Malo, Courtney S.

Abstract α-PD-1 checkpoint blockade is a promising immunotherapeutic approach for many solid tumors; however, this strategy is minimally effective for GBM patients as a monotherapy. One potential reason why α-PD-1 checkpoint blockade is ineffective is the tumor microenvironment and systemic derangements of the immune system which are prevalent in GBM patients. We have successfully modeled peripheral immune derangements in experimental murine GBM models. These include low peripheral blood CD4 T cell counts, reduced MHC class II expression on monocytes, and atrophy of primary immune organs in animals harboring gliomas. We therefore hypothesized that extended half-life IL-2, a potent cytokine which promotes the proliferation, differentiation, and killing activity of T cells, could overcome immune derangements and could potentially synergize with α-PD-1 checkpoint blockade therapy. In cohorts of C57Bl/6 mice, we supplemented α-PD-1 checkpoint blockade with multiple bio-engineered extended half-life IL-2 molecules to treat GL261 and CT2A orthotopic glioma models which are normally resistant to monotherapy. We observe a statistically significant increase in survival in GL261-bearing animals receiving the combination of extended half-life IL-2 reagents and αPD-1 therapy relative to control-treated mice. Some animals receiving this combination therapy had a reduction in tumor burden to levels below detection. These long-term survivors underwent a rechallenge experiment with a second inoculation of GL261-Luc and effectively cleared tumor without further therapeutic intervention. This intervention causes a significant shift in the immune profile, including the formation of mature T cell responses, as well as restoring the diminished peripheral immune cell counts in GL261-bearing animals. Finally, combining extended half-life IL-2 with α-PD-1 checkpoint blockade is effective independently of the CD8 T cell response. Together, these data suggest that combination immunotherapies employing extended half-life IL-2 fusion proteins are translational to the clinic and do not require the identification of tumor specific antigens for efficacy

Abstract Glioblastoma (GBM) is the most common malignant brain tumor in adults, responsible for approximately 225,000 deaths per year. Despite preclinical successes, most interventions have failed to extend patient survival by more than a few months. Treatment with anti—programmed cell death protein 1 (anti–PD-1) immune checkpoint blockade (ICB) monotherapy has been beneficial for malignant tumors such as melanoma and lung cancers but has yet to be effectively employed in GBM. This study aimed to determine whether supplementing anti–PD-1 ICB with engineered extended half-life IL2, a potent lymphoproliferative cytokine, could improve outcomes. This combination therapy, subsequently referred to as enhanced checkpoint blockade (ECB), delivered intraperitoneally, reliably cures approximately 50% of C57BL/6 mice bearing orthotopic GL261 gliomas and extends median survival of the treated cohort. In the CT2A model, characterized as being resistant to CBI, ECB caused a decrease in CT2A tumor volume in half of measured animals similar to what was observed in GL261-bearing mice, promoting a trending survival increase. ECB generates robust immunologic responses, features of which include secondary lymphoid organ enlargement and increased activation status of both CD4 and CD8 T cells. This immunity is durable, with long-term ECB survivors able to resist GL261 rechallenge. Through employment of depletion strategies, ECB's efficacy was shown to be independent of host MHC class I–restricted antigen presentation but reliant on CD4 T cells. These results demonstrate ECB is efficacious against the GL261 glioma model through an MHC class I–independent mechanism and supporting further investigation into IL2-supplemented ICB therapies for tumors of the central nervous system.

Abstract Glioblastoma multiforme (GBM) is a deadly CNS malignancy with an average survival of ~12–18 months post diagnosis. The unique immune environment of the brain and the immunosuppressive features of GBM complicate the use of immunotherapies against these tumors. We aimed to promote the efficacy of αPD-1 against the GL261 murine model of GBM by supplementing this therapy with an engineered IL-2 cytokine. IL-2 therapy has been used clinically for metastatic melanoma and renal cancer, but its short half-life in circulation necessitated high doses to keep the drug concentration sufficiently high. Fusing IL-2 to the mouse serum albumin molecule (MSA-IL-2), extends the cytokine’s half-life. Neither αPD-1 nor MSA-IL-2 are effective monotherapies for GL261 tumors, but the combination provides durable tumor clearance and prevents subsequent rechallenge. Most strikingly, this combination therapy cleared established GL261 tumors even in mice incapable of MHC class I restricted antigen presentation. Instead, therapeutic efficacy was abrogated by the depletion of CD4 T cells. Our combination therapy also reversed multiple key features of GBM-associated peripheral immune suppression, including depressed CD4 T cell counts and dampened MHC class II expression. Our combination therapy’s ability to overcome these immune derangements associated with glioma growth and to function independent of MHC class I restricted antigen presentation enhance the appeal of this therapy for future translational adaptation. This work was supported by the National Institutes of Health (R01 NS103212) and the Mayo Clinic-Koch Institute Cancer Solutions Team Grant

Abstract Cytotoxic T cell infiltration into the central nervous system (CNS) is required for clearance of pathogen or tumor. Whether microglia, the resident innate immune cells of the brain, are involved in orchestrating CD8 T cell infiltration of the brain in an antigen dependent manner remains to be elucidated. We hypothesized that antigen presentation by major histocompatibility complex (MHC) class I molecules expressed by microglia is required to orchestrate antigen specific CD8 T cell infiltration into the brain parenchyma. To test this hypothesis, we generated a novel tamoxifen-inducible microglial MHC class I conditional knockout mouse (CX3CR1-creERT2H-2Kb cKO). In these mice, administration of tamoxifen mediates deletion of the H-2Kbmolecule in cells expressing CX3CR1. Short-lived peripheral CX3CR1+ cells repopulate within 4–6 weeks after tamoxifen treatment and restore expression of H-2Kb, leaving only long-lived microglia deficient for MHC class I. We challenged these mice with intracranial infection of Theiler’s murine encephalomyelitis virus expressing the model antigen ovalbumin (TMEV-OVA) to assess antigen specific CD8 T cell responses within the brain parenchyma. In microglia MHC I cKO mice, the frequency of brain infiltrating OVA-specific CD8 T cells was attenuated while the frequency of intravascular OVA-specific CD8 T cells was increased when compared to cre− littermate controls. These results suggest that microglial antigen presentation is important for antigen specific CD8 T cell infiltration into the brain parenchyma. These studies directly address the critical role of microglial MHC class I on antigen specific CD8 T cell responses to an in vivo neuroinflammatory insult.

Abstract CD8 T cell priming after viral infection of the central nervous system (CNS) is incompletely understood. In peripheral infections, T cell priming generally relies upon stimulation by antigen-bearing dendritic cells (DCs) migrating into draining lymph nodes. The parenchyma of the brain, however, is largely devoid of dendritic cells and evidence demonstrating that activated DCs leave the brain parenchyma is lacking. We therefore sought to understand the role of DCs and specific MHC class I molecules in priming CD8 T cells following CNS infection. We employed H-2Db−/− H-2Kb−/− (MHC-I deficient) mice that express a floxed H-2Db or floxed Kb molecule, a unique transgenic mouse system developed by our research program. Crossed with an MHC class I deficient CD11c-Cre expressing animal results in DC-specific ablation of Db or Kb Class I restricted antigen presentation ability. We found that CD11c+ APCs are critical for the early priming of CD8 T cells against the Db restricted VP2121–130 epitope of the model pathogen, TMEV. However, DCs are ultimately dispensable for long-term development of these antigen specific T cells. Discordantly, the loss of MHC class I restricted antigen presentation ability on CD11c+ APCs does not impact the priming of T cells against the H-2Kb restricted OVA257–264 epitope following intracranial TMEV-OVA infection, but does impact the ability of these T cells to successfully occupy the CNS. This work suggests that differing patterns of MHC class I restriction for CNS pathogen-derived epitopes might have functional consequences on the APC requirements for a successful CD8 T cell response against neurotropic viruses.

Abstract Cerebral malaria (CM) is a severe complication arising from Plasmodium falciparum infection which results in high mortality. Lack of mechanistic insight into this pathology has resulted in ineffective therapeutic strategies. Using the Plasmodium berghei ANKA model of experimental cerebral malaria (ECM) our lab recapitulated hallmarks of human CM using small animal imaging, including blood-brain barrier (BBB) permeability, severe edema and microhemorrhage visible via T1 and T2-weighted MRI. Previously it had been demonstrated that infiltration of CD8 T cells with cytotoxic capacity is required to induce BBB permeability and ECM. The observation that not only infiltration, but cytotoxic killing is required provoked us to analyze differential MHC class I-restricted priming among discrete antigen presenting cell (APC) types. Using new MHC class I conditional knockout mice generated by our team, we addressed CD8 T cell priming and capacity to disrupt the BBB in ECM. Our findings revealed that mice lacking the H-2Kb MHC class I molecule expression in CD11c+ cells, but not LysM+ cells, were protected from the development of lethal ECM. Subsequently, conditional ablation of the H-2Db on these same subsets yielded similar results. Therefore, we demonstrate that both Db and Kb class I molecule expression by CD11c+ cells enable CD8 T cell priming that results in BBB permeability during ECM. In contrast, CD8 T cells primed by H-2Kb or H-2Db expressing LysM+ APCs are not capable of inducing BBB disruption. These results imply discrete APC subsets prime functionally distinct CD8 T cell responses in the CNS and modulate the outcome of cerebral malaria. This work begins to bridge the mechanistic gap and will impact future therapeutic strategies.

Abstract Glioblastoma multiforme (GBM) is incurable and among the most lethal of cancers. Immunotherapy for GBM is hampered by the fact that GBM patients are severely immunosuppressed with dangerously low T cell counts. Importantly, immunotherapy for GBM will be ineffective without defining the mechanism of immunosuppression. Given that GBM does not metastasize, we investigated the possibility of neuro-immune interactions using the GL261 glioma model in C57BL/6 mice. C57BL/6 mice harboring GL261 gliomas have abnormal thymuses which are acutely and significantly involuted compared to healthy control mice. Glioma bearing mice also have significant thymic atrophy which increases with tumor burden. Further structural analysis of the thymus in glioma bearing mice revealed disrupted thymic epithelial cells and reduced EPCAM+ cell expression of MHC II. Thymic subsets in tumor bearing mice were also developmentally skewed with increased generation of single positive T cells and a block at double negative 2 (DN2) to DN3 transition. Peripherally, MHCII expression was also significantly decreased in blood derived cells in glioma bearing mice. This correlated with an overall reduction in total CD4 T cell numbers. The results of this study demonstrate the GL261 model is recapitulating the immune suppression observed in GBM patients. These results also support the presence of a brain-thymus interaction that could negatively impact the development of immunotherapy strategies to treat GBM.

Abstract Glioblastoma multiforme (GBM) is a deadly CNS malignancy with an average survival of ~12 months post diagnosis despite aggressive treatment. Immunotherapy approaches to treat GBM are being developed, but the unique immune environment of the brain and the immunosuppressive features of GBM have complicated early efforts. Even αPD-1 checkpoint blockade, highly useful in the context of melanoma, has proved ineffective in the treatment of GBM. In order to promote the efficacy of αPD-1 against the GL261 murine model of GBM, we supplemented this therapy with an engineered IL-2 cytokine. IL-2 therapy has been used clinically for metastatic melanoma and renal cancer, but its short half-life in circulation necessitated dangerously high doses to keep the drug concentration sufficiently high. Fusing IL-2 to the mouse serum albumin molecule (MSA-IL-2), extends the cytokine’s half-life. While neither αPD-1 nor MSA-IL-2 are effective monotherapies against the GL261 model tumor, we have found this combination immunotherapy to provide durable tumor clearance and to create immunologic memory capable of resisting GL261 rechallenge. Most strikingly, this combination therapy is independent of the CD8 T cell function, clearing established GL261 tumors in mice incapable of MHC class I restricted antigen presentation. Independence of MHC class I restricted antigen presentation makes this combination immunotherapy using αPD-1 checkpoint blockade both highly novel and translatable to clinical use.

Abstract Peripheral immune suppression is a hallmark of highly malignant Glioblastoma multiforme (GBM). Given that effective immunotherapy for GBM relies on a functional immune system, it is crucial to define the underlying mechanism of peripheral immunosuppression. We therefore evaluated immunosuppression using the GL261 model of GBM. We first examined cytokine production by microglia and brain-infiltrating lymphocytes for immune suppressing factors. We observed relatively unchanged levels of IL-10 and TGF-β in glioma bearing mice compared to controls. Therefore, we next investigated cellular components of the blood for evidence of suppression. However, we did not observe enhanced frequencies of Tregs (Foxp3+ CD4+) in tumor bearing mice. Finally, we analyzed cytokine levels in the serum of mice harboring gliomas compared to glioma negative animals. This assay revealed fluctuations in chemokine and cytokine levels between these two groups, prompting the analysis of a potential serum derived factor in promoting immunosuppression. We therefore evaluated T cell activation and proliferation in the presence of serum isolated from glioma harboring mice and healthy controls. CFSE-labeled T cells stimulated with CD3/CD28 beads in the presence of glioma negative mouse serum had normal proliferation. In contrast, CD4+ and CD8+ T cell proliferation was completely inhibited when cultured with serum from glioma bearing mice. These studies demonstrate the importance of a serum derived factor in mediating immunosuppression. Defining and inhibiting this circulating factor could be the first step towards improving immunotherapy strategies for GBM.

Abstract Blood brain barrier (BBB) disruption is a feature of multiple neurological disorders for which there is an increasingly recognized role for CD8+ T cells. Theiler’s murine encephalomyelitis virus (TMEV) infection serves as an excellent model to study CD8+ T cell mediated BBB disruption because of the well-characterized immune response to the immunodominant VP2121–130 peptide presented in the context of the H-2Db class I molecule. Intravenous injection of VP2121–130 peptide seven days post TMEV infection results in severe BBB disruption. BBB disruption is dependent on CD8+ T cell interaction with the Db: VP2121–130 epitope, but the antigen presenting cell required for BBB disruption is undefined. We hypothesized that dendritic cells are responsible for presentation of the immunodominant VP2121–130 peptide. We tested this hypothesis using a novel H-2Db Cre-LoxP conditional knockout mouse bred to MHC class I deficient CD11c-Cre expressing mice. We then evaluated these animals at 7 days post infection for induction of BBB disruption, following VP2121–130 peptide administration, using gadolinium enhanced MRI, fluorescent albumin leakage into the brain, and confocal microscopy for the tight junction protein occludin. Control CD11c-Cre(−) mice generated robust Db: VP2121–130 epitope specific CD8+ T cell responses and displayed extensive BBB disruption. In contrast, CD11c-Cre(+) littermates generated no brain infiltrating antiviral CD8+ T cell responses and were protected from BBB disruption. We conclude that CD11c+ dendritic cells are required for priming of CD8+ T cell responses and are necessary for BBB disruption during neuroinflammation in this model.