Molecular Chaperone Calnexin Regulates the Function of Drosophila Sodium Channel Paralytic

Abstract

Neuronal activity mediated by voltage-gated channels provides the basis for higher-order behavioral tasks that orchestrate life. Chaperone-mediated regulation, one of the major means to control protein quality and function, is an essential route for controlling channel activity. Here we present evidence that Drosophila ER chaperone Calnexin colocalizes and interacts with the α subunit of sodium channel Paralytic. Co-immunoprecipitation analysis indicates that Calnexin interacts with Paralytic protein variants that contain glycosylation sites Asn313, 325, 343, 1463, and 1482. Downregulation of Calnexin expression results in a decrease in Paralytic protein levels, whereas overexpression of the Calnexin C-terminal calcium-binding domain triggers an increase reversely. Genetic analysis using adult climbing, seizure-induced paralysis, and neuromuscular junction indicates that lack of Calnexin expression enhances Paralytic-mediated locomotor deficits, suppresses Paralytic-mediated ghost bouton formation, and regulates minature excitatory junction potentials (mEJP) frequency and latency time. Taken together, our findings demonstrate a need for chaperone-mediated regulation on channel activity during locomotor control, providing the molecular basis for channlopathies such as epilepsy.

Keywords: Calnexin chaperone; NMJ synaptogenesis; locomotor behavior; paralytic; sodium channel.

Figure 1

Cnx colocalizes and interacts with Para. (A–C) Adult brains carrying Elav>ER-GFP were co-stained with Cnx (magenta) and Para (blue in A” and A”’, green in B, or magenta in C). Note that most of the GFP-positive cells were also Cnx- and Para- positive (white arrows in A”’ and C) and that Cnx and Para were colocalized (white arrows in A”’,B). Scale bar: 50 μm. (D,E) Protein lysates from S2 cells transfected with 3xFlag-GFP (control) or 3xFlag-cnx plasmids were subjected to Co-IP analysis. Our results revealed that Para interacts with Cnx and they are in the same protein complex. ^* non-specific band. (F) Co-IP analysis revealed that Cnx interacts with the first Para N-terminal intracellular segment (Para-N, amino acid 1–149). (G,H) Co-IP analysis on the interaction between Cnx and three Para protein variants: 6xMyc-Para_IS5-S6, 6xMyc-Para_IIIS5-S6, or 6xMyc-Para_IIS6-IIIS1. Proteins sizes of three para variants were about 25–30 kDa shown on the blots. Note that Cnx interacts with the two Para fragments containing glycosylation sites (6xMyc-Para_IS5-S6 and 6xMyc-Para_IIIS5-S6), but not the variant without one. Co-IP analyses were done by both anti-Flag- or anti-Myc-antibody conjugated beads.

Figure 2

Para protein levels are reduced when down regulating Cnx expression. (A–H”’) Adult brains expressing Elav>ER-GFP were dissected and stained with Para (Magenta) and HRP (Cyan). White rectangles represent areas enlarged and shown directly underneath. Genotypes shown: Elav; ER-GFP>LacZ (A–B”’), Elav; ER-GFP>cnxR⁴²³⁹⁷ (C–D”’), Elav; ER-GFP>cnxR¹⁰⁰⁷⁴⁰ (E–F”’), and Elav; ER-GFP>cnx^5-2, cnxR¹⁰⁰⁷⁴⁰ (G–H”’). (I,J) Statistics for the intensities of Para/HRP (I) and HRP (J) in adult brains. Note a significant decrease in Para/HRP intensities when cnxR⁴²³⁹⁷ or cnxR¹⁰⁰⁷⁴⁰ was expressed. (K) Western blot analysis on Para protein levels in adult brains (n = 50) carrying the following genotypes: Elav>LacZ, Elav>cnxR⁴²³⁹⁷, Elav>cnxR¹⁰⁰⁷⁴⁰, Elav>cnx^5-2, Elav>cnx^5-2; cnxR⁴²³⁹⁷. Input control: α-Tubulin. (L–M”) VNC motor axons were stained with Para (Green) and HRP (Magenta) in w¹¹¹⁸ and cnx² larval NMJs. (N) Statistics for the intensities of Para/HRP in w¹¹¹⁸ and cnx² axons. Note a decrease in Para levels in cnx² axons. (O) Western blot analysis revealed a decrease in Para protein levels in cnx² mutant adult brains (n = 50). Input control: α-Tubulin. Data were shown as mean ± SEM. Scale bar: 20 μm (NMJs) and 50 μm (adult brains). Number of adult brains dissected per genotype was shown in the bars. At least 13 brains were analyzed per genotype and six different areas in the brains were selected. ^**p < 0.01, ^***p < 0.001. Data were shown as mean ± SEM. At least three independent experiments were performed.

Figure 3

Para protein levels are upregulated when overexpressing the Cnx C-terminal calcium-binding domain. (A–F”’) Elav>ER-GFP adult brains expressing LacZ, cnx-N, or cnx-C were dissected and stained with Para (Magenta) and HRP (Cyan). Scale bar: 50 μm. White rectangles represent areas enlarged and shown directly underneath. (G,H) Quantification of Para/HRP intensities (G) or HRP (H) for the above genotypes. Note a significant increase in the Para/HRP intensities when Cnx-C was overexpressed. (I) Western blot analysis revealed an increase in Para protein levels when Cnx-C was overexpressed. ^***p < 0.001. Data were shown as mean ± SEM. Number of adult brains dissected per genotype was shown in the bars. At least 13 brains were analyzed per genotype and six different areas in the brains were selected. At least three independent experiments were performed.

Figure 4

Cnx regulates Para-mediated adult climbing and paralysis. Adult climbing ability was analyzed by a multi-cylinder electrical system that allows synchronization of both time and activity. w¹¹¹⁸ or LacZ was used as a control, and the climbing distance was measured at a time interval of 6–7 s. At least 90 flies were analyzed, with 10 in each cylinder, and three independent experiments were performed. Climbing distance for each genotype was plotted as bar graphs. Only male flies were shown. (A) Climbing distance was measured for flies carrying the following genotypes listed from left to right: Elav; cnx²>LacZ, Elav; cnx²>cnx^5-2, Elav; cnx²>cnx², Elav; cnx²>cnx^5-2; cnx², Elav; cnx²>cnx-N; cnx², Elav; cnx²>cnx-C; cnx². Note a significant decrease in climbing activity in cnx² mutant (third gray bar from the left). Overexpression of full-length Cnx (cnx^5-2) or Cnx containing only its C-terminus (cnx-C) rescued the reduction in climbing activity in cnx² mutant, whereas overexpression of Cnx containing only its N-terminus (cnx-N) failed to do so. (B) Climbing distance was measured for flies carrying the following genotypes listed from left to right: Elav; Dicer2>LacZ, Elav; Dicer2>cnxR⁴²³⁹⁷, Elav; Dicer2>cnxR¹⁰⁰⁷⁴⁰, Elav; Dicer2>cnx^5-2, Elav; Dicer2>cnx^5-2; cnxR⁴²³⁹⁷, Elav; Dicer2>cnx^5-2; cnxR¹⁰⁰⁷⁴⁰. cnx RNAi expression driven by Elav; Dicer2 leads to a decrease in climbing distance (second and third gray bars from the left), whereas Cnx overexpression does not alter climbing ability. (C) Adult climbing ability was assessed similarly for the following genotypes: Elav>LacZ, Elav>cnxR⁴²³⁹⁷, Elav>cnxR¹⁰⁰⁷⁴⁰, Elav>paraR⁶¹³¹, Elav>cnxR⁴²³⁹⁷, paraR⁶¹³¹, and Elav>cnxR¹⁰⁰⁷⁴⁰; paraR⁶¹³¹. (D) Cnx regulates the recovery kinetic of Para-mediated paralysis. Percentage of recovered flies from paralysis (Recovered/Total flies, %) was plotted against time cumulatively at an interval of 10 s using SPSS and Prism software. Purple curves: para^ts1; Sca>cnx^5-2, blue curves: para^ts1; Sca>LacZ, green curves: para^ts1; Sca>cnxR⁴²³⁹⁷, red curves: para^ts1; Sca>cnxR¹⁰⁰⁷⁴⁰. (E) Recovery time was defined differently as the time required for half of the flies recovered, climbed up, and passed the fixed height (4 cm). The time was calculated and plotted for the following genotypes: para^ts1; Sca>LacZ, para^ts1; Sca>cnx^5-2, para^ts1; Sca>cnxR⁴²³⁹⁷, and para^ts1; Sca>cnxR¹⁰⁰⁷⁴⁰, ^*p < 0.05, ^**p < 0.01, ^***p < 0.001, Data were shown as mean ± SEM.

Figure 5

Cnx regulates Para-mediated ghost bouton formation in NMJs. Number of boutons (A), branches (B), and boutons with diameter > 5 μm (C) were quantified for muscle 6/7 NMJs. Results were shown for the following genotypes: Elav>LacZ, Elav>cnxR⁴²³⁹⁷, Elav>cnxR¹⁰⁰⁷⁴⁰, Elav>paraR⁶¹³¹, Elav>cnxR⁴²³⁹⁷, paraR⁶¹³¹, and Elav>cnxR¹⁰⁰⁷⁴⁰; paraR⁶¹³¹. (D) Muscle 6/7 NMJs stained with HRP (Magenta) and DLG (Green) were shown for the above genotypes. Note an increase in the number of ghost boutons stained positively for HRP but not DLG in Elav>paraR⁶¹³¹ NMJs (white arrows). (E) Quantification of ghost bouton numbers for NMJs carrying the above genotypes. Note a decrease in the number of ghost boutons in the presence of both cnx and para RNAi expression, ^*p < 0.05, ^**p < 0.01, ^***p < 0.001, Scale bar: 20 μm. Data were shown as mean ± SEM. Number of NMJs per genotype was shown in the bars. At least 12 NMJs were analyzed and three independent experiments were performed for each genotype.

Figure 6

Cnx and Para regulate the latency time and mEJP frequency during synaptic function. (A–F) Samples traces of EJPs and mEJPs were shown for the following genotypes: Elav>w¹¹¹⁸, Elav>cnxR⁴²³⁹⁷, Elav>cnxR¹⁰⁰⁷⁴⁰, Elav>paraR⁶¹³¹, Elav>cnxR⁴²³⁹⁷, paraR⁶¹³¹, and Elav>cnxR¹⁰⁰⁷⁴⁰; paraR⁶¹³¹. (G–J) Statistics for electrophysiological parameters such as mEJP amplitude (mV, G), mEJP frequency (Hz, H), EJP amplitude (mV, I), and latency time (msec, J) were shown. Note that mEJP frequency and the latency time were generally affected when Cnx or Para expression was downregulated. A further decrease in mEJP frequency (H) or increase in the latency time (J) was detected when both cnx and para RNAi were expressed. ^*p < 0.05, ^**p < 0.01, ^***p < 0.001, Data were shown as mean ± SEM. At least 7 NMJs were analyzed and three independent experiments were performed for each genotype.

Figure 7

Schematic diagram on Cnx-mediated Para regulation in two distinct modes. Cnx regulates Para expression and function via two distinct pathways. First, Cnx interacts with glycosylation sites on Para, suggesting a regular quality control check-up point in ER (A). In addition, Cnx C-terminal domain buffers intracellular Ca²⁺ ion levels, creating a Ca²⁺gradient both inside and outside plasma membrane and the ER membrane (B). This difference in Ca²⁺ ion levels in turn regulates the sodium ion flows via the sodium-calcium exchanger on the plasma membrane or promotes ER vesicular trafficking, both routes affecting Para expression and function.