Brain structure, chemistry and function investigations in aging using MRI/MRS

通过分析高清晰度3D结构MRI图像获得的体积测量捕捉到了神经解剖学上的可变性，这可能与长期存在的特征有关，或者可能有助于解释为什么某些大脑区域或个人容易受到神经退化过程的影响。去年，我将一种名为组级独立分量分析(ICA)的数据简化技术应用于结构性MRI图像，以揭示结构性协方差(独立分量)的模式。我们将ICA应用于ADNI研究中收集的MRI数据，并研究了它们如何区分认知正常、轻度认知障碍(MCI)和阿尔茨海默病(AD)的参与者。我们表明，ICs可以作为未来AD诊断或从MCI转变为AD的分类器和预测者(手稿目前在《精神病学研究：神经成像》上出版)。 此外，考虑到胰岛素抵抗和阿尔茨海默病之间的关联，我们研究了AD几个关键感兴趣区的体积和FDG-PET摄取测量与外周胰岛素抵抗之间的关系。我们发现，胰岛素抵抗促进了两种异常和病理性的代偿：在轻度认知损害阶段，它增加了海马区和内侧颞叶的糖代谢；在AD阶段，它增加了默认模式网络节点的糖代谢。先前的功能磁共振研究表明，代谢的代偿性增加与疾病的进展有关；因此，胰岛素抵抗似乎促进了适应不良的补偿。这篇手稿目前正在《神经学》杂志上发表评论。 我们与我的博士后导师奥里尔·威莱特博士一起，对神经成像研究进行了系统的回顾，研究了肥胖与不同年龄段脑体积之间的关系。我们发现，肥胖增加主要与额叶萎缩有关。我们的综述发表在《老龄化研究综述》杂志上。 我使用一种新的方法进行了一项功能磁共振研究，以基于格兰杰因果关系原理对功能磁共振数据进行有效的连通性分析。这种方法使我们能够发现大脑网络节点之间的因果路径。我在2009年进行的一项关于宗教信仰的功能磁共振研究的数据集中实施了这一分析。这些发现支持了这样一种理论，即心理理论中涉及的区域是宗教信仰产生的关键。该研究发表在《大脑连通性》杂志上。 最后，我们与Chia博士和Egan博士合作，完成了RISE研究，这是一项多方面、随机、安慰剂对照的交叉临床研究，研究CB1受体激动剂和拮抗剂对外周代谢、脑功能和脑代谢控制的影响。这项研究包括一个强大的神经成像部分，包括fMRI和MRS。我们进行了两项激活范式fMRI研究，一项是发现头部胰岛素分泌与CB1受体的作用之间的大脑相关性，另一项是评估CB1受体对食物食欲的影响。第一项研究的目的是证明，由于某些大脑区域(脑岛、前扣带回、下丘脑、腹侧被盖区等)的激活，食物视觉刺激(头部胰岛素反应)导致胰岛素水平上升。此外，考虑到候选区域存在CB1受体，我们的目的是证明它们与CB1激动剂和拮抗剂的激活水平不同。第二项研究的目标是证明CB1受体刺激对食物价值(食物选择)和显着性(此类选择的强度)的分离影响。此外，我们进行了一项静息fMRI研究，以评估CB1对各种大脑网络功能连接的调节。最后，我们进行了MRS来评估CB1对大脑代谢(葡萄糖、乳酸)和神经传递(谷氨酸、GABA、甘氨酸)的调节。我们目前正在分析和解释这些研究的数据。

Volumetric measures obtained with analysis of high-definition 3D structural MRI images capture neuroanatomical variability which may be associated with long-standing traits or may help explain why certain brain areas or individuals are vulnerable to neurodegenerative processes. This last year, I applied a data reduction technique called group-level independent component analysis (ICA) to structural MRI images to uncover patterns of structural covariance (Independent Components). We applied ICA to MRI data collected in the ADNI study and examined how they can discriminate between participants with normal cognition, mild cognitive impairment (MCI) and Alzheimer's disease (AD). We showed that ICs can be useful as classifiers and predictors of future AD diagnosis or conversion to AD from MCI (the manuscript is currently in press in "Psychiatry Research: Neuroimaging"). In addition, given the association between insulin resistance and Alzheimer's disease, we examined how volumetric and FDG-PET uptake measures for several key regions of interest for AD relate to peripheral insulin resistance. We found that insulin resistance promotes two abnormal and pathogenic compensations: it increases glucose metabolism in the hippocampus and medial temporal lobe at the stage of mild cognitive impairment and increases glucose metabolism in default mode network nodes at the stage of AD. Prior fMRI studies suggest that these compensatory increases in metabolism are associated with disease progression; therefore, insulin resistance seems to promote a maladaptive compensation. The manuscript is currently under review in the journal "Neurology". Together with a post-doctoral fellow I mentor, Dr. Auriel Willette, we conducted a systematic review of neuroimaging studies looking at associations of obesity and brain volume across the age-span. We found that increased obesity is associated with atrophy mainly in the frontal lobes. Our review was published by the journal "Aging Research Reviews". I conducted an fMRI study using a novel methodology for effective connectivity analysis to fMRI data based on the principles of Granger causality. This methodology allows us to uncover causal pathways between nodes of a brain network. I implemented this analysis in a dataset from a fMRI study on religious beliefs conducted in 2009. The findings support the theory that areas implicated in Theory of Mind are key to generation of religious beliefs. The study was published in the journal "Brain Connectivity". Finally, in collaboration with Drs. Chia and Egan, we completed the "RISE study", a multi-faceted randomized placebo-controlled cross-over clinical study of the effects of a CB1 receptor agonist and antagonist on peripheral metabolism, brain function and brain metabolic control. The study included a robust neuroimaging component including fMRI and MRS. We performed two activation-paradigm fMRI studies, one to discover brain correlates of cephalic insulin secretion and the effects of CB1 receptors, the second to assess the effects of CB1 receptors on food appetitiveness. The goal of the first study was to demonstrate a rise in insulin levels in response to food visual stimuli (cephalic insulin response) as a result of activation of certain brain areas (insula, anterior cingulate, hypothalamus, ventral tegmental area, etc). Moreover, given the presence of CB1 receptors in the candidate areas, we aimed to demonstrate a difference in their level of activation with CB1 agonists and antagonists. The goal of the second study was to demonstrate dissociable effects of CB1 receptor stimulation on food value (food choices) and salience (intensity of such choices). In addition, we performed a resting fMRI study to assess CB1 modulation of functional connectivity of the various brain networks. Finally, we performed MRS to assess CB1 modulation of brain metabolism (glucose, lactate) and neurotransmission (glutamate, GABA, glycine). We are currently in the process of analyzing and interpreting the data from these studies.