Alternative splicing of apoptosis genes promotes human T cell survival
Curation statements for this article:-
Curated by eLife
Evaluation Summary:
Apoptotic regulators have long been known to often be expressed in pairs of pro- and anti-apoptotic isoforms. This demonstration of how a program of these splicing changes contributes to immune responses adds an important new understanding of both apoptosis and T cell biology.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. The reviewers remained anonymous to the authors.)
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
Abstract
Alternative splicing occurs in the vast majority of human genes, giving rise to distinct mRNA and protein isoforms. We, and others, have previously identified hundreds of genes that change their isoform expression upon T cell activation via alternative splicing; however, how these changes link activation input with functional output remains largely unknown. Here, we investigate how costimulation of T cells through the CD28 receptor impacts alternative splicing in T cells activated through the T cell receptor (TCR, CD3) and find that while CD28 signaling alone has minimal impact on splicing, it enhances the extent of change for up to 20% of TCR-induced alternative splicing events. Interestingly, a set of CD28-enhanced splicing events occur within genes encoding key components of the apoptotic signaling pathway; namely caspase-9, Bax, and Bim. Using both CRISPR-edited cells and antisense oligos to force expression of specific isoforms, we show for all three of these genes that the isoform induced by CD3/CD28 costimulation promotes resistance to apoptosis, and that changes in all three genes together function combinatorially to further promote cell viability. Finally, we show that the JNK signaling pathway, induced downstream of CD3/CD28 costimulation, is required for each of these splicing events, further highlighting their co-regulation. Together, these findings demonstrate that alternative splicing is a key mechanism by which costimulation of CD28 promotes viability of activated T cells.
Article activity feed
-
-
Evaluation Summary:
Apoptotic regulators have long been known to often be expressed in pairs of pro- and anti-apoptotic isoforms. This demonstration of how a program of these splicing changes contributes to immune responses adds an important new understanding of both apoptosis and T cell biology.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. The reviewers remained anonymous to the authors.)
-
Reviewer #1 (Public Review):
Blake and colleagues examine programs of alternative splicing controlled during T cell activation. Using CD4+ T cells from human donors, cells were stimulated with anti-CD28, anti-CD3, and combined anti-CD3/28 antibodies. RNA was then isolated at 2 time points, sequenced, and analyzed for changes in spliced isoform ratios. T Cell Receptor stimulation alone via anti-CD3 is known to induce the anergic state resulting from suboptimal stimulation, while CD28 costimulation with CD3 induces many genes to a higher level of expression similar to stimulation by antigen-presenting cells. Analyzing the splicing responses to these stimuli, the authors find that CD28 costimulation also enhances the splicing changes that accompany T cell activation. A subset of these splicing targets encode apoptotic regulators including …
Reviewer #1 (Public Review):
Blake and colleagues examine programs of alternative splicing controlled during T cell activation. Using CD4+ T cells from human donors, cells were stimulated with anti-CD28, anti-CD3, and combined anti-CD3/28 antibodies. RNA was then isolated at 2 time points, sequenced, and analyzed for changes in spliced isoform ratios. T Cell Receptor stimulation alone via anti-CD3 is known to induce the anergic state resulting from suboptimal stimulation, while CD28 costimulation with CD3 induces many genes to a higher level of expression similar to stimulation by antigen-presenting cells. Analyzing the splicing responses to these stimuli, the authors find that CD28 costimulation also enhances the splicing changes that accompany T cell activation. A subset of these splicing targets encode apoptotic regulators including Caspase-9, Bax, and Bim. They show that forced expression of the isoforms that are increased by costimulation results in reduced apoptosis in Jurkat cells treated with apoptotic inducers. Using kinase inhibitor treatments they show that Jnk kinase activity is required for the splicing changes in the three apoptotic regulators.
-
Reviewer #2 (Public Review):
Blake and colleagues study how co-stimulation of T cells through the TCR and CD28 (a necessary event for full T cell response) impacts on alternative splicing and how these post-transcriptional regulation events promote resistance to apoptosis and thus enhance the viability of activated T cells. They identify and validate the functional impact of some of these alternative splicing events and also argue that JNK signaling is involved in their regulation. Revealing a program of alternative splicing induced by co-stimulation of CD3 and CD28, important for T cell activation, is novel and relevant. The results support the conclusions, although the extent to which the quantitative changes in splicing induced by CD28 co-stimulation contribute to anergy protection is not fully clear.
-
Reviewer #3 (Public Review):
Blake et al. describe a comprehensive analysis of alternative splicing changes that accompany the activation of primary human T cells with anti-CD3 and anti-CD3/CD28 antibodies. They then focused their attention on 3 genes involved in the regulation of apoptosis that exhibited anti-CD28 enhanced alternative splicing, culminating in functional studies suggesting that the 3 splicing changes make important contributions to T-cell apoptosis/cell survival. They further document a role for JNK signaling in activating the splicing changes. These results should be of considerable interest to both the alternative splicing and T-cell activation fields.
Despite the substantial merits of both the initial comprehensive analysis and the subsequent targeted analysis of genes involved in the regulation of T cell apoptosis …
Reviewer #3 (Public Review):
Blake et al. describe a comprehensive analysis of alternative splicing changes that accompany the activation of primary human T cells with anti-CD3 and anti-CD3/CD28 antibodies. They then focused their attention on 3 genes involved in the regulation of apoptosis that exhibited anti-CD28 enhanced alternative splicing, culminating in functional studies suggesting that the 3 splicing changes make important contributions to T-cell apoptosis/cell survival. They further document a role for JNK signaling in activating the splicing changes. These results should be of considerable interest to both the alternative splicing and T-cell activation fields.
Despite the substantial merits of both the initial comprehensive analysis and the subsequent targeted analysis of genes involved in the regulation of T cell apoptosis and survival, the manuscript has one major limitation (#4 below) and a few lesser limitations. The major limitation makes it difficult to accurately assess the CRISPR-based functional experiments included in the manuscript.
1. The initial analysis in Figure 1D could have been strengthened by the inclusion of additional quantitative information about the distribution of alternative splicing changes. For example, the authors set a threshold of >10% dPSI to be considered a significant event. To appreciate the findings, it would have been helpful to know how many of these start at 0-10 PSI prior to stimulation, how many start at 10-20 PSI, 20-30 PSI, etc. In addition, the distribution of dPSI magnitudes would have been of interest (the scatter plots in Figures 2A and 2B are difficult to evaluate quantitatively).
2. Similar to the above, an evaluation of the data in Figures 2E and 2F would have benefited from a closer look. For example, only a subset of the "significant alternative splicing" events will have the potential to be enhanced 2-fold by CD28 stimulation because the dPSI value with CD3 alone may be in the range of 40 or 50 or more at some genes. It therefore would have been of interest to know the extent to which the distributions shown in Figures 2E and 2F are influenced by the CD3 dPSI. (One thought would be to examine dPSI ratio distributions after separating the splicing events into a few different bins based on CD3 dPSI.)
3. An evaluation of the data in Figure 3 would have benefited from the inclusion of the PSI value from unstimulated cells for each gene.
4. My most significant concern about the results is that, from the data in Figures 5A, 5D, and S5, it isn't clear that the remaining wild-type allele in the CASP9 and BIM heterozygous clones is generating full-length transcripts and protein (unless I'm misunderstanding the experiment). In the images shown, the full-length mRNAs and proteins appear to be entirely absent, despite the genetic evidence that an undeleted allele remains. One possibility is that a CRISPR guide RNA damaged the second wild-type clone without resulting in a large deletion. The strategy employed to create heterozygous clones to examine the impact of moderate changes in protein ratio is admirable, but the results appear to show dramatic changes (rather than moderate changes) in protein ratio due to the absence of transcripts and protein from the undeleted alleles.
-
