June 2016 Volume 3, Issue 4

The Effect of Temperature, Chelators, and Denaturants on the Activity of Botulinum Neurotoxin LC/A-424

Sydney Mason1* and Michael Krebs2
Student1: SEAP students, McDonogh School, 8600 McDonogh RD Owings Mills, MD 21117
Mentor2: USAMRICD, 3100 Ricketts Point Road Aberdeen Proving Ground, MD 21010
*Corresponding author: smason@mcdonogh.org

Abstract

Botulinum neurotoxin (BoNT), being one of the most lethal substances known to man, is a large concern as a potential chemical weapon. BoNT is produced by bacteria of the genus Clostridium and consists of a light chain (LC), responsible for enzymatic activity, and a heavy chain (HC), responsible for the translocation and binding of the toxin. The two chains are connected by a single disulfide bond. BoNT LC also contains zinc within its active site integral to the cleavage of its substrate. BoNTs specifically cleave SNARE proteins in the presynaptic terminal of the neuromuscular junction, which are essential for neurotransmitter release. Inhibition of these proteins causes flaccid paralysis and, depending on the serotype, this can last from weeks to years. Using a cell free FRET-based reporter assay, we tested the effects of temperature, chelators, and denaturants on the activity of a truncated LC of serotype A, which cleaves the SNARE protein SNAP-25. BoNT LC/A-424 activity was sensitive to temperatures above 50 ͦC and showed no activity at 60 ͦC when incubated for 30 minutes. It was also shown in further studies that LC/A-424 inactivation by high temperature is irreversible. In addition, two chelators were tested, TPEN and EDTA. Although EDTA is a universal metal chelator and TPEN is more zinc specific, they both have similar effects on the toxin activity, with complete inhibition at 20 uM and above. Finally, the two denaturants used were urea and guanidinium chloride (guanidine hydrochloride). Urea was found to cause complete inhibition only in the molar range, while the guanidinium began to show complete inhibition in the upper millimolar range. These insights can provide us with knowledge of the toxin and how it works under various environmental conditions. The knowledge of the biochemistry of Botulinum neurotoxin is crucial for drug development and the discovery of a treatment for botulism.

Introduction

Botulinum neurotoxin (BoNT) is one of the most lethal toxins known to man and has been classified as a Category A bio threat agent by the Centers for Disease Control and Prevention (CDC) (1). The toxin is a large concern as a potential chemical weapon that can be utilized in terrorist attacks either as an aerosol or a contaminant in the food supply (1).

BoNT is produced by bacteria of the genus Clostridium (2) and is separated into seven different serotypes designated by letters A through G. Each serotype’s chemical structure is similar consisting of a heavy chain (HC) and a light chain (LC) linked by a single disulfide bond. The HC is made up of two domains: the translocation domain and the binding domain. The translocation domain is responsible for the translocation of the LC into the cytosol, and the binding domain is responsible for binding to polysialoganglioside (PSG) receptors and synaptic protein receptors on the presynaptic membrane (3). The LC comprises a zinc metalloprotease domain that cleaves SNARE proteins (4). Through the cleavage of these SNARE proteins, which are integral to the exocytotic process (2) within the synapse, BoNTs inhibit the release of neurotransmitter at the neuromuscular junction thus causing flaccid paralysis. We use serotype BoNT/A in these experiments because it is the most potent and causes the longest-lasting and most severe cases of paralysis.  However, only the LC was used since this is a cell free assay and, therefore, it is not necessary for the BoNT to have the translocation and binding domains to enter the cytosol. The LC used is also a truncated version with an amino acid chain of 424 rather than the native 438.

There are several steps in the process of botulinum intoxication. The toxin is most often introduced into the body through ingestion or an open wound (1). It will remain in the bloodstream for some time until reaching a synapse. The carboxy-terminal end of the HC, responsible for binding to receptors, will then bind to a polysialoganglioside (PSG) receptor that is present on the presynaptic membrane, followed by binding to a protein receptor that is located either inside the exocytosed synaptic vesicle or on the presynaptic membrane. BoNT/A specifically binds to a PSG and SV2. The BoNT, now inside the synaptic vesicle, is carried along as the vesicle goes through endocytosis (3). Once inside the presynaptic terminal, the translocation domain of the HC inserts into the endosome membrane of the synaptic vesicle it is contained within and forms a pore that allows for the translocation of the LC into the cytosol (4). The disulfide bond is then broken and the LC is free to cleave its designated SNARE protein. The LC also contains zinc within its active site that is crucial for the cleavage of the SNARE proteins.

In this study we investigated the effects of temperature extremes, chelators, and denaturants on the activity of the LC of the serotype BoNT/A. These insights can provide us with further knowledge of the toxin and how it works under various environmental conditions, which is important information for future drug development.

The method used in these experiments was the Forster Resonance Energy Transfer (FRET) cell free assay (BioSentinal, Madison, WI) described by Ruge et al in 2011 (5). The reporter consisted of residues 141-206 of SNAP-25 sandwiched between two fluorescent proteins: a cyan fluorescent protein (CFP) and a yellow fluorescent protein (YFP). Since the CFP and YFP are in such close proximity, when the CFP is excited its energy is transferred to the YFP through FRET. Consequentially, the CFP fluorescence emission is quenched while the YFP fluorescence emission increases (5). However, when the BoNT cleaves the SNAP-25 in the middle, the YFP and CFP are no longer close, and the CFP emission is restored, while the YFP emission is quenched (5). The cleavage of the SNAP-25 is tracked by measuring the ratio of the YFP emission to CFP emission throughout a set amount of time using relative fluorescence units (RFU).

BoNT/A LC activity was found to be sensitive to temperatures above 50 ͦC with no activity at 60 ͦC. The toxin activity was also tested using varying concentrations of two chelators: N,N,N′,N′-Tetrakis(2-pyridylmethyl)ethylenediamine (TPEN) and Ethylenediaminetetraacetic acid (EDTA). Although EDTA is a universal chelator and TPEN is zinc specific, it was found that they both have extremely similar effects on the toxin activity, with complete inhibition at 20 uM and above. Two denaturants were also tested: urea and guanidinium chloride (guanidine hydrochloride). Urea was found to cause complete inhibition only in the molar range, while the guanidinium began to show complete inhibition in the upper millimolar range.

This knowledge of the biochemistry of the toxin and how it works under various environmental conditions can provide us with important and useful knowledge for future drug development towards a cure for botulism.

Materials and Methods

Reagents: The BoNT LC/A-424, provided by Dr. Subramanyam Swaminathan (Brookhaven National Laboratory, Upton, NY), was used throughout all the experiments. BoTest Botulinum Neurotoxin Detection Kit (Biosentinel Pharmaceuticals, Madison, WI) was used to detect the LC/A-424 activity under different conditions. This kit consists of a 10X BoTest Reaction Buffer (Part A1002, Lot 1964A) and an A/E Reporter (Part A1oo4, Lot 1964B). The composition of the reaction buffer was 500 mM Hepes-NaOH, pH 7.1, 50 mM NaCl, 1% Tween-20, and 100 uM ZnCl/2. The composition of the reporter was 20 uM in 50 mM Hepes-NaOH, 10 mM NaCl, and 15% glycerol. The reporter consisted of two fluorescent proteins, one cyan fluorescent protein and one yellow fluorescent protein, with residues 141-206 of SNAP-25 sandwiched in between. This kit was used in all experiments along with a bond-breaker tris(2-carboxyethyl) phosphine solution (TCEP, Thermo Scientific, Waltham, MA).

All chelators and denaturants used were purchased from Sigma (St. Louis, MO). The two chelators used were Ethylenediamine tetracetic acid (EDTA) and N,N,N’,N’-tetrakis (z-pyridyl-methyl) Ethylenediamine (TPEN). The two denaturants used were guanidine hydrochloride (guanidinium chloride) and urea.

Assays: A cell free FRET-based reporter assay described by Ruge et al in 2011 (5) was used. Each 96 well plate was incubated for 30 minutes and then run for 60 minutes. When testing chelators and denaturants, the plate was both incubated and run at 37 ͦC. When testing temperature, the plate was incubated and run at the temperature being tested. The plate reader used was a Molecular Devices SpectraMax M5 Fluorescence Plate Reader (San Francisco, CA). The CFP was excited at a wavelength of 435 nM and its emission was tracked at a wavelength of 470 nM. The YFP emission was tracked at a wavelength of 527 nM.

For all conditions, a concentration of 1 nM LC/A-424 was made in the experiment using a stock of 100 nM. The varying concentrations of chelators and denaturants were all done at half log units in the mM or µM range. Molar range concentrations were not done at half log units.

Data analysis: The numbers used to perform data analysis were the YFP/CFP ratios, which represent the reporter cleavage by the LC/A-424. The data was organized and processed using a Microsoft Excel based script created by Mr. Richard Sweeney (USAMRICD, Aberdeen Proving Ground, MD). Graphs were created using Prism 5 for Windows version 5.04 (Graph Pad Software Inc. LaJolla, CA). To determine statistical significance the program Instat (Graph Pad Software Inc. Lajolla, CA) was used to run Tukey-Kramer Multiple Comparisons tests comparing the results of the control to the results of the various temperatures and concentrations at certain times. One-way ANOVAs were run to compare all groups. A significance level of 0.05 was used.      

Results

Temperature
Six different temperatures were tested: 24°C, 33°C, 41°C, 50°C, 55°C, and 60°C, with 37°C, being around human body temperature, as the control. It was found that at temperatures above 50°C, the LC/A-424 activity began to be inhibited and at 60°C and above completely inhibited. Temperatures below 50°C were similar to the control, at 55°C inhibition is 90-95% and complete inhibition is observed at 60°C. At 24°C the LC/A-424 had 20-30% activity inhibited. At 33°C, the LC/A-424 activity was close to the control, and at 41°C, LC/A-424 activity was identical to the control. At 50°C, activity was 20% inhibited, and at 55°C the LC/A-424 activity was diminished, and at 60°C it was completely inhibited. An experiment was also run in which a plate was heated up to 60°C for an hour, then cooled to 37°C for 30 minutes, and then run at 37°C. The results were the same as when a plate was heated to 60°C and run at 60°C: complete inhibition (Fig 2).  These results reflect the overall trend of the activity as recorded on the graphs. Table 1 shows the results with standard deviations at specific time points within each experiment.

  • fig1a
  • fig1b
  • fig1c

FIGURE 1. LC/A 424 ACTIVITY IS TEMPERATURE DEPENDENT: A) This graph shows a representative of LC activity dependency on temperature between 0 and 60 minutes. B) This graph shows the toxin activity at 10 minutes. 60 ͦC is shown to be the same as the reporter alone with similar result at 55 ͦC. Most others temperatures are close to the control all being under 50 % inhibition. C) This graph shows the toxin activity at 45 minutes. 60 ͦC is still the same as the reporter alone, which shows no decrease in substrate. The 55 ͦC is still close with about 90 % inhibition. The others are fairly close to the control. ***significant difference from the control with a p value of <0.001. **significant difference from the control with a p value of <0.01 .

  • fig1a
    FIGURE 1. LC/A 424 ACTIVITY IS TEMPERATURE DEPENDENT A) This graph shows a representative of LC activity dependency on temperature between 0 and 60 minutes.
  • fig1b
    FIGURE 1. LC/A 424 ACTIVITY IS TEMPERATURE DEPENDENT B) This graph shows the toxin activity at 10 minutes. 60 ͦC is shown to be the same as the reporter alone with similar result at 55 ͦC. Most others temperatures are close to the control all being under 50 % inhibition.
  • fig1c
    FIGURE 1. LC/A 424 ACTIVITY IS TEMPERATURE DEPENDENT C) This graph shows the toxin activity at 45 minutes. 60 ͦC is still the same as the reporter alone, which shows no decrease in substrate. The 55 ͦC is still close with about 90 % inhibition. The others are fairly close to the control.

 

  • fig2

FIGURE 2. SWAMI LC/A-424 HEATED TO 60 ͦC AND RUN AT 37 ͦC: When heated up to 60 ͦC for an hour and then subsequentially cooled down to 37 ͦC for 30 minutes the LC did not regain activity as represented by the straight line in the graph.

  • fig2
    FIGURE 2- SWAMI LC/A-424 HEATED TO 60 ͦC AND RUN AT 37 ͦC: When heated up to 60 ͦC for an hour and then subsequentially cooled down to 37 ͦC for 30 minutes the LC did not regain activity as represented by the straight line in the graph above.

 

Chelators
Two different chelators were tested: EDTA and TPEN. Although EDTA is a universal chealtor and TPEN is a zinc specific chelator (BoNT A contains zinc in its active site) the two chelators were found to have similar effects on the LC/A-424 activity. At 20 uM and above there was complete inhibition. At 10 uM there was less than 50% inhibition. At 3 uM and below there was no inhibition (Table 2). The bar graphs show the two chelators similarity at 3 different concentrations (Fig 3B and Fig 3C). These results reflect the overall trend of the activity as recorded on the graphs.

 

  • fig3a
  • fig3b
  • fig3c

FIGURE 3. LC/A-424 ACTIVITY IS AFFECTED BY CHELATORS: A) This graph shows that the EDTA and TPEN had similar effects on the activity of the LC/A-424 between 0 and 60 minutes. B) This graph shows the toxin activity at 10 minutes. The bars give an obvious view of the similar effects of TPEN and EDTA with the bars for each chelator at each concentration being very even. C) This graph shows even bars for each chelator at each concentration once again. ***significant difference from the control with a p value of <0.001.

  • fig3a
    FIGURE 3. LC/A-424 ACTIVITY IS AFFECTED BY CHELATORS: A) This graph shows that the EDTA and TPEN had similar effects on the activity of the LC/A-424 between 0 and 60 minutes.
  • fig3b
    FIGURE 3. LC/A-424 ACTIVITY IS AFFECTED BY CHELATORS: B) This graph shows the toxin activity at 10 minutes. The bars give an obvious view of the similar effects of TPEN and EDTA with the bars for each chelator at each concentration being very even.
  • fig3c
    FIGURE 3. LC/A-424 ACTIVITY IS AFFECTED BY CHELATORS: C) This graph shows even bars for each chelator at each concentration once again.

 

Denaturants
The two denaturants tested were guanidinium chloride and urea. The guanidinium was found to cause complete inhibition at 300 mM and higher concentrations while a concentration of 2 M urea or above was needed to cause 100 % inhibition. At 100 mM guanidinium and 1 M urea there was around 50% inhibition. At 10 mM guanidinium and 300 mM urea there was no inhibition. It is also interesting to note that at 300 mM guanidinium there was complete inhibition while at 300 mM urea there was no inhibition (Table 3). These results reflect the overall trend of the activity as recorded on the graphs.

 

  • fig4a
  • fig4b
  • fig4c

FIGURE 4. LC/A 424 ACTIVITY IS AFFECTED BY DENATURANTS: A) This graph shows the similar effects of the two denaturants are differing conecntrations between 0 and 60 minutes. B) This graph shows the toxin activity at 10 min. The higher concentrations of each denaturant are the same as the reporter which shows no reduction in substrate and, therefore, full inhibition of the toxin. The lower concentrations are around the same as the control. C) This graph shows the toxin activity at 45 minutes. The higher concentrations are still the same as the reporter and the lower concentrations are still the same as the control.***significant difference from the control with a p value of <0.001.*significant difference from the control with a p value of <0.05

  • fig4a
    FIGURE 4. LC/A 424 ACTIVITY IS AFFECTED BY DENATURANTS: A) This graph shows the similar effects of the two denaturants are differing conecntrations between 0 and 60 minutes.
  • fig4b
    FIGURE 4. LC/A 424 ACTIVITY IS AFFECTED BY DENATURANTS: B) This graph shows the toxin activity at 10 min. The higher concentrations of each denaturant are the same as the reporter which shows no reduction in substrate and, therefore, full inhibition of the toxin. The lower concentrations are around the same as the control.
  • fig4c
    FIGURE 4. LC/A 424 ACTIVITY IS AFFECTED BY DENATURANTS: C) This graph shows the toxin activity at 45 minutes. The higher concentrations are still the same as the reporter and the lower concentrations are still the same as the control.

 

Discussion

Temperature             

BoNT, like most proteins, is thermally sensitive and its activity is, therefore, affected by drastic changes in temperature (6). In our temperature studies we found that at temperatures above 50 ͦC the toxin begins to be dramatically inhibited. At 60 ͦC and above the toxin is completely inhibited with no substrate being cleaved. All temperatures below 50 ͦC were around the same as the control with, at the most, about 30% inhibition of the toxin (Fig 1). This inhibition at high temperatures is most likely due to small structural changes in the active site of the enzyme (6). For most enzymes, there is a certain threshold above which the activity will be lost and the speed of the reaction will drop sharply due to temperature extremes (7). Once we knew that the activity was lost at a temperature of 60 ͦC, we ran an experiment to see if this denaturation could be reversed by cooling our plate back down to 37 ͦC. (Fig 2) It was found, though, that the denaturation of the protein could not be reversed simply by lowering the temperature back to 37 ͦC for 30 minutes. Most likely chaperone proteins would be needed to restore the activity of the toxin; however, it may be useful in future studies to see if activity could be restored by allowing it to cool down for a longer amount of time. Some other future studies in this area may involve finding the exact threshold above which the LC loses activity. In a study by Encinar et al in 1998, serotype A was found to denature at 50.5 ͦC, although it is not clear if the toxin was completely void of activity at this temperature and the threshold could be more precisely determined (8). 

Chelators
Chelators remove the metals from solution and the LC of BoNT contains a zinc cofactor in its active site necessary for the cleavage of SNAP-25. Our studies with EDTA and TPEN show that, although EDTA is a universal chelator and TPEN is a more zinc specific chelator, both display the exact same amount of inhibition at each concentration. It was found that at 20 uM and above both showed 100% inhibition of the toxin (Fig 3). This range makes sense as it was also shown in a study by Kalandakanond et al in 2000 that chelation of zinc by TPEN in the 20 uM to 30 uM range “significantly prolonged the onset of paralysis in the hemidiaphragm” of mice (9). Although this study was performed in vivo rather than in vitro, it still serves as an example of the chelation of zinc having a significant effect on the activity of botulinum. The fact that the TPEN was not any more efficient at chelating the zinc in the LC’s active site may be due to the fact that TPEN is lipid soluble, while EDTA is only water-soluble. Since TPEN is lipid soluble it is more often used in cell assays as an intracellular chelator. Since this is a cell free assay the TPEN and EDTA showed the same amount of efficacy at chelating the zinc. A possible future study in this area may be testing the restoration of activity after the chelation of the zinc. A study by Simpson et al in 2001 (10) showed that the stripped version of the toxin can rebind to exogenous zinc in the cytosol and regain complete activity. Experiments testing this ability to regain its zinc cofactor may be interesting and useful to run on the LC.

Denaturants
Denaturants are chemicals which disturb the weak bonds, such as hydrogen bonds, that hold the secondary structure of proteins together. When not folded properly, proteins lose their ability to function correctly. In our studies with denaturants we used guanidinium chloride and urea. The guanidinium was found to be a more potent denaturant showing full inhibition at 300 mM, while the urea required a concentration of 2 M or above to reach full inhibition. These results show a sharp contrast to a previous study of the same nature by Shantz et al in 1960. They found a loss of toxicity with urea between 3 and 4 M with complete inhibition at 6 M and above. For guanidinium, they found a loss of toxicity to begin at around 1 M with complete inhibition at 3 M or higher. This study did not use the isolated LC, which could explain the difference in results (11).  Another interesting result in our study was while the guanidinium show 100% inhibition at 300 mM, the urea showed no inhibition whatsoever at the same concentration (Fig 4). The greater potency of guanidinium is most likely due to the fact that guanidinium is a stronger base with more nitrogen groups than urea. It is also known to be a very strong chaotrope, so it would be more efficient at disturbing the weak bonds holding together the secondary structure of the protein. 

Significance
All three conditions tested in this study provide us with important knowledge of the toxin and its activity under the pressure of various environmental factors. This knowledge of the effects of temperature, chelators, and denaturants on the activity of the BoNT LC/A-424 can aid in the future development of drugs to cure botulism in humans.

References

1. Romano, Jr., J. A., & Lukey, B. & Salem, H. (2008). Chemical Warfare Agents Chemistry, Pharmacology, Toxicology, and Therapeutics Boca Raton, Florida: CRC Press Taylor & Francis Group.

2. Turton, K., & Chaddock, J. A., Acharya, K. R. (2002). Botulinum and tetanus neurotoxins: structure, function and therapeutic utility. Trends in Biochemical Sciences, 27(11), pp. 552-558.  SOURCE

3. Rossetto, O., & Pirazzini, M., & Montecucco, C. (2014). Botulinum neurotoxins: genetic, structural, and mechanistic insights. Nature Reviews Microbiology, 12, pp. 535-549.  SOURCE

4. Baldwin, M. R., & Barbieri, J. T. (2009). Association of botulinum neurotoxins with synaptic vesicle protein complexes. Toxicon, 54(5), pp. 570-574.  SOURCE

5. Ruge, D. R., & Dunning, M. F., & Piazza, T. M., & Molles, B. E., & Adler, M., & Zeytin, F. N., & Tucker, W. C. (2011). Detection of six serotypes of botulinum neurotoxin using fluorogenic reporters. Analytical Biochemistry, 411(2), pp. 200-209. SOURCE

6. Ibañez, C., & Blanes-Mira, C., & Fernández-Ballester, G., & Planells-Cases, R., & Ferrer-Montiel, A. (2004). Modulation of botulinum neurotoxin A catalytic domain stability by tyrosine phosphorylation. FEBS Letters, 578(1-2), pp. 121-127.  SOURCE

7. Campbell, N. A., & J. Reece, & L. Urry, & M. Cain, & S. Wasserman, & P. Minorsky, & R. Jackson. (2008).Biology San Francisco, California: Pearson Benjamin Cummings.

8. Encinar, J. A., & Fernández, A., & Ferragut, J. A., & González-Ros, J. M., & DasGupta, B. R., & Montal, M., & Ferrer-Montiel, A. (1998). Structural stabilization of botulinum neurotoxins by tyrosine phosphorylation. FEBS Letters, 429(1), pp. 78-82.  SOURCE

9. Kalandakanond, S., & Coffield, J. (2001). Cleavage of SNAP-25 by botulinum toxin type A requires receptor-mediated endocytosis, pH-dependent translocation, and zinc. The Journal of Pharmacology and Experimental Therapeutics, 296(3), pp. 980-986.  SOURCE

10. Simpson, L. L., & Maksymowych, A. B., & Hao, S. (2001). The role of zinc binding in the biological activity of botulinum toxin. The Journal of Biological Chemistry, 276, pp. 27034-27041.  SOURCE

11. Schantz, E. J., & Stefanye, D., & Spero, L. (1960). Observations on the fluorescence and         toxicity of botulinum toxin. The Journal of Biological Chemistry, 235(12), pp. 3489-3491.  SOURCE