Act1, Scene Brain: Astrocytes Play a Lead Role

Interleukin-17 (IL-17) is crucial for the progression of experimental autoimmune encephalomyelitis. In this issue of Immunity, Kang, et. al (Kang et al., 2010) report that neuroectoderm-derived astrocytes are the critical cellular element that respond to IL-17.

Act1 is a crucial adapter protein in the interleukin-17 receptor (IL-17R) signaling complex. Upon IL-17 binding, the heteromeric transmembrane receptor (IL-17RA and IL-17RC) recruits and interacts with Act1 by its SERIF domains (Chang et al., 2006). A variety of cells express the IL-17R signaling complex, facilitating inflammatory responses to locally produced IL-17 in most tissues (Gaffen, 2009). IL-17-responsive cells are capable of expanding local tissue inflammation because IL-17 induces Act1-dependent upregulation and mRNA stabilization of pro-inflammatory cytokines and CXC chemokines (Hartupee et al., 2007).

Several inflammatory autoimmune disease models including experimental autoimmune encephalomyelitis (EAE), a mouse model of multiple sclerosis (MS), have been shown to be mediated to a substantial extent by Th17 cells, although both Th1 and Th17 cells are capable of inducing EAE individually, but likely by different mechanisms and with different efficiency (Qian et al., 2007). As previously reported (Qian et al., 2007), mice deficient in Act1 exhibited delayed onset and less severe myelin oligodendrocyte glycoprotein (MOG) peptide-induced EAE. Given that there are multiple stages and cell types involved in EAE pathology, it was unclear at which stage of disease pathogenesis IL-17, signaling via Act1 was necessary, and it is this question which was approached by Kang, et al. (Kang et al., 2010) in this issue of Immunity..

EAE is a multi-step inflammatory process initiated by Th1 and Th17 cells. During actively-induced EAE, myelin-specific Th1 or Th17 cells are activated and expand in the peripheral lymphoid tissues in response to myelin peptide and complete Freund’s adjuvant (CFA) immunization. These activated T cells extravasate across the blood brain barrier (BBB) secondary to interaction of very late activation antigen 4 (VLA-4) expressed on activated T cells with vascular cell adhesion molecule (VCAM) expressed on cerebrovascular endothelial cells (Yednock et al., 1992). In the central nervous system (CNS), infiltrating myelin-specific T cells must be re-activated by brain-resident antigen presenting cells (including microglia, macrophages and myeloid dendritic cells) presenting myelin antigen (Bailey et al., 2007; Tompkins et al., 2002). Release of Th1 or Th17 cell cytokines induced by this interaction results in local inflammation and demyelination of white matter tracts, reducing the ability of axons to conduct electrical signals, and eventually neuronal transection. Astrocytes are involved in nearly all processes within the brain, but their contribution to EAE pathogenesis is not certain. In vitro studies suggest that they produce cytokines that affect inflammatory cytokine production by T cells as well as chemokines that may attract additional peripheral inflammatory leukocytes.

EAE can also be induced in mice by adoptive transfer of myelin-specific T cells from primed mice to naïve recipients. Kang et al. utilized adoptive transfer EAE to determine the role of Act1 in both Th1 and Th17cell-mediated EAE and to distinguish at which stage of disease this crucial signaling adapter protein operates. It is shown that transfer of activated myelin-specific Th1 cells into either naïve wild-type or Act1-deficient mice resulted in a normal EAE disease course. However, transfer of activated myelin-specific Th17 cells induced disease in wild-type, but not Act1-deficient mice, due in part to impaired CNS infiltration of Th17 cells in Act1-deficient mice. Thus Act1 only regulates Th17 cell-mediated EAE.

Data presented in this paper also suggest that Act1-deficient Th17 cells are not defective, but perhaps even more pathogenic in induction of disease when adoptively transferred into wild-type mice. Further supporting the conclusion that there were no defects in T cell activation or differentiation in Act1-deficient mice, it is shown that the kinetics of initial CNS infiltration of Act1-deficient T cells was normal and that these T cells displayed normal expression of proliferation and survival factors in wild-type recipients. Together these data clearly demonstrate that Act1 is not necessary for normal activation, peripheral expansion, infiltration or effector function of the initial infiltrating set of T cells. Not surprisingly, the authors demonstrated that mice in which Act 1 deficiency was restricted to endothelial cells by breeding floxed Act1 mice with TIE2e-Cre mice were normally susceptible to EAE ruling out the possibility of endothelial cells as a major cellular compartment dependent on Act1.

Microglia are F4/80⁺CD11b^lo CNS-resident cells of hematopoietic origin, which may play a role in both local presentation of myelin epitopes as well as in carrying out the effector stages of demyelination. In addition, large numbers of CD11b^hi peripherally-derived macrophages and myeloid DCs (CD11b^hiCD11c⁺) are recruited into the CNS during EAE pathogenesis and play important roles in antigen presentation and epitope spreading, recruitment of inflammatory cells, and effector demyelination (Bailey et al., 2007). Thus, CD11b⁺ cells were a possible cellular compartment that may be dependent on Act1. However, it is also shown that mice develop EAE in a normal fashion when Act1 deficiency was targeted to CD11b⁺ cells by breeding floxed Act1 mice with CD11b-Cre mice.

Neurons, oligodendrocytes, and astrocytes are considered brain resident cells because they originate from a common progenitor in the neuroectoderm during development (Gilyarov, 2008). During early development, neuroectodermal cells express nestin (NES), an intermediate filament. Kang, et al. thus used NES-Cre mice mated with floxed Act1 mice to target Act1 deficiency to neuroectoderm-derived brain resident cells. Similar to Act1 null mice, these brain resident cell-specific Act1-deficient mice were susceptible to EAE transferred with Th1 cells, but displayed a markedly delayed EAE onset and severity upon active MOG peptide immunization and following transfer of Th17 cells. These mice also had a reduced IL-17-induced gene profile in the lumbar spinal cord, where lesions are common during EAE. These data clearly identify brain resident cells as the key IL-17-induced, Act1-dependent cellular responders.

Based on the previously discussed data, the authors hypothesized that astrocytes may play a key role in EAE pathogenesis by IL-17 induced, Act1-dependent upregulation of chemokines necessary for the recruitment of the second wave of peripheral leukocyte infiltration during EAE pathogenesis (Figure 1). In support of their hypothesis, they stimulated primary astrocytes from NesCreAct1 floxed mice and control mice with the cytokines IL-17, IFN-γ, and TNF-α separately, or in combination. They describe chemokine profiles specific to stimulation with either IL-17 and TNF-α (CXCL1, CXCL2, CCL20) or IFN-γ and TNF-α (CXCL9, CXCL10, CXCL11), highlighting cytokine-specific responses in CNS-resident astrocytes. However, only IL-17+TNFα synergistic responses were decreased by the nestin-specific deletion of Act1, supporting an astrocyte-specific Act1-dependent response to IL-17.

Figure 1. IL-17-stimulated, Act1-dependent Stimulation of CNS Astrocytes is Required for Progression of EAE Via Chemokine-Mediated Recruitment of Peripheral Inflammatory Cells.

Kang et al suggest a necessary role for astrocytes in EAE pathology. After activated myelin-specific Th17 cells infiltrate the CNS and are re-stimulated to produce IL-17, astrocytes respond via expression of IL-17R are stimulated to produce leukocyte attracting chemokines in an Act1-dependent manner. Astrocyte-derived chemokines then recruit a second wave of peripheral inflammatory cells, which mediate EAE progression via Th17cell -mediated bystander demyelination.

In summary, Kang, et al. (Kang et al., 2010) have identified astrocytes as a key CNS-resident cell responsive to IL-17-driven signaling (Figure 1). Upon stimulation with IL-17, astrocytes, at least in vitro, appear to respond to IL-17 by recruiting Act1 to the IL-17R followed by recruitment of the TAK1 (a kinase) and TRAF6 (an E3 ubiquitin ligase) which mediate NF-κB activation. The subsequent upregulation of a variety of chemokine genes required to direct the second wave of recruitment of peripheral inflammatory cells characterizes the inflammatory demyelinating process in EAE and presumably MS. Despite long recognition of astrogliosis as a key feature of neuroinflammatory processes, the significance of this study is that it is the first in vivo evidence for a pathologic role of astrocytes in EAE pathogenesis. This study should stimulate future experiments focused on exploring the in vivo response of astrocytes to IL-17 and other inflammatory cytokines involved in neuroinflammatory diseases and the role of these glial cells in recruitment of peripheral inflammatory cells. Astrocytes may receive signals from both Th1 (IFN-γ) and Th17 (IL-17, etc.) cells resulting in upregulation of distinct patterns of cytokines and/or chemokines ultimately contributing to distinct pathological patterns of inflammation. It will also be interesting to explore the response of tissue resident cells to IL-17 in other Th17 cell-dependent disease models such as arthritis and asthma. However, it is important to keep in mind that targeting Act1 in astrocytes may have limited therapeutic implications in MS because ablation of Act1 has no significant influence on Th1 cell-mediated neuroinflammation and that Act1 is a negative regulator of B cells through CD40L and BAFF signaling (Qian et al., 2004).