Single molecule detection of intermediates during botulinum neurotoxin translocation across membranes

Abstract

The dynamics of Clostridium botulinum neurotoxins (BoNTs) protein-translocation across membranes was investigated by using a single molecule assay with millisecond resolution on excised patches of neuronal cells. Translocation of BoNT/A light chain (LC) by heavy chain (HC) was observed in real time as an increase of channel conductance: the HC channel is occluded by the LC during transit, then unoccluded after completion of translocation and release of LC-cargo. We identified an entirely unknown succession of intermediate conductance stages during LC translocation. For the single-chain BoNT/E, by contrast to the di-chain BoNT/A, we demonstrate that productive translocation requires proteolysis of the LC cargo from the HC chaperone. We propose a model for the set of protein-protein interactions between translocase and cargo at each step of translocation that supports the notion of an interdependent, tight interplay between the HC chaperone and the LC cargo preventing LC aggregation and dictating the outcome of translocation: productive passage of cargo or abortive channel occlusion by cargo.

Fig. 1.

BoNT/A holotoxin and HC channels in excised patches of Neuro 2A cells. Single-channel currents for holotoxin (A) and HC (B) at −100 mV and the indicated times. Holotoxin (A) and HC (B) channel activity begins 20 min and 21 min after GΩ seal formation. The letters C and O denote the closed and open states. (C) Time course of γ change of holotoxin (black circle) and HC (blue circle) for the representative experiments (average N per data point = 570). γ values associated with raw data (colored numbers) from A and B are indicated. (D) Structure of BoNT/A holotoxin (10). Shown are LC (purple), translocation domain (orange), and receptor binding domain (red) before insertion in the membrane (gray bar with magenta boundaries); then is shown a schematic of the membrane inserted BoNT/A during translocation of the LC (purple) through the HC channel (orange) with intact disulfide bridge (green) while located in the cis-compartment. Conventions hold for all figures.

Fig. 2.

Analysis of BoNT/A holotoxin and HC channels in Neuro 2A cells. (A) Average time course of channel γ change for holotoxin (n = 6, average N per data point = 4,650). (B) γ amplitude histogram (gray) and Gaussian fit calculated from the combined data set of six separate BoNT/A holotoxin experiments (red) and one representative BoNT/A HC experiment (blue). BoNT/A HC has a γ = 66.4 ± 2.4 pS (N = 10,200); holotoxin has γ values = 12.6 ± 2.0 pS, 24.1 ± 3.4 pS, 46.8 ± 1.6 pS, 55.3 ± 3.4 pS (N = 55,890). The endpoint γ = 67.1 ± 2.0 pS is equivalent to that of HC. (C) Average occupancy time of conductance states of BoNT/A holotoxin (n = 6, average N per data point = 4,650). (D) P_o as a function of voltage for holotoxin (black circle) (V_½ = −58.3 ± 1.7 mV) and for HC (blue circle) (V_½ = −63.4 ± 2.4 mV) (3 ≤ n ≤ 11 per data point; average N per data point = 2,630).

Fig. 3.

LC/A translocation arrest by a LC/A specific Fab. Shown are single-channel currents for HC (A) and holotoxin (B) at −100 mV and the indicated times. HC (A) and holotoxin (B) channel activity begins 5 min after GΩ seal formation. (B) Trace-① displays the pattern of activity at the beginning of the record. Multiple channel insertions occur with time (traces② and ③). (C) Time course of holotoxin channel γ change (average N per data point = 600). Thin red line represents results for unmodified holotoxin/A. (D) γ amplitude histogram (gray) and Gaussian fit (black) with γ = 14.8 ± 4.1 pS and 24.0 ± 1.6 pS (N = 22,113). HC indicates the unoccluded HC conductance. (E) Average occupancy time of conductance states (n = 5, average N per data point = 5,170). (F) Schematic representation of BoNT/A holotoxin translocation arrested by Fab (red).

Fig. 4.

The single-chain BoNT/E holotoxin requires proteolytic cleavage to complete translocation of LC. Shown are representative single-channel currents at the indicated voltages and times; consecutive voltage pulses applied to the two different patches. (A) Holotoxin channel activity begins 5 min (A) and 25 min (B) after GΩ seal formation. (B) Trypsin addition (4 mM) to the trans-side occurs at 300 s. (C) Time course of channel γ change of trypsin-nicked BoNT/E before exposure to the cells (red circle), single-chain holotoxin/E (pink circle), and single-chain holotoxin/E after trypsin addition to the trans-side (black circle) (green arrow) for representative experiments (average N per data point = 275). (D) γ amplitude histogram and Gaussian fit of single-chain holotoxin/E, γ = 7.7 ± 1.8 pS and 12.7 ± 1.0 pS; Gaussian fit illustrated in pink, data not shown (N = 10,840). Holotoxin/E, after trypsin addition to the trans-side, has γ = 9.3 ± 1.3 pS, 13.3 ± 2.0 pS, 28.4 ± 1.6 pS, 42.4 ± 1.3 pS, 49.2 ± 0.5 pS, 55.0 ± 0.5 pS, and 65.4 ± 2.8 pS; shown are raw data (gray) and Gaussian fit (black) (N = 19,431). (E) Average occupancy time of conductance states for trypsin-nicked holotoxin/E (black) and single-chain holotoxin/E (pink) (n = 4 for each experimental condition, average N per data point: single-chain holotoxin/E = 460, trans-trypsin nicked holotoxin/E = 1,470). (F) P_o as a function of voltage for holotoxin/E (black circle) and holotoxin/A (red circle); V_½ for holotoxin/E is −45.0 ± 5.4 mV (3 ≤ n ≤ 11 per data point; average N per data point = 2,275). (G) Schematic of single-chain holotoxin BoNT/E occluded channels, the subsequent cleavage of the LC from the HC by trypsin, and the consequent LC release into the neutral pH compartment.

Fig. 5.

Sequence of events underlying BoNT LC translocation through the HC channel. Step 1, Crystal structure of BoNT/A holotoxin (10) before insertion in the membrane. Then is shown a schematic representation of the membrane inserted BoNT/A during an entry event (step 2), a series of transfer steps (steps 3 and 4), and an exit event (step 5), under conditions that recapitulate those across endosomes.