Specific cell populations may have unique contributions to schizophrenia but may be missed in studies of homogenate tissue. Here laser capture microdissection followed by RNA sequencing (LCM-seq) was used to transcriptomically profile the granule cell layer of the dentate gyrus (DG-GCL) in human hippocampus and contrast these data to those obtained from bulk hippocampal homogenate. We identified widespread cell-type-enriched aging and genetic effects in the DG-GCL that were either absent or directionally discordant in bulk hippocampus data. Of the ~9 million expression quantitative trait loci identified in the DG-GCL, 15% were not detected in bulk hippocampus, including 15 schizophrenia risk variants. We created transcriptome-wide association study genetic weights from the DG-GCL, which identified many schizophrenia-associated genetic signals not found in transcriptome-wide association studies from bulk hippocampus, including GRM3 and CACNA1C. These results highlight the improved biological resolution provided by targeted sampling strategies like LCM and complement homogenate and single-nucleus approaches in human brain.

特定的细胞群可能对精神分裂症有独特的贡献，但在匀浆组织的研究中可能会被遗漏。在这里，激光捕获显微切割和 RNA 测序 (LCM-seq) 用于转录组分析人类海马齿状回 (DG-GCL) 的颗粒细胞层，并将这些数据与从大量海马匀浆中获得的数据进行对比。我们在 DG-GCL 中发现了广泛的细胞类型丰富的衰老和遗传效应，这些效应在海马体数据中不存在或方向不一致。在 DG-GCL 中鉴定的约 900 万个表达数量性状基因座中，15% 未在海马体中检测到，包括 15 种精神分裂症风险变异体。我们从 DG-GCL 创建了全转录组关联研究遗传权重，GRM3和CACNA1C。这些结果突出了目标采样策略（如 LCM）以及人脑中的补体匀浆和单核方法提供的改进的生物分辨率。