第一部分 睡眠障碍与肥胖的关联：基于医务人员队列的横断面研究 目的：本研究旨在分析医务人员的睡眠状况与超重肥胖现象之间的关系，探索睡眠质量对超重肥胖发生的潜在影响，为制定有效的健康干预措施提供科学依据。

方法：采用方便抽样方法，对北京协和医院和天津东丽区中医院的医务人员开展睡眠特征与超重肥胖状况的相关调查。通过网络问卷收集数据，调查内容涵盖人口统计学指标、匹兹堡睡眠质量指数量表（PSQI）、人体测量参数、睡眠行为、运动和饮食习惯等，研究对象的身高、体重、行为学数据均采用自陈式报告法记录。通过构建单因素和多因素logistic回归模型，重点探究研究医务人员的睡眠质量与体重异常之间的关联。本研究共发放问卷402份，有效回收率为100%。

结果：调查对象的平均BMI值为23.22±3.87kg/m2。其中，超重及肥胖共144例，占比35.8%（144/402）。医务人员较差睡眠质量发生率为27.4%（110/402），平均PSQI得分为8.37±3.624分，超重肥胖组较正常组较差睡眠质量的发生率显著增高（36.1% vs 22.5%，P=0.003）。多因素分析结果显示，睡眠时间（OR = 1.411，95%CI 1.043–1.910，P = 0.026），睡眠障碍（OR = 1.574，95% CI 1.123–2.206，P = 0.008）是医务人员超重肥胖的显著危险性因素。

结论：超重或肥胖医务人员较体重正常者睡眠质量较差，睡眠时间，睡眠障碍是医务人员超重或肥胖的独立危险因素，增加睡眠时间和改善睡眠障碍对于控制医务人员的超重肥胖可能具有积极作用。

第二部分 荷叶抗肥胖成分及靶点的筛选：网络药理学-机器学习-分子模拟的协同验证

目的：本章旨在筛选治疗肥胖的荷叶活性化合物及潜在靶点，之后对靶点进行网络药理学分析，筛选出抗肥胖的最主要靶点和所涉及的信号通路。将活性化合物与关键靶点蛋白进行分子对接筛选高亲和荷叶活性化合物，通过机器学习模型验证其有效性。利用分子动力学模拟研究在分子层面上Sitogluside和Cycloartenol对 PPARγ的作用机制及抑制效果。

方法：利用中药系统药理学数据库与分析平台(TCMSP)对荷叶的主要活性成分进行了筛选，随后利用网络药理学，深入探讨了三种代表化合物抗肥胖的作用机制及其靶点相互作用网络。将活性化合物与关键靶点蛋白进行分子对接筛选高亲和荷叶活性化合物。建立机器学习模型用于预测靶点蛋白抑制剂。对活性成分与靶蛋白复合体系和空蛋白体系分别进行分子动力学模拟。

结果：共筛选到的18种活性化合物进行聚类分析，得到3种代表活性化合物，聚类分析确定了胡萝卜苷(Sitogluside), 山奈酚(Kaempferol)和荷叶碱(Nuciferine)为主要活性成分。通过网络药理学分析确定PPARγ为共同靶基因。此外，基因本体（Gene Ontology，GO）数据库和基因组百科全书（Kyoto Encyclopedia of Genes and Genomes，KEGG）数据库对潜在靶点进行富集分析结果显示，活性化合物影响关键的生物过程和通路，包括炎症调节、细胞信号传导和代谢过程。分子对接结果显示胡萝卜苷(Sitogluside)和环阿屯醇(Cycloartenol)与PPARγ具有很强的结合亲和力。机器学习模型的开发包括使用随机森林 (Random Forest，RF) 和极端梯度提升 (Extreme Gradient Boosting，XGBoost) 算法，以五种分子指纹(MACCS, Morgan, RDKit, Topological Torsion, AtomPairsFP)作为输入，最后产生 10个预测模型。机器学习模型将Sitogluside和Cycloartenol预测为PPARγ的抑制剂，再次强调了荷叶活性化合物Sitogluside和Cycloartenol与关键靶点蛋白PPARγ之间的生物活性关系。分子动力学模拟结果显示，Sitogluside与PPARγ复合体系相比Cycloartenol与PPARγ复合体系而言：配体-受体结合体系更加稳定和紧密；分子间正负相关运动较大，可能更有利于受体与配体的深度结合；Sitogluside的结合自由能(-36.49 ± 3.69 kcal/mol)高于Cycloartenol(-33.41 ± 2.34 kcal/mol)。

结论：荷叶提取物中，Sitogluside和Cycloartenol两种代表活性化合物可能通过多方面的方式发挥其抗肥胖作用，包括抑制PPARγ通路、调节抗炎、改善代谢等。而Sitogluside对蛋白PPARγ的抑制效果可能比Cycloartenol更好，为荷叶活性化合物的抗肥胖治疗提供了分子基础，并为开发新的、有效的、更安全的肥胖治疗方法提供了新思路。

第三部分 胡萝卜苷（Sitogluside）抗肥胖作用的体内外验证

目的：本研究旨在评估PPARγ在人体脂肪组织中的表达情况，并通过体外细胞实验和体内动物模型评估 Sitogluside 的抗肥胖活性，探索其改善肥胖的潜在分子机制。

方法：首先，通过HE和IHC染色检测人体脂肪组织样本中 PPARγ 的表达水平情况。对小鼠3T3-L1前脂肪细胞系进行Sitogluside干预，通过油红O染色观察脂滴形成和脂肪细胞分化情况，并检测细胞内甘油三酯聚集程度。建立高脂饮食诱导肥胖（DIO）小鼠模型，进行为期6周Sitogluside（10mg/kg和15mg/kg）灌胃，观察体重以及肝脏脂质沉积变化，进行葡萄糖耐量实验(GTT)和胰岛素耐量实验(ITT)，再通过MRI扫描体脂成分，检测血清学脂代谢指标，取材后利用免疫印记、免疫组织化学方法检测肝脏、脂肪组织PPARγ的表达，利用16S测序观察肠道菌群变化情况。

结果：PPARγ在肥胖人群的脂肪组织中表达较高。Sitogluside在10uM浓度下可显著下调3T3-L1细胞的PPARγ mRNA及蛋白表达水平（分别降至对照组的27.03%和30.2%），并减少脂滴面积和甘油三酯蓄积达50%以上，且未表现细胞毒性（存活率＞95%）。在高脂饮食诱导的肥胖（DIO）小鼠模型中，15 mg/kg Sitogluside干预6周显著降低体重增幅（较模型组减少26.6%），减少内脏脂肪组织重量（63.8%）及肝脏脂质沉积（肝脏脂肪分数下降65.5%），同时改善糖耐量异常（OGTT-AUC下降39.1%）和胰岛素抵抗（ITT-AUC降低42.2%），血清总胆固醇水平从6.67 mmol/L降至3.12 mmol/L。肠道菌群分析进一步揭示，Sitogluside可重塑肥胖相关菌群失调，显著提升阿克曼氏菌（Akkermansia muciniphila）丰度，并降低厚壁菌门/拟杆菌门比值。

结论：Sitogluside可能通过调控PPARγ信号通路、改善糖脂代谢及影响肠道菌群，实现其抗肥胖作用，为代谢疾病的潜在治疗提供新的研究思路。

Part I Poor Sleep Quality and Overweight/Obesity in Healthcare Professionals: A Cross-Sectional Study

Objective: This study aims to analyze the relationship between the sleep conditions of healthcare professionals and the incidence of overweight and obesity, exploring the potential impact of sleep quality on the onset of overweight and obesity, in order to provide a scientific basis for formulating effective health intervention measures.

Methods: A convenience sampling method was employed to conduct a survey on the sleep characteristics and obesity status among healthcare professionals at Peking Union Medical College Hospital and Tianjin Dongli District Traditional Chinese Medicine Hospital. The survey was conducted via online questionnaires, which included demographic data, the Pittsburgh Sleep Quality Index (PSQI), height, weight, and related sleep, exercise, and dietary habits. Univariate and multivariate logistic regression analyses were applied to study the relationship between sleep quality and overweight/obesity among healthcare professionals. A total of 402 questionnaires were distributed, with a 100% retrieval rate, yielding 402 valid questionnaires.

Results: The average BMI of the subjects was 23.22±3.87 kg/m^2. Among them, 144 cases were overweight or obese, accounting for 35.8% (144/402) of the total. The prevalence of poor sleep quality among healthcare professionals was 27.4% (110/402), with an average PSQI score of 8.37±3.624. The rate of poor sleep quality was significantly higher in the overweight and obese group compared to the normal-weight group (36.1% vs 22.5%, P=0.003). Multivariate analysis indicated that sleep duration （OR = 1.411，95%CI 1.043–1.910，P = 0.026), and sleep disorders (OR = 1.574，95% CI 1.123–2.206，P = 0.008) were significant risk factors for overweight and obesity among healthcare professionals.

Conclusion: Overweight or obese healthcare professionals had poorer sleep quality compared to those with normal weight. Sleep duration and sleep disorders were identified as independent risk factors for overweight or obesity in healthcare professionals. Increasing sleep duration and improving sleep disorders may play a positive role in controlling overweight and obesity among healthcare professionals.

Part II Identification of Anti-Obesity Compounds and Targets in Lotus Leaf Extract through Integrated Approaches of Network Pharmacology, Machine Learning, and Molecular Dynamics Simulation

Objective: The aim of this chapter is to screen the active compounds and potential targets of Lotus leaf for the treatment of obesity. Subsequently, network pharmacology was applied to identify key targets and associated signaling pathways in anti-obesity mechanisms. Molecular docking was performed between active compounds and key target proteins to identify high-affinity lotus leaf components, with their effectiveness confirmed through machine learning models. Molecular dynamics simulations were employed to investigate the molecular mechanisms and inhibitory effects of Sitogluside and Cycloartenol on PPARγ.

Methods: The primary active components of lotus leaves were screened using the Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP), followed by network pharmacology to investigate the anti-obesity mechanisms and target interaction networks of three representative compounds. Molecular docking was performed between active compounds and key target proteins to screen for high-affinity lotus leaf components. Machine learning models were developed to predict target protein inhibitors. Both active compound-target protein complexes and apo-protein systems underwent molecular dynamics simulations.

Results: We screened a total of 18 active compounds and cluster analysis resulted in 3 representative active compounds, including Sitogluside, Kaempferol and Nuciferine. PPARγ was identified as a common target gene by network pharmacological analysis. Additionally, enrichment analysis of potential targets using Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) databases showed that active compounds influence crucial biological processes and pathways, including inflammation regulation, cell signaling, and metabolic processes. Molecular docking results showed that Sitogluside and Cycloartenol exhibited strong binding affinity with PPARγ. Machine learning models were developed using RF and XGBoost algorithms, incorporating five molecular fingerprints (MACCS, Morgan, RDKit, Topological Torsion, AtomPairsFP) as inputs, resulting in 10 predictive models. Sitogluside and Cycloartenol were predicted as PPARγ inhibitors by the machine learning models, highlighting the bioactive connection between these lotus leaf compounds and the key target protein PPARγ. Molecular dynamics simulations showed that the ligand-receptor binding system of Sitogluside and PPARγ composite system was more stable and tightly packed, with larger positive-negative correlated motions, which might promote deeper receptor-ligand interactions, compared with that of the Cycloartenol and PPARγ composite system. The free energy of binding of Sitogluside (-36.49 ± 3.69 kcal/mol) was higher than that of Cycloartenol (-33.41 ± 2.34 kcal/mol).

Conclusion: In lotus leaf extract, the two representative active compounds, Sitogluside and Cycloartenol, may exert anti-obesity effects through multiple mechanisms, including PPARγ pathway inhibition, anti-inflammatory regulation, and metabolic improvement. Moreover, Sitogluside may demonstrate superior inhibitory effects on PPARγ compared to Cycloartenol.Our study provide a molecular basis for the anti-obesity treatment of lotus leaf active compounds and offering new insights for developing novel, effective, and safer obesity therapies.

Part III In Vivo and In Vitro Validation of the Anti-Obesity Effects of Sitogluside

Objective: To investigate the expression pattern of PPARγ in human adipose tissue. Furthermore, this study aims to evaluate the anti-obesity effects of Sitogluside and explore its underlying mechanisms for improving obesity using both in vitro cellular experiments and in vivo animal models.

Methods: Human adipose tissue samples were analyzed using HE staining and IHC to detect the expression level of PPARγ.3T3-L1 preadipocytes were treated with Sitogluside, with subsequent assessment of lipid droplet formation and adipocyte differentiation via Oil Red O staining, and quantification of intracellular triglyceride levels. A diet-induced obesity (DIO) mouse model was established, followed by 6-week oral administration of Sitogluside (10mg/kg and 15mg/kg) to monitor body weight and hepatic lipid deposition changes. Glucose tolerance test (GTT) and insulin tolerance test (ITT) were performed, with body fat composition analyzed through MRI scanning. Serum lipid metabolism indicators were measured, and PPARγ expression in liver and adipose tissues was examined using Western blot and immunohistochemistry. Gut microbiota alterations were observed via 16S sequencing.

Results: Immunohistochemical analysis indicated relatively higher expression levels of PPARγ in the adipose tissue of obese individuals. Sitogluside at 10 μM concentration markedly reduced PPARγ mRNA and protein expression in 3T3-L1 cells (to 27.03% and 30.2% of control, respectively), suppressed lipid droplet area and triglyceride accumulation by more than 50%, while maintaining cell viability above 95%. In DIO mouse models, 6-week intervention with 15 mg/kg Sitogluside significantly reduced weight gain (26.6% decrease compared to model group), visceral adipose tissue weight (63.8% reduction), and hepatic lipid deposition (65.5% decrease in hepatic fat fractions), while improving glucose tolerance (39.1% reduction in OGTT-AUC) and insulin resistance (42.2% decrease in ITT-AUC). Serum total cholesterol levels dropped from 6.67 mmol/L to 3.12 mmol/L. Gut microbiota analysis further revealed that Sitogluside could remodel obesity-related dysbiosis, significantly increasing the abundance of Akkermansia muciniphila and reducing the Firmicutes/Bacteroidetes ratio.

Conclusion: Sitogluside may exert its anti-obesity effects through modulating the PPARγ signaling pathway, improving glucose and lipid metabolism, and influencing gut microbiota, providing new research perspectives for potential metabolic disease treatments.