Winden Kellen D, Oldham Michael C, Mirnics Karoly, Ebert Philip J, Swan Christo H, Levitt Pat, Rubenstein John L, Horvath Steve, Geschwind Daniel H
Interdepartmental Program for Neuroscience, University of California Los Angeles, Los Angeles, CA, USA.
Mol Syst Biol. 2009;5:291. doi: 10.1038/msb.2009.46. Epub 2009 Jul 28.
Genome-wide expression profiling has aided the understanding of the molecular basis of neuronal diversity, but achieving broad functional insight remains a considerable challenge. Here, we perform the first systems-level analysis of microarray data from single neuronal populations using weighted gene co-expression network analysis to examine how neuronal transcriptome organization relates to neuronal function and diversity. We systematically validate network predictions using published proteomic and genomic data. Several network modules of co-expressed genes correspond to interneuron development programs, in which the hub genes are known to be critical for interneuron specification. Other co-expression modules relate to fundamental cellular functions, such as energy production, firing rate, trafficking, and synapses, suggesting that fundamental aspects of neuronal diversity are produced by quantitative variation in basic metabolic processes. We identify two transcriptionally distinct mitochondrial modules and demonstrate that one corresponds to mitochondria enriched in neuronal processes and synapses, whereas the other represents a population restricted to the soma. Finally, we show that galectin-1 is a new interneuron marker, and we validate network predictions in vivo using Rgs4 and Dlx1/2 knockout mice. These analyses provide a basis for understanding how specific aspects of neuronal phenotypic diversity are organized at the transcriptional level.
全基因组表达谱分析有助于理解神经元多样性的分子基础，但要获得广泛的功能见解仍然是一项巨大的挑战。在这里，我们使用加权基因共表达网络分析对来自单个神经元群体的微阵列数据进行了首次系统水平分析，以研究神经元转录组组织与神经元功能和多样性之间的关系。我们使用已发表的蛋白质组学和基因组数据系统地验证网络预测。几个共表达基因的网络模块对应于中间神经元发育程序，其中已知枢纽基因对中间神经元的特化至关重要。其他共表达模块与基本细胞功能相关，如能量产生、放电频率、运输和突触，这表明神经元多样性的基本方面是由基本代谢过程的定量变化产生的。我们鉴定出两个转录上不同的线粒体模块，并证明其中一个对应于富含神经元突起和突触的线粒体，而另一个代表仅限于胞体的群体。最后，我们表明半乳糖凝集素-1是一种新的中间神经元标记物，并且我们使用Rgs4和Dlx1/2基因敲除小鼠在体内验证了网络预测。这些分析为理解神经元表型多样性的特定方面如何在转录水平上组织提供了基础。
