Neutrophil extracellular traps in COVID-19

Let's start with the image in your head: a neutrophil, one of those first-responder white blood cells, unspooling its own DNA like a tangled fishing net. That web, studded with histones, enzymes like myeloperoxidase, and oxidants, is a neutrophil extracellular trap or NET. It's a clever move when you're trying to corral bacteria. But a net is indiscriminate. When it snags the wrong things, it can rile up inflammation and thicken blood. That's the double-edged sword. In the lungs, NETs can worsen injury. In blood vessels, they can seed clots. Early in the pandemic, when doctors were seeing both raging inflammation and strange sticky clots in severe COVID-19, it raised a simple, high-stakes question: are NETs part of this story? Zuo, Yalavarthi, Shi, and colleagues at the University of Michigan took a clean, serum-first approach to find out. They assembled a cohort of 50 hospitalized patients with confirmed COVID-19 and a parallel group of 30 healthy controls, ending up with 84 total patient samples because some people were sampled more than once. These weren't bespoke research draws; they were leftover clinical sera, saved for up to two days at refrigerator temperature before freezing, which will matter later. The team focused on three markers in serum. Cell-free DNA as a broad indicator of injury and NET burden. Myeloperoxidase-DNA complexes, which are myeloperoxidase bound to DNA, as a neutrophil-specific NET remnant. And citrullinated histone H3, or Cit-H3, a footprint of a particular NET-making route that depends on the enzyme PAD4. The first thing they checked was whether NET remnants were actually higher in COVID-19. Yes, all three markers—cell-free DNA, myeloperoxidase-DNA, and Cit-H3—were elevated in patient sera compared with healthy controls, with strong statistical support. Think of that as the big headline. The blood of hospitalized COVID-19 patients carries more of the stuff you'd expect to see when neutrophils cast their DNA nets. When they looked at how the markers relate to each other, cell-free DNA tracked with myeloperoxidase-DNA, while its relationship with Cit-H3 wasn't significant. That subtle split will come back. Numbers alone don't tell you what's happening clinically, so the team lined up the NET markers with standard labs drawn on the same day. Cell-free DNA rose in step with C-reactive protein, D-dimer, and lactate dehydrogenase—classic readouts of inflammation, coagulation, and tissue damage—and it also tracked with the absolute neutrophil count. Myeloperoxidase-DNA painted a different picture. It strongly associated with the neutrophil count but didn't meaningfully line up with those inflammatory or coagulation markers. Cit-H3 took its own path, correlating with platelet counts and not much else. When they repeated the analysis using just the first sample from each patient to avoid overcounting, the pattern held. If NETs are tied to worse illness, you should see more of them in the sickest patients. That's what they found. Among samples from people who needed mechanical ventilation compared with those breathing room air, cell-free DNA and myeloperoxidase-DNA were clearly higher. Cit-H3 didn't separate those groups. Interestingly, the absolute neutrophil count itself—the number of foot soldiers—wasn't significantly different between ventilated and nonventilated groups. That suggests it's not just how many neutrophils you have, but what they're doing and what they're leaving behind. A tiny but telling longitudinal window sharpened the point. In a subset of 22 patients with serial samples, three had steadily worsening oxygen needs over time—room air to nasal cannula to high-flow and then intubation—and in all three, NET markers tended to climb as breathing got harder. It's not proof of cause, but the directionality matters: as the lungs faltered, the NET signal rose. Blood markers can be epiphenomena. To test whether patient sera actually push neutrophils toward NETosis, the team moved into the dish. They purified neutrophils from healthy donors and did the unglamorous work to get them to at least 95 percent purity. Then they bathed them in 10 percent serum from either COVID-19 patients or healthy controls. Two independent readouts told the same story. Using a dye that only binds DNA once it's outside the cell, they saw a clear bump in extracellular DNA with COVID-19 sera compared with controls, based on 27 patient samples versus 20 control sera. When they measured myeloperoxidase activity released from NETs after digesting the DNA scaffold with a nuclease, COVID-19 sera again drove a higher signal in 27 patient versus 17 control comparisons. Under the microscope, those neutrophils exposed to patient serum sprouted the telltale web: strands of chromatin—DNA plus histones—decorated with neutrophil elastase, draped across the field. If you're listening for potential confounders, one common medication at the time was hydroxychloroquine. About half the patients received it. On days with versus without hydroxychloroquine, NET markers didn't budge in a consistent way in this dataset. That doesn't settle the drug's effect on neutrophils, but it argues these NET signals weren't simply a pharmacologic artifact. Now, why do those split patterns among the markers matter? Because they hint that COVID-19 isn't driving NETosis through a single on-ramp. Myeloperoxidase-DNA moved with the number of neutrophils, which makes sense for a general "more soldiers, more nets" model. Cit-H3, the PAD4-dependent signature, linked to platelet counts, which points toward a platelet-neutrophil axis. In platelets, activation can flip on neutrophils; in neutrophils, PAD4 rewrites histones, loosening DNA so it can be cast. Zuo and colleagues didn't claim a mechanism from correlations alone, but the divergence between myeloperoxidase-DNA and Cit-H3 sits comfortably with the idea of multiple NET-making programs engaged at once. Step back, and you see how this fits the broader thromboinflammatory picture clinicians were seeing. NETs can serve as scaffolds for thrombin generation by activating the contact arm of coagulation. They can snare and activate platelets, fuel complement, and injure endothelium. The rise in NET remnants alongside D-dimer, the split with platelets, the higher levels in ventilated patients—those are the blood echoes of that biology. What kicks NETosis off in COVID-19? Take your pick of plausible triggers: virus-damaged epithelial cells, activated platelets, perturbed endothelium, and a cytokine soup including interleukin-one beta, interleukin-six, and interleukin-eight. The in vitro experiment, where COVID-19 serum alone pushed healthy neutrophils to NET, tells you potent soluble triggers are present in the blood of these patients. There are caveats—important ones. These were leftover clinical sera, not collected to preserve delicate complexes, and they sat up to 48 hours at four degrees Celsius before freezing. That's reasonable in a busy hospital, but it raises the possibility that some NET remnants degraded, which could blur differences. The study design is mostly cross-sectional; you can line up markers with labs and with ventilation status, but you can't declare NETs as the driver. Even that longitudinal hint—three patients whose NET markers rose as oxygenation fell—is just that, a hint. And because different NET assays capture different slices of the phenomenon, standardizing multi-marker panels will matter if we want to compare across centers or over time. So what do we do with this? First, treat circulating NET remnants as serious candidates for risk stratification. In this cohort, cell-free DNA and myeloperoxidase-DNA signaled inflammation and illness severity. That's useful for triage, especially if it holds up in larger, standardized, prospective cohorts where you can ask: do NET markers on day three predict who will need a ventilator on day five? Second, consider whether dialing NETs down changes outcomes. There are three broad levers. You can cut the nets that already exist—deoxyribonuclease or DNase does that by chopping up extracellular DNA. You can block formation upstream—PAD4 inhibitors reduce histone citrullination, and neutrophil elastase inhibitors blunt chromatin decondensation. Or you can modulate the signals that push neutrophils toward NETosis. One intriguing example comes from a small study in China where dipyridamole, an adenosine receptor agonist better known as an antiplatelet drug, was associated with improved platelet counts and lower D-dimer in severe COVID-19; in other work, dipyridamole inhibits NET formation. These are leads, not prescriptions, and they need proper randomized trials with thromboinflammatory endpoints. If you're hearing a theme, it's this: NETs sit at the crossroads of the two things that made severe COVID-19 so hard—runaway inflammation and a blood system inclined to clot. Zuo and colleagues didn't settle causality, and they didn't need to. They mapped the terrain. They showed that hospitalized patients carry more NET remnants in their blood than healthy people, that certain NET signatures move with inflammation and with platelets in distinctive ways, that patients on ventilators carry a heavier NET burden, that in a handful of people getting sicker by the day, the NET signal climbed with them, and that the serum of those patients can push a healthy neutrophil to cast its net. The next moves are obvious and demanding. Validate these markers prospectively in well-phenotyped cohorts. Nail down which NET readouts—cell-free DNA, myeloperoxidase bound to DNA, PAD4-dependent histone citrullination, or a combination—best predict dangerous turns. And in carefully controlled trials, ask whether cutting or calming NETs changes the arc of illness, particularly the thromboinflammatory spiral that lands people in intensive care. Until then, this study gives us a clear, evidence-based lens: when you see severe COVID-19, you're looking at a disease where the body's own traps may be tightening.