Nitrite oxidation in the Namibian oxygen minimum zone

Abstract

Nitrite oxidation is the second step of nitrification. It is the primary source of oceanic nitrate, the predominant form of bioavailable nitrogen in the ocean. Despite its obvious importance, nitrite oxidation has rarely been investigated in marine settings. We determined nitrite oxidation rates directly in (15)N-incubation experiments and compared the rates with those of nitrate reduction to nitrite, ammonia oxidation, anammox, denitrification, as well as dissimilatory nitrate/nitrite reduction to ammonium in the Namibian oxygen minimum zone (OMZ). Nitrite oxidation (≤372 nM NO(2)(-) d(-1)) was detected throughout the OMZ even when in situ oxygen concentrations were low to non-detectable. Nitrite oxidation rates often exceeded ammonia oxidation rates, whereas nitrate reduction served as an alternative and significant source of nitrite. Nitrite oxidation and anammox co-occurred in these oxygen-deficient waters, suggesting that nitrite-oxidizing bacteria (NOB) likely compete with anammox bacteria for nitrite when substrate availability became low. Among all of the known NOB genera targeted via catalyzed reporter deposition fluorescence in situ hybridization, only Nitrospina and Nitrococcus were detectable in the Namibian OMZ samples investigated. These NOB were abundant throughout the OMZ and contributed up to ~9% of total microbial community. Our combined results reveal that a considerable fraction of the recently recycled nitrogen or reduced NO(3)(-) was re-oxidized back to NO(3)(-) via nitrite oxidation, instead of being lost from the system through the anammox or denitrification pathways.

Figure 1

(a) Location of sampling stations (solid circles with station numbers) over the Namibian shelf in relation to the bathymetry of the region. The full water depths at these stations ranged between 103 and 131 m. (b) Vertical distribution of oxygen, (c) nitrite, (d) ammonia, (e) nitrate and (f) chlorophyll a (measured as fluorescence in arbitrary units, without cross calibration with absolute quantities of chlorophyll), along a NW-SE transect over the Namibian shelf. Sampling sites are indicated by black dots along the water column (the figure has been plotted with Ocean Data View (ODV) (Schlitzer, 2011)).

Figure 2

Stations 252 and 225: High-resolution vertical profiles of oxygen and 10 × its concentration in the OMZ and BBL (O₂ gray circles, O₂ × 10 open diamonds) (left panels) and of nutrients (NO₂⁻, black solid circles; NH₄⁺, gray triangles; NO₃⁻, open circles) (central panels). NO₂⁻ oxidation rates measured as ¹⁵NO₃⁻ production are indicated by horizontal bars (right panels). At each depth, cellular abundance of Nitrococcus (triangle) and Nitrospina (circle) are shown. Most nitrite oxidation rates presented are derived from significant slopes (P<0.05) in corresponding linear regression. On occasions, ¹⁵NO₃⁻ was produced so rapidly that a maximum was reached within as little as ∼12 h, therefore rates were calculated only from these first 12 h. Despite their substantial and instantaneous ¹⁵NO₃⁻ production and the apparently high correlation between ¹⁵NO₃⁻ and time (r²>0.90), the low degrees of freedom resulted in slightly higher P-values (0.07–0.20) from one-way ANOVA in a few experiments—St 225: 85 m and 118 m; St 252: 105, 109 and 110.7 m. ‘n.d.' denotes non-detectable reaction rates.

Figure 3

(a) Influence of O₂ concentration on nitrite oxidation: ¹⁵NO₃⁻ production rates were determined in incubations with various controlled O₂ levels in two water samples—station 206 at 100 m and station 252 at 105 m. (b) NOB abundance as detected with 16S rRNA-based CARD-FISH in relation to the measured nitrite oxidation rates.

Figure 4

Epifluorescence micrographs showing dense populations of Nitrospina (green) (a) and Nitrococcus (green) (b) cells in large aggregates with other microorganisms (blue) at station 206 in 110 m water depth, as detected with 16S rRNA-based CARD-FISH.

Figure 5

Comparison of nitrite sources and sinks at stations 243 and 252. Rates of NO₂⁻ consuming processes, include anammox (dark green), nitrite oxidation (light green) and DNRA (red). NO₂⁻ producing processes, include ammonia oxidation (light blue) and nitrate reduction (dark blue). The rates were determined from parallel incubation experiments that have been conducted with the same set of samples. Most rates presented are derived from significant slopes (P<0.05) in corresponding linear regression. On occasions, ¹⁵NO₃⁻/¹⁵NO₂⁻ was produced so rapidly that a maximum was reached within as little as ∼12 h, so rates were calculated only from these first 12 h. Despite their substantial and instantaneous ¹⁵NO₃⁻/¹⁵NO₂⁻ production and the apparently high correlation between ¹⁵NO₃⁻/¹⁵NO₂⁻ and time (r²>0.90), the low degrees of freedom resulted in slightly higher P-values (0.07–0.20) from one-way ANOVA in a few experiments—Nitrite oxidation: St 243: 97m; St 252: 105, 109 and 110.7 m; Nitrate reduction: St 252, 76 and 109 m. ‘n.d.' denotes non-detectable reaction rates.