Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 10 (2), e0117246
eCollection

Identifying a Kinase Network Regulating FGF14:Nav1.6 Complex Assembly Using Split-Luciferase Complementation

Affiliations

Identifying a Kinase Network Regulating FGF14:Nav1.6 Complex Assembly Using Split-Luciferase Complementation

Wei-Chun Hsu et al. PLoS One.

Abstract

Kinases play fundamental roles in the brain. Through complex signaling pathways, kinases regulate the strength of protein:protein interactions (PPI) influencing cell cycle, signal transduction, and electrical activity of neurons. Changes induced by kinases on neuronal excitability, synaptic plasticity and brain connectivity are linked to complex brain disorders, but the molecular mechanisms underlying these cellular events remain for the most part elusive. To further our understanding of brain disease, new methods for rapidly surveying kinase pathways in the cellular context are needed. The bioluminescence-based luciferase complementation assay (LCA) is a powerful, versatile toolkit for the exploration of PPI. LCA relies on the complementation of two firefly luciferase protein fragments that are functionally reconstituted into the full luciferase enzyme by two interacting binding partners. Here, we applied LCA in live cells to assay 12 kinase pathways as regulators of the PPI complex formed by the voltage-gated sodium channel, Nav1.6, a transmembrane ion channel that elicits the action potential in neurons and mediates synaptic transmission, and its multivalent accessory protein, the fibroblast growth factor 14 (FGF14). Through extensive dose-dependent validations of structurally-diverse kinase inhibitors and hierarchical clustering, we identified the PI3K/Akt pathway, the cell-cycle regulator Wee1 kinase, and protein kinase C (PKC) as prospective regulatory nodes of neuronal excitability through modulation of the FGF14:Nav1.6 complex. Ingenuity Pathway Analysis shows convergence of these pathways on glycogen synthase kinase 3 (GSK3) and functional assays demonstrate that inhibition of GSK3 impairs excitability of hippocampal neurons. This combined approach provides a versatile toolkit for rapidly surveying PPI signaling, allowing the discovery of new modular pathways centered on GSK3 that might be the basis for functional alterations between the normal and diseased brain.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Fig 1
Fig 1. Using LCA to measure real-time interaction between FGF14 and Nav1.6 C-tail in live cells.
(A) Using the bioluminescence-based luciferase complementation assay (LCA) to measure protein:protein interactions. Two proteins of interest (FGF14, and CD4-Nav1.6-Ctail in this example) are fused to Cluc and Nluc fragments of Photinus luciferase. Upon interaction of the protein components, Nluc and Cluc fragments reconstitute into functional luciferase enzyme, which produces luminescence in the presence of luciferin substrate. The intensity of luminescence is linear to the strength of the protein:protein interaction identified. (B) Schematic of constructs used for LCA experiments, to scale. Top: Cluc (AAs 398–550), linker (GGGSSGGGQISYASRG), FGF14-b (AAs 1–252). Bottom: CD4-ΔCtail (AAs 1–395), Nav1.6-Ctail (AAs 1763–1976), linker (QISYASRGGGSSGGG), Nluc (AAs 2–416). (C) Schematic of protein inhibition by kinase inhibitors. ATP-competitive kinase inhibitors block the ATP-binding site of the target kinase, preventing the transfer of phosphate groups to the substrate. Non-ATP competitive kinase inhibitors work through other mechanisms, such as changing the conformation of the ATP-binding site to prevent docking of ATP.
Fig 2
Fig 2. Split luciferase assay-based screen identifies new phospho-regulatory pathways upstream of the FGF14:Nav1.6 channel complex.
(A) Schematic diagram of the large-scale kinase inhibitor libraries-based screen detected by bioluminescence. Counter and secondary screenings used include validation of cell viability, measurement of inhibitor activity against intact Photinus luciferase, and siRNA validation of target kinases. (B) 3D plot of screening results by split-luciferase signal (LCA), Photinus luciferase assay, and cell viability. Results that passed screening are shown in blue; results that passed screening but were rejected based on counter and secondary screens are shown in green. (C) Example time series data for increased (triciribine), unchanged (ML-7 Hydrochloride) or decreased (Wee1 inhibitor) luminescence relative to control (Cluc-FGF14 / CD4-Nav1.6-Nluc) shown, at a final concentration of 50 μM. (D) Results of the screen are represented as peak areas (±1 time point of peak) normalized to % luminescence values of DMSO treated controls (FGF14:Nav1.6 complex) at a final concentration of 50 μM. Data shown is mean ± 95% CI. Significant increase (red) or decrease (green) of luminescence, relative to the control signal (p < 0.05) shown. (E) Western blot analysis of HEK293 and transfected HEK293 cells, visualized with anti-calnexin (Calnexin) and anti-luciferase (CD4-Nav1.6-Nluc and Cluc-FGF14) antibodies. Transfected HEK293 cells were treated with 0.5% DMSO or inhibitors dissolved in 0.5% DMSO at an empirically determined concentration to minimize effects on cell viability (calnexin). Use concentrations are as follows: Triciribine (25 uM), Cdk4 inhibitor (15 uM), Cdk1 inhibitor (25 uM), BAY 11–7085 (25 uM), Wee1 inhibitor (15 uM).
Fig 3
Fig 3. Dose-response studies of identified full inverse agonists of FGF14:Nav1.6 regulatory pathways.
Fitting was performed with nonlinear regression using Graphpad Prism 6 (Statistics). Indicated pIC50 and pEC50 were derived from best fit nonlinear regression after a maximum of 1000 iterations. Full inverse agonists were defined as compounds that act as inverse agonists (inhibit FGF14:Nav1.6 complementation with increasing dose) and have an efficacy value of greater than 1. Plot, X-axis: log10([Inhibitor]), Y-axis: Percent of intensity, normalized to peak observed raw intensity for each compound.
Fig 4
Fig 4. Dose-response studies of identified partial inverse agonists of FGF14:Nav1.6 regulatory pathways.
Fitting, pIC50/EC50 calculation, axes as in Fig. 3. Partial inverse agonists were defined as compounds that act as inverse agonists (inhibit FGF14:Nav1.6 complementation with increasing dose) and have an efficacy value of less than 1.
Fig 5
Fig 5. Dose-response studies of identified partial agonists of FGF14:Nav1.6 regulatory pathways.
Fitting, pIC50/EC50 calculation, axes as in Fig. 3. Partial agonists were defined as compounds that act as agonists (promote FGF14:Nav1.6 complementation with increasing dose).
Fig 6
Fig 6. Hierarchical clustering of inhibitors for FGF14:Nav1.6 regulatory pathways.
A) Heatmap and hierarchical clustering for individual inhibitors. Red, increased intensity relative to DMSO control. Green, decreased intensity. Left, Hierarchical clustering, based on differences in normalized interaction strength between Nav1.6 and FGF14 for each inhibitor, with equal weighing of all categories. B) Heatmap and hierarchical clustering for primary kinase targets of each inhibitor, derived from geometric averaging of all inhibitors of each primary kinase.
Fig 7
Fig 7. Principal component analysis of inhibitor/kinase dose response profiles.
Computed principal components are visualized with a Scree plot, with the top 3 principal components (PC #1, PC #2, PC #3) denoted by orthogonal axes and line segments denoting individual inhibitors or kinases. The first principal component contributes the most (54.5% for inhibitors and 70.5% for kinases) and is denoted by the horizontal axis. Spheres denote the composite response of a single dose category (1, 5, 25, 50 uM). A) Response profile of inhibitors. B) Response profile of kinases.
Fig 8
Fig 8. Bioinformatics analysis of a GSK-3 centered kinase regulatory network.
For protein interaction networks, Ingenuity Pathway Analysis (IPA) was applied to the list of identified major kinase targets for all of the inhibitors tested, in addition to GSK-3, and an unbiased network was generated using the “Connect” algorithm and subsequently submitted to the pathway analysis engine. Both direct (solid) and indirect (dashed) interactions, as classified by Ingenuity, are shown; the sub-network with edge length of 1 to GSK-3 is additionally highlighted (purple). A) Interaction network, showing direct interactions only. B) Interaction network, showing both direct (solid) and indirect (dashed) interactions. C) Western blot analysis of co-immunoprecipitation (IP:myc) and cell lysate from HEK293-Nav1.6 cells transfected with FGF14–6×myc. GSK3 inhibitor XIII (25 μM) treatment reduces the co-immunoprecipitated fraction of Nav1.6 without affecting FGF14–6×myc. D) and E) representative traces showing effect of 12 hour treatment with ether DMSO 0.25% (D) or CHIR99021 5μM (E) on neuronal excitability in cultured hippocampal neurons DIV 12–15. Single action potentials were evoked by brief (2.5 ms) depolarizing current injections. Grey squares indicate action potential threshold. F) CHIR99021 increases action potential threshold in cultured hippocampal neurons. Results represent mean ± SEM. n = 7 (DMSO), n = 6 (CHIR990221). *p<0.05, Student t-test. G) and H) CHIR99021 has no effect on input resistance (G) and action potential half width (H) threshold in cultured hippocampal neurons. Results represent mean ± SEM. n = 7 (DMSO), n = 6 (CHIR990221), NS—non significant, p>0.05, Student t-test.

Similar articles

See all similar articles

Cited by 19 articles

See all "Cited by" articles

References

    1. Kramer T, Lo Monte F, Goring S, Okala Amombo GM, Schmidt B (2012) Small molecule kinase inhibitors for LRRK2 and their application to Parkinson's disease models. ACS Chem Neurosci 3: 151–160. 10.1021/cn200117j - DOI - PMC - PubMed
    1. Smith B, Medda F, Gokhale V, Dunckley T, Hulme C (2012) Recent advances in the design, synthesis, and biological evaluation of selective DYRK1A inhibitors: a new avenue for a disease modifying treatment of Alzheimer's? ACS Chem Neurosci 3: 857–872. 10.1021/cn300094k - DOI - PMC - PubMed
    1. Malhi GS, Tanious M, Das P, Coulston CM, Berk M (2013) Potential mechanisms of action of lithium in bipolar disorder. Current understanding. CNS Drugs 27: 135–153. 10.1007/s40263-013-0039-0 - DOI - PubMed
    1. Chiu CT, Chuang DM (2010) Molecular actions and therapeutic potential of lithium in preclinical and clinical studies of CNS disorders. Pharmacol Ther 128: 281–304. 10.1016/j.pharmthera.2010.07.006 - DOI - PMC - PubMed
    1. Jope RS, Roh MS (2006) Glycogen synthase kinase-3 (GSK3) in psychiatric diseases and therapeutic interventions. Curr Drug Targets 7: 1421–1434. - PMC - PubMed

Publication types

MeSH terms

Feedback