In silico analysis of alternative splicing events implicated in intracellular trafficking during B-lymphocyte differentiation

Abstract

There are multiple regulatory layers that control intracellular trafficking and protein secretion, ranging from transcriptional to posttranslational mechanisms. Finely regulated trafficking and secretion is especially important for lymphocytes during activation and differentiation, as the quantity of secretory cargo increases once the activated cells start to produce and secrete large amounts of cytokines, cytotoxins, or antibodies. However, how the secretory machinery dynamically adapts its efficiency and specificity in general and specifically in lymphocytes remains incompletely understood. Here we present a systematic bioinformatics analysis to address RNA-based mechanisms that control intracellular trafficking and protein secretion during B-lymphocyte activation, and differentiation, with a focus on alternative splicing. Our in silico analyses suggest that alternative splicing has a substantial impact on the dynamic adaptation of intracellular traffic and protein secretion in different B cell subtypes, pointing to another regulatory layer to the control of lymphocyte function during activation and differentiation. Furthermore, we suggest that NERF/ELF2 controls the expression of some COPII-related genes in a cell type-specific manner. In addition, T cells and B cells appear to use different adaptive strategies to adjust their secretory machineries during the generation of effector and memory cells, with antibody secreting B cell specifically increasing the expression of components of the early secretory pathway. Together, our data provide hypotheses how cell type-specific regulation of the trafficking machinery during immune cell activation and differentiation is controlled that can now be tested in wet lab experiments.

Keywords: B cell differentiation; COPII; NERF/ELF2; alternative splicing; antibody-secreting cells; memory B cell; secretory pathway.

Figure 1

Gene expression during B cell differentiation. (A) Summarized schematic representation of the steps of B cell differentiation. Step 1. Antigen recognition induces the expression of effector molecules by T cells, which then activate B cells. Step 2. Activated B-cell proliferation. Step 3. Differentiation to resting memory cells or antibody-secreting cells (PB and PC). Adapted from “Steps in B-cell Differentiation”, by BioRender.com (2022). (B) PCA of all genes expressed in at least one subtype of B cells. (C) Heatmap showing all differentially expressed genes in B cells. Pajd < 0.001, log2FC > 0.5 (or < -0.5). (D) The 10 strongest enriched GO terms in PB vs MBC. The GO term enrichment was performed using the enrichGO function (see methods). The size of the dots represents the number of genes in the significant differentially expressed gene list associated with the GO term and the color of the dots represents the P-adjusted values (FDR). GeneRatio: the number of differentially expressed genes divided by the total number of genes in the given GO term.

Figure 2

Expression of RNA processing-related and intracellular trafficking-related genes. The expression of RNA processing-related and intracellular trafficking related-genes was compared between MBC and PC. The following GO terms were analyzed: (A) mRNA processing. (B) RNA splicing. (C) intracellular protein transport. (D) COPII-coated vesicle budding. (E) retrograde vesicle-mediated transport Golgi to ER. (F) protein exit from ER. In the volcano plots, log2 fold change (log2FC) was plotted against Pajd < 0.001, fold changes were considered significant if log2FC > 0.5 (or < -0.5). Genes up-regulated in plasma cells are shown in red (down-regulated in memory). Genes down-regulated in plasma cells are shown in blue (up-regulated in memory). (G) Comparison of gene expression levels of COPII-components in PC compared to prePB and T-cell activation [T-cell data was taken from(34)].

Figure 3

Global splicing analysis in B-Cell differentiation. (A) Heatmap showing the PSI of all significantly altered splicing events in B-cell populations (P-value < 0.001, DPSI > |0.4|). (B) PCA of all annotated splice events in B-cell differentiation. (C) Enriched GO terms in AS genes in MBC; only the GO terms related to intracellular transport are shown. P-adjusted values (FDR). (D) Heatmap showing all significant PSI changes for splicing events in genes in the GO-term “COPII vesicle coat”. (E) Sashimi plots for selected frameshift splice events in COPII compartments. (F) Sashimi plot showing splicing of SEC16A Exon 29 and 30.

Figure 4

Splice events with protein-coding isoforms. (A) Barplot showing PSI of two different splice events in KLHL12. Schematic summary of KLHL12 alternative splicing and the resulting change of stop codon. (B) Protein structure prediction with Robetta (https://robetta.bakerlab.org/) of the KLHL12 isoforms (FL and Δ10,11). The Δ10,11 causes a deletion of the kelch domain (colored magenta) (C) Bargraph showing the PSIs of exon 13 and exon 17 of PICALM in differentiation stages of B-Lymphocytes. (D) Schematic summary of PICALM isoforms. Significance is indicated by asterisks (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001).

Figure 5

NERF/ELF2 controls expression of COPII-related genes in MBCs. (A) Venn diagram showing the TFs that had a high Pearson correlation coefficient (>0.9) with SEC24B, CUL3, and CSNK1D in MBCs (left). Dotplot showing the Pearson correlation coefficient of the 135 TFs and the SEC24B, CUL3, and CSNK1D genes in publicly available RNA-seq data of 17 human tissues (SRA study ERP003613). Shown in the graph is the mean pearson correlation coefficient against the mean p-value of the correlation in -log10 scale of each TF with the three COPII genes (right). (B) Dotplot with added linear regression line for the correlation between NERF/ELF2 and SEC24B, CUL3, and CSNK1D in human tissue data. Each dot is color coded for originated tissue. (C) NERF/ELF2 gene expression in B cells. (D) NERF/ELF2 alternative splicing pattern of Exon 7 3'SS in B cells.

Figure 6

Alternative splicing of the transcription factor NERF/ELF2. (A) Possible NERF/ELF2 isoforms according to USCS genome browser and UniProt. (B) Sashimi plot of NERF/ELF2 transcript in MBC, prePB, PB, and PC. (C) Stacked barplot of NERF/ELF2 isoforms in B-Lymphocyte differentiation.