Large-conductance Ca²⁺ and voltage-activated (BK) channels are key regulators of cellular excitability and are modulated by auxiliary subunits including β1-4^1,2, γ1-4² and LINGO1-3^3-5. Co-expression of LINGO1-3 proteins with BK causes inactivation and shifts the voltage dependence of activation (V_1/2ACT), towards more negative potentials^3,4 in BK:LINGO1 and BK:LINGO2, but to more positive potentials⁵ in BK:LINGO3, compared to BK alone. Previous work has identified the intracellular C-terminal tail⁶ as the most likely determinant of these effects and sequence alignment revealed a non-conserved residue in the proximal tail that differs between LINGO isoforms (E594 in LINGO1, D578 in LINGO2, S564 in LINGO3 and T569 in LINGO4), suggesting that this position may contribute to differences in V_1/2ACT shifts.

 

To test this, BK and LINGO cDNA was transiently co-transfected into HEK293 cells and macroscopic currents were recorded at 37^oC from inside-out patches using patch-clamp electrophysiology. Pipette and bath solutions contained 140mM K⁺ and the [Ca²⁺] at the cytosolic surface ranged from 100nM-10mM. Conductance-voltage (G-V) relationships were fitted with a Boltzmann function to determine V_1/2ACT. Site-directed mutagenesis of D578 in LINGO2 and E594 in LINGO1 was examined in both full-length and inactivation-deficient (ΔMKMI) constructs³. Data is presented as mean±SEM, all experiments were carried out on five to eight patches and statistical comparisons were performed using one-way ANOVA.

 

In full length BK:LINGO1 channels, the V_1/2ACT was 121 ± 2 mV in 100 nM Ca²⁺ as shown in previous studies^3,5. Substitution of the equivalent residue from LINGO2 (BK:LINGO1_E594D) resulted in a V_1/2ACT of 129 ± 4 mV, (ns vs BK:LINGO1). In contrast, BK:LINGO1_E594S significantly shifted V_1/2ACT to 164 ± 4 mV in 100 nM Ca²⁺ (p<0.01 vs WT BK:LINGO1), consistent with the idea that swapping in the equivalent residue from LINGO3 at this position could predictably alter V_1/2ACT.

 

Co-expression of BK:LINGO2 also produced a negative shift in V_1/2ACT to 131 ± 2 mV in 100 nM Ca²⁺ as shown previously^4-6. Swapping in the equivalent residue from LINGO1 (ie BK:LINGO2_D578E ) resulted in a V_1/2ACT of 114 ± 2 mV, whereas swapping in the equivalent residue from LINGO3 (BK:LINGO2_D578S, V_1/2ACT=167 ± 2 mV) or LINGO4 (BK:LINGO2_D578T V_1/2ACT=157 ± 2 mV) produced significant positive shifts in V_1/2ACT (p<0.05 vs WT BK:LINGO2).

 

When the LINGO2 inactivation particle was removed (BK:LINGO2_ΔMKMI), the V_1/2ACT in 100 nM was significantly shifted to 109 ± 3 mV (p<0.01 vs BK:LINGO2), suggesting that inactivation affected the accurate determination of V_1/2ACT in full length BK:LINGO2. When the equivalent residue to LINGO1 was substituted into LINGO2 (BK:LINGO2_D578EΔMKMI) the V_1/2ACT was 96 ± 2 mV in 100 nM Ca²⁺ but this failed to reach statistical significance. In contrast, when the equivalent residue from LINGO3 was substituted in, to make BK:LINGO2_D578SΔMKMI, the V_1/2ACT predictably shifted positively to 152 ± 2 mV in 100 nM Ca²⁺ (p<0.01 vs BK:LINGO2_ΔMKMI).

 

These findings suggest that a single conserved residue within the juxta-membrane region of the LINGO intracellular tail is a key determinant of BK channel V_1/2ACT in both LINGO1 and LINGO2 since substitutions at this position produced predictable, isoform-specific shifts in V_1/2ACT.