In this project, a range of technologies, such as HPLC, malonylation peptide enrichment and MS-based proteomics technologies, were combined to study the qualitative proteomics of malonylation in S. sanghuang (Fig. 1a). The results showed that the peptide score was between − 10 and 10 (Fig. 1b). The tolerance of peptides was in a reasonable range. The distribution of identified peptide lengths was examined, and the lengths of most peptides were between 7 and 22 (Fig. 1c), meeting the requirements of proteomic analysis. The MS results of malonylated peptides are summarized in Additional file 1: Fig. S1. Consequently, 713 malonyl-modified sites matched to 255 different proteins were identified in S. sanghuang (Additional file 2: Table S1). Among them, many are related to triterpene synthesis. Farnesyl pyrophosphate synthases (FPPs), which are are pivotal enzymes in the main pathway of triterpene synthesis, the mevalonate (MVA) pathway, were found to be malonylated.
Pattern analysis of malonylated sites
For the purpose of valuing the distribution of malonylation sites in S. sanghuang, the number of identified modification sites was calculated for each protein. As shown in Fig. 2a, 47% of the proteins had one malonylation site, while only 18%, 7%, 12%, 3%, and 13% of the proteins contained 2, 3, 4, 5, and 6 or more modification sites, respectively. It has been documented that modification positions are prioritized at specific sites for lysine (Additional file 2: Table S3). Therefore, the compositional frequencies of the amino acids surrounding lysine were examined. In Fig. 2c, lysine (K) had the highest frequencies in the − 10 to + 10 position, whereas arginine (R) and glutamate (E) had the lowest frequencies. Hence, proteins with this group are the preferred substrates for malonyltransferases in S. sanghuang. Consistent with the results of the motif enrichment heatmap (Fig. 2b), only one motif was detected. K indicates modified lysine sites and * denotes random amino acid sites (Fig. 2d). To elucidate the secondary structure of proteins and the correlation between modified lysines, the secondary structures of all malonylated proteins in S. sanghuang were examined (Fig. 2e). The malonylation sites are located more in the coiled-coil regions (P = 0.18) than in the α-helical (P = 0.01) and β-strand (P = 0.48) regions, suggesting that malonylation may favour the disordered structures of S. sanghuang. In addition, we assessed the surface accessibility on lysine malonylated sites and found 39.62% of the unmodified lysine residues were located on the protein surface, compared to 39.54% of the modified lysine sites (Fig. 2f). As such, the protein's surface accessibility may be influenced by lysine malonylation.
Functional annotation and cellular localization of malonylated proteins in S. sanghuang
To obtain a better comprehension of the malonylated proteins in S. sanghuang. Upon their corresponding biological processes and molecular functions, we annotated and classified the identified proteins. GO analysis showed that malonylated proteins had extensive activity molecular functions and biological processes in S. sanghuang. The group of malonylated proteins in the largest classification of biological processes consists of enzymes related to metabolism (53%) (Fig. 3a). The majority of malonylated proteins were associated with organocyclic compound binding (15%), heterocyclic compound binding (15%) and structural constituent of ribosome (10%) within the molecular functional classification (Fig. 3b). Characterization of the subcellular localization of malonylated proteins showed the modified proteins were found in the cytoplasm (36%), mitochondria (31%), and nucleus (21%) (Fig. 3c). Consequently, malonylated proteins have multiple functions and are broadly found in S. sanghuang.
Functional enrichment analysis of malonylated proteins
To summarize and analyse the proteins and their functions, we performed functional enrichment analysis of the obtained malonylome by GO and KEGG pathway and protein domains analyses (Additional file 2: Table S5, Additional file 2: Table S6). Proteins associated with structural components of the ribosome were highly enriched by functional analysis of GO molecules (Additional file 2: Table S4). Based on GO cellular component classification, proteins that are located in the ribosomal subunit, ribosome, large ribosomal subunit, small ribosomal subunit, and cytosol are more likely to be malonylated (Additional file 1: Fig. S2). The domain enrichment studies indicated that these proteins are core histone H2A/H2B/H3/H4, proteasome, beta-ketoacyl synthase, 1-cys peroxiredoxin, acyl transferase domain, isocitrate/isopropyl malate dehydrogenase, and oxidoreductase flavin adenine dinucleotide (FAD) binding domain (Additional file 1: Fig. S3). These enriched domains play a crucial role in glycolysis, polysaccharide synthesis and the tricarboxylic acid (TCA) cycle in S. sanghuang. Thus, malonylated proteins participate in a myriad of cellular processes. To probe the process of malonylation regulation, we performed enrichment analysis of proteins corresponding to malonylation modification sites in KEGG pathways (Fig. 4). Several pathways of the enriched proteins in the ribosome, glucuronide and dicarboxylic acid metabolism, TCA cycle, glycolysis/gluconeogenesis and pyruvate metabolism were shown. Conclusively, malonylated proteins are enriched in several types of proteins and pathways and play a pivotal role in the lysine malonylation of S. sanghuang..
PPI network of malonylated proteins in S. sanghuang
To determine how the identified proteins are associated with multiple pathways, a PPI network was constructed. Ninety proteins were detected in the PPI database. (Fig. 5, Additional file 2: Table S7), presenting a global vision of how the identified malonyl proteins are involved in the multiple pathways of S. sanghuang. Analysis of the STRING PPI network with Cytoscape identified three heavily correlated clusters of malonylated proteins, including those associated with ribosomes, metabolic pathways, and biosynthesis of secondary metabolites in S. sanghuang. Above all, we conclude that malonylation is a critical PTM for proteins in S. sanghuang and helps interactions and coordination with diverse pathways.
Malonylated proteins associated with the biosynthesis of bioactive compounds in S. sanghuang
Malonylated proteins were identified based on functional enrichment, and proteins related to ribosomes, glucuronide and dicarboxylic acid metabolism, glycolysis/gluconeogenesis, TCA cycle, methane metabolism, oxidative phosphorylation, and pyruvate metabolism were found to be greatly enriched (Fig. 4). These findings suggested that the malonylation of lysine may be essential in the biosynthesis of bioactive compounds in S. sanghuang. To further confirm these findings, we analysed malonylated proteins associated with triterpene and polysaccharide biosynthesis in S. sanghuang. Consistent with these hypotheses, the TCA cycle, glycolysis supplied compounds for the biosynthesis of triterpenoids and polysaccharides. A grand total of 26 enzymes associated with triterpene and polysaccharide biosynthesis were found to be malonylated (Fig. 6, Additional file 2: Table S8). In Fig. 6, a large number of enzymes are supported by malonylation in glycolysis and TCA cycle, suggesting that malonylation may be associated with multiple levels of intracellular metabolism. Furthermore, our results also showed that 51 malonyl modified sites detected on ribosomes, such as ribosomal proteins L24, L13a, and S3, were closely tied up bioactive functions (Fig. 7).