Integrated Whole Transcriptome and DNA Methylation Analysis Identifies Gene Networks Specific to Late-Onset Alzheimer's Disease
Integrated Whole Transcriptome and DNA Methylation Analysis Identifies Gene Networks Specific to Late-Onset Alzheimer's Disease
复制标题
DOI：
10.3233/jad-141989
复制
发表时间：
2015-01-01
影响因子：
4
通讯作者：
Gilbert, John
中科院分区：
文献类型：
--
作者：
Humphries, Crystal E.;Kohli, Martin A.;Gilbert, John
先前的转录组研究观察到晚发性阿尔茨海默病（LOAD）的细胞过程中断，但尚不清楚这些变化是LOAD特异性的，还是一般神经退行性变的共同特征。在这项研究中，我们通过检查LOAD中的转录并将其与认知正常对照和一组“疾病对照”进行比较来解决这个问题。使用RNA-seq检查差异转录，这允许检查蛋白质编码基因，非编码rna和剪接。在五个基因中观察到LOAD特异性的显著转录差异：C10orf105， DIO2, a lincRNA， RARRES3和WIF1。这些发现在两个独立的公开的微阵列数据集中得到了重复。对LOAD中具有中等转录差异的2504个基因进行的网络分析显示，这些基因聚集成7个网络。参与髓鞘形成和先天免疫反应的两个网络与LOAD特异性相关。FRMD4B和ST18是髓鞘形成网络中的枢纽基因，先前与LOAD有关。在这5个重要基因中，WIF1和RARRES3直接参与髓鞘形成过程；另外三个基因位于网络中。LOAD特异性的DNA甲基化变化位于整个基因组中，并且在髓鞘形成网络中确定了甲基化的实质性变化。在整个基因组中观察到LOAD特有的剪接差异，并且在所有七个网络中都减少了。与两个对照组相比，DNA甲基化对髓鞘网络中LOAD内转录的影响降低。这些结果提示了LOAD的分子基础，并指出了该疾病特有的几个关键过程、基因和网络。
Previous transcriptome studies observed disrupted cellular processes in late-onset Alzheimer's disease (LOAD), yet it is unclear whether these changes are specific to LOAD, or are common to general neurodegeneration. In this study, we address this question by examining transcription in LOAD and comparing it to cognitively normal controls and a cohort of "disease controls." Differential transcription was examined using RNA-seq, which allows for the examination of protein coding genes, non-coding RNAs, and splicing. Significant transcription differences specific to LOAD were observed in five genes: C10orf105, DIO2, a lincRNA, RARRES3, and WIF1. These findings were replicated in two independent publicly available microarray data sets. Network analyses, performed on 2,504 genes with moderate transcription differences in LOAD, reveal that these genes aggregate into seven networks. Two networks involved in myelination and innate immune response specifically correlated to LOAD. FRMD4B and ST18, hub genes within the myelination network, were previously implicated in LOAD. Of the five significant genes, WIF1 and RARRES3 are directly implicated in the myelination process; the other three genes are located within the network. LOAD specific changes in DNA methylation were located throughout the genome and substantial changes in methylation were identified within the myelination network. Splicing differences specific to LOAD were observed across the genome and were decreased in all seven networks. DNA methylation had reduced influence on transcription within LOAD in the myelination network when compared to both controls. These results hint at the molecular underpinnings of LOAD and indicate several key processes, genes, and networks specific to the disease.