Metabolomic and Signaling Programs Induced by Immobilized versus Soluble IFN γ in Neural Stem Cells

Abstract

Neural stem cells (NSCs) provide a strategy to replace damaged neurons following traumatic central nervous system injuries. A major hurdle to translation of this therapy is that direct application of NSCs to CNS injury does not support sufficient neurogenesis due to lack of proper cues. To provide prolonged spatial cues to NSCs IFN-γ was immobilized to biomimetic hydrogel substrate to supply physical and biochemical signals to instruct the encapsulated NSCs to be neurogenic. However, the immobilization of factors, including IFN-γ, versus soluble delivery of the same factor, has been incompletely characterized especially with respect to activation of signaling and metabolism in cells over longer time points. In this study, protein and metabolite changes in NSCs induced by immobilized versus soluble IFN-γ at 7 days were evaluated. Soluble IFN-γ, refreshed daily over 7 days, elicited stronger responses in NSCs compared to immobilized IFN-γ, indicating that immobilization may not sustain signaling or has altered ligand/receptor interaction and integrity. However, both IFN-γ delivery types supported increased βIII tubulin expression in parallel with canonical and noncanonical receptor-signaling compared to no IFN-γ. Global metabolomics and pathway analysis revealed that soluble and immobilized IFN-γ altered metabolic pathway activities including energy, lipid, and amino acid synthesis, with soluble IFN-γ having the greatest metabolic impact overall. Finally, soluble and immobilized IFN-γ support mitochondrial voltage-dependent anion channel (VDAC) expression that correlates to differentiated NSCs. This work utilizes new methods to evaluate cell responses to protein delivery and provides insight into mode of action that can be harnessed to improve regenerative medicine-based strategies.

Figure 1.

Immobilized and soluble IFN-γ support neurogenesis through JAK1/JAK2 signaling at 7 days. A) representative images of NSCs grown on coverslips coated in hydrogel with 300 ng/ml immobilized or soluble IFN-γ with(+) or without 100 μM of Ruxolitinib, scale bar = 20 μm. B) quantification of βIII tubulin expression normalized to nuclei staining, n=4, s (areas analyzed per biological replicate)=4) C) β III tubulin mRNA expression expressed as fold change from no IFNγ group, n=3 one way ANOVA with Tukey’s post hoc test. *** p<0.001, ** p < 0.005 * p<0.05

Figure 2.

Immobilized and Soluble IFN-γ activate STAT1 expression but do not maintain STAT1 phosphorylation equivalently A) mRNA expression of STAT1 at 7 days, n=3, one way ANOVA with Tukey’s post hoc **** p value <0.0001. B) Short term expression of STAT1p in NSCs plated on glass coverslips treated with 300 ng/ml soluble IFN-γ. C) Protein expression of STAT1p and total STAT1 in NSCS seeded on hydrogels treated with and without JAK1/JAK2 inhibitor, Ruxolitinib, at 7 days of treatment and 24 hours after last dose of soluble IFN γ 300 ng/ml

Figure 3 :

Immobilized and soluble IFN-γ dysregulate non-canonical signaling in NSCs grown on hydrogels for 7 days. A) multiplex data of ratio of phosphorylated to total protein in each treatment group, n=4 for all groups except day 0 NSCs n=2. One way ANOVA tukey’s post hoc *p<0.05, **p<0.01, ***<0.001. B) visualization of measured phospho-protein network. Small boxes reveal the ratio of phosphorylated protein for immobilized IFN-γ, soluble IFN-γ and no IFN-γ (left to right).C) Clustering of pathway analysis of STAT1, AKT, JNK and JAK1/2 signaling show high impact on purine binding activity, metabolism and response to reactive species

Figure 4.

Global metabolomics visualization and pathway analysis reveal similarities and differences between soluble and immobilized IFN-γ treated hydrogel embedded NSCs A) PCA plot analysis of treatment groups B) circos plot visualization of pathway derived genes for each treatment group, size of outer circle segment correlates to the number of hits for that pathway, red lines connect genes that are present in each group, green lines represent enrichment. See Table. S2 for full key of pathways C) Heat map of pathway impact per treatment groups, green represents high impact, blue low impact and white no impact E) Heat map of identified metabolites (Fig. S2), red indicates highest expression and blue indicates lowest expression F) example of two identified metabolites significantly dysregulated in IFN γ compared to no IFN γ, n=4, one way ANOVA with Tukey’s post hoc test ** p-value <0.005, *** p-value <0.001.

Figure 5.

JAK1/JAK2 signaling and metabolic changes indicate dysregulated energy metabolism supported by altered VDAC expression A) VDAC analysis in NSCS plated on 2D hydrogels N=4 s=4 B) co-staining of β III tubulin and VDAC, one way ANOVA Tukey’s post hoc C) spearman correlation of relationship between β III tubulin and VDAC, N=20. Scale bar = 20 μm.