Pearl powder reduces sleep disturbance stress response through regulating proteomics in a rat model of sleep deprivation

Abstract Aims This study aimed to explore whether pearl could help prevent cognitional morbidity and improve the metabolic processes of hippocampus. Methods Rats were divided into group of control (CTL), sleep deprivation (SD) and pearl powder (PP). The sleeplessness was introduced to all rats except control. Before and after administration with vehicle or pearl powder, cognition was evaluated by Morris water maze (MWM). The protein expression in hippocampus among all groups was examined using iTRAQ‐based global proteomic analysis. Results Morris water maze tests revealed improvements of insomnia‐induced cognitive deficit in both PP‐ and ES‐treated rats, as compared to SD rats. However, proteomic analysis indicates that the pharmacological impact on gene expression of these two medicines is quite different: pearl is more capable of correcting aberrant gene expression caused by SD than estazolam. Therefore, pearl is more suitable for treatment of insomnia. These data, together with protein‐protein interaction analysis, indicate that several pathways, affected by sleep deprivation, may be rescued by pearl powder: retrograde endocannabinoid signalling pathway, and the protein interaction or network enrich in oxidative phosphorylation Parkinson's disease and Huntington disease, etc Conclusions Sleep deprivation can mimic cognition decline caused by insomnia with altered protein expression in the hippocampus; such behavioural and pathological changes can be significantly ameliorated by pearl powder.


| INTRODUC TI ON
Pearl, in addition to improving skin beauty, is a well-known traditional Chinese medicine (TCM) for various disorders, including insomnia, palpitations, convulsions or epilepsy. According to Compendium of Materia Medica, a famous historical Chinese Herb Pharmacology, patients in stressful status can take pearl powder as tranquillizer and sedative to protect and nourish YIN by calming endogenous WIND and removing toxic substances. 1,2 Pearl is mainly composed of calcium carbonate and magnesium carbonate, which accounted for 91% of total weight. Other inorganic molecules such as silica, calcium phosphate, aluminium oxide and ferric oxide, as well as some trace elements such as sodium, magnesium, manganese, selenium, aluminium and copper can also be found in pearl powder. It also contains essential amino acids including lysine (Lys), valine (Val), threonine (Thr), methionine (Met), leucine (Leu), phenylalanine (Phe), tryptophan (Trp) and histidine (His). 3,4 The nutrients in pearl powder are enriched with proteins, peptides and amino acids, making great contributions to the body's bioactivity including improving antioxidant defence system. 5 For example, amino acid Asp and Glu possess antioxidant properties, and Cys exhibits free radical quenching ability. 6 Indeed, there are different usages of pearl when it is added to different combination of Chinese herbs which can selectively targeting different part of body sites to balance YIN. 7 Estazolam, a triazolobenzodiazepine, has been widely used as hypnotic medication. 8,9 It belongs to class of benzodiazepines, a GABAa receptor agonist in the brain. Estazolam can cause sedation and relaxation, therefore facilitating sleep. However, estazolam is extremely addictive with very severe withdrawal symptoms and can cause a similar black-out effect, and some studies reported that it also confers the risk of dementia. 10,11 Therefore, pearl and nacre are valuable traditional medicines with a lot of potential for more clinical treatments, if we understand their mechanism of action and signalling pathway.
In this study, we have been suggested that pearl contains many bio-protective reagents that target several molecules and play critical roles in multiple metabolic pathways. To test this hypothesis, we used a previously validated insomnia model in rats, induced by sleep deprivation (SD), resulting in learning and memory impairment which had been thought to be caused by pathological changes in hippocampus. 12,13 Based on it, we conducted an advance technology iTRAQ for a broad screening and comparing on differential proteomic expression in hippocampus. Our results indicate that sleep deprivation can induce learning and memory deficit which associated with oxidative phosphorylation, ribosome and proteasome proteins. Pearl is able to lowering or reversing some, if not all of the mal-alternation.

| Animals and ethics statement
The experimental protocols, care and handling of animals used in this study were approved by the Institutional Animal Care and Use Six-week-old rats received standard laboratory rat chow and tap water ad libitum feeding. After a week of adaptation, rats were randomly assigned into four groups (n = 9/group) as control (CTL), sleep deprivation (SD), pearl powder (PP) and estazolam (ES). The sleep deprivation was conducted for about 20 hours/d (23:00-19:00) and lasted for total 7 days: rats were placed in a platform (height 20 cm and 5 × 5 square centimetre area) which was in a container of water (depth 18 cm). The Ctr, PP and ES groups were administrated repeatedly with 0.9% saline, pearl powder (2.5 mg/kg/d) and estazolam (0.13 mg/kg/d) starting from 1st day of sleep deprivation once per day for 14 days. The schedule of administration is illustrated in Figure 1.

| Morris Water Maze (MWM) test
Before behavioural tests, rats were adapted in our animal facility for at least 1 week in the housing room which has the same condition as behavioural room. 14 MWM training and test were performed in a circular water tank (120 cm in diameter) containing opaque water (22 ± 1°C) at a depth of 25 cm and were divided into four quadrants. A hidden escape platform (9 cm in diameter) was placed in the centre of one quadrant, with its surface 1 cm below the water. The rats were subjected to an acquisition trial four times a day for five consecutive days. During each trial, the rats were placed in water at one of the four quadrants and the starting position was randomly selected. Each rat was trained to swim for locating the hidden platform. Rats that failed to find the hidden platform within 60 seconds were placed on it for 30 seconds. The same platform location was used for all rats. The platform was removed on the sixth day, and the rats were subjected to the spatial probe trial test for 60 seconds.
The time and distance spent in the target quadrant were recorded.

| Protein extraction and trypsin digestion
Rats were killed in a box connected to CO 2 tank. Then, rat brains were quickly removed from skulls and left on cold platform. Two pieces of hippocampal tissues underneath cortex were carefully dis-

| Peptide isobaric labelling and HPLC fractionation
After trypsin digestion, peptide was desalted by Strata-X SPE column and vacuum-dried. Peptide was reconstituted in 20 μL 500 mmol/L TEAB and processed according to the manufacturer's protocol for 8-plex iTRAQ kit. Briefly, one unit of iTRAQ reagent was all added to peptide solution after thawed and dissolved in 50 μL isopropanol.

| Proteomic data analysis and bioinformatics
The resulting MS/MS raw data were searched against the Rattus norvegicus proteome database (taxon identifier: 10 116 including 29 975 protein sequences) downloaded from UniProt database using Sequest software integration in Proteome Discoverer (version 1.3, Thermo Scientific). Trypsin was chosen as enzyme, and two missed cleavages were allowed. Carbamidomethylation (C) was set as a fixed modification and oxidation (M), and acetylation in N-term was F I G U R E 1 Workflow. Six-week-old rats were conducted MWM test from day 3 to day 7 with the hidden platform in the water tank, and then, spatial memory and cognition were examined by MWM at day 8 without the hidden platform. From day 9 to day 16, all rats were administrated with or without sleep deprivation, as well as treatment, respectively, with PP/ES or saline from day 9 to day 23. Pearl powder (2.5 mg/kg/d) and estazolam (0.13 mg/kg/d) in saline, the same volume of saline was administered in groups of CTL and SD. Afterwards, the MWM training (days [18][19][20][21][22] and test (day 23) were performed for another 6 d again. Then, the rats were killed and hippocampus tissues were collected for proteomic and other analyses set as variable modification. The searches were performed using a peptide mass tolerance of 20 ppm and a product ion tolerance of 0.05 D, resulting in 1% false discovery rate (FDR). The GO (gene ontology) was used for the annotation of the identified proteins, which was composed of BP (biological processes), CC (cellular components) and MF (molecular functions), and the analysis was based on the UniProt-GOA database (http://www.ebi.ac.uk/GOA/). The pathway analysis of differently expressed proteins was based on the KEGG (Encyclopedia of Genes and Genomes) database. The online KEGG Automatic Annotation Server (KAAS) service tools were used to annotate the KEGG database description for each protein and to map the annotation results using other KEGG online service tools and the KEGG mapper. The interaction between differential expression protein (DEPs) groups was derived from the Search Tools for the Retrieval of Interacting Genes/Proteins (STRING) database.
A high confidence (0.7) for the required interaction score and the active interaction sources, including text mining, experiments and databases, were chosen to draw the protein-protein interaction map using Cytoscape 3.2.1.

| Western blotting
The protein expression levels of PLK1, PGP and HGS were analysed

| Statistical analysis
One-way ANOVA and Bonferroni post hoc test were used in behavioural tests (Figure 2) and Western blot quantification ( Figure 5).
Data were expressed as mean ± SEM P < .05 is considered as statistically significant difference between groups. To analysis the data of proteomics, paired comparisons of mean number between two groups were performed using t test, and the proteins with fold change (ratio) ≥1.2 and those with fold change (ratio) ≤0.83 were considered difference (DEPs) with statistically significant when P < .05.

| Behavioural tests
To evaluate the effect of pearl powder on sleeplessness stress, the spatial learning and memory ability were tested using MWM test, which primarily depends on the hippocampus function. 15,16 Prior to any treatment, all rats were firstly trained for swimming and the measurements were accessed as shown in Figure 2A

| Proteomic quality and differential quantitative analysis
In screening for protein(s) expressed differently from SD vs to other groups in proteomic study, there were total 3745 proteins identified, and 69% (2592 proteins) were quantified with high correlation in biological replicas, and six differential proteins identified from Venn analysis ( Figure 3A Figure 3D). And 22 differential expressing proteins were identified (Table 1). These proteins had different expressing trend after sleep deprivation treatment but rescued towards back to CTL expression direction after pearl was added to these rats.

| Functional, pathway and protein domain enrichment analysis
To obtain a functional overview of differentially expressed protein, gene ontology (GO) annotation analysis was applied. Proteins were functionally classified based on UniProt GO annotation and demonstrated through heat map ( Figure 4A). Although as many as 353 differentially expressed proteins were identified, their dynamic changes can be assigned into 5 main patterns ( Figure 4B).

| Western blot
To validate the expression data from proteomic analyses, five proteins related to sleep deprivation were selected for Western blot- were decreased and 2 (SOD1 and Ppp2cb) were increased in SD as compared to other groups. Western blotting assay was consistent with the results of the mass spectrometry analysis. F I G U R E 2 MWM test result comparison before and after modelling (data expressed as mean ± SEM, p < .05 represents statistical significant). A, Escape latency is the time (s) in which the rat swims from the starting point to the hidden platform point at the test day (without the hidden platform). The rats in assigned groups (CTR, SD, PP and ES) were tested before (day 8) and after (day 23) modelling (sleep deprivation + treatment). Escape latency in SD was increased significantly from 9.2 ± 1.8 (before modelling) to 19.4 ± 2.6 (after modelling), whereas the escape latencies in the PP and ES groups show no difference before and after modelling. Therefore, sleep deprivation injured the rats' capability of finding the platform; pearl powder and estazolam reverse the injury. B, Distance ratio: ratio of swimming distance (m) in target quadrant (the hidden platform stood in at the training days) versus total swimming distance in water pool within 60 s is used to reflect the animal memory. The distance ratio in the SD group was significantly decreased from 60.0 ± 5.0 (before SD) to 35.5 ± 5.6 (after SD), whereas the distance ratio of rats in other groups was not significantly changed before and after modelling. Therefore, SD rats were injured in memory, but PP and ES were capable to prevent memory injury. C, Number of entries into the quadrant with withdrawn platform within 60 s as measurement of spatial memory. Before modelling, the rats in different group show the similar entry times: that is, 2.2 ± 0.28 (CTL), 2.1 ± 0.20 (SD), 2.2 ± 0.33 (PP) and 2.0 ± 0.33 (ES). After modelling, the travel times in SD were 1.4 ± 0.27, decreased by 31.8% (*P < .01). However, the travel times in other groups were not changed or slightly increased after modelling. This indicates the rats treated with PP or ES ameliorate spatial memory deficit induced by SD

| Pathway-protein crosslinking network construction
After GO analysis was performed, we would like to explore interaction relationship formed by these differentially expressed proteins.
Cytoscape software v2.8.3 (http://www.cytos cape.org) was used to visualize the pathway-protein crosslinking network ( Figure 6). In this network, protein Ppp2cb has most edges which means it has most interaction relationship. Oxidative phosphorylation Parkinson's disease pathway, ribosome and ribonucleoprotein pathway were the two most enriched pathway in this interaction network.

| D ISCUSS I ON
In this study, we used a previously validated rat insomnia model, Hierarchical clustering showed that the top 22 temporal proteins and genes that could be divided into four groups exhibited different expression patterns. From the four patterns, we generalized from differentially expressed proteins, the changing trend in the PP group and the ES group was not always the same. This means the pharmacological effect of these two medicines is not the same. Our results indicate that PP shows higher rescue capability, and ES has bigger side effect. Figure 4 shows that pearl powder displays specific rescue effect in pattern 1 and pattern 4. In pattern 1, the pathways most sensitive to PP treatment are retrograde endocannabinoid signalling 17,18 and autophagy; in pattern 4, the most sensitive pathways to PP treatment are the proteins enriched relative to Huntington disease and Parkinson disease (pattern 4).
Our experimental results have human correlates with studies that support the application of pearl powder to stressful behaviour disorder. 19 There is substantial literature correlating depression symptoms and psychosocial dysfunction in patients with insomnia. [20][21][22][23] More direct evidence also comes from a randomized   24 It has also been suggested that humans who experience an adverse life event have a hyper-responsive HPA and that HPA reactivity may in turn modulate memory and learning impairment. 25 It is known that chronic insomnia patients show abnormal substructure, 26 which may be caused by decreased neurogenesis. 27 Endocannabinoid signalling (ECS) is altered in preclinical and clinical models of depression, with one of the common symptoms as insomnia. Several studies indicating that hippocampal progenitor F I G U R E 5 Validation of five differentially expressed proteins in CTL and treatment groups. A, RIMS3, Ppp1r14a, MGR3, SOD1, Ppp2cb and β-actin bands, respectively, expressed from rat groups of CTL, SD, PP and estazolam. B, The relative density of the validated proteins was normalized with β-actin as the internal reference. One-way ANOVA, n = 3, *P < .05 F I G U R E 6 Protein-protein interaction between DEPs. Proteins' expression ratio of PP/SD was marked with continuous colour mapper. Proteins enriched in oxidative phosphorylation Parkinson's disease and ribosome and ribonucleoprotein were marked in grey cycling proliferation and neurogenesis require intact ECS. 28,29 Therefore, ECS pathway deficit is one of the underline mechanisms for abnormal hippocampal structure and activity in insomnia patients, which can be rescued by pearl treatment.
Further studies will be required to understand the mechanisms regarding the regulation of hippocampus such as impaired feedback signalling from epigenetic changes regulating RIMS3, Ppp1r14a, MGR3, SOD1 and Ppp2cb or other forms of expression. Such as the impact of pearl treatment on endocannabinoid signalling pathway in hippocampus of SD rats. What initiates this cascade of events during stress is also unknown. Insomnia may alter the HPA axis through activation of cytokines, which is another biological targeted that may be rescued by pearl powder.

| CON CLUS IONS
In conclusion, our results demonstrate, for the first time, that sleep deprivation causes hippocampus injury through several pathways including endocannabinoid signalling, autophagy, Huntington disease, oxidative phosphorylation Parkinson disease and ribosome and proteasome, which can be corrected (at least partially) by pear powder.
Clinically, the cause of insomnia is very complicated, and pearl powder had been used in combination with other Chinese medicine as treatments for insomnia, with unclear functionality. Interestingly, the identified pathways had been implicated in the pathology of anxiety and depression. Therefore, we speculate that pearl powder by itself may be useful for anxiety-or depression-induced insomnia. Further studies can be focused on dissecting effect of pearl powder on each biological pathway, in the context of insomnia and cognitive decline.

CO N FLI C T O F I NTE R E S T
All authors declare no conflict of interests.

AUTH O R CO NTR I B UTI O N S
Jiang Lin designed the study. Meng Xia, Delun Huang and Yuangming Tong performed experiments. Meng Xia analysed the data. Delun Huang and Yuangming Tong carried out critical revision of the manuscript for important intellectual content.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data used to support the findings of this study are available from the corresponding author upon request.