Brainnetome Atlas Of Parahippocampal Region

The parahippocampal region(PHR) is an interface region between the hippocampus and the neocortex located in the medial temporal lobe(MTL). The PHR has been implicated in many functions, including long-term memory, working memory, and perception. Given these important and complex functions of the PHR, accurately identification of the PHR sub-regions in vivo human brains is one of the major obstacles in understanding human PHR function.However, there is a lack of fine-grained and accurate PHR atlas based on the neuroimaging data. Traditional brain atlases only have rough description of this area(such as AAL, FreeSurfer DK atlas et al.), and the major drawbacks are that these atlases are not consistent with cytoarchitectonic studies. Therefore, current studies on PHR function usually used guidelines for distinguishing the PHR sub-regions is based on visible landmarks in MRI images, based on cytoarchitectonic studies. The traditionally accepted cytoarchitectonic parcellation scheme for the PHR identified three sub-regions: entorhinal cortex(ERC), perirhinal cortex(PRC), and parahippocampal cortex(PHC). However, this anatomical description of the human PHR is coarse, so functional heterogeneity may exist within the PRC or PHC. Thus, we used a well-accepted anatomical connection-based parcellation method to obtain a well-defined parcellation of the human PHR, which can be used in the neuroimaging data.The PHR is mainly responsible for processing memory, the brain region that is most affected by neurological disease, especially Alzheimer’s Disease(AD). Thus, the T1-weighted structural MRI and resting-state functional MRI(rs-fMRI) data are used to explore the structural and functional abnormalities of each subregion of PHR in AD, and Mild Cognitive Impairment(MCI). In addition, many previous studies have reported that the APOE ε4 allele is the most significant genetic risk factor of AD. Therefore, the current study assesses the main effect of APOE ε4 allele on the functional network of each subregion of PHR. The main achievements of this study are as follows:1. By applying a well-accepted anatomical connection-based parcellation method to DTI data, we consistently identified four major subregions in the PHR in two independent spatial resolutions and population datasets. Based on cross-species similarities in their topological distribution and the names of PHR subregion of monkey brain, these 4 subregions were named as: the rostral perirhinal cortex(PRCr), caudal perirhinal cortex(PRCc), parahippocampal cortex(PHC), and entorhinal cortex(ERC). Especially, PRCr, PRCc, and PHC three subregions constitute the traditional PHR subregion parahippocampal gyrus(PHG). The PHR brainnetome atlas for the first time obtained a well-defined parcellation of the human PHR, which can be used on the neuroimaging data. Furthermore, we use the DTI and rs-fMRI data to explore the connectivity profiles of each PHR subregion. The anatomical and functional connectivity profiles for each subregion showed that the PRCr as a core component of an anterior temporal(AT) memory system, the PHC as a core component of a posterior medial(PM) memory system, and PRCc integrated the AT-PM memory systems. And also, we founded that ERC is an important information channel between hippocampus and the three subregions of PHG. These results may expand our understanding of the medial temporal lobe memory system and the two cortical systems for memory guided behaviour.2. We use T1-weighted structural MRI and rs-fMRI data to explore the grey matter atrophy and disease related functional connectivity pattern for each PHR subregion in AD/MCI datasets from two sites. The result showed that the grey matter volume decreased progressively from normal aging subject to MCI, and to AD in ERC and PRCr. These results are consistent with previous histopathological findings, which showed that the two subregions are the first several pathologically affected brain regions in AD. In addition, we founded that there is no significant difference in grey matter volume between healthy aging population and AD. These findings indicated the heterogeneity of the changes of each PHR subregion in AD/MCI. Furthermore, the analysis of functional connectivity pattern of each PHR subregion showed that the connected brain regions of the default mode network were mostly affected by AD. In particular, we found that the three subregions, i.e., PRCr, PRCc, and PHC, connected with different posterior cingulate subregions, showing altered connectivity strength in AD/MCI. These results are consistent with a model that the posterior cingulate gyrus is the echo of the brain and each posterior cingulate subregion corresponds to the different large-scale brain network. Finally, we observed abnormal grey matter volume and functional connectivity pattern of the whole PHR as ROI and also each PHR subregion affected by AD/MCI data. We found that both grey matter volume and functional connectivity of the whole PHR are the integrated the results from each PHR subregion, but did not reflect the difference between each PHR subregion results. These results demonstrate necessity of parcellating big brain region into fine-grained subregions, and the significance of our parcellation scheme for the PHR in the disease research.3. We used rs-fMRI data to detect the functional connectivity profiles of each PHR subregion affected by APOE ε4 allele. We found that the APOE ε4 allele not only impacted the functional connectivity between the PRCr and AT system brain regions, but also impacted the functional connectivity between the PHC and PM system brain regions. Interestingly, APOE ε4 allele is linked with the functional connectivity between the PRCc and brain regions of the AT or PM systems. These results verified our proposed model that the PRCc integrates the AT-PM memory system by using the method of imaging genetics.