Toxic behaviour: why do tuberculosis bacteria poison themselves?

Stealthy bacteria slow down their division when they invade the body to avoid drawing the immune system's attention. Mycobacterium tuberculosis, the world's leading bacterial infectious killer, takes a seemingly counterintuitive approach to that end. M. tuberculosis expresses self-toxins that damage its DNA and shut down growth as well as antitoxins to later help recuperate and resume proliferation. By studying these toxin-antitoxin pairs, Durham University microbiologist Professor Tim Blower aims to find ways to mimic the self-toxins with new therapeutics.

The six-piece complex consists of two toxin proteins (orange) and four antitoxin proteins (green). The light green antitoxins adopt a different shape from the dark green ones, even though they share the same underlying sequence. Image credit: Izaak Beck, Durham University.

By conducting X-ray crystallography at Diamond’s I04 beamline, Blower and his colleagues uncovered the structure of toxin-antitoxin complexes, providing insight into how they regulate DNA damaging activity. The findings reveal that the protein pair potentially form two types of complexes. A grouping of two toxins and four antitoxins dominated at body temperature, whereas an equal pairing of two and two were more common in colder conditions, which may reflect how the proteins come together when bacteria live in the environment. These findings change our perspective on how the toxins and antitoxins operate, bringing researchers closer to designing new drugs against a pandemic microbe that continuously evolves resistance to existing antibiotics. 

Each year, Mycobacterium tuberculosis leads approximately 10 million people to endure a bloody cough, exhaustion, and fever, and it causes over one million deaths. Doctors typically prescribe patients a course of four to six antibiotics to clear the infection, but the bacteria evolve mechanisms to resist the effects of the drugs. As many as 2.5 percent of tuberculosis patients carry variants of the bacteria resistant to the four most common first-line antibiotics, and that proportion is expected to climb if researchers don't develop other therapeutics that could kill resistant strains.    

Blower and the team from Durham University and Newcastle University study mechanisms the bacteria use to limit their own growth in pursuit of inspiration for new drug candidates. Specifically, they focus on an enzyme that controls DNA organisation in the cell and a pair of toxins and antitoxins that regulate this enzyme’s function. 

Bacteria and eukaryotes (for instance, humans), organise DNA in the cell differently. Eukaryotic DNA is tightly packaged in the nucleus by histone proteins that wind it up into compact chromosomes. Bacteria, on the other hand, lack histones and rely on DNA to undergo a process called supercoiling. Like how a wound-up rubber band contracts into a small volume, bacterial DNA winds up into a condensed coil to save space. However, supercoiled DNA needs constant maintenance, which involves occasional unwinding and rewinding of the molecules. To this end, an essential enzyme called DNA gyrase cuts the DNA, allows it to untwist, and glues the cut ends back together again, so they can coil again. 

Repairing the DNA breaks is essential to the bacteria’s survival because it avoids the build-up of harmful DNA damage and mutations, but sometimes M. tuberculosis interferes with the process. It achieves this using a toxin-antitoxin system that inhibits DNA gyrase. Scientists are still uncertain about the biological role of the toxin, Blower said, but one hypothesis is that by partially shutting down bacterial growth, it prevents antibiotics that target growth machinery from working. Another is that the toxin helps quiescent bacteria evade immune detection as slow-growing microbes tend to slip under the radar. The antitoxin relieves the bacteria, allowing those that survived the accumulation of DNA breaks to seal them back together and resume growth when conditions in the body become favourable. 

Researchers developing new therapeutics are drawn to these systems. Suggesting scientists could develop copycat drugs, Professor Blower said: 

Blower’s team began by purifying copious amounts of the toxin, antitoxin, and DNA gyrase. Confirming that the proteins still worked when isolated from the bacteria, they found that DNA gyrase successfully chopped and glued 90 percent of DNA strands in vitro, and when they added the gyrase-inhibiting toxin to the mix, the yield dropped to 50 percent. 

Having produced ordered crystals of the purified proteins, the researchers shipped their samples to Diamond to perform X-ray crystallography experiments at beamline I04, resolving the toxin and antitoxin to a resolution of 2.35 Ångströms — detailed enough to see the shape of the amino acids that make up the molecules. The crystal structure revealed how the antitoxin might counteract the toxin’s activity. At one end of the toxin protein lies a helical (corkscrew shaped) region that likely binds to and inhibits the DNA gyrase. The structure revealed that the antitoxin interferes with this helix. “It’s actually pushing that helix out of place so that the toxin can no longer be functional,” Blower said. The antitoxin also forms a complex with the toxin, sequestering it such that it cannot inhibit the enzyme. 

In similar systems, complexes carry an equal weight of toxin and antitoxin proteins; however, these complexes contained two toxins and four antitoxins, which came as a surprise to the researchers. One pair of antitoxins adopt a different shape from the other pair in the complex. “The same protein is folding differently, which is unusual,” Blower said. 

They next performed size exclusion chromatography to see if they could detect other types of complexes. This method separates protein complexes of different sizes by passing them through a column filled with beads: small complexes evade the beads and move through the column more quickly than larger ones. They found that two types of complexes trickled through the column. One was the complex that they resolved by crystallography and the other was a smaller complex that had an even split of two toxins and two antitoxins—what they initially expected to find. 

They wondered if the proteins change back and forth between the two complexes. Using in vitro experiments, they found that increasing the temperature drove the even complex to remodel into the unequal one with twice as many antitoxin proteins. One possible interpretation is that when M. tuberculosis is in a warm setting, such as inside the human body, rather than outside in the colder environment, it may accumulate more of the antitoxin in the complex. However, it’s not clear how this might affect the function of either the toxin or antitoxin. What’s more, these experiments were done using purified proteins rather than inside cells, so the researchers still need to determine which of the two complexes is the predominant form in M. tuberculosis and whether temperature-driven remodelling of the complexes occurs inside bacteria. 

“The next structural project is to see the toxin bound to the gyrase,” Blower said, allowing his team deeper insight into the important regions of the protein. “Then we can start designing drugs.” 

To find out more about I04 or discuss potential applications, please contact Principal Beamline Scientist Ralf Flaig: [email protected]. 

Beck IN, Arrowsmith TJ, Grobbelaar MJ, Bromley EHC, Marles-Wright J, and Blower TR. Toxin release by conditional remodelling of ParDE1 from Mycobacterium tuberculosis leads to gyrase inhibition. Nucleic Acid Research. 52(4),1909-1929 (2023). doi: 10.1093/nar/gkad1220 

Biotechnology and Biological Sciences Research Council Newcastle-Liverpool-Durham Doctoral Training Partnership studentship [BB/M011186/1 to I.N.B.]; Engineering and Physical Sciences Research Council Molecular Sciences for Medicine Centre for Doctoral Training studentship [EP/S022791/1] to T.J.A. and M.J.G.]; and by the Biophysical Sciences Institute at Durham University (to T.R.B.). Funding for open access charge: BBSRC funds from Durham University.