Use of reagents for detecting metabolic markers in the preparation of sleep disorder screening or diagnosis products

Abstract

The present invention provides the use of a reagent for detecting metabolite markers in the preparation of a sleep disorder screening or diagnostic product, and relates to the field of biomedicine. The metabolite marker is selected from one or more of triglyceride metabolites, phosphatidylcholine metabolites, cholesterol lipid metabolites, phosphatidylethanolamine metabolites, ceramide metabolites, amino acids, polypeptides and their analogs, benzoic acid and its derivatives, fatty acid metabolites, fatty acid ester metabolites, steroid and derivative metabolites, and other types of metabolites. The present invention discloses the risk metabolites of sleep disorders and studies the relationship between these risk metabolites and sleep disorder comorbidities. These risk metabolites can be used as markers for screening or diagnosing sleep disorders, and are used for screening or diagnosing sleep disorders, especially for the screening or diagnosis of sleep disorders affected by different comorbidities, and have good application prospects.

Description

Technical Field

[0001] The present invention relates to the field of biomedicine, and in particular to the use of reagents for detecting metabolic markers in the preparation of sleep disorder screening or diagnosis products. Background Art

[0002] Sleep is very important to humans. About one-third of a person's life is spent sleeping. Sleep is closely related to human health. First of all, lack of sleep and sleep disorders may be caused by certain diseases, reflecting certain health problems in the human body; at the same time, lack of sleep and sleep disorders will also have a significant impact on people's cognitive functions and emotions, easily causing inattention, weakened alertness, memory loss, slow movement, slow reaction, visual and auditory hallucinations, irritability and mania and other anxiety symptoms, and even lead to more serious diseases. Surveys show that many people suffer from sleep disorders or sleep-related diseases. The proportion of adults with sleep disorders is as high as 30%, and it is more common among the elderly.

[0003] At present, the most important method for diagnosing sleep disorders is to use an EEG multi-lead recording device to monitor the whole night's sleep process; at the same time, various scales are used to assess sleep quality. However, sleep monitoring is expensive and time-consuming, and the use of various scales to assess sleep quality is affected by the patient's subjective influence, and it is impossible to quickly and accurately diagnose sleep disorders; it is even more impossible to accurately and effectively screen and prevent people with a high risk of sleep disorders.

[0004] In addition, sleep disorders in the elderly may be due to an increase in age-related comorbidities and medication use, as well as changes in sleep structure. Although the close connection between sleep disorders, related comorbidities, and aging is well known, the connection between them has not yet been explained at the molecular level. As a result, the true cause of sleep disorders cannot be correctly screened or diagnosed, and thus sleep disorders cannot be prevented or treated in a targeted manner.

[0005] Patent WO2016044338A3 discloses a method and system for diagnosing sleep disorders. The method discloses a variety of biomarkers that can be used to diagnose sleep disorders. Although these biomarkers can be used to diagnose sleep disorders, they cannot further diagnose the type of sleep disorder, determine the comorbidity of sleep disorders, and find the cause of sleep disorders. Finding a marker that can determine the comorbidity of sleep disorders and determine the cause of sleep disorders for screening or diagnosing sleep disorders is of great significance for the targeted prevention or treatment of sleep disorders. Summary of the invention

[0006] In order to solve the above problems, the present invention provides the use of a reagent for detecting metabolite markers in the preparation of a sleep disorder screening or diagnosis product. The purpose of the present invention is to provide a metabolite marker related to sleep disorders, and to determine whether a patient suffers from a sleep disorder or is at risk of suffering from a sleep disorder, and what causes the sleep disorder, by detecting the level of the metabolite marker, thereby providing a new means for early screening and diagnosis of sleep disorders.

[0007] The present invention provides the use of a reagent for detecting metabolic markers in the preparation of a sleep disorder screening or diagnosis product;

[0008] The metabolite marker is selected from one or more of triglyceride metabolites, phosphatidylcholine metabolites, cholesterol lipid metabolites, phosphatidylethanolamine metabolites, ceramide metabolites, amino acids, polypeptides and their analogs, benzoic acid and its derivatives, fatty acid metabolites, fatty acid ester metabolites, steroid and derivative metabolites, and other types of metabolites.

[0009] Further,

[0010] The triglyceride metabolites are selected from one or more of TG (53:2), TG (52:2), TG (52:7), TG (54:2), and TG (50:3);

[0011] and / or, the phosphatidylcholine metabolites are selected from one or more of PC(37:2), PC(33:2), PC(44:6), PC(35:2), PC(36:2), PC(38:7), PC(34:3), PC(36:4), PC(34:2);

[0012] and / or, the cholesterol lipid metabolites are selected from one or more of ChE (22:6), ChE (20:4), and ChE (18:3);

[0013] and / or, the phosphatidylethanolamine metabolites are selected from one or more of PE(42:10), PE(40:5), PE(38:4), PE(40:4), PE(40:7), PE(36:4), PE(38:8), and PE(38:5);

[0014] And / or, the ceramide metabolites are selected from one or more of Cer(42:0), Cer(40:1), and Cer(42:1);

[0015] and / or, the amino acid, polypeptide and analogs thereof are selected from one or more of L-serine, glutamylglycine, creatine, diaminocaproate, 2-phenylglycine, L-aspartic acid, guanidinoacetic acid, 3-hydroxy-N6,N6,N6-trimethyl-L-lysine;

[0016] And / or, the benzoic acid and its derivatives are selected from one or more of salicylic acid, hexylresorcinol, adenylate succinate, and hydroxybenzoic acid;

[0017] And / or, the fatty acid metabolites are selected from one or more of undecanoic acid and tridecanoic acid;

[0018] And / or, the fatty acid ester metabolites are selected from one or more of stearoylcarnitine, palmitoylcarnitine, linoleylcarnitine, and hydroxybutyrylcarnitine;

[0019] and / or, the steroid and derivative metabolites are selected from one or more of calcidiol and pregnenolone;

[0020] And / or, the other types of metabolites are selected from one or more of L-kynurenine, LysoPE (16:0), 5-hydroxyindoleacetic acid, and sphingosine.

[0021] Further,

[0022] One or more of the metabolite markers ChE (20:4), ChE (22:6), PC (44:6), PE (40:4), TG (54:2), stearoylcarnitine, palmitoylcarnitine, 2-phenylglycine, and hydroxybutylcarnitine are characteristic metabolite markers for screening or diagnosing sleep disorders;

[0023] And / or, one or more of the metabolite markers PE (42:10), L-kynurenine, 5-hydroxyindoleacetic acid, and pregnenolone are metabolite markers used for screening or diagnosing the occurrence of aging-related sleep disorders;

[0024] and / or, one or more of the metabolite markers ChE (18:3), PC (33:2), PC (34:2), PC (35:2), PC (36:2), PC (37:2), TG (52:7) and hexylresorcinol are metabolite markers for screening or diagnosing the occurrence of sleep disorders associated with cognitive dysfunction;

[0025] And / or, one or more of the metabolite markers PC (38:7), PE (40:5), 3-hydroxy-N6, N6, N6-trimethyl-L-lysine are metabolite markers used to screen or diagnose sleep disorders caused by hypertension.

[0026] Furthermore, the content of the metabolite marker is up-regulated or down-regulated in the human body.

[0027] Further, the reagent is an enzyme-linked immunosorbent assay reagent;

[0028] And / or, the reagent is a western blot reagent;

[0029] And / or, the reagent is a reagent for a protein chip detection method.

[0030] Furthermore, the reagent detects metabolite levels by one or more of the following methods: chromatography, spectroscopy, mass spectrometry, chemical analysis.

[0031] Furthermore, the sample detected by the reagent is human peripheral blood.

[0032] Furthermore, the sample detected by the reagent is plasma.

[0033] Further,

[0034] The reagent for detecting one or more of the metabolic markers ChE (20:4), ChE (22:6), PC (44:6), PE (40:4), TG (54:2), stearoylcarnitine, palmitoylcarnitine, 2-phenylglycine, and hydroxybutylcarnitine is used to prepare a product for characteristic screening or diagnosis of sleep disorders;

[0035] And / or, the reagent for detecting one or more of the metabolic markers PE (42:10), L-kynurenine, 5-hydroxyindoleacetic acid, and pregnenolone is used to prepare a product for screening or diagnosing aging-related sleep disorders;

[0036] and / or, the reagent for detecting one or more of the metabolic markers ChE (18:3), PC (33:2), PC (34:2), PC (35:2), PC (36:2), PC (37:2), TG (52:7) and hexylresorcinol is used to prepare a product for screening or diagnosing sleep disorders associated with cognitive dysfunction;

[0037] And / or, the reagent for detecting one or more of the metabolic markers PC (38:7), PE (40:5), 3-hydroxy-N6, N6, N6-trimethyl-L-lysine is used to prepare a product for screening or diagnosing sleep disorders caused by hypertension.

[0038] The present invention discloses risk metabolites of sleep disorders and studies the relationship between these risk metabolites and sleep disorder comorbidities. These risk metabolites can be used as markers for screening or diagnosing sleep disorders, and are used for screening or diagnosing sleep disorders, especially for screening or diagnosing sleep disorders affected by different comorbidities, and have good application prospects.

[0039] Obviously, according to the above contents of the present invention, in accordance with common technical knowledge and customary means in the art, without departing from the above basic technical ideas of the present invention, other various forms of modification, replacement or change may be made.

[0040] The above contents of the present invention are further described in detail below through specific implementation methods in the form of embodiments. However, this should not be understood as the scope of the above subject matter of the present invention being limited to the following examples. All technologies realized based on the above contents of the present invention belong to the scope of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 Results of screening for risk metabolites for sleep disorders: a is the distribution of PSQI scores of 500 people randomly selected from the Western China Health and Aging Trends Study Cohort (WCHAT) for metabolomics testing; b is the result of drawing a principal component diagram using metabolomics data; c is the result of drawing a principal component diagram using lipidomics data; d is 52 metabolites significantly associated with sleep quality after logistic regression analysis, with 28 hydrophobic lipid metabolites on the left and 24 hydrophilic metabolites on the right. The metabolites are sorted from top to bottom in the figure according to the crude odds ratio (OR), and the width of each metabolite represents the 95% confidence interval. TG denotes triglyceride, LysoPE denotes lysophosphatidylethanolamine, ChE denotes cholesterol ester, Cer denotes ceramide, PC denotes phosphatidylcholine, PE denotes phosphatidylethanolamine; 5-HIAA denotes 5-hydroxyindoleacetic acid; HTMLA denotes 3-hydroxy-N6, N6, N6-trimethyl-L-lysine.

[0042] Figure 2 Figure 3 Correlation results between risk metabolites and PSQI components. The network diagram shows the correlation between risk metabolites of sleep disorders and the six components of PSQI, including subjective sleep quality, sleep duration, sleep latency, habitual sleep efficiency, daytime dysfunction and sleep disturbance. The Spearman correlation coefficient obtained by the WGCNA package of R software is expressed as the connecting line (edge) between PSQI components and metabolite nodes (correlation coefficient >0.1 or <0.1, algorithm is AS 89, p <0.05). The edge width is proportional to the correlation strength. The node size is proportional to the number of connections with other nodes. Red segments indicate positive correlation, and green segments indicate negative correlation. The node color indicates the type of metabolite and the component of PSQI. The graph was generated in Gephi0.9.2. ChE is cholesterol ester; PC is phosphatidylcholine; PE is phosphatidylethanolamine; LysoPE is lysophosphatidylethanolamine; TG is triglyceride; 5-HIAA is 5-hydroxyindoleacetic acid.

[0043] Figure 3To identify the prediction model for sleep disorders: a is a visualization of the coefficient of each penalty; b and c are the areas under the curve (AUC) of the training set (b) and the test set (c); a prediction model was established by lasso regression analysis of 52 sleep disorder risk metabolites; d and e are principal component analysis plots of the external validation set (34 people), after removing two outliers outside the 95% confidence interval; f is the AUC of the external validation set.

[0044] Figure 4 The impact of diseases and symptoms on sleep quality assessment: a is that all participants were divided into four groups according to questionnaire assessment and metabolic model prediction: the first group was assessed as good sleep quality by PSQI, and metabolomics predicted good sleep quality; the second group was assessed as poor sleep quality by PSQI, and metabolomics predicted good sleep quality; the third group was assessed as good sleep quality by PSQI, and metabolomics predicted poor sleep quality; the fourth group was assessed as poor sleep quality by PSQI, and metabolomics predicted poor sleep quality; b is the difference in age and blood routine indicators among the four groups; all y-axis values ​​were scaled to the range of 0 to 1 by "min-max" standardization; c is the difference in the number of chronic diseases among the four groups; d and e are the differences in the epidemiology of hypertension (d) and osteoarthritis (e) among the metabolomics prediction groups. f is the difference in the epidemiology of participants with cognitive dysfunction among the metabolomics prediction groups, and cognitive dysfunction was assessed by a 10-point Short Portable Mental Status Questionnaire (SPMSQ) combined with cultural level; g is the difference in clinical symptoms among the four groups. The p values ​​for blood biochemical tests, diseases, and clinical symptoms have been adjusted for age and sex, * represents p < 0.05, ** represents p < 0.01, and *** represents p < 0.001.

[0045] Figure 5 Figure 3 Analysis of the difference in intensity of risk metabolites between group I and group IV: a The box plot shows the difference in intensity of risk metabolites between group I (PSQI assessment was good, metabolomics prediction was good) and group IV (PSQI assessment was poor, metabolomics prediction was poor); b is the intensity of HTMLA in groups I-IV, HTMLA is 3-hydroxy-N6,N6,N6-trimethyl-L-lysine.

[0046] Figure 6Common risk metabolites between sleep disorders, aging and related comorbidities: Correlation of 52 sleep disorder-related risk metabolites with age and comorbidities; Correlation analysis between metabolites and age adjusted for gender, and linear regression and logistic regression were used for analysis, in which the median age was 69 years old in logistic regression; Correlation of 52 sleep-related risk metabolites with chronic diseases, cognitive disorders, anxiety, depression and anxiety-depressive disorders were adjusted for age and gender; SD is sleep disorder, CI is cognitive impairment, HP is hypertension, DIA is diabetes, CHD is coronary heart disease, STR is stroke, COPD is chronic obstructive pulmonary disease, ANX is anxiety, DEP is depression, ANX-DEP is anxiety-depressive disorder, GD is gastrointestinal disease, LD is liver disease, KD is kidney disease, OST is osteoarthritis, and the number of participants used for correlation analysis is shown in brackets.

[0047] Figure 7 Figure 2: Changes in sleep risk metabolites at different ages and in sleep-related diseases. a is a box plot showing the difference in the intensity of risk metabolites between participants under 69 years old and over 69 years old. bl are box plots showing the difference in the intensity of risk metabolites between participants with cognitive impairment, hypertension, diabetes, stroke, chronic obstructive pulmonary disease, depression, anxiety and depression, gastrointestinal diseases, liver diseases, kidney diseases and osteoarthritis and those without the above diseases. DETAILED DESCRIPTION

[0048] The raw materials and equipment used in the specific embodiments of the present invention are all known products and are obtained by purchasing commercially available products.

[0049] Example 1. Determination of biomarkers for diagnosing sleep disorders

[0050] 1. Methods

[0051] (1) Helsinki Declaration of Ethical Principles

[0052] The study samples were from the baseline of the Western China Health and Aging Trends (WCHAT) study, an ongoing prospective cohort study designed to assess health status and its influencing factors in western China. The study was registered with the Chinese Clinical Trial Registry (ChiCTR1800018895) and approved by the Ethics Committee of West China Hospital, Sichuan University (No. 2017-445), and started in 2018. In accordance with the Declaration of Helsinki Ethical Principles, all participants signed written informed consent to participate in the trial. Subject information was collected through face-to-face interviews. The Pittsburgh Sleep Quality Index (PSQI) includes 7 aspects (subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbance, sleep medication use, and daytime dysfunction) to assess sleep quality, with a total score of 21 points. In the study, PSQI>5.0 was defined as poor sleep quality. Higher rates of comorbidities were observed among participants with sleep disturbances, including coronary heart disease, hypertension, chronic obstructive pulmonary disease (COPD), diabetes, gastrointestinal disease, stroke, liver disease, kidney disease, osteoarthritis, cognitive impairment, depression, anxiety, and anxiety-depression.

[0053] (2) Blood sample collection

[0054] When the subjects arrived at the research center in the morning, three tubes of fasting blood were collected by trained nurses for routine blood tests on the same day, and the other two tubes of blood were centrifuged at 3500g for 15 minutes within 30 minutes after venipuncture, and the plasma was collected and stored at -80℃. The entire blood processing process was carried out under a strict standardized protocol.

[0055] (3) Non-targeted metabolomics analysis

[0056] Hydrophilic metabolites were extracted with methanol, and lipids were extracted with dichloromethane / methanol (v / v=2:1) ​​liquid-liquid extraction. 13 C6 L-Lysine Salt Powder (Silantes) and 13 C6 15N4 L-arginine salt powder (Silantes) was added to monitor the extraction efficiency of hydrophilic metabolites, and PE (16:0-d31-18:1) was added to monitor the extraction efficiency of lipids. Mass spectrometry metabolomics and lipidomics analysis was performed by the Center for Metabolomics and Lipidomics, Tsinghua University. BEH amide column (Waters, USA) and BEH C18 column (Waters, USA) were used for metabolomics analysis in positive and negative ion modes, respectively. CORTECS C18 column (Waters, USA) was used for lipidomics in positive mode. Mixed quality controls (QCs) were added for every 15-20 plasma samples injected. Polar metabolites were assigned using Tracker (Thermo, CA) based on an in-house database. Standard MS / MS spectra of more than 1500 metabolites were included in the database. Lipids were identified using Lipid Research (Thermo, CA) software. Only lipids with reliable MS / MS were used in the following statistical analysis.

[0057] (4) Data processing and statistical analysis

[0058] All retained metabolites were present in at least 80% of the discovery set samples, and the coefficient of variation (CV) of the quality control (QC) samples was less than 30%. The total relative intensity of the metabolites was normalized. The intensity values ​​were log2 transformed to reduce skewness and stabilize variance. Statistical analysis was performed using R or SPSS software version 26 (IBM Corporation, Chicago, IL, USA). Categorical variables were expressed as counts and percentages. Normally distributed continuous variables were explored using the Kolmogorov-Smilov test, and normally distributed variables were expressed as mean ± standard deviation. Non-normally distributed variables are expressed as median and interquartile range (IQR). Differences between two groups were analyzed using the Pearson chi-square test (categorical variables), the independent sample T test (normally distributed continuous variables), and the Manny U test (non-normally distributed continuous variables). Equal variance analysis and Kruskal-Wallis test were used to test differences between three or more groups. In the prediction model, 433 people were divided into training set and test set in the ratio of 8:2, and then cross-validation was performed, and hyperparameters were optimized for internal validation.

[0059] 2. Results

[0060] (1) Clinical characteristics of sleep disorders in the WCHAT cohort

[0061] 6887 WCHAT cohort participants contained complete Pittsburgh Sleep Quality Index (PSQI) information. Their ages ranged from 50 to 95 years old, with an average age of 62.43±8.27 years old. Statistical analysis showed that 47.2% of the 6887 participants had sleep disorders, with PSQI scores higher than 5.0, indicating that sleep disorders are very common in the elderly. In addition, participants with sleep disorders had a higher rate of comorbidities, including coronary artery disease, hypertension, chronic obstructive pulmonary disease, diabetes, gastrointestinal disease, stroke, liver disease, kidney disease, osteoarthritis, cognitive impairment, depression, anxiety, and anxiety-depression.

[0062] (2) Identification of risk metabolites for sleep disorders

[0063] In order to reveal the relationship between sleep disorders and metabolic disorders, 500 Han Chinese elderly people were randomly selected as the initial discovery set for metabolic analysis. The PSQI score distribution of these 500 people was consistent with that of the WCHAT cohort ( Figure 1 a), indicating that this initial discovery set is well representative. After removing 16 participants who were taking medication and 51 outliers in the principal component analysis (PCA) ( Figure 1 b-1c), metabolic data of 433 participants were used to screen metabolites and establish a risk prediction model for sleep disorders. Logistic regression analysis was used to determine metabolites significantly associated with PSQI scores, and the abundance of all metabolites was log2 transformed and Z-score normalized. As a result, by using the glm function in R, 52 metabolites (including 28 hydrophobic and 24 hydrophilic metabolites) were significantly associated with sleep quality (p<0.05) ( Figure 1 d). Among the hydrophobic metabolites, triglycerides (TGs) are risk factors for sleep disorders, while phosphatidylcholine (PC), cholesterol esters (ChEs), phosphatidylethanolamine (PEs), and ceramides (Cers) are protective factors for sleep disorders ( Figure 1 d left). For hydrophilic metabolites, some such as L-kynurenine, 5-hydroxyindoleacetic acid, calcitriol, and stearoylcarnitine increased the risk of sleep disorders, while others such as creatine, L-serine, and undecanoic acid protected participants from sleep disorders ( Figure 1 d right).

[0064] Next, we investigated how the 52 risk metabolites derived from logistic regression analysis affected sleep quality. The relationships between risk metabolites and each domain of PSQI except medication were analyzed separately by Spearman correlation analysis in the WGCNA software package (algorithm AS 89, p < 0.05, correlation coefficient > 0.1 or < -0.1) ( Figure 2). The study found that more risk metabolites were associated with subjective sleep quality, sleep delay (sleep latency) and daytime dysfunction. Specifically, TG (53:2), TG (52:7), PC (34:2), PC (44:6), PC (35:2), PC (36:4), PC (34:3), ChE (20:4), ChE (18:3), ChE (22:6), salicylic acid, LysoPE (16:0), L-kynurenine, 5-hydroxyindoleacetic acid and guanidinoacetic acid metabolites were associated with subjective sleep quality; TG (52:7), PC (36:2), PC (37:2), PC (44:6), PC (35:2), PC (34:2), PE (40:5), PE (36:4), PE (38:8), Cer (4 Metabolites such as Cer(42:0), LysoPE(16:0), salicylic acid, sphingosine, diaminocaproate, tridecanoic acid, and undecanoic acid affect sleep latency; metabolites TG(50:3), LysoPE(16:0), PC(38:7), Cer(42:0), PE(38:8), PE(36:4), PE(40:7), PE(38:4), PE(38:5), PE(42:10), glutamylglycine, creatine, diaminocaproate, aspartic acid, adenosylsuccinic acid, hexylresorcinol, tridecanoic acid, pregnenolone, and undecanoic acid affect daytime dysfunction. Among them, some risk metabolites such as salicylic acid, tridecanoic acid, undecanoic acid, LysoPE (16:0), PC (34:2) and PE (38:8) were significantly correlated with the three aspects of sleep quality in PSQI. The above classification partially reveals the role of metabolites in sleep regulation.

[0065] (3) Prediction model of sleep disorders

[0066] To establish a metabolic model for participants diagnosed with sleep disorders, ridge regression analysis was performed using the glmnet package in R with 52 risk metabolites. The 443 samples were randomly divided into training and test sets in an 8:2 ratio. The quality of the trained model was validated by 10-fold cross validation to optimize the penalty parameter ( Figure 3 a), and the model quality was evaluated by calculating the sensitivity, specificity, and area under the receiver operating characteristic curve (ROC). The area under the curve (AUC) of the training set was 72.24% (95% CI: 66.94% to 77.54%) ( Figure 3 b), the area under the curve (AUC) of the test group was 71.62% (95% CI: 60.25% to 82.99%) ( Figure 3 c). 32 out of 34 participants were used as an external validation set after removing outliers ( Figure 3d-3e), the AUC statistical value was 75.42% (95% CI: 57.17%-93.66%) ( Figure 3 f). Cronbach's alpha is a statistical indicator commonly used to measure reliability. Studies have found that the prediction accuracy of the metabolic model is consistent with the Cronbach's alpha value of PSQI (0.70-0.83)37. This indicates that the metabolites of the present invention can be used to construct a prediction model for sleep disorders.

[0067] (4) Impact of comorbidities and symptoms on sleep quality assessment

[0068] To evaluate the impact of comorbidity on sleep quality, participants were divided into four groups: Group I (n = 133), assessed as good by the questionnaire and predicted as good by the model; Group II (n = 73), assessed as poor and predicted as good; Group III (n = 71), assessed as good and predicted as poor; Group IV (n = 156), assessed as poor and predicted as poor ( Figure 4 a).

[0069] Statistical analysis showed that participants who were assessed as having poor sleep but were predicted to have good sleep (Group II) were significantly younger ( Figure 4 b) and suffer from much fewer chronic diseases ( Figure 4 c). Participants who were evaluated as good but predicted to have poor sleep (Group III) had higher TG and VLDL, and lower HDL, indicating that sleep quality is closely related to plasma lipid profiles. In addition, they had more types of comorbidities ( Figure 4 c). Among relevant comorbidities, subjects predicted to be good sleepers had much lower prevalence of hypertension, osteoarthritis, and cognitive impairment ( Figure 4 d-4f), indicating that comorbidity may significantly affect metabolic balance and the predictive accuracy of the model. Next, the clinical symptoms of all participants were examined and it was found that those who were rated as poor but predicted to be good sleepers (Group II) suffered more headaches, palpitations, chest tightness, precordial pain, diarrhea, bitter taste, dry eyes, back pain, and toothache, indicating that mild symptoms may affect questionnaire-based assessments but have little impact on the balance of sleep disorder risk metabolites ( Figure 4 g).

[0070] In addition, the abundance of 52 risk metabolites in groups I and IV were compared ( Figure 5 a), found that most of the metabolites except 3-hydroxy-N6,N6,N6-trimethyl-L-lysine (HTMLA) were significantly different between the two groups. Interestingly, those participants who were assessed as poor but predicted to be poor had much lower levels of HTMLA (Supplementary Figure 5b) Overall, the studies revealed the impact of aging, comorbidities, and clinical symptoms on metabolic balance and sleep quality assessments.

[0071] (5) Study on metabolic markers of sleep disorders, aging and related comorbidities

[0072] To establish a direct link between sleep disturbance, related comorbidities, and aging, we next performed a fusion analysis of 52 risk metabolites for sleep disturbance ( Figure 6 and Figure 7 ). The study found that the risk metabolites ChE (20:4), ChE (22:6), PC (44:6), PE (40:4), TG (54:2), stearoylcarnitine, palmitoylcarnitine, 2-phenylglycine, hydroxybutyrylcarnitine and hydroxybenzoic acid were not related to aging and / or certain related comorbidities, but were only related to sleep disorders. They are markers of sleep disorders and can be characteristically used to screen or diagnose sleep disorders ( Figure 6 ).

[0073] Meanwhile, many risk metabolites that were closely associated with sleep disturbance also showed close associations with aging and / or certain related comorbidities.

[0074] The study found that among the 52 sleep disorder risk metabolites, changes in four metabolites, including PE (42:10), L-kynurenine, 5-hydroxyindoleacetic acid, and pregnenolone, indicated a higher risk of aging than sleep disorders and related comorbidities ( Figure 6 ), suggesting that changes in these metabolites may be caused by aging, but also have an important impact on sleep disorders and related comorbidities. These metabolic markers can be used to screen or diagnose aging-related sleep disorders.

[0075] Among the related comorbidities, cognitive impairment, hypertension, and diabetes had the most common risk metabolites with sleep disorders. Among them, the metabolites ChE (18:3), PC (33:2), PC (34:2), PC (35:2), PC (36:2), PC (37:2), TG (52:7), and hexylresorcinol had a higher risk of cognitive impairment than sleep disorders and other comorbidities, but also had an important impact on sleep disorders and related comorbidities. These metabolite markers can be used to screen or diagnose sleep disorders related to cognitive impairment.

[0076] Metabolites PC (38:7), PE (40:5), and 3-hydroxy-N6,N6,N6-trimethyl-L-lysine are biomarkers that can be used to screen or diagnose sleep disorders caused by hypertension.

[0077] The present invention discloses risk metabolites of sleep disorders and studies the relationship between these risk metabolites and sleep disorder comorbidities. These risk metabolites can be used as markers for screening or diagnosing sleep disorders, and are used for screening or diagnosing sleep disorders, especially for screening or diagnosing sleep disorders affected by different comorbidities, and have good application prospects.

Claims

1. Use of reagents for detecting metabolic markers in the preparation of sleep disorder screening or diagnostic products, Features: The metabolite markers are triglyceride metabolites, phosphatidylcholine metabolites, cholesterol lipid metabolites, phosphatidylethanolamine metabolites, ceramide metabolites, amino acids, peptides and their analogs, benzoic acid and its derivatives, fatty acid metabolites, fatty acid ester metabolites, steroid and derivative metabolites and other types of metabolites; The triglyceride metabolites are TG(53:2), TG(52:2), TG(52:7), TG(54:2) and TG(50:3); The phosphatidylcholine metabolites are PC(37:2), PC(33:2), PC(44:6), PC(35:2), PC(36:2), PC(38:7), PC(34:3), PC(36:4) and PC(34:2); The cholesterol lipid metabolites are ChE(22:6), ChE(20:4) and ChE(18:3); The phosphatidylethanolamine metabolites are PE(42:10), PE(40:5), PE(38:4), PE(40:4), PE(40:7), PE(36:4), PE(38:8) and PE(38:5); The ceramide metabolites are Cer(42:0), Cer(40:1) and Cer(42:1); The amino acids and polypeptides and their analogs are L-serine, glutamylglycine, creatine, diaminocaproate, 2-phenylglycine, L-aspartic acid, guanidinoacetic acid and 3-hydroxy-N6,N6,N6-trimethyl-L-lysine; The benzoic acid and its derivatives are salicylic acid, hexylresorcinol, adenylic acid succinic acid and hydroxybenzoic acid; The fatty acid metabolites are undecanoic acid and tridecanoic acid; The fatty acid ester metabolites are stearoylcarnitine, palmitoylcarnitine, linoleylcarnitine and hydroxybutyrylcarnitine; The steroid and derivative metabolites are calcidiol and pregnenolone; The other classes of metabolites are L-kynurenine, LysoPE(16:0), 5-hydroxyindoleacetic acid and sphingosine.

2. The use according to claim 1, Features: One or more of the metabolite markers ChE (20:4), ChE (22:6), PC (44:6), PE (40:4), TG (54:2), stearoylcarnitine, palmitoylcarnitine, 2-phenylglycine, and hydroxybutylcarnitine are characteristic metabolite markers for screening or diagnosing sleep disorders; And / or, one or more of the metabolite markers PE (42:10), L-kynurenine, 5-hydroxyindoleacetic acid, and pregnenolone are metabolite markers used for screening or diagnosing the occurrence of aging-related sleep disorders; and / or, one or more of the metabolite markers ChE (18:3), PC (33:2), PC (34:2), PC (35:2), PC (36:2), PC (37:2), TG (52:7) and hexylresorcinol are metabolite markers for screening or diagnosing the occurrence of sleep disorders associated with cognitive dysfunction; And / or, one or more of the metabolite markers PC (38:7), PE (40:5), 3-hydroxy-N6, N6, N6-trimethyl-L-lysine are metabolite markers used to screen or diagnose sleep disorders caused by hypertension.

3. The use according to claim 1, Features: The content of the metabolic marker is up-regulated or down-regulated in the human body.

4. The use according to claim 1 or 2, Features: The reagent is an enzyme-linked immunosorbent assay reagent; And / or, the reagent is a western blot reagent; And / or, the reagent is a reagent for a protein chip detection method.

5. The use according to claim 1 or 2, Features: The reagent detects metabolite levels by one or more of the following methods: chromatography, spectroscopy, mass spectrometry, chemical analysis.

6. The use according to any one of claims 1 to 3, Features: The sample detected by the reagent is human peripheral blood.

7. The use according to claim 6, Features: The sample detected by the reagent is plasma.

8. The use according to claim 1, Features: The reagent for detecting one or more of the metabolic markers ChE (20:4), ChE (22:6), PC (44:6), PE (40:4), TG (54:2), stearoylcarnitine, palmitoylcarnitine, 2-phenylglycine, and hydroxybutylcarnitine is used to prepare a product for characteristic screening or diagnosis of sleep disorders; And / or, the reagent for detecting one or more of the metabolic markers PE (42:10), L-kynurenine, 5-hydroxyindoleacetic acid, and pregnenolone is used to prepare a product for screening or diagnosing aging-related sleep disorders; and / or, the reagent for detecting one or more of the metabolic markers ChE (18:3), PC (33:2), PC (34:2), PC (35:2), PC (36:2), PC (37:2), TG (52:7) and hexylresorcinol is used to prepare a product for screening or diagnosing sleep disorders associated with cognitive dysfunction; And / or, the reagent for detecting one or more of the metabolic markers PC (38:7), PE (40:5), 3-hydroxy-N6, N6, N6-trimethyl-L-lysine is used to prepare a product for screening or diagnosing sleep disorders caused by hypertension.

Images