Back in March, Michigan’s Covid-19 cases exploded — leaping from zero to 3,657 in just two weeks. Detroit’s three big automakers closed factories temporarily, and the state’s largest health care system warned it was reaching capacity.
In the midst of this crisis, Joseph Roche, an associate professor in the physical therapy program at Wayne State University, had an idea.
From his research into muscular dystrophies, Roche understood that inflammation can do significant damage to the body. When he read that in severe Covid-19 cases, runaway inflammation was causing damage to tissues and organ failure, he dove into the data as well as older research on SARS.
Initially, it appeared that the virus might cause immune cells to overproduce molecules called cytokines, causing a severe inflammatory response known as a cytokine storm. But what Roche suspected as he sifted through early case studies was that it wasn’t the immune system’s cytokines causing so much of the damage but an entirely different pathway in the circulatory system knocked off balance by the virus: bradykinin signaling.
He believed that an accumulation of two peptides, des-Arg(9)-bradykinin, abbreviated to DABK, and bradykinin — both part of a system that regulates blood pressure and other functions — were starting a feedback loop of inflammation and tissue injury. By stopping this reaction, he argued in an open letter to the scientific community in April and in a May paper published in the Journal of the Federation of American Societies of Experimental Biology, doctors could prevent some of Covid-19’s worst effects.
Several months later and 500 miles away, a group of researchers unaware of Roche’s work started feeding the world’s second-fastest computer data from about 17,000 genetic samples from 1,300 Covid-19 patients. The team, based at the Oak Ridge National Laboratory in Tennessee, asked the $200 million computer to look for patterns in how Covid-19 was changing genes and impacting different systems in the body.
After almost a week of data crunching, the supercomputer landed on something they found surprising: bradykinins. “I was literally at home on a Sunday afternoon looking at different visualizations, and it just jumped out at me,” Daniel Jacobson, a computational systems biologist at Oak Ridge, says.
He calls these haywire reactions a “bradykinin storm,” and like Roche, believes they may help researchers treat severely ill Covid-19 patients, possibly staving off damage to organ systems or even preventing deaths. Outside researchers agree: Elements of the supercomputer’s analysis have been corroborated since it was published in July, and researchers say it could help lead the way to more effective treatments.
Here’s a deep dive into what has been published on bradykinin signaling since the pandemic began, and what we know about how this compound might be instigating some of the worst Covid-19 damage.
Why bradykinin signaling might be making Covid-19 so much worse
How Covid-19 can prompt an inflammatory cascade gets complicated, but Roche and other experts now think bradykinin might be the key to the vascular changes, lung damage, and even neurological symptoms the disease can cause.
The virus usually enters the body through the airways and lands on cells, where a protein called ACE2 functions as a doorway. As the virus replicates in the body, it finds other cells that have ACE2 receptors, such as those in the lungs, hearts, intestines, kidneys, and brain.
“The virus not only uses ACE2 as an entryway into cells but also tells that cell’s nucleus to start reducing ACE2 expression,” Roche says. This causes an accumulation of an enzyme called DABK, which creates conditions for inflammation.
This is where bradykinin might come in. When the virus binds with ACE2 receptors, DABK piles up, and bradykinin levels increase—causing an inflammatory cascade. “It creates a vicious feedback loop,” Roche says, amplifying inflammatory processes, including producing more cytokines.
Scientists initially thought that Covid-19 caused the immune system to release an overwhelming flood of cytokines — as often happens in response to a viral infection. In fact, promising treatments like remdesivir lower cytokine production. But recent evidence suggests that Covid-19 patients may not have particularly elevated levels of cytokines compared to people critically ill with other respiratory conditions, and other interventions attempting to lower cytokine production failed to reduce mortality — suggesting something else is going on.
That something, says Jacobson, might be a bradykinin storm instead. This hypothesis fits with a surprising number of Covid-19’s bizarre symptoms.
Researchers have observed many vascular symptoms, but previously blamed cytokine storms’ inflammation or direct damage from the virus. But bradykinin can impact how your blood coagulates — possibly explaining the strange clotting problems reported in Covid-19 patients and the high percentage of Covid-19 deaths from heart attacks, strokes, and deep vein thrombosis. As the virus causes bradykinin to accumulate in the cells it has hijacked, it makes your blood vessels permeable, letting your blood leak out. This could also explain the “Covid toes,” that have been linked to blood circulation.
In the lungs, increasing gaps in the cells of blood vessels can spell further damage. Lungs are covered in capillaries, so these gaps start leaking blood and immune cells into the interior surface of the lungs, potentially providing the reason for Covid-19 patients’ respiratory distress.
To make things worse, according to the supercomputer analysis, the virus might also increase the natural production of hyaluronic acid—a biopolymer familiar to skincare aficionados, as it can absorb more than 1,000 times its weight in water. As bradykinin causes blood vessels to leak water into your lungs, it hits the hyaluronic acid in your lungs and forms a hydrogel. “It’s like trying to breathe through Jell-O,” Jacobson says. “At that point, unfortunately no matter how much oxygen you’re pumping through a ventilator, you can’t get a gas exchange through the hydrogel.”
Bradykinin dysregulation may also be behind the thyroid problems some Covid-19 patients are reporting. Previous research has found that, in addition to influencing the circulatory system, bradykinin is an important regulator of thyroid hormones.
Ilaria Muller, an endocrinologist at the Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico in Milan, and colleagues recently found that many patients who were hospitalized had abnormally low levels of thyroid-stimulating hormones, suggesting thyrotoxicosis and at least temporary thyroid damage. She says this damage could come from direct damage from the virus through the thyroid’s ACE2 receptors or from systemic inflammation.
More surprisingly, bradykinin storms also help offer an explanation for some of Covid-19’s neurological symptoms — from headaches to long-term nerve damage — which in one study afflicted 57 percent of Covid-19 patients. High levels of bradykinin in particular can cause the blood-brain barrier to break down, potentially allowing the virus into the brain and causing inflammation and damage.
Finally, as Elemental reports, the theory may even explain why men seem to be more likely to have worse cases of Covid-19. Some aspects of the RAS systems have receptors on the X chromosome, meaning that women have twice the levels of these stop-gap proteins, possibly giving them extra protection against the virus.
The supercomputer model also found different gene expression patterns in the lavage fluid from the lungs of COVID-19 patients. This is rare data, in part because getting that fluid can be dangerous to healthcare professionals, who may get infected while taking the samples, so this procedure is no longer carried out. A clinical trial measuring actual bradykinin levels in samples from Covid-19 patients’ lungs would provide a lot of valuable information but is unlikely to happen because of the transmission risk.
When something like a virus tweaks part of the body’s intertwined systems, you often end up with rippling consequences—in this case, a dire trend toward inflammation, possibly through both bradykinin pathways and cytokine production. Essentially, the bradykinin pathway gets off the track—and then it’s like a runaway train, potentially causing damage in many locations around your body.
What do bradykinin storms mean for possible Covid-19 treatments?
After finding the potential role of bradykinins in severe Covid-19 in March, Roche went looking for a way to halt this inflammatory cascade. “It’s like a set of gear wheels—inflammation, injury, inflammation—and you’re trying to jam up the wheels,” he says. Along with his wife, Renuka Roche, an assistant professor in occupational therapy at Eastern Michigan University, he started to explore potential treatments that were ready to use.
As clinicians trained to pay a lot of attention to recovery through rehabilitation, he says, “We know that health care does not end with just saving a person’s life.” Roche says life quality is important too, meaning any intervention that could minimize damage would be a true advancement in the fight against Covid-19’s ravages.
Treatment targeting bradykinin signaling wouldn’t have to be perfect to improve lung damage and long-term outcomes. “If you’re able to even dampen the cycle by 50 percent, that means that much tissue may be spared,” Roche says.
In the medical literature, the Roches found a medication called icatibant that is both known to be safe and inhibits bradykinin signaling. It was already approved by the FDA, with the added benefit of an expired patent, meaning generic versions could be made much more affordably. They reached out to the Canadian and Indian governments about starting rapid research on icatibant in late March, wrote an open letter to the scientific community in April, and published a paper on their hypothesis in May.
At the same time, Frank van de Veerdonk, an infectious disease specialist at Radboud University Medical Center in the Netherlands, was reaching similar conclusions. He knew that ACE2 is an important part of the RAS, and in April, hypothesized that a dysregulated bradykinin system was causing blood vessels to leak into Covid-19 patients’ lungs.
More recently, “We published data in patients with icatibant targeting bradykinin in Covid-19 as a treatment,” van de Veerdonk wrote Vox in an email. While not a controlled clinical trial, van de Veerdonk published a study where nine hospitalized patients were treated with icatibant and matched to similar Covid-19 patients who were not; the patients who’d received icatibant needed less supplemental oxygen and experienced no adverse effects from the drug.
In the US, Quantum Leap Healthcare Collaborative has started a clinical trial of five potential treatments, including icatibant. (They are still currently enrolling patients.) “The safety of the drug is well understood, and it’s fast-acting,” says Paul Henderson, director of collaboration at Quantum Leap.
In general, he says bradykinin receptors are interesting because they are upstream of most of the inflammatory response, including cytokines. If proven effective, he says, these treatments will probably also be useful for influenza and other diseases that cause acute respiratory distress.
Henderson doesn’t discount cytokines’ inflammatory impact altogether but suggests that interventions targeting cytokines may have been “taking out too little of all the processes going on to have much impact.” Imagine how much easier it is to dam a river at its headwater than closer to its mouth—similarly, interventions further “upstream” in biological pathways could have a larger impact.
In some ways, this work could be as important as finding a vaccine. “Reducing the burden on the health care system and preventing the very sickest from dying is really important,” Henderson says.
But he also cautions that, like with cancer, there is unlikely to be one “magic bullet drug.” Instead, it’s more likely a combination therapy, including anti-inflammatory medications and antivirals, will be necessary. “You’ll likely need different interventions in different stages of infection,” he says. “It is extremely complicated.”
Nevertheless, since Jacobson’s paper came out, his hypotheses have been supported by other research. For example, vitamin D is known to regulate RAS, and vitamin D deficiencies have been associated with severe cases of Covid-19.
This fits with a part of the supercomputer analysis that suggested the virus activates genes that break down more vitamin D. Lo and behold, at the end of August, a clinical trial in Spain on vitamin D found that it significantly reduced the need for ICU treatment in Covid-19 patients.
Similarly, another analysis, run by the World Health Organization, which incorporates seven different clinical trials, found that corticosteroids, which inhibit a protein activated by the bradykinin receptor, reduced the risk of Covid-19 death — fitting the computer model’s prediction neatly.
Bradykinin storms may also have implications for long-haul Covid-19 patients. Jacobson is now collaborating with Covid patient groups to gather data. “We’re looking at the top 100 symptoms and trying to map them to this mechanism,” he says, adding that several of his fellow researchers are long-haulers.
He says one of the next questions they hope to address is whether bradykinin dysfunction continues even after the virus has cleared, if the virus itself is persisting in different organ systems or some combination of both.
When new information raises more questions
The notion of bradykinin storms are appealing because they offer a tantalizingly unified theory that would explain so many of Covid-19’s inscrutable impacts. Joshua Zimmerberg, a biophysical virologist at the Eunice Kennedy Shriver National Institute of Child Health and Human Development at the National Institutes of Health, who was not involved in any of the bradykinin research, says the evidence is now compelling. “When you have independent confirmation, when people come to the same conclusion for different reasons—that’s very good evidence.”
But he warns against raising hopes for immediate treatments. “We all crave simple pathways and simple ideas, but inflammation is really complicated. There are still a lot of inflammatory diseases without good treatments.” Dampening bradykinin production too much, or at the wrong time—for example, early in the infection, when the natural inflammation cycle is needed to fight the virus—might actually be harmful.
Roche says the next steps are for large-scale randomized placebo-controlled clinical trials on potential drugs that inhibit bradykinin. “The hypothesis, [Jacobson’s] gene expression data, [van de Veerdonk’s] small-scale case series—these won’t move the needle,” he says. Data is needed to add drugs to doctors’ arsenal against the pandemic. But he’s gracious about more widespread attention only being directed toward bradykinin now, after he’s spent months trying to raise its profile.
“The pandemic has exposed key weaknesses in health care itself,” he says. “We need to empower ourselves with as much knowledge as we can, so we can serve our patients and protect ourselves.”
Lois Parshley is a freelance investigative journalist. Follow her Covid-19 reporting on Twitter @loisparshley.