Extended JAK activation and delayed STAT1 dephosphorylation contribute to the distinct signaling profile of CNS neurons exposed to interferon-gamma

Abstract

Although interferon-gamma (IFN-γ) plays a critical role in the noncytolytic elimination of many neurotropic viral infections, the signaling response to this cytokine has not been extensively characterized in primary CNS neurons. We previously demonstrated that the IFN-γ response at the signaling and gene expression levels is temporally extended in primary mouse hippocampal neurons, as compared to the transient response of primary mouse embryonic fibroblasts (MEF). We hypothesize that the protracted kinetics of STAT1 phosphorylation in IFN-γ-treated neurons are due to extended receptor activation and/or delayed STAT1 dephosphorylation in the nucleus. Here, we show that in response to IFN-γ, the Janus kinases (JAK1/JAK2) associated with the neuronal IFN-γ receptor complex remain active for an extended period as compared to MEF. Experimental inactivation of JAK1/JAK2 in neurons after IFN-γ treatment did not reverse the extended STAT1 phosphorylation phenotype. These results suggest that the extended kinetics of neuronal IFN-γ signaling are a product of distinct negative feedback mechanisms operating at both the receptor and within the nucleus.

Figure 1. IFN-γ exposure and sample collection

Schematic for the experimental setup used to establish IFN-γ-stimulated JAK1 and JAK2 activation kinetics in neurons and MEF. CM: conditioned medium.

Figure 2. JAK1 phosphorylation is extended in IFN-γ-treated neurons as compared to MEF

Neurons and MEF were treated as described (Figure 1), and equal volumes of cell lysates were examined by immunoblotting for a) phospho-JAK1 (pY1022/1023); b) total JAK1; and GAPDH signals. The double bands in the phospho-JAK1 blot are due to hyperphosphorylation of JAK1 (Irie-Sasaki et al., 2001); both bands were included in the densitometry. c) Blot signals were quantified using densitometry, and phospho-JAK1 signals were normalized to total JAK1 expression and GAPDH for loading. Values are expressed as a percent of control. Data shown are from a representative experiment. Ø: cells not exposed to IFN-γ.

Figure 3. JAK2 phosphorylation in IFN-γ-treated neurons is significantly extended as compared to MEF

Neurons and MEF were treated as described (Figure 1), and equal volumes of cell lysates were examined by immunoblotting for a) phospho-JAK2 (pY1007/1008); b) total JAK2; and GAPDH signals. c) Blot signals were quantified using densitometry, and phospho-JAK2 signals were normalized to total JAK2 expression and to GAPDH for loading. Values are expressed as a percent of control. Data shown are from a representative experiment. Ø: cells not exposed to IFN-γ.

Figure 4. JAK inhibitor I effectively prevents IFN-γ-stimulated STAT1 activation in neurons

a) Neurons and MEF were treated with IFN-γ (100U/ml) or without IFN-γ (Ø) in the presence or absence of JAK inhibitor I (10μM). Cell lysates were collected at the indicated timepoints post-treatment. b) Equal volumes of lysates were examined by immunoblotting for phospho-STAT1 (pY701) and GAPDH signals.

Figure 5. STAT1 dephosphorylation is delayed in IFN-γ-treated neurons as compared to MEF

a) Neurons and MEF were treated with IFN-γ (100U/ml) or without IFN-γ (Ø) for 30 min, washed, and then incubated with CM supplemented with or without JAK inhibitor I (10μM) for the indicated timepoints. b) Equal volumes of whole cell lysates were examined by immunoblotting for phospho-STAT1 (pY701) and GAPDH signals. c) Blot signals were quantified using densitometry, and phospho-STAT1 signals were normalized for loading. Values are expressed as a percent of maximum phospho-STAT1 signal. Data shown are from a representative experiment.