Performance of Marine Anammox Candidatus Scalindua sp. under High Nitrate Conditions in a Biofilm Reactor

Abstract

To investigate the NO₃^– tolerance of Candidatus Scalindua sp., a continuous reactor was gradually exposed to increasing NO₃^– concentrations up to 3,200 mg N L^–1. High NH₄⁺ and NO₂^– removal efficiencies were maintained up to 2,600 mg N L^–1, above which performance declined and Ca. Scalindua relative abundance decreased to 0.8%. After one year of recovery, removal efficiencies exceeded 97%, whereas Ca. Scalindua relative abundance only reached 6.5%. EC₅₀ values for NH₄⁺ and NO₂^– were both 3,000 mg N L^–1. We demonstrated that our enriched Ca. Scalindua population tolerated NO₃^– up to 2,600 mg N L^–1, far exceeding the levels typically encountered in most human-derived wastewaters.

The anammox (anaerobic ammonium oxidation) process is a chemoautotrophic nitrogen removal process by which ammonium (NH₄⁺), the electron donor, is transformed into dinitrogen gas (N₂) using nitrite (NO₂^–) as the electron acceptor (Strous et al., 1999; Kartal et al., 2012). This process is performed by anaerobic ammonium-oxidizing bacteria (AnAOB) belonging to the phylum Planctomycetes and the order Candidatus Brocadiales. Among these, Candidatus Scalindua is the only known marine anammox genus (Zhang and Okabe, 2020). Due to its limited need for external carbon and its low emission of greenhouse gases, this process has been recognized as a cost-effective and environmentally friendly nitrogen removal method from wastewater, including municipal (Zhao et al., 2022), industrial (Li et al., 2018), and, more recently, aquaculture origins (Micolucci et al., 2023).

However, AnAOB are sensitive to changes in environmental conditions, including a suboptimal concentration of nitrogen waste originating from the environment or locally produced by cohabitating organisms in the biofilm. Anammox bacteria are generally not present as a monoculture, they cohabit with other bacteria, such as ammonia-oxidizing bacteria (AOB), nitrite-oxidizing bacteria (NOB), and heterotrophic bacteria (HB). Cooperation and competition among these microorganisms are critical to the stability and performance of the anammox process within a reactor (Li et al., 2018). Since anammox bacteria have a slow growing rate (Awata et al., 2013), any changes in key nitrogen parameters, including nitrate (NO₃^–) produced by NOB (within the granule or upstream of the reactor) or already present in industrial wastewater, may impair its normal functioning and change its abundance withing the granule (Roques et al., 2021; Roques et al., 2024). Therefore, the specific requirements for optimal wastewater treatment under all system conditions, including NO₃^– using Ca. Scalindua, must be identified. We previously investigated the impact of NO₃^– levels typically found in marine recirculating aquaculture systems (RAS), showing that the removal efficiencies for both NH₄⁺ and NO₂^– remained high and constant (above 95%) at the highest concentration tested (400 mg N L^–1) (Roques et al., 2024). However, we also observed a decline in the relative abundance of our Ca. Scalindua population in favor of an expansion of HB potentially capable of NO₃^– reduction (Roques et al., 2024).

Fish farmers typically initiate actions when NO₃^– levels reach approximately 100 mg N L^–1 to avoid health and welfare impairments (Banerjee et al., 2023; Roques et al., 2024); however, concentrations approaching 1,000 mg N L^–1 have been reported in poorly controlled RAS (Van Rijn, 2013). Similar concentrations have been found in industrial wastewater from cellophane, fertilizer, pectin, metal finishing, and explosives industries, reaching up to 3,000 mg N L^–1 for the latter (Fernández-Nava et al., 2008; Cyplik et al., 2012). Therefore, the aim of the present study was to investigate the maximum tolerance of Ca. Scalindua to NO₃^– with a view to its application to wastewater treatment systems in which NO₃^– levels may be elevated and, thus, a limiting factor for the anammox process.

The granular biomass (6.5 g wet weight) from a Ca. Scalindua-dominated enrichment reactor cultivated at the University of Gothenburg (Sweden) since 2019 (Micolucci et al., 2023), which was originally derived from a long-term up-flow anammox enrichment reactor at Hiroshima University (Japan) (Kindaichi et al., 2011), was used as the inoculum for a new reactor, operated for 956 days in five experimental phases (Table 1, Fig. 1).

Table 1.Operational conditions of the column reactor in three experimental phases. Values represent the average (±standard deviation). HRT: hydraulic retention time

Phase	Period (day)	[NO3– N] (mg L–1)	HRT (h)	pH influent	pH effluent	Nitrogen loading rate (g TN L–1 day–1)	Nitrogen removal rate (g TN L–1 day–1)	NH4+ and NO2– loading rate (g TN L–1 day–1)	NH4+ and NO2– removal rate (g TN L–1 day–1)
1	0–52	0	4.9 (0.4)	7.13 (0.13)	7.47 (0.15)	0.30±0.04	0.21±0.07	0.30±0.04	0.22±0.06
2	53–170	200 → 3,200	5.3 (1.0)	7.19 (0.23)	7.55 (0.21)	6.06±4.29	0.28±0.57	0.28±0.03	0.23±0.07
3	171–368	1,600	6.3 (1.8)	7.34 (0.21)	7.65 (0.18)	6.59±1.37	0.45±0.40	0.26±0.05	0.21±0.07
4	369–589	2,000 → 3,000	4.7 (0.5)	7.33 (0.32)	7.50 (0.29)	13.69±2.58	0.23±0.86	0.35±0.05	0.25±0.07
5	589–956	0	3.9 (0.6)	7.65 (0.39)	7.78 (0.47)	0.77±0.14	0.71±0.15	0.77±0.14	0.75±0.14

Fig. 1. Anammox performance in the reactor. (A) NH₄⁺ (closed diamonds) and NO₂^– (open triangles) removal efficiencies (%). Purple arrows indicate biomass sampling on days 38, 138, 167, 340, 469, and 956. (B) Expected NO₃^– concentrations (dotted lines and open circles).

During Phase 1 (Stabilization, days 0–52), the reactor was fed with the same standard synthetic marine wastewater feed supplemented with NH₄⁺, NO₂^–, inorganic carbon (KHCO₃), and a mineral and trace elements mix (van de Graaf et al., 1996; Roques et al., 2021). During Phase 2 (days 53–170), the reactor was exposed to gradually increasing concentrations of NO₃^–, supplied as sodium nitrate (NaNO₃), starting from 200 mg N L^–1 and reaching 3,200 mg N L^–1 on day 141. Following a reduction in removal efficiencies after 30 days at 3,200 mg N L^–1, NO₃^– concentrations were decreased and subsequently maintained at 1,600 mg N L^–1 for recovery during Phase 3 (days 171–368). During Phase 4 (days 369–589), the reactor was exposed to gradually increasing concentrations of NO₃^–, starting from 2,000 mg N L^–1 and reaching 3,000 mg N L^–1 on day 513. During Phase 5 (Recovery, days 590–956), the reactor was operated without the addition of NO₃^–, while NH₄⁺ (56 mg N L^–1), NO₂^– (68 mg N L^–1), and KHCO₃ (1,500 mg L^–1) were increased. All other environmental parameters were kept unchanged (Table 1, Fig. S1).

During Phase 1, NH₄⁺ and NO₂^– removal efficiencies reached 75–100% after 18 days (Fig. 1), while NO₃^– exhibited a high negative removal efficiency (i.e., it was produced) of ca. –600%. Ca. Scalindua was also the most dominant species in the granule (28.3%, Fig. 2), indicating the successful establishment of the anammox process (Kindaichi et al., 2011). During Phase 2, high NH₄⁺ (87.7±7.8%) and NO₂^– (95.4±4.1%) removal efficiencies were maintained until NO₃^– concentrations reached 3,200 mg N L^–1, at which point the reactor crashed (Fig. 1 and S2). During this phase, the relative abundance of Ca. Scalindua decreased to 12.6% when chronically exposed to 1,600 mg N L^–1 (day 138) and further declined to 0.8% after 30 days at 3,200 mg N L^–1 (day 168). NO₃^– removal efficiencies during these phases were 2.2±5.8 and 2.4±3.7%, respectively, indicating net NO₃^– removal, representing only‍ ‍a small fraction of the total NO₃^– pool under these conditions. In parallel, there were increases in Firmicutes‍ ‍(Bacillaceae), Alphaproteobacteria (mainly Rhodobacteraceae), and Bacteroidota (Flavobacteriaceae and Saprospiraceae), which replaced Ca. Scalindua and Actinobacteria (order Actinomarinales) (Fig. 2). Several of these enriched taxa include lineages reported to possess NO₃^–-reducing or denitrifying capabilities under anoxic conditions (Parkes et al., 2014; Mandic-Mulec et al., 2016; Shan et al., 2024). However, since denitrification intermediates and functional genes were measured in this study, it was not possible to directly quantify the contribution of denitrification.

Fig. 2. Microbial community composition at the end of Phase 1 (day 38), during Phase 2 after 48 days of exposure to 1,600 mg N L^–1 (day 138), after 30 days of exposure to 3,200 mg N L^–1 (day 167), at the end of Phase 3 (after 5.5 months of constant exposure to 1,600 mg N L^–1, day 340), in the middle of Phase 4 (2,800 mg N L^–1 day 469), and at the end of Phase 5 (0 mg N L^–1, recovery, day 956). An anal­ysis based on 16S rRNA gene amplicon sequencing. Red percentages correspond to the relative abundance of the marine anammox Ca. Scalindua.

NH₄⁺ and NO₂^– removal efficiencies were restored to >75% after three months under 1,600 mg N L^–1 during Phase 3 (Fig. 1). Ca. Scalindua relative abundance increased to 13.1%, while the other groups, with the exception of Melioribacteraceas, returned to pre- NO₃^– exposure levels (Fig. 2).

In Phase 4, the reactor started to become unstable again from exposure to 2,600 mg N L^–1 onward, where NH₄⁺ and NO₂^– removal efficiencies fell below 50% (Fig. 1). Consequently, the relative abundance of Ca. Scalindua decreased to 5.3% (Fig. 2). Net NO₃^– removal efficiencies were negligible and highly variable during this phase. Following the removal of NO₃^– during Phase 5, high NH₄⁺ (97.2±2.6%) and NO₂^– (99.7±0.7%) removal efficiencies were achieved again (Fig. 1), and were accompanied by large net NO₃^– production (–920±265%). This stoichiometric pattern is consistent with the reestablishment of anammox as the dominant nitrogen transformation pathway in the reactor (Kindaichi et al., 2011).

Despite NH₄⁺ and NO₂^– being the sole nitrogen sources supplied in Phase 5, Ca. Scalindua relative abundance remained moderate (6.5%) rather than being further enriched. Firmicutes quasi disappeared, while the percentages of some Proteobacteria, Bacteroidota (including Flavobacteriaceae), Chloroflexi, and Planctomycetes (excluding Ca. Scalindua) were higher than the pre- NO₃^– exposure phase (Fig. 2). The inoculum originated from a long-term enrichment reactor in which Ca. Scalindua was dominant but coexisted with other taxa. Throughout operation, all major bacterial groups were already present from the start, and no new taxa emerged; their relative abundance shifted in response to operational conditions. Strong biomass retention provided by the non-woven fabric carrier promoted attached-growth biofilm formation, minimizing washout and allowing multiple functional guilds to persist within the reactor (Ren et al., 2018).

Importantly, no alternative anammox lineage increased, indicating that fluctuations reflected changes in the abundance of Ca. Scalindua rather than clade replacement. The marked decline in Ca. Scalindua during high NO₃^– exposure followed by partial recovery may reflect a bottleneck event after which a more tolerant fraction of the population persisted. Given the slow growth of anammox bacteria (Awata et al., 2013), full return to initial dominance may be limited. However, strain-level selection cannot be resolved from our 16S rRNA gene data.

Regarding non-anammox community changes, Firmicutes (Bacillaceae) are mostly fast-growing generalist heterotrophs that are capable of fermentation and NO₃^– reduction or denitrification and may form spores, making them relatively resistant to environmental stress (Parkes et al., 2014; Mandic-Mulec et al., 2016). They are known to thrive during early community shifts under perturbations, such as that observed during Phase 2, before disappearing once conditions become more stable or favorable for other bacterial groups thriving on high NO₃^–. A similar pattern was reported by Zhang et al. (2022) where 25–100 mg N L^–1 led to a decrease in the abundance of Firmicutes in groundwater. This is consistent with our previous study where Firmicutes also almost disappeared after a chronic exposure to 400 mg N L^–1 (Roques et al., 2024). The anaerobic versatile Alphaproteobacteria (HB) may have benefited from the higher NO₃^– sensitivity of Firmicutes to outcompete this phylum in the long term (Qu et al., 2021; Roques et al., 2024). Bacteroidota, Flavobacteriaceae, Saprospiraceae, and, to a lesser extent, photosynthetic Chloroflexi likely served as organic matter recyclers, scavenging dead Firmicutes or other bacteria to flourish after chronic exposure to high NO₃^– levels (Yang et al., 2018; Roques et al., 2024).

Based on the removal efficiencies of both NH₄⁺ and NO₂^–, concentration–response curves for NO₃^– were generated (Fig. 3). Half maximal effective concentration values (EC₅₀) and their 95% confidence intervals were 3,000 (2,930–3,090) and 3,000 (2,910–3,110) mg N L^–1 for NH₄⁺ and NO₂^–, respectively, indicating an exceptionally high tolerance of Ca. Scalindua to elevated NO₃^– concentrations.

Fig. 3. Concentration-response curves for NH₄⁺ and NO₂^– as a function to ambient NO₃^– concentrations (in mg N L^–1). The curves were calculated using data obtained from day 18 (end of the stabilization period in Phase 1) to day 589 (end of Phase 4). EC₅₀ values (95% confidence intervals) were 3,000 (2,930–3,090) and 3,000 (2,910–3,110) mg N L^–1 for NH₄⁺ and NO₂^–, respectively. The fitted models were y=88.8*(1+exp[12.1*{log x–8.0}])^–1 (R²=0.932) for NH₄⁺ and y=97.4 * (1+exp[9.3*{log x–8.0}])^–1 (R²=0.952) for NO₂^–.

Although we noted a reduction in the relative abundance of our population of Ca. Scalindua in the granule (from 28.3% at the beginning of the experiment to 5.3% after 19 months of exposure to fluctuating high NO₃^– concentrations and to 6.5% after one year of recovery), high NH₄⁺ and NO₂^– removal efficiencies coupled with net anammox-consistent NO₃^– production and fluorescence in situ hybridization (FISH) observations (Fig. S3) indicated the presence of a functioning, stable population of Ca. Scalindua throughout the experiment. Based on our results, we conclude that the enriched Ca. Scalindua population investigated in this study is highly tolerant to high levels of NO₃^– up to concentrations of 2,600 mg N L^–1, far exceeding the levels typically encountered in most human-derived wastewaters. Collectively, these results suggest that marine anammox systems enriched with Ca. Scalindua are broadly applicable for nitrogen removal from a wide variety of wastewater sources containing high NO₃^– levels. However, the precise cellular and molecular mechanisms underlying NO₃^–-associated inhibition and acclimation remain unclear and warrant further investigation.

Acknowledgements

The present study was conducted within the frame of the MIRAI project and was supported by FORMAS (2020-00867), Kungl. Skogs-och Lantbruksakademien (GFS2024-0148), STINT (mobility grant for internationalization, MG2019-8483) in Sweden; JSPS KAKENHI (JP23KJ1642, JP24KK0197), JSPS Bilateral Program (JPJSBP120259928) and FY2025 JSPS Invitational Fellowship for Research in Japan (S25124) in Japan. The authors thank Lise Brault, Linda Frank Hasselberg, and Thanh Nguyen Duc at the University of Gothenburg for their technical assistance.

Conflicts of interest

The authors declare that there are no conflicts of interest.

Citation

Roques, J. A. C., Fujii, N., Unegbu, E., Marqué, A., Johansson, E., Yamamoto, K., et al. (2026) Performance of Marine Anammox Candidatus Scalindua sp. under High Nitrate Conditions in a Biofilm Reactor. Microbes Environ 41: ME25094.

https://doi.org/10.1264/jsme2.ME25094

References

•  Awata,  T.,  Oshiki,  M.,  Kindaichi,  T.,  Ozaki,  N.,  Ohashi,  A., and  Okabe,  S. (2013) Physiological characterization of an anaerobic ammonium-oxidizing bacterium belonging to the “Candidatus Scalindua” group. Appl Environ Microbiol 79: 4145–4148.
•  Banerjee,  P.,  Garai,  P.,  Saha,  N.C.,  Saha,  S.,  Sharma,  P., and  Maiti,  A.K. (2023) A critical review on the effect of nitrate pollution in aquatic invertebrates and fish. Water Air Soil Pollut 234: 333.
•  Cyplik,  P.,  Marecik,  R.,  Piotrowska-Cyplik,  A.,  Olejnik,  A.,  Drożdżyńska,  A., and  Chrzanowski,  Ł. (2012) Biological denitrification of high nitrate processing wastewaters from explosives production plant. Water Air Soil Pollut 223: 1791–1800.
•  Fernández-Nava,  Y.,  Maranon,  E.,  Soons,  J., and  Castrillón,  L. (2008) Denitrification of wastewater containing high nitrate and calcium concentrations. Bioresour Technol 99: 7976–7981.
•  Kartal,  B.,  van Niftrik,  L.,  Keltjens,  J.T.,  Op den Camp,  H.J., and  Jetten,  M.S. (2012) Anammox—growth physiology, cell biology, and metabolism. Adv Microb Physiol 60: 211–262.
•  Kindaichi,  T.,  Awata,  T.,  Tanabe,  K.,  Ozaki,  N., and  Ohashi,  A. (2011) Enrichment of marine anammox bacteria in Hiroshima Bay sediments. Water Sci Technol 63: 964–969.
•  Li,  J.,  Li,  J.,  Gao,  R.,  Wang,  M.,  Yang,  L.,  Wang,  X., et al. (2018) A critical review of one-stage anammox processes for treating industrial wastewater: optimization strategies based on key functional microorganisms. Bioresour Technol 265: 498–505.
•  Mandic-Mulec,  I.,  Stefanic,  P., and  van Elsas,  J.D. (2016) Ecology of bacillaceae. In The Bacterial Spore: From Molecules to Systems. Driks, A., and Eichenberger, A. (eds). Washington, DC: ASM Press, pp. 59–85.
•  Micolucci,  F.,  Roques,  J.A.C.,  Ziccardi,  G.S.,  Fujii,  N.,  Sundell,  K., and  Kindaichi,  T. (2023) Candidatus Scalindua, a biological solution to treat saline recirculating aquaculture system wastewater. Processes 11: 690.
•  Parkes,  R.J.,  Cragg,  B.,  Roussel,  E.,  Webster,  G.,  Weightman,  A., and  Sass,  H. (2014) A review of prokaryotic populations and processes in sub-seafloor sediments, including biosphere: geosphere interactions. Mar Geol 352: 409–425.
•  Qu,  J.,  Yang,  H.,  Liu,  Y.,  Qi,  H.,  Wang,  Y., and  Zhang,  Q. (2021) The study of natural biofilm formation and microbial community structure for recirculating aquaculture system. IOP Conference Series: Earth and Environmental Science, IOP Publishing, 742: 012018.
•  Ren,  L.-F.,  Lv,  L.,  Kang,  Q.,  Gao,  B.,  Ni,  S.-Q.,  Chen,  Y.-H., and  Xu,  S. (2018) Microbial dynamics of biofilm and suspended flocs in anammox membrane bioreactor: The effect of non-woven fabric membrane. Bioresour Technol 247: 259–266.
•  Roques,  J.A.C.,  Micolucci,  F.,  Hosokawa,  S.,  Sundell,  K., and  Kindaichi,  T. (2021) Effects of recirculating aquaculture system wastewater on anammox performance and community structure. Processes 9: 1183.
•  Roques,  J.A.C.,  Unegbu,  E.,  Fujii,  N.,  Marqué,  A.,  Micolucci,  F.,  Sundell,  K.S., and  Kindaichi,  T. (2024) Tolerance of the marine anammox Candidatus Scalindua to high nitrate concentrations: Implications for recirculating aquaculture systems. Water 16: 3705.
•  Shan,  H.,  Du,  Y.,  Li,  T.,  Wang,  F.,  Li,  H., and  Wang,  H. (2024) Bacterial community characteristics and roles in nitrogen transformation in industrial farming systems of litopenaeus vannamei. J Mar Sci Eng 12: 787.
•  Strous,  M.,  Kuenen,  J.G., and  Jetten,  M.S. (1999) Key physiology of anaerobic ammonium oxidation. Appl Environ Microbiol 65: 3248–3250.
•  van de Graaf,  A.A.,  de Bruijn,  P.,  Robertson,  L.A.,  Jetten,  M.S., and  Kuenen,  J.G. (1996) Autotrophic growth of anaerobic ammonium-oxidizing micro-organisms in a fluidized bed reactor. Microbiology 142: 2187–2196.
•  Van Rijn,  J. (2013) Waste treatment in recirculating aquaculture systems. Aquac Eng 53: 49–56.
•  Yang,  W.,  He,  S.,  Han,  M.,  Wang,  B.,  Niu,  Q.,  Xu,  Y., et al. (2018) Nitrogen removal performance and microbial community structure in the start-up and substrate inhibition stages of an anammox reactor. J Biosci Bioeng 126: 88–95.
•  Zhang,  L., and  Okabe,  S. (2020) Ecological niche differentiation among anammox bacteria. Water Res 171: 115468.
•  Zhang,  X.-Y.,  Li,  Z.-L.,  Chen,  F.,  Wang,  S.-P.,  Nan,  J.,  Huang,  C., et al. (2022) Influence of nitrate concentration on trichloroethylene reductive dechlorination in weak electric stimulation system. Chemosphere 295: 133935.
•  Zhao,  Q.,  Peng,  Y.,  Li,  J.,  Gao,  R.,  Jia,  T.,  Deng,  L., and  Du,  R. (2022) Sustainable upgrading of biological municipal wastewater treatment based on anammox: From microbial understanding to engineering application. Sci Total Environ 813: 152468.