Substantial oxygen consumption by aerobic nitrite oxidation in oceanic oxygen minimum zones

Oceanic oxygen minimum zones (OMZs) are globally significant sites of biogeochemical cycling where microorganisms deplete dissolved oxygen (DO) to concentrations <20 µM. Amid intense competition for DO in these metabolically challenging environments, aerobic nitrite oxidation may consume significant amounts of DO and help maintain low DO concentrations, but this remains unquantified. Using parallel measurements of oxygen consumption rates and 15N-nitrite oxidation rates applied to both water column profiles and oxygen manipulation experiments, we show that the contribution of nitrite oxidation to overall DO consumption systematically increases as DO declines below 2 µM. Nitrite oxidation can account for all DO consumption only under DO concentrations <393 nM found in and below the secondary chlorophyll maximum. These patterns are consistent across sampling stations and experiments, reflecting coupling between nitrate reduction and nitrite-oxidizing Nitrospina with high oxygen affinity (based on isotopic and omic data). Collectively our results demonstrate that nitrite oxidation plays a pivotal role in the maintenance and biogeochemical dynamics of OMZs.


nature research | reporting summary
April 2020 Field-specific reporting Please select the one below that is the best fit for your research. If you are not sure, read the appropriate sections before making your selection. Reporting for specific materials, systems and methods Nitrite oxidation rate measurements (using 15NO2-), OCR measurements (using optical sensor spots), and ammonia oxidation rate measurements (using 15NH4+) were made along depth profiles at six stations in the ETNP, including three OMZ stations and three AMZ stations. Other measurements included water column properties, stable isotopic composition of dissolved nitrate, and 16S rRNA and metagenome sequencing.
At each station, we sampled in the upper 100 m to capture the primary nitrite maximum and an expected peak in nitrite oxidation rates at the base of the euphotic zone (EZ). We then sampled across a range of DO levels and nitrite levels to quantify rate variations in response to vertical gradients in DO. Samples were collected at 200, 100, 50, 20, 10, 5, and 1 µM [DO] at all stations.
OCR measurements were made with five replicates, providing sufficient replication to accurately measure rates. Ammonia/nitrite oxidation rate measurements, isotopic measurements, and 16S/metagenome sequencing represent single measurements owing to the expense of the analyses and their general reproducibility.
Data were recorded in field and lab notebooks while at sea or in the lab by JMB, JMW, and EPC, with assistance from the other authors, and then immediately transferred to spreadsheets.
Samples were collected in April 2017 and June 2018 (in order to capture any annual variation in conditions) aboard the R/V Oceanus. The sampling region ranged from 16 to 27.4 N and 106.5 to 117.5 W.
For water column profiles, OCR rate values calculated at 10-14 hour and 20-24 hour measurement time points were highly correlated with each other (r2= 0.968-0.995; slopes=1.06-1.19; all P<0.0001 across different stations), indicating that OCR did not accelerate or decrease substantially over the course of the incubations. The only exceptions were three sampling depths from station 3 (77, 87, 97 m) showing nonlinearity; following the suggestions of the reviewers, we instead use data from duplicate OCR measurements conducted at similar depths (75, 88, 100 m) during the previous 24 hour period (when nitrite oxidation rates were not measured in tandem).
Oxygen manipulation experiments were conducted in 500 mL serum bottles with attached FireSting sensor spots. Experiments were repeated successfully 7 times. For each experiment, a total of 24 bottles were filled with water collected at a specific depth, sealed, and then bubbled with ultrapure He gas while DO was monitored. 8 bottles had tracer-level 15NO2-additions, 8 bottles had tracerlevel 15NH4+ additions, and 8 were unlabeled. In each set of 8 bottles, we established initial DO values typically ranging from 10s to 1000s of nM. OCR was measured in all bottles based on starting and ending DO values, and samples for nitrite oxidation were collected from 15NO2-labeled bottles at the end of the experiments. For all experiments, dedicated bottles were used for the different 15N labels.
In all cases, different depths/water samples were collected randomly into differently numbered sample bottles. For experiments, bottles were allocated randomly into treatment groups. Bottle numbers were recorded while sampling, and only bottle numbers (not sample numbers) were noted during data collection phase. After analysis, bottle numbers and samples were matched.
All measurements in the field and lab were made on numbered samples, such that all analyses were blinded.
Air and water temperatures increased from north to south given the significant differences in latitude. Conditions were calm in 2017; however, in 2018, a Category 4 hurricane disrupted sampling.
Station locations are provided in Figure 1, and details of sample depths are provided in the Supplement.