Botulism is a rare, potentially fatal paralyzing illness. It’s the reason we shouldn’t feed infants honey and why we need to take care in consuming home-canned foods: they can potentially contain nerve-damaging toxins produced by Clostridium botulinum. Botulinum toxin is classified as one of the six most dangerous potential bioterrorism agents.
There are seven known types of botulinum toxin. Toxins A and B were first identified in 1919, and first purified in 1946 and 1947, respectively. (Both are also used medically.) Toxins C, D, E and F eventually followed. The last, toxin G, was identified in 1969 in soil bacteria in Argentina.
And that’s where it’s stood until now. But to truly defend against botulism, we need to know all the toxins made by the various C. botulinum strains, since each requires a separate antibody to neutralize it.
“For a long time, no new toxins have been found,” says Min Dong, PhD, an assistant professor in Boston Children’s Hospital’s Department of Urology and Harvard Medical School’s Department of Microbiology and Immunobiology. “We have found new subtypes, but not a totally new toxin. The question has been when we would find a one, and where to look for it.”
In 2013, a group in California had what seemed like a new toxin, type H, but it proved to be a false alarm: when the protein was eventually sequenced, it was found to be a combination of two existing toxins (a subtype of toxin F with a piece of toxin A).
Last week in Nature Communications, Dong and colleagues report the first new botulinum toxin to be found in close to 50 years. Provisionally called toxin X, it has some unusual properties that set it apart from the others.
“Sequence-wise it doesn’t look like any other toxin, and it cannot be recognized by antibodies to any other known botulinum toxin,” says Dong.
Reopening a cold case
The bacteria that produce toxin X had been isolated in 1990s in Japan. The strain, which had caused cases of infant botulism, was duly categorized, and its toxicity was attributed to toxin B. The bacterium was sequenced, and the sequence encoding toxin B was found.
That seemed to be the end of the story. “It was set aside,” says Dong.
But in 2015, another Japanese group sequenced the bacterium’s genome and put the sequence in a public database.
“What they missed within this genomic sequence was a piece that contains this new toxin gene,” says Dong.
Pål Stenmark at Stockholm University in Sweden first noticed this in a bioinformatics analysis. The new gene bore all the characteristics of encoding a functional toxin.
“We have been collaborating with Pål on structure-function of botulinum toxins for a long time,” says Dong. “He came to me with this information and we decided to join forces and categorize the toxin functionally.”
With postdoctoral fellow Sicai Zhang, PhD, leading the work, the researchers validated the toxin’s activity by assembling it artificially in the lab. “We decided to avoid generating the full-length active toxin gene, as introducing a toxin gene into organism or cellular system is always a significant biosafety concern,” says Dong. “Instead, we developed an approach to generate a limited amount of toxin in test tubes by joining two non-toxic fragments.”
This approach provided all the elements needed to understand how toxin X works. Jie Zhang, PhD, a senior scientist in Dong’s lab, was able to show that it causes paralysis in mice similar to other botulinum toxins.
The surprise did not stop here. In further studies, lab member Sicai Zhang, PhD, found that botulinum toxin X cleaves the same set of nerve proteins targeted by other botulinum toxins. But it also cleaves a group of proteins that none of other toxins touch.
“Type X has this unique capability to cleave VAMP4, VAMP5 and Ykt6,” elaborates Dong. “Some of these proteins are poorly characterized, so type X toxin will be a valuable tool for defining their functions.”
The additional targets could potentially endow toxin X with different properties when used medically. Botulinum toxins A and B are currently used for spasticity, chronic pain overactive bladder and removing wrinkles, to name a few applications. They work by cutting proteins in nerve endings that affect the secretion of neurotransmitters, in turn affecting neuron communication. The effects of cutting the additional proteins has yet to be explored.
“Can this new toxin add additional therapeutic benefit? This is an exciting question that we don’t have the answer to right now,” says Dong. “We also still don’t know the potency of this toxin. That needs to be studied as well, by a CDC-approved lab.”
To Dong, the process of discovery is as compelling as the results.
“Traditionally, you discover bacterial virulence factors by looking at infection consequences and finding the proteins and the genes,” he says. “In this case, the toxin was discovered by whole-genome sequencing of the bacteria. This illustrates the importance of genetics and bioinformatics approaches for understanding the microbial world.”