肥胖是一种多系统疾病，几乎影响到包括中枢神经系统在内的所有器官系统。肥胖与认知功能的关系复杂，由于肥胖是基因和环境相互作用的结果，肥胖与认知功能之间的关系可能受到多种因素的影响，认知功能活动准确进行需要相对完整的大脑结构。肥胖老年人的大脑结构也发生改变。虽然肥胖与认知之间的复杂关系尚未从现有研究中得到充分理解，但探索肥胖与老年人认知功能的这种关系对于推进预防认知功能障碍具有重要的价值。

本研究基于国内外数据库，系统地整合行为学、影像学和基因组学研究等多维度指标，全面地阐明肥胖对老年人认知功能影响背后的神经基础，以期为老年预防认知退化提供新的思路。

 

第一部分 探讨肥胖对老年人认知功能的影响

本部分包括两节内容，在社区老年居民中探究肥胖与认知障碍的关系。

第一节  基于NHANES数据库探讨肥胖对美国老年人认知功能的影响

目的：本研究旨在利用美国国家健康与营养调查（NHANES）的数据，描述不同 BMI和腰围状态的老年人在不同认知领域的认知功能，评估肥胖、生活方式因素和认知不良之间的复杂联系。

材料与方法：研究采用2011至2014年美国国家健康与营养调查（NHANES）的数据进行横断面分析。从NHANES数据库中提取了肥胖指标、认知功能、身体活动、饮食习惯、睡眠、慢性病状况以及社会人口学信息的数据。采用倾向性匹配分析、线性回归分析、结构方程模型等研究各项肥胖指标与认知测试之间的关系。

结果：本研究共纳入 2,222 名参与者。在 CERAD 单词列表学习任务（CERAD-WL）测试中，超重和肥胖人群在三次学习试验中的表现均不如正常体重的老年人。肥胖组的数字符号转换测验（DSST）得分低于超重组。在控制了年龄、性别、教育程度等因素后，较高的 BMI与 CERAD-WL 和 DSST 得分呈负相关，腰围与 CERAD-WL、动物词语流畅性测验（AFT）和 DSST 的得分呈负相关。在结构方程模型中，腰围状况完全介导了从生活方式到 CERAD-WL 的路径，中介效应为 0.018。

结论：肥胖会损害学习能力和工作记忆。同时，我们证明了生活方式对认知的影响是通过与肥胖相关的身体指标来介导的。睡眠、身体活动和饮食影响肥胖程度，而肥胖程度又决定了认知水平。积极的体重管理和保持健康的生活方式可能有助于预防认知能力下降。

 

第二节  基于BABRI社区队列数据探讨肥胖对北京社区非痴呆老年人认知功能的影响

目的：本节研使用来源于中国本土社区的BABRI （Beijing Aging Brain Rejuvenation Initiative）队列数据探究肥胖对老年人各认知域认知功能的影响；构建结构方程模型，以探讨生活方式和伴发疾病等因素在肥胖与认知功能关系之中的作用，为建立符合中国人群特征的肥胖相关认知障碍预警体系提供证据支持。

材料与方法：此部分的研究对象来自于 BABRI社区队列数据。采用独立双样本t检验或卡方检验比较两组特征的差异。通过相关分析和线性回归研究BMI与行为结果之间的图形关联。双侧p < 0.05认为有统计学意义。使用结构方程建来评估生活方式（身体活动，饮食或疾病）与肥胖和认知表现的关联程度。统计学分析使用R语言完成，结构方程模型分析使用Mplus 8完成。

结果：肥胖成人在注意力和记忆力方面的表现较非肥胖对照组差。日常身体活动和肥胖相关的慢性疾病可以介导了肥胖与AVLT或SDMT评分的关系，而健康的饮食习惯没有表现出显著的中介作用。

结论：北京社区居民中肥胖老年人的在记忆力和注意力方面的认知能力表现较差。老年人的肥胖程度与记忆力、执行功能、处理速度和语言能力等认知能力成负相关关系。身体活动和慢性疾病可以介导肥胖与认知功能之间的关系。避免过度肥胖，保持合适体重对保护老年人认知功能具有重要意义。

 

 

 

第二部分 肥胖老年人大脑结构完整性和功能活动的神经影像学机制研究

目的：许多神经影像学研究表明肥胖会对大脑产生显著影响，但结果仍存在一定争议，目前依旧缺乏关于中国老年人口中肥胖人群大脑结构完整性和功能连接程度的相关研究。本研究利用BABRI社区队列中的神经影像数据部分，探讨肥胖影响大脑的神经影像学机制。

材料与方法：影像部分研究纳入了506名参与者，进行核磁数据的灰质结构和功能连接分析。基于体素形态学分析（VBM）来分析核磁数据。使用CAT12工具箱预处理结构T1加权的MRI数据。选择VBM分析的结果灰质差异显著的脑作为种子点进行功能连接分析。控制年龄、性别、受教育年数和颅内总体积等协变量，采用dpabi工具箱进行脑影像数据的统计分析，比较肥胖组与非肥胖组大脑差异。多重比较校正采用GRF校正，体素水平p < 0.001、团块水平p < 0.05为阳性结果。差异脑区与认知结果进行偏相关分析以探究大脑改变与认知功能的关系。将显著脑区结果添加到结构方程模型中，以评估肥胖、生活方式、灰质体积和认知之间的关系。所有统计分析采用SPSS 26.0版、Mplus 8 和R 4.1.2版进行。

结果：与非肥胖对照组相比，肥胖老年人大脑左侧缘上回和颞上回的灰质体积减少，右侧额中回和额上回灰质体积增加。SDMT成绩与右侧额中回的灰质体积呈负相关 （β = -0.253，p = 0.026）。AVLT成绩与左侧颞上回的灰质体积呈正相关（β = 0.279，p = 0.014）。结构方程模型中，身体活动和疾病均介导了肥胖与大脑结构之间的关联, 从大脑结构到数字符号转换测试（SDMT）的路径具有显著性。肥胖通过疾病影响右侧额中回的灰质体积，然后右侧额中回较大的灰质体积与注意力表现（SDMT）更好显著相关（β=0.167±0.044，p<0.001）。功能分析中，肥胖老年人左缘上回与左侧脑岛、左侧中央沟盖和右侧枕上回之间的功能连通性增加；左侧颞上回与双侧扣带或右侧角回之间的连通性增加的模式。

结论：肥胖会损害老年人的记忆力、注意力、处理速度等高级认知功能。肥胖老年人左侧边缘上回和颞上回萎缩，萎缩脑区与突显模式网络所在脑区功能连接增强，提示肥胖老年人存在脑损伤，存在代偿网络机制。

 

 

 

第三部分 肥胖对老年人大脑结构影响的神经影像转录组学研究

本部分包括两节内容，从基因表达角度探究肥胖对大脑结构影响的生物学机制，完成了从神经影像到生物学视角的整合。

第一节  孟德尔随机化法探讨肥胖与大脑形态学特征的关系

目的：将肥胖与人类大脑中的灰质和遗传信息联系起来的研究工作依旧不多。全脑基因表达图谱已将基因表达的空间变化与大脑结构和功能联系起来，有助于更清晰地理解肥胖与大脑结构之间的关联。在本研究中，我们利用孟德尔随机化的方法确定肥胖与大脑皮层之间的因果关联，随后获取了肥胖相关脑区的基因表达情况，进行富集分析以检测了相关的生物学过程或功能通路。

材料与方法：单变量两样本孟德尔随机化分析，以评估肥胖对大脑皮质结构的因果效应。基于两样本孟德尔随机化分析的结果额外进行了多变量孟德尔随机化分析，以进一步探究肥胖对大脑皮质的影响。随后从艾伦人脑图谱中提取了在 MR 分析中具有统计学意义的大脑区域的特异性表达基因。我们利用这些基因进行了生物信息学分析。

结果：较高 BMI 或内脏脂肪组织水平与全脑皮质厚度变薄有关。对于皮层厚度这一指标，受肥胖影响的脑回涉及内嗅皮质、梭状回、颞下回、顶下回、眶额回、眶额三角部、顶上小叶、缘上回和颞极。对于皮质表面积这一指标，受影响的脑区主要包括顶下小叶、颞下回、扣带回峡部和颞横回。使用来自艾伦人图谱中的数据进行了一系列生物信息学分析。GO 和 KEGG 分析表明，肥胖相关脑区特异性表达的基因参与化合物转运、激素分泌、免疫反应、神经肽信号通路和神经活性配体受体相互作用等神经生理过程。在 DO 富集分析中，基因在认知疾病（如阿尔茨海默病、tau 蛋白病和路易体痴呆）中和帕金森病、偏头痛、癫痫和情绪障碍等疾病中富集。蛋白质-蛋白质相互作用（PPI）分析确定了多个与神经系统疾病密切相关的枢纽蛋白，包括APOE、IL-1β、PVALB和HTR2C。这些枢纽蛋白与其他基因蛋白形成了高度连接的相互作用网络。

结论：本研究提供了肥胖与大脑皮质形态学在整体和区域层面存在因果关联的遗传学证据。受肥胖影响的脑区特异性改变也与阿尔茨海默病和其他神经退行性疾病有关。

 

第二节  基于BABRI社区队列结果的神经影像转录组学研究

目的：肥胖是环境因素和先天生物因素相互作用的结果。神经影像转录组学分析通过将艾伦脑转录组信息与脑影像表型相关联，揭示与认知功能、大脑发展和疾病有关的分子基础。

材料与方法：本研究使用第二部分BABRI社区队列人群中肥胖老年人大脑灰质体积统计值脑图，与艾伦人脑图谱中的人脑空间转录组数据进行关联分析, 随后对鉴定出的与老年人肥胖脑结构变化的相关基因进行一系列富集分析。

结果：肥胖相关大脑灰质体积 t值脑图与1207个基因的表达量相关，我们使用Metascape平台进行功能富集分析来探究与这些基因相关的生物学功能。GO分析中基因在膜功能、蛋白质分解代谢过程、突触信号传导和细胞程序性死亡等方面富集。对肥胖相关基因进行KEGG富集分析共发现25条信号通路，其中大多数通路与衰老、炎症和细胞死亡有关，如神经退行性疾病、催产素信号通路、多巴胺能突触、MAPK信号通路、内吞作用等。细胞类型富集分析结果显示，这些基因在多种类型的兴奋性神经元（Ex1、Ex3b、Ex3e和Ex4）中特异性表达。

结论：对老年肥胖者大脑异常改变相关基因的富集分析表明，这些基因涉及到了这些基因主要富集于突触和生物组织膜相关的通路、神经退行性疾病、多巴胺能突触、细胞衰老、细胞程序性死和MAPK信号通路等通路。肥胖大脑损伤的基因定位在兴奋性神经元。这些发现有助于我们从遗传机制的角度去理解肥胖对大脑灰质体积的影响。

 

 

 

 

 

 

 

 

 

Obesity is a multisystem disease that affects almost all organ systems, including the central nervous system. The relationship between obesity and cognitive function is complex. Because obesity is the result of interactions between genes and the environment, the relationship between obesity and cognitive function may be affected by a variety of factors. Accurate cognitive function requires a relatively complete brain structure. The brain structure also changes in obese elderly people. Although the complex relationship between obesity and cognition has not been fully understood by existing studies, exploring the relationship between obesity and cognitive function in elderly individuals is highly valuable for the promotion and prevention of cognitive dysfunction.

On the basis of domestic and international databases, this study systematically integrated multidimensional indicators such as behavioral, imaging and genomics studies to comprehensively elucidate the neural basis behind the effect of obesity on cognitive function in elderly individuals to provide new directions for the prevention of cognitive degradation in elderly individuals. Ideas.

 

Part I. Investigated the effect of obesity on cognitive function in elderly people

 

This section includes two sections to investigate the association between obesity and cognitive impairment among elderly community residents.

 

Section 1 investigated the effect of obesity on the cognitive function of American elderly people via the NHANES database.

Objective: This study aimed to use data from the U.S. National Health and Nutrition Survey (NHANES) to describe cognitive function in different cognitive domains of elderly people with different BMIs and waist circumference statuses and to evaluate the associations among obesity, lifestyle factors and poor cognition. Complex relationships.

Materials and methods: This study used data from the 2011–2014 National Health and Nutrition Survey (NHANES) for cross-sectional analysis. Data on obesity indicators, cognitive function, physical activity, dietary habits, sleep, chronic disease status, and sociodemographic information were extracted from the NHANES database. Propensity matching analysis, linear regression analysis, and structural equation modeling were used to investigate the relationships between obesity indicators and cognitive tests.

Results: A total of 2,222 participants were enrolled in this study. In the Consortium to Establish a Registry for AD - word list learning task (CERAD-WL) test, the performance of the overweight and obese participants in the three learning trials was not as good as that of the normal weight elderly participants. The digit symbol substitution test (DSST) score of the obese group was lower than that of the overweight group. After controlling for age, sex, education level and other factors, a higher BMI was negatively correlated with CERAD-WL and DSST scores, and waist circumference was negatively correlated with CERAD-WL, animal fluency test (AFT) and DSST scores. In the structural equation model, waist circumference status completely mediated the path from lifestyle to CERAD-WL, with a mediating effect of 0.018.

Conclusion: Obesity impairs learning ability and working memory. Moreover, we demonstrated that lifestyle effects on cognition were mediated through obesity-related body indicators. Sleep, physical activity and diet affect the degree of obesity, which in turn affects cognitive ability. Active weight management and maintaining a healthy lifestyle may help prevent cognitive decline.

 

In Section 2, on the basis of data from the BABRI community cohort, the effect of obesity on the cognitive function of elderly individuals without dementia in the Beijing community was investigated.

 

Objective: This study used BABRI (Beijing Aging Brain Rejuvenation Initiative) cohort data from indigenous Chinese communities to investigate the effect of obesity on the cognitive function of elderly people in various cognitive domains and to construct a structural equation model to investigate lifestyle and comorbidities. The role of disease and other factors in the relationship between obesity and cognitive function provides evidence supporting the establishment of an early warning system of obesity-related cognitive impairment that is in line with the characteristics of the Chinese population.

Materials and methods: The study subjects in this part were from the BABRI community cohort data. The characteristics of the two groups were compared via an independent two-sample t test or the chi-square test. Graphic associations between BMI and behavioral outcomes were investigated via correlation analysis and linear regression. Two-sided p < 0.05 was considered statistically significant. Structural equations were used to assess the extent of the associations between lifestyle (physical activity, diet, or disease) and obesity and cognitive performance. Statistical analysis was performed via R language, and structural equation modeling analysis was performed via Mplus 8.

Results: The attention and memory performance of obese adults was worse than that of nonobese controls. Daily physical activity and obesity-related chronic diseases mediated the relationship between obesity and AVLT or SDMT scores, whereas healthy eating habits did not have a significant mediating effect.

Conclusion: Among Beijing community residents, obese elderly people presented poorer cognitive performance in memory and attention. The degree of obesity in elderly people is negatively correlated with cognitive abilities such as memory, executive function, processing speed, and language ability. Physical activity and chronic diseases can mediate the relationship between obesity and cognitive function. Avoiding obesity and maintaining an appropriate weight are important for protecting cognitive function in elderly individuals.

 

 

 

Part II. Neuroimaging mechanism of brain structural integrity and functional activities in obese elderly people

 

Objective: Many neuroimaging studies have shown that obesity has significant effects on the brain, but the results are still controversial. Relevant studies on the structural integrity and functional connectivity of the brain in the obese Chinese elderly population are still lacking. This study used neuroimaging data from the BABRI community cohort to investigate the neuroimaging mechanisms of the effects of obesity on the brain.

Materials and methods: In the imaging part of the study, 506 participants were enrolled, and gray matter structure and functional connectivity analyses of MMR data were performed. The NMR data were analyzed via voxel-based morphological analysis (VBM). Structural T1-weighted MR data were preprocessed via the CAT12 toolbox. The brains with significant differences in gray matter according to the VBM analysis results were selected as seed points for functional connectivity analysis. Covariates such as age, sex, years of education, and total intracranial volume were controlled, and the dpabi toolbox was used for statistical analysis of brain imaging data to compare the differences between the brains of the obese and nonobese groups. GRF was used to correct for multiple comparisons, and p < 0.001 at the voxel level and p < 0.05 at the clump level were considered positive. Partial correlation analysis was performed between the differences in brain regions and the cognitive results to explore the relationships between brain changes and cognitive function. The results of the significant brain regions were added to the structural equation model to assess the relationships among obesity, lifestyle, gray matter volume, and cognition. All the statistical analyses were performed via SPSS version 26.0 (SPSS Inc., Chicago, Illinois, USA), Mplus 8 and R version 4.1.2.

Results: Compared with those of nonobese controls, the gray matter volumes of the left superior marginal gyrus and superior temporal gyrus were decreased, and the gray matter volumes of the right middle frontal gyrus and superior frontal gyrus were increased in obese elderly people. The SDMT score was negatively correlated with gray matter volume in the right middle frontal gyrus (β = -0.253, p = 0.026). The AVLT score was positively correlated with the gray matter volume of the left superior temporal gyrus (β = 0.279, p = 0.014). According to the structural equation model, both physical activity and disease mediated the associations between obesity and brain structure, and the pathway from the brain structure to the SDMT was significant. Obesity affected the gray matter volume of the right middle frontal gyrus throughout the disease course, and a larger gray matter volume in the right middle frontal gyrus was significantly associated with better attention performance (SDMT) (β=0.167 ± 0.044, p <0.001). According to the functional analysis, the functional connectivity between the left superior marginal gyrus and the left insula, the left central sulcus and the right superior occipital gyrus was increased in obese elderly people; patterns with increased connectivity were also observed.

Conclusion: Obesity can impair cognitive functions such as memory, attention, and processing speed in elderly people. The left superior marginal gyrus and superior temporal gyrus are atrophied in obese elderly people, and the functional connectivity between the atrophic brain regions and the brain regions where the salience pattern network is located is increased, suggesting that obese elderly people have brain injury and that there is a compensatory network mechanism.

 

 

 

Part 3.  A neuroimaging transcriptomic study on the effect of obesity on the brain structure of elderly people

 

This section includes two sections. It investigates the biological mechanisms underlying the effects of obesity on brain structure from the perspective of gene expression and integrates the findings from neuroimaging with those from a biological perspective.

 

Section 1: The relationships between obesity and brain morphological characteristics were investigated via the Mendelian randomization method

 

Objective: Few studies have linked obesity with gray matter and genetic information in the human brain. The whole-brain gene expression atlas has linked spatial changes in gene expression with brain structure and function, which helps to elucidate the associations between obesity and brain structure more clearly. In this study, we used the Mendelian randomization method to ascertain the causal association between obesity and the cerebral cortex, obtained the gene expression status in obesity-related brain regions and performed enrichment analysis to detect related biological processes or functions. Pathway.

Materials and methods: Univariate two-sample Mendelian randomization was used to assess the causal effect of obesity on cerebral cortical structure. On the basis of the results of the two-sample MR analysis, we additionally performed multivariate MR (MVMR) analysis to further explore the effects of obesity on the cerebral cortex. The specifically expressed genes in brain regions that were statistically significant in the MR analysis were subsequently extracted from the Allen human brain atlas. We performed bioinformatics analysis on these genes.

Results: Increased BMI and visceral adipose tissue levels were associated with decreased global cortical thickness. The brain gyri affected by obesity are the entorhinal cortex, fusiform gyrus, inferior temporal gyrus, inferior parietal gyrus, orbitofrontal gyrus, orbitofrontal triangle, superior parietal lobule, superior marginal gyrus, and temporal pole. For the indicator of cortical surface area, the affected brain regions included mainly the infraparietal lobule, inferior temporal gyrus, cingulate gyrus and transverse temporal gyrus. A series of bioinformatics analyses were performed using data from the Allen atlas. GO and KEGG analyses revealed that genes specifically expressed in obesity-related brain regions were involved in neurophysiological processes such as compound transport, hormone secretion, the immune response, neuropeptide signaling pathways, and neuroactive ligand‒receptor interactions. In the DO enrichment analysis, genes were enriched in cognitive diseases (such as Alzheimer's disease, tauopathies, and dementia with Lewy bodies) and diseases such as Parkinson's disease, migraine, epilepsy, and mood disorders. Protein‒protein interaction (PPI) analysis revealed multiple hub proteins closely related to neurological diseases, including APOE, IL-1β, PVALB, and HTR2C. These hub proteins form a highly connected interaction network with other gene proteins.

Conclusion: This study provides genetic evidence for a causal association between obesity and cerebral cortical morphology at the global and regional levels. Brain region-specific changes caused by obesity are also associated with Alzheimer's disease and other neurodegenerative diseases.

 

Section 2: Neuroimaging transcriptomic study based on the results of the BABRI community cohort

 

Objective: Obesity is the result of the interaction between environmental factors and innate biological factors. Neuroimaging transcriptomic analysis correlates the Allen brain transcriptome information with brain imaging phenotypes to reveal the molecular basis associated with cognitive function, brain development and disease.

Materials and methods: This study used brain atlases of gray matter volume statistics for obese elderly people in the second part of the BABRI community cohort population and conducted an association analysis with human brain spatial transcriptome data from the Allen Human Brain Atlas. A series of enrichment analyses were performed on genes associated with brain structural changes in obese elderly people.

Results: The t value of obesity-related brain gray matter volume was correlated with the expression levels of 1207 genes. To explore the biological functions of these genes, we performed functional enrichment analysis via the Metascape platform. In the GO analysis, genes associated with membrane function, protein catabolic processes, synaptic signaling, and programmed cell death were enriched. KEGG enrichment analysis of obesity-related genes revealed 25 signaling pathways, most of which were associated with aging, inflammation and cell death, such as neurodegenerative diseases, the oxytocin signaling pathway, dopaminergic synapses, the MAPK signaling pathway, endothelial synapses, and the swallowing effect. The results of cell type enrichment analysis revealed that these genes were specifically expressed in various types of excitatory neurons (Ex1, Ex3b, Ex3e, and Ex4).

Conclusion: The enrichment analysis of genes associated with abnormal brain changes in elderly obese patients revealed that these genes were involved in pathways related to synapses and biological tissues, neurodegenerative diseases, dopaminergic synapses, cellular senescence, programmed cell death and the MAPK signaling pathway. The genes associated with obesity-related brain injury are located in excitatory neurons. These findings help us understand the effects of obesity on brain gray matter volume from the perspective of genetic mechanisms.