Collibri NGS Case Study: Transcriptome Reprogramming in CD8 T Cells

Cytotoxic CD8 T cells recognize cells infected with pathogens or transformed into cancer cells and kill them. So CD8 T cells are crucial to our defense against infections and cancer. The first step of this defense is to activate naïve CD8 T cells, which makes them differentiate into effector CD8 T cells capable of killing other cells. The Turner lab at the Babraham Institute focuses on understanding the contribution of molecular mechanisms that control gene expression at the posttranscriptional level that are important for the development, activation, and function of lymphocytes such as CD8 T cells.

The Collibri Stranded RNA Library Prep Kits for Illumina Systems were selected to study changes in the transcriptome during the first 24 hours of T cell activation. In total, 11,837 genes were found to be expressed (>1FPKM, >5 base mean reads). Of these, 4,676 genes show differential expression upon CD8 T cell activation (Figure 1A), with 2,471 and 2,205 genes with increased and decreased mRNA abundance when comparing activated to naïve CD8 T cells, respectively. This data shows a significant reprogramming of gene expression in CD8 T cells in the first 24 hours of activation. Importantly, there was a greater than 6-fold increase in the abundance of important effector molecules of CD8 T cells such as granzyme B (Gzmb), interferon gamma (Ifng), and the transcription factors IRF8 and IRF4 (Figure 1A).

Figure 1. Transcriptome reprogramming upon T cell activation. A. Significantly differentially expressed genes in activated compared with naïve mouse CD8 T cells (FDR-adjusted p-value < 0.05 and absolute log2 fold change > 1) were identified using DESeq2, and are shown in red. Selected genes with high log2 fold change are labeled. B. Reads from libraries prepared with Collibri Stranded RNA Library Prep Kit are highly stranded. The percentage of all reads mapping to gene, intron, and ribosomal (r)RNA features are shown. Bars represent means, while points represent biological replicates.

The Collibri Stranded RNA Library Prep Kit generated very high-quality total stranded RNAseq data. More than 95% of reads mapped to annotated genes, and > 95% of them are in the same orientation as the gene, while only < 0.6% reads mapped to ribosomal RNA (Figure 1B). Reflecting the capture of pre-mRNA in the total RNAseq libraries, 20-40% of the reads mapped to introns (Figure 1B).

Request sample

The Turner lab compared the total RNAseq libraries prepared with the Collibri Stranded RNA Library Prep Kit to mRNAseq libraries prepared from the same RNAs with the SMART-Seq® v4 Ultra® Low Input RNA Kit (Cat. No. 634891, Takara) and the Nextera XT DNA Library Preparation Kit (Cat. No. FC-131–1096, Illumina). The Collibri kit revealed more even coverage over genes in the libraries, with less bias towards the 3’ end of genes (Figure 2A). This data suggests that the use of the Collibri helper adaptors for cDNA synthesis results in a reduced 3’-end bias in gene coverage compared to oligo(dT)-primed cDNA synthesis.

Figure 2. Collibri Total RNAseq libraries reliably detect histone and non-coding transcripts.A. Relative coverage of reads averaged across alignment of all genes, from the 5’ to the 3’ end, comparing activated CD8 T cell libraries prepared using the Collibri and SMART-Seq v4 kits (replicates averaged for each). B. Gene expression level measured in fragments per kilobase per million reads (FPKM) for different classes of coding or non-coding genes in activated CD8 T cells, using Collibri or SMART-Seq v4 kits. C. The total number of histone genes detected as expressed (>1 FPKM; >5 mean reads) in naïve or activated CD8 T cells from libraries prepared using the Collibri compared with the SMART-Seq v4 kit. D. Log2 fold change in gene expression in activated compared with naïve CD8 T cells for histone genes detected with the Collibri kit, assessed using DESeq2.

The number of expressed genes detected with both methods was comparable (11,837 with Collibri and 11,381 with SMART-Seq with a 78% overlap), with a similar representation of protein-coding and long non-coding RNAs (Figure 2B). However, there was a substantially higher representation of some classes of small non-coding genes and histone encoding genes that are not polyadenylated in the Collibri libraries (Figure 2B). In the total RNAseq libraries prepared with the Collibri kit, over 5-fold more histone genes were detected compared to the mRNAseq libraries prepared with the SMART-Seq v4 kit (Figure 2C). The increased sensitivity of detection allowed identification of a 10-fold increase on average in the abundance of histone transcripts upon CD8 T cell activation (Figure 2D). Altogether the total RNAseq data generated with the Collibri Stranded RNA Library Prep Kit allowed the Babraham team to address the expression of genes that are not well represented in mRNAseq libraries.

RNA-binding proteins are emerging as central regulators of gene expression at the post-transcriptional level in the immune system. The Turner lab plan to use the Collibri Stranded RNA Library Prep Kits to understand the roles of RNA-binding proteins in controlling mRNA abundance in lymphocytes. These kits will provide an advantage over mRNAseq methods as they allow interrogation of the expression of non-coding or non-polyadenylated RNAs such as small nuclear RNAs and RNAs encoding histones.