Structural Basis for Activation of Cholera Toxin

Cholera is a serious disease that claims thousands of victims each year in third-world, war-torn, and disaster-stricken nations. The culprit is the bacterium Vibrio cholerae, which can be ingested through contaminated food or water and colonizes the mucous membrane of the human small intestine. There, it secretes cholera toxin (CT), a protein whose A1 subunit (CTA1) triggers a series of events culminating in the massive efflux of electrolytes and water into the intestinal cavity, causing the watery diarrhea characteristic of cholera that can lead to severe dehydration and death. Crystal structures of the CTA1 subunit in complex with its activator molecule (ARF6) reveal that binding of the ARF6 "switch" elicits dramatic changes in CTA1 loop regions, exposing the toxin's active site. The extensive CTA1–ARF6 interface mimics recognition of ARF6's normal cellular protein partners, which suggests that the toxin has evolved to exploit the molecular switch's promiscuous binding properties.

Cholera in the 21st Century

We are still living, as it turns out, in the time of cholera. There have been seven cholera pandemics since we began keeping track in the early 1800s. The ongoing seventh pandemic originated in Indonesia in 1961 and spread rapidly through Asia, Europe, and Africa; it finally reached the New World in 1991, causing over 4000 deaths in 16 Latin American countries that year. A possible eighth pandemic may be incubating in Bangladesh, where a new strain of the bacterium V. cholerae was identified in 1992. The risks are greatest in impoverished, overcrowded, or refugee communities characterized by poor sanitation and an unsafe water supply. However, recent tsunamis, hurricanes, and earthquakes have demonstrated quite graphically that even well-developed areas are not immune to the dangers of a compromised water supply. In the 19th century, the first epidemiologists realized that cholera outbreaks could be traced to a single source of contaminated water. From this simple insight flowed many advances in medicine and improvements in public health. In the 21st century, the world is more aware than ever of the need for similar epiphanies about the molecular mechanisms employed by persistent, resistant scourges like cholera.

The complex between bacterial CTA1 (gray) and human ARF6 (yellow). Important loop regions are delineated in gold (activation loop) and red (active-site loop).

After a molecule of cholera toxin binds to an epithelial (lining) cell of the human small intestine, CTA1 can enter the cell via several organellar trafficking systems. Once inside, CTA1 interacts with a protein known as adenosine diphosphate (ADP)–ribosylation factor 6 (ARF6), which enhances the activity of the toxin. Almost thirty years ago, ARFs were originally discovered and defined as activators of CTA1. They have since been revealed to also play essential roles in the trafficking of vesicles within cells during normal physiological conditions. While this multispecific binding strategy gives ARF proteins broad flexibility, it also leaves them vulnerable to interaction with unintended partners.

CTA1 before (left) and after (right) binding to ARF6-GTP (not shown). ARF6-GTP binding causes conformational changes in the activation loop (gold) and the active-site loop (red), leading to exposure of the active site (green).

Several crystal structures of both cholera toxin and ARFs have been published; however, it is not clear how these two proteins interact, and more interestingly, how ARF interaction enhances the cholera toxin's activity. To investigate, researchers from the Howard Hughes Medical Institute, the University of Washington, and the University of Colorado Health Sciences Center turned to ALS Beamline 8.2.2 to determine the 1.8-Å-resolution crystal structure of a CTA1 variant bound to a complex of human ARF6 and guanosine triphosphate (GTP), a mediator of the interaction.

CTA1 makes two large conformational changes when interacting with ARF6-GTP, both in loop regions indicated by previous structures to be flexible. The CTA1 activation loop changes from a structured loop to an amphipathic (polar) helix to make direct contacts to ARF6-GTP. Most strikingly, the active-site loop, which normally occludes the active site and prevents the binding of substrate (target) molecules, swings out of the active site when CTA1 is bound to ARF6-GTP, exposing areas along the active-site cleft implicated in substrate binding. The active-site loop does not directly contact ARF6-GTP; instead, the presence of the activator is communicated from a distance by several contacts with the activation loop in its altered conformation. The structure thus implies that ARF6 acts as an allosteric activator of CTA1; that is, it forces the active site open by binding at a site other than the enzymatically active one. This conclusion was confirmed by soaking CTA1:ARF6-GTP crystals with nicotinamide adenine dinucleotide (NAD⁺), the substance that binds to the active site. The resultant structure was solved at Beamline 19-ID of the Advanced Photon Source and provides the first glimpse of substrate binding by CTA1.

The CTA1 active site (green) occupied by NAD⁺ (sticks), viewed from above. ARF6-GTP (yellow) binds to CTA1 (gray) far from the active site. The knob formed by the active-site loop (red) when CTA1 is ARF6-bound and the ADP-ribosylating turn-turn (ARTT) motif (brown) form a surface for potential recruitment of the toxin's target protein.

Further, the open active-site loop conformation creates a knob on the surface of CTA1 near a motif (the "ADP-ribosylating turn-turn" motif) known in functionally related toxins to be involved in binding to the human protein targeted by the toxin. This suggests that, in addition to activation, interaction with ARF6 also generates a surface on CTA1 enabling the toxin to recognize its human protein target. Together, these results elegantly solve the long-standing question of how the human ARF6 protein activates CTA1, while the substrate-bound structure also provides a starting point for structure-based drug design specifically aimed at cholera.

Research conducted by C.J. O'Neal (University of Washington), M.G. Jobling and R.K. Holmes (University of Colorado Health Sciences Center), and W.G.J. Hol (Howard Hughes Medical Institute and University of Washington).

Research funding: National Institutes of Health and Howard Hughes Medical Institute. Operation of the ALS and the APS is supported by the U.S. Department of Energy, Office of Basic Energy Sciences.

Publication about this research: C.J. O'Neal, M.G. Jobling, R.K. Holmes, and W.G.J. Hol, "Structural basis for the activation of cholera toxin by human ARF6-GTP," Science 309, 1093 (2005).