Memory T cell RNA rearrangement by hn RNP LL

Immunological Memory is characterized by a heightened recall response to a previously encountered antigen, resulting from clonal expansion of antigen-specific lymphocytes and differentiation into specialized memory cells. Differentiation of memory B cells involves irreversible encountered antigen, resulting from clonal expansion of antigen-specific lymphocytes and differentiation into specialised memory cells. Differentiation of memory B cells involves irrversible DNA rearrangements and mutations, but differentiation of memory T cells is less understood. In an ENU screen for mouse mutations affecting the proportion of T cells with known memory cell markers, such as the alternative spliced Ptprc isoform CD45RO, we identified an inducible RNA-binding protein of previously unknown function, hnRNPLL (gene symbol Hnrpll), whose mutation abolishes the regulated silencing of CD45RABC exons in T cells and other leukocytes. In the mutant, a single amino acid substitution destabilizes an RNA-recognition domain that binds with micromolar affinity to RNA containing the Ptprc ARS exon silencing sequence. Hnrpll mutations selectively diminishes T cell accumulation in peripheral lymphoid tissues but not proliferation. Exon array analysis of Hnrpll mutant naive and memory T cells reveals an extensive program of alternative mRNA splicing in memory T cells coordinated by hnRNPLL. A remarkable overlap with alternative splicing in neural tissues may reflect a co-opted strategy of RNA rearrangement for achieving immunological memory in T cells. Keywords: Memory; T cell; Immunological memory; Cell differentiation; Alternative splicing; T cell survival; Neuron; Brain; RNA-binding; Exon silencing sequence; Mouse mutation; Exon Microarray Single cell suspensions of spleen and lymph node cells were prepared from either wildtype or thu/thu mice, stained with anti-CD4, anti-CD8, and anti-CD44 antibodies, and sorted into sterile ice-cold RPMI+10%FCS. Four independent pools of mRNA from naïve CD4 (CD44lo), memory CD4 (CD44hi), naive CD8 (CD44lo), and memory CD8 (CD44hi) T cells, and from flow sorted CD4 and CD8 single positive thymocytes, were prepared from wildtype and thu/thu mice. For each pool, total RNA was extracted from 2-5 million sorted cells using a RNeasy plus Mini kit (Qiagen), purity analyzed using an Agilent Bioanalyzer Chip, 1ug RNA in vitro transcribed to cRNA, and after the first round amplification 10ug cRNA was used in a second round of amplification. A hybridization cocktail that included 5.5ug of fragmented end labelled ssDNA was applied to Affymetrix GeneChip Mouse Exon 1.0 ST Array chips. Hybridization was performed using F450-001 fluidics was and stain script using the Affymetrics Gene Chip Fluidics station 450. Array were scanned using the Affymetrix GCS3000 7G and Gene Chip Operating software to produce CEL files. The probeset intensities were calculated from the CEL files of the 48 samples in Partek Genomics Suite 6.3 using the core probe sets as defined by Affymetrix. 15,788 informative genes were included in the analysis, defining an informative gene as a unique transcript ID with mean signals for individual probesets greater than 3 in the T cell samples analyzed. Alternative splicing was analysed in each T cell subset using two way (genoytpe versus probeset ID) ANOVA using Partek algorithms with no normalisation. False discovery rates for each T cell subset 2-way ANOVA comparison were calcualted by the Benjamini & Hochger step-up method in Parteck. Gene vies were created with Partek using known trancripts in the UCSC Mouse Genome database (version July 2007).