LFA-1 Activation on NK Cells: A Sticky Situation See article p.1944

Antenatal IL-1 Exposure and Postnatal Development See article p.2047

MLKL Activation and NLRP3-Mediated IL-1β Release See article p.2156

Defining Th22 Cells as Distinct See article p.2182

Improved Database for Human Immune Receptor Genes See article p.2202

Preterm birth (PTB), a leading cause of infant mortality and morbidity, is commonly accompanied by in utero fetal inflammation, but the only treatments currently available do not affect inflammatory mediators. Although IL-1 production appears central to PTB, studies to date primarily describe the harmful role of IL-1 in the postnatal period. In this issue, Nadeau-Vallée et al. (p. 2047) elucidated the effects of antenatal exposure to IL-1 on postnatal development and determined the efficacy of IL-1 receptor antagonists as potential therapeutics to treat PTB. Compared with untreated controls, preterm delivery induced by intrauterine administration of IL-1β to pregnant mice during late gestation resulted in elevated expression of genes encoding proinflammatory factors such as Tnfα, Il-6, Ccl2, and Cox2 in the placenta, amniotic fluid, and fetal membranes. Maternal inflammation appeared to disseminate to the fetus, as surviving neonates of IL-1β–induced PTB demonstrated noticeable morphological alterations in the lung, intestine, and brain in addition to increased IL1b, Il6, Il8, Il10, Pghs2, Tnfa, and Crp expression in WBCs and elevated levels of IL-1β, IL-6, and IL-8 in the plasma and tissues. Treatment of pregnant dams late in gestation with 101.10, a noncompetitive IL-1 receptor antagonist, reversed inflammation and adverse outcomes in both LPS- and IL-1β–induced PTB, whereas treatment with anakinra (Kineret), a competitive IL-1 receptor antagonist, demonstrated only modest effects on gestational outcome and placental inflammation. Additionally, pups of dams receiving 101.10 demonstrated no changes in postnatal development when compared with pups from untreated dams. Taken together, these data implicate IL-1 as a significant inflammatory mediator of morbidity in PTB and demonstrate that suppression of IL-1 signaling using peptide inhibitors such as 101.10 may be a safe and effective therapeutic for the prevention of PTB.

Formation of a stable immunological synapse between NK cells and target cells, initiated by the integrin LFA-1, is an important step in NK cell cytotoxicity. A unique property of integrins is that adhesiveness, controlled by the molecule’s conformation and clustering within the membrane, can be regulated by inside-out signaling of activating receptors. In this issue, Urlaub et al. (p. 1944) analyze the signals that are involved in activation of LFA-1 on NK cells. Using a ligand complex–based adhesion assay to detect changes in LFA-1 avidity and affinity for ICAM-1, the authors demonstrated that although LFA-1 has low binding activity in resting human NK cells, LFA-1 can be stimulated by triggering of the SLAM family activating receptors 2B4 and NTB-A, or pairwise stimulation of 2B4 and NTB-A with other activating receptors such as NKp30, NKp46, or NKG2D. Incubation of freshly isolated NK cells with cytokines IL-15, IL-12, and IL-18, either alone or in combination, did not activate LFA-1 when compared with untreated cells. Alternatively, pretreatment of NK cells with cytokines prior to stimulation of activating receptors such as 2B4 enhanced LFA-1 activation. Coengagement of the inhibitory NK cell receptor KIR2DL/3 on NK cells stimulated via 2B4 reduced activation of LFA-1, demonstrating that inhibitory KIR receptors can interfere with inside-out signaling of activating receptors. The CD56bright, CD56dim/CD57, and CD56dim/CD57+ subsets of NK cells express similar levels of activating receptors and LFA-1, but differ in their ability to activate LFA-1 following 2B4 stimulation. Interestingly, the two CD56dim subsets, which possess higher cytotoxic potential than the others, showed the highest activation of LFA-1. Proteomic analysis of the NK cell subsets revealed that expression of the small calcium binding protein S100A4, which is involved in signal transduction and control of the cytoskeleton, increased as NK cells matured and that this upregulated expression correlated with the activation status of LFA-1. These findings suggest that S100A4 is needed to properly convert the inside-out signal from activating NK receptors to cytoskeleton remodeling and LFA-1 activation. Taken together, these data show that differentiation status of NK cells modulates LFA-1 binding activity by regulating inside-out signaling pathways of activating receptors.

Necroptosis, a form of lytic programmed cell death, is defined by the activation of receptor interacting protein kinase (RIPK)1 and RIPK3 and the downstream effector mixed lineage kinase domain-like (MLKL). Once activated, MLKL translocates to the cell membrane where it triggers pore formation and membrane disruption. Although activation of MLKL has been associated with the processing and release of bioactive IL-1β, potentially via activation of the NLRP3 inflammasome, studies to characterize this association have proved difficult, as induction of necroptosis requires caspase inhibition, which in turn precludes the activation of inflammasome effector NLRP3-mediated caspase-1. To circumvent this issue, Gutierrez et al. (p. 2156) constructed an activatable form of MLKL (acMLKL) that could be directly stimulated using a small molecule ligand. Activation of acMLKL in LPS-primed THP-1 cells triggered cleavage of IL-1β, which appeared to be caspase-1 dependent, as cleaved IL-1β was absent in supernatants of identically treated THP-1 cells lacking caspase-1. acMLKL-induced cell death, while shown to be independent of a potassium gradient across the cell membrane, was found to trigger potassium efflux, membrane disruption, and release of IL-1β following acMLKL activation that was reduced by an increase in extracellular potassium. When compared with cells activated using canonical inflammasome activators, acMLKL-activated cells showed a similar frequency of inflammasome formation. Additionally, acMLKL-induced IL-1β processing was abolished in cells deficient in either NLRP3 or ASC, suggesting that potassium efflux triggered by MLKL-induced cell death activates the NLRP3 inflammasome and downstream IL-1β processing. Finally, deletion of gasdermin-D (GSDMD), a key effector in the secretion of IL-1β following inflammasome activation, had no effect on IL-1β processing following acMLKL-induced activation, indicating that NLRP3 inflammasome activation following MLKL-induced cell death is independent of GSDMD. Together, these findings suggest a mechanism for IL-1β release following MLKL activation and demonstrate a mechanistic link between the necroptotic program, NLRP3 inflammasome activation, and IL-1β release.

Next generation sequencing of genes encoding human TCRs and Igs allows analysis of immune repertoires by aligning sequences from an individual with a reference database of immune receptor genes and can generate important insights into adaptive immunity. Multiple alignment algorithms have been investigated, and most reference data come from the International ImMunoGeneTics Database (IMGT). However, the majority of the sequences in IMGT were obtained 20 or more years ago and the resultant database appears to be incomplete and to contain a variety of sequence errors. Yu et al. (p. 2202) have worked to establish a more comprehensive and accurate database by developing a pipeline to acquire immune receptor sequence data from population resequencing studies. Using genomic data from the 1000 Genomes Project and the AlleleMiner pipeline, the authors collected TCR and Ig sequences from 2504 individuals and merged these with the sequences in IMGT to create the Lym1K database of immune receptor alleles. TCR and Ig alleles obtained from genomic data, when compared with those included in IMGT, were found to include the vast majority of known immune receptor alleles, as well as large numbers of previously unidentified putative alleles. V and J gene segments of human immune receptors from short-read sequence datasets more successfully aligned with the Lym1K database than the IMGT database, suggesting that the Lym1K database may be a more useful reference database than IMGT. The 1000 Genomes Project includes genomes from 26 different global populations of individuals and, as such, the Lym1K database was found to capture more diversity in immune receptor gene sequences, particularly of African populations, than could be found in the IMGT database. Development of this more inclusive reference database of human immune receptor gene sequences, and description of the pipeline used to generate these data, should greatly facilitate future studies of immune repertoires and their effects on a variety of immune situations.

Compared with other Th cell subsets, much less is known about the differentiation requirements and functions of the IL-22–producing Th22 cell subset, mainly due to difficulties in obtaining populations of these cells free of contaminating Th17 cells. To more thoroughly characterize the Th22 population, Plank et al. (p. 2182) developed dual IL-22/IL-17A reporter mice that allowed them to isolate and compare Th22 and Th17 cells. Testing a variety of in vitro differentiation conditions on cells from these mice while monitoring cytokine expression revealed that naive CD4+ T cells could be polarized into Th22 cells using IL-6, IL-23, IL-1β, FICZ, and the TGF-β inhibitor Galunisertib, along with anti–IFN-γ and anti–IL-4 to block Th1 and Th2 cell differentiation. The resultant Th22 cells showed no traces of having expressed IL-17 during differentiation, suggesting that the Th22 cell population is fully distinct from the Th17 lineage. Comparison of transcriptional profiles of in vitro–differentiated Th17 and Th22 cells identified 287 differentially expressed genes, and those upregulated in Th22 cells included Il22, Tbx21, genes encoding granzymes (particularly Gzmb), and Il13. Use of cells from mice deficient in either RORγt or T-bet revealed that loss of RORγt partially blocked Th22 cell differentiation while strongly inhibiting Th17 cell differentiation, whereas loss of T-bet strongly promoted Th22 cell differentiation but did not affect that of Th17 cells. In vitro–differentiated Th22 cells maintained their phenotype when cultured under Th0 or Th17 conditions but demonstrated plasticity toward the Th1 and Th2 phenotypes when cultured under Th1- and Th2-promoting conditions, respectively. This plasticity was also observed in vivo, as in vitro–differentiated OVA-specific Th22 cells adoptively transferred into the Th1 environment of mice infected with an OVA-expressing virus stopped producing IL-22 and upregulated IFN-γ production. This study has advanced our understanding of the differentiation requirements, potential functions, and plasticity of Th22 cells differentiated in vitro and has defined them as a cell lineage that is fully distinct from Th17 cells.