Aerobic bacteria produce nitric oxide via denitrification and promote algal population collapse

Abstract

Microbial interactions govern marine biogeochemistry. These interactions are generally considered to rely on exchange of organic molecules. Here we report on a novel inorganic route of microbial communication, showing that algal-bacterial interactions between Phaeobacter inhibens bacteria and Gephyrocapsa huxleyi algae are mediated through inorganic nitrogen exchange. Under oxygen-rich conditions, aerobic bacteria reduce algal-secreted nitrite to nitric oxide (NO) through denitrification, a well-studied anaerobic respiratory mechanism. The bacterial NO is involved in triggering a cascade in algae akin to programmed cell death. During death, algae further generate NO, thereby propagating the signal in the algal population. Eventually, the algal population collapses, similar to the sudden demise of oceanic algal blooms. Our study suggests that the exchange of inorganic nitrogen species in oxygenated environments is a potentially significant route of microbial communication within and across kingdoms.

Fig. 1. Algal nitrite secretion during exponential growth is linked to the timing of algal death in co-cultures.

A Assimilatory nitrate reduction to ammonium during algal biomass assimilation produces nitrite as an intermediate. Dissimilatory nitrate reduction to di-nitrogen during bacterial denitrification also includes nitrite as an intermediate, which is further reduced to NO. The main nitrogen species discussed in the manuscript are shaded in gray. B Nitrate and (C) nitrite detected in filtrates of axenic algal cultures on the indicated days. D Daily monitoring of nitrite in filtrates of axenic algal cultures (light green) and co-cultures (dark green). E, F Algal growth (bars) and bacterial growth (black line) in axenic algal cultures (light green bars) and algal-bacterial co-cultures (dark green bars). E Cultures initiated with an algal inoculum of 10³ algae/ml. F Cultures initiated with an algal inoculum of 10⁵ cells/ml. Black arrows indicate the day of the extracellular nitrite peak as shown in (D) and in Fig. S1D. Each data point in the figure consists of at least 3 biological replicates, error bars designate ±SD. Statistical significance was calculated using a two-sample t test to compare algal cell counts in axenic cultures versus in co-cultures. E Algal cell counts on day 17 and (F) on day 10 were significantly different between axenic and co-cultures and resulted in p < 0.01.

Fig. 2. Bacteria produce and secrete NO under oxic conditions.

Relative gene expression of P. inhibens denitrification genes under exposure to 10 µM nitrite for the indicated times; (A) nirK, (B) norB. Each data point consists of 3 biological replicates, error bars designate ±SD. C Microscopy images of WT and Δ262 bacteria cured from the native 262 kb plasmid, stained with the fluorescent NO-indicator diacetate DAF-FM, and incubated with or without 100 µM nitrite for 2 h. Scale bar corresponds to 1 µm. D Extracellular nitric oxide concentrations of WT and Δ262 bacteria incubated with the indicated nitrite concentrations. Each data point consists of 3 biological replicates, each containing 10⁷ cells/ml. Error bars designate ±SD. ND stands for not detected.

Fig. 3. Bacterial NO production by denitrification is involved in triggering algal death.

A Algal growth (bars) and bacterial growth (lines) in co-cultures with WT bacteria (green bar and line) and Δ262 bacteria cured from the native 262 kb plasmid (gray bar and black line). Each data point consists of 3 biological replicates, error bars designate ±SD. Statistical significance was calculated using a two-sample t test to compare algal cell counts in co-cultures with WT bacteria versus in co-cultures with Δ262 bacteria. Algal cell counts were significantly different on day 14 and 17 and resulted in p < 0.01. Statistical significance was calculated using a two-sample t test to compare the growth of WT bacteria versus Δ262 bacteria in co-cultures and did not result in significant differences. B Algae were co-cultured with either WT, ΔnirK or ΔnnrS→nirV bacteria. Co-culturing was conducted as described in materials and methods. Algal death was monitored manually every 6–8 h and death was defined as the complete change of the culture color from green to white. The timing of algal death was determined as follows: Co-cultures grew for 10 days and then algal death was monitored. The first culture that exhibited algal death was enumerated as 1 h, and the timing of algal death in all other co-cultures was measured from that point on (i.e. a culture that exhibited algal death 48 h after the first culture, was enumerated as 48). Each type of co-culture (with WT, ΔnirK or ΔnnrS→nirV bacteria) included 6 biological replicates. Box-plot elements are: center line—median; box limits—upper and lower quartiles; whiskers—min and max values, point—outlier. Statistical significance was calculated using a two-sample t test to compare timing of algal death between co-cultures with WT versus ΔnirK bacteria and between WT versus ΔnnrS→nirV bacteria. The difference between the timing of algal death in co-cultures with WT versus ΔnnrS→nirV bacteria was significant and resulted in p < 0.01. C Relative expression of nirK in bacterial cells from an algal-bacterial co-culture (corresponding to co-cultures in A), sampled on the indicated days. Data is relative to the expression on day 10. Each data point consists of 4 biological replicates, error bars designate ±SD. ND stands for not detected. D Fluorescence of algal cells stained with the fluorescent NO-indicator diacetate DAF-FM, incubated with live or dead bacteria for 2 h under the following conditions: control (– nitrite, light green), exposure to nitrite ( + nitrite [10 µM], dark green), or exposure to a chemical NO donor ( + NO donor, [300 µM] DEANO, black). Results represent 3 biological replicates, each containing 10,000 algal cells and 10⁷ bacterial cells when indicated. Error bars designate ±SD. Statistical significance was calculated using a two-sample t test to compare DAF-FM fluorescence between treatments. DAF-FM fluorescence between algal treatment with or without nitrite versus NO donor treatment was significant and resulted in p < 0.01. DAF-FM fluorescence of algae treated with live bacteria with nitrite versus without nitrite was significant and resulted in p < 0.01. E The effect on the death of axenic algal cultures upon adding (+) a NO donor (a single dose of 100 µM DEANO) and a NO scavenger (20 µM c-PTIO). Red checkmark indicates observed algal death. Results represent at least 3 biological replicates. ND stands for not detected.

Fig. 4. Extracellular NO triggers a PCD-like process in algae, followed by algal NO production and secretion.

A Relative gene expression of algal oxidative stress and PCD-like genes following incubation with a NO donor ([100 µM] DEANO for 18 h). Annotated gene products are depicted (Table 4). B Extracellular NO concentrations of axenic algal cultures and algal-bacterial co-cultures on day 10. C Algal growth in axenic algal cultures (light green), co-cultures (dark green) and co-cultures supplemented with a NO scavenger ([20 µM] c-PTIO added on day 9 of the co-culture, black). All data points in the figure consist of 3 biological replicates, error bars designate ±SD. Statistical significance was calculated using a two-sample t test to compare algal cell counts on day 13 in co-cultures versus algal cell counts in co-cultures treated with NO scavenger. Algal cell counts were significantly different and resulted in p < 0.05.

Fig. 5. Expression of nirK in oxygenated marine environments.

Transcript abundances of selected denitrification genes (napA, norB, nosZ and nirS) decreased with increasing oxygen concentrations. In contrary, transcript abundances of nirK were frequently high in phytoplankton-rich water layers (represented by the deep chlorophyll maximum, green dots) in which oxygen levels of 100–400 µM were detected, likely produced by phytoplankton. In the same samples, nitrite levels of up to 1.5 µM were measured. Transcript abundances were fit with a linear regression model (black line) and a 95% confidence interval (yellow shade). Mixed water layer samples originated from the epipelagic region but could not be classified to the surface or deep chlorophyll maxima. Nitrite concentrations <0.001 µM are not shown. In brackets: KEGG orthologous group; napA: periplasmic nitrate reductase; norB: nitric oxide reductase subunit B; nosZ: nitrous-oxide reductase; nirS: nitrite reductase (NO-forming) / hydroxylamine reductase; nirK: nitrite reductase (NO-forming). Data adapted from Salazar et al., 2019 [67], see Table 5.

Fig. 6. A model depicting inorganic nitrogen exchange during algal-bacterial interactions.

Inorganic nitrogen exchange underlies the algal-bacterial dynamic interaction. Algae secrete nitrite in a growth-phase dependent manner, with a nitrite peak during mid exponential growth. Bacteria reduce algal-secreted nitrite to NO via denitrification. NO secreted by bacteria triggers algal death at stationary phase, during which algae produce NO. Algal-secreted NO propagates the death signal among algal cells, causing the collapse of the algal population.