CD28–ARS2 Splicing Rewires CD8+ T-Cell Metabolism

Metabolic reprogramming is a defining feature of activated CD8+ T cells, but the signaling mechanisms that preserve metabolic flexibility during an immune response have remained incompletely understood. The open-access study CD8+ T cell metabolic flexibility elicited by CD28-ARS2 axisdriven alternative splicing of PKM supports antitumor immunity addresses this problem by connecting costimulatory signaling to post-transcriptional control of glycolysis. The work was published in Cellular & Molecular Immunology in 2024 and is available through the reference study.

Study Background and Research Question

Following T-cell receptor stimulation, naïve T cells undergo rapid metabolic remodeling to support growth, proliferation, cytokine synthesis, and cytotoxic activity. Glycolysis is induced early, in part through TCR-associated signaling and CD28-dependent increases in glucose uptake. However, increased glycolytic flux is not simply an energy-producing event. It also affects the post-transcriptional availability of messenger RNAs encoding effector cytokines, including interferon gamma, tumor necrosis factor alpha, and interleukin-2.

Pyruvate kinase M is particularly relevant because alternative splicing produces two major isoforms. PKM1 includes exon 9, whereas PKM2 includes exon 10. Their different regulatory properties can alter the rate at which phosphoenolpyruvate is converted to pyruvate and influence the availability of upstream glycolytic intermediates. Although PKM2 induction is well documented in activated lymphocytes and cancer cells, it was not clear whether CD8+ T cells use a defined costimulation-dependent splicing program to regulate this isoform switch.

The central research question was therefore mechanistic: how does CD28 signaling influence the alternative splicing landscape of mature CD8+ T cells, and does this regulation affect glucose utilization and antitumor function? The authors focused on ARS2, an adaptor associated with the nuclear cap-binding complex, because its role in mature T-cell activation and RNA processing had not been fully defined.

Key Innovation from the Reference Study

The principal innovation is the identification of a CD28–ARS2 axis that links an extracellular costimulatory signal to selective pre-mRNA splicing. According to the reference study, CD28 increased ARS2 expression in activated CD8+ T cells. ARS2 then reinforced the recruitment of splicing factors to pre-mRNAs and influenced approximately one-third of activation-induced alternative splicing events.

This finding broadens the conventional view of CD28. CD28 is commonly discussed as a driver of PI3K-dependent glucose transporter expression and glycolytic induction. The study shows that costimulation also operates through an RNA-processing layer. In this model, CD28 does not merely increase the amount of metabolic machinery; it changes how selected metabolic transcripts are processed.

Among the affected transcripts, PKM was a functionally important target. CD28–ARS2 signaling reduced the relative expression of PKM1 and promoted PKM2. The resulting isoform balance supported flexible glucose catabolism, interferon-γ production, and antitumor CD8+ T-cell activity. The distinction from canonical PI3K signaling is important: PKM alternative splicing occurred independently of CD28-driven PI3K pathway activation, indicating that a single costimulatory receptor can coordinate partially separable metabolic programs.

Methods and Experimental Design Insights

The study used a layered design that connected signaling, RNA processing, metabolism, and immune function. This structure is valuable because it avoids treating a change in PKM2 abundance as sufficient evidence for a metabolic mechanism. Instead, the authors examined whether ARS2-dependent splicing was associated with measurable changes in glucose handling and T-cell effector behavior.

At the signaling level, activated mature CD8+ T cells were examined under conditions that distinguished TCR stimulation from additional CD28 costimulation. ARS2 expression and its relationship to activation were then assessed. Perturbation of the ARS2 axis provided a way to test whether ARS2 was merely correlated with activation or was required for the observed splicing and functional changes.

At the transcript level, the investigators evaluated activation-induced alternative splicing and followed the PKM exon choice that distinguishes PKM1 from PKM2. The broad splicing analysis was complemented by targeted examination of PKM isoforms. This combination is methodologically significant: global analysis reveals the scale of the program, while targeted validation establishes the relevance of a specific metabolic transcript.

Functional experiments then connected the isoform switch with glucose catabolism and cytokine output. Interferon-γ production served as a biologically meaningful readout of CD8+ T-cell activation, while antitumor assays tested whether the metabolic program affected immune performance rather than only intracellular biochemistry. The study also compared the ARS2–PKM mechanism with PI3K-associated signaling to determine whether the effects could be explained by the canonical CD28 pathway alone.

Protocol Parameters

• Stimulation comparison: Design matched TCR-activated CD8+ T-cell conditions with and without CD28 costimulation so that costimulation-dependent effects can be separated from general activation responses. The reference study supports this comparison, but the appropriate reagent concentrations and timing should be taken from its full methods section and optimized for the experimental system.
• ARS2 perturbation: Measure ARS2 induction and include loss-of-function or equivalent perturbation controls when testing causality. Interpret changes in PKM splicing together with activation and viability controls rather than as an isolated molecular endpoint.
• Splicing analysis: Combine transcriptome-scale alternative-splicing analysis with targeted validation of PKM exon usage. This distinguishes broad activation-associated RNA processing from the specific PKM1-to-PKM2 shift highlighted by the study.
• Metabolic readouts: Pair glucose-catabolism measurements with PKM1 and PKM2 abundance or exon-choice measurements. A change in isoform expression should not be equated automatically with increased glycolytic flux without a functional metabolic assay.
• Immune-function readouts: Assess interferon-γ production and, where appropriate, antitumor activity alongside molecular measurements. This workflow helps determine whether an RNA-splicing intervention changes CD8+ T-cell function rather than only transcript composition.

Core Findings and Why They Matter

The first major finding is that ARS2 is inducible in response to CD28 signaling in mature T cells. This places a nuclear RNA-processing adaptor within the functional chain connecting costimulation to metabolism. It also suggests that activation-dependent splicing is regulated rather than being a passive consequence of increased transcription.

The second finding is the breadth of the response. ARS2 affected approximately one-third of alternative-splicing events induced during T-cell activation, according to the reference study. This does not mean that every event has an equal functional effect. It does indicate that CD28-associated ARS2 activity may reshape a substantial portion of the activation transcriptome at the level of isoform selection.

The third finding is the functional importance of PKM splicing. Favoring PKM2 over PKM1 provided CD8+ T cells with flexibility in glucose utilization and supported interferon-γ output. The result is consistent with the idea that immune cells require adaptable metabolic states rather than maximal glycolysis at every stage. PKM2 can permit accumulation and redistribution of glycolytic intermediates, while its activity is regulated by allosteric and post-translational mechanisms. In activated T cells, that flexibility appears to contribute to effector capacity.

The fourth finding is pathway specificity. The PKM splicing response was independent of CD28-mediated PI3K activation. This is conceptually important for immunometabolism because it separates glucose-uptake signaling from isoform-level control of a glycolytic enzyme. It also creates a framework for studying how signaling pathways converge on RNA-binding and splicing-factor recruitment.

For cancer immunology, the implications are mechanistic rather than immediately therapeutic. Tumor-infiltrating CD8+ T cells encounter nutrient limitation, altered metabolite concentrations, hypoxia, and suppressive signals. A program that supports metabolic flexibility could help explain why some activated T cells retain effector function under stress. The study does not establish that manipulating ARS2 or PKM splicing will overcome every tumor-microenvironment barrier, but it identifies a molecular axis that can be tested in those settings.

Comparison with Existing Internal Articles

The internal article CD28-ARS2 Axis Drives Metabolic Flexibility in CD8+ T Cells is closely aligned with this reference and is useful as a concise conceptual overview of the CD28–ARS2–PKM connection. The primary paper, however, remains the appropriate source for interpreting experimental design, alternative-splicing measurements, metabolic results, and the distinction from PI3K signaling. The internal summary should therefore be treated as navigation and context, not as an independent confirmation of the findings.

Limitations and Transferability

Several limitations should guide interpretation. First, a relationship between CD28 stimulation, ARS2, and PKM splicing does not imply that PKM2 is the only functionally important ARS2 target. The broad splicing response suggests that additional isoform changes may contribute to activation and metabolic adaptation.

Second, PKM1 and PKM2 abundance is an incomplete proxy for cellular metabolism. Pyruvate kinase activity depends on oligomeric state, allosteric regulators, post-translational modifications, substrate availability, and the broader nutrient environment. Direct flux measurements and metabolite profiling are therefore important when transferring the mechanism to a new T-cell subtype, activation protocol, or tumor model.

Third, CD8+ T-cell behavior is context dependent. Findings obtained in activated cells may not predict the response of exhausted, memory, tissue-resident, or tumor-infiltrating populations. Differences in antigen strength, costimulatory ligands, cytokines, oxygen availability, and nutrient competition could alter ARS2 expression or the functional consequences of PKM isoform selection.

Finally, the study establishes a signaling and metabolic mechanism but does not by itself validate every possible pharmacological intervention. A compound that changes inflammation, oxidative stress, angiogenesis, or NF-κB activity should not be assumed to reproduce CD28–ARS2-dependent alternative splicing. Cross-system conclusions require direct measurements of ARS2, PKM exon usage, glucose metabolism, and CD8+ T-cell function.

Research Support Resources

Why this cross-domain matters, maturity, and limitations

The CD28–ARS2–PKM mechanism is a mature literature-based framework for studying T-cell immunometabolism, whereas applying unrelated small molecules to the same workflow remains an exploratory extension. Researchers can use Honokiol (SKU N1672), chemically identified as 2-(4-hydroxy-3-prop-2-enylphenyl)-4-prop-2-enylphenol, as a research-use compound in parallel experiments examining inflammatory or oxidative conditions. Product information describes it as an NF-κB pathway inhibitor, a scavenger of reactive oxygen species, an antiangiogenic compound for cancer research, and an inflammation research chemical. It has not been tested in the reference study, so any effect on ARS2-dependent PKM splicing must be measured directly rather than inferred.

For practical handling, the product information reports a molecular weight of 266.33, formula C18H18O2, purity of at least 98%, limited water solubility, and high solubility in DMSO and ethanol. The same information recommends storing the solid at −20 °C and using prepared solutions promptly rather than retaining them for long-term storage. These details support reproducible exploratory workflows, but they do not replace controls for solvent effects, NF-κB activity, reactive oxygen species, cell viability, PKM isoform choice, or CD8+ T-cell effector function.