Vapors Produced by Electronic Cigarettes and E-Juices with Flavorings Induce Toxicity, Oxidative Stress, and Inflammatory Response in Lung Epithelial Cells and in Mouse Lung

A small vial of e-liquid sits under a fluorescent lab light. A researcher draws vapor through a chemical solution and watches a number climb on a spectrophotometer — a fluorescent signal for oxidants, rising with every flavor of juice they test. The question that forces itself into the room is what exactly is in that vapor, and what does it do once it gets inside a lung? That's the question Chad Lerner, Irfan Rahman, and their colleagues at the University of Rochester set out to answer. Their approach was systematic — starting with the chemistry of the aerosol itself, then moving to cells, and finally to living animals. What they found at each step was consistent, and that consistency is the story. The foundation of the study is a cell-free assay using a fluorescent dye called DCFH — 2',7'-dichlorodihydrofluorescein. When oxidants or reactive oxygen species are present, DCFH converts to a form that glows green under light. It's a sensitive chemical detector, and the team used it to measure what's actually in e-cigarette vapor before it ever touches a lung cell. The answer was unambiguous. Both vaporized aerosols and unvaporized e-liquids produced measurable oxidants. Compared to an air sham control sitting at around 3.2 micromolar hydrogen peroxide equivalents, aerosols from the base humectants — the carrier liquids that make up most e-juice — were far higher. Propylene glycol aerosol averaged 125.7 micromolar, and glycerin averaged 255.9. A fifty-fifty mix of the two hit 306.6. These are not trace signals. The oxidant chemistry is happening in the vapor itself, before any biology is involved. Two sources drive that oxidant load. The first is the liquid — specifically, the flavoring. When the team tested many commercial e-liquids in the DCFH assay, non-tobacco flavors, including dessert, fruit, and candy varieties, were significantly more reactive than tobacco-type flavors. One fruit flavor, "Very Berry," measured 101.3 fluorescence units, with most sweet and candy flavors clustering in that same range. The second source is the device itself. Running a heating element with no liquid at all still produced oxidants well above background — about 175.6 micromolar from a pre-used coil running dry. The coil's history matters too. A brand-new element, activated repeatedly, dropped from 33.3 micromolar on its first use to 5.7 by the third. Then the researchers added a drop of e-liquid directly onto the wick — a practice called dripping — and the number spiked to 135.6 micromolar. The method of delivery changes the dose dramatically. Dripping consistently produced far higher oxidant readings than filling a standard clearomizer tank, often requiring the researchers to dilute their samples tenfold just to measure them. One more variable worth noting is that nicotine itself wasn't the main culprit. When the team compared nicotine-containing liquids to nicotine-free versions, the nicotine-containing ones actually showed less DCFH reactivity in several comparisons. The oxidant story was being written by the flavors and the hardware, not the nicotine. Now the team moved the experiment from chemistry to biology. Human lung fibroblasts — the structural cells that support and repair lung tissue — were treated directly with e-liquids for twenty-four hours. Under a microscope, the cells looked different. They were enlarged, vacuolated, and losing their typical elongated shape, with some conditions producing what the researchers described as large adhered circular cells. The damage tracked with concentration: five percent e-liquid produced worse effects than one percent, and cigarette smoke extract produced the worst. Cell viability told a similar story. At 2.5 percent concentration, most e-liquids didn't cause statistically significant cell death on their own — though twenty-four milligram nicotine tobacco e-liquid dropped viability to 78.6 percent versus a control around 90.5 percent. Cigarette smoke extract at the same concentration was catastrophic, dropping viability to 12.7 percent. But at lower concentrations in smaller culture dishes, three commercial e-liquid brands reduced viability below 50 percent. The context of the exposure shapes the outcome. The inflammatory signal from these fibroblasts was flavor-specific in a striking way. Baseline IL-8 — a cytokine that recruits inflammatory immune cells to a site of damage — was about 15.9 picograms per milliliter in control cells. The base humectants didn't move it. Tobacco-flavored e-liquid with no nicotine barely moved it. But cinnamon roll e-liquid drove IL-8 to 458 picograms per milliliter. That's a roughly twenty-nine fold increase from a single flavor additive. Something in cinnamon flavoring is triggering a pronounced inflammatory response that other flavors are not. The team also noticed a partial mismatch between the cell-free chemistry and the cell-based biology. In the DCFH assay, nicotine-containing e-liquids had lower oxidant reactivity. Yet in fibroblasts, nicotine-containing products still worsened morphological damage and, in some cases, boosted IL-8. Pure humectants, which showed no reactivity in the cell-free assay at all, still caused physical stress to fibroblasts under the microscope. The chemical sensor and the living cell are not measuring the same thing. Both sets of data are real; they're just capturing different aspects of what the exposure does. From fibroblasts, the team moved to airway epithelial cells — the cells that actually line the surface of the airways. This is the first tissue that inhaled vapor contacts. Here they used an air-liquid interface exposure system, a setup where cells sit on a membrane with their top surface exposed to air and aerosol, mimicking the geometry of an actual airway far more closely than submerging cells in liquid. H292 human airway epithelial cells were exposed to aerosols from a Blu e-cigarette — tobacco flavor — in three to four second puffs for five, ten, or fifteen minutes, then allowed to rest for sixteen hours before cytokine measurement. The result was a clear inflammatory signal. IL-8 was significantly elevated in all aerosol-exposed groups compared to air sham controls. So was IL-6, a cytokine that modulates multiple immune and inflammatory pathways. The IL-6 response was dose-dependent with time: ten-minute exposures produced significantly more IL-6 than five-minute ones. The cells were reading the aerosol as a threat and sending out distress signals in proportion to how long they'd been exposed. The final step was the whole animal. The team exposed C57BL/6J mice to side-stream Blu e-cigarette aerosols — approximately 200 milligrams per cubic meter of total particulate matter — for five hours a day over three consecutive days. Systemic uptake was confirmed: plasma cotinine, a marker of nicotine absorption, averaged 10.78 nanograms per milliliter in mice tested right after the third exposure, and was undetectable twenty-four hours later. In the lungs, two cytokines in bronchoalveolar lavage fluid — the fluid collected by washing out the airspace — were significantly elevated: MCP-1 and IL-6. A broader cytokine panel showed IL-1 alpha and IL-13 also significantly increased compared to air controls. The cell count data trended upward but didn't reach statistical significance, which tells you this was a short acute exposure — three days, not three months. The inflammatory signal was there, but structural damage would take longer to accumulate. The glutathione data are where the in vivo picture gets most pointed. Glutathione is the lung's primary antioxidant — the molecule that neutralizes reactive oxygen species before they damage tissue. In air-exposed mice, total glutathione in lung tissue averaged 5.89 nanomoles per milligram of protein. In e-cigarette-exposed mice, that number dropped to 1.69. The lung's chemical defense system was being depleted. Lerner and colleagues frame glutathione as critical in maintaining cellular redox balance, and three days of e-cigarette exposure was enough to measurably disrupt it. What do you do with all of this? The paper is careful not to claim e-cigarettes are as harmful as combustible cigarettes — it notes that some carbonyls in e-cig aerosols are nine to eight hundred seven times lower than in cigarette smoke. But that framing cuts both ways. Lower than cigarette smoke is not the same as safe. The team also detected nanoscale particles and metals including lead, chromium, and nickel in the aerosols — materials that come from the hardware, not the liquid. The biological evidence is consistent across every experimental level: oxidants in the chemistry, stress and cell death in fibroblasts, inflammatory cytokines in airway epithelial cells, and depleted antioxidant defenses in living lungs. These products reached millions of users before any of this data existed. The flavor-specific risks — cinnamon in particular — the long-term consequences of repeated glutathione depletion, and the dose-response relationship with real-world vaping behavior are questions that are still open. But the premise that e-cigarettes are biologically inert has been tested here, across three levels of biological organization, and it didn't hold. This lecture was created by ennepō. Go to https://ennepo.ai to Discover, Create and Follow the latest research in your field. Read when you can. Listen when you want to.