Chronic nicotine exposure induces up-regulation of α7nAChRs, β2nAChRs, and PDLIM5
To clarify the role of PDLIM5 in nAChRs’ upregulation with chronic nicotine treatment, 1µM nicotine, a concentration in the range of peak serum levels found in smokers[32], was chosen in our study. Primary cultured hippocampal neurons were treated with control medium (Con) or with medium containing 1µM nicotine (Nic) for 1, 3, and 7 days respectively, and then quantitated with immunoblotting at DIV10. Nicotine treatment did not affect the amount of total α7nAChRs, β2nAChRs, and PDLIM5 proteins for one day, but increased the amount of these proteins for 3 and 7 days gradually, with 7-day nicotine treatment the most obvious (Fig. 1a). The relative protein levels showed that 7-day exposure to 1µM nicotine significantly increased the expression of total α7nAChRs for about 2.7-fold (Con vs Nic7d, 1.01 ± 0.11 vs 2.79 ± 0.23, p < 0.001), β2nAChRs for about 2.9-fold (Con vs Nic7d, 1.08 ± 0.18 vs 2.93 ± 0.30, p < 0.001), and PDLIM5 for about 2.5-fold (Con vs Nic7d, 0.89 ± 0.19 vs 2.26 ± 0.31, p < 0.001) (Fig. 1b). Immunofluorescence imaging performed at DIV10 also showed that 7-day nicotine exposure increased the puncta number of α7nAChRs, β2nAChRs and fluorescence intensity of PDLIM5. The puncta number of α7nAChRs and β2nAChRs increased about 1.7-fold (Fig. 1c, e). The fluorescence intensity of PDLIM5 was about 1.3-fold increase (Fig. 1d-e). These results indicated that chronic nicotine exposure could induce the up-regulation of α7nAChRs, β2nAChRs, and PDLIM5 in primary cultured hippocampal neurons.
To illustrate whether nicotine-induced up-regulation of PDLIM5 is dependent on AChRs, α7nAChRs and β2nAChRs were inhibited by methyllycaconitine (MLA, 200 nM) and dihydro-β-erythroidine (DhβE, 10 µM) respectively. The antagonists were added with nicotine at DIV3 and treated for 7-day in primary hippocampal neurons. As shown in Fig. 1f-g, nicotine-induced up-regulation of PDLIM5 protein levels could be blocked by MLA (Con vs Nic + MLA, 1.09 ± 0.28 vs 1.48 ± 0.35, p > 0.05), but not by DhβE (Con vs Nic + DhβE, 1.09 ± 0.28 vs 2.01 ± 0.17, p < 0.05). These results showed that the nicotine-induced up-regulation of PDLIM5 has a similar pattern with α7nAChRs, which implied a close relationship between these two proteins.
PDLIM5 associates with α7nAChRs, but not β2nAChRs in hippocampal neurons
It has been reported that α7nAChRs are presented at both presynaptic and postsynaptic sites[33], similar cellular localization was also found in PDLIM5[34, 35]. To reveal the relationship between these two proteins, we performed immunostaining to detect the subcellular colocalization of PDLIM5 and nAChRs in primary hippocampal neurons (DIV10). As the results shown in Fig. 2a, PDLIM5 colocalized more with α7nAChRs than β2nAChRs. Next, lysates from primary cultured hippocampal neurons were assessed by co-immunoprecipitation with nAChRs or PDLIM5 antibody respectively. The results showed that PDLIM5 interacted with α7nAChRs, but not β2nAChRs (Fig. 2b, left panel). Similar results have been observed in the lysates prepared from the rat hippocampi (Fig. 2b, right panel).
Since α7nAChRs and PDLIM5 are both located on the plasma membrane and in the cytoplasm of a neuron, then we examined the distribution of these two proteins in membrane and cytoplasm fractions in DIV10 cultured neurons. Immunoblot assay showed that nicotine exposure for 7 days caused α7nAChRs at the cell membrane for about 2.4-fold increase (Con vs Nic7d, 1.00 ± 0.13 vs 2.39 ± 0.11, p < 0.001) and in the cytoplasm for about 1.5-fold increase (Con vs Nic7d, 1.02 ± 0.05 vs 1.55 ± 0.02, p < 0.05) (Fig. 2c-d). Similarly, the upregulation of PDLIM5 caused by nicotine treatment appeared to be more on the membrane than the cytoplasm (Membrane: Con vs Nic7d, 0.99 ± 0.08 vs 1.99 ± 0.26, p < 0.001; Cytoplasm: Con vs Nic7d, 1.04 ± 0.23vs 1.45 ± 0.11, p < 0.05) (Fig. 2c-d). Our data suggest that nicotine-induced up-regulation of α7nAChRs and PDLIM5 is mainly located on the membrane.
PDLIM5 is necessary for the up-regulation and surface expression of α7nAChRs
To investigate the role of PDLIM5 in nicotine-induced α7nAChRs up-regulation, SH-SY5Y cells were transfected with a scramble-GFP plasmid, or a GFP-tagged PDLIM5 shRNA plasmid and then treated with 1 µM of nicotine for 3 days. The knockdown effect was validated through western blot analyses, as indicated in Fig. 3a-b (scramble vs PDLIM5 shRNA, 0.95 ± 0.08 vs 0.13 ± 0.01, p < 0.001). Compared to the β2nAChRs, chronic nicotine induced up-regulation of α7nAChRs was significantly reduced in PDLIM5 knockdown cells (Fig. 3b, α7nAChRs: Con vs Nic3d, 1.08 ± 0.11 vs 1.09 ± 0.09, p > 0.05; β2nAChRs: Con vs Nic3d, 0.92 ± 0.13 vs 2.88 ± 0.32, p < 0.01). To confirm the abolished effect of nicotine induced α7nAChRs up-regulation in PDLIM5 knockdown cells, we next tested the distribution of α7nAChRs in SH-SY5Y cells. As the data showed in Fig. 3c and 3d, up-regulation of α7nAChRs was impaired both in cell membrane and cytoplasm in nicotine treated PDLIM5 knockdown cells (Membrane: Con vs Nic3d, 1.00 ± 0.02 vs 0.94 ± 0.15; Cytoplasm: Con vs Nic3d, 1.03 ± 0.04 vs 1.01 ± 0.19, p > 0.05).
Since nicotine-induced up-regulation of α7nAChRs and PDLIM5 was mainly at the membrane, we hypothesized that PDLIM5 might induce surface α7nAChRs changes in hippocampal neurons. To test this hypothesis, knockdown effect against PDLIM5 in primary cultured hippocampal neurons was first evaluated by immunofluorescence. Compared to scramble, relative PDLIM5 protein intensity was significantly reduced in neurons expressing PDLIM5 shRNA (Fig. 4a,c, from 0.95 ± 0.06 to 0.32 ± 0.07, p < 0.001). Then, surface α-Bgtx labeling of α7nAChRs in neurons infected with PDLIM5 shRNA was observed (Fig. 4b-c). Knockdown of PDLIM5 significantly reduced the surface α-Bgtx labeling of α7nAChRs puncta number for about 35% and area for about 41% (puncta number, from 3.40 ± 0.27 to 2.20 ± 0.20, p < 0.01; area, from 0.49 ± 0.03 to 0.29 ± 0.04,p < 0.01) (Fig. 4c). As the expression of α7nAChRs is associated with its functional regulation, we next recorded the whole-cell nicotinic currents medicated by α7nAChRs. As expected, once knockdown the PDLIM5 in primary neurons, the α7nAChRs currents was reduced for about 50% compared to the scramble cells (Fig. 4d). Collectively, these results suggest that PDLIM5 plays an important functional role in α7nAChRs.
PDLIM5 interacts with α7nAChRs through the PDZ domain
As protein-interaction domains, PDZ domains are usually found in multi-domain scaffolding proteins[36]. To test whether PDZ domain in PDLIM5 plays such a role with α7nAChRs interaction, we generated the full-length PDLIM5, the corresponding truncations PDZ domain, LIM1-3 and D1 fragment (Fig. 5a), and verified by GST pulldown assays. PDLIM5-GST, PDZ-GST and D1-GST could be detected in α7nAChRs-His complexes, but not in LIM1-3-GST (Fig. 5b). It appeared that PDLIM5 and α7nAChRs interacted with each other in a fashion independent of the LIM domain. To test the specificity of the PDZ domain for α7nAChRs binding, the mutants target at the GLGF sequence was generated as MUTANT1 (GLGF → GAGA), MUTANT2 (GLGF → GAAA), and MUTANT3 (GLGF → AAAA) (Fig. 5a), and verified by GST pulldown assays (Fig. 5c). With the increase of mutant amino acids at the GLGF, PDLIM5 mutants found in the α7nAChRs-His complex were gradually decreased. This was also confirmed by the CO-IP assay, in which Flag-tagged α7nAChRs co-transfected with HA-tagged PDLIM5 or MUTANT3 in cultured HEK293T cells was used. HA-PDLIM5 was coimmunoprecipitated with Flag-α7nAChRs, but MUTANT1, MUTANT2, MUTANT3 had a much weaker interaction, with MUTANT3 had the weakest interaction (Fig. 5d). Besides, immunofluorescent staining also showed that the fluoresce intensity of surface α7nAChRs was decreased in the HEK293T cells expressing MUTANT3 (Fig. 5e). Therefore, our results strongly support that the interaction between PDLIM5 and α7nAChRs was mainly mediated by the PDZ domain of PDLIM5.
Nicotine promotes the interaction between PDLIM5 and α7nAChRs
We have shown that nicotine can induce the up-regulation of PDLIM5, it remains unclear how nicotine affects this up-regulation. To determine whether nicotine promotes stronger interaction between α7nAChRs and PDLIM5, binding thermodynamics of the interaction incubated with nicotine was analyzed using isothermal titration calorimetry (ITC). PDLIM5-GST and α7nAChRs-His was used for the ITC experiments. All interactions between PDLIM5-GST and α7nAChRs-His showed exothermic reactions. Binding affinity is typically measured and reported by the equilibrium dissociation constant (Kd), which is used to evaluate and rank order strengths of bimolecular interactions. In comparison with control, the binding affinity of the nicotine group was about 1 order of magnitude higher (Fig. 6 a, b). The Kd (µM) for the control group and nicotine group was 0.54 ± 0.19 vs 1.07 ± 0.52, respectively. This was also confirmed by the CO-IP assay, in which Flag-tagged α7nAChRs co-transfected with HA-tagged PDLIM5 in cultured HEK293T cells was used. The CO-IP assays also showed that nicotine significantly increased the protein level of PDLIM5 in α7nAChRs complexes and vice versa. (Fig. 6c). Collectively, these results demonstrated that nicotine promoted the affinity of a7nAChRs and PDLIM5.