Microglia shield the murine brain from damage mediated by the cytokines IL-6 and IFN-α

Sustained production of elevated levels of the cytokines interleukin (IL)-6 or interferon (IFN)-α in the central nervous system (CNS) is detrimental and directly contributes to the pathogenesis of neurological diseases such as neuromyelitis optica spectrum disorders or cerebral interferonopathies, respectively. Using transgenic mice with CNS-targeted production of IL-6 (GFAP-IL6) or IFN-α (GFAP-IFN), we have recently demonstrated that microglia are prominent target and effector cells and mount stimulus-specific responses to these cytokines. In order to further clarify the phenotype and function of these cells, we treated GFAP-IL6 and GFAP-IFN mice with the CSF1R inhibitor PLX5622 to deplete microglia. We examined their ability to recover from acute microglia depletion, as well as the impact of chronic microglia depletion on the progression of disease. Following acute depletion in the brains of GFAP-IL6 mice, microglia repopulation was enhanced, while in GFAP-IFN mice, microglia did not repopulate the brain. Furthermore, chronic CSF1R inhibition was detrimental to the brain of GFAP-IL6 and GFAP-IFN mice and gave rise to severe CNS calcification which strongly correlated with the absence of microglia. In addition, PLX5622-treated GFAP-IFN mice had markedly reduced survival. Our findings provide evidence for novel microglia functions to protect against IFN-α-mediated neurotoxicity and neuronal dysregulation, as well as restrain calcification as a result of both IL-6- and IFN-α-induced neuroinflammation. Taken together, we demonstrate that CSF1R inhibition may be an undesirable target for therapeutic treatment of neuroinflammatory diseases that are driven by elevated IL-6 and IFN-α production.

Contribution of STAT1 to innate and adaptive immunity during type I interferon-mediated lethal virus infection

Signal transducers and activators of transcription (STAT) 1 is critical for cellular responses to type I interferons (IFN-Is), with the capacity to determine the outcome of viral infection. We previously showed that while wildtype (WT) mice develop mild disease and survive infection with lymphocytic choriomeningitis virus (LCMV), LCMV infection of STAT1-deficient mice results in a lethal wasting disease that is dependent on IFN-I and CD4⁺ cells. IFN-Is are considered to act as a bridge between innate and adaptive immunity. Here, we determined the relative contribution of STAT1 on innate and adaptive immunity during LCMV infection. We show that STAT1 deficiency results in a biphasic disease following LCMV infection. The initial, innate immunity-driven phase of disease was characterized by rapid weight loss, thrombocytopenia, systemic cytokine and chemokine responses and leukocyte infiltration of infected organs. In the absence of an adaptive immune response, this first phase of disease largely resolved resulting in survival of the infected host. However, in the presence of adaptive immunity, the disease progressed into a second phase with continued cytokine and chemokine production, persistent leukocyte extravasation into infected tissues and ultimately, host death. Overall, our findings demonstrate the key contribution of STAT1 in modulating innate and adaptive immunity during type I interferon-mediated lethal virus infection.

The mammalian immune system is divided into innate and adaptive immunity. In response to harmful agents, innate immunity acts first, followed by late-acting, specialized, adaptive immunity. Type I interferons (IFN-Is) are important means of communication between innate and adaptive immunity. IFN-Is mediate their effects via a number of signaling molecules, principally including signal transducers and activators of transcription 1 (STAT1). The importance of STAT1 to the immune response is evident from our previous finding that mice deficient in STAT1 develop a lethal, host immunity-mediated disease following infection with the otherwise harmless lymphocytic choriomeningitis virus (LCMV). In the present study, we characterized the role of STAT1 in protecting against harmful host immune responses against LCMV. We report that STAT1 plays a significant role in lessening both the early, inflammatory responses of innate immunity and the sustained, destructive actions of adaptive immunity. These findings exemplify the extent of STAT1’s role as a key immune response modulating factor.

Zika virus encephalitis in immunocompetent mice is dominated by innate immune cells and does not require T or B cells

Background

Until the end of the twentieth century, Zika virus (ZIKV) was thought to cause a mostly mild, self-limiting disease in humans. However, as the geographic distribution of ZIKV has shifted, so too has its pathogenicity. Modern-day ZIKV infection is now known to cause encephalitis, acute disseminated encephalomyelitis, and Guillain-Barré syndrome in otherwise healthy adults. Nevertheless, the underlying pathogenetic mechanisms responsible for this shift in virulence remain unclear.

Methods

Here, we investigated the contribution of the innate versus the adaptive immune response using a new mouse model involving intracranial infection of adult immunocompetent mice with a moderately low dose of ZIKV MR766. To determine the contribution of type I interferons (IFN-Is) and adaptive immune cells, we also studied mice deficient for the IFN-I receptor 1 (Ifnar1^−/−) and recombination-activating gene 1 (Rag1^−/−).

Results

We show that intracranial infection with ZIKV resulted in lethal encephalitis. In wild-type mice, ZIKV remained restricted predominantly to the central nervous system (CNS) and infected neurons, whereas astrocytes and microglia were spared. Histological and molecular analysis revealed prominent activation of resident microglia and infiltrating monocytes that were accompanied by an expression of pro-inflammatory cytokines. The disease was independent of T and B cells. Importantly, unlike peripheral infection, IFN-Is modulated but did not protect from infection and lethal disease. Lack of IFN-I signaling resulted in spread of the virus, generalized inflammatory changes, and accelerated disease onset.

Conclusions

Using intracranial infection of immunocompetent wild-type mice with ZIKV, we demonstrate that in contrast to the peripheral immune system, the CNS is susceptible to infection and responds to ZIKV by initiating an antiviral immune response. This response is dominated by resident microglia and infiltrating monocytes and macrophages but does not require T or B cells. Unlike in the periphery, IFN-Is in the CNS cannot prevent the establishment of infection. Our findings show that ZIKV encephalitis in mice is dependent on the innate immune response, and adaptive immune cells play at most a minor role in disease pathogenesis.

Electronic supplementary material

The online version of this article (10.1186/s12974-019-1566-5) contains supplementary material, which is available to authorized users.

Microglia have a more extensive and divergent response to interferon-α compared with astrocytes

Type I interferons (IFN-I) are crucial for effective antimicrobial defense in the central nervous system (CNS) but also can cause severe neurological disease (termed cerebral interferonopathy) as exemplified by Aicardi-Goutières Syndrome. In the CNS, microglia and astrocytes have essential roles in host responses to infection and injury, with both cell types responding to IFN-I. While the IFN-I signaling pathways are the same in astrocytes and microglia, the extent to which the IFN-I responses of these cells differ, if at all, is unknown. Here we determined the global transcriptional responses of astrocytes and microglia to the IFN-I, IFN-α. We found that under basal conditions, each cell type has a unique gene expression pattern reflective of its developmental origin and biological function. Following stimulation with IFN-α, astrocytes and microglia also displayed a common core response that was characterized by the increased expression of genes required for pathogen detection and elimination. Compared with astrocytes, microglia had a more extensive and diverse response to IFN-α with significantly more genes with expression upregulated (282 vs. 141) and downregulated (81 vs. 3). Further validation was documented for selected IFN-I-regulated genes in a murine model of cerebral interferonopathy. In all, the findings highlight not only overlapping but importantly divergent responses to IFN-I by astrocytes versus microglia. This suggests specialized roles for these cells in host defense and in the development of cerebral interferonopathy.

Hypoxia reduces ER-to-Golgi protein trafficking and increases cell death by inhibiting the adaptive unfolded protein response in mouse beta cells

AIMS/HYPOTHESIS

Hypoxia may contribute to beta cell failure in type 2 diabetes and islet transplantation. The adaptive unfolded protein response (UPR) is required for endoplasmic reticulum (ER) homeostasis. Here we investigated whether or not hypoxia regulates the UPR in beta cells and the role the adaptive UPR plays during hypoxic stress.

METHODS

Mouse islets and MIN6 cells were exposed to various oxygen (O2) tensions. DNA-damage inducible transcript 3 (DDIT3), hypoxia-inducible transcription factor (HIF)1α and HSPA5 were knocked down using small interfering (si)RNA; Hspa5 was also overexpressed. db/db mice were used.

RESULTS

Hypoxia-response genes were upregulated in vivo in the islets of diabetic, but not prediabetic, db/db mice. In isolated mouse islets and MIN6 cells, O2 deprivation (1-5% vs 20%; 4-24 h) markedly reduced the expression of adaptive UPR genes, including Hspa5, Hsp90b1, Fkbp11 and spliced Xbp1. Coatomer protein complex genes (Copa, Cope, Copg [also known as Copg1], Copz1 and Copz2) and ER-to-Golgi protein trafficking were also reduced, whereas apoptotic genes (Ddit3, Atf3 and Trb3 [also known as Trib3]), c-Jun N-terminal kinase (JNK) phosphorylation and cell death were increased. Inhibition of JNK, but not HIF1α, restored adaptive UPR gene expression and ER-to-Golgi protein trafficking while protecting against apoptotic genes and cell death following hypoxia. DDIT3 knockdown delayed the loss of the adaptive UPR and partially protected against hypoxia-induced cell death. The latter response was prevented by HSPA5 knockdown. Finally, Hspa5 overexpression significantly protected against hypoxia-induced cell death.

CONCLUSIONS/INTERPRETATION

Hypoxia inhibits the adaptive UPR in beta cells via JNK and DDIT3 activation, but independently of HIF1α. Downregulation of the adaptive UPR contributes to reduced ER-to-Golgi protein trafficking and increased beta cell death during hypoxic stress.

The balance between adaptive and apoptotic unfolded protein responses regulates β-cell death under ER stress conditions through XBP1, CHOP and JNK

Endoplasmic reticulum (ER) stress and the subsequent unfolded protein response (UPR) have been implicated in β-cell death in type 1 and type 2 diabetes. However, the UPR is also a fundamental mechanism required for β-cell adaptation and survival. The mechanisms regulating the transition from adaptive to apoptotic UPR remain to be clarified. Here, we investigated the relationships between XBP1, CHOP and JNK in the transition from adaptive to apoptotic UPR and β-cell death in models of type 1 and type 2 diabetes. XBP1 inhibition potentiated cell death induced by pro-inflammatory cytokines or the saturated fatty acid palmitate in MIN6 β-cells. This response was prevented by CHOP inhibition. IRE1/XBP1 inhibition led to alterations in islets from diabetes-resistant ob/ob mice that resemble those found in diabetes, including increases in cell death and inflammation and antioxidant gene expression. Similarly, IRE1/XBP1 inhibition increased cell death in islets from NOD mice. On the other hand, JNK inhibition: 1) increased adaptive UPR and reduced cell death in islets from diabetic db/db mice, and 2) restored adaptive UPR while protecting against apoptotic UPR gene expression and β-cell death and dysfunction following cytokine exposure. These findings suggest that the balance between XBP1-mediated adaptive and CHOP-dependent apoptotic UPR is critically important for β-cell survival during ER stress. JNK activation regulates the transition from adaptive to apoptotic UPR, thus providing a mechanism for β-cell propensity to cell death rather than ER stress adaptation in type 1 and type 2 diabetes.