Diversified expression of pcdhg isoforms in neocortical neurons.

(A) UMAP analysis of all cells (n=17438) collected from 5’-end single-cell sequencing after the cleanup and doublets removing. Neurons were labeled as green dots and non-neurons as red. (B) UMAP clusters of all neurons further labeled by the UMI cutoff. Red dots represented cells with no more than 10 total UMIs of pcdhg (n=3671), and green dots represented cells with more than 10 UMIs (n=2834). (C) Fractions of neurons with more than 10 UMIs in different clusters. (D) Fractions of neurons expressing different pcdhg isoforms in the neocortex. (E) Fraction distribution for different similarity levels in the combinatorial expression of pcdhg variable isoforms among neurons. The similarity level was calculated as. . (F) The observed distribution (black) for the fraction of cells with a various number of isoforms for all neurons. The shuffled distribution (red) was generated by the hypothesis that all the isoform expressed stochastically. (G) The difference between observed and shuffled variance of the fraction distribution of cells from all clusters.

The absence of γ-PCDH increased the synaptic connectivity.

(A) Sample traces (red/green, average traces; gray, 10 original traces) of a multiple electrode whole-cell patch-clamp recording on six neurons in the layer 2/3 of the barrel cortex. Positive evoked postsynaptic responses were indicated by arrows. Orange/green arrows: unidirectional/ bidirectional synaptic connections. Scale bars, 100 mV (green), 50 pA (red), and 50 ms (black). (B) Connectivity probability among nearby pyramidal cells in the layer 2/3 of the barrel cortex in pcdhg cKO mice and their littermate WT controls. γKO: pcdhg flox/flox::nex-cre mice; Ctrl: pcdhg+/+::nex-cre. (C) Connectivity probability among vertically or horizontally aligned neurons in the layer 2/3 of the barrel cortex in pcdhg cKO mice and their littermate WT controls. (D) Connectivity probability among vertically aligned pyramidal cells in the layer 2/3 of the barrel cortex in pcdha cKO mice and their littermate WT controls. αKO: pcdha flox/flox::nex-cre mice; Ctrl: pcdha+/+::nex-cre mice. (E) Connectivity probability among vertically aligned pyramidal cells along with the distance between recorded pairs in pcdhg cKO mice or WT mice. (F) Developmental profiling of the connectivity probability among vertically aligned neurons in pcdhg cKO mice or WT mice. Chi-square tests were used in B-F to calculate the statistical difference.

Overexpressing identical variables, but not C-types γ-PCDHs in neurons decreased their synaptic connectivity.

(A) The sketch to show the brain regions used for RT-qPCR for both experimental and control groups. (B) The overexpression levels were tested by RT-qPCR of electroporated regions. Electroporated isoforms: red; control isoforms: blue. The contralateral sides were used as controls shown in black. Student’s t test was used, *: p<0.05, **: p<0.01, ***: p<0.001, ****: p<0.0001. (C) The diagram of extracting cells for the single-cell RT-PCR assay. (D) Single-cell RT-PCR for γ-PCDH isoforms after the electroporation. Green numbers indicated neurons with fluorescence, and black ones represented nearby neurons without fluorescence. N1-N6 were negative controls for the PCR. Isoforms used in the electroporation were shown in green. Red stars indicate faint signals in negative controls. (E) The effect of overexpressing one or six γ-PCDH isoforms on the synaptic connectivity in WT mice. 1 isoform: pcdhga2; 6 isoforms: pcdhga2, pcdhga8, pcdhga10, pcdhgb1, pcdhgb2, and pcdhgb6; gC4: pcdhgC4; Control: the plasmid vector without pcdhg insertion. (F) The effect of overexpressing 6 γ-PCDH isoforms on the synaptic connectivity in pcdhg cKO mice. 6 used isoforms were same with the ones in (E). pcdhg cKO: pcdhg conditional knockout mice; Control: WT littermates. Chi-square test and false discovery rate (FDR) correction were used to determine the statistical differences between groups in (E) and (F).

Diversified γ-PCDHs were critical for synapse formation in cortical neurons.

(A) The diagram of a sequential in utero electroporation at E14.5 and E15.5. (B) The overexpressed γ-PCDH isoforms in different sets of experiments to achieve different similarity levels between neurons. S.l., Similarity level; EP, Electroporation; P., plasmids mixture; s./d., same/different isoforms in two electroporations. (C) A sample image of recorded neurons after two times of electroporation at E14.5 with plasmids carrying mNeongreen and E15.5 with plasmids carrying mRuby3. In this sample, cells 3 and 4 are mNeongreen positive, cell 6 is mRuby3 positive, and cells 1, 2, and 5 are negative ones without fluorescence. Neurobiotin was added into the internal solution to indicate recorded neurons. The translucent arrows showed the positions of electrodes. Scale bar, left, 100μm, right, 25μm. (D) The connectivity probability for neuron pairs overexpressed different batch of γ-PCDHs isoforms (labeled with different fluorescence) after a sequential in utero electroporation. Chi-square test and false discovery rate (FDR) correction were used to determine the statistical difference. (E) The correlation between the similarity level of overexpressed γ-PCDHs combination and the probability of synaptic connection rate (The source data were the same as d). (F) The graph summary of γ-PCDHs’ effect on synapse formation.