Understanding Hippocampal Neuroplasticity in Schizophrenia by Targeting Molecular Pathways Through Physical Exercise Training (PsychTrain)

有氧运动已被证明可以改善精神分裂症患者的认知、海马体积和功能结果。研究表明，改善可能是由遗传风险因素调节的神经可塑性脑再生介导的。然而，先前研究的一个局限性是，它没有确定运动与基因和特定神经可塑性机制之间的分子途径。我们将通过研究有氧运动的基因相关调节因子的影响来解决这一研究空白。本研究的主要结果将是通过全基因组关联研究评估的精神分裂症多基因风险评分（PRS）与进行3个月有氧运动干预后海马亚区CA4 /齿状回（DG）体积变化的相关性，与平衡和张力训练（对照条件）相比。之前，我们发现了有氧运动训练后PRS对CA4/DG分区体积变化影响的初步证据。次要结果将是突触相关通路、大脑结构和静息状态功能改变、症状、认知和表观遗传参数。我们将调查142名精神分裂症患者的样本。共有65名精神分裂症患者在bmbf资助的“通过创新治疗增强精神分裂症预防和康复”（ESPRIT）联盟的有氧运动研究中完成了为期3个月的训练。在这些患者中，34人接受了3-特斯拉多模态磁共振成像（MRI）扫描，并提供了血液用于遗传和分子生物标志物调查，并将纳入本研究。此外，为了进一步评估次要结果，ESPRIT研究中收集了31名仅具有临床表型的患者的血液。根据我们的功率计算，假设退出率为30%，我们还需要在计划的研究中再入组156例患者。我们将构建PRSs，并纳入介导突触可塑性的分子途径，正如我们最近的有氧运动研究所表明的，包括突触部分（GO: 0045202）和化学突触传递（GO: 0007268），并将它们与与突触过程无关的途径进行比较。我们将使用机器学习根据CA4/DG体积对患者进行分层，从而接近神经可塑性的复杂结构。此外，我们将确定PRS对结构MRI、弥散张量成像、静息状态功能MRI和脑电图参数的影响。丰富的多模态纵向成像数据将通过多变量技术进行分析，包括独立成分分析和先进的纵向建模技术。我们将利用干预前后获得的遗传、表观遗传、神经影像学、临床和认知数据来识别预测性多模态特征。

Aerobic exercise has been shown to improve cognition, hippocampal volume, and functional outcomes in individuals with schizophrenia. Research indicates that improvements may be mediated by neuroplastic brain regeneration moderated by genetic risk factors. However, a limitation of prior research is that it did not identify the molecular pathways linking exercise with genes and specific neuroplastic mechanisms. We will address this research gap by investigating the impact of genetics-related moderators of aerobic exercise. The primary outcome of our study will be the correlation of a schizophrenia polygenic risk score (PRS), assessed by genome-wide association studies, with the change in volume of the hippocampal subfield cornu ammonis 4 (CA4)/dentate gyrus (DG) after a 3-month aerobic exercise intervention compared with balance and tone training, the control condition. Previously, we found preliminary evidence for effects of PRS on volume change in the CA4/DG subregion after aerobic exercise training. Secondary outcomes will be synapse-related pathways, brain structural and resting-state functional alterations, symptoms, cognition, and epigenetic parameters. We will investigate a sample of 142 patients with schizophrenia. A total of 65 patients with schizophrenia have already completed the 3-month training in our aerobic exercise study of the BMBF-funded “Enhancing Schizophrenia Prevention and Recovery through Innovative Treatments” (ESPRIT) consortium. Of these patients, 34 underwent 3-Tesla multimodal magnetic resonance imaging (MRI) scans and provided blood for genetic and molecular biomarker investigations and will be included in this study. In addition, for additional assessment of secondary outcomes, blood has been collected from 31 patients of the ESPRIT study with only clinical phenotyping. According to our power calculation, assuming a drop-out rate of 30% we additionally need to enroll a further 156 patients in the planned study. We will construct PRSs and include molecular pathways that mediate synaptic plasticity, as indicated by our recent aerobic exercise studies, including Synapse Part (GO: 0045202) and Chemical Synaptic Transmission (GO: 0007268), and compare them with pathways not related to synaptic processes. We will use machine learning to stratify patients according to CA4/DG volume and thus approach the complex construct of neuroplasticity. Furthermore, we will determine the influence of PRS on structural MRI, diffusion tensor imaging, resting-state functional MRI, and electroencephalography parameters. Rich, multimodal longitudinal imaging data will be analyzed by multivariate techniques, including independent component analyses and advanced longitudinal modeling techniques. We will harness genetic, epigenetic, neuroimaging, clinical, and cognitive data acquired before and after the intervention to identify predictive multimodal signatures.