Click here to close
Hello! We notice that you are using Internet Explorer, which is not supported by Xenbase and may cause the site to display incorrectly.
We suggest using a current version of Chrome,
FireFox, or Safari.
Int J Mol Sci
2024 Aug 15;2516:. doi: 10.3390/ijms25168879.
Show Gene links
Show Anatomy links
The Ubiquitin Ligase Adaptor NDFIP1 Interacts with TRESK and Negatively Regulates the Background K+ Current.
Pergel E
,
Tóth DJ
,
Baukál D
,
Veres I
,
Czirják G
.
???displayArticle.abstract???
The TRESK (K2P18.1, KCNK18) background potassium channel is expressed in primary sensory neurons and has been reported to contribute to the regulation of pain sensations. In the present study, we examined the interaction of TRESK with NDFIP1 (Nedd4 family-interacting protein 1) in the Xenopus oocyte expression system by two-electrode voltage clamp and biochemical methods. We showed that the coexpression of NDFIP1 abolished the TRESK current under the condition where the other K+ channels were not affected. Mutations in the three PPxY motifs of NDFIP1, which are responsible for the interaction with the Nedd4 ubiquitin ligase, prevented a reduction in the TRESK current. Furthermore, the overexpression of a dominant-negative Nedd4 construct in the oocytes coexpressing TRESK with NDFIP1 partially reversed the down-modulating effect of the adaptor protein on the K+ current. The biochemical data were also consistent with the functional results. An interaction between epitope-tagged versions of TRESK and NDFIP1 was verified by co-immunoprecipitation experiments. The coexpression of NDFIP1 with TRESK induced the ubiquitination of the channel protein. Altogether, the results suggest that TRESK is directly controlled by and highly sensitive to the activation of the NDFIP1-Nedd4 system. The NDFIP1-mediated reduction in the TRESK component may induce depolarization, increase excitability, and attenuate the calcium dependence of the membrane potential by reducing the calcineurin-activated fraction in the ensemble background K+ current.
Figure 1. NDFIP1 reduced the TRESK background K+ current. (A) Two-electrode voltage clamp (TEVC) measurement of three groups of Xenopus oocytes expressing human TRESK (n = 7), coexpressing TRESK with human NDFIP1 (+NDFIP1 (1×), n = 6), or coexpressing the channel with a higher amount of NDFIP1 (+NDFIP1 (2.5×), n = 6; two and half times more NDFIP1 cRNA was microinjected). Inward (negative) currents were measured at −100 mV. Extracellular [K+] was increased from 2 to 80 mM, as shown above the curves, and the oocyte was then challenged with ionomycin (0.5 μM), as indicated by the horizontal black bar. The average currents and SD are plotted. (B) Statistical analysis of the effect of NDFIP1 on the basal TRESK current. The basal TRESK currents were calculated as the difference in 2 and 80 mM [K+], before the application of ionomycin (see panel (A)). (C) Statistical analysis of the effect of NDFIP1 on the TRESK current stimulated with ionomycin. The stimulated TRESK currents were measured at the end of the application of ionomycin in 80 mM [K+] (see panel (A)), and the small currents in 2 mM [K+] were subtracted. * p < 5 × 10−4, ns: not significant (Dunn’s test after Kruskal–Wallis ANOVA). Although not significant with Dunn’s test, the difference in the basal currents between the TRESK and NDFIP1 (1×) groups in panel B was significant with the Mann–Whitney U test (p < 0.005), which gave p < 0.015 after Bonferroni’s correction. TRESK vs. NDFIP1 (1×) in panel (C) gave p < 0.05 with the Mann–Whitney U test after Bonferroni’s correction.
Figure 2. TRESK was regulated by NDFIP1 more effectively than TREK-1 and GIRK4. (A) K+ currents in two groups of oocytes expressing TRESK, or coexpressing TRESK with NDFIP1 (0.67×), as indicated below the plot. (B) Similar experiment as in panel A with TREK-1. (C) Similar experiment as in panel A with mutant GIRK4. Data are from the same oocyte preparation, n = 5 in all groups. * p < 0.01 with Mann–Whitney U test, ns: not significant.
Figure 3. Mutation of PY motifs in NDFIP1 prevented the effect of the adaptor protein on TRESK. K+ currents in three groups of oocytes expressing TRESK, coexpressing TRESK with NDFIP1, or coexpressing TRESK with a mutant version of NDFIP1 lacking PY motifs (NDFIP1-noPPxY), as indicated below the plot. Data are from the same oocyte preparation, n = 7 in all groups. * p < 0.04, ** p < 0.02 with Dunn’s test after Kruskal–Wallis ANOVA, ns: not significant.
Figure 4. Dominant negative Nedd4 counteracted the NDFIP1-mediated reduction in the TRESK current. K+ currents in four groups of oocytes expressing TRESK, coexpressing TRESK with NDFIP1, or coexpressing wild-type (wt) or dominant negative (dn) Nedd4 with TRESK and NDFIP1 (triple coexpression), as indicated below the plot. Data are from the same oocyte preparation, n = 7 in all groups. * p < 0.015, ** p < 10−4 with Dunn’s test after Kruskal–Wallis ANOVA, ns: not significant.
Figure 5. NDFIP1 reduced the K+ current of all the tested lysine-mutant TRESK constructs. The K+ currents in the oocytes expressing the different lysine-mutant TRESK constructs (see above the panels), in the absence or presence of NDFIP1 coexpression, as indicated below the graphs. Data points on the same panel always come from the same oocyte preparation, n = 5–7 in all groups. * p < 0.02, ** p < 0.01 with Mann–Whitney U test. # RFRK motif of TRESK was mutated to NFNN, including the K168N mutation. In the Δ172–248 and Δ188–275 constructs, the appropriate regions were replaced by a lysine-free linker. In the ΔKK383,384 construct, the C-terminal double lysine was deleted.
Figure 6. Human TRESK was ubiquitinated in an NDFIP1-dependent manner. The HA2-N70Q-hTRESK protein was pulled down with anti-HA resin from the solubilized membrane preparations of three groups of oocytes expressing HA-tagged TRESK or NDFIP1, or coexpressing both proteins, as indicated below the images. On the left side, a Western blot was performed with an anti-ubiquitin primary antibody (anti-Ub, as indicated above the image). On the right side, an anti-HA immunoblot of the same membrane is shown. Representative of three similar experiments. * monomeric TRESK protein, ** mono-ubiquitinated TRESK in lane 2.
Figure 7. NDFIP1 is co-immunoprecipitated with TRESK. Three groups of oocytes expressed HA2-N70Q-hTRESK, His8-hNDFIP1, or both proteins, as indicated below the image. The HA2-N70Q-hTRESK protein was pulled down with anti-HA resin from the solubilized membrane preparations of these oocyte groups, and the co-immunoprecipitation of His8-hNDFIP1 was detected with an anti-His immunoblot (upper panel). Aliquots of the solubilized membrane preparations (Input) were also analyzed with an anti-His immunoblot on the same blotting membrane (middle panel). The presence of the bait protein was verified in the samples after anti-HA immunoprecipitation with the anti-HA immunoblot (lower panel; dimeric TRESK was detected). Representative of three similar experiments. IP: immunoprecipitation; WB: Western blot/immunoblot; anti-HA: antibody against the influenza hemagglutinin epitope; anti-His: antibody against the hexahistidine tag.
Figure 8. Coexpression of NDFIP1 decreased the immunofluorescence signal of HA-tagged TRESK on the oocyte surface. Three groups of oocytes expressed HA2-N70Q-hTRESK, coexpressed this construct with NDFIP1, or remained as a non-injected control (non-inj), as indicated in the representative images and below the graph. White scale bars indicate 100 μM. Data are from the same oocyte preparation, n = 10 in all groups. * p < 0.02, ** p < 0.005 with Dunn’s test after Kruskal–Wallis ANOVA.
