Single cell transcriptomics identifies a Hedgehog-mediated immunomodulatory signaling circuit in the eye choroid
The activity and survival of retinal photoreceptors depend on support functions performed by the retinal pigment epithelium (RPE) and on oxygen and nutrients delivered by blood vessels in the underlying choroid. By combining single cell and bulk RNA sequencing, we categorized mouse RPE/choroid cell types and characterized the tissue-specific transcriptomic features of choroid endothelial cells. We found that choroid endothelium adjacent to the RPE expresses very high levels of Indian Hedgehog, and identified its downstream target as stromal GLI1+ mesenchymal stem cell-like cells. Genetic impairment of Hedgehog signaling in vivo induced significant loss of choroidal mast cells, as well as an altered inflammatory response and exacerbated visual function defects after retinal damage. Our studies reveal the cellular and molecular landscape of adult RPE/choroid and uncover a Hedgehog-regulated choroidal immunomodulatory signaling circuit. These results open new avenues for the study and treatment of retinal vascular diseases and choroid-related inflammatory blinding disorders.
For more informations please refer to the manuscript: Lehmann GL, Hanke-Gogokhia C, Hu Y, Bareja R, Salfati Z, Ginsberg M, et al. Single-cell profiling reveals an endothelium-mediated immunomodulatory pathway in the eye choroid. Journal of Experimental Medicine. 2020;217(6).
Chromium single-cell RNA-seq outputs were processed by Cell Ranger analysis pipelines. Data is currently unavailable to the public before publication.
R version 3.4.3 http://cran.us.r-project.org/bin/macosx/R-3.4.3.pkg
dplyr_0.7.4
Seurat_2.1.0 https://cran.r-project.org/src/contrib/Archive/Seurat/Seurat_2.1.0.tar.gz (other Seurat versions will generate slightly different results )
After pulling this repository, create folders data and output in the top working folder. Move Cell Ranger analysis results into data folder.
Tree structure of directory:
|── LICENSE.md
|── R
| |── Differential_analysis.R
| |── Identify_Cell_Types_Manually.R
| |── Seurat_functions.R
| └── Seurat_setup.R
|── README.md
|── data
| |── 129_B6
| | └── outs
| | |── filtered_gene_bc_matrices
| | | └── mm10
| | | |── barcodes.tsv
| | | |── genes.tsv
| | | └── matrix.mtx
| |── B6
| | └── outs
| | |── filtered_gene_bc_matrices
| | | └── mm10
| | | |── barcodes.tsv
| | | |── genes.tsv
| | | └── matrix.mtx
|── doc
|── output
└── scRNAseq-MouseEyes.Rproj
QC.R A brief overview for raw data. No data pre-processing will be made in this step.
Seurat_setup.R Unsupervised cell clustering analysis was carried out using the Seurat 2.2 R package. Cells with <500 genes and genes detected within <3 cells were excluded from the analysis. Gene expression raw counts were normalized following a global-scaling normalization method with a scale factor of 10,000 and a log transformation, using the Seurat NormalizeData function. The top 1000 highly variable genes from young C57BL/6J and aged C57BL/6J datasets were selected, followed by canonical correlation analysis (CCA) to identify common sources of variation between the two datasets and minimize the batch effect. The first 20 CCA results were chosen for principal component analysis (PCA). Cells were used for 2-dimensional t-Distributed Stochastic Neighbor Embedding (tSNE) (ref van der maaten and hinton 2008) with 0.8 resolution.
After running this script, a mouse_eyes_alignment.Rda file will be generated inside data folder.
Do not modify any files in data folder.
Identify_Cell_Types_Manually.R All clusters are examed against 122(number may change) CD marker genes. All cell types are predicted by at least two marker genes with the adjusted p-value(FDR) smaller than 10^-30.
Endothelial cells were identified by Cdh5, Flt1, Kdr, Pecam1, Plvap, Ptprb, and Vwf.
Pericytes were identified by Dcn, Des, Mylk, Pdgfrb, and Rgs5.
Hematopoietic cells were identified by Laptm5 and Ptprc.
Melanocytes were identified by Mlana and Pmel.
Myelinating Schwann cells were identified by Mbp and Mpz.
Retinal pigment epitheliums were identified by Rlbp1 and Rpe65.
DotPlot is implemented for visualizing differential expressed marker genes.
Multiple plots and table will be generated, save them if you want. I prefer to keep the original identity of mouse_eyes_alignment.Rda intact for further downstream analysis.
Differential_analysis.R
Modified FindAllMarkers() FindAllMarkers.UMI() will generate similar dataframe plus two extra columns pct.1_UMI and pct.2_UMI to record nUMI. pct.1_UMI is nUMI of current cluster, pct.2_UMI is average nUMI of rest of clusters.
FindAllMarkers(object, test.use = "bimod") : Likelihood-ratio test for single cell gene expression, (McDavid et al., Bioinformatics, 2013)
p.adjust(p, method = "BH"):Benjamini & Hochberg (1995) ("BH" or its alias "fdr").
Generate CSV file in output folder.
Further, subset the cell types into small clusters.
Re-run RunPCA(), FindClusters(),RunTSNE()
Generate CSV file in output folder.
Below is a example of ./output/129_B6.csv file with first 6 rows.
| row.name | p_val | avg_logFC | pct.1 | pct.2 | p_val_adj | pct.1_UMI | pct.2_UMI | cluster | gene |
|---|---|---|---|---|---|---|---|---|---|
| Lum | 0.000 | 1.596 | 0.983 | 0.236 | 0.000 | 3.048 | 0.567 | 0)Pericytes | Lum |
| Cygb | 0.000 | 1.531 | 0.981 | 0.311 | 0.000 | 2.721 | 0.651 | 0)Pericytes | Cygb |
| Igfbp4 | 0.000 | 1.507 | 1.000 | 0.720 | 0.000 | 3.897 | 1.773 | 0)Pericytes | Igfbp4 |
| Serpine2 | 0.000 | 1.462 | 0.999 | 0.506 | 0.000 | 3.488 | 1.191 | 0)Pericytes | Serpine2 |
| Dcn | 0.000 | 0.964 | 0.968 | 0.347 | 0.000 | 3.400 | 0.983 | 0)Pericytes | Dcn |
| Cxcl12 | 0.000 | 1.222 | 0.981 | 0.531 | 0.000 | 2.957 | 1.148 | 0)Pericytes | Cxcl12 |
The results data frame has the following columns :
p_val: p_val (unadjusted) is calculated using likelihood-ratio test for single-cell gene expression, (McDavid et al., Bioinformatics, 2013)
avg_logFC: log fold-change of the average expression between the two groups. Positive values indicate that the gene is more highly expressed in the first group.
pct.1: The percentage of cells where the gene is detected in the first group.
pct.2: The percentage of cells where the gene is detected in the second group.
p_val_adj: Adjusted p-value, based on Benjamini & Hochberg (1995) ("BH" or its alias "fdr")
pct.1_UMI is nUMI of the current cluster.
pct.2_UMI is average nUMI of rest of clusters.
cluster : either cell types or original clusters in ./data/mouse_eyes_alignment.Rda.
row.name and gene column are identical.
Main_Figures.R Source code to produce all main figures.
sessionInfo() R version 3.4.4 (2018-03-15) Platform: x86_64-apple-darwin15.6.0 (64-bit) Running under: macOS High Sierra 10.13.4
Matrix products: default BLAS: /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libBLAS.dylib LAPACK: /Library/Frameworks/R.framework/Versions/3.4/Resources/lib/libRlapack.dylib
locale: [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
attached base packages: [1] parallel stats graphics grDevices utils datasets methods base
other attached packages: [1] dplyr_0.7.4 Seurat_2.1.0 Biobase_2.38.0 BiocGenerics_0.24.0 Matrix_1.2-14 cowplot_0.9.2 ggplot2_2.2.1
loaded via a namespace (and not attached):
[1] diffusionMap_1.1-0 Rtsne_0.14 VGAM_1.0-5 colorspace_1.3-2
ggridges_0.5.0 class_7.3-14 modeltools_0.2-21 mclust_5.4
htmlTable_1.11.2 base64enc_0.1-3 proxy_0.4-22 rstudioapi_0.7
DRR_0.0.3 flexmix_2.3-14 lubridate_1.7.4 prodlim_2018.04.18
mvtnorm_1.0-7 ranger_0.9.0 codetools_0.2-15 splines_3.4.4
R.methodsS3_1.7.1 mnormt_1.5-5 doParallel_1.0.11 robustbase_0.93-0
knitr_1.20 tclust_1.3-1 RcppRoll_0.2.2 Formula_1.2-2
caret_6.0-79 ica_1.0-1 broom_0.4.4 gridBase_0.4-7
ddalpha_1.3.3 cluster_2.0.7-1 kernlab_0.9-26 R.oo_1.22.0
sfsmisc_1.1-2 compiler_3.4.4 backports_1.1.2 assertthat_0.2.0
lazyeval_0.2.1 lars_1.2 acepack_1.4.1 htmltools_0.3.6
tools_3.4.4 bindrcpp_0.2.2 igraph_1.1.0 gtable_0.2.0
glue_1.2.0 reshape2_1.4.3 Rcpp_0.12.16 NMF_0.21.0
trimcluster_0.1-2 gdata_2.18.0 ape_5.1 nlme_3.1-137
iterators_1.0.9 fpc_2.1-11 psych_1.8.3.3 timeDate_3043.102
gower_0.1.2 stringr_1.3.0 irlba_2.3.2 rngtools_1.2.4
gtools_3.5.0 DEoptimR_1.0-8 MASS_7.3-50 scales_0.5.0
ipred_0.9-6 RColorBrewer_1.1-2 yaml_2.1.19 pbapply_1.3-4
gridExtra_2.3 pkgmaker_0.22 segmented_0.5-3.0 rpart_4.1-13
latticeExtra_0.6-28 stringi_1.1.7 foreach_1.4.4 checkmate_1.8.5
caTools_1.17.1 ggjoy_0.4.0 lava_1.6.1 geometry_0.3-6
dtw_1.18-1 SDMTools_1.1-221 rlang_0.2.0 pkgconfig_2.0.1
prabclus_2.2-6 bitops_1.0-6 lattice_0.20-35 ROCR_1.0-7
purrr_0.2.4 bindr_0.1.1 recipes_0.1.2 htmlwidgets_1.2
tidyselect_0.2.4 CVST_0.2-1 plyr_1.8.4 magrittr_1.5
R6_2.2.2 gplots_3.0.1 Hmisc_4.1-1 dimRed_0.1.0
sn_1.5-2 withr_2.1.2 pillar_1.2.2 foreign_0.8-70
mixtools_1.1.0 survival_2.42-3 scatterplot3d_0.3-41 abind_1.4-5
nnet_7.3-12 tsne_0.1-3 tibble_1.4.2 KernSmooth_2.23-15
grid_3.4.4 data.table_1.10.4-3 FNN_1.1 ModelMetrics_1.1.0
digest_0.6.15 diptest_0.75-7 xtable_1.8-2 numDeriv_2016.8-1
tidyr_0.8.0 R.utils_2.6.0 stats4_3.4.4 munsell_0.4.3
registry_0.5 magic_1.5-8