The oxidative potential of differently charged silver and gold nanoparticles on three human lung epithelial cell types

For years, gold nanoparticles were the safe choice. When researchers needed to deliver drugs, image tissue, or probe cellular biology without poisoning the patient, they reached for gold. Its chemical inertness was almost a given — a background assumption so solid it rarely got tested directly. Silver, by contrast, had a well-documented toxicity profile that kept researchers cautious. Then, Schlinkert and colleagues coated gold nanoparticles with a common polymer and watched them kill primary human lung cells. The gold hadn't changed; the coating had. That single finding reshapes how we think about nanoparticle safety because it means the core material alone doesn't determine biological risk. The outer surface does. Here's the experimental logic. The team synthesized monodisperse silver and gold nanoparticles in the 7 to 10 nanometer size range, then coated them with one of two substances: sodium citrate, which makes the surface negatively charged, or chitosan, a polymer that makes it positively charged. By varying the amount of chitosan, they could dial the surface charge — measured as zeta potential — across a range from about negative 50 millivolts all the way up to positive 70. Same core, same size, same basic chemistry — just different surface charges. That design lets you isolate what charge alone does to a living cell. But they didn't stop at one cell type, which is where the study gets interesting. They tested three different human lung epithelial models, each representing a different point on the spectrum from convenient to realistic. A549 is a cancer-derived alveolar cell line — fast-growing, widely used, and when seeded at the right density, it forms a neat, stable confluent monolayer within four days. BEAS-2B is an immortalized bronchial line that behaves differently: it piles up into multilayers rather than a single sheet, though it still develops functional tight junctions. Then there are normal human bronchial epithelial cells taken from healthy donors, known as NHBE cells. These primary cells are the closest thing to what actually lines a human airway. Under the study's culture conditions, they never formed a confluent layer at all. The team characterized each cell type's barrier using two tools: transepithelial electrical resistance, or TEER, a measure of how tightly cells seal against each other, and immunofluorescence staining for claudin-1, the protein that stitches tight junctions together. A549 cells reached a TEER of about 67 ohm-centimeter-squared, BEAS-2B hit around 75, and NHBE topped out at just 12. In the NHBE cells, claudin-1 was being produced but stayed trapped in the cell center rather than migrating to the membrane where it would actually form a junction. So, you have three cell types with very different barrier architectures — and as it turns out, very different vulnerabilities. When the nanoparticles arrived, the pattern was clear: the more biologically realistic the cell model, the more damage the particles caused. A549 cells, the standard workhorse of lung toxicology, were the least susceptible regardless of which nanoparticle they received. Their intact monolayer and tight junctions seemed to buffer them against even the most aggressively charged particles. BEAS-2B cells showed increased cytotoxicity, but only when exposed to highly positively charged gold nanoparticles — those in the positive 65 to 75 millivolt range. And NHBE primary cells were hit hardest. Both silver and gold nanoparticles with a surface charge above positive 40 millivolts induced cytotoxicity in those cells. Schlinkert and colleagues measured this using two assays run in tandem. The lactate dehydrogenase release assay reports membrane rupture, because dying cells leak this enzyme into the surrounding medium. The CellTiter-Blue assay reports metabolic viability, which measures how well cells are still functioning. Together, they provide both a death count and a health check. The pattern that emerged wasn't a smooth dose-response ramp. It looked more like a threshold effect: below about positive 40 millivolts, the nanoparticles were largely tolerated. Above that, in primary cells, toxicity kicked in. The authors describe the positive 30 to 40 millivolt range as one that might be considered harmless under cell culture conditions, and charge above positive 60 millivolts as something that has to be designated as problematic. The cytotoxicity data answers what is happening. The reactive oxygen species data answers why. Reactive oxygen species are chemically unstable oxygen-containing molecules that cells normally keep under tight control. When their production overwhelms the cell's antioxidant defenses — things like glutathione and antioxidant enzymes — the result is oxidative stress, which damages DNA, proteins, and cell membranes. The team measured reactive oxygen species using the DCFH-DA assay, a fluorescent dye that lights up in the presence of reactive oxygen species, both inside living cells and in cell-free conditions. The cell results threw up a paradox. Reactive oxygen species production was most prominent in A549 cells, the very cell type that showed the least cytotoxicity. The highest signal came from highly positively charged gold nanoparticles, specifically the formulation at positive 65 millivolts, in those A549 cultures. So, the cells generating the most oxidative stress were also the most resistant to dying from it. That's not a contradiction so much as a reminder that oxidative stress and cell death are not the same thing — resilient, rapidly dividing cancer-derived cells may handle reactive oxygen species differently than primary airway cells. The more striking finding came from the cell-free measurements. When the researchers ran the reactive oxygen species assay without cells at all — just nanoparticles in solution with the fluorescent dye — reactive oxygen species production still increased as the amount of chitosan coating increased, with the gold nanoparticles generating far more than silver at equivalent charges. Crucially, when they tested the coating solvents alone, no chitosan-concentration-dependent increase appeared. The reactive oxygen species wasn't coming from free chitosan floating in solution; it was coming from chitosan presented on a nanoparticle surface. The coating is chemically reactive in its own right. There's an important complication, though. When the team incubated nanoparticles in cell culture medium containing fetal calf serum, the surface charge changed dramatically. The formulation went from positive 65 millivolts as synthesized to negative 23 millivolts after 24 hours in A549 medium. That shift happens because serum proteins adsorb onto the nanoparticle surface, forming what's called a protein corona, and essentially bury the original charge. When that occurs, reactive oxygen species production in the cell-free assay dropped to minimal levels. Incubating in medium without serum didn't produce the same effect, confirming that adsorbed proteins are doing the masking. This means that inside a biological environment, the nanoparticle a cell actually encounters may look chemically very different from the nanoparticle that went in. Taken together, these findings carry a specific practical warning. Studies relying exclusively on A549 cells — still among the most commonly used models in pulmonary nanotoxicology — may systematically underestimate nanoparticle toxicity. A549 cells produced robust reactive oxygen species signals but survived. Primary NHBE cells, which lack the tight barrier those cancer cells maintain, were killed by the same particles at lower charge thresholds. The cell model you choose isn't a neutral methodological detail; it determines what you see. The second warning is about material identity. Gold's reputation for inertness is real but conditional. It holds for bare or citrate-coated gold, but it does not hold for chitosan-coated gold pushed above a critical surface charge. The core material matters, but the coating is the part that actually meets the cell, and it needs to be evaluated just as rigorously. As Schlinkert and colleagues put it, chitosan functionalization with high surface charge plays an important role in nanoparticle toxicity, and these effects depend on both the core material and the cell type. Finally, the paper is candid about what these results don't cover. All of this was measured in submerged liquid culture — cells sitting under a bath of nanoparticle suspension. That's not how you actually inhale particles. Verifying these findings under air-liquid interface conditions, where cells sit at the boundary of air and liquid the way they do in a real lung, remains the next necessary step. The biology observed here is real and reproducible, but the exposure scenario is a simplification, and the authors know it. What the study establishes firmly is that surface charge is a toxicity dial, and the particle core is only part of the story. 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.