Clarkson, Ben D.S.

Abstract We recently reported normalization of CSF cytokine levels following successful treatment of drug-resistant epilepsy (DRE) with recombinant interleukin-1 receptor antagonist (rIL1RA). To further understand the mechanism of this therapeutic outcome, we interrogated the ability of serum and CSF factors that are elevated in drug resistant-seizure disorders to promote inflammatory responses in astrocytes and human brain microvessel endothelial cells (HBMVECs). We report that CCL2, TNFa, IL6, and CXCL8/1 were strongly induced in 1) astrocytes and HBMVECs stimulated with IL1β (blocked by rIL1RA) and 2) astrocytes but not HBMVECs in response to DRE patient serum but not control serum (partially blocked by rIL1RA). This suggested that other serum factors contribute to astrocyte activation in this DRE. HMGB1 and IL18 were also dysregulated in DRE patient serum; however, IL18 failed to stimulate cytokine secretion from astrocytes. In contrast, HMGB1 induced cytokine secretion from astrocytes, which was blocked by the TLR4 inhibitor TAK242. Remarkably, TAK242 also prevented cytokine production by astrocytes in response to serum from some patients with DRE. Finally, IL1β induced the polarized uptake and secretion of HMGB1 across transwell inserts harboring confluent monolayers of HBMVECs suggesting that IL1b signaling may cause endothelial cells to transport serum HMGB1 into the perivascular space—promoting TLR4-dependent cytokine release from astrocytes. This translation of an “outside” inflammatory signal into an “inside” inflammatory response may underlie seizure disorders that are resistant to anti-seizure drugs.

Abstract In diseases such as multiple sclerosis and autoimmune encephalitides, activated CD8+ cytotoxic T cells infiltrate the brain and are found in clonally expanded clusters that outnumber CD4+ cells in progressive stages of disease. At present, the pathophysiological role of these cells is poorly understood and therapeutic strategies for preventing or ameliorating autoinflammatory disease progression are limited. T cells exhibit profound metabolic reprogramming that facilitates the energetic and proliferative demands of effector T cells. Key metabolic pathways in this reprogramming include upregulation of glycolysis and the pentose phosphate pathway (PPP). Here we show that 1) pharmacologic inhibition of the PPP suppresses effector phenotype acquisition and proinflammatory cytokine secretion by CD8+ T cells; 2) engagement of the PPP is enhanced by pharmacologic Nrf2 activation, implicating reactive oxygen species-induced activation as a transcriptional regulator of PPP-dependent immune activation; 3) Nrf2-driven PPP flux promotes the acquisition of specific CD8+ effector phenotypes; 4) limiting the metabolic flux of glucose-derived metabolites through the PPP prevents CD8+ T cell-mediated injury of neurons expressing cognate antigen and preserves their electrophysiological function; and 5) pharmacologic inhibition of the PPP in B6 mice at symptom onset is sufficient to delay and ameliorate motor disability in the MOG EAE preclinical model of MS. Overall, these results underscore the potential of immunometabolic therapies to limit T cell mediated injury in autoinflammatory and autoimmune diseases across organ systems including the prevention of neurologic deficits in patients with T cell driven neurologic disease. This work was supported by the NIH (R01NS115126 to CLH; T32GM065841 to EMG), the Mayo Clinic Graduate School of Biomedical Sciences, the Mayo Clinic Center for MS and Autoimmune Neurology, and generous donations from Eugene and Marcia Applebaum, Donald J. and Frances I. Herdrich, Bob and Loralee Marvin, and Rob and Debbie Mudge.

Abstract Cognitive impairment in MS is associated with diffuse gray matter injury, which remains poorly understood. Given that axons span both white and gray matter regions, it’s possible that signals within these axons contribute to the spread of diffuse injury in the brain. We show that retrograde IFNg signaling in axons (in vitro human and murine) and demyelination (murine in vivo) cause transcriptional / translational changes in neuronal cell bodies. Chief among the candidate signaling pathways identified is ISGylation, a process whereby a multitude of cellular proteins are modified by the attachment of one or more ISG15 molecules. Using custom adeno-associated viral vectors we tested how altering ISGylation in neurons affects neuronal synaptic function and how it affects neuronal responsiveness to specific inflammatory factors in cell cultures. We also measured ISGylation in MS brain tissues to see if neuronal ISGylation correlates with gray matter injury in MS. We report early evidence that increased neuron ISGylation alters the composition of neuron-derived extracellular vesicles. Microglia treated with extracellular vesicles from neurons overexpressing ISGylation pathway genes exhibited morphology consistent with increased phagocytic activity and increased mRNA expression of inflammatory cytokines including: CCL2, IL1b, TNFa, iNOS, CXCL10, CCL5, and IL6. At this point, it is still unclear how these changes are related to gray matter pathology. However, it is possible that microglial activation by these vesicles is involved with driving synapse loss as activated microglia are known to be capable of stripping neuronal synapses.

Abstract Cytotoxic CD8+ T cells outnumber CD4+ T cells within multiple sclerosis (MS) lesions, yet the pathogenic role for these cells and the relevant target antigen(s) involved in neuropathology remain unknown. We predict that a subset of lesion-infiltrating CD8+ T cells are specific for neuronal antigens and contribute to axon or neuron injury, thereby driving disease progression and accumulation of disability. To test this, we are developing protocols to measure MS patient-derived T cell reactivity against neuronal and control antigens. Peripheral blood mononuclear cells (PBMCs) isolated from whole blood were incubated with GM-CSF and IL-4 to selectively expand dendritic cells (DCs) or with IL-7 to promote survival and maintenance of T cells. After resting, patient T cells were combined 1:1 with autologous DCs in the presence of vaccine antigens or peptides derived from common pathogens. We tested responses against a pool of cytomegalovirus, Epstein Barr virus, and influenza peptides (CEF) and against mumps viral protein extract. We also tested responses in T cells derived from patients with autoimmune neurological disorders following incubation with autologous DCs and disease-specific proteins. Activation profiles were assessed using CD69, CD44, and LFA-1 by conventional flow cytometry. Cellular responses were also assessed using mass cytometry (cytometry by time of flight; CyTOF). Our next steps include testing MS patient-derived T cell responses against neuronal and axonal protein libraries, with the goal of identifying potential autoantigens that underlie disease progression.