Single-Cell RNA-Seq of Mouse Olfactory Bulb Reveals Cellular Heterogeneity and Activity-Dependent Molecular Census of Adult-Born Neurons
题目：小鼠嗅球的单细胞rna-seq揭示了细胞的异质性和成体神经元活性依赖的分子普查
作者及单位：
Burak Tepe, Matthew C. Hill, Brandon T. Pekarek, Patrick J. Hunt, Thomas J. Martin, James F. Martin, Benjamin R. Arenkiel
1 Program in Developmental Biology, Baylor College of Medicine, Houston, TX 77030, USA
2 Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX 77030, USA
发表刊物及时间：
December 4, 2018, Cell Reports Vol. 25, Issue 10, p2689–2703.e3
Highlights
- •Single-cell sequencing reveals cellular heterogeneity in the mouse olfactory bulb
- •Differential gene expression uncovers selective markers for cell types
- •Pseudotemporal ordering of adult-born neurons reveals developmentally governed genes
- •Olfactory experience changes the cellular composition of olfactory bulb circuits
亮点：
•单细胞测序揭示了小鼠嗅球的细胞异质性
•差异基因表达揭示细胞类型的选择性标记
•从出生到成体的神经元的伪时间排序揭示了发育上控制的基因
•嗅觉经历改变了嗅球回路的细胞组成
Summary
Cellular heterogeneity within the mammalian brain poses a challenge toward understanding its complex functions. Within the olfactory bulb, odor information is processed by subtypes of inhibitory interneurons whose heterogeneity and functionality are influenced by ongoing adult neurogenesis. To investigate this cellular heterogeneity and better understand adult-born neuron development, we utilized single-cell RNA sequencing and computational modeling to reveal diverse and transcriptionally distinct neuronal and nonneuronal cell types. We also analyzed molecular changes during adult-born interneuron maturation and uncovered developmental programs within their gene expression profiles. Finally, we identified that distinct neuronal subtypes are differentially affected by sensory experience. Together, these data provide a transcriptome-based foundation for investigating subtype-specific neuronal function in the olfactory bulb (OB), charting the molecular profiles that arise during the maturation and integration of adult-born neurons and how they dynamically change in an activity-dependent manner.
摘要
哺乳动物脑细胞的异质性对于理解其复杂功能来说一直是一种 挑战， 在嗅球中， 气味信息由抑制性中间神经元的亚型处理， 其异质 性和功能被不断形成的成体神经发生所影响。 为了研究其细胞异质性以及 更好的理解 adult-born 神经元发育过程， 我们利用单细胞 RNA 测序 和计算模型来揭示不同的和转录不同的神经元和非神经元细胞类型。 我们还分析了 adult-born 中间神经元成熟期间的分子变化以及在其基 因表达谱中发现的发育程序。 最后， 我们发现不同的神经元亚型受感 觉经验的不同影响。 总之， 这些数据提供了基于转录组的基础， 用于 研究嗅球（OB） 中的亚型特异性神经元功能， 绘制成人出生神经元 的成熟和整合期间出现的分子谱以及它们如何以活动依赖性方式动 态变化。
图表选析
Single-Cell Sequencing Establishes a Molecular Census of Olfactory Bulb Cells To elucidate the overall cellular heterogeneity and activity dependent changes in olfactory bulb composition, we profiled the transcriptomes of 51,246 single cells collected from the olfactory bulbs of wild-type adult mice (Figure 1A). Mice were naive, olfactory deprived through naris occlusion, or olfactory enriched through training on an olfactory-discrimination learning paradigm. To block olfactory sensory input, we performed unilateral naris occlusion, with the occluded side serving as the sensory-deprived sample and the open side as the control (Najbauer and Leon, 1995, Quast et al., 2017, Yamaguchi and Mori, 2005). Mice were trained to discriminate various odorants using an olfactory-cued learning paradigm (Liu et al., 2017, Liu et al., 2018). This form of olfactory training exposed mice to several different odorants while also actively engaging the olfactory system to facilitate olfactory-discrimination learning. Cells from naive, olfactory-deprived, and enriched mice were clustered together after single-cell sequencing based on similarities in their transcriptional profiles using an unsupervised principal-component analysis (Macosko et al., 2015) and visualized using t-distributed stochastic neighbor embedding (t-SNE) (Van der Maaten, 2014, Van Der Maaten and Hinton, 2008) (Figure 1B). We identified 38 distinct cellular clusters, each composed of cells from different olfactory experience paradigms, indicating that experimental conditions did not bias the identity of the clusters (Figure S1). Using the expression patterns of cluster-enriched genes, we next assigned identities to each cluster. In total, we observed 16 neuronal (Syt1+/Tubb3+), three astrocytic (Gfap+), five olfactory ensheathing cell-based (Sox10+), six hematopoietic (all Aif1+; three Siglich+ microglia, one CD52+ macrophage, one CD74+ monocyte, and one Hba-a1+ red blood cell), four blood-vessel-based (two Slco1c1+ endothelial and two Pdgfrb+ mural), one oligodendrocyte-precursor-based (Olig2+), one myelinating-oligodendrocyte-based (Mag+), and two mesenchymal clusters (Figures 1B and 1D). Together, these data reveal the overall transcriptional heterogeneity of the cell types that comprise the mammalian olfactory bulb and identify molecular markers to further investigate the diversity and function of olfactory bulb cell types. Transcriptome-Based Clustering of Neurons Identifies Markers of Neuronal Subtypes To assign specific identities to neuronal subtypes within the olfactory bulb, we filtered and subclustered neurons identified from initial clustering (Figure 1B, Neuron01–Neuron16). Neurons were initially identified utilizing gene enrichment data (Figure 1C; Table S1) by selective expression of known neuronal markers such as Syt1 and Tubb3.
单细胞测序建立了嗅球细胞的分子普查 为了阐明嗅球组成的整体细胞异质性和活性依赖性变化，我们分 析了从野生型成年小鼠的嗅球中收集的 51,246 个单细胞的转录 组（图 1A）。通过鼻窦阻塞或通过嗅觉辨别学习范例的训练丰富 嗅觉，小鼠是缺乏嗅觉经历的； 通过单侧鼻孔闭塞导致嗅觉剥夺； 或者通过嗅觉识别学习模型的训练，丰富了嗅觉。为了阻断嗅觉 感官的输入，我们进行了单侧鼻孔闭塞，闭塞侧作为感觉缺失的 样本，开放侧作为对照(Najbauer and Leon, 1995, Quast et al., 2017, Yamaguchi and Mori, 2005)。 训练小鼠区分各种气味使用嗅觉暗 示的学习模型(Liu et al., 2017, Liu et al., 2018)。这种形式的嗅觉 训练使小鼠暴露于几种不同的气味，同时还积极地参与嗅觉系统 以促进嗅觉辨别学习。 来自嗅觉无经验，嗅觉剥夺和丰富嗅觉经 验小鼠的细胞在单细胞测序后基于其转录谱中的相似性，使用无 监督的主成分分析聚集在一起(Macosko et al., 2015)，并使用 t 分 布随机邻域嵌入（t-SNE）可视化(Van der Maaten, 2014, Van Der Maaten and Hinton, 2008)（图 1B）。 我们鉴定了 38 个不同的细胞 簇，每个由来自不同嗅觉经验模型的细胞组成，表明实验条件对 聚类的识别没有偏倚（图 S1）。接下来，我们使用簇富集的基因 的表达模式，为每个群集分配了身份。 总共，我们观察到 16 个 神经元（Syt1 + / Tubb3 +）， 3 个星形细胞（Gfap +）， 5 个嗅鞘细 胞（Sox10 +）， 6 个造血（所有都是 Aif1 + ; 3 个 Siglich+ 小胶质 细胞， 1 个 CD52+ 巨噬细胞，一个 CD74+ 单核细胞和一个 Hbaa1+ 红细胞），四个基于血管的（两个 Slco1c1+ 血管内皮和两个 Pdgfrb+ 外壁），一个基于少突胶质细胞前体（Olig2 +），一个基 于髓鞘-少突胶质细胞（Mag +）和两个间充质簇（图 1B 和 1D）。 总之，这些数据揭示了构成哺乳动物嗅球的细胞类型的整体转录 异质性，并鉴定分子标记以进一步研究嗅球细胞类型的多样性和 功能。 基于转录组的神经元聚类识别神经元亚型的标记 为了将特定身份分配给嗅球内的神经元亚型，我们过滤并从初始 聚类中鉴定出亚群集神经元（图 1B， Neuron01-Neuron16）。最初 通过选择性表达已知的神经元标志物如 Syt1 和 Tubb3，利用基 因富集数据（图 1C ; 表 S1）鉴定神经元。
Figure 1Single-Cell Transcriptome Analysis Delineates Mouse Olfactory Bulb Cellular Heterogeneity
图1. 单细胞转录组测序揭示小鼠嗅 球细胞异质性
(A) Schematic view of the experimental workflow.
(B) Cellular composition of the olfactory bulb was visualized using t-distributed stochastic neighbor embedding (t-SNE). Individual single-cell transcriptomes were colored according to cluster identity in (B)–(D).
(C) Dendrogram depicting hierarchical relationships between distinct cell populations. Heatmap illustrating the genes most highly enriched in each cluster, with each column representing a gene and each row representing average expression level of that gene in each cluster.
(D) Graph showing number of cells per cluster, number of unique molecular identifiers (UMIs) per cluster (mean ± SEM; scale is in thousands), and number of genes detected per cluster (mean ± SEM; scale is in thousands). Violin plots show expression of cell-type-specific marker genes for each cluster.
(A) 实验流程的示意图 (B) 用 t-SNE 可视化嗅球的细胞组成。 根据（B） –（D） 中的簇 同一性对每个单细胞转录本进行着色。 (C) 描述不同细胞群体之间的层次关系的树状图。 热图显示了每 个簇中富集程度高的基因， (D) 每一列代表一个基因， 每一行代表该基因在每个簇中的平均 表达水平。 (E) 显示每个簇的细胞数、 每个簇的独特分子标识符(UMI)的数 目(平均±SEM； 标度为千)的图表， 以及每簇检测到的基因（平均 ±SEM； 规模以千计） 。 小提琴图显示了每个簇的细胞类型特异性标 记基因的表达。
Figure 3Pseudo-Timeline Analysis Reveals Transcriptional Changes during Maturation and Integration along Distinct Developmental Axes
(A) Schematic sagittal view of the mouse olfactory system. Inset: summary diagram of olfactory bulb adult-neurogenesis illustrating broad morphological and developmental changes throughout maturation of adult-born neurons. LV, lateral ventricle; RMS, rostral migratory stream. 小鼠嗅觉系统 的示意图矢状位图。内嵌：嗅球成体神经发生简图，说明成年出生神经元在成熟过程中的广泛形态和发 育变化。左室，侧脑室；均方根，吻移流 2018/12/25 Untitled Document 7/16
(B) (Top left) Monocle2 pseudotime trajectory of adult-born neurons. Cells are colored by pseudotime score, with dark colors representing immature cell stages and light colors representing mature cell stages. (Top right) Monocle2 pseudotime trajectory of adult-born neurons with cells colored by cluster identity according to Figure 2. (Bottom) Adult-born neuron cluster density plot projected to the x axis of the bifurcating Monocle2 pseudotime trajectory, indicating which arm of the timeline each cell type is located. (左上角)成年神经元的单时间假时间轨迹。细胞用假时间记分染色， 深色代表不成熟的细胞阶段，浅色代表成熟的细胞期。(右上)按图2所示，带有团簇同一性的细胞的成年 出生神经元的单时间假时间轨迹。(底部)成年神经元团簇密度图投射到分叉单时间轨迹的x轴，指示每种 细胞类型的时间线的哪个臂。
(C) (Top) Axis A of Monocle2 pseudotime trajectory colored according to cluster identity. (Bottom) Axis A pseudotime trajectory colored by pseudotime score, with the dark color representing an immature cell stage and the light color representing a mature cell stage. (顶)轴a为单点2伪时轨迹， 按簇同一性着色。(下)轴用假时间分数着色的伪时间轨迹，暗颜色代表不成熟的细胞阶段，浅色代表成 熟的细胞期。
(D) Axis A: 4 distinct groups of pseudotime-dependent genes with dynamic expression patterns plotted across pseudotime as heatmaps, with blue indicating low levels and red indicating high levels of expression. (Middle) Gene expression trends for each gene (black) with the trend line highlighted in red. (Right) Top 6 enriched gene ontology (GO) terms for each temporal cluster. 轴a：4组具有动 态表达模式的假时相关基因作为热图，蓝色表示低水平，红色表示高表达水平。(中)每个基因(黑色)的 基因表达趋势，趋势线以红色突出显示。(右)前6个丰富基因本体论(GO)术语为每个时间簇。
(E) Differential expression patterns of one example gene from each group of genes along developmental axis A. 一例基因与每组基因沿发育轴a的差异表达模式。
(F) (Top) Axis B of Monocle2 pseudotime trajectory colored according to cluster identity. (Bottom) Axis B pseudotime trajectory colored by pseudotime score, with the dark color representing an immature cell stage and the light color representing a mature cell stage. (顶部)单点2伪时轨迹的轴b 按簇同一性着色。(下)轴b伪时间轨迹用伪时间记分着色，暗色代表不成熟细胞期，浅色代表成熟细胞 期。
(G) Axis B: 4 distinct groups of pseudotime-dependent genes, with dynamic expression patterns plotted across pseudotime as heatmaps. Blue indicates low levels and red indicates high levels of expression. (Middle) Gene expression trends for each gene (black), with the trend line highlighted in red. (Right) Top 6 enriched gene ontology (GO) terms for each temporal cluster. 轴b：4组不同的假 时间依赖基因，其动态表达模式以热图的形式绘制。蓝色代表低层次，红色代表高表达水平。(中)每个 基因(黑色)的基因表达趋势，趋势线以红色突出。(右)前6个丰富基因本体论(GO)术语为每个时间簇。
(H) Differential expression patterns of one example gene from each group of genes along developmental axis B. See also Tables S3 and S4. 一例基因与每组基因沿发育轴b的差异表达模式。 另见表S3和S4
Figure 5. Olfactory Activity Alters Adult-Born Interneuron Subtype Composition (A) Schematic view of olfactory bulb experimental procedures, which are detailed in methods. (B)Two-dimensionalt-SNE representation of 16, 302 adult-born interneurons colored according to experimental group. OC, narisoccluded; TR, olfactory trained; WT, wild-type. (C) t-SNE representation of 16,302 adult-born interneurons colored according to cluster identity. (D) Shifts in adult-born neuron cluster composition for each indicated experimental condition plotted across expression-based t-SNE (left and middle) and along the Monocle2 pseudotime trajectory (right). Pearson’s chi-square test residuals were calculated for the corresponding experimental group (left). Discrete values were determined from Pearson’s chi-square test (middle); if the p value was < 0.05, then clusters were assigned appropriate designation of increased or decreased (red or blue) based on their residual score. Clusters with no significant composition shift are highlighted in gray (no change, p value > 0.05).
图 5.嗅觉活动改变成人新生中间神经元亚型组成 (A)嗅球实验过程详细方法示意图。 (B)根据实验组着色的 16， 302 例成人新生中间神经元的二维 SNE 表示。 OC，麻醉闭塞； TR， 嗅觉训练； WT，野生型。 (C)依据聚类特征着色的 16， 302 个成人新生中间神经元的 t-SNE 表示。 (D)在基于表达的 t-SNE(左边和中间)上和单峰 2伪时间轨迹(右)旁绘制的每个指示实验条件的 成人新生神经元簇组成的变化。 计算相应实验组(左边)的皮尔逊(Pearson)卡方检验残值。 离 散值由 Pearson‘s 卡方检验(中间)确定，如果 p 值<0.05，则根据其残值给聚类分配适当的增 减值(红或蓝)。无明显组分变化的簇呈灰色(无变化， p 值>0.05)。
翻译小组：
李碧琪、王俊豪、陈志荣、黄敬潼、陈凯星、黄子亮、郑凌伶