iTRAQ-based proteomic analysis reveals the effect of ribosomal proteins on essential-oil accumulation in Houttuynia cordata Thunb.

Plant material and sample source

H. cordata wild type plants were grown in medical plant garden of pharmacy college, Navy Medical University, under the natural climate. Different tissues samples for flower, stem, leave and underground-stem were collected from flourishing plants in summer. They were put in a 1.5 ml tube and immediately in liquid nitrogen and stored at −80 °C until further use.

CDS library prediction

The RNA of flower, stem, leave and underground-stem tissues were extracted by Trizol kit as described. Total amounts and integrity of RNA were assessed by the RNA Nano 6000 Assay Kit of the Bioanalyzer 2100 system (Agilent Technologies, Santa Clara, CA, USA). cDNA library construction started with mRNA purification and then doble- strand cDNA synthesis and was quantified to ensure the quality of library. The separate libraries are pooling equivalently according to the concentration of samples, then being sequenced by the use of the Illumina NovaSeq 6000. The transcriptome assembly was performed using Trinity software (version 2.6.6) (Grabherr et al., 2011). Several databases were used to annotate gene functional annotation: Nr (NCBI non-redundant protein sequences); Nt (NCBI non-redundant nucleotide sequences); Pfam (Protein family); KOG/COG (Clusters of Orthologous Groups of proteins); Swiss-Prot (A manually annotated and reviewed protein sequence database); KO (KEGG Ortholog database); GO (Gene Ontology).

CDS represents the sequence encoding protein. CDS prediction firstly is mapped into NR and Swissprot. protein library. If mapped successfully, the open reading frame (ORF) of the transcript is isolated and the coding region sequence is translated into amino acid sequence (5′–>3′). For the sequences without successfully, the ORF was predicted by TransDecoder (3.0.1) software.

Protein extraction

Protein extraction was carried out according to the protocol of Isaacson et al. (2006) with some modifications. Frozen tissues were homogenized with a plant-tissue grinder. Protein extraction buffer was added to the sample and mixed gently in a vortex for 5∼10 min. Adding two volume of phenol saturated with Tris–HCl (pH 8.0) to each sample and then mixing for 10 min, followed by centrifugation at 14,000 g for 10 min at 4 °C. The upper phenolic phase was picked and mixed with 1.2 volume of protein extraction buffer, and then centrifuged. The upper phenolic phase was again collected and mixed with pre-cooled 0.1M ammonium acetate in methanol, followed by the incubation for 6 h at −20 °C. The precipitate was obtained by centrifugation at 14,000 g for 10 min and washed 1–2 times by pre-cooled methanol. Then the precipitate was washed 2–3 times by acetone containing 0.07% β-Mercaptoethanol and dried at room temperature. Protein concentration was determined by the BCA method (Smith et al., 1985).

Protein digestion

Protein digestion was carried out according to the method of FASP (Wiśniewski et al., 2009). Dried protein powder sample was resolved in 8M urea with Tris-Hcl (pH 8.0). 1M reducing reagent (DTT) was added to cell protein lysates and incubated at 55 °C for 1 h, followed by 5 µL 1M cysteine-blocking reagent (IAA) for 30 min at 25 °C in dark. Then the samples were transferred into 10 KDa ultrafiltration tube and centrifuged 14,000 g for 15 min at 4 °C. Then 100 µL 8M urea was added twice to the ultrafiltration membrane to protein denaturation completely. 0.5 M TEAB was added and centrifuged three times at 14,000 g for 15 min to clean denaturation reagent. Sequencing grade trypsin (Enzyme: Substrate = 1:50) was incubated with samples at 37 °C overnight. Next day, the peptide samples were collected and added 1% formic acid to stop the reaction.

iTRAQ labeling and high pH reverse phase separation by HPLC

The peptides were subjected to vacuum centrifugation to facilitate drying. Subsequently, the dried peptides were suspended in TEAB buffer and labeled by iTRAQ 8-plex kits in accordance with the manufacturer’s instructions (AB SCIEX Inc., Framingham, MA, USA, AB SCIEX Inc., USA). The labeling reaction was initiated by the addition of a reagent vial to the digested peptides, which were then incubated at room temperature for a period of two hours. Following this, 100 µL of water was added to halt the reaction. The labeling scheme was as follows: The tags used were 113 and 114, F1 and F2; Tags 115 and 116, L1 and L2; Tags 117 and 118, S1 and S2; Tags 119 and 121, Us1 and Us2. Following this, all the samples were pooled for mass spectrometry analysis. High pH reverse phase separation by HPLC was performed in accordance with the previously decribed methodology (You et al., 2017), with certain modifications. The Agilent 1100 HPLC with Agilent Zorbax Extend C18 column (2.1 × 150 mm, 5 µm) were used for peptide separation at a flow rate of 1.0 ml/min. The mobile phases A was 20 mM ammonium formate (pH = 10), while the mobile phases B was 80% acetonitrile with 20 mM ammonium formate (pH = 10). The entire gradient course lasted for 65 min, with the following profile: 5 min at 5% B, 30 min at 15% B, 45 min at 38% B, 46 min at 90% B, 54.5 min at 90% B, 55 min at 5% B, and finally 65 min at 5% B.

LC–MS/MS analysis

Microflow LC-MS/MS analysis was conducted by coupling Eksigent Micro LC-1D plus system to Triple TOF 5600 System (AB SCIEX Inc., Framingham, MA, USA). Samples for the proteome analysis were resuspended in 2% acetonitrile containing 0.1% formic acid. Peptide loading and washing were done on a trap column (ChromeXP C18_CL-3 µm, 120A, 350 µm × 0.5 µm, AB,SCIEX) at a flow rate of 10 µL/min in 2% acetonitrile (0.1% formic acid) for 7 min. Peptide separation was performed on an analytical column (0.075 × 150 mm, 3 µm, 120A, AB SCIEX Inc.) at a flow rate of 5 µL/min using a 120 min gradient from 5% to 80% solvent B (solvent A: 2% acetonitrile with 0.1% formic acid in LC-MS grade water; solvent B: 98% acetonitrile with 0.1% formic acid in acetonitrile) for proteome analysis. A spray voltage of 2.3 kV and ESI ion source temperature of 150 °C were employed to facilitate the ionisation of peptides. A switch from MS to MS2 scanning was automatically initiated based on the data collected from the instrument. For the full proteome samples, full scan MS spectra (m/z 350–1,500) were acquired. In the Time of Fight detector, high-resolution MS2 spectra were acquired for the 40 most abundant precursor ions. This analysis was conducted with the objective of fragmenting precursors with charge states between 2 and 5.

Analysis of proteomic data

Peptide and protein identification and quantification data files were searched with Protein Pilot software (v.5.0) using default parameters (Zhang et al., 2014). MS/MS spectra were searched against CDS library (Hc. blast.pep.fasta). Trypsin was the only specific protease and iTRAQ-8-plex (peptide labelled) was chosen as sample labeling type (FDR ≤ 1%). The annotations of the identified proteins, combined with the result of BLAST alignments. The screening criteria for differential expression proteins (DEPs) was —Fold Change (FC)— ≥ 2 and a p-value ≤0.05 in aerial parts (Leaf, Stem, Flower) compared to underground -stem. The Weighted Gene CoExpression Network Analysisn (WGCNA) of identified proteins was performed with the R package WGCNA (https://cran.rstudio.com/web/packages/WGCNA), and further bioinformatic analysis of DEP was performed using the OECloud tools (https://cloud.oebiotech.cn/), based on the Gene Ontology (GO) categories and KEGG Ortholog database.

Bioinformatic analysis of different expression proteins

The KEGG pathway enrichment profile of DEPs was analyzed by OECloud tools with FDRs less than 0.05, and DEPs data was displayed using volcano plots and Wayne diagram with the help of the image GP. The prediction of functional protein association networks was analyzed by the STRING database (http://string-db.org/).

Plant morphology and chemical composition

The general morphology of the H. cordata plant during its flowering period is illustrated (Fig. 1A). The plant is divided into four sections: flowers, stems, leaves, and underground stems (Us). Flowers, stems, and leaves are classified as aerial parts (AP). Medicinal components application in both AP and Us, were recognized for their heat-clearing and detoxifying properties in traditional Chinese medicine. Among the constituents of H. cordata essential oil, the most predominant compounds were 2-undecanone (23.96–36.07%), β-myrcene (12.57–14.29%), bornyl acetate (6.03−8.61%), and β-pinene (trace - 23.29%) (chemical structures shown in Fig. 1B). The previous findings indicated that the essential oil content is higher in the aerial parts of H. cordata compared to its underground stems (Lu et al., 2006a). These findings suggest that the substantial variation of essential oil content could potentially be linked to the differential expression of proteins across distinct plant parts in H. cordata.

Assembled transcriptome analysis

To perform proteomics analysis, RNA-seq is carried out to build non-model plant proteins identification search database. Based on transcriptome result, 90,067 transcripts were assembled in tissues pool (flower, stem, leave and underground-stem) of H. cordata, of which 33,493 were unigenes that represents the longest transcript of each gene. More than 50% of unigenes were distributed between 300–1000 bp (Fig. S1A). 21,661 unigenes (64.67%) were annotated by Nr (NCBI non-redundant protein sequences) database, while annotated 13,290 unigenes were screened out by Nt (NCBI nucleotide sequences) database. 15,756 were annotated by GO (Gene Ontology) and involved in cellular and metabolic process (Fig. S1B and Fig. 1C). 8,928 unigenes were mapped to KO and KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway which mainly participated in translation and signal transduction (Figs. 1C and 1D). In total of 4,177 unigenes existed in multiple databases. 19,271 peptides sequence of CDS were predicted through Nr and SwissProt. Assembled transcriptome database serves as an extensive public repository, serving as a valuable search resource for proteomic studies of H. cordata.

Overview of proteomics analysis

The proteomics analysis identified a total of 3,261 proteins across the four parts (flower, stem, leaf, and underground-stem) of H. cordata. Notably, the expression clusters exhibited significant distribution shifts among the various plant parts, as shown in Fig. 2A. In terms of GO categories, 1,470 proteins were involved in biological processes, 1,194 proteins were linked to cellular components, and 2,247 proteins were associated with molecular functions (Fig. S2A). Moreover, 1,684 proteins were assigned to KEGG pathways, with 31 notable signalling pathways exhibiting significance (p value ≤ 0.05). The three leading pathways were those of dihydrolipoamide dehydrogenase, photosystem I P700 chlorophyll apoprotein A1, and the 26S proteasome complex subunit DSS1 (Fig. S2B). To assess the correlation coefficients among the expressed proteins across various parts of H. cordata, WGCNA analysis was conducted. This analysis classified the 3,930 proteins into five distinct modules, revealing correlations that ranged from high to low among the MEturquoise proteins. Notably, the proteins expressed in the underground-stem exhibited a lower correlation coefficient with those in the aerial parts (Fig. 2B and Fig. S3).

DEPs analysis of aerial parts and underground-stem

A significance cutoff was established with criteria including a fold change ≥ 1.5 and a p-value ≤ 0.05 (t-test), along with a false discovery rate of ≤ 1%. Compared with the proteins in underground-stem, 306 proteins were up-regulated in leaf, 192 were up-regulated in flower, and 266 were up-regulated in stem, there were 61 proteins increased both in leaf and stem (Fig. 3A). In comparison to the underground-stem, ribosomal proteins L23A and S28 displayed a notable upregulation in flower tissue. Similarly, proteins L1, L30e, L23A, L4, and L29 exhibited upregulation in both the stem and leaf. Conversely, P3-like and S8 were found to be downregulated in the stem and leaf (Figs. 3B–3D). Furthermore, the differential abundance proteins (DAPs) were associated with KEGG pathways, with 251 proteins being mapped to these pathways. Notably, this study revealed 12 significant signalling pathways (p-value ≤ 0.05), with the differentially expressed proteins (DEPs) being most prominently enriched within the large subunit ribosomal protein group (Fig. 3E). The findings suggest that large subunit ribosomal proteins might play a role in influencing essential oil accumulation in aerial parts (AP).

The expression profile of interesting proteins

In this study, a total of 34 ribosomal proteins were identified, including 10 small subunit ribosomal proteins and 24 large subunit ribosomal proteins (Table 1). The majority of these exhibited higher expression levels in the leaf and stem, a pattern consistent with the spatial distribution of active components (Fig. 4A) (Lu et al., 2006a). The internet analysis revealed that the majority of ribosomal proteins in H. cordata exhibited robust interactions with those in Arabidopsis thaliana according to STRING model data. (Fig. 4B, Table 1).