Neuronal MHC Class I Expression Is Regulated by Activity Driven Calcium Signaling

Abstract

MHC class I (MHC-I) molecules are important components of the immune system. Recently MHC-I have been reported to also play important roles in brain development and synaptic plasticity. In this study, we examine the molecular mechanism(s) underlying activity-dependent MHC-I expression using hippocampal neurons. Here we report that neuronal expression level of MHC-I is dynamically regulated during hippocampal development after birth in vivo. Kainic acid (KA) treatment significantly increases the expression of MHC-I in cultured hippocampal neurons in vitro, suggesting that MHC-I expression is regulated by neuronal activity. In addition, KA stimulation decreased the expression of pre- and post-synaptic proteins. This down-regulation is prevented by addition of an MHC-I antibody to KA treated neurons. Further studies demonstrate that calcium-dependent protein kinase C (PKC) is important in relaying KA simulation activation signals to up-regulated MHC-I expression. This signaling cascade relies on activation of the MAPK pathway, which leads to increased phosphorylation of CREB and NF-κB p65 while also enhancing the expression of IRF-1. Together, these results suggest that expression of MHC-I in hippocampal neurons is driven by Ca2+ regulated activation of the MAPK signaling transduction cascade.

Fig 1. KA induced expression of MHC-I by hippocampal neurons play a role in the destabilization of synapse.

Expression levels of MHC-I mRNA (A) and protein (B) were not varied during the development of cultured hippocampal neurons in vitro. (C) The expression of c-fos increased soon after adding 100μM KA to 8 div hippocampal neurons for 30 min. (D) Eight hours after adding 100μM KA to hippocampal neurons, the expression level of H-2K^b and H-2D^b mRNA significantly increased. (E) Expression of MHC-I mRNA was quantified as the ratio of band density to that of Tuj1. Data was presented as the ratio compared to control and was calculated from three independent experiments. (F) The expression level of MHC-I protein was increased significantly at 8 hours after KA treatment. (G) Expression of MHC-I protein was quantified as the ratio of band density to that of Tuj1. Data was presented as the ratio compared to control and was calculated from three independent experiments. (H) Forty-eight hours after 100μM KA stimulation for 30min, the expression of the pre-synaptic marker protein SYP (red) and post-synaptic marker protein PSD-95 (green) as well as their colocalization was decreased. The effect was partially rescued by using MHCI antibody ox18. Scale bar: 50 μm. (I) Western blot analysis showed increased expression of SYP and PSD-95 after KA stimulation, which was blocked by ox18. (J) Expression of SYP and PSD-95 was quantified as the ratio of band density to that of Tuj1. Data was presented as the ratio compared to control and was calculated from three independent experiments. *p<0.05, **p<0.01 vs control.

Fig 2. Enh A, ISRE and X-box are important elements for the endogenous and KA induced MHC-I promoter activity in neuro2a cell lines.

(A) Schematic illustration of constructs used for HLA-A heavy chain promoter-luciferase reporter assay. (B) At 6, 8 and 10 hours post 100 μM KA treatment for 30min increased the expression level of H-2K mRNA and protein (C). (D) Luciferase activity of different promoter-reporter plasmids in neuro2a cell lines. Mean luciferase activity was calculated from three independent experiments and shown with the SD. (E) Relative luciferase activity of different promoter-reporter plasmids in neuro2a cell lines by using 100μM KA for 30min.

Fig 3. Involvement of NF-κB, CREB and IRF-1 in neuronal MHC-I expression.

(A) Western blot analysis of time-dependent activation of NF-κB p65, CREB as well as the increased expression of IRF-1 by adding 100μM KA to 8 div hippocampal neurons for 30min. (B) The densitometric analyses of p-p65/p65, p-CREB/CREB, and IRF-1/Tuj1 from three separate experiments were taken, and the data was shown as ratio compared to control. All the data are indicated as mean±SD. (C) Western blot analysis of the expression level of of NF-κB p65, CREB as well as the expression of IRF-1 during the development stages of mouse hippocampus. (D) The densitometric analyses of p-p65/p65, p-CREB/CREB, and IRF-1/Tuj1 from four separate experiments were taken, and the data was shown as ratio compared to P4. All the data are indicated as mean±SD.

Fig 4. Calcium-dependent PKA and PKC activation participate in KA induced MHC-I expression.

(A) Two representative cells showed the increased concentration of intracellular calcium soon after adding 100μM KA. Scale bar: 50 μm. (B) Average normalized fluorescence intensity before and after KA (100μM) treatment. Each bar represents the mean ± SD of 50 cells. (C) Pretreatment of primary hippocampal neurons with PKA inhibitor (H89, 40μM) resulted in inhibition of KA-induced expression of MHCI. (D) Pretreatment of primary hippocampal neurons with PKC inhibitor (Staurosporine, 0.1μM) blocked KA-induced expression of MHC-I. (E) Densitometric analyses of MHC-I/Tuj1 at 8 hours after KA treatment from three separate experiments were taken, and the data was shown as ratio compared to control. All the data are indicated as mean±SD. (F) Pretreatment of primary hippocampal neurons with PKA inhibitor (H89, 40μM) resulted in inhibition of KA-induced expression of IRF-1, but had no effect on the activation of NF-κB p65 and CREB. (G) Pretreatment of primary hippocampal neurons with PKC inhibitor (Staurosporine, 0.1μM) blocked KA-induced expression of p-p65, p-CREB and IRF-1. n.s: no significance, *p<0.05, **p<0.01 vs control.

Fig 5. PKC mediated activation of MAPK and AKT pathways by KA stimulation.

(A) Western blot analysis of time-dependent activation of JAK1, STAT1, AKT, MAPK ERK1/2 and p38 after adding 100μM KA to 8 div hippocampal neurons for 30min. (B) Pretreatment of 8 div hippocampal neurons with PKC inhibitor (Staurosporine, 0.1μM) compromised the KA-induced activation of AKT, MAPK ERK1/2 and p38, but had no effect on the activation of JAK1 and STAT1. (C) Pretreatment of 8 div hippocampal neurons with PKA inhibitor (H89, 40μM) had no effect on the activation of JAK1, STAT1, AKT, MAPK ERK1/2 and p38 induced by KA. (D) Pretreatment of primary hippocampal neurons with MAPK-p38 inhibitor (Skepinone-L, 20μM) resulted in inhibition of KA-induced expression of MHC-I. (E) Pretreatment of primary hippocampal neurons with MAPK-ERK inhibitor (U0126, 10μM) blocked KA-induced expression of MHC-I. (F) Pretreatment of primary hippocampal neurons with MAPK-p38 inhibitor (Skepinone-L, 20μM) blocked KA-induced expression of p-p65, p-CREB and IRF-1. (G) Pretreatment of primary hippocampal neurons with MAPK-ERK inhibitor (U0126, 10μM) resulted in inhibition of KA-induced expression of IRF-1 and activation of NF-κB p65, but had no effect on the activation of CREB.

Fig 6. Model depicting the calcium-dependent pathway mediated MHC-I expression induced by KA.

Exposure of hippocampal neurons to KA leads to activation of calcium-dependent PKA and PKC, which results in the subsequent activation of the MAPK pathways and the downstream transcription factors NF-κB, CREB and IRF-1. Activation of these molecules finally leads to enhanced expression of MHC-I by binding to its promoter elements. JAK1/STAT1 and AKT pathways are also activated by KA stimulation.