Chimeric antigen receptor macrophages target and resorb amyloid plaques.

Substantial evidence suggests a role for immunotherapy in treating Alzheimer's disease (AD). While the precise pathophysiology of AD is incompletely understood, clinical trials of antibodies targeting aggregated forms of ß amyloid (Aß) have shown that reducing amyloid plaques can mitigate cognitive decline in patients with early-stage AD. Here, we describe what we believe to be a novel approach to target and degrade amyloid plaques by genetically engineering macrophages to express an Aß-targeting chimeric antigen receptor (CAR-Ms). When injected intrahippocampally, first-generation CAR-Ms have limited persistence and fail to significantly reduce plaque load, which led us to engineer next-generation CAR-Ms that secrete M-CSF and self-maintain without exogenous cytokines. Cytokine secreting "reinforced CAR-Ms" have greater survival in the brain niche and significantly reduce plaque load locally in vivo. These findings support CAR-Ms as a platform to rationally target, resorb, and degrade pathogenic material that accumulates with age, as exemplified by targeting Aß in AD.

Chimeric Antigen Receptor Macrophages Target and Resorb Amyloid Plaques in a Mouse Model of Alzheimer's Disease.

Substantial evidence suggests a role for immunotherapy in treating Alzheimer's disease (AD). Several monoclonal antibodies targeting aggregated forms of beta amyloid (Aß), have been shown to reduce amyloid plaques and in some cases, mitigate cognitive decline in early-stage AD patients. We sought to determine if genetically engineered macrophages could improve the targeting and degradation of amyloid plaques. Chimeric antigen receptor macrophages (CAR-Ms), which show promise as a cancer treatment, are an appealing strategy to enhance target recognition and phagocytosis of amyloid plaques in AD. We genetically engineered macrophages to express a CAR containing the anti-amyloid antibody aducanumab as the external domain and the Fc receptor signaling domain internally. CAR-Ms recognize and degrade Aß in vitro and on APP/PS1 brain slices ex vivo; however, when injected intrahippocampally, these first-generation CAR-Ms have limited persistence and fail to reduce plaque load. We overcame this limitation by creating CAR-Ms that secrete M-CSF and self-maintain without exogenous cytokines. These CAR-Ms have greater survival in the brain niche, and significantly reduce plaque load locally in vivo. These proof-of-principle studies demonstrate that CAR-Ms, previously only applied to cancer, may be utilized to target and degrade unwanted materials, such as amyloid plaques in the brains of AD mice.

Wiskott-Aldrich syndrome protein restricts cGAS/STING activation by dsDNA immune complexes.

Dysregulated sensing of self-nucleic acid is a leading cause of autoimmunity in multifactorial and monogenic diseases. Mutations in Wiskott-Aldrich syndrome protein (WASp), a key regulator of cytoskeletal dynamics in immune cells, cause autoimmune manifestations and increased production of type I IFNs by innate cells. Here we show that immune complexes of self-DNA and autoantibodies (DNA-ICs) contribute to elevated IFN levels via activation of the cGAS/STING pathway of cytosolic sensing. Mechanistically, lack of endosomal F-actin nucleation by WASp caused a delay in endolysosomal maturation and prolonged the transit time of ingested DNA-ICs. Stalling in maturation-defective organelles facilitated leakage of DNA-ICs into the cytosol, promoting activation of the TBK1/STING pathway. Genetic deletion of STING and STING and cGAS chemical inhibitors abolished IFN production and rescued systemic activation of IFN-stimulated genes in vivo. These data unveil the contribution of cytosolic self-nucleic acid sensing in WAS and underscore the importance of WASp-mediated endosomal actin remodeling in preventing innate activation.

Isoform-Specific Expression and Feedback Regulation of E Protein TCF4 Control Dendritic Cell Lineage Specification.

The cell fate decision between interferon-producing plasmacytoid DC (pDC) and antigen-presenting classical DC (cDC) is controlled by the E protein transcription factor TCF4 (E2-2). We report that TCF4 comprises two transcriptional isoforms, both of which are required for optimal pDC development in vitro. The long Tcf4 isoform is expressed specifically in pDCs, and its deletion in mice impaired pDCs development and led to the expansion of non-canonical CD8+ cDCs. The expression of Tcf4 commenced in progenitors and was further upregulated in pDCs, correlating with stage-specific activity of multiple enhancer elements. A conserved enhancer downstream of Tcf4 was required for its upregulation during pDC differentiation, revealing a positive feedback loop. The expression of Tcf4 and the resulting pDC differentiation were selectively sensitive to the inhibition of enhancer-binding BET protein activity. Thus, lineage-specifying function of E proteins is facilitated by lineage-specific isoform expression and by BET-dependent feedback regulation through distal regulatory elements.

Mechanism of processing of the NF-kappa B2 p100 precursor: identification of the specific polyubiquitin chain-anchoring lysine residue and analysis of the role of NEDD8-modification on the SCF(beta-TrCP) ubiquitin ligase.

Processing of the NF-kappa B2 precursor p100 to the mature p52 subunit is regulated via a unique pathway. NF-kappa B-inducing kinase (NIK) induces I kappa B kinase alpha (IKK alpha)-mediated phosphorylation of specific serine residues in the C-terminal domain of p100, leading to recruitment of the SCF(beta-TrCP) ubiquitin ligase. We identified a single lysine residue, K855, that serves as the ubiquitin-anchoring residue required for signal-induced processing of p100. In a reconstituted system containing purified components, p100-K855R could not be ubiquitinated. In a crude extract and cells, only residual, signal-independent ubiquitination and processing were retained. Importantly, K855 is located in a site homologous to K22 that serves as an ubiquitination site in I kappa B alpha. This suggests a common recognition mechanism for the two molecules. In contrast, p105, the p100 homologue, lacks a similar Lys residue. We also demonstrate that the NEDD8 pathway is essential for the SCF(beta-TrCP) activity. In a reconstituted system, efficient ubiquitination of p100 required all three components of the pathway - E1, the UBC12 E2 and NEDD8. Experiments in reconstituted systems and in cells demonstrated that SCF(beta-TrCP), which contains a mutant Cul-1 that cannot be NEDDylated, cannot stimulate ubiquitination and processing. Similarly, dominant negative UBC12 inhibits, in a reversible manner, both ubiquitination and processing of p100.