Data_Sheet_1_Identification of Natural Antisense Transcripts in Mouse Brain and Their Association With Autism Spectrum Disorder Risk Genes.XLSX

Genome-wide sequencing technologies have greatly contributed to our understanding of the genetic basis of neurodevelopmental disorders such as autism spectrum disorder (ASD). Interestingly, a number of ASD-related genes express natural antisense transcripts (NATs). In some cases, these NATs have been shown to play a regulatory role in sense strand gene expression and thus contribute to brain function. However, a detailed study examining the transcriptional relationship between ASD-related genes and their NAT partners is lacking. We performed strand-specific, deep RNA sequencing to profile expression of sense and antisense reads with a focus on 100 ASD-related genes in medial prefrontal cortex (mPFC) and striatum across mouse post-natal development (P7, P14, and P56). Using de novo transcriptome assembly, we generated a comprehensive long non-coding RNA (lncRNA) transcriptome. We conducted BLAST analyses to compare the resultant transcripts with the human genome and identified transcripts with high sequence similarity and coverage. We assembled 32861 de novo antisense transcripts mapped to 12182 genes, of which 1018 are annotated by Ensembl as lncRNA. We validated the expression of a subset of selected ASD-related transcripts by PCR, including Syngap1 and Cntnap2. Our analyses revealed that more than 70% (72/100) of the examined ASD-related genes have one or more expressed antisense transcripts, suggesting more ASD-related genes than previously thought could be subject to NAT-mediated regulation in mice. We found that expression levels of antisense contigs were mostly positively correlated with their cognate coding sense strand RNA transcripts across developmental age. A small fraction of the examined transcripts showed brain region specific enrichment, indicating possible circuit-specific roles. Our BLAST analyses identified 110 of 271 ASD-related de novo transcripts with >90% identity to the human genome at >90% coverage. These findings, which include an assembled de novo antisense transcriptome, contribute to the understanding of NAT regulation of ASD-related genes in mice and can guide NAT-mediated gene regulation strategies in preclinical investigations toward the ultimate goal of developing novel therapeutic targets for ASD.

全基因组测序技术极大推动了我们对自闭症谱系障碍（Autism Spectrum Disorder, ASD）等神经发育障碍遗传基础的认知。有趣的是，多个与自闭症谱系障碍相关的基因可表达天然反义转录本（natural antisense transcripts, NATs）。在部分案例中，这类天然反义转录本已被证实可对有义链基因的表达发挥调控作用，进而参与大脑功能的维持。然而，目前仍缺乏针对自闭症谱系障碍相关基因与其天然反义转录本伙伴之间转录调控关系的系统性详细研究。本研究采用链特异性深度RNA测序技术，对小鼠出生后发育阶段（出生后第7天、第14天及第56天）内侧前额叶皮层（medial prefrontal cortex, mPFC）与纹状体中的100个自闭症谱系障碍相关基因的有义链与反义链测序读段表达谱进行分析。我们通过从头转录组组装技术，构建了一套全面的长链非编码RNA（long non-coding RNA, lncRNA）转录组。通过BLAST序列比对分析，将新组装得到的转录本与人类基因组进行比对，筛选出序列相似度与覆盖度均较高的转录本。本次组装共得到32861条从头反义转录本，对应12182个基因，其中1018条被Ensembl数据库注释为长链非编码RNA。我们通过聚合酶链式反应（Polymerase Chain Reaction, PCR）验证了部分筛选出的自闭症谱系障碍相关转录本的表达情况，其中包括Syngap1与Cntnap2基因。分析结果显示，本次检测的100个自闭症谱系障碍相关基因中，超过70%（72/100）存在至少一条可表达的反义转录本，这表明小鼠体内受天然反义转录本介导调控的自闭症谱系障碍相关基因数量较此前认知更多。我们还发现，在整个发育阶段中，反义重叠群的表达水平大多与其同源编码有义链RNA转录本的表达水平呈正相关。另有少量被检测的转录本呈现出脑区特异性富集现象，提示其可能在特定神经环路中发挥功能。本次BLAST比对分析还发现，271条自闭症谱系障碍相关的从头转录本中，有110条与人类基因组的序列相似度超过90%且覆盖度超过90%。本研究构建的从头反义转录组及上述研究结果，有助于深入解析小鼠体内天然反义转录本对自闭症谱系障碍相关基因的调控机制，同时可为临床前研究中基于天然反义转录本的基因调控策略提供指导，最终助力自闭症谱系障碍新型治疗靶点的开发。