doi:10.1016/j.bmc.2004.10.026 
Published by Elsevier Ltd.
Conformational sampling of the botulinum neurotoxin serotype a light chain: implications for inhibitor binding
James C. Burnetta, James J. Schmidtb, Connor F. McGratha, Tam L. Nguyena, Ann R. Hermonea, Rekha G. Panchala, Jonathan L. Vennerstromc, Krishna Kodukulad, Daniel W. Zaharevitza, Rick Gussioa, 
, 
and Sina Bavarib, ,
aDevelopmental Therapeutics Program, NCI Frederick, Frederick, MD 21702, USA
bUS Army Medical Research Institute of Infectious Diseases, Frederick, MD 21702, USA
cCollege of Pharmacy, University of Nebraska Medical Center, 986025 Nebraska Medical Center, Omaha, NE 68198, USA
dOffice of Technology and Strategy, Innovation and Acquisitions, Sarnoff Corporation, 201 Washington Road, Princeton, NJ 08543, USA
Received 19 August 2004;
revised 9 October 2004;
accepted 9 October 2004.
Available online 10 November 2004.
References and further reading may be available for this article. To view references and further reading you must
purchase this article.
Abstract
Botulinum neurotoxins (BoNTs) are the most potent of the known biological toxins, and consequently are listed as category A biowarfare agents. Currently, the only treatments against BoNTs include preventative antitoxins and long-term supportive care. Consequently, there is an urgent need for therapeutics to counter these enzymes––post exposure. In a previous study, we identified a number of small, nonpeptidic lead inhibitors of BoNT serotype A light chain (BoNT/A LC) metalloprotease activity, and we identified a common pharmacophore for these molecules. In this study, we have focused on how the dynamic movement of amino acid residues in and surrounding the substrate binding cleft of the BoNT/A LC might affect inhibitor binding modes. The X-ray crystal structures of two BoNT/A LCs (PDB refcodes = 3BTA and 1E1H) were examined. Results from these analyses indicate that the core structural features of the examined BoNT/A LCs, including α-helices and β-sheets, remained relatively unchanged during 1 ns dynamics trajectories. However, conformational flexibility was observed in surface loops bordering the substrate binding clefts in both examined structures. Our analyses indicate that these loops may possess the ability to decrease the solvent accessibility of the substrate binding cleft, while at the same time creating new residue contacts for the inhibitors. Loop movements and conformational/positional analyses of residues within the substrate binding cleft are discussed with respect to BoNT/A LC inhibitor binding and our common pharmacophore for inhibition. The results from these studies may aid in the future identification/development of more potent small molecule inhibitors that take advantage of new binding contacts in the BoNT/A LC.
Graphical abstract
Molecular dynamics simulations were used to explore how residue motion in and around the botulinum neurotoxin serotype A light chain (BoNT/A LC) substrate binding cleft might affect inhibitor binding. Results from these studies indicate that surface loop reorientations toward the substrate binding cleft may facilitate small molecule binding by creating additional inhibitor-residue contacts. Based on molecular docking studies, our common pharmacophore model for BoNT/A LC inhibitors has been refined via the inclusion of these potential contact residues.
Keywords: Bioterrorism; Botulinum neurotoxin; Drug discovery; Inhibitors; Molecular dynamics; Molecular modeling; Pharmacophore; Metalloprotease
Abbreviations: BoNT, botulinum neurotoxin; BoNT/A, botulinum neurotoxin serotype A; BoNT/B, botulinum neurotoxin serotype B; LC, light chain; HC, heavy chain; Rmsd, rms deviation
Figure 1. Two-dimensional structures of BoNT/A LC metalloprotease inhibitors michellamine B and Q2-15. Michellamine B potency: 62% inhibition at 20 μM concn; Q2-15 potency: 60% inhibition at 20 μM concn.
Figure 2. Comparisons of BoNT/A LC models. (a) All backbone atoms superimposition of the BoNT/A LCs from X-ray crystal structures PDB refcodes = 3BTA (red) and 1E1H (blue); (b) all backbone atoms superimposition of the 3BTA BoNT/A LC X-ray crystal structure (green) and its dynamics average structure (magenta). Loops 1–3 are shown as ribbons; (c) all backbone atoms superimposition of the 1E1H BoNT/A LC X-ray crystal structure (green) and its dynamics average structure (magenta). Loops 1–3 are shown as ribbons.
Figure 3. (a) BoNT/A LC inhibitor Q2-15 docked in the molecular dynamics model for the 1E1H BoNT/A LC. Enzyme atom colors: subsite 1 carbons (light blue); subsite 2 carbons (magenta); polar contact region carbons (orange); all other carbons (green); oxygen (red); nitrogen (blue); sulfur (yellow). Q2-15 carbons are white and chlorine atoms are light green. BoNT/A LC loops 1, 2, and 3, as well as the side chains of specified residues of these loops are shown in thicker stick. Residues with brown stripes are new contacts that are observed when docking Q2-15 in the dynamics BoNT/A LC (as opposed to contacts made by Q2-15 docked in the molecular mechanics refined X-ray structure); (b) BoNT/A LC inhibitor Q2-15 docked in the molecular mechanics refined X-ray crystal structure of the 1E1H BoNT/A LC. All colors and stick thickness are as described for (a). A comparison of (a) and (b) shows how loop 1 reorientation provides additional contacts for the inhibitor.
Figure 4. (a) Michellamine B docked in the molecular dynamics structure for the 3BTA BoNT/A LC. All colors and stick thickness are as described in the caption for Figure 3a; (b) michellamine B docked in the molecular mechanics refined X-ray crystal structure of the 3BTA BoNT/A LC. All colors and stick thickness are the same as described in Figure 3a. A comparison of (a) and (b) shows that their are more favorable electrostatic and hydrophobic contacts between michellamine B and the BoNT/A LC dynamics structure, and that these interactions are the result of loop reorientations toward the substrate binding cleft.
Figure 5. A proposed binding site for silver ion in the dynamics structure of the 3BTA BoNT/A LC. Oxygen atoms are red and nitrogen atoms are blue. The silver ion is shown as a light blue sphere and the zinc ion is a magenta sphere. All other atoms are green. Loops 1, 2, and 3 are shown in thicker stick. Loop 1 reorientation partially shields the polar contact region from solvent, creating a pocket that may potentially trap a silver ion.
Figure 6. Refined pharmacophore for BoNT/A LC inhibition. Planar components A and B are blue dashed rectangles. The dashed circle in plane A represents a heteroatom. Hydrophobic components C and D are shown as light blue circles. The positive ionizable component E of the pharmacophore is shown as a red circle. Residues that remained consistent when docking inhibitors in predicted binding subsites of both dynamics and molecular mechanics ‘only’ refined models are shown as gray spheres. Residues E63, V67, and E170 are shown as a gray spheres with dashed black boarders––to indicate that these amino acids were found to participate when docking inhibitors in dynamics structures.
Table 1.
Comparisons of BoNT/A LC models
a All superimpositions were performed using backbone atoms.
b L.E. confs. = lowest-energy conformers from the trajectory.
c H.E. confs. = highest-energy conformers from the trajectory.
d The average rmsd for 100 L.E. confs. from the 3BTA dynamics trajectory compared with the average structure from the same trajectory (rmsd range = 0.64–0.94 Å).
e The average rmsd for 10 H.E. confs. from the 3BTA dynamics trajectory compared with the average structure from the same trajectory (rmsd range = 0.70–0.99 Å).
f The average rmsd for 100 L.E. confs. from the 1E1H dynamics trajectory compared with the average structure from the same trajectory (rmsd range = 0.61–0.94 Å).
g The average rmsd for 10 H.E. confs. from the 1E1H dynamics trajectory compared with the average structure from the same trajectory (rmsd range = 0.63–0.79 Å).
Table 2.
Torsional ranges for potential inhibitor contact residues shown in Figure 6
Abstract
Purchase PDF (918 K)
E-mail Article
Add to my Quick Links
Cited By in Scopus (11)
Purchase PDF (360 K)
150-kDa single-chain protein, is nicked proteolytically either endogenously or exogenously. The 
