Nitrous oxide is one of the most consequential greenhouse gases in the atmosphere, roughly 270 times more powerful than carbon dioxide over a century, and it also attacks the stratospheric ozone layer. Most human-caused emissions come from microbes that carry out denitrification, the stepwise conversion of nitrate to dinitrogen gas through nitrite, nitric oxide, and finally nitrous oxide. For decades, scientists have tried to trace exactly which microbial pathways produce this gas by reading the isotopic fingerprints written into its molecules. A new study published in Biogeosciences now shows that these fingerprints are far more slippery than the field has assumed, and that the physiology of the producing organism can rewrite the signature almost at will.
The research, led by Noémy Chénier of Empa, the Swiss Federal Laboratory for Science and Technology, together with colleagues at the University of Basel, focused on two closely related Pseudomonas species that are workhorses of denitrification research. Pseudomonas chlororaphis subsp. aureofaciens carries the copper-containing nitrite reductase NirK, while Pseudomonas chlororaphis subsp. chlororaphis carries the structurally distinct cytochrome cd1-type enzyme NirS. Both strains lack the enzyme NosZ, which normally consumes nitrous oxide, so the gas simply accumulates and can be measured cleanly. The team grew both organisms under strictly anoxic conditions and tracked the isotopic composition of the emitted nitrous oxide in real time using quantum cascade laser absorption spectroscopy, complemented by conventional isotope ratio mass spectrometry of the accumulated product.
Two isotopic quantities sit at the heart of the study. The first is site preference, the difference in nitrogen-15 abundance between the central and terminal nitrogen atoms within the linear nitrous oxide molecule. Because bacterial denitrification is widely assumed to yield site preference values near zero per mil through the canonical nitric oxide reductase NorB, deviations from this value are routinely interpreted as evidence for other production pathways such as fungal denitrification, hydroxylamine oxidation, or abiotic nitrite reactions. The second quantity is the oxygen-18 signature, which reflects how much of the oxygen in nitrous oxide came from the original nitrate versus from exchange with surrounding water before nitrite was reduced.
The most striking result concerns site preference. When the researchers measured the accumulated gas at the end of closed batch incubations, both species produced values clustered near zero per mil, exactly as the textbook picture predicts. But the continuous laser measurements told a different story. During the earliest phase of nitrous oxide production, site preference transiently climbed to values between roughly plus eight and plus twenty per mil before converging back toward zero as the reaction progressed. These excursions were reproducible, most clearly in the NirK-bearing strain, yet they would be completely invisible in any experiment that only measures the pooled product at the end.
The authors propose a plausible enzymatic explanation. Early in denitrification, rapid nitrate and nitrite reduction can transiently accumulate nitric oxide faster than NorB can handle it, while the expression of NorB itself may still be ramping up. Under such high nitric oxide stress, alternative detoxification enzymes, particularly flavohemoglobins, can step in and reduce nitric oxide to nitrous oxide with a distinctly elevated site preference, an effect demonstrated directly in recent work on engineered bacteria. Other candidate systems, including the hybrid cluster protein pathway and flavodiiron proteins, may also contribute. The implication is unsettling for isotope-based source apportionment: a site preference of five to fifteen per mil, often read as evidence of mixed or non-denitrification sources, could in principle arise entirely within a denitrifying cell simply because different nitric oxide reductases took turns.
The oxygen isotope results challenge an even more entrenched assumption. In the widely used denitrifier method, which converts nitrate to nitrous oxide for isotope analysis, the NirS-bearing species is known to exchange a large fraction of its oxygen atoms with water, while the NirK-bearing species was long thought to exchange almost none. That contrast has been extrapolated into a general rule that oxygen exchange behavior is dictated by nitrite reductase identity. The new experiments dismantled this rule. Under active growth, the NirK-bearing Pseudomonas aureofaciens exhibited oxygen exchange ranging from about 38 percent in gas-flushed bioreactor incubations to essentially complete exchange in closed batch cultures, vastly exceeding the values below nine percent reported under denitrifier-method conditions.
The NirS-bearing strain told a complementary story. Pseudomonas chlororaphis maintained consistently high but less variable exchange, around 66 percent, during active growth. Crucially, when the researchers harvested actively growing cells and resuspended them in a defined nitrate medium, mimicking the stationary-phase, washed-cell conditions of the denitrifier method, both strains reproduced exactly the canonical low- and high-exchange behavior reported in the classical literature. This demonstrates that the famous NirK-versus-NirS exchange dichotomy is not an intrinsic property of the enzymes at all, but an artifact of the specific physiological regime in which the denitrifier method operates.
What governs the exchange, the authors argue, is the lifetime of the nitrite intermediate pool. Oxygen exchange between nitrite and water requires nitrite to persist long enough to approach isotopic equilibrium with its surroundings. During active growth, when biomass synthesis and enzyme expression proceed simultaneously, the coupling between nitrate reduction and nitrite reduction loosens, nitrite lingers, and exchange increases. In resuspension assays, fully induced cells reduce nitrate rapidly and tightly, leaving nitrite little time to equilibrate, so nitrate-derived oxygen is retained. The oxygen-18 signature of nitrous oxide therefore behaves as a composite signal shaped by nitrate source, exchange dynamics, and branching isotope effects, all of which fluctuate with metabolic state rather than with enzyme identity.
The consequences ripple outward across biogeochemistry. Models that assign fixed isotopic endmember values to denitrification, nitrification, and abiotic pathways may misattribute sources in soils, oceans, groundwater, and wastewater treatment plants whenever the physiology of the microbial community shifts. The authors do not suggest discarding existing reference ranges, but they insist that physiological state, intermediate dynamics, and environmental context must be built explicitly into any interpretation of nitrous oxide isotopocules. Time-resolved measurements, they show, are essential, because pooled endpoint values can average away the very signals that reveal which enzymes were actually at work.
For a field that has relied on static isotopic fingerprints for more than two decades, the message is both humbling and energizing. Nitrous oxide isotope signatures are not fixed constants stamped by enzyme identity; they are conditional expressions of living metabolism. Pairing laser-based isotopocule monitoring with transcriptomics, proteomics, and targeted inhibitor experiments could finally connect the isotopic record to the specific enzymes operating at each moment, opening the door to genuinely mechanistic source attribution for one of the planet’s most important greenhouse gases.
Subject of Research: Isotopic variability of microbially produced nitrous oxide during bacterial denitrification
Article Title: High nitrous oxide isotopic variability during denitrification by Pseudomonas species bearing NirK and NirS
Article References: Chénier, N., Magyar, P. M., Zopfi, J., Frey, C., Kuhn, T., Lehmann, M. F., & Mohn, J. (2026). High nitrous oxide isotopic variability during denitrification by Pseudomonas species bearing NirK and NirS. Biogeosciences, 23(18), 6705-6724. https://doi.org/10.5194/bg-23-6705-2026
Image Credits: AI Generated
Keywords: nitrous oxide, denitrification, isotopocules, site preference, Pseudomonas, NirK, NirS, oxygen isotope exchange, greenhouse gas, nitrogen cycle, biogeochemistry, QCLAS
News Source: Morgan Morrow. (October 9, 2026). Microbial Physiology, Not Just Enzymes, Drives Nitrous Oxide Isotope Signatures. Scienmag.
