Electrophysiological, transcriptomic and morphologic profiling of single neurons using Patch-seq

Abstract

Despite the importance of the mammalian neocortex for complex cognitive processes, we still lack a comprehensive description of its cellular components. To improve the classification of neuronal cell types and the functional characterization of single neurons, we present Patch-seq, a method that combines whole-cell electrophysiological patch-clamp recordings, single-cell RNA-sequencing and morphological characterization. Following electrophysiological characterization, cell contents are aspirated through the patch-clamp pipette and prepared for RNA-sequencing. Using this approach, we generate electrophysiological and molecular profiles of 58 neocortical cells and show that gene expression patterns can be used to infer the morphological and physiological properties such as axonal arborization and action potential amplitude of individual neurons. Our results shed light on the molecular underpinnings of neuronal diversity and suggest that Patch-seq can facilitate the classification of cell types in the nervous system.

Figure 1

Two morphologically and electrophysiologically distinct neuronal classes in neocortical layer 1. (a) Schematic of experimental approach. (b) Representative examples of the morphology (top) and firing pattern (bottom) of the two main types of neurons found in L1: elongated neurogliaform cells (eNGCs, orange) and single bouquet cells (SBCs, cyan). For morphological reconstructions, the darker outline represents the somatodendritic region, and the lighter color is the axonal arbor; scale bar, 100 μm. For firing patterns, gray lines represent current steps used to elicit the firing patterns shown above; scale bars, 300 ms (horizontal bar), 40 mV and 500 pA (vertical bar); arrows denote prominent after-depolarization in SBCs. (c) Neurons recorded using Patch-seq protocol display similar firing responses as seen using standard electrophysiological techniques, as shown in b. (d) Output of automated cell type classifier robustly predicts morphological class based on electrophysiological features. (e) Weights of features used in the automated cell type classifier. (f) Results of the automated classifier highly correlate with an independent, blinded expert classification of the electrophysiological properties as “eNGC-like” or “SBC-like,” r = 0.91. (g) Example cells before and after RNA extraction.

Figure 2

Single-neuron transcriptome profiles predict cell type and electrophysiological properties. (a) Clustering analysis separates interneurons (blue dendrogram subtree) from other neuronal classes (green dendrogram subtree; includes four pyramidal neurons, and one astrocyte) based on marker gene expression. Two L1 interneurons clustered with non-interneuron cell types, indicating possible contamination of these samples, and so these two cells were excluded from our analysis of interneuron subtypes. (b) Number of genes detected per neuron using two different expression thresholds for interneurons patch-clamp recorded ex vivo and in vivo. (c) Pairwise Spearman correlation across all detected genes for ex vivo and in vivo patch-clamp recorded interneurons. (d) Two-dimensional t-SNE representation of gene expression for all L1 interneurons. Cells are colored according to affinity propagation-based clustering in gene-space spanned by the 3,000 most variable genes, prior to dimensionality reduction. (e) The same two-dimensional map as in d, but with cells color-coded according to expert classification of cell type based on electrophysiological properties. Performance of GLMs using single-neuron gene expression to predict cell type (f), ADP (g), AHP (h), or action potential (AP) amplitude (i).

Figure 3

Differential gene expression analysis reveals novel markers for L1 interneuron classes. (a) Boxplots summarize the cell-type expression level of previous marker genes (Vip and Reelin). (b) Boxplots with expression levels across cell types for differentially expressed genes identified between the two affinity propagation clusters. (c) Gene categories that were significantly enriched in SBCs or eNGCs based on gene set enrichment analysis. The gene matrix illustrates gene overlap among categories; the bar plot shows the false discovery rates and the numbers indicate normalized enrichment scores per category from GSEA.