We report a role for the 3-untranslated region in control of mRNA splicing and show that human TNF-␣ 3 UTR harbors a cis-acting element that renders splicing of precursor transcripts dependent on activation of PKR, the RNA-activated protein kinase that phosphorylates eukaryotic initiation factor 2 (eIF2). When this element, designated 2-APRE, is present, splicing becomes sensitive to inhibition by the PKR inhibitor, 2-aminopurine, or by coexpression of transdominant-negative mutant PKR. Our results reveal that activation of PKR is required for splicing of mRNA when precursor transcripts contain the 2-APRE and that increased expression of wild-type PKR enhances their splicing efficiency. Thus, PKR responds as trans-acting factor to the 2-APRE. 2-APRE RNA forms a stable, 17-bp stem-loop structure and strongly activates PKR in vitro, inducing eIF2␣ phosphorylation. Despite its ability to activate PKR during splicing, the 2-APRE within the 3 UTR does not affect translation efficiency of the resulting TNF-␣ mRNA in transfected cells. PKR and the 3 UTR thus interact during mRNA splicing to confer a novel type of regulation on expression of the TNF-␣ gene.

Interferon-gamma mRNA activates the RNA-dependent protein kinase PKR, which in turn strongly attenuates translation of interferon-gamma mRNA. Unlike riboswitches restricted to noncoding regions, the interferon-gamma RNA domain that activates PKR comprises the 5' UTR and 26 translated codons. Extensive interferon-gamma coding sequence is thus dedicated to activating PKR and blocking interferon-gamma synthesis. This implies that the PKR activator is disrupted by ribosomes during translation initiation and must refold promptly to restore PKR activation. The activator structure harbors an essential kink-turn, probably to allow formation of a pseudoknot that is critical for PKR activation. Three indispensable short helices, bordered by orientation-sensitive base pairs, align with the pseudoknot stem, generating RNA helix of sufficient length to activate PKR. Through gain-of-function mutations, we show that the RNA activator can adopt alternative conformations that activate PKR. This flexibility promotes efficient refolding of interferon-gamma mRNA, which is necessary for its dual function as translation template and activator of PKR, and which thus prevents overexpression of this inflammatory cytokine.

Full T-cell activation requires interaction between the costimulatory receptors B7-2 and CD28. By binding CD28, bacterial superantigens elicit harmful inflammatory cytokine overexpression through an unknown mechanism. We show that, by engaging not only CD28 but also its coligand B7-2 directly, superantigens potently enhance the avidity between B7-2 and CD28, inducing thereby T-cell hyperactivation. Using the same 12-aa β-strand-hinge-α-helix domain, superantigens engage both B7-2 and CD28 at their homodimer interfaces, areas remote from where these coreceptors interact, implying that inflammatory signaling can be controlled through the receptor homodimer interfaces. Short B7-2 dimer interface mimetic peptides bind diverse superantigens, prevent superantigen binding to cellsurface B7-2 or CD28, attenuate inflammatory cytokine overexpression, and protect mice from lethal superantigen challenge. Thus, superantigens induce a cytokine storm not only by mediating the interaction between MHC-II molecule and T-cell receptor but also, critically, by promoting B7-2/CD28 coreceptor engagement, forcing the principal costimulatory axis to signal excessively. Our results reveal a role for B7-2 as obligatory receptor for superantigens. B7-2 homodimer interface mimotopes prevent superantigen lethality by blocking the superantigen-host costimulatory receptor interaction.superantigen | cytokine storm | costimulatory receptor | B7-2 dimer interfaceA s principal costimulatory receptor, CD28 is a critical regulator of the immune response (1-3). Expressed constitutively on T cells, CD28 is a homodimer that interacts with its B7 coligands, transducing the signal essential for an immediate T-cell response (2-5). CD28 coligand B7-2 (CD86) is expressed constitutively on antigen-presenting cells whereas B7-1 (CD80) is induced only later (5, 6); thus, B7-2/CD28 interaction regulates early antigen signaling (7,8).The inflammatory cytokine response is indispensable for protective immunity, yet bacterial and viral infections often elicit an exaggerated response ("cytokine storm") harmful to the host. Thus, superantigens from Staphylococcus aureus and Streptococcus pyogenes induce toxic shock by activating an immune response, orders of magnitude beyond that elicited by regular antigens. Superantigens exploit the main axis of T-cell activation by binding directly as intact proteins to most major histocompatibility class II (MHC-II) and T-cell receptor (TCR) molecules outside their antigen-binding domains, linking them and bypassing restricted presentation of conventional antigens that typically activate <1% of T cells, thereby activating up to 20 to 30% of T cells (9-11). Moreover, T-cell activation by superantigens requires their direct binding to CD28 (12), the second signaling molecule mandatory for T-cell activation, which results in massive induction of inflammatory cytokines that mediate toxic shock, including IL-2, IFN-γ, and TNF.Induction of human inflammatory cytokine gene expression by divergent superantigens is inhibited by a short p...

Short elements in mammalian mRNA can control gene expression by activating the RNA-dependent protein kinase PKR that attenuates translation by phosphorylating cytoplasmic eukaryotic initiation factor 2α (eIF2α). We demonstrate a novel, positive role for PKR activation and eIF2α phosphorylation in human globin mRNA splicing. PKR localizes in splicing complexes and associates with splicing factor SC35. Splicing and early-stage spliceosome assembly on β-globin pre-mRNA depend strictly on activation of PKR by a codon-containing RNA fragment within exon 1 and on phosphorylation of nuclear eIF2α on Serine 51. Nonphosphorylatable mutant eIF2αS51A blocked β-globin mRNA splicing in cells and nuclear extract. Mutations of the β-globin RNA activator abrogated PKR activation and profoundly affected mRNA splicing efficiency. PKR depletion abrogated splicing and spliceosome assembly; recombinant PKR effectively restored splicing. Excision of the first intron of β-globin induces strand displacement within the RNA activator of PKR by a sequence from exon 2, a structural rearrangement that silences the ability of spliced β-globin mRNA to activate PKR. Thus, the ability to activate PKR is transient, serving solely to enable splicing. α-Globin pre-mRNA splicing is controlled likewise but positions of PKR activator and silencer are reversed, demonstrating evolutionary flexibility in how PKR activation regulates globin mRNA splicing through eIF2α phosphorylation.

Tumor necrosis factor alpha (TNF-α) is expressed promptly during inflammatory responses. Efficient TNF-α mRNA splicing is achieved through a 3' UTR element that activates RNA-dependent eIF2α protein kinase (PKR). The TNF-α RNA activator, we show, folds into a pseudoknot conserved from teleost fish to humans, critical for PKR activation and mRNA splicing. The pseudoknot constrains the RNA into two double-helical stacks having parallel axes, permitting facile PKR dimerization and trans-autophosphorylation needed for kinase activation. Mutations show that the PKR activator potently enhances splicing without inhibiting translation. eIF2α phosphorylation represses translation and is essential for coping with cellular stress, yet PKR-enabled TNF mRNA splicing depends strictly on eIF2α phosphorylation. Indeed, eIF2α phosphorylation at Serine51 is necessary and sufficient to achieve highly efficient splicing, extending its role from negative control of translation to positive control of splicing. This mechanism, operational in human peripheral blood mononuclear cells (PBMCs), links stress signaling to protective immunity through TNF mRNA splicing rendered efficient upon eIF2α phosphorylation.