Inner Workings: How bacteria cause pain and what that reveals about the role of the nervous system
Amber Dance
Flesh-eating bacteria employ a devious strategy. When the bacteria, known as Streptococcus pyogenes, slip under the skin via a small cut or bug bite, they eat away at the tissue underneath, causing a condition called necrotizing fasciitis. One of the early, defining symptoms is that a minor wound or abrasion becomes intensely painful, far more painful than it ought to feel.
No one knew why this type of infection hurt so much until neuroimmunologist Isaac Chiu and colleagues at Harvard Medical School in Boston showed that S. pyogenes hijacks the body’s pain-sensing neurons, dialing up the misery. These neurons, in turn, communicate with the immune system, greatly reducing its response to the infection, according to research published last May in Cell (1). This communication creates a double whammy for the host: intense pain and a tough battle to fight off the bug. Approximately 700 to 1,200 people in the United States get necrotizing fasciitis from Streptococcus every year; one-third of them die, and many others suffer scarring and even amputations.
Over the past several years, Chiu and other researchers have started to build a picture of how certain bacteria, fungi, and viruses can ramp up pain — or in other cases numb their hosts. These interactions are complex: Sometimes they benefit the microbes, sometimes the host. But this work has already made one thing clear: There are three, not two, major players in infection. There’s the pathogen itself, the immune system, and the nervous system. And it’s the cross-talk among members of this trio that may often determine the body’s response to infection.
Researchers are just beginning to understand those conversations. But their discoveries about bacterial tactics are already suggesting new approaches to treating microbial infections.
Game Changer
Chiu’s team has “made huge discoveries in understanding what’s happening at the frontline of host defense against infection,” says Kevin Tracey, a neurosurgeon and president of the Feinstein Institute for Medical Research in Manhasset, NY, who studies neuroimmune interactions. “This is a game-changer.”
The work has resulted in a big shift in how researchers think about infection. A decade ago, conventional wisdom held that the nervous system operates separately from interactions between pathogens and the immune system. Biologists believed that infections hurt because of the inflammation — an immune-system response and call to arms — causing redness and swelling. “That literally was what the textbook said, and what everyone has thought, until Isaac tested it directly,” says Clifford Woolf, a neurobiologist at Boston Children’s Hospital in Boston and Chiu’s former postdoctoral advisor.
In a series of experiments they detailed in 2013, Chiu and Woolf revealed not only that bacteria were interacting directly with neurons to cause pain but also that the interaction had the effect of dampening inflammation (2).
Overturning Dogma
These early experiments by Chiu and Woolf focused on Staphylococcus aureus, which causes diverse infections in the skin, bone, and the lungs. Injecting a mouse’s paw with methicillin-resistant S. aureus (MRSA) caused the animal to act as if it was hypersensitive to pain: yanking back the paw when exposed to heat, cold, or a poke with a thin wire. But the instances when mice showed the highest pain sensitivity didn’t correlate with inflammation features such as swelling. Instead, the animals were most sensitive to pain when the bacterial load was highest.
Chiu tried the same experiment in mice lacking key genes in the immune system, as well in mice missing immune cells such as neutrophils, T cells, and B cells. MRSA was still painful. The immune system wasn’t necessary for the microbes to produce pain.
So what was causing the pain? Normally, when pain-sensing neurons called nociceptors encounter something noxious such as heat or acid, they open up channels in their membranes. Positive ions flow in, changing the cell’s membrane potential and making the neuron fire. That sends a pain signal to the spinal cord and brain.
Chiu discovered that S. aureus α-hemolysin, a pore-forming toxin, pokes holes in nociceptor membranes. Those pores let in positive ions, mimicking the process by which neurons depolarize and causing the cells to fire. He also found that the nociceptors had receptors that directly recognized bacterial molecules called formyl peptides as noxious and fired pain signals in response.
Moreover, when Chiu infected mice lacking nociceptors, their paws swelled more with incoming neutrophils and other immune cells, indicating a stronger immune response. The researchers formulated an explanation: The nociceptors, when activated, normally release compounds that stifle the immune system. Without that suppression, immune cells were flooding in to do their work. “What we found was very surprising,” recalls Chiu. “It suggests that nerves actually are very similar to immune cells…they’re poised to detect bacteria, or damaging molecules from bacteria.”
Three-Way Chatter
Around the same time that Chiu was zeroing in on these intriguing phenomena in S. aureusinfections, other scientists began reporting interactions among different pathogens and the nervous and immune systems. Multiple groups, for example, found that the common bacterial component lipopolysaccharide can activate nociceptors much as the S. aureus formyl peptides do (3⇓–5).
Effects on the host vary by microbe, however. When nociceptors detect fungal infections caused by Candida albicans, the recognition causes the host’s nociceptors to respond defensively. They release a neuropeptide called CGRP that can be pro- or antiinflammatory, depending on the circumstances (6). With Candida, CGRP amps up the host’s immune response, helping resist the fungus (7).
But another microbe’s interaction with neurons can also blunt pain, rather than causing it, perhaps buying the intruder extra time to infect undetected. Skin lesions caused by Mycobacterium ulcerans, called Buruli ulcers, are famously painless, especially during the early stages of infection. Doctors had assumed that a M. ulcerans toxin, mycolactone, destroyed pain-sensing nerves. But researchers at Inserm in Angers and Lille, France determined in 2014 that mycolactone actually changes the nociceptor membrane potential so the cells can’t fire (8).
Focus on Flesh-Eating Bacteria
Building on these findings, Chiu’s postdoc Felipe Ribeiro decided to investigate S. pyogenes, work that would result in the 2018 Cell article (1). Ribeiro selected S. pyogenes because it causes a stinging sensation well before it has eaten away at tissues. “That was, for us, an indicator that there was a huge neuronal activation there,” says Ribeiro.
Working in mice, Ribeiro found that S. pyogenes’s pore-forming toxin, streptolysin S, activated nociceptors. Bacteria missing the toxin didn’t cause damaging skin lesions and disappeared on their own. Similarly, mice lacking nociceptors developed small lesions and recovered quickly.
The explanation, Ribeiro discovered, was that nociceptors activated by streptolysin S release CGRP, which in this circumstance dampens immune responses. Without the toxin, or without nociceptors for the toxin to act on, the bacteria couldn’t repress inflammation, and the host’s immune response won the day.
But location mattered when it came to interfering with the microbe’s pain-causing mechanisms, as Ribeiro found via experiments using botulinum neurotoxin A. The toxin prevents neurons from releasing vesicles full of CCRP and other signaling molecules — but doesn’t prevent neurons from firing. When Ribeiro injected the neurotoxin under the skin, mice had smaller abscesses; blocking CGRP produced a similar effect. But the mice still acted like they were in pain. If, however, Ribeiro injected the neurotoxin directly into the spinal cord, it blocked pain signals transmitted onward by vesicle release. The mice were no longer bothered, but they did develop large skin lesions, indicating that the bacteria were still able to stifle the immune response.
“It was really potent. The QX-314 basically completely blocked pain.”
— Isaac Chiu
The results neatly separated the two effects of streptolysin S: to cause pain directly by making nearby neurons fire and to repress immunity by promoting the release of immunosuppressive molecules such as CGRP (1).
Taking on Microbial Tactics
It’s early days for research on communication among pathogens, neurons, and the immune system, but already the work has hinted at new therapeutic approaches. For people with necrotizing fasciitis, for instance, the ideal treatment would block the pain while allowing the immune response to pursue the bacteria. Chiu hopes botulinum neurotoxin A or drugs that act on CGRP could treat not only rare necrotizing fasciitis but also more common skin infections.
Chiu and his graduate student Kimbria Blake have also experimented with exploiting bacterial tactics to treat infection-related pain. They figured that if the bacteria were making additional holes in nociceptors, the researchers could use those same pores to insert an anesthetic.
They tried it with QX-314, a lidocaine derivative that can specifically target nociceptors without affecting other neurons. Woolf and his colleagues have been working to develop the drug as a targeted painkiller since 2007 when they showed that using the noxious pepper derivative capsaicin to open up channels in nociceptors allowed QX-314 to squeeze into those same channels to numb pain (9). QX-314 won’t fit through the smaller channels of other types of neurons — and once the drug is inside a nociceptor, it’s stuck, causing the painkilling effects to last for hours.
In a study published last year (10), Blake showed that QX-314 could also get into nociceptors by taking advantage of the holes punched by bacteria. In mice with S. aureus, Woolf’s drug worked much better than lidocaine or ibuprofen. “It was really potent,” says Chiu. “The QX-314 basically completely blocked pain.” And while some painkillers can interfere with immune responses to infection, Blake saw no evidence that QX-314 did so.
Microbes such as Staphylococcus and Streptococcus, Candida, and Mycobacterium may be just the first wave of pathogens found to talk directly to both the nervous and immune systems. Chiu is now investigating others, including viruses such as influenza. And Tracey notes that the toxins involved in microbial modulation of the host’s pain are made by a wide variety of bacteria. That makes him think there will be more than just a few pathogens that tweak pain and immune responses by interacting with neurons. “I predict it’ll be many,” he says. “Many.”
Published under the PNAS license.
