An unrelated amphipathic helix functionally substitutes for the last 19 amino acids of CpxII CTD. (A)

Primary sequence of CpxIÍs CTD (left) and the mutant CpxII 1-115 amphipathic α helix (right). (B) Mean [Ca]i levels (upper panel) and the corresponding membrane capacitance response (lower panel) from CpxII ko cells (red n=15), and those expressing either wt CpxII (blue n=15), the CpxII 1-115 mutant (purple n=14), or the CpxII 1-115 amphipathic helix mutant (orange n=16). (C) Amplitudes of the readily releasable pool (RRP), the slow releasable pool (SRP), and the rate of sustained release (SR). CpxII 1-115 amphipathic helix mutant fully supports exocytosis like the wt protein. (D) The time constants for the EB components (τRRP and τSRP), and the exocytotic delay are rescued by the CpxII mutants like the wt protein. (E) Mean premature secretion of the tested groups at similar submicromolar [Ca]i before triggering the flash response shown in (B). (F) The rate of tonic exocytosis (determined at similar [Ca2+]i: in nM, CpxII ko: 789 ± 120; CpxII ko + CpxII: 777 ± 57; CpxII ko+ CpxII 1-115: 739 ± 84; CpxII ko+ CpxII1-115 amph helix: 796 ± 64) is significantly reduced with CpxII and CpxII 1-115 amph helix mutant but not with the truncated CpxII 1-115 mutant. ANOVA or Kruskal-Wallis followed by corresponding post-hoc test. **p<0.01; ***p<0.001. Error bars indicate mean ± SEM.

Disrupting the hydrophobic face of the amphipathic helix impairs the ability of the CpxII CTD to prevent premature release. (A, D)

Primary sequences depicting the point mutations L124E-L128E (A) and L124W-L128W (D) within the amphipathic region of CpxII CTD. (B, E) Mean [Ca]i levels (upper panel) and the corresponding ΔCM (lower panel) from CpxII ko cells (red n=25), and those expressing CpxII wt (blue n=25), or the mutants CpxII L124E-L128E (yellow n=26) in (B), and CpxII ko cells (red n=18) or those expressing the CpxII wt protein (blue n=15) or CpxII LL-WW (dark green n=12) in (D). Mean amplitudes of the RRP, SRP and SR show a significant reduction in the size of the EB by the CpxII L124E-L128E mutant, whereas CpxII L124W-L128W rescued the synchronized exocytosis like CpxII wt. (C, F) Average tonic exocytosis at similar submicromolar [Ca]i levels. Note that the mutant CpxII L124W- L128W fully clamps premature secretion like the CpxII wt protein, but the CpxII L124E- L128E mutant fails to do so. ANOVA or Kruskal-Wallis followed by corresponding post- hoc test. *p<0.05; **p<0.01; ***p<0.001. Error bars indicate mean ± SEM.

CpxII-CTD and its mutant variants differentially interact with SNAREs and SytI. (A, C)

The GST-CpxII CTD fusion protein co-precipitates SytI and Syx1a from detergent extract of mouse brain. Substitution of leucine residues with glutamate residues (CpxII-CTD L124E-L128E, LE) abolishes any binding to CpxII CTD (A). Substitution with tryptophan residues (CpxII-CTD L124W-L128W, LW), instead, or replacement of the entire CpxII CTD with an unrelated amphipathic helix (CpxII-CTD-ampH, C) is tolerated. Samples were eluted from the column by thrombin cleavage. Equal volumes of supernatant (Sbh), the thrombin-eluted (STh) and the non-eluted fraction (P) were analyzed by SDS-PAGE (12% gel) and Western blotting with antibodies against the indicated antigens. No binding to GST alone could be detected. (B, D) Corresponding Coomassie gels documenting the integrity of the GST fusion proteins, remaining in the non-eluted fraction (P) after thrombin cleavage.

The cluster of glutamate residues in CpxII CTD facilitates Ca2+ triggered exocytosis. (A)

Schematic representation of CpxII highlighting the position of the glutamate cluster within the CTD. Primary sequence of CpxII CTD showing the exchange of glutamate into alanine residues (CpxII E-A mutant). (B) Mean [Ca]i levels (upper panel) and CM response from wt cells (black n=18), CpxII ko cells (red n=19), and CpxII ko cells expressing either CpxII wt (blue n=17), or the mutant CpxII E-A (green n=24). (C) The mutant CpxII E-A only partially restores the fast component (RRP) of synchronized exocytosis. (D) The CpxII E-A mutation significantly slows down the speed of the RRP and SRP and delays the stimulus secretion coupling. (E) Average tonic exocytosis at similar submicromolar [Ca]i levels with the quantification in (F) showing that the mutant CpxII E-A effectively prevents premature vesicle loss at submicromolar [Ca]i levels like CpxII wt. ANOVA or Kruskal-Wallis followed by corresponding post-hoc test. *p<0.05; **p<0.01; ***p<0.001. Error bars indicate mean ± SEM.

CpxII and SytI act interdependently in triggering synchronized exocytosis.

(A) Mean [Ca]i levels (upper panel) and CM responses from wt cells (black n=50), SytI ko cells (green n=46), CpxII ko (red n=35), CpxII/SytI dko (dark blue n=26). (B) Quantification of the RRP and SRP size shows that SytI deletion in the absence of CpxII does not further aggravate the secretory deficit. (C) Normalized CM (as shown in A) scaled to the wt response 1s after the flash. Inset, extended scaling of normalized CM during the first 20ms after flash (arrow) depicting the delayed onset of secretion. (D) Kinetics of RRP and SRP exocytosis and the secretory delay. (E) Loss of SytI does not alter premature vesicle fusion. Additional loss of CpxII increases premature exocytosis to the level of CpxII ko (F). (G) Mean [Ca]i levels (upper panel) and CM responses from wt cells (black n=17), SytVII ko cells (orange n=20), CpxII ko (red n=20), CpxII/SytVII dko (blue n=24). (H) The RRP and SRP sizes are reduced by either CpxII or SytVII single ko, while the combined loss of CpxII and SytVII abolishes the EB. (I) Normalized CM (as shown in G) scaled to the wt response 1s after the flash. Inset, extended scaling of normalized CM during the first 20ms after flash (arrow). (J) Loss of CpxII but not of SytVII slows the time constants of RRP and SRP exocytosis (τRRP, τSRP) and prolongs the secretory delay. (K, L) SytVII deficiency diminishes the elevated premature exocytosis of CpxII ko cells (SytVII CpxII dko) and reduces it compared to wt cells. ANOVA or Kruskal- Wallis followed by corresponding post-hoc test. *p<0.05; **p<0.01; ***p<0.001. Error bars indicate mean ± SEM.

CpxII NTD exclusively modulates the kinetics of SytI mediated exocytosis.

(A) Mean [Ca]i levels (upper panel) and CM response from wt cells (black n=13), SytI ko cells (red n=19), and those overexpressing either CpxII wt (dark blue n=16), or the mutant CpxII ΔN (green n=20). In the absence of SytI, Cpx ΔN fails to slow down exocytosis timing as observed in SytVII cells (I). (B-D) Neither CpxII nor CpxII ΔN expression in SytI ko cells alters the size of the pools, the SR rate (B) or the kinetics of synchronous exocytosis (C, normalized CM; D, time constants of RRP and SRP exocytosis and secretory delay). (E, F) Both, CpxII and CpxII ΔN expression, hinders asynchronous release in the absence of SytI (F). (G) Mean [Ca]i levels (upper panel) and CM response of wt cells (black n=13), SytVII ko cells (orange n=17), and those expressing CpxII ΔN (Light blue n=29). (H) Both the RRP and SRP sizes are reduced in the absence of SytVII. (I) Normalized CM responses (of data shown in G) scaled to the wt response 1s after the flash. Note that CpxII ΔN mutant slows kinetics of release in SytVII ko cells. (J) Loss of SytVII speeds up exocytosis timing, which in turn is slowed down by additional expression of CpxII ΔN. (K, L) Analysis of premature exocytosis showing reduced premature exocytosis in SytVII ko cells compared to wt. CpxII ΔN suppresses asynchronous release also in the absence of SytVII. ANOVA or Kruskal-Wallis followed by corresponding post- hoc test. *p<0.05; **p<0.01; ***p<0.001. Error bars indicate mean ± SEM.

CpxII overexpression rescues the slow secretory rates of SytI R233Q ki cells. (A)

Mean [Ca]i levels (upper panel) and CM responses from wt (black n=35), SytI R233Q ki (violet n=39), and SytI R233Q ki cells expressing either CpxII wt (blue n=18), or the mutant CpxII ΔN (olive green n=11). (B) Compared to wt cells, SytI R233Q ki cells show an increased the RRP and SRP size, which are restored by CpxII or CpxII ΔN expression. (C) While CpxII speeds up exocytosis timing of SytI R233Q ki cells to the level of wt cells, CpxII ΔN slows it further down. Normalized CM responses (of data shown in A) scaled to the wt response 1s after the flash. (D) CpxII and CpxII ΔN expression oppositely regulate the kinetics of exocytosis (τRRP, τSRP) and the secretory delay. (E, F) Mean premature exocytosis, determined at similar submicromolar [Ca]i (in nM, SytI R233Q ki:623±31; SytI R233Q ki + CpxII:569±28; SytI R233Q ki + CpxII ΔN:600±26; wt: 582±22), showing that CpxII wt or CpxII ΔN expression effectively hinders premature vesicle secretion. ANOVA or Kruskal-Wallis followed by corresponding post-hoc test *p<0.05; **p<0.01; ***p<0.001. Error bars indicate mean ± SEM.

CpxII NTD increases the Ca2+ affinity of secretion. (A)

Exemplary capacitance measurement of a chromaffin cell (lower panel) in response to a ramp like increase in intracellular Ca2+ (top panel). (B) The slope of the ΔCM (top panel), remaining pool (middle panel) and rate of fusion (lower panel) were determined from the data shown under (A) with a temporal interval of 40 ms to match the timing of the ratiometric Ca2+- measurement. The rate was determined by dividing the slope [fF/s] by the remaining pool size [fF]. (C) Mean profile of pool depletion as the function of [Ca]i (left panel) for SytI R233Q ki cells (violet n=15) and those overexpressing CpxII wt (blue n=14) or CpxII ΔN (olive green n=11). Note that expression of CpxII wt lowers [Ca]i for half-maximal pool depletion (P50), whereas CpxII ΔN increases it (right panel). (D) Double-logarithmic plot of the mean exocytotic rate as the function of [Ca]i for the indicated groups. CpxII overexpression increases the rate of secretion from SytI R233Q ki cells, whereas loss of CpxII NTD further slows it down. Hill equations with similar coefficient, but different KD value approximate the data (dashed lines). Extrapolation of the rates at low [Ca]i coincides with the RRP kinetics (Figure 7D) determined in the flash experiment (hollow circle). ANOVA or Kruskal-Wallis followed by corresponding post-hoc test. *p<0.05; ***p<0.001. Error bars indicate mean ± SEM.

The CpxII CTD mutants CpxII L124E-L128E and CpxII L124W- L128W rescue the kinetics of synchronized exocytosis.

(A, B) The kinetics of synchronized exocytosis as well as the delay of the stimulus-secretion coupling of CpxII ko cells (red) are fully restored by the CpxII L124E-L128E (yellow) as well as the CpxII L124W-L128W mutant (green) when compared with expression of the CpxII wt protein (blue), consistent with the view that exocytosis timing is determined by the functionally intact NTD. ANOVA or Kruskal-Wallis followed by corresponding post-hoc test. *p<0.05; ***p<0.001. Error bars indicate mean ± SEM.

The hydrophobic face of CpxII’s amphipathic helix is essential to suppress premature release.

(A) Primary sequences showing point mutations L117W- L121W and L117E-L121E within the amphipathic α helix of CpxII CTD. (B) Mean [Ca]i levels (upper panel) and the corresponding average increase in the membrane capacitance from CpxII ko cells (red, n=26), and those expressing CpxII wt (blue, n=25), or the mutants CpxII L117E-L121E (yellow, n=25), or CpxII L117W-L121W (dark green, n=14). (C) Mean amplitudes of the RRP, SRP and SR show a significant reduction for the CpxII L117E-L121E mutant while CpxII L117W-L121W rescues the synchronized exocytosis like the expressed CpxII wt protein. (D) All mutant variants restore exocytosis timing of CpII ko cells like the CpxII wt protein. (E, F) Average tonic exocytosis at similar submicromolar [Ca]i levels. The CpxII L117E-L121E mutant failed to rescue premature secretion phenotype of CpxII ko cells, but CpxII L117W-L121W mutant hindered it like the CpxII wt protein. ANOVA or Kruskal-Wallis followed by corresponding post-hoc test. ***p<0.001. Error bars indicate mean ± SEM.

Alanine substitutions on the polar face of CpxII CTD have no impact on its clamping ability.

(A)Primary sequences of SNAP-25 SN1 and CpxII CTD showing the positions of the alanine substitutions (subsequent panels refer to the same color coding). (B) Mean [Ca]i levels (upper panel) and the corresponding membrane capacitance increase from CpxII ko cells (red n=42), and those expressing either CpxII wt (blue n=58), or the mutants CpxII D118A (yellow, n=17) CpxII Q129A (pink, n=19), CpxII D130A (violet, n=15), CpxII K133A (green, n=17). (C) When compared with expression of the CpxII wt protein, all CpxII mutants fully support an undiminished exocytotic burst, without changing either the RRP, SRP or the SR. (D) Kinetic parameters of the synchronized exocytosis are rescued by CpxII and its mutant variants. (E,F) CpxII mutants suppress premature vesicle exocytosis at submicromolar calcium concentrations with similar efficacy as the CpxII wt protein. ANOVA or Kruskal-Wallis followed by corresponding post-hoc test.. **p<0.01; ***p<0.001. Error bars indicate mean ± SEM.

The charge reversal mutation D118K-D130K does not affect the inhibitory function of the CpxII CTD.

(A) Primary sequence of CpxII CTD showing the double point mutation D118K-D130K. (B) Mean [Ca]i levels (upper panel) and the corresponding average increase in CM from CpxII ko cells (red n=27), and those expressing either CpxII wt (blue n=29) or the mutant CpxII DD-KK (green n=14). (C) The D118K-D130K mutant fully rescued all component of the exocytotic burst and the SR, like the expression of the CpxII wt protein. (D-F) Kinetic parameters of the synchronized exocytosis (D) and the suppression of premature vesicle exocytosis (E, F) were unaffected by the CpxII mutant. ANOVA or Kruskal-Wallis followed by corresponding post-hoc test. *p<0.05; ***p<0.001. Error bars indicate mean ± SEM.

Expression analysis of CpxII and its mutants reveals similar levels of protein expression.

(A) Exemplary images of CpxII ko, wt,or CpxII ko cells expressing CpxII wt or its mutant variants. Exposure times were adjusted to avoid the saturation of the sensor (CpxII ko 2s, wt 300ms CpxII ko over expressing CpxII or the CpxII mutants 30 ms). (B) Mean total fluorescence intensity (corrected for the exposure time) shows that CpxII wt and its mutants are expressed to similar levels in CpxII ko cells. Number of cells for each group is indicated in the corresponding bar. Kruskal-Wallis followed by Dunn’s post-hoc test. ***p<0.001. Error bars indicate mean ± SEM.

CpxII and its mutants concentrate at vesicular membranes.

(A) Exemplary immunostaining images for cells stained against CpxII or its mutant variants (green), and the vesicular marker SybII (red). The corresponding merged images (yellow) illustrate the overlap of CpxII and SybII signals (yellow). (E) Exemplary cytofluorogram for wt (boxed area) showing a strong correlation between SybII and CpxII stains (Pearson’s Coeff. = 0.85). (C) Pearson’s co-localization coefficient (threshold 6xSD of the background fluorescence) for CpxII or its mutants versus SybII shows that virally expressed CpxII variants concentrate at vesicles like the endogenous protein. Error bars indicate mean ± SEM. Kruskal-Wallis followed by Dunn’s post-hoc test. *p<0.05.

CpxII clamps premature exocytosis independently of SytI.

(A) Mean time course of premature exocytosis from CpxII ko cells (red n=35), CpxII/SytI dko (violet n=26), SytI ko (light green n=46), wt (black n=49), SytIko+CpxII (dark green n=22), CpxII ko+CpxII (blue n=15). (B, C) Quantification of the premature exocytosis rate per 10 seconds at similar calcium levels shown in (C) for indicated groups. (D, E) The EB size (RRP+SRP component, E) correlates inversely with the extent of the preceding premature exocytosis upon changes of CpxII expression in the presence and in the absence of SytI (solid lines linear regressions, R2=0.99). ANOVA or Kruskal-Wallis followed by corresponding post-hoc test. **p<0.01; ***p<0.001. Error bars indicate mean ± SEM.

CpxII boosts exocytosis in the absence of SytVII.

(A) Mean [Ca]i levels (upper panel) and CM responses from wt cells (black n=17), SytVII ko cells (orange n=18), and SytVIIko cells expressing CpxII (green n=18). (B) SytVII deletion reduces the RRP and SRP size without affecting the sustained rate of secretion. (C) Kinetics of RRP and SRP exocytosis and the secretory delay. (D) Time course of premature exocytosis at submicromolar [Ca]i. (E, F) Loss of SytVII reduces premature vesicle fusion (E), which is further diminished upon CpxII expression at nearly comparable [Ca]i (F). ANOVA or Kruskal-Wallis followed by corresponding post-hoc test. *p<0.05; **p<0.01; ***p<0.001. Error bars indicate mean ± SEM.