EphB receptors co-distribute with a nicotinic receptor subtype and regulate nicotinic downstream signaling in neurons

Abstract

Activation of nicotinic acetylcholine receptors (nAChRs) on neurons engages calcium-dependent signaling pathways regulating numerous events. Receptors containing alpha7 subunits (alpha7-nAChRs) are prominent in this because of their abundance and high relative calcium permeability. We show here that EphB2 receptors are co-localized with postsynaptic alpha7-nAChRs on chick ciliary ganglion neurons and that treatment of the cells with an ephrinB1 construct to activate the EphB receptors exerts physical restraints on both classes of receptors, diminishing their dispersal after spine retraction or lipid raft disruption. Moreover, the ephrinB1/EphB receptor complex specifically enhances the ability of alpha7-nAChRs to activate the transcription factor CREB, acting through a pathway including a receptor tyrosine kinase, a Src family member, PI3 kinase, and protein kinase A most distally. The enhancement does not appear to result from a change in the alpha7-nAChR current amplitude, suggesting a downstream target. The results demonstrate a role for ephrin/EphB action in nicotinic signaling.

Figure 1

Developmental timecourse for EphB2R appearance in chick ciliary ganglia. (A) Western blots of embryonic ciliary ganglia dissected at the indicated times, solubilized, electrophoresed, and probed with anti-EphB2R Ab. Size standards (left): 116 (βgalactosidase) and 200 (myosin) kD. (B) Quantification of the Western blots, normalizing for the amount of ganglionic protein (open squares) or for the number of neurons present at each age (filled circles). Peak levels of EphB2R occur at E12–E14 and subsequently decline.

Figure 2

Surface distribution of EphB2R on CG neurons. Embryonic ganglia of the indicated ages were dissociated and co-stained for EphB2Rs (left; red) and α7-nAChRs (middle; green) , and the images overlaid (right; yellow). (A) E10. (B) E14. (C) E18. (D) E14 with non-immune IgG substituted for the anti-EphB2R Ab as a negative control. Scale bar: 10 μm. EphB2Rs co-distribute with α7-nAChRs on the neuron surface, and are most prominent at E14.

Figure 3

Retention of α7-nAChRs by ephrinB1/EphB2Rs after spine collapse or lipid raft dispersal. Freshly dissociated E14 CG neurons were either treated with ephrinB1-Fc/Ab to activate EphBRs (Ephrin) or with Fc/Ab as a control (Con) and then treated with methyl-βcyclodextrin (MβCD) to extract cholesterol and disperse lipid rafts or with latrunculin A (Lat) to collapse F-actin and induce spine retraction. The cells were then labeled for α7-nAChRs (left; green) and either (A) EphB2Rs, (B) F-actin, or (C) Src (middle; red), and the images merged (right; yellow). Alternatively, the cells were immunostained with cholera toxin for the ganglioside GM1, confirming that MβCD treatment dispersed the lipid rafts (C, far right). Scale bar: 10 μm. EphrinB1-Fc/Ab protected both EphB2Rs and α7-nAChRs from dispersal after spine retraction or raft disruption; it did not protect Src from dispersal.

Figure 4

Effects of ephrinB1 treatment on nicotine-induced whole-cell currents. (A) Representative traces of nicotine-induced whole-cell responses in E14 neurons after treatment with Fc/Ab as a control (Fc, left) or with ephrinB1-Fc/Ab (EphrinB, right) and then incubated in culture medium (Con, top) or extracted with MβCD to extract cholesterol and disperse lipid rafts (MβCD, bottom). (B) Quantification of the peak currents, normalized for capacitance to correct for differences in cell size. (C) Quantification of the peak current components contributed by α7-nAChRs (left) and α3*-nAChRs (right), separated based on decay rates (fast vs. slow, respectively). Values represent the mean ± SEM of 12–18 cells. Asterisks, p < 0.05. The ephrinB1 treatment does not significantly enhance the nicotine-induced whole-current but does partially protect the rapidly decaying α7-nAChR component of the response against MβCD-induced loss.

Figure 5

EphrinB1 enhancement of α7-nAChR-induced long-lasting pCREB. E14 CG neurons were treated with Fc/Ab (Con) or ephrinB1-Fc/Ab (Ephrin) and the indicated compounds followed by 10 μM nicotine (Nic) for 5 min in the presence of 200 μM cadmium to block voltage-gated calcium channels, and then incubated an additional 20 min in the absence of nicotine before fixing, staining, and immunostaining for nuclear pCREB. Values represent the mean ± SEM for the proportion of labeled cells. Asterisk, p < 0.05; double asterisk, p < 0.01; multi-line comparisons: group members compared to “No Blkr”. EhprinB1 treatment increased the proportion of cells showing long-term pCREB, and the increase was entirely attributable to α7-nAChRs (blocked by αBgt) and dependent on a tyrosine receptor kinase (blocked by K252a), a Src family member (blocked by SU6656; also blocked by PP2 but not PP3, not shown), PI3 kinase (blocked by LY294002) and PKA (blocked by KT5720) but not on calcium release from internal stores (not blocked by thapsigargin).

Figure 6

Enzyme pathway mediating the ephrinB1 enhancement of nicotine-induced pCREB. E14 neurons were treated as in Fig. 5, including the addition of cadmium to block voltage-gated calcium channels, except that the 20 min rinse was omitted so that short-term pCREB could be assessed. Values represent the mean ± SEM for the proportion of labeled cells (n = 200–300 cells per condition from each of 3–6 separate experiments). Asterisk, p < 0.05; double asterisk, p < 0.01; multi-line comparisons: group members compared to “No Blkr”. (A) EphrinB1-Fc/Ab treatment increased the proportion of cells with short-term pCREB, acting specifically through α7-nAChRs (stimulated by nicotine; blocked by αBgt), and the effect depended on a receptor tyrosine kinase, a Src family member, PI3 kinase, and PKA. None of these kinases were required for short-term pCREB induced by nicotine via either α3*-nAChRs or α7-nAChRs in the absence of ephrinB1 treatment. CaMKII/IV was required for both α7- and α3*-nAChR-mediated pCREB (blocked by KN93 but not by control KN92). (B) Combining the PKA agonist 8Br-cAMP (8Br) with either SU6656 or LY294002 demonstrated that both the Src family member and PI3 kinase act upstream of PKA in mediating ephrinB1 enhancement of α7-nAChR-produced pCREB. The receptor or kinase being blocked is indicated in parentheses.