Roles of IKKβ and TLR3 in Inflammation and Axon Regeneration after Spinal Cord Injury

Abstract

Traumatic spinal cord injury (SCI) is accompanied by widespread neuronal cell death, consecutive inflammation processes and glial scar formation. Damage in an injured spinal cord is modulated by the progressive functions of microglia, neutrophils, macrophages and astrocytes. Many studies have reported that neutrophils and macrophages have an effect on the damage and recovery of the injured spinal cord. However, it is still controversial whether these cells play a beneficial or harmful role in the spinal cord injury. In the first chapter, to address the in vivo role of these immune cells in SCI, I utilized myeloid cell-specific IκB kinase (IKK)-β conditional knockout (ikkβΔmye) mice, in which ikkβ gene is specifically deleted from myeloid cells including neutrophils and macrophages. IkkβΔmye mice showed significantly reduced inflammatory responses of neutrophils and macrophages in SCI compared to ikkβ+/+ controls: SCI-induced expression of proinflammatory mediators and a potent neutrophil-attractant, C-X-C motif (CXC) ligand 1, were reduced in ikkβΔmye mice. In addition, neuronal loss, axonal damage, and behavioral deficits in motor activity were attenuated. These results demonstrate that IKK-β-dependent neutrophil and macrophage activation potentiates neuroinflammation and increases neuronal damage after SCI. In the second chapter, I characterized the in vivo role of TLR3 in SCI using TLR3 knockout mice to test whether TLR3 signaling is involved in the inflammatory responses observed in the injured spinal cord. TLR3 plays important roles in innate immune cell activation in response to tissue damage. However, it is not clear whether TLR3 functions beneficial or detrimental in SCI. At 5 days post-injury (dpi), axon demyelination and vestibulospinal tract (VST) axon did not differ in the wild-type and TLR3 knockout mice. However, in TLR3 knockout mice, behavioral recovery was improved at 14 dpi, and the axon demyelination and lesion volume were attenuated at 35 dpi compared to the wild-type mice. In addition, more axons in the corticospinal tract (CST) and VST were detected in the TLR3 knockout mice at 24 dpi or 35 dpi, respectively. Upon SCI, TLR3 expression increased mainly in astrocytes and oligodendrocytes. Furthermore, astrocytes were vigorously activated in SCI-induced wild-type mice, which were relatively attenuated in TLR3 knockout mice. In addition, there was less chondroitin sulfate proteoglycans (CSPGs) deposited in the TLR3 knockout mice compared to the wild-type mice. Interestingly, expression and activation of matrix metalloproteinase-2 (MMP-2), a protease involved in CSPG degradation, increased more in the TLR3 knockout mice compared to the wild-type mice after SCI. MMP-2 expression was detected mostly in astrocytes in the injured spinal cord. In the in vitro assays, recombinant CSPG was degraded by MMP-2. The expression level of MMP-2 was higher in astrocytes from the TLR3 knockout mice than from the wild-type mice. Furthermore, TLR3 stimulation inhibited TGF-β-induced MMP-2 expression in wild-type astrocytes. These data suggest that the TLR3 signaling inhibits MMP-2 expression in astrocytes in the spinal cord during SCI. Taken together, my data show that TLR3 has a detrimental role in SCI at least by partly inhibiting astrocytic MMP-2 expression, which results in the accumulation of CSPG and the impairment of axon regeneration and functional recovery after SCI. In this study, I conclusively found that IKK-dependent macrophage and neutrophil activation exacerbates SCI and that TLR3-mediated inhibition of MMP-2 expression in astrocytes impedes recovery from SCI.