Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 Mar 15.
Published in final edited form as: Cell Host Microbe. 2011 Sep 15;10(3):237–247. doi: 10.1016/j.chom.2011.06.012

Receptor Binding Enables Botulinum Neurotoxin B to Sense Low pH for Translocation Channel Assembly

Shihu Sun 1,2,5, Swetha Suresh 3,5, Huisheng Liu 1,2, William H Tepp 4, Eric A Johnson 4, J Michael Edwardson 3, Edwin R Chapman 1,2,*
PMCID: PMC3243646  NIHMSID: NIHMS327875  PMID: 21925111

SUMMARY

Botulinum neurotoxins (BoNTs, serotypes A-G), elaborated by Clostridium botulinum, can induce lethal paralysis and are classified as category-A bioterrorism agents. However, how BoNTs translocate from endosomes into the cytosol of neurons to gain access to their intracellular targets remains enigmatic. We discovered that binding to the ganglioside GT1b, a toxin co-receptor, enables BoNT/B to sense low pH, undergo a significant change in secondary structure, and transform into a hydrophobic oligomeric membrane protein. Imaging of the toxin on lipid bilayers using atomic force microscopy revealed donut-shaped channel-like structures that resemble other protein translocation assemblies. Toosendanin, a drug with therapeutic effects against botulism, inhibited GT1b-dependent BoNT/B oligomerization, and in parallel truncated BoNT/B single-channel conductance, suggesting that oligomerization plays a role in the translocation reaction. Thus, BoNT/B functions as a coincidence detector for receptor and low pH to ensure spatial and temporal accuracy for toxin conversion into a translocation channel.

INTRODUCTION

Botulinum neurotoxins (BoNTs) are the most deadly toxins known, and cause the disease botulism (Schiavo et al., 2000). This family of toxins - of which there are seven serotypes (BoNT/A-G) - are potential biological weapons, and are categorized as category-A bioterrorism agents by the Centers for Disease Control (CDC) (Arnon et al., 2001; Rotz et al., 2002). Despite their extreme potency and lethality, BoNTs possess great therapeutic potential. For instance, two serotypes, BoNT/A and B, are widely used in a broad range clinical applications (Montecucco and Molgó, 2005), including the treatment of muscle dysfunction and neurological pain.

Each BoNT consists of an N-terminal light chain (LC, 50 kDa) and a C-terminal heavy chain (HC, 100 kDa) (Schiavo et al., 2000). The LC acts as a zinc-dependent protease that cleaves SNARE (soluble NSF attachment protein receptor) proteins in the cytosol of pre-synaptic nerve terminals, thereby blocking neurotransmitter release. Loss of neurotransmission can result in paralysis and death. The HC is connected to the LC via a disulfide bond; the HC also wraps around the LC via a single stranded “belt” loop (Montal, 2010). The HC is composed of an N-terminal translocation domain and a C-terminal receptor binding domain. The receptor binding domain interacts with specific receptors on the surface of presynaptic nerve terminals, and the toxin is endocytosed, usually via recycling synaptic vesicles (Dong et al., 2008; Schiavo et al., 2000). In order to access SNARE proteins, the LC must be translocated from the lumen of endosomes/synaptic vesicles into the cytosol. This process, translocation, is the least understood step in the action of BoNT on target cells (Montal, 2010; Schiavo et al., 2000).

Low pH in the lumen of endosomes/synaptic vesicles serves as the trigger for BoNT LC translocation. Presumably, low pH converts the translocation domain within the BoNT HC into a membrane channel that mediates LC translocation (Schiavo et al., 2000). Evidence for this model includes the well-established single-channel activity of BoNTs in planar bilayers and in membranes excised from PC12 or N2A cells (Hoch et al., 1985; Sheridan, 1998; Montal, 2010). Addition of BoNTs to the low-pH, oxidizing side of a membrane results in the formation of single-channel conductances that are initially small but grow over time, reaching a plateau. It was suggested that the early low conductance state represents a HC channel that is partially occluded by the nascent, translocating LC, and that the large conductance state at the plateau represents an un-occluded channel after the LC has left (Montal, 2010). Despite these advances, the mechanisms underlying LC translocation are still poorly understood. For instance, the mechanism by which BoNT transforms from a soluble to an integral membrane protein is unclear, and the molecular architecture of the toxin translocation channel is completely unknown.

The goal of the current study was to use a combination of cell biological, biochemical, structural and electrophysiological approaches to probe the assembly of the BoNT translocation channel. We focused on BoNT/B, a clinically-used serotype (Montecucco and Molgó, 2005), for which the crystal structure has been determined (Swaminathan and Eswaramoorthy, 2000), and for which interactions with co-receptors, the ganglioside GT1b and synaptotagmin (syt) I/II, have been well characterized (Chai et al., 2006; Dong et al., 2003; Jin et al., 2006; Nishiki et al., 1996; Rummel et al., 2007). The data reported here indicate that BoNT/B must interact with its neuronal co-receptor, GT1b, in order to sense the low pH in endosomes/recycling synaptic vesicles and transform into a membrane associated oligomeric channel-like structure. The ability of BoNT/B to act as a coincidence detector for receptor and low pH ensures that the toxin converts into a translocation machine at the right time and place.

RESULTS

Low-pH Pretreatment Does Not Inactivate BoNT/B

Low pH in the lumen of endosomes/synaptic vesicles triggers the transformation of BoNTs from soluble proteins into translocation-competent membrane channels (Montal, 2010; Schiavo et al., 2000). For some bacterial toxins that also utilize low endosomal pH to translocate, such as anthrax toxin, exposure to low extracellular pH prior to binding to cellular receptors results in conformational changes - that would normally underlie translocation - in solution, leading to aggregation such that the toxin becomes inactive (i.e. enters an “off-pathway”, Figure 1A) (Sun et al., 2007; Young and Collier, 2007). To test whether BoNT/B responds to low extracellular pH and enters an off-pathway, we preincubated BoNT/B for 2 h in high-K+ buffer at either pH 7.4 or pH 4.4, before treating rat hippocampal neurons with the toxin in the same buffers. High-K+ buffer stimulates synaptic vesicle recycling, thus facilitating uptake of BoNT/B into neurons (Dong et al., 2007; Schiavo et al., 2000). BoNT/B that was not taken up by neurons was washed away, and translocation was monitored by assaying cleavage of synaptobrevin II (syb), the intracellular target of BoNT/B (Schiavo et al., 2000). We found that pH 4.4-pretreatment had no effect (Figures 1B and 1C), and thus does not drive BoNT/B into an off-pathway. Either the isolated toxin is unable to sense the drop in pH, or pH-induced conformational changes - unlike the case of anthrax toxin - failed to inactivate the toxin.

Figure 1. Low pH Pretreatment Does Not Inactivate BoNT/B.

Figure 1

Open in a new tab

(A) Illustration of the “off-pathway” potentially induced by low-pH pretreatment of BoNT/B. In principle, exposure to low pH in solution could transform BoNT/B into a hydrophobic protein, causing aggregation and consequent inactivation.

(B) Immunoblot analysis of syb cleavage by BoNT/B. Toxin was exposed to either pH 7.4 or pH 4.4 for 2 h before incubation with rat hippocampal neurons. Syntaxin served as a loading control.

(C) Quantitation of syb cleavage, normalized to syntaxin levels. Data are means ± SEM (n=3).

Gangliosides are Required for Efficient Cell Surface Entry of BoNT/B

The ideal sites to study BoNT/B translocation are endosomes/synaptic vesicles, where translocation normally occurs. However, it is difficult to measure and manipulate pH inside these organelles. Therefore, in order to study the effect of pH on the behavior of BoNT/B, we devised a method to reconstitute BoNT/B translocation across the plasma membrane of cultured rat hippocampal neurons. To this end, normal translocation of BoNT/B from the vesicle lumen into the cytosol was blocked by addition of bafilomycin (Figure 2A), a specific inhibitor of vacuolar H+-ATPase (Keller et al., 2004; Simpson et al., 1994). Neurons were then treated with BoNT/B in extracellular buffers of varying pH, and LC translocation into the cytoplasm was determined, again, by monitoring cleavage of syb (Figure 2B). After exposure of neurons to BoNT/B at an extracellular pH of 7.4 for 8 min, the toxin efficiently cleaved syb in the absence, but not the presence, of bafilomycin during a subsequent 24-h incubation. As extracellular pH was lowered, cleavage in the presence of bafilomycin increased progressively; at pH 4.4, which was the lowest pH value possible without compromising cell viability, cleavage of syb reached ~50% (Figures 2B and 2C). Hence, BoNT/B LC can be translocated across the plasma membrane at low pH.

Figure 2. Gangliosides are Required for Efficient Cell Surface Entry of BoNT/B.

Figure 2

Open in a new tab

(A) Illustration of potential BoNT/B LC translocation pathways. Normal translocation of BoNT/B LC across the membrane of endocytosed synaptic vesicles is blocked by bafilomycin A1 (Baf), an inhibitor of the vesicular H+-ATPase. After BoNT/B holotoxin binds to its co-receptors, syt and gangliosides, on the neuronal plasma membrane, low extracellular pH treatment enables BoNT/B to deliver its LC directly into the neuronal cytoplasm. Translocation is monitored by assaying cleavage of syb.

(B and C) pH dependence of BoNT/B LC translocation across membranes. (B) Immunoblot analysis of syb cleavage after exposure of rat hippocampal neurons to BoNT/B for 20 h. Neurons were treated with BoNT/B with and without Baf, as indicated. SNAP-25 served as an internal loading control. The black line indicates lanes, from the same blot, that were juxtaposed. (C) Quantitation of syb cleavage in (B), normalized to SNAP-25 levels. Data are means ± SEM (n=3).

(D–G) Gangliosides are required for BoNT/B LC translocation across membranes. Hippocampal neurons from ganglioside KO (Gan−/−) mice (D) or FB1-treated (+FB1) rat hippocampal neurons (F) were exposed to BoNT/B with and without Baf at the indicated pH for 10 min. Cleavage of syb after 20 h was analyzed by immunoblotting. Cleavage of syb in wild-type (WT) mouse neurons (D) or in rat neurons without FB1 treatment (−FB; F) was also monitored. Syntaxin served as a loading control. Cleavage of syb was quantified after normalization to the levels of syntaxin (E and G). Data are means ± SEM (n=3).

We note that the time-averaged pH value reported for synaptic vesicles is 5.6 (Miesenböck et al., 1998). However, given the minute volume of these organelles (diameter of ~42 nm) (Heuser and Reese, 1981), the instantaneous intra-luminal pH will be significantly altered by the flux of even a single proton, and pH values of 4.4 - or lower - seem likely to occur, at least for brief periods.

BoNT/B normally enters neurons via receptor-dependent endocytosis (Schiavo et al., 2000), but whether interactions with receptor molecules play a role in the translocation reaction is a question that has not been addressed. To begin to study this problem, we monitored translocation in neurons lacking gangliosides, the low-affinity co-receptors for BoNT/B (Schiavo et al., 2000). For this purpose, we used both ganglioside knockout (KO) mouse hippocampal neurons (Figure 2D, quantified in Figure 2E) and rat hippocampal neurons treated with fumonisin B1 (FB1) (Figure 2F, quantified in Figure 2G). FB1 effectively inhibits synthesis of gangliosides in cultured rat neurons without impairing axonal growth; the ganglioside GT1b is decreased to ~26% of control level in the presence of 25 μM FB1 (de Chaves et al., 1997). In both ganglioside KO and FB1-treated neurons, BoNT/B - in low pH extracellular buffer - failed to translocate across the plasma membrane to cleave syb (Figures 2D–2G). Hence, gangliosides are required for efficient surface translocation of BoNT/B. We note that in a previous study, it was shown that loading ganglioside KO neurons with exogenous gangliosides rescued the entry of BoNT/B (Dong et al., 2007).

Ganglioside GT1b Enables BoNT/B to Change Conformation and Transform into a Hydrophobic Membrane Protein at Low pH

One fundamental question regarding BoNT/B translocation concerns the mechanism by which BoNT/B transforms from a soluble protein into a membrane channel in response to low pH. The cell surface translocation/entry experiments described above indicate that gangliosides facilitate efficient translocation of BoNT/B across membranes. We therefore tested whether binding to the ganglioside GT1b, the lipid receptor of BoNT/B, triggers conformational changes in the toxin that allow it form a membrane channel at low pH.

Channel formation by pore-forming toxins, such as anthrax toxin, is associated with secondary structural rearrangements at low pH, which can be probed by circular dichroism (CD) spectroscopy (Kintzer et al., 2010). At pH 7.4, the CD spectra of BoNT/B with or without GT1b largely overlapped from 197 nm to 260 nm (Figure 3A, left panel), and the helical content estimated from molar ellipticity at 222 nm (Chen et al., 1972) was ~20% both for BoNT/B alone, and for BoNT/B plus GT1b. BoNT/B alone yielded similar spectra at pH 7.4 and pH 4.4, with a helical content of ~23% at pH 4.4 (Figure 3A, right panel). These findings are in agreement with crystallography studies indicating that exposure to low pH alone does not cause a structural change in BoNT/B (Eswaramoorthy et al., 2004), and are also consistent with the lack of effect of low pH in our off-pathway assays. However, addition of GT1b at pH 4.4 resulted in a significant shift in the spectrum, with a large reduction in signal at ~208 nm, and a reduction in helical content to only ~12%. These results indicate that BoNT/B undergoes a major structural rearrangement in the presence of GT1b at pH 4.4, a pH value that enables efficient translocation.

Figure 3. GT1b Enables BoNT/B to Undergo a Change in Secondary Structure and to Transform into a Hydrophobic Membrane-associated Protein at Low pH.

Figure 3

Open in a new tab

(A) Low pH triggers major conformational changes in BoNT/B in the presence of GT1b. Far UV-CD spectra of BoNT/B in the absence (dashed lines) or presence (solid lines) of GT1b at pH 7.4 (left) and pH 4.4 (right). CD spectra of GT1b alone at pH 7.4 and pH 4.4 were included as controls (dotted lines).

(B) Cosedimentation of BoNT/B with liposomes. BoNT/B was incubated with liposomes that either harbored or lacked GT1b at pH 7.4 or pH 4.4. Liposomes were sedimented, and BoNT/B in the total input (T), supernatant (S) and pellet (P) fractions was assayed by immunoblotting using anti-BoNT/B antibodies.

(C) Nature of the association of BoNT/B with liposomes. BoNT/B efficiently bound to GT1b-containing liposomes at pH 4.4. The liposome-protein complexes were suspended in the indicated buffers, incubated for 20 min at room temperature and then resedimented. BoNT/B in the total input (T) and initial supernatant (1S), and in the second supernatant (2S) and pellet (2P) fractions was again assayed by immunoblotting.

(D) CD spectra showing that the GT1b-induced conformational change in BoNT/B at pH 4.4 was largely reversed at pH 7.4. Representative CD spectra of BoNT/B alone at pH 4.4 (long dashed line), BoNT/B with GT1b at pH 4.4 (solid line), and BoNT/B with GT1b after raising the pH from 4.4 to 7.4 (short dashed line), are shown. The spectrum of GT1b alone (dotted line) serves as a control.

(E) Increased hydrophobicity of BoNT/B at low pH in the presence of GT1b. Triton X-114 partitioning assays were performed at pH 7.4, pH 5.4 and pH 4.4 in the absence or presence of GT1b. BoNT/B in the total input (T), aqueous (A) and detergent (D) phases was detected by immunoblot analysis. The behavior of full-length syb (an archetypal membrane protein) and syb cytosolic domain (cd-syb, an archetypal soluble protein) were analyzed by immunoblotting or staining with Coomassie blue, respectively. See also Figure S1.

To further test whether the GT1b-dependent structural rearrangement of BoNT/B at low pH enables it to transform into a hydrophobic protein, we examined the interaction of BoNT/B with membranes. We found that the toxin did not cosediment with GT1b-free liposomes at either pH 7.4 or pH 4.4 (Figure 3B), consistent with the notion that low pH alone is insufficient to convert BoNT/B into a hydrophobic species. When the liposomes contained GT1b, only low levels of binding were observed at pH 7.4; however, at pH 4.4 efficient cosedimentation of BoNT/B with GT1b-containing liposomes occurred.

To characterize the interactions that mediate cosedimentation at pH 4.4, we collected the BoNT/B-bound liposomes and performed extraction experiments with either 1M NaCl at pH 4.4 or Na2CO3 at pH 11.0 (Figure 3C). Typically, only peripheral proteins, and not transmembrane or lipid-anchored proteins, are extracted by these reagents (Bai et al., 2000). Interestingly, 1M NaCl failed to extract BoNT/B from liposomes, but Na2CO3 caused complete extraction. These results suggest that GT1b enables BoNT/B to sense low pH and tightly associate with membranes, and further, that the interaction can be reversed by exposure to higher pH. Consistent with this idea, incubation of the liposomes in a buffer (phosphate-buffered saline; PBS) at pH 7.4 caused partial extraction of the BoNT/B (Figure 3C). As expected, BoNT/B was completely extracted by treatment of the liposomes with Triton X-100. We further tested whether the observed partial reversibility in the liposome co-sedimentation assay was related to conformational changes in BoNT/B via CD, and found that ~76% of the measured-conformational change triggered by GT1b and pH 4.4 was reversed at pH 7.4 (Figure 3D).

To further test whether BoNT/B is converted into a hydrophobic species in the presence of GT1b at low pH, we performed Triton X-114 partitioning assays. Triton X-114 is a detergent that is homogeneous at 0°C in solution, but partitions into aqueous and detergent phases above 20°C. This behavior provides a simple way to distinguish integral or lipid-anchored proteins (which partition into the detergent phase), from soluble or peripheral proteins (which partition into the aqueous phase) (Bordier, 1981). We found that in the presence of GT1b, BoNT/B partitioned into the aqueous phase at pH 7.4, but was almost equally divided between the two phases at pH 5.4, and was found exclusively in the detergent phase at pH 4.4 (Figure 3E). In contrast, the toxin remained in the aqueous phase at all pH values when GT1b was absent. As controls, a known membrane protein, syb, remained in the detergent phase, whereas the cytoplasmic domain of syb (cd-syb) remained predominantly in the aqueous phase. The partitioning of BoNT/B into the detergent phase at low pH in the presence of GT1b was not due to more avid binding of the toxin to GT1b at low pH, as the receptor binding domain of BoNT/B (HCR/B) (tested at the same concentration as BoNT/B in Triton X-114 partitioning assay) exhibited similar GT1b binding activity at pH 7.4 and 4.4 (Figure S1) in a solid-phase binding assay (Schmitt et al., 2010). In addition, the interaction of another BoNT serotype with gangliosides is known to be less, rather than more, avid at low pH than at neutral pH (Kamata et al., 1988). We also note that BoNT/B was stable under all conditions tested; the partitioning behavior of BoNT/B at low pH in the presence of GT1b was not due to degradation or aggregation of the toxin. Together, the experiments reported in this section reveal that BoNT/B is converted into hydrophobic species in the presence of GT1b at low pH.

BoNT/B Assembles into Oligomeric Channel-like Structures in GT1b-containing Bilayers at Low pH

To visualize the structure adopted by BoNT/B on GT1b-containing bilayers at low pH, we turned to atomic force microscopy (AFM). This technique offers single-molecule resolution under near-physiological conditions (i.e. under fluid) (Saslowsky et al., 2002; Suresh and Edwardson, 2010). We first imaged BoNT/B holotoxin adsorbed onto mica in air. The toxin appeared as homogenous particles, which consisted of adjoined large and small components, likely representing the HC and the LC (Figures 4A and 4C). We measured the dimensions of a number of particles and calculated their molecular volumes, using Equation 1, as detailed in Experimental Procedures. The frequency distribution of molecular volumes had a peak around 250 nm3 (Figure 4B), consistent with the predicted molecular volume for a 150-kDa protein (around 285 nm3, according to Equation 2).

Figure 4. BoNT/B HC Forms Donut-shaped Structures on GT1b-containing Bilayers at Low pH.

Figure 4

Open in a new tab

(A) AFM image of BoNT/B holotoxin on mica, with sections taken at the positions indicated by the lines. Scale bar, 20 nm.

(B) Frequency distribution of molecular volumes of BoNT/B holotoxin on mica (right).

(C) Representative image of BoNT/B holotoxin on mica, showing the two-lobed structure of the LC and the HC. Scale bar, 20 nm.

(D) AFM images of non-reduced (left) and DTT-reduced (right) BoNT/B on GT1b-containing bilayers at pH 5.0. Scale bar, 15 nm.

(E) Diagram indicating the dimensions of single donuts that were measured.

(F) Dimensions of donuts. The left-hand panel shows an AFM image of an individual single donut. The right-hand panel shows a section taken at the position indicated by the lines. The height and hole depth of the donut are indicated.

(G–J) Frequency distributions of heights h (G), hole depths d (H), outer radii R (I), and hole radii r (J) for single donuts.

See also Figure S2.

We next examined the interaction of BoNT/B with lipid bilayers via AFM imaging under fluid. Both at pH 7.0 and pH 5.0, the mean molecular volume of the particles was around 300 nm3, very close to the value for holotoxin. Although the measured molecular volume suggests that the toxin bound to the bilayer predominantly as monomers, it is unwise to rely too heavily on volume measurements to deduce stoichiometry. For instance, it is well known that the geometry of the scanning AFM probe introduces a tendency to overestimate particle radii, a convolution that becomes especially significant when imaging under fluid, because fluid-imaging tips are blunter than air-imaging tips. In addition, volume measurements of particles bound to lipid bilayers do not take into account penetration of the bilayer by the protein, which is likely to be an issue here (see below). At pH 5.0, BoNT/B holotoxin appeared as a dome-shaped structure on GT1b-containing bilayers (Figure 4D, left panel). When the LC was released by reduction of the toxin with DTT, the appearance of some of the BoNT/B particles changed from dome-shaped to donut-shaped, channel-like structures (Figure 4D, right panel). The presence of GT1b in the bilayers clearly promoted the formation of these donut-like structures. Specifically, 33% (218 of 650) of bound particles were donuts in the presence of GT1b, compared with 19% (42 of 225) without GT1b.

Sections through single donuts were used to construct frequency distributions for various dimensions (Figures 4E and 4F). The peak height above the bilayer was 0.3 nm (Figure 4G); the peak depth of the hole was 0.2 nm (Figure 4H); the peak radius of the donut was 17 nm (Figure 4I); and the peak radius of the hole (at the highest point of the donut) was 7 nm (Figure 4J). We note that the radius of the donut is very similar to the radius of the holotoxin, indicating that donut formation does not require oligomerization of the toxin. The rather small protrusion of the toxin molecule from the bilayer surface (0.3 nm) suggests that the toxin is embedded in the bilayer. Consistent with this finding, increasing the pH from pH 5.0 to pH 7.0 resulted in the dissociation of some of the toxin particles, leaving indentations where toxin molecules were previously bound (Figure S2). This result provides direct support for the idea that BoNT/B inserts, at least partially, into membranes at low pH.

Some of the donuts self-associated to form dimers and trimers (Figure 5A). Of the donuts seen in the presence of GT1b, 60±1% were monomers, 17±1% were dimers, and 8±1% were trimers, with the remainder being higher-order assemblies of donuts (data from three independent experiments; total donut number = 218). In addition to increasing the percentage of particles that appeared as donuts, GT1b also supported oligo-donut formation. Specifically, in the absence of GT1b, the percentage of donut structures that were oligomers fell from 40% to 22%. Significantly, the septa between associating donuts had only a single and not a double width, suggesting that self-association resulted in a reorganization of toxin to generate unique structures. Sections through typical double and triple donuts are shown in Figure 5B. We found that the double-donut structures have a long axis of 65 nm and a short axis of 52 nm (n = 35); the long axis was less than twice the length of the short axis, consistent with the presence of a single septum between the two adjoined donuts. In addition, the mean radius of the holes in the double donuts was 7 nm, identical to the value for single donuts, indicating that self-association did not generate larger holes.

Figure 5. GT1b Triggers BoNT/B Oligomerization at Low pH.

Figure 5

Open in a new tab

(A) Gallery of AFM images showing double (top) and triple (bottom) donut structures formed by reduced BoNT/B on GT1b-containing bilayers at pH 5.0. Scale bar, 15 nm.

(B) Dimensions of donuts. The left-hand panels show AFM images of individual double (top) and triple (bottom) donuts. The right-hand panels show sections taken at positions indicated by the lines. Heights and hole depths of the donuts are indicated.

(C) BN-PAGE analysis of BoNT/B holotoxin at pH 7.4, pH 5.0 and pH 4.4 in the absence and presence of GT1b. Protein was detected by silver staining.

(D) BN-PAGE analysis of BoNT/B holotoxin at pH 4.4 in the absence and presence of the control ganglioside, GM1.

We were unable to generate stable lipid bilayers for AFM imaging at pH 4.4, a pH value used in the other experiments to drive more efficient translocation (Figures 2B and 2C). As an alternative method for investigating the effect of GT1b on BoNT/B oligomerization, we incubated BoNT/B with the ganglioside (below its critical micelle concentration) at various pH values, and performed blue native-PAGE (BN-PAGE; Figures 5C and 5D). We found that in the absence of GT1b, BoNT/B behaved as a monomer (molecular mass ~150 kDa) between pH 7.4 and pH 4.4 (Figure 5C). In the presence of GT1b, BoNT/B was predominantly monomeric at pH 7.4; however, when the pH was reduced to 5.0, BoNT/B behaved as a mixture of monomers and oligomers. These results are consistent with our observation, by AFM, that on GT1b-containing bilayers at pH 5.0, BoNT/B HC exists as single and multiple donut-shaped structures (Figures 4D and 5A). Interestingly, at pH 4.4, BoNT/B trimers became dominant (Figure 5C). In contrast to GT1b, a control ganglioside - GM1 - at the same concentration, did not induce oligomerization of BoNT/B at low pH (Figure 5D). These results indicate that BoNT/B holotoxin exists mainly as an oligomer, most likely a trimer, in the presence of GT1b at low pH.

Toosendanin (TSN) Hinders BoNT/B Oligomerization and Restricts LC Translocation

The drug TSN has been shown to inhibit translocation of the BoNT/A LC (Fischer et al., 2009; Li and Shi, 2006). Given the data described above suggesting that BoNT/B might assemble into multimers - and trimers in particular - at low pH, we tested whether TSN affects trimerization of BoNT/B. We first confirmed that TSN blocked BoNT/B intoxication of neurons by incubating the drug with holotoxin (20 nM) for 2 h prior to addition of the mixture to rat hippocampal neurons. We found that BoNT/B alone completely cleaved syb, and that cleavage was reduced ~50%, and never reached ~100%, in the presence of ≥ 2 μM TSN (Figure 6A). We then used BN-PAGE to determine whether trimerization of BoNT/B holotoxin was also disrupted by TSN. TSN prevented GT1b-promoted oligomerization of BoNT/B at pH 5.0 but not at pH 4.4 (Figures 6B and 6C); TSN also lowered the pH requirements for BoNT/B oligomerization in solution from ~pH 5.0 to ~pH 4.6 (Figure S3A). Inhibition of oligomerization (at pH 5.0) occurred only when TSN was added before, or concurrently with, GT1b (Figure 6D). These results suggest that TSN is a moderate affinity partial antagonist of BoNT/B intoxication that might act by interfering with the assembly of oligomeric BoNT/B translocation channels. Also, the observation that TSN prevented oligomerization at pH 5.0 but not at pH 4.4 may help to explain why TSN provides only partial protection against BoNT/B in neurons (Figure 6A), since the effects of TSN appear to be pH dependent, and the intraluminal pH of synaptic vesicles is likely to fall transiently to relatively low values, as detailed above.

Figure 6. TSN Hinders BoNT/B Translocation and Assembly Into Oligomers.

Figure 6

Open in a new tab

(A) Immunoblot analysis of syb cleavage after exposure of rat hippocampal neurons to BoNT/B in the presence of TSN. Syntaxin served as a loading control.

(B–E) TSN hinders the oligomerization of BoNT/B. (B) BN-PAGE analysis of BoNT/B holotoxin with GT1b at pH 5.0 in the presence of TSN. Proteins were detected by silver staining. (C) BN-PAGE analysis of BoNT/B holotoxin with GT1b at pH 7.4, pH 5.0 and pH 4.4 in the absence and presence of TSN. (D) BN-PAGE analysis of the time-course of the effect of TSN at pH 5.0. TSN was added either before (−1.5 h), at the same time (0 h) or after (1.5 h) incubation of BoNT/B with GT1b. (E) BN-PAGE analysis of BoNT/B holotoxin at various times after addition of GT1b, in the absence (left) and presence (right) of TSN.

(F–J) TSN modulates BoNT/B channel activity at pH 5.0. BoNT/B channel activity in PC12 cell membranes in the absence (left) or the presence (right) of TSN at pH 5.0. BoNT/B was added to the pipette and channel activity was recorded in inside-out membrane patches (F). Effect of TSN on the time-course of BoNT/B channel conductance growth. Conductances in the absence (black symbols) and presence (gray symbols) of TSN were plotted against time and fitted using sigmoid curves (G). Effect of TSN on minimum conductance (H), maximum conductance (I) and t1/2 of channel conductance growth (J). Data are means ± SEM.

See also Figure S3.

We also used AFM to test whether TSN affected the behavior of BoNT/B on GT1b-containing bilayers, and found that TSN did not significantly change the distribution of molecular volumes of BoNT/B at pH 5.0. As noted above, due to the instability of the supported lipid bilayers at lower pH, we were not able to determine the effects of TSN at lower pH values via AFM; however, the drug did increase the percentage of donut-shaped structures formed by reduced BoNT/B at pH 5.0 from 28±2% to 50±1% (data from three independent experiments; total particle numbers = 463 [TSN] and 816 [control]). The significance of these observations is discussed below in the context of the single channel recordings.

Given the finding that BoNT/B might form oligomeric translocation “machines” and the fact that TSN hindered the assembly of BoNT/B oligomers in solution, we sought to determine whether the channel properties of BoNT/B are modulated by TSN. Our BN-PAGE analysis indicated that, in the presence of GT1b, BoNT/B assembles into trimers in a time-dependent manner at pH 5.0 (Figure 6E, left panel). When exposed at pH 5.0 to membrane patches excised from PC12 cells, BoNT/B induced a small-conductance channel (10 pS), which grew gradually over time until a maximum conductance of around 400 pS was reached (Figure 6F, left panel; Figure 6G). When TSN was present in addition to GT1b, the time-dependent assembly of trimers was inhibited (Figure 6E, right panel). On PC12 cell membrane patches, TSN facilitated the initial conductance growth induced by BoNT/B, but reduced the plateau conductance (Figure 6F, right panel; Figure 6G). Although the minimum conductance of the BoNT/B channel was unaffected by TSN (Figure 6H), both the maximum conductance (Figure 6I) and the t1/2 for conductance growth (Figure 6J) were reduced by ~50%. Since cosedimentation assays demonstrated that TSN does not reduce binding of BoNT/B to GT1b-containing membranes at low pH (data not shown), the drug must inhibit some aspect of translocation. Our results are consistent with the notion that TSN might inhibit translocation by preventing GT1b-dependent, low pH-induced oligomerization of BoNT/B.

DISCUSSION

A remarkable feature of the BoNTs concerns their ability to convert, in response to low pH, into membrane-associated proteins that are capable of delivering the LC into the cytosol of host neurons. This process also represents the least understood step in the action of these agents, so in the current study we carried out experiments with the goal of gaining insight into how this transformation occurs.

Initially, we found that low pH alone was unable to drive the conversion of the BoNT/B-HC into a hydrophobic, membrane-bound species that would be capable of mediating LC translocation. Further experiments revealed that the pH-triggered transformation of BoNT/B into a membrane protein requires interactions with a co-receptor molecule, the ganglioside GT1b. These findings suggest that the toxin is a coincidence detector: that is, binding to a receptor molecule, and exposure to low pH, both play a role in the conversion of the toxin into a putative translocation-competent form. Interestingly, interactions with receptors also prime some viruses for subsequent low pH-triggered fusion with host cells (Côté et al., 2009; Mothes et al., 2000; Nurani et al., 2003).

CD spectroscopy revealed that the conversion of BoNT/B into a hydrophobic species at low pH in the presence of GT1b involves a significant reduction in the α-helical content of the toxin; remarkably, this conformational change was largely reversed at neutral pH. Such reversibility in translocation-related conformational changes appears to be a property that is shared by other bacterial toxins, including diphtheria toxin (Ladokhin et al., 2004) and Clostridium difficile toxin B (Qa’Dan et al., 2000), but the significance of this reversibility remains unclear.

We then turned our attention to experiments designed to gain insights into the structure of the low-pH membrane-form of the toxin using AFM imaging; virtually nothing is known concerning the structure of the toxin in this crucial state. We found that the low-pH membrane-bound form of BoNT/B HC resembles other protein translocation complexes that self-associate to form either double- or triple-donut structures. These include the mitochondrial TIM (Rehling et al., 2003) and TOM complexes (Ahting et al., 1999; Künkele et al., 1998; Model et al., 2008), the endoplasmic reticulum protein Sec61p (Becker et al., 2009; Beckmann et al., 2001), and PhoE porin (Jap et al., 1991). There is some debate as to whether the assembly of these proteins into double- or triple-donut structures plays a role in translocation, or whether individual donuts are functionally active. For Sec61p, for example, early cryo-electron microscopy analysis indicated the formation of trimeric channels (Beckmann et al., 2001); however, more recent analysis led to the conclusion that the protein conducting channel consists of a single Sec61p molecule (Becker et al., 2009). In the case of BoNT/B, earlier studies are consistent with functionally relevant oligomerization of the toxin in membranes. For example, the rate of BoNT channel formation on membranes was shown to depend on the square of the toxin concentration, indicating cooperative assembly of the translocation competent form of the toxin (Donovan and Middlebrook, 1986; Hoch, 1985). Our observation that TSN inhibits BoNT/B trimerization and in parallel truncates channel growth, lends further support to the idea that trimers may represent the optimally functional translocation unit.

We note that changes in BoNT/B conformation, and assembly into oligomers in the presence of GT1b, occurred over a very narrow pH range, and that formation of oligomers at the same pH (pH 5.0) was more efficient in solution than on membranes. In addition, the low pH requirements for BoNT/B oligomerization in solution (pH 4.8~5.0) (Figure S3A) were less stringent than for effective surface entry into neurons (pH 4.4) (Figures S3B and S3C). These results are consistent with observations made using other pore-forming bacterial toxins, such as diphtheria toxin and anthrax toxin. Diphtheria toxin also changes conformation over a very narrow pH range (0.2 units) once the pH drops below pH 5.0 (Blewitt et al., 1985). Further, for anthrax toxin, pore formation requires a lower pH (~1 unit) on membranes than in solution (Miller et al., 1999). These findings imply that membranes also play a crucial role in shaping the behavior of bacterial toxins.

We also probed the assembly of the translocation machinery with TSN and found that this drug truncates the growth of channels formed by BoNT/B (peaking at 200 pS instead of 400 pS) and prevents trimer formation at pH 5.0. Interestingly, AFM imaging further revealed that TSN increases the proportion of toxin particles that appear as donuts following reduction. These findings are consistent with the increased rate of appearance of low-conductance BoNT/B channels in the presence of TSN. Therefore, TSN may prevent BoNT/B translocation by hindering oligomeric channel assembly, and such inhibitory effects are likely to be achieved - at least in part - by stabilizing the monomeric and low-conductance form of the toxin.

The AFM images revealed that the oligomeric BoNT/B donut structures have central holes that are similar in radius to those of single donuts, making it difficult to account for the very large increase in single-channel conductance as the pore grows on the membranes of PC12 cells (from 10 to 400 pS). However, it should be emphasized that the two structures we have imaged likely represent the beginning (membrane-associated holotoxin) and end (HC with LC removed) of the normal LC translocation process, and we have not been able to image the toxin in the act of translocation. It is possible that during translocation, three holotoxin molecules form a structure that transiently generates a very large pore, which later relaxes back to the triple-donut structure seen here when the holotoxin is reduced.

Given our finding that interactions with GT1b enable BoNT/B to sense a drop in pH, we tested whether preincubation of the toxin with GT1b, in solution at low extracellular pH, leads to premature activation of the toxin such that it enters an “off-pathway”. Indeed, the action of BoNT/B on neurons was significantly reduced following such a preincubation step (data not shown). However, it should be noted that binding of exogenous GT1b to BoNT/B can also competitively inhibit binding to cellular receptors. In addition to GT1b, BoNT/B also specifically binds to the protein co-receptor, syt I/II, to undergo receptor-mediated endocytosis, prompting the question of whether syt I/II can also affect the pH-sensing activity of BoNT/B. We attempted to address this issue using P21, a soluble peptide of syt II that contains the binding site for BoNT/B (Dong et al., 2003), and failed to observe any effect. However, the lack of effect could be due to the fact that P21 binds to BoNT/B only weakly. High-affinity binding appears to require a syt I/II fragment that contains an intact transmembrane region (Dong et al., 2003), and the presence of a hydrophobic segment precludes analysis in our partitioning and co-sedimentation assays.

In summary, the data presented here provide a glimpse into the most enigmatic state of a Clostridial neurotoxin: the low pH-induced membrane-bound form of the toxin that mediates translocation. Further studies are needed to image translocation of nascent LC. This might be achieved by arresting translocation with antibodies (Fischer and Montal, 2007), and using cryo-electron microscopy and single particle reconstruction to gain a higher resolution view of this translocation machine. Moreover, given the function for GT1b identified here, it will be important to determine what happens when the GT1b binding domain of a BoNT is removed. In the case of BoNT/A, deletion of the receptor binding domain results in a toxin that immediately forms large conductance channels without a requirement for low pH (Fischer et al., 2008). Taking into account our current findings, in conjunction with previous data, the following model emerges: the receptor binding domain normally plays an inhibitory role in the conversion of the toxin into a translocation-competent form; this inhibition would be relieved by low pH and binding of GT1b, or by deletion of this domain. Hence, for the holotoxin, translocation occurs at the correct time and place.

EXPERIMENTAL PROCEDURES

Antibodies, BoNT/B and Recombinant Proteins

Details of the above reagents are provided in Supplemental Experimental Procedures.

Animals and Mice Lines

All procedures involving animals were performed according to US National Institutes of Health guidelines, as approved by the Animal Care and Use Committee of the University of Wisconsin, Madison. See Supplemental Experimental Procedures for details.

Off-pathway Assay

Cultures of rat hippocampal neurons and cell lysates were prepared as previously described (Dong et al., 2007); see Supplemental Experimental Procedures for further details. BoNT/B was incubated in high-K+ buffer at pH 7.4 or pH 4.4 for 2 h at 37°C, before addition to neurons for 5 min. High-K+ buffer stimulates recycling of synaptic vesicles, thus facilitating uptake of toxins into neurons by endocytosis. Neurons were washed with low-K+ buffer at pH 7.4 or pH 4.4, respectively, and incubated in neuronal media at 37°C for 20 h. Cells were lysed, and cleavage of syb by BoNT/B was analyzed by SDS-PAGE and immunoblotting using anti-syb and anti-syntaxin (loading control) antibodies. The cleavage of syb was quantified using data from three independent trials.

Translocation of BoNT/B Across the Plasma Membrane

Where appropriate, hippocampal neurons were preincubated with bafilomycin A1 (1 μM, Sigma-Aldrich) for 30 min at 37°C. Unless otherwise specified, bafilomycin was added to all buffers used subsequently. Neurons were treated with high-K+ buffer (85 mM NaCl, 60 mM KCl, 2 mM CaCl2, 2 mM MgCl2, 5.5 mM glucose, 10 mM Hepes, 10 mM sodium 2-(N-morpholino)ethanesulfonate (Na-MES), pH 7.4) for 5 min at 37°C, before exposure to BoNT/B (30 nM) for 10 min at 0°C. Note: with 30 nM, rather than 10 nM BoNT/B (as used in off-pathway assays), we observed reliable cleavage of syb in neurons for all batches of toxin, so this concentration was used in all subsequent cell entry experiments. Neurons were washed and treated with low-K+ buffer (140 mM NaCl, 5 mM KCl, 2 mM CaCl2, 2 mM MgCl2, 5.5 mM glucose, 10 mM Hepes, 10 mM Na-MES, pH as indicated) for 8 min at 37°C. For Gan KO mouse hippocampal neurons and rat hippocampal neurons treated with FB1 (25 μM, added to neurons at 4 days in vitro, Sigma-Aldrich), BoNT/B (30 nM; in low low-K+ buffer, pH as indicated) was added for 10 min after high-K+ buffer treatment (pH 7.4). The neurons were then washed and incubated without bafilomycin for 24 h. As tested by Trypan blue staining, >90% cells were viable after this 24 h incubation. Cells were harvested as described above, and cell lysates were analyzed by SDS-PAGE and immunoblotting using anti-syb, anti-SNAP-25 and anti-syntaxin antibodies. The amount of syb in each sample was normalized against SNAP-25, and cleavage of syb was quantified in three independent trials. Cells without bafilomycin treatment served as negative (minus BoNT/B) or positive (plus BoNT/B) controls for syb cleavage.

CD Spectroscopy, Cosedimentation and Triton X-114 Partitioning Assays

CD experiments were performed using an Aviv Model 202SF spectrometer. Cosedimentation assays were performed essentially as described previously (Bai et al., 2000). Triton X-114 partitioning assays were performed as described previously (Bordier, 1981). Details are also provided in Supplemental Experimental Procedures.

AFM Imaging

AFM imaging was performed using a Veeco Digital Instruments Multimode instrument, controlled by a Nanoscope IIIa controller. Samples were prepared for AFM imaging as described in Supplemental Experimental Procedures.

All AFM images were plane-fitted before any analysis to remove tilt. For dry imaging, particles were analyzed using the Nanoscope V5.31rl software. Fifty particles were first analyzed manually. These measurements were used to define the detection level in the program, which in turn defines the minimum z-value to be considered as part of the particle. After particles were thus defined, height and radius (radius of a circle fitted to data obtained for each particle) were determined. For fluid imaging, particles were first identified from the phase image and their heights and radii were measured manually using the section tool.

Bound particles assumed the approximate shape of a spherical cap. The radius and height of each particle were used to calculate its molecular volume using the formula

Vm=(πh/6)(3r2+h2) (1)

where h is the particle height and r is the radius, as described previously (Schneider et al., 1998). For comparison, molecular volume based on molecular mass was calculated using the equation

Vc=(M0/N0)(V1+dV2) (2)

where M0 is the molecular mass, N0 is Avogadro s number, V1 and V2 are the partial specific volumes of particle (0.74 cm3/g) and water (1 cm3/g), respectively, and d is the extent of protein hydration (taken as 0.4 g water/g protein).

BN-PAGE Assay of BoNT/B

The NativePAGE Novex Bis-Tris gel system for BN-PAGE was obtained from Invitrogen. BoNT/B (30 or 100 nM) was incubated alone or with the ganglioside GT1b or GM1 (10 μM, unless specifically indicated) at pH 7.4, pH 5.0 and pH 4.4 at 37°C for 2 h. For samples containing TSN, TSN (concentration as indicated) was preincubated with BoNT/B for 2 h before GT1b was added. In Figure 6D, TSN was either preincubated with BoNT/B for 1.5 h, added the same time as BoNT/B, or 1.5 h after BoNT/B, followed by incubation at 37°C for 3 h. Samples were analyzed by BN-PAGE and silver staining.

Electrophysiological Recording of BoNT/B Channels

Recordings were performed essentially as described previously (Sheridan, 1998), see Supplemental Experimental Procedures for details.

Supplementary Material

01

HIGHLIGHTS.

  • Gangliosides are required for cell surface entry of BoNT/B at low pH

  • The ganglioside GT1b enables BoNT/B to sense low pH and change conformation

  • BoNT/B assembles into channel-like structures on GT1b-containing bilayers at low pH

  • TSN, a botulism therapeutic, inhibits GT1b-dependent BoNT/B oligomerization at low pH

Acknowledgments

We thank M. Dong for helpful discussions, D. Wang for helping to prepare rat hippocampal neurons, and the Chapman lab for critical comments regarding this manuscript. We also acknowledge J.T. Barbieri for preparing HCR/B, with support from the GLRCE U54 AI057153. This work was supported by a grant from the NIH (AI057744) to E.R.C. S. Suresh was supported by a Cambridge Nehru Fellowship and a Cambridge Overseas Research Studentship. E.R.C. is an Investigator of the Howard Hughes Medical Institute.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Ahting U, Thun C, Hegerl R, Typke D, Nargang FE, Neupert W, Nussberger S. The TOM core complex: the general protein import pore of the outer membrane of mitochondria. J Cell Biol. 1999;147:959–968. doi: 10.1083/jcb.147.5.959. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Arnon SS, Schechter R, Inglesby TV, Henderson DA, Bartlett JG, Ascher MS, Eitzen E, Fine AD, Hauer J, Layton M, et al. Botulinum toxin as a biological weapon: medical and public health management. JAMA. 2001;285:1059–1070. doi: 10.1001/jama.285.8.1059. [DOI] [PubMed] [Google Scholar]
  3. Bai J, Earles CA, Lewis JL, Chapman ER. Membrane-embedded synaptotagmin penetrates cis or trans target membranes and clusters via a novel mechanism. J Biol Chem. 2000;275:25427–25435. doi: 10.1074/jbc.M906729199. [DOI] [PubMed] [Google Scholar]
  4. Becker T, Bhushan S, Jarasch A, Armache JP, Funes S, Jossinet F, Gumbart J, Mielke T, Berninghausen O, Schulten K, et al. Structure of monomeric yeast and mammalian Sec61 complexes interacting with the translating ribosome. Science. 2009;326:1369–1373. doi: 10.1126/science.1178535. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Beckmann R, Spahn CM, Eswar N, Helmers J, Penczek PA, Sali A, Frank J, Blobel G. Architecture of the protein-conducting channel associated with the translating 80S ribosome. Cell. 2001;107:361–372. doi: 10.1016/s0092-8674(01)00541-4. [DOI] [PubMed] [Google Scholar]
  6. Blewitt MG, Chung LA, London E. Effect of pH on the conformation of diphtheria toxin and its implications for membrane penetration. Biochemistry. 1985;24:5458–5464. doi: 10.1021/bi00341a027. [DOI] [PubMed] [Google Scholar]
  7. Bordier C. Phase separation of integral membrane proteins in Triton X-114 solution. J Biol Chem. 1981;256:1604–1607. [PubMed] [Google Scholar]
  8. Chai Q, Arndt JW, Dong M, Tepp WH, Johnson EA, Chapman ER, Stevens RC. Structural basis of cell surface receptor recognition by botulinum neurotoxin B. Nature. 2006;444:1096–1100. doi: 10.1038/nature05411. [DOI] [PubMed] [Google Scholar]
  9. Chen YH, Yang JT, Martinez HM. Determination of the secondary structures of proteins by circular dichroism and optical rotatory dispersion. Biochemistry. 1972;11:4120–4131. doi: 10.1021/bi00772a015. [DOI] [PubMed] [Google Scholar]
  10. Côté M, Zheng YM, Liu SL. Receptor binding and low pH coactivate oncogenic retrovirus envelope-mediated fusion. J Virol. 2009;83:11447–11455. doi: 10.1128/JVI.00748-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. de Chaves EI, Bussiere M, Vance DE, Campenot RB, Vance JE. Elevation of ceramide within distal neurites inhibits neurite growth in cultured rat sympathetic neurons. J Biol Chem. 1997;272:3028–3035. doi: 10.1074/jbc.272.5.3028. [DOI] [PubMed] [Google Scholar]
  12. Dong M, Liu H, Tepp WH, Johnson EA, Janz R, Chapman ER. Glycosylated SV2A and SV2B mediate the entry of botulinum neurotoxin E into neurons. Mol Biol Cell. 2008;19:5226–5237. doi: 10.1091/mbc.E08-07-0765. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Dong M, Richards DA, Goodnough MC, Tepp WH, Johnson EA, Chapman ER. Synaptotagmins I and II mediate entry of botulinum neurotoxin B into cells. J Cell Biol. 2003;162:1293–1303. doi: 10.1083/jcb.200305098. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Dong M, Tepp WH, Liu H, Johnson EA, Chapman ER. Mechanism of botulinum neurotoxin B and G entry into hippocampal neurons. J Cell Biol. 2007;179:1511–1522. doi: 10.1083/jcb.200707184. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Donovan JJ, Middlebrook JL. Ion-conducting channels produced by botulinum toxin in planar lipid membranes. Biochemistry. 1986;25:2872–2876. doi: 10.1021/bi00358a020. [DOI] [PubMed] [Google Scholar]
  16. Eswaramoorthy S, Kumaran D, Keller J, Swaminathan S. Role of metals in the biological activity of Clostridium botulinum neurotoxins. Biochemistry. 2004;43:2209–2216. doi: 10.1021/bi035844k. [DOI] [PubMed] [Google Scholar]
  17. Fischer A, Montal M. Crucial role of the disulfide bridge between botulinum neurotoxin light and heavy chains in protease translocation across membranes. J Biol Chem. 2007;282:29604–29611. doi: 10.1074/jbc.M703619200. [DOI] [PubMed] [Google Scholar]
  18. Fischer A, Mushrush DJ, Lacy DB, Montal M. Botulinum neurotoxin devoid of receptor binding domain translocates active protease. PLoS Pathog. 2008;4:e1000245. doi: 10.1371/journal.ppat.1000245. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Fischer A, Nakai Y, Eubanks LM, Clancy CM, Tepp WH, Pellett S, Dickerson TJ, Johnson EA, Janda KD, Montal M. Bimodal modulation of the botulinum neurotoxin protein-conducting channel. Proc Natl Acad Sci USA. 2009;106:1330–1335. doi: 10.1073/pnas.0812839106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Heuser JE, Reese TS. Structural changes after transmitter release at the frog neuromuscular junction. J Cell Biol. 1981;88:564–580. doi: 10.1083/jcb.88.3.564. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Hoch DH. Doctoral Dissertation. Albert Einstein College of Medicine; 1985. Botulinum, tetanus, and diphtheria toxin channels in planar lipid bilayers. [Google Scholar]
  22. Hoch DH, Romero-Mira M, Ehrlich BE, Finkelstein A, DasGupta BR, Simpson LL. Channels formed by botulinum, tetanus, and diphtheria toxins in planar lipid bilayers: relevance to translocation of proteins across membranes. Proc Natl Acad Sci USA. 1985;82:1692–1696. doi: 10.1073/pnas.82.6.1692. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Jap BK, Walian PJ, Gehring K. Structural architecture of an outer membrane channel as determined by electron crystallography. Nature. 1991;350:167–170. doi: 10.1038/350167a0. [DOI] [PubMed] [Google Scholar]
  24. Jin R, Rummel A, Binz T, Brunger AT. Botulinum neurotoxin B recognizes its protein receptor with high affinity and specificity. Nature. 2006;444:1092–1095. doi: 10.1038/nature05387. [DOI] [PubMed] [Google Scholar]
  25. Kamata Y, Kozaki S, Sakaguchi G. Effects of pH on the binding of Clostridium botulinum type E derivative toxin to gangliosides and phospholipids. FEMS Microbiol Lett. 1988;55:71–76. [Google Scholar]
  26. Keller JE, Cai F, Neale EA. Uptake of botulinum neurotoxin into cultured neurons. Biochemistry. 2004;43:526–532. doi: 10.1021/bi0356698. [DOI] [PubMed] [Google Scholar]
  27. Kintzer AF, Sterling HJ, Tang II, Abdul-Gader A, Miles AJ, Wallace BA, Williams ER, Krantz BA. Role of the protective antigen octamer in the molecular mechanism of anthrax lethal toxin stabilization in plasma. J Mol Biol. 2010;399:741–758. doi: 10.1016/j.jmb.2010.04.041. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Künkele KP, Heins S, Dembowski M, Nargang FE, Benz R, Thieffry M, Walz J, Lill R, Nussberger S, Neupert W. The preprotein translocation channel of the outer membrane of mitochondria. Cell. 1998;93:1009–1019. doi: 10.1016/s0092-8674(00)81206-4. [DOI] [PubMed] [Google Scholar]
  29. Ladokhin AS, Legmann R, Collier RJ, White SH. Reversible refolding of the diphtheria toxin T-domain on lipid membranes. Biochemistry. 2004;43:7451–7458. doi: 10.1021/bi036157w. [DOI] [PubMed] [Google Scholar]
  30. Li MF, Shi YL. Toosendanin interferes with pore formation of botulinum toxin type A in PC12 cell membrane. Acta Pharmacol Sin. 2006;27:66–70. doi: 10.1111/j.1745-7254.2006.00236.x. [DOI] [PubMed] [Google Scholar]
  31. Miesenböck G, De Angelis DA, Rothman JE. Visualizing secretion and synaptic transmission with pH-sensitive green fluorescent proteins. Nature. 1998;394:192–195. doi: 10.1038/28190. [DOI] [PubMed] [Google Scholar]
  32. Miller CJ, Elliott JL, Collier RJ. Anthrax protective antigen: prepore-to-pore conversion. Biochemistry. 1999;38:10432–10441. doi: 10.1021/bi990792d. [DOI] [PubMed] [Google Scholar]
  33. Model K, Meisinger C, Kühlbrandt W. Cryo-electron microscopy structure of a yeast mitochondrial preprotein translocase. J Mol Biol. 2008;383:1049–1057. doi: 10.1016/j.jmb.2008.07.087. [DOI] [PubMed] [Google Scholar]
  34. Montal M. Botulinum neurotoxin: a marvel of protein design. Annu Rev Biochem. 2010;79:591–617. doi: 10.1146/annurev.biochem.051908.125345. [DOI] [PubMed] [Google Scholar]
  35. Montecucco C, Molgó J. Botulinal neurotoxins: revival of an old killer. Curr Opin Pharmacol. 2005;5:274–279. doi: 10.1016/j.coph.2004.12.006. [DOI] [PubMed] [Google Scholar]
  36. Mothes W, Boerger AL, Narayan S, Cunningham JM, Young JA. Retroviral entry mediated by receptor priming and low pH triggering of an envelope glycoprotein. Cell. 2000;103:679–689. doi: 10.1016/s0092-8674(00)00170-7. [DOI] [PubMed] [Google Scholar]
  37. Nishiki T, Tokuyama Y, Kamata Y, Nemoto Y, Yoshida A, Sato K, Sekiguchi M, Takahashi M, Kozaki S. The high-affinity binding of Clostridium botulinum type B neurotoxin to synaptotagmin II associated with gangliosides GT1b/GD1a. FEBS Lett. 1996;378:253–257. doi: 10.1016/0014-5793(95)01471-3. [DOI] [PubMed] [Google Scholar]
  38. Nurani G, Lindqvist B, Casasnovas JM. Receptor priming of major group human rhinoviruses for uncoating and entry at mild low-pH environments. J Virol. 2003;77:11985–11991. doi: 10.1128/JVI.77.22.11985-11991.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Qa’Dan M, Spyres LM, Ballard JD. pH-induced conformational changes in Clostridium difficile toxin B. Infect Immun. 2000;68:2470–2474. doi: 10.1128/iai.68.5.2470-2474.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Rehling P, Model K, Brandner K, Kovermann P, Sickmann A, Meyer HE, Kühlbrandt W, Wagner R, Truscott KN, Pfanner N. Protein insertion into the mitochondrial inner membrane by a twin-pore translocase. Science. 2003;299:1747–1751. doi: 10.1126/science.1080945. [DOI] [PubMed] [Google Scholar]
  41. Rotz LD, Khan AS, Lillibridge SR, Ostroff SM, Hughes JM. Public health assessment of potential biological terrorism agents. Emerg Infect Dis. 2002;8:225–230. doi: 10.3201/eid0802.010164. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Rummel A, Eichner T, Weil T, Karnath T, Gutcaits A, Mahrhold S, Sandhoff K, Proia RL, Acharya KR, Bigalke H, et al. Identification of the protein receptor binding site of botulinum neurotoxins B and G proves the double-receptor concept. Proc Natl Acad Sci USA. 2007;104:359–364. doi: 10.1073/pnas.0609713104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Saslowsky DE, Lawrence J, Ren X, Brown DA, Henderson RM, Edwardson JM. Placental alkaline phosphatase is efficiently targeted to rafts in supported lipid bilayers. J Biol Chem. 2002;277:26966–26970. doi: 10.1074/jbc.M204669200. [DOI] [PubMed] [Google Scholar]
  44. Schiavo G, Matteoli M, Montecucco C. Neurotoxins affecting neuroexocytosis. Physiol Rev. 2000;80:717–766. doi: 10.1152/physrev.2000.80.2.717. [DOI] [PubMed] [Google Scholar]
  45. Schmitt J, Karalewitz A, Benefield DA, Mushrush DJ, Pruitt RN, Spiller BW, Barbieri JT, Lacy DB. Structural analysis of botulinum neurotoxin type G receptor binding. Biochemistry. 2010;49:5200–5205. doi: 10.1021/bi100412v. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Schneider SW, Larmer J, Henderson RM, Oberleithner H. Molecular weights of individual proteins correlate with molecular volumes measured by atomic force microscopy. Pflugers Arch. 1998;435:362–367. doi: 10.1007/s004240050524. [DOI] [PubMed] [Google Scholar]
  47. Sheridan RE. Gating and permeability of ion channels produced by botulinum toxin types A and E in PC12 cell membranes. Toxicon. 1998;36:703–717. doi: 10.1016/s0041-0101(97)00131-1. [DOI] [PubMed] [Google Scholar]
  48. Simpson LL, Coffield JA, Bakry N. Inhibition of vacuolar adenosine triphosphatase antagonizes the effects of clostridial neurotoxins but not phospholipase A2 neurotoxins. J Pharmacol Exp Ther. 1994;269:256–262. [PubMed] [Google Scholar]
  49. Sun J, Vernier G, Wigelsworth DJ, Collier RJ. Insertion of anthrax protective antigen into liposomal membranes: effects of a receptor. J Biol Chem. 2007;282:1059–1065. doi: 10.1074/jbc.M609869200. [DOI] [PubMed] [Google Scholar]
  50. Suresh S, Edwardson JM. The endophilin N-BAR domain perturbs the structure of lipid bilayers. Biochemistry. 49:5766–5771. doi: 10.1021/bi100760e. [DOI] [PubMed] [Google Scholar]
  51. Swaminathan S, Eswaramoorthy S. Structural analysis of the catalytic and binding sites of Clostridium botulinum neurotoxin B. Nat Struct Biol. 2000;7:693–699. doi: 10.1038/78005. [DOI] [PubMed] [Google Scholar]
  52. Young JA, Collier RJ. Anthrax toxin: receptor binding, internalization, pore formation, and translocation. Annu Rev Biochem. 2007;76:243–265. doi: 10.1146/annurev.biochem.75.103004.142728. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

01
NIHMS327875-supplement-01.pdf (673KB, pdf)

RESOURCES