It has been suggested that overactivated microglia surrounding neuritic plaques produce proinflammatory cytokines and chemokines that recruit more neurotoxic microglia and lead to further neuroinflammation [1, 3–6]. Therefore, attenuation of overactivated microglia at neuritic plaques may be a therapeutic strategy for AD. In the present study, we used a time-lapse recording approach to examine the effects of TGF-β1 on Aβ-induced microglial clustering. Our data suggests that TGF-β1 significantly inhibits microglial chemotaxis toward Aβ. Phosphorylated SMAD2 and down-regulation of CCL5 were at least partially involved in mediating the effects of TGF-β1. These results suggest that TGF-β1 may have therapeutic potential for AD by preventing excessive microglial clustering at neuritic plaques, since microglial overactivation at neuritic plaques actively contributes to neurodegeneration .
Although microglia have been shown to be recruited and activated by nearby plaques , the trajectory of microglial movement in vivo has not been examined because of the difficulty in monitoring microglial clustering at neuritic plaques. Our study provides, for the first time, real-time tracing data of microglial migration toward Aβ aggregates. BV-2 cells have been widely used to study microglial migration induced by alpha-synuclein and cannabinoid [34, 35]. The migratory behavior of BV-2 cells is similar to that of primary microglia . Furthermore, the morphology and behavior of migrating BV-2 cells in our study resembled that of microglia in brain slices . Although the clustering of BV-2 microglia around Aβ aggregates in vitro may not accurately mimic microglia around neuritic plaques in AD, our approach provides a tractable system for the study of molecular mechanisms of TGF-β1 attenuation of microglia chemotaxis toward neuritic plaques.
TGF-β1 overexpression has been shown to prominently reduce plaque formation and Aβ accumulation in hAPP/TGF-β1 double transgenic mice . It is suggested that TGF-β1 enhances the uptake of Aβ by microglia. However, Town et al. showed that the blockade of TGF-β1-SMAD2/3 signaling in peripheral macrophages reduces Aβ deposits in hAPP mice, and they demonstrated that such blockade enhances infiltration of macrophages into the central nervous system . Furthermore, it has been shown that TGF-β1 knockout mice develop spontaneous neurodegeneration [37, 38]. Blockade of neuronal TGF-β1 signaling has been shown to reduce neuronal survival rate and enhance amyloid deposition in AD mice . These studies suggest that the function of the TGF-β1-SMAD2/3 pathway is cell-type specific. Therefore, the neuroprotective effects of TGF-β1 in AD may directly promote neuronal survival as well as enhance Aβ uptake and prevent the clustering of overactivated microglia at senile plaques.
Phosphorylation of SMAD2/3, the downstream effector of ALK4, 5, and 7, has been shown to be responsible for the transmission of TGF-β signaling in mammalian cells . Several studies have suggested that SMAD2/3 signaling mediates the down-regulation of chemokines during inflammatory responses [26, 27]. SB431542, a specific ALK5 inhibitor, is commonly used to inhibit SMAD2/3 phosphorylation [39–41], and has been used to attenuate the neuronal protective effects of TGF-β1 in Aβ-injected rats . A key finding in this study is the inhibition of SB431542 on TGF-β1-suppressed microglia clustering and its concurrent effects on inhibiting SMAD2 phosphorylation and restoring CCL5 expression. Based on the specificity of SB431542 for the TGF-β1-SMAD2/3 pathway [39–41], a logical conclusion is that TGF-β1 reduces microglial clustering and CCL5 expression by activating the ALK5-SMAD2 pathway. Due to the lack of a SMAD2-specific inhibitor, it is impossible to exclude SMAD3 or other effectors as responsible for the inhibitory effects on microglial chemotaxis toward Aβ. CCL5 can serve as a microglia activator and a chemokine in dendritic cells . Moreover, the deletion of the CCL5 gene significantly reduces glial activation induced by different stimuli, suggesting that CCL5 can activate microglia . Thus, down-regulation of CCL5 by TGF-β1 at neuritic plaques may prevent sequential microglial clustering and microglia activation.
Although numerous studies suggest that microglia surrounding senile plaques express pro-inflammatory cytokines, Colton et al. has shown that genes involved in tissue repair are upregulated in microglia of AD patients and AD mice . Microglia can be beneficial as well as deleterious due to the various phenotypes of microglia under different microenvironments in CNS . For example, the ligation of CD40-CD40L induces TNF-α production. This proinflammatory phenotype is attenuated by TGF-β1 and IL-10 . Different types of stimuli will lead to distinct microglial responses; therefore, the broad spectrum of microglial activation should be evaluated during the development of therapeutic strategies of AD.