Anthrax toxin protective antigen protein (PA) binds to receptors on the surface of mammalian cells and transports two other toxin proteins, lethal factor (LF) or edema factor (EF) to the cytosol. EF is a potent calmodulin-dependent adenylyl cyclase that causes large increases in intracellular cAMP concentrations. LF is a metalloprotease that cleaves several mitogen-activated protein kinase kinases (MEKs). Anthrax lethal toxin (LT, the combination of PA and LF) is considered the primary virulence factor of B. anthracis, immunization against either of its components provides full protection against challenge with anthrax spores. Injection of this toxin or anthrax edema toxin (ET, the combination of PA and EF) into animals induces a unique vascular leakage in animal models in a manner similar to that seen in anthrax disease. Inbred mice and rats vary in the susceptibility of their macrophages to anthrax LT. Macrophages from certain strains either undergo a rapid caspase-1 dependent lysis (pyroptosis) while those from other rodent strains are completely resistant. We recently demonstrated that in rats, LT cleaves the inflammasome sensor NLRP1 within an 8 amino acid polymorphic region. Nlrp1 is a NOD-like receptor (NLR) protein which is part of the inflammasome, a multiprotein complex that activates caspase-1 in response to cytoplasmic danger signals. A consequence of inflammasome activation is macrophage death with concurrent IL-1beta/IL-18 maturation and release. We discovered that LT only cleaves rat Nlrp1 in rat strains with macrophages sensitive to pyroptosis and that the cleavage event was required for inflammasome activation. In the current year of 2013 we extended our studies to mice, where 3 different polymorphic Nlrp1b alleles exist. We found that mouse Nlrp1b proteins are also cleaved by LF. In contrast to the situation in rats, sensitivity and resistance of Balb/cJ (sensitive) and NOD/LtJ (resistant) macrophages was not determined by the sensitivity of their Nlrp1 proteins to cleavage by LF, as both proteins are cleaved. Two LF cleavage sites, at residues 38 and 44, were identified in mouse Nlrp1b proteins. Our results suggest that the resistance of NOD/LtJ macrophages to LT, and the inability of the inflammasome sensor in these mice to be activated by toxin is due to polymorphisms other than those at the LF cleavage sites. In 2013 we extended our analyses of mouse Nlrp1 proteins to study the expression and sequence of two paralogs, Nlrp1a and Nlrp1c which lie in immediate proximity to the mouse Nlrp1b gene. The high level of homology seen among the paralogs suggests they could potentially compete for the same binding partners involved in inflammasome activation. Survey of their expression in a large number of inbred mice having varying LT sensitivity showed that Nlrp1a was only expressed in LT-resistant macrophages, with the exception of the macrophages from the LT-sensitive Cast/EiJ mouse strain. Nlrp1c was only expressed in LT-resistant macrophages from five mouse strains. Unlike the highly polymorphic Nlrp1b, Nlrp1a was conserved across all inbred mouse strains. Interestingly, our studies also revealed novel splice variants of Nlrp1b which were simultaneously expressed in certain inbred mice. These findings reveal the complexity of Nlrp1 inflammasome locus in mice and also suggest that Nlrp1a has been conserved during evolution because it recognizes an important stimulus, one that remains to be identified. In other work done during 2013 with researchers at the FDA, we investigated the effects of LT in mice on endothelial cell junctions. LT-induced endothelial barrier dysfunction in vitro was shown to be temporally linked to a reduced expression of the tight junction protein claudin-5, with no changes observed in other tight junction or adherens junction proteins. LT-induced loss of claudin-5 was independent of cell death, dependent on toxin targeting of MEKs and occurred at the level of mRNA reduction. Mice challenged with LT also showed significant reduction in claudin-5 expression. These studies revealed that LT inhibition of claudin-5 expression may play a role in endothelial tight junction disruption which can impact vascular leakage and shock in animals. Finally, during 2013 we extended our understanding of ET action in vivo. We found that the activity of this adenylate cyclase is dependent on the N-end rule of proteasome degradation, where EF variants having different N-terminal residues can vary 100-fold in potency both in cells and in mice. EF appears to be highly sensitive to ubiquitination and breakdown. An understanding of the fate of ET in cells aids in explaining the highly variable capabilities of toxin preparations from different laboratories, as well as providing an example of the in vivo relevance of the N-end rule.
Showing the most recent 10 out of 39 publications
