Vaccines and Therapeutics for Anthrax

National Institute Of Allergy And Infectious Diseases

Investigators

Linked publications & trials

Abstract

Our long-term basic research on anthrax vaccines has included work on next generation candidates that induce antibody to the poly-gamma-D-glutamic acid (PGA) capsule as well as to the protective antigen protein (PA). A candidate conjugate vaccine developed in collaboration with Drs. John Robbins and Rachel Schneerson (NICHD) in 2003 consists of PGA peptides chemically conjugated to PA. This candidate was later licensed by a local company, Biologic Resources LLC. NIAID awarded this company a 5-year contract for development of this vaccine, and the data needed for submission of an IND was collected. We supported the company throughout this work and a joint publication is being prepared. The PA protein forms a channel in endosomal membranes through which LF and EF translocate to the cytosol. This system can translocate heterologous peptides to the cytosol when these are linked to the N-terminal domains of either LF (LFn) or EF. Translocation through the PA channel requires that polypeptides unfold and then refold, events that depend on properties of the PA channel. During FY2021, we sought to characterize a poorly understood behavior of the PA channel seen when it is studied in artificial lipid membranes. The ion current flowing through this channel shows 1/f noise, meaning that there are fast, voltage-independent current interruptions between conducting and nonconducting states. Notably, similar current fluctuations are observed in the channel-forming components of the clostridial binary C2 and iota toxins, which share functional and structural similarities with the PA channel. Like PA, these toxins have evolved to translocate the enzymatic components of the toxins into the cytosol. Using high-resolution single-channel lipid bilayer experiments and all-atom molecular dynamic simulations, we deduced that the 1/f noise in the PA channel occurs because of hydrophobic gating at the psi-clamp region. This psi-clamp is a narrow hydrophobic ring in the PA channel lumen formed by seven or eight phenylalanine residues, a structure that facilitates protein translocation. Notably, the 1/f noise is not seen in a PA mutant having the ring-forming phenylalanine residue replaced by alanine. This mutant also cannot translocate peptides to the cytosol of cells, indicating that detection of the1/f noise is indicative of a functional channel. This lab has developed the anthrax toxin delivery system as a platform for targeting solid tumors. This system employs a unique strategy to achieve tumor specificity, one that relies on the over-expression of urokinase plasminogen activator (uPA) and matrix metalloproteases (MMP) in the tumor environment. Replacing the furin cleavage site of PA, RKKR, with sites cleaved by uPA, MMPs, or other proteases, endows PA with strong anti-tumor activity in mouse models when administered with an effector. In earlier work, we found that this drug is host-directed, acting specifically on the abnormal, rapidly growing endothelial cells that feed tumors, while having little effect on normal, quiescent blood vessels. The consequence is that this therapy will be indifferent to the tumor type, its receptor repertoire, genetic heterogeneity, or variation over time and space. During FY2021, we continued to assemble pre-clinical data to validate this approach to cancer therapy. In this case, we focused on osteosarcomas (OSA), which have similar properties in dogs and humans, and which are difficult to treat in both. Two OSA cell lines (canine D17 and human MG63) and a non-neoplastic canine osteoblastic cell line (COBS) were studied. Cells were treated with different concentrations of the recombinant engineered anthrax toxin and cell viability, cell cycle, apoptosis, and necrosis were analyzed. D17 and MG63 cells had significantly decreased viability after 24 h of treatment, and were apoptotic, whereas COBS cells arrested in the G1 phase. These results demonstrate the inhibitory effects of the reengineered anthrax toxin on OSA cells and suggest it could be considered as a new therapy for OSA. Bacilli like B. anthracis have the genetic potential to produce many virulence factors, although not all are expressed. The closest B. anthracis relative, Bacillus cereus, produces two hemolysins that we have studied in recent years, hemolysin B (HBl) and non-hemolytic enterotoxin (Nhe). (The latter name is a misnomer since the protein is hemolytic.) We have produced these proteins in large amounts in the avirulent B. anthracis strain BH500, thereby enabling characterization of their interaction with target cells, which include activation of inflammasomes, and identification of an unexpected protein receptor for HBl. HBl and Nhe are unusual in that they each contain three rather similar proteins that assemble sequentially on the surface of cells to produce a large lytic pore. Interestingly, many bacteria produce single component hemolysins where the protein structure is very similar to that of all six HBl and Nhe proteins, yet they act as a single component to make lytic pores. In FY2021, as a first step to determining the mechanism of pore assembly and the structures of the pores formed by these three-component hemolysins, we worked with a collaborator to obtain an X-ray crystal structure of the HBl L1 component. The structure obtained confirms L1 to be a member of the alpha-helical pore-forming toxin family. In comparison to other members of this group, it has an extended hydrophobic beta tongue region that may be involved in pore formation.