the suitability of the new enzymatic Nat test. Expression of catalytic subunit of the NatA and the candidate Nat, AtNAA70, resulted in significant increase of acetylated peptides. (B–E) Web logos of N-termini from E. coli proteins that were found to be specifically acetylated after expression of AtNAA10 (B, C) or AtNAA70 (D, E). These proteins start with the iMet (B, D) or were subject of iMet removal (C, E). (N = number of acetylated peptides).

Figure 4

Comparison of AtNAA70 substrate specificity with N-termini of known acetylated plastid localized proteins. Web logo of substrate specificity for His₆-MBP-NAA70 (N = 121, A) is compared with experimentally identified acetylated N-termini of chloroplastic stromal proteins (B, C). The proteins (N = 195; [19]) are either imported from the cytosol (C) or translated (N = 10) in the chloroplast (D). The red arrow indicates similarity between substrates acetylated by AtNAA70 in E. coli and acetylated chloroplastic proteins in Arabidopsis.

Figure 5

AtNAA70 possesses N^ε-acetylation activity on internal Lys residues. (A) MS spectrum of the AtNAA70 internal LIWKAEALAKNWGCR peptide with three distinct isotope distributions corresponding to the peptide with A₁) two heavy acetylations (d3-Acetyl, chemical modification), A₂) a mix of heavy and light acetylations (d3-Acetyl/Ac), A₃) two light acetylations (endogenous). The MS signal extractions displayed in A₁ to A₃ are used to determine the relative abundance of each form (unmodified peptide: 84%, partial Lys acetylation (on Lys 217 or 223): 14%, Acetylation of both Lys 217 and 233: 2%. (B) Annotated synthetic MSMS of the peptide (combination of spectra with precursor mass of 953.45 ± 0.5 Da). (C) Immunological detection of ac-Lys on AtNAA70 in absence (–) or presence (+) of acetyl-CoA for indicated time points using an N^ε-acLys specific antiserum. Staining of AtNAA70 with amido black serves as loading control. (D) Determination of AtNAA70 auto-Kat activity by incorporation of [³H]-acetyl moieties into purified AtNAA70 (black circles) over time. Incubation of [³H]-acetyl-CoA with heat inactivated AtNAA70 (white circles) for 60 min served as a negative control. (N = 3)