Report

Cell types in the mouse cortex and hippocampus revealed by single-cell RNA-seq

See allHide authors and affiliations

Science  06 Mar 2015:
Vol. 347, Issue 6226, pp. 1138-1142
DOI: 10.1126/science.aaa1934

Cellular diversity in the brain revealed

The mammalian brain has an extraordinarily large number of cells. Although there are quite a few different cell types, many cells in any one category tend to look alike. Zeisel et al. analyzed the transcriptomes of mouse brain cells to reveal more than meets the eye. Interneurons of similar type were found in dissimilar regions of the brain. Oligodendrocytes that seemed to be all of one class were differentiated by their molecular signatures into a half-dozen classes. Microglia associated with blood vessels were distinguished from look-alike perivascular macrophages. Thus, the complex microanatomy of the brain can be revealed by the RNAs expressed in its cells.

Science, this issue p. 1138

Abstract

The mammalian cerebral cortex supports cognitive functions such as sensorimotor integration, memory, and social behaviors. Normal brain function relies on a diverse set of differentiated cell types, including neurons, glia, and vasculature. Here, we have used large-scale single-cell RNA sequencing (RNA-seq) to classify cells in the mouse somatosensory cortex and hippocampal CA1 region. We found 47 molecularly distinct subclasses, comprising all known major cell types in the cortex. We identified numerous marker genes, which allowed alignment with known cell types, morphology, and location. We found a layer I interneuron expressing Pax6 and a distinct postmitotic oligodendrocyte subclass marked by Itpr2. Across the diversity of cortical cell types, transcription factors formed a complex, layered regulatory code, suggesting a mechanism for the maintenance of adult cell type identity.

The brain is built from a large number of specialized cell types, enabling highly refined electrophysiological behavior, as well as fulfilling brain nutrient needs and defense against pathogens. Functional specialization allows fine-tuning of circuit dynamics and decoupling of support functions such as energy supply, waste removal, and immune defense. Cells in the nervous system have historically been classified using location, morphology, target specificity, and electrophysiological characteristics, often combined with molecular markers (15). Systematic in situ hybridization has revealed extensive regional heterogeneity (6). However, none of these properties carry enough information to result, in every case, in a definitive cell type identification (7). Single-cell RNA sequencing (RNA-seq) has been used to classify cells in spleen (8), lung epithelium (9), and embryonic brain (10). However, the adult nervous system has greater complexity and more cell types, presenting a challenge both to sample preparation methods and computational analysis.

Here, we have used quantitative single-cell RNA-seq (11) to perform a molecular census of the primary somatosensory cortex (S1) and the hippocampal CA1 region, based on 3005 single-cell transcriptomes (Fig. 1A and fig. S1, A to C). Individual RNA molecules were counted using unique molecular identifiers (UMIs) (essentially tags that identify individual molecules) (12) (figs. S1, D to J, and S2, A to E) and confirmed by single-molecule RNA fluorescence in situ hybridization (FISH) (fig. S2, G to I).

Fig. 1 Molecular census of somatosensory S1 cortex and hippocampus CA1 by unbiased sampling and single-cell RNA-seq.

(A) Workflow for obtaining and analyzing single-cell RNA-seq from juvenile mouse cortical cells, from dissection to single-cell RNA-seq and biclustering. (B) Visualization of nine major classes of cells using t-distributed stochastic neighbor embedding (tSNE). Each dot is a single cell, and cells are laid out to show similarities. Colored contours correspond to the nine clusters in (A) and fig. S3. Expression of known markers is shown using the same layout (blue, no expression; white, 1% quantile; red, 99% quantile). (C) Hierarchical clustering analysis on 47 subclasses. Bar plots show number of captured cells in CA1 and S1, number of detected polyA+ RNA molecules per cell, and total number of genes detected per cell.

We used clustering to discover molecularly distinct classes of cells. Standard hierarchical clustering resulted in fragmented clusters (fig. S4), because most genes were not informative in most pairwise comparisons and contributed at best only noise. Biclustering can overcome this problem by simultaneously clustering genes and cells. We developed BackSPIN (see the supplementary materials), a divisive biclustering method based on sorting points into neighborhoods (SPIN) (13), which revealed nine major classes of cells: S1 and CA1 pyramidal neurons, interneurons, oligodendrocytes, astrocytes, microglia, vascular endothelial cells, mural cells (that is, pericytes and vascular smooth muscle cells), and ependymal cells (Fig. 1, A and B, and fig. S3).

The data set allowed us to identify the most specific markers for each class, many of which are known to play a functional role in these cells (fig. S5). S1 pyramidal cells were marked by Tbr1, a transcription factor required for the final differentiation of cortical projection neurons; oligodendrocytes by Hapln2, encoding a protein required for proper formation of nodes of Ranvier; mural cells by Acta2, a key component of actin thin filaments; and endothelial cells by Ly6c1 [expressed by monocytes peripherally, and endothelial cells in the brain (14)]. Some were novel, such as Gm11549 (a long noncoding RNA specific to S1 pyramidal neurons), Spink8 (a serine protease inhibitor specific to hippocampal pyramidal cells), and Pnoc (prepronociceptin, here identified as an interneuron marker).

By repeating biclustering on each of the nine major classes (Fig. 1C and figs. S5 to S8), we identified a total of 47 molecularly distinct subclasses of cells. Every subclass was detected in multiple mice (fig. S1K), arguing that cell identity was preserved across these genetically outbred (CD-1) mice. Neurons contained more RNA than glia and vascular cells and a larger number of detectable genes (Fig. 1C and fig. S1E). Mitochondrial mRNAs were less variable, although mitochondrial tRNAs were highly specifically enriched in endothelial cells (fig. S1E).

We identified seven subclasses of S1 pyramidal cells (Fig. 2A and figs. S6A and S7), which were largely layer-specific. The superficial layers II/III and IV were represented by single populations, whereas layer V showed two distinct subclasses. Layers VI and VIb were represented by single populations, but in addition we found a subclass lacking specific markers but expressing common deep-layer markers such as Pcp4. A distinct subclass expressed Synpr and Nr4a2, which are abundant in the adjacent claustrum, with some cells extending into S1.

Fig. 2 Neuron subclasses in the somatosensory cortex.

(A) Subclasses of pyramidal neurons in the somatosensory cortex (S1) identified by BackSPIN clustering. Bar plots show mean expression of selected known and novel markers (error bars show standard deviations). Layer-specific expression shown by in situ hybridization (Allen Brain Atlas). S1PyrL23, layer II-III; S1PyrL4, layer IV; S1PyrL5a, layer Va; S1PyrL5, layer V; S1PyrL6, layer VI; S1PyrL6b, layer VIb; S1PyrDL, deep layers; ClauPyr, claustrum. (B) Identification of interneuron subclasses. Bar plots show selected known and novel markers. Fraction of S1/CA1 cells is depicted at bottom: blue, S1; yellow, CA1; white, flow-sorted Htr3a+ cells from S1. (C) Immunohistochemistry demonstrating the existence and localization of novel PAX6+/5HT3aEGFP+ interneurons, Int11. Bar plots show the layer distribution of these neurons. (D) Intrinsic electrophysiology and morphology of PAX6+ interneurons in S1 layer I, identified by post hoc staining.

We found two types of CA1 glutamatergic cells (fig. S8), plus cells derived from the adjacent CA2 (as defined by Pcp4) and subiculum (as defined by Ly6g6e). Genes highly expressed in type 2 CA1 pyramidal neurons were associated with mitochondrial function (fig. S8), which has been shown to correlate with the firing rate and length of projections in cortical neurons (15). Orthogonal to the two main classes, we found CA1 layer–specific markers (i.e., Calb1 and Nov), as well as dorsoventrally patterned genes (i.e., Wfs1 and Grp) (16), in both of the two main types of CA1 cells. These may correspond to functional differences between layers (17).

We found 16 subclasses of interneurons (Fig. 2B and fig. S6, C and D), but there are likely more subclasses because we achieved only shallow sampling of Sst- and Pvalb-expressing cells. In superficial layers of S1, we identified an Htr3a- and Pax6-expressing interneuron subclass, confirmed by immunohistochemistry (Fig. 2C) [13.9 ± 2.4% of serotonin (5HT) receptor 3a-enhanced green fluorescent protein (5HT3aEGFP) cells in layer I, n = 4 mice, 636 cells analyzed]. These interneurons specifically expressed Myh8, Fut9, and Manea. In whole-cell current clamp recordings of layer I neurons, subsequently stained for PAX6, these cells exhibited intrinsic electrophysiological and morphological characteristics of late-spiking neurogliaform cells (6 PAX6+ out of 40 recorded cells) (Fig. 2D and fig. S6E). Pax6 is not expressed in the ventral forebrain during development, further suggesting that neurogliaform cells are developmentally heterogeneous (18).

CA1 and S1 regions both contained interneurons of almost every subclass (Fig. 2B), showing that interneurons residing in functionally distinct cortical structures are transcriptionally closely related. The two exceptions were cells expressing Vip, Penk, Calb2, and Crh (which were confined to S1) and cells expressing Lhx6, Reln, and Gabrd [which were confined to CA1 and may be medial ganglionic eminence–derived Ivy cells and neurogliaform cells (18)].

Astrocytes formed two subclasses (Fig. 3A and fig. S9A) distinguished by differential expression of Gfap (type 1) and Mfge8 (type 2). Immunostaining showed that type 1 astrocytes were derived from layer I, particularly from the glia limitans, a thin layer made up mostly of astrocytes that is arranged against the pia (Fig. 3B). In contrast, type 2 astrocytes were more uniformly distributed in the cortex and were smaller and less ramified.

Fig. 3 Characterization of glial subclasses.

(A) Two types of astrocytes (Astro1 and Astro2) identified by common and distinct markers. (B) Immunohistochemistry for glial fibrillary acidic protein (red, Astro1) and MFGE8 (green, Astro2). Scale bar, 50 μm. (C) Genes showing expression restricted to microglia (Mgl), perivascular macrophages (Pvm), and peritoneal macrophages (Pmac). Error bars show standard deviations. (D) Cartoon illustrating the morphology and localization of microglia and perivascular macrophages. (E) Immunostaining for AIF1 (previously known as Iba-1, blue) marking microglia, and for MRC1 (green) and LYVE1 (red) marking perivascular macrophages. Asterisk, a microglia cell. Arrow, a perivascular macrophage aligned to a vessel (not stained). Scale bar, 20 μm. (F) Heat map showing progressive changes in gene expression along oligodendrocyte differentiation, illustrated below. (G) Single-molecule RNA FISH for Itpr2 and Cnksr3 mark a strict subset of oligodendrocytes (as identified by Plp1). Scale bar, 11 μm.

We identified two types of immune cells: microglia (the tissue-resident macrophages of the brain) and perivascular macrophages. Although closely related, these cell types have distinct developmental origin (19). Both expressed brain macrophage markers Aif1 and Cx3cr1, whereas perivascular macrophages were distinguished by expression of Mrc1 and Lyve1, characteristic of pro-angiogenic perivascular type 2 macrophages (20). Immunohistochemistry for the corresponding proteins confirmed that microglia (AIF1+/LYVE1/MRC1) had a classical, ramified morphology and were located throughout the cortex (Fig. 3, D and E). In contrast, perivascular macrophages (AIF+/LYVE1+/MRC1+) were located only along vessels and showed an ameboid morphology. They were distinct from mural and endothelial cells (fig. S10). Comparison with peritoneal macrophages confirmed their identity (fig. S9A). The correlation between brain and peripheral macrophages (0.67) was similar to that between neurons and glia (0.62), underscoring the functional divergence of this immune cell class.

Six subpopulations of oligodendrocytes were identified (Fig. 3F and fig. S9C), likely representing stages of maturation: immature (Oligo4), premyelinating (Oligo2), myelinating (Oligo5), and terminally differentiated postmyelination (Oligo6) oligodendrocytes. An intermediate population, Oligo3, was almost exclusively observed in somatosensory cortex and may represent a distinct cellular state specific for this tissue. The subclass Oligo1, which did not express the prototypical genes associated with oligodendrocyte precursor cells (OPCs), may represent a postmitotic cellular state, associated with the first steps of oligodendrocyte differentiation. Oligo1 cells expressed a distinct set of genes, including Itpr2, Prom1, Gpr17, Tcf7l2, 9630013A20Rik, Idh1, Cnksr3, and Rnf122. Single-molecule RNA FISH confirmed that Itpr2 and Cnksr3 were expressed in strict subsets of cells expressing Plp1, a pan-oligodendrocyte marker (4.5% and 7.5%, respectively) (Fig. 3G). Together, the Oligo1 to Oligo6 populations may represent sequential steps in the process of maturation from an OPC to a terminally differentiated oligodendrocyte.

Across this diverse set of cell types, we found many transcription factors with highly restricted expression patterns (Fig. 4A and supplementary materials). For example, interneurons expressed key interneuron regulators Dlx1, Dlx2, Dlx5, and Arx, and pyramidal layer II/III neurons expressed Neurog2, which can directly reprogram human embryonic stem cells to excitatory neurons of layer II/III phenotype with near 100% efficiency (21). Lyl1 and Spic were specific to perivascular macrophages; Spic is essential for the maintenance of red pulp macrophages (22), suggesting that it may play a similar role in brain perivascular macrophages.

Fig. 4 Expression of regulatory genes across 47 subclasses.

(A) Transcription factors showing restricted expression across cell types. Asterisks denote genes with additional expression in distinct subclasses: Sp8 in Int11, Msx1 in vascular cells and microglia. (B) Genes specific to ependymal cells. Transcription factors Foxj1, Rfx2, and Rfx3 (with asterisk to indicate its wider expression) and their known targets are shown in red, green, and blue, respectively. Arrows indicate known direct interactions between transcription factors. Only genes with known ciliary function are included.

Expanding this analysis to all genes, we found extensive functional specialization between cellular subclasses. Ependymal cells (multiciliated cells lining the ventricles) expressed the largest set of subclass-restricted genes, including transcription factors Foxj1, Myb, and Rfx2, the master regulators of motile ciliogenesis (23) (24), and Zmynd10, which causes ciliopathy when mutated in humans (25). Nearly every structural component of cilia was also represented (Fig. 4B), including the 2+9 microtubule core and radial spokes, the dynein and kinesin motors, the filamentous shell, the basal body that anchors cilia to the cytoplasm, and two adenylate kinases (Ak7 and Ak8) that generate adenosine triphosphate energy supporting cilia motility. Many of these structural genes are directly regulated by Foxj1, Rfx2, or Rfx3 (23, 26) (Fig. 4B).

In summary, our findings reveal the diversity of brain cell types and transcriptomes. Across the full set of cell types, transcription factors formed a complex, layered regulatory code, suggesting a plausible mechanism for the maintenance of adult differentiated cell types. More broadly, these results showcase the power of explorative single-cell RNA-seq and point the way toward future whole-brain and even whole-organism cell type discovery and characterization. Such data will deepen our understanding of the regulatory basis of cellular identity, in development, neurodegenerative disease, and regenerative medicine.

Supplementary Materials

www.sciencemag.org/content/347/6226/1138/suppl/DC1

Materials and Methods

Supplementary Text

Figs. S1 to S11

Tables S1 and S2

References (2736)

References and Notes

  1. Acknowledgments: The raw data have been deposited with the Gene Expression Omnibus (www.ncbi.nlm.nih.gov/geo) under accession code GSE60361. Annotated data are available at http://linnarssonlab.org/cortex. We thank P. Ernfors, K. Harris, and R. Sandberg for useful comments on the manuscript; F. Ginhoux for helpful discussions on microglia and macrophages; A. Johnsson for laboratory management and support; ALM/SciLife (H. G. Blom) for technical support; and Fluidigm Inc. (R. C. Jones and M. Lynch) for generous technical and instrument support. S.L. was supported by the European Research Council (261063, BRAINCELL) and the Swedish Research Council (STARGET); A.Z. was supported by the Human Frontier Science Program; A.B.M.-M. was supported by the Karolinska Institutet (BRECT); C.R. was supported by the Swedish Cancer Society (CAN2013/852); G.C.-B. was supported by the Swedish Research Council, the European Union (FP7/Marie Curie Integration Grant EPIOPC), the Åke Wiberg Foundation, the Karolinska Institutet Research Foundations, Svenska Läkaresällskapet, Clas Groschinskys Minnesfond, and Hjärnfonden; J.H.-L. was supported by the Swedish Research Council, the European Union [FP7/Marie Curie Actions (322304, Adolescent Development)], StratNeuro, and the Jeanssons, Åke Wibergs, and Magnus Bergvalls Foundations; C.B. was supported by the European Research Council (294556, BBBARRIER), a Knut and Alice Wallenberg Scholar Grant, the Swedish Cancer Society, and Swedish Research Council. Supplementary materials contain additional data.
View Abstract

Navigate This Article