Despite studies highlighting a critical role for lysine-specific demethylase 1 (LSD1) in the formation of plasmablasts, the role of this enzyme in B cell development has not been explored. Haines et al. (p. 1867) demonstrated that B cell–conditional deletion of LSD1 decreased marginal zone B cell (MZB) populations, whereas follicular B cell (FoB) and bone marrow B cell populations were minimally impacted. RNA sequencing and principal component analysis revealed that LSD1 regulates the MZB transcriptional program by repressing genes normally upregulated in FoB and myeloid lineages. This repression was achieved via regulation of chromatin accessibility at transcription factor binding motifs known to guide splenic B cell development, including NF-κB. Additional studies demonstrated that LSD1-deficient B cells in which NF-κB was inhibited failed to undergo full MZB development, highlighting an important role for noncanonical NF-κB in this process. Consistent with these observations, LSD1 also was shown to regulate downstream target genes of noncanonical NF-κB signaling. Thus, this study establishes a clear role for LSD1 in MZB development and identifies cooperation between LSD1 and noncanonical NF-κB signaling in regulating downstream MZB developmental genes.

Heparin-induced thrombocytopenia (HIT) occurs when autoantibodies are produced against the platelet factor 4 (PF4)/heparin complex. Because regulatory T (Treg) cells are known to prevent the production of autoantibodies, Zheng et al. (p. 1786) sought to elucidate their role in HIT. Compared with wild-type (WT) controls, unmanipulated mice lacking Treg cells produced significantly higher numbers of functionally active, PF4/heparin-specific Abs. Adoptive transfer of Treg cells into newborn mice deficient in Treg cells suppressed production of PF4/heparin-specific Abs. To confirm these observations, the authors transferred either conventional CD4+ T cells or Treg cells into WT mice and challenged the recipients with the PF4/heparin complex. Whereas Treg cells reduced PF4/heparin-specific Ab production in recipient mice, mice given conventional CD4+ T cells showed expected PF4/heparin-specific Ab production 7 d after challenge. In support of a role for Treg cells in suppressing production of PF4/heparin-specific Abs, IL-10–deficient mice spontaneously produced these Abs. Additionally, CD4-specific deletion of IL-10 in mice led to significantly higher levels of PF4/heparin-specific Abs than controls challenged with PF4/heparin. These data suggest that Treg cells may have therapeutic potential for the treatment of HIT via their production of IL-10.

Previous studies have demonstrated an increase in proinflammatory macrophages and the microRNA 155 (miR-155) in patients with rheumatoid arthritis (RA). TNF inhibitors (TNFi) have been used as RA therapeutics, and effective treatment has been correlated with depletion of synovial macrophages. In this issue, Paoletti et al. (p. 1766) sought to elucidate the molecular mechanisms governing the reduction of synovial macrophages following TNFi treatment. Membrane TNF (mTNF) expression was increased on monocytes from RA patients and correlated with disease activity. Monocytes isolated from RA patients displayed impaired maturation into alternatively activated macrophage phenotypes (M2-like), whereas differentiation into classically activated macrophage phenotype (M1-like) was maintained. The defect in M2-like macrophages positively correlated with mTNF expression and was reversed by a monoclonal anti-TNF Ab, but not by a TNF soluble receptor. miR-155 was increased in M2-like macrophages from RA patients, except those treated with adalimumab. Transfection of monocytes from healthy donors with miR-155 resulted in a decrease in M2-like markers, whereas silencing of miR-155 in RA monocytes restored the defect of M2 polarization. Thus, this study demonstrates that an increase in monocyte mTNF and a defect in their capacity to differentiate into anti-inflammatory macrophages are associated with RA severity. Enhanced miR-155 expression in monocytes may account for this defect and may be reversed by monoclonal anti-TNF Abs. Targeting of miR-155 may, therefore, represent a new therapeutic strategy for the treatment of RA.

Natural killer cells are a powerful component of innate immunity, with roles in viral clearance and tumor immunity. In this issue, Piersma et al. (p. 1981) elucidate the timing of signals required for NK cell production of IFN-γ. Stimulation of NK cells with either soluble mAbs against activation receptors or with ligand-expressing cells also required IL-2 and IFN-β for robust IFN-γ production. Additional studies demonstrated that the cytokines acted directly on NK cells and that receptor activation after cytokine signaling resulted in optimal IFN-γ production. Consistent with these observations, cytokines induced Ifng transcription, while receptor stimulation induced translation of the IFN-γ protein. Finally, receptor-dependent translation of IFN-γ requires the proteasome–ubiquitin–IKK–TPL2–ERK pathway. Together, these data indicate that IFN-γ production via NK cells is tightly regulated to provide maximum effectiveness in infection and tumor prevention, while preventing autoimmunity and immunopathology.

The evolution of mammalian adaptive immunity from a primordial immune complex (PIC) to a complex system of MHC and Ag receptors is thought to have occurred in two major recombination events, 1R and 2R. In this issue, Ohta et al. (p. 1882) use comparative analysis of Xenopus laevis genes to identify how and when MHC and Ag receptors evolved. The authors propose that the CD1 region in mammals resulted from an en block translocation (MHCtrans) into huMHCpara-1. Additional analysis suggested that MHC class I (MHC I) emerged first in evolution, because MHC class II genes were not found outside of the bona fide MHC or paralogous regions. Ag receptor–like genes are present in both major and minor MHCpara regions and MHCtrans, indicating that they were present in the PIC prior to 1R. The linkage of Ag receptor–like and MHC I genes on chromosome regions that emerged after 1R provides further evidence that Ag receptor precursors present in the PIC predate the emergence of bona fide MHC I genes. All rearranging Ag receptors are likely derived from the huMHCpara-19 precursor and that invasion of the RAG transposon, likely after 2R, is responsible for the separation of the VJ region into separate V and J elements. Together, these data elucidate the evolutionary differences in the adaptive immune system of jawed versus jawless vertebrates.

Genome-wide association studies have highlighted a link between the autophagy genes ATG16L1 and ATG16L2 with autoimmune diseases such as Crohn disease and systemic lupus erythematosus. Interestingly, ATG16L2 is only present in mammals, suggesting a unique role in human disease for this homolog. Despite this known association, the contribution of ATG16L2 to canonical autophagy pathways and other cellular pathways is poorly understood. Khor et al. (p. 1820) generated mice deficient in ATG16L2 (Atg16l2−/−) and demonstrated that it is not a required component for the canonical autophagy pathway or the development and function of major hematopoietic subsets. Furthermore, deletion of ATG16L2 did not exacerbate T cell defects observed in ATG16L1-deficient animals. Similar to previous reports, loss of epithelial ATG16L1 conferred resistance to Citrobacter rodentium infection. Whereas loss of ATG16L2 alone did not affect susceptibility to infection, loss of both ATG16L1 and ATG16KL2 enhanced immunity to C. rodentium. Single-cell RNA sequencing of the intestinal epithelium demonstrated that ATG16L2-deficient mice have a reduction in the number of Paneth cells, with a concomitant increase in the number of early progenitors in enterocyte and enteroendocrine populations. Taken together, this study demonstrates that ATG16L2 may not have an obvious function in the lymphoid compartment but may have a previously unappreciated function in the intestinal epithelium.