Norwegian version of this page

Bacterial toxins

Bacterial infections have historically killed more people than any other disease and the annual burden of diarrheal diseases alone is estimated to 3 million deaths, according to the World Health Organization (WHO). Cholera, which is caused by the human pathogen Vibrio cholerae, gives rise to the most severe form of secretory diarrhea, whereas the most frequently isolated enteropathogen is enterotoxigenic Escherichia coli (ETEC), accounting for 210 million diarrhea episodes annually.

Surface representation of the cholera toxin B-pentamer. The blood group A antigen analog GalNAca3(Fuca2)Galb4(Fuca3)Glcb is shown in blue/yellow stick representation. A model of the GM1 pentasaccharide (green) is included for comparison.

About the project

The bacterial toxins cholera toxin (CT) and heat-labile enterotoxin (LT) produced by enterotoxigenic Escherichia coli are the causative agents of these diarrheal diseases. The two toxins are structurally as well as functionally related and consist of one enzymatically active A-subunit, non-covalently anchored in the center of a homo B-pentamer.

The non-toxic B pentamers, CTB and LTB, respectively, carry lectin domains responsible for binding the toxin to the glycoconjugate receptors present on the target cells. Cholera toxin and Escherichia coli heat-labile enterotoxin share the same primary receptor, the GM1 ganglioside. However, despite the extensive similarities (including ca. 85% sequence identity), these two toxins exhibit distinct ligand specificities, with LT being much more promiscuous than CT.


In order to find out which of the amino acids are responsible for the broader binding specificity of heat-labile enterotoxin, our collaborators constructed hybrids between the two carbohydrate-binding domains and tested their binding specificities. In this way, they serendipitously created a CTB/LTB hybrid, which exhibited a novel binding specificity to blood group A and B determinants (Ångström et al., 2000).


Docking blood group determinants to this hybrid strongly suggested a novel and distinct binding site located at the interface between two B subunits. The crystal structure (at 1.9 Å resolution) indeed revealed the presence of a second binding site (Holmner et al., 2004). Unexpectedly, however, this binding site exhibited only a single amino acid substitution in the blood group binding region, compared to the native LTB structure. Inspired by this knowledge, we created a single mutant of LTB (S4N), which exhibited the same binding characteristics as the hybrid (Holmner et al., 2004), and we showed that even native LTB contains this binding site (Holmner et al., 2007). The binding site is probably biologically relevant, as there are correlations between the severity of cholera infections and the blood group of infected individuals, which can be explained on the basis of the crystal structure (Holmner et al., 2007 and 2010).

Cholera toxin and blood-group dependence

Cholera has long been known to be a blood-group dependent disease, with blood group O individuals experiencing more severe symptoms. By studying the link between the cholera toxin and human blood groups, we are working towards improving the current cholera treatments. Today's vaccine is not optimal and there are no effective drugs against the disease. Blood group is determined by molecules expressed on the surface of red blood cells. In addition, these are also expressed on intestinal cells, as well as in the mucus layer lining our intestine. Since the cholera toxin exerts its effect in the intestine, we are focusing on these blood group molecules. Using X-ray crystallography, we have solved 3D structures of the cholera toxin binding to blood group determinants (Heggelund et al., 2016). In addition, we have used SPR to determine the binding strengths of the interactions. The blood group O determinant binds stronger than blood group A or B, leading us to the following working hypothesis: After the toxin is produced in the intestine, it might bind to the blood group molecules close to the cell surface, and if you have blood group O more toxins will be able to reach your intestinal cells and you get more sick. These results can be used to tailor treatments based on a patient's blood group, or for making a new drug that for example blocks the toxin binding to the intestinal cells. 


Heggelund et al. (2016) PLoS Pathogens 12(4): e1005567

Heggelund (2015) Cholera toxin – carbohydrate interaction as the basis for cholera blood-group dependence. PhD thesis.

Vasile et al. (2014), Glycobiology 24 (8), 766-778.

Heggelund et al. (2012), BBRC 418, 731-735. 

Holmner et al. (2011), J. Mol. Biol. 406, 387-402. 

Holmner et al. (2010), FEBS Lett. 584, 2548-2555. 

Holmner et al. (2007), J. Mol. Biol. 371, 754-764. 

Holmner et al. (2004), Structure 12, 1655-1667; Erratum in: Structure (2007) 15, 253. 

Ångström et al. (2000), J. Biol. 275, 3231-3238.



Published Dec. 29, 2010 12:39 PM - Last modified Apr. 26, 2016 4:57 PM