Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 Nov 18:9:120.
doi: 10.3389/fnmol.2016.00120. eCollection 2016.

Structural and Functional Consequences of Connexin 36 (Cx36) Interaction with Calmodulin

Ryan C F Siu  1 , Ekaterina Smirnova  2 , Cherie A Brown  1 , Christiane Zoidl  3 , David C Spray  4 , Logan W Donaldson  1 , Georg Zoidl  3
Affiliations

Structural and Functional Consequences of Connexin 36 (Cx36) Interaction with Calmodulin

Ryan C F Siu et al. Front Mol Neurosci. .

Abstract

Functional plasticity of neuronal gap junctions involves the interaction of the neuronal connexin36 with calcium/calmodulin-dependent kinase II (CaMKII). The important relationship between Cx36 and CaMKII must also be considered in the context of another protein partner, Ca2+ loaded calmodulin, binding an overlapping site in the carboxy-terminus of Cx36. We demonstrate that CaM and CaMKII binding to Cx36 is calcium-dependent, with Cx36 able to engage with CaM outside of the gap junction plaque. Furthermore, Ca2+ loaded calmodulin activates Cx36 channels, which is different to other connexins. The NMR solution structure demonstrates that CaM binds Cx36 in its characteristic compact state with major hydrophobic contributions arising from W277 at anchor position 1 and V284 at position 8 of Cx36. Our results establish Cx36 as a hub binding Ca2+ loaded CaM and they identify this interaction as a critical step with implications for functions preceding the initiation of CaMKII mediated plasticity at electrical synapses.

Keywords: CaMKII; calmodulin; connexins; electrical synapse; plasticity; protein interaction.

PubMed Disclaimer

Figures

FIGURE 1
FIGURE 1
Expression, localization, and gap junction plaque (GJP) formation of Cx36 in Neuro2a cells. (A) Sequence alignment of Cx36 orthologs from major vertebrate phyla showing a conserved CaM-binding domain overlapping with the CaMKII pseudosubstrate binding site. The 1–8–14 positions of the CaM-binding motif are highlighted. (B) Western blot analysis of transiently transfected Neuro2a cells expressing Cx36 wild-type, Cx36 mutants, CaM, and CaME1234Q. Expressed proteins were detected using an anti-EGFP antibody (in red), and an anti-beta-actin antibody (in green) as the control. M indicated the protein standard. (C) Rectangles indicate representative GJPs in Neuro2a cells double transfected with CaM (blue), Cx36 (red) wild-type, and mutant Cx36 proteins. Scale bar: 10 μm. (D,E) Quantification of gap junction size and frequency demonstrating that gap junction size is not affected by mutations of the core CaM binding region, but mutation W277A affects GJP frequency. Error bars box plot: maximum and minimum range with median; Mann–Whitney U test, ∗∗∗p < 0.001.
FIGURE 2
FIGURE 2
Colocalization of Cx36 with organelle markers indicating interaction occurring as early as at the endoplasmic reticulum. (A) Representative 3D images of Neuro2A cells expressing wild-type Cx36 or the W277A mutant. Note that cell pairs derived from single independent transfections tagging the proteins of interest with EGFP or DsRed monomer. The increased presentation of mixed red and turquoise vesicles in W277A transfected cell pairs suggests reduced GJP stability and increased protein turnover. Insets represent magnifications of the GJP region. (B) Neuro2a cells expressing wild-type Cx36–ECFP (in red) and DsRed monomer tagged organelle markers (endoplasmic reticulum, Golgi apparatus, and mitochondria, in light turquoise). The localization of Cx36 plaques is indicated with white arrows. (C) Mander’s overlap coefficients calculated with the ZEN 2010 program for Cx36 WT (n = 30) and W277A (n = 40) with the different organelle markers are shown with error bars indicating the maximum and minimum range. 3D images in (A,B) were created with Bitplane Imaris 7.6.4. Scale bars in (A,B): 10 μm. Error bars box plot: maximum and minimum range with median; Mann–Whitney U test, ∗∗∗p < 0.01.
FIGURE 3
FIGURE 3
FRET analysis of Cx36 and CaM interaction. Neuro2a cells were transiently co-transfected, and FRET efficiencies (FRETeff) determined 48 h post transfection. (A) FRETeff of Neuro2a cells expressing active FRET control pairs. (B) DsRed and ECFP protein tags are interchangeable and with minimal impact on FRETeff. (C,D) FRETeff at GJPs and vesicles without and with 2 mM extracellular calcium [Ca2+]E and 2 μM ionomycin in the presence of the pharmacological calmodulin blocker W-7 (10 μM), the calcium chelator BAPTA-AM (24 μM), or the Ca2+ insensitive CaME1234Q mutant. (E,F) FRET distance (FRETdis) highlighting that calcium-dependent binding of Cx36 and CaM is significantly shortened at vesicles. Abbreviations: Iono, treatment with 2 mM [Ca2+]E and 2 μM ionomycin. V, vesicles; GJP, gap junction plaque. Error bars box plot: maximum and minimum range with median; Mann–Whitney U test significance, ∗∗∗p < 0.01, NS, not significant.
FIGURE 4
FIGURE 4
Alanine scanning mutagenesis reveals W277 as the critical site for Cx36–CaM binding. (A) FRETeff at vesicles in Neuro2a cells. Cells expressed fluorescently tagged Cx36–ECFP and Cx36–DsRed monomer as the control. Wild-type Cx36–DsRed monomer and Cx36 mutants (G276A, W277A, R278A) in the presence of CaM were stimulated with 2 μM of ionomycin for 10 min. (B) FRET distance with the same parameters as FRET efficiency. The dashed line indicates the FRET distance threshold at 10 nm. Error bars box plot: maximum and minimum range with median; Mann–Whitney U test, ∗∗p < 0.05, ∗∗∗p < 0.01.
FIGURE 5
FIGURE 5
Ethidium bromide dye uptake and redistribution after photobleaching. (A–D) Cell pairs expressing Cx36 wild-type proteins or the W277A mutant. The left column highlights examples for each treatment condition, with the cells selected for photobleaching encircled (large circles) and cell positions of regions of interest indicated (small circles; R1, R2, and R0 = background). The right column shows traces representing fluorescence over time for regions of interests. Scale bars: 10 μm. (E) Quantification of fluorescence recovery in regions of interest. Treatments: Iono (2 mM [Ca2+]E and 2 μM ionomycin), W-7 (10 μM W-7, calmodulin inhibitor), CBX (50 μM carbenoxolone, gap junction blocker). Error bars display the maximum and minimum range. NS, not significant. Paired t-test, p-values ∗∗ < 0.01, ∗∗∗ < 0.005.
FIGURE 6
FIGURE 6
Structural features of the Cx36–CaM complex. (A) A schematic representation of the hybrid Cx36–CaM protein used for the NMR structural determination. The N- and C-terminal lobes are colored for reference, followed by a short linker leading into a region in the carboxy-terminal cytoplasmic tail of Cx36. (B) Cartoon representation of the lowest energy model in the ensemble of twenty solutions. Four calcium ions are depicted as balls and other features colored consistently with previous panels. (C) Helical wheel representation of Cx36 bound to CaM as determined by experimental observations. W277 and V284 at anchor positions 1 and 8 are colored red and other hydrophobic amino acids colored black. (D) Sequence alignment of Cx36, smooth muscle myosin light chain kinase (smMLCK), and NMDA receptor NR1 (NMDAR NR1). Key contact sites between CaM and these ligands are made at positions 1, 8, and 14. Boxes indicate the helical amino acids in each ligand. The ligand may extend amino-terminally before anchor position 1 but for clarity, only amino acids from position 1 are shown. (E) Ligand plot (Wallace et al., 1995) of key interactions observed in the Cx36–CaM complex. Green lines denote hydrogen bonds and ionic interactions.
FIGURE 7
FIGURE 7
Isothermal titration calorimetry (ITC) of Cx36 peptides and CaM. Calorimetric traces and integrated isotherms acquired at 30°C for (A) Calcium saturated CaM titrated with a Cx36 derived peptide, and (B) Calcium saturated CaM titrated with the same peptide bearing a W277A substitution. No binding was detected between CaM and W277A mutant peptide. Abbreviations: Kd, dissociation constant; n.d., not detected.
FIGURE 8
FIGURE 8
Model of calmodulin (CaM), Ca2+/calmodulin dependent kinase 2a (CaMKIIa) and connexin 36 (Cx36) in plasticity of the electrical synapse. Different to our original model, Cx36 can either bind calcium-activated CaM (right) or interact with CaMKIIa (left) at the vesicular level. The initial binding of CaM and Cx36 is critically dependent on elevated calcium concentrations. As the Cx36–CaM loaded vesicles traffic to the cellular membrane, the response to intracellular calcium activation shifts toward CaMKIIa interaction by calcium elevation. At the nexus, CaM serves as a kinase activator bound directly to CaMKIIa, and CaMKIIa plays an active part in modulating Cx36 nexus conductance (Illustration by Lesia Szyca, 2016).
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Abbaci M., Barberi-Heyob M., Stines J. R., Blondel W., Dumas D., Guillemin F., et al. (2007). Gap junctional intercellular communication capacity by gap-FRAP technique: a comparative study. Biotechnol. J. 2 50–61. 10.1002/biot.200600092 - DOI - PubMed
    1. Alev C., Urschel S., Sonntag S., Zoidl G., Fort A. G., Höher T., et al. (2008). The neuronal connexin36 interacts with and is phosphorylated by CaMKII in a way similar to CaMKII interaction with glutamate receptors. Proc. Natl. Acad. Sci. U.S.A. 105 20964–20969. 10.1073/pnas.0805408105 - DOI - PMC - PubMed
    1. Ataman Z. A., Gakhar L., Sorensen B. R., Hell J. W., Shea M. A. (2007). The NMDA receptor NR1 C1 region bound to calmodulin: structural insights into functional differences between homologous domains. Structure 15 1603–1617. 10.1016/j.str.2007.10.012 - DOI - PMC - PubMed
    1. Bayer K. U., De Koninck P., Leonard A. S., Hell J. W., Schulman H. (2001). Interaction with the NMDA receptor locks CaMKII in an active conformation. Nature 411 801–805. 10.1038/35081080 - DOI - PubMed
    1. Bennett M. V., Zukin R. S. (2004). Electrical coupling and neuronal synchronization in the Mammalian brain. Neuron 41 495–511. 10.1016/S0896-6273(04)00043-1 - DOI - PubMed