iTRAQ-based proteomics analysis on insomnia rats treated with Mongolian medical warm acupuncture

We purchased 30 healthy male Sprague-Dawley rats aged 8–10 weeks and weighing 220 ± 10 g from the Department of Laboratory Animals of Peking University Health Science Center. The protocol was approved by the Laboratory Animal Welfare Committee. The animal experiments conformed the U.S. National Institutes of Health guidelines and the guidelines from Directive 2010/63/EU of the European parliament on the protection of animals used for scientific purposes.

Animal experiments were completed at the animal experiment center of Inner Mongolia medical university. All animals were maintained under controlled conditions (21–22°C; 55–65% humidity; 12-h light, 12-h dark cycle) with free access to rodent soy-free food and water. The animals were acclimated to their new circumstances for one week. Then, the animals were randomly divided into three groups (n=10/group): healthy controls (C), insomnia model group (M) and the warm acupuncture-treated insomnia group (W).

Rats in the M and W groups use peritoneal injection of p-chlorophenylalanine (300 mg/ kg) to construct the insomnia rat model, and the dosage was adjusted with the increase in body weight. The concentrations have been described previously. In the warm acupuncture group, the rats were adaptively raised for 7 days and intraperitoneally injected with normal saline (0.l ml/kg) between 8:30 and 9:00 a.m. Each day for two consecutive days from the eight day. Starting from the 10th day. Acupuncture treatment was performed at 08:00 every morning. Stainless steel acupuncture needles (0.35 mm × 20 mm, supplied by Inner Mongolia Yuanyang Traditional Chinese and Mongolian Science and Technology Development LLC) were used. Depth of needling at each point was 5 mm. The Dinghui Acupoint, Heyi Acupoint, and Xin Acupoint of each rat were stimulated with warm acupuncture for 15 min each time (Mongolian Model MY-I electric heating needle warmers) at ∼40°C (be careful not to burn the needling site) once each day for seven consecutive days [5,10].

All rats were killed at the anestrus period following anesthesia under 1% sodium pentobarbital (40 mg/kg intraperitoneal injection). Collected rats hypothalamic tissue. Finally, rats were euthanized by an excessive dose of 1% sodium pentobarbital (intraperitoneal injection).

The hypothalamus of all rat groups were harvested, SDT buffer was added to the sample, and then transferred into 2 ml centrifuge tubes, prefilled with appropriate amount of quartz sand and an additional 1/4 inch of ceramic bead. Following, the samples were homogenized in a MP tissue homogenizer (24×2, 6.0 M/S, 60 s, twice), ultrasonificated (80 W, 10-s ultrasound, 15-s interval, 10 cycles), and then incubated in boiling water for 15 min. After centrifugation at 14,000 g for 40 min, supernatants were filtered through a 0.22-µm filter. Protein concentrations were determined using the BCA method. The sample was stored a −80°C [11].

For each sample, 20 μg of proteins were mixed with 5× loading buffer, boiled for 5 min, and then separated by 12.5% SDS-PAGE (constant current of 14 mA, 90 min). Protein bands were visualized using Coomassie Blue staining.

For each group, 200 μg of proteins for each sample were incorporated into 30 μl SDT buffer (4% SDS, 100 mM DTT, 150 mM Tris-HCl pH 8.0). Following, the mixtures were boiled for 5 min, and then cooled down to room temperature. Next, 200 ml of UA buffer was added, mixed up, and transferred into 10 kDa ultrafiltration tubes and centrifuged at 14,000 g for 15 min, the filtrate was discarded (this step was repeated once). About 100 μl of iodoacetamide (IAA) buffer (100 mM IAA in UA) was added, then vortexed at 600 rpm for 1 min, and the mixture reacted for 30 min at room temperature, under dark conditions; then centrifuged at 14,000 g for 15 min. About 100 μl of UA buffer was added and the sample was centrifuged at 14,000 g for 15 min; the process was repeated twice. About 100 μl of 10 times diluted dissolution medium was added, and the sample was centrifuged at 14,000 g for 15 min; this step was repeated twice. Following, 40 μl of trypsin buffer (4 μg trypsin in 40 μl dissolution buffer) was added, mixed by vortexing at 600 rpm for 1 min, and then incubated at 37°C for 16–18 h. The sample column was transferred into a new collection tube, and centrifuged at 14,000 g for 15 min; then the sample was eluted by adding 40 μl of 10 times diluted dissolution buffer, followed by centrifugation at 14,000 g for 15 min. Peptides were desalinated using a C18 cartridge, and after lyophilization were re-suspended in 40 μl of dissolution buffer. Peptide concentration was estimated by measuring absorbance at OD280 [12].

About 100 μg peptide mixture of each sample was labeled using iTRAQ reagent according to the manufacturer’s instructions (Applied Biosystems). The experiment was carried out three times.

Labeled peptides from each group were mixed and separated using the AKTA Purifier 100. About 10 mM KH₂PO₄ with 25% ACN (pH 3.0) was used as buffer A, and 10 mM KH₂PO₄, 500 mM KCl with 25% ACN (pH 3.0) was used as buffer B. The chromatographic column was equilibrated with buffer A, and samples were loaded into the column through a sample injector with flow rate adjusted at 1 ml/min. The gradient was set as follows: linear gradient of 0–8% buffer B for 0–22 min; linear gradient of 8–52% buffer B for 22–47 min; linear gradient of 52–100% buffer B for 47–50 min; after 58 min buffer B concentration was set to 0%. During elution, the absorbance at 214 nm was recorded, and eluted solutions were collected every 1 min; totally, 30 eluted solutions were collected. After lyophilization the samples were desalinated using a C18 cartridge.

MASCOT is considered the golden standard in the proteomics qualitative analysis, thus, Mascot 2.2 was used for database searching in this study. The Uniprot protein Rattus norvegicus database (35,838 sequences) was used for search. Caramidomethylation of cysteines, iTRAQ labeling at the N-term and the lysine side chain amion groups were set as fixed modification. The oxidation of methionine and iTRAQ 4 plex(Y) were set as variable modification. The false discovery rate (FDR) for peptide was set to 1%. The minimum peptide length was set to 6. Enzyme specificity was set to be Trypsin and up to two missed cleavages per peptide were allowed. Mass tolerance for precursor ions was set to 20 ppm and for fragment ions at 0.1Da.

The process of GO annotations was performed on the targeted protein clusters using Blast2GO, and was divided into four steps: sequence alignment, extraction of GO annotations, GO annotation and supplementary annotation [13]. First, the targeted protein clusters were aligned with the corresponding protein sequences in the database, using the localization sequence alignment tool NCBI BLAST+ (ncbi-blast-2.2.28+-win32.exe), and then the first 10 aligned sequences with E-value ≤ 1e-3 were saved for further analysis. Following, the GO annotations that were associated with the proteins with high similarities with the target protein clusters or with homologous proteins, which had been blasted by the Blast2GO Command Line (dataset version: go_201608.obo, www.geneontology.org↗), were extracted. During annotation, Blast2GO Command Line annotated the extracted GO annotations to target proteins through comprehensive examination of the sequence similarities, GO annotation resource reliabilities, and the existence of circle plots between target and aligned proteins. After annotation, to further improve the annotation efficiency, conserved motifs in EBI database that matched with target proteins were scanned using InterProScan [14], and the motif-related functional information were annotated to target protein sequences. ANNEX was used to further supplement the annotated information, and linkages between different GO types were established to improve the annotation accuracy.

In the KEGG pathway database, KO (KEGG Orthology) terms were used for classification of genes and their product. Orthologous genes and their products were classified in the same pathway group, and were annotated with the same KO (or K) term. During KEGG pathway analysis of target protein clusters, first KEGG genes data were mapped, using KAAS (KEGG Automatic Annotation Server) [15], to KO terms in order to classify the target protein sequences, and then information of pathways containing these target proteins was automatically extracted according to KO classification.

While conducting enrichment analysis of the GO annotations or KEGG pathway annotations of target protein clusters, Fisher’s exact test was used to compare the distribution of each GO classification or KEGG pathway between target protein clusters and total protein clusters, and for evaluating the significance of enrichment of the corresponding GO term or KEGG pathway.

Protein expression related data were analyzed using hierarchical cluster analysis. For hierarchical cluster analysis, clusters were selected and hot-plots were drafted using Cluster3.0 software (http://bonsai.hgc.jp/∼mdehoon/software/cluster/software.htm↗), and similarity definition was calculated using Java Treeview (http://jtreeview.sourceforge.net↗) with Euclidean distance and average linkage hierarchical clustering.

The results indicated that a total of 7477 corresponding proteins (Supplementary Table S1) and fold change of differentially expressed proteins among C, W and M groups were determined based on the intergroup ratio in iTRAQ reports. Proteins that conformed to the following screening criteria were deemed as differentially expressed: expression fold change > 1.2 (up- or down-regulation) and P-value<0.05, while the protein difference multiple <0.8 and P<0.05 are the difference significant proteins. Comparisons between M/C, W/C and W/M groups were carried out, with altogether 81 (including 36 up- and 45 down-regulated), 116 (72 up- and 44 down-regulated) and 45 (28 up- and 17 down-regulated) differentially expressed proteins identified, respectively (Supplementary Table S2).

Optimized screening criteria, for those proteins where in differences could be seen between W/M and W/C groups, with no difference being seen M/C groups, or those differences showing opposite variation trend. The identified proteins were further screened to yield a total of 45 optimized differentially expressed proteins (Table 1).

First, cluster analysis was conducted (Figure 2). Hierarchical clustering results were expressed as a tree heat map, with red representing up-regulation and green indicating down-regulation. As determined through horizontal comparison, samples could be classified into three categories: C1-C3, M1-M3and W1-W3. Such vertical comparison indicated that proteins could be classified into two categories with opposite directional variation, which displayed the expression patterns of differentially expressed proteins in three groups, demonstrating the rationality of the selected differentially expressed proteins. The cluster analysis process thus supported that the differentially expressed proteins screened out in this experiment were reasonable and accurate.

As shown in Figure 3A, in the model group, the proteins with differential levels, as compared with the control groups, were enriched in the biological processes of growth regulation, signal transduction and hormone regulation. Molecular function enrichment was observed in molecular signal transduction, transcription factor activation, protein binding and catalytic reactions. The cellular compositions of the altered proteins were mainly membrane supramolecular complexes and intermembrane macromolecular complexes, which participate in the cell division and intercellular connection (Supplementary Table S3).

As shown in Figure 3B, in the warm acupuncture-treated group, when compared with the control groups, the proteins with altered levels were enriched in the GO biological processes of growth regulation, metabolism process, movement regulation and behavioral regulation. GO molecular function enrichment of differential level proteins was observed in catalytic activity activation, signal transduction activation, antioxidant activity activation, transcription factor activation, protein binding and molecular function regulation. The GO cellular composition of the altered proteins was enriched in intercellular connection of intermembrane, neural synapse and extracellular domains, and the synaptic regulation (Supplementary Table S4).

As shown in Figure 3C, in the warm acupuncture treated group, when compared with the model groups, the proteins with altered levels were enriched in the GO biological processes of signal transduction, regeneration, regulation, growth, immunological reaction and metabolism process; molecular function GO enrichment was observed in nerve conduction electrical carrier activation, antioxidant activity activation, catalytic activation and intermembrane transduction activation. The GO cellular composition of the altered proteins was enriched in cell membrane, synapse, intermembrane tissue, supramolecular complex and intermembrane space, participation in cell junctions (Supplementary Table S5).

As shown in Figure 4A, in the model group, proteins with differential levels, as compared with the control groups, were enriched in the KEGG pathways of tight junctions, atherosclerosis, endocytosis, choline metabolism in cancer, inflammatory mediator regulation of TRP channels, insulin resistance, insulin signal transduction, ribosome, herpes simplex infection, hippocampal signaling, RNA transportation, glycerol phospholipid metabolism, glycerolipid metabolism, complement and coagulation cascades, protein processing in endoplasmic reticulum, synaptic vesicle recycling, chemical carcinogenesis, influenza A and MAPK signaling (Supplementary Table S6).

As shown in Figure 4B, in the warm acupuncture-treated group, when compared with the control groups, proteins with differential levels were enriched in the KEGG pathways of ribosome, RNA transportation, chemical carcinogenesis, protein processing in endoplasmic reticulum, 5-hydroxytryptamine synapse, adhering junctions, inflammatory mediator regulation of TRP channels, platinum resistance, influenza A, Hippo signaling, glycolysis, apelin signaling, mTOR signaling, olfactory transduction, of vascular smooth muscle contraction, insulin resistance, protein binding, Rapl signaling and central carbon metabolism in cancer (Supplementary Table S7).

As shown in Figure 4C, in the warm acupuncture-treated group, when compared with the model groups, the proteins with differential levels were enriched in the KEGG pathways of cardiac muscle contraction, endocytosis, platelet activation, actin cytoskeleton regulation, complement and coagulation cascades, apelin signaling, oxytocin signaling, amino acid biosynthesis, autophagy, axon singling, Alzheimer disease, non-alcoholic fatty liver disease (NAFLD), cytokine–cytokine receptor interaction, spliceosome, chemokine signaling, TNF-α signaling, neuroactive ligand–receptor interaction, phagocytosis, oxidative phosphorylation and autoimmune thyroid disease (Supplementary Table S8).

As shown in Figure 5, compared with the protein expression change between the three groups, the warm acupuncture-treated group protein expression verge on control group suggests that warm acupuncture treatment is efficacious in improving sleep by regulating the protein expression process in an experimental rat model.

In our study, integration of the three groups of clusters analysis, revealed GO and KEGG pathway enrichment for four nerve–related proteins with differential levels in the warm acupuncture treated group, as compared with the model groups. These proteins were Prolargin (PRELP), NMDA receptor synaptonuclear-signaling and neuronal migration factor (NSMF), Transmembrane protein 41B (TMEM41B) and Microtubule-associated protein 1B (MAP1B) (shown in Table 2). These results provided new clues and theoretical basis for the further studies on the treatment mechanism of Mongolian medical warm acupuncture. Moreover, these four proteins were found to be closely related to neuroregulation or insomnia treatment. Among them, prolactin protein prolactin (prelp), which has been shown to be closely related to insomnia treatment, is the focus of our future research.