Contrasting effects of osmolytes on nitrifying biofilms under salinity stress

Several industries produce high or variable salinity effluents. This can be challenging for the microorganisms involved in the biological water treatment of these effluents, especially the nitrifying microorganisms. Some microorganisms can adapt to a salinity increase through the uptake of certain molecules called osmolytes (or osmoprotectants) from the environment. This salinity acclimation strategy has been effective over a range of microorganisms. Thus, osmolyte addition could be a sustainable strategy for osmoregulation, but it has never been investigated in nitrifying biofilms. In this study, we investigated the impact of adding an osmolyte cocktail (1 mM each of trehalose, sucrose, glycine betaine, proline, carnitine, and ectoine) on the functionality of nitrifying biofilms undergoing a salinity increase from freshwater to seawater. The experiment was conducted on moving bed biofilm reactors (MBBR) operated in a sequencing batch mode. The osmoprotectants did not improve the nitrification activity on the first day after seawater transfer. Moreover, after 2 days in seawater, the treatment with osmolytes showed a severe reduction in the nitrification activity. This was accompanied by the growth of heterotrophic microorganisms in the medium facilitated by the uptake of osmolytes as substrate. Thus, the reduction in nitrification activity was likely due to competition between the heterotrophs and nitrifiers for resources (such as oxygen) and/or osmolytes. This study highlights the complex effects of the addition of osmoprotectants on biofilms undergoing a salinity change. Future studies should investigate the impact of individual osmoprotectants, as their potential as growth substrate and as osmoregulators may vary.

About 5% of total annual wastewater produced worldwide is saline (Vyrides, 2015). Several industries produce high or variable salinity effluents, such as oil refineries, aquaculture, seafood processing and tanneries (Vyrides, 2015; Lefebvre and Moletta, 2006). Seawater is also increasingly being used to alleviate water shortages leading to increased salinity in municipal wastewater treatment systems (Fu et al., 2022). Variable salinity is challenging for biological wastewater treatment processes because it can affect the metabolism and activity of the microorganisms by causing osmotic stress (Madigan et al., 2018; Sleator and Hill, 2002). In particular, hyperosmotic (salinity increase) changes are more detrimental than hypoosmotic (salinity decrease) changes (Csonka, 1989). The most common strategy is to gradually adapt the microorganisms to the new salinity, but the adaptation period can be very long (weeks to months) (Vyrides, 2015; Bassin et al., 2012; Gonzalez-Silva, 2016). Moreover, in many systems, it is common to have great variations due to process fluctuations rather than a gradual increase in salinity (Lefebvre and Moletta, 2006) and salinity may be different during different periods (Vyrides, 2015). Inoculation with salt-adapted microorganisms or with salt-adapted sludge has been shown to improve salinity acclimation (Cui et al., 2016; Sudarno et al., 2010; Vyrides and Stuckey, 2017). However, this strategy may not work during sudden increases in salinity and suitable inocula can be expensive or difficult to procure (Vyrides, 2015). Therefore, a more effective strategy is required for rapid salinity adaptation in microorganisms.

In principle, microbial cells must maintain an intracellular osmotic pressure greater than that of the environment to enable growth and cell division (Sleator and Hill, 2002). Microorganisms use two main strategies to adapt to salinity increase–i) the “salt-in” strategy and ii) the compatible solute strategy (Sleator and Hill, 2002; Csonka, 1989). In the “salt-in” strategy, the microorganisms increase their intracellular ion concentration to balance the external osmolarity (Vyrides and Stuckey, 2017). Owing to the high intracellular ionic strength, extensive structural adaptions are required, thus making this strategy exclusive to strictly halophilic bacteria (Sleator and Hill, 2002). In the second strategy, the primary response is an increase in the concentration of K + (and its counter-ion glutamate), followed by a secondary response of an increase in the cytoplasmic concentration of compatible solutes through synthesis or uptake (Sleator and Hill, 2002). This strategy offers a greater degree of flexibility and is commonly used by halotolerant bacteria. Halotolerant microorganisms have a bi-phasic response to salinity increase. Compatible solutes, also known as osmolytes, are highly soluble molecules that do not interact with proteins, thus enabling them to accumulate at high intracellular concentrations without interfering with the cell function (Sleator and Hill, 2002). Not just one, but several osmolytes may be associated with salinity adaption, depending on the magnitude of salinity change and exposure time (Vyrides and Stuckey, 2017; Saum and Müller, 2007). The synthesis of osmolytes depends not only on the salinity but also on the availability of nutrients (Schimel et al., 2007). A significant amount of carbon and nitrogen are consumed during osmolyte production, leading to a 2-3 fold increase in the energetic cost for the microorganisms (Schimel et al., 2007). Thus, the synthesis of osmolytes is energetically expensive, and the uptake of osmolytes from the environment is an energetically cheaper option (Oren, 2011). As opposed to osmolytes that accumulate inside the cells, osmoprotectants are compounds that stimulate bacterial growth at high osmolality when provided in the growth medium (Wood, 2007). Adding the osmoprotectant in the beginning of a salinity exposure is preferable to a delayed addition to prevent the irreversible inhibition effects of salinity stress on cells (Vyrides and Stuckey, 2017). Among the commonly used osmolytes and osmoprotectants are sugars (sucrose, trehalose, etc.) and amino acids (glycine betaine, carnitine, proline, ectoine, choline, etc.) (Sleator and Hill, 2002; Vyrides and Stuckey, 2017; Oren, 1999). Several studies have shown that the exogenous addition of osmolytes can alleviate salinity stress across a wide variety of microorganisms (Vyrides and Stuckey, 2017).

While several microbial communities such as methanogens, denitrifiers and anammox have been found capable of osmoprotectant uptake, little is known about the salinity acclimation mechanisms in nitrifying microorganisms (Vyrides and Stuckey, 2017). Nitrification is a two-step process consisting of ammonia oxidation by ammonia oxidizing bacteria (AOB) and archaea (AOA), followed by nitrite oxidation by nitrite oxidizing bacteria (NOB). Within the genus Nitrospira, some species can perform both the nitrification steps (comammox) (Daims et al., 2015). Compared to heterotrophs, relatively less energy is produced during the autotrophic oxidation by nitrifiers, and the energy produced may not be sufficient to sustain growth at very high salt concentrations (Oren, 2011). Nonetheless, there are several marine nitrifiers that have an obligatory requirement for saline environments. In particular, AOA are known to thrive in seawater. Moreover, nitrifiers are found at high abundance in estuaries that frequently encounter salinity fluctuations, suggesting that several genera of nitrifying microorganisms are halotolerant (Bernhard and Bollmann, 2010; Santos et al., 2018; Ward et al., 2007). This is also supported by studies on nitrifying bioreactors where the same taxa were present across different salinities (Gonzalez-Silva, 2016; Navada et al., 2020a). A proteomic study suggested that the osmolytes sucrose and glycine were produced in response to a salinity increase in Nitrosomonas europaea (Ilgrande et al., 2018).

We hypothesize that the uptake of osmolytes may alleviate salinity stress in nitrifying microorganisms as it is energetically more favorable than de novo synthesis. We are aware of only a few studies on the effect of external osmolytes on nitrifiers, and with mixed results. In one study, the addition of a cocktail of osmolytes (betaine, trehalose, proline, 3-[N-morpholino]propanesulfonic acid, taurine, and γ-amino- n -butyric acid) did not enhance microcolony formation of N. europaea at salt concentration >0.1 M NaCl (∼6‰ salinity) (Wood and Sørensen, 1998). However, the study used a solution of sodium chloride, and was thus lacking K + ions present in seawater that play an important role in osmoregulation (Sleator and Hill, 2002; Vyrides and Stuckey, 2017). In another study, Nitrobacter was capable of trehalose production and uptake of glycine betaine and sucrose from the medium (Vyrides and Stuckey, 2017). In a third study, the concentration of glutamine and proline increased in aerobic nitrifying granules, but no other osmolytes were analyzed (Wan et al., 2014). Recent studies show that the addition of osmolytes, such as glycine betaine and mannitol, can alleviate the inhibitory effects of salinity on AOB in activated sludge (Fu et al., 2022; Zuo et al., 2021). However, as far as we know, there exist no studies on the exogenous addition of osmolytes on nitrification in aerobic microbial communities with both nitrifying and heterotrophic bacteria, such as nitrifying biofilms.

Nitrifying biofilms are widely used in wastewater treatment, and nitrification is an important water treatment process for removing ammonia and nitrite that can be toxic to aquatic life. But nitrifying biofilms can be sensitive to salinity changes (Gonzalez-Silva, 2016; Navada and Vadstein, 2022). Therefore, it is important to minimize the negative impact of salinity increase on these biofilms. One potential strategy could be through the addition of osmolytes during a salinity increase. This is relevant, for instance, in recirculating aquaculture systems for anadromous species, such as Atlantic salmon, where the salinity must be changed during specific life stages of the fish (Navada et al., 2021), or during salinity fluctuations in industrial or municipal wastewater systems. The genome of some nitrifying bacteria have been reported to contain genes encoding the transport of osmolytes such as proline, ectoine, glutamate and betaine, suggesting that these can potentially be taken up for salinity acclimation (Wu et al., 2024). Thus, if the added osmolytes are effectively taken up by the nitrifiers for osmoregulation, it would improve the functionality of biofilm reactors, thus rendering water treatment systems more robust to salinity changes.

The goal of this study was to investigate the effect of the exogenous addition of osmolytes on nitrifying biofilms undergoing salinity increase from fresh- to seawater. Important osmolytes involved in bacterial osmoadaptation include glycine betaine, carnitine and proline (Sleator and Hill, 2002). The intracellular concentration of proline was reportedly increased during salinity increase in a study on aerobic nitrifying granules (Vyrides and Stuckey, 2017). N. europaea and Nitrosomonas winogradskyi have genes for production of glycine betaine and sucrose (Ilgrande et al., 2018), and Nitrobacter was found to accumulate these osmolytes from the medium under high salinity (Vyrides and Stuckey, 2017). Trehalose and ectoine are well known common osmoprotectants in several organisms (Elbein et al., 2003; Jebbar et al., 1992). Thus, a cocktail of the following common osmoprotectants was chosen: trehalose, sucrose, glycine betaine proline, carnitine, and ectoine. We hypothesized that the exogenous addition of osmolytes would alleviate salinity stress and lead to a lower reduction in nitrification rate upon salinity increase.

Experimental design and setup

Aerated lab beakers (0.5 L water volume) were used as moving bed biofilm reactors (MBBR). These were operated in a sequencing batch mode with synthetic medium exchanged every day (∼24 h hydraulic retention time (HRT)). Three treatments were operated in triplicate: 1. control, no salinity change (C), 2. salinity change from fresh- to seawater without the addition of osmolytes (S), and 3. salinity change from fresh- to seawater with the addition of an osmolyte cocktail (O). All the reactors were started in freshwater medium with mature biofilm carriers (AnoxK™ Chip P, Krüger Kaldnes AS, Norway). After 2 days in freshwater medium (days 1–2), the freshwater medium was replaced with synthetic seawater medium in treatments S and O, thus providing a hyperosmotic shock. In treatment O, the seawater medium contained an osmolyte cocktail. This osmolyte cocktail provided 1 mM each of trehalose, sucrose, glycine betaine (betaine hydrochloride), proline, carnitine, and ectoine. The concentration of 1 mM was chosen based on previous studies on osmoregulation. Although the external osmolyte concentration in previous studies on different microorganisms ranges from 0.1 to 150 mM, most studies used concentrations of the order of 1 mM (Vyrides and Stuckey, 2017). Moreover, higher concentrations (order of 100 mM or above) may also inhibit the uptake of osmolytes by microorganisms (Vyrides and Stuckey, 2017). Treatments S and O were operated for 3 days (days 3–5) in the saline media. Treatment C was operated in freshwater throughout the study.

The biofilm carriers were obtained from a nitrifying MBBR in a commercial freshwater recirculating aquaculture system (RAS) for Atlantic salmon smolt. The MBBR had been operated at 12 °C, pH 7.4, and 1‰ salinity at the time of collection, and had never been exposed to salinities higher than 5‰. Due to restrictions caused by the COVID-19 pandemic, these carriers had to be stored for 2 months at 4 °C. To revive the carriers before the experiment, they were transferred (∼25% fill by volume) into a large aerated reactor (4.5 L water volume). After 5 days, the zero-order ammonia oxidation rate reached a plateau, and these carriers were transferred to the experimental beakers. These beakers were maintained at room temperature (17 °C–18 °C). As there was no automatic pH control, the pH varied between 8.3 and 8.8 during the study, which is slightly above the pH optimum of 7.0–8.5 for nitrification (Chen et al., 2020). To minimize evaporation, the air was humidified by bubbling through water. The dissolved oxygen saturation in the reactors was 80%–90% during the study. Each reactor was filled with 15 biofilm carriers (∼20% fill by volume). The synthetic medium was modified from (Bassin et al., 2011) and designed to have an ammonia concentration of 100 mgN L -1. The medium comprised of NH 4 Cl (7.14 mM), MgSO 4 ·7H 2 O (0.72 mM), KCl (0.94 mM), K 2 HPO 4 (0.84 mM), and KH 2 PO 4 (0.42 mM), and a trace element solution (1 mL L -1) as described in (Vishniac and Santer, 1957). A stoichiometric quantity of alkalinity required for nitrification was added to the synthetic medium as NaHCO 3 (14.28 mM). The seawater medium had the same composition as the freshwater medium, with 35 g L -1 Instant Ocean® sea salt in addition to the other chemicals (∼35‰ salinity). To maintain similar alkalinity and pH in the freshwater medium as in the seawater medium, extra NaHCO 3 (2.40 mM) was added to the freshwater medium.

Sampling and analysis

Nitrification capacity tests were conducted on days 2, 3, and 5 to calculate the maximum specific ammonia oxidation rate (AOR) i.e., the zero-order rate (independent of the substrate concentration). During each 24 h test, 4-7 water samples were collected in 2 mL Eppendorf tubes. These samples were collected immediately after changing the medium, with a time interval of 1–2 h for all the samples except the last one. The last sample was collected the next day in the morning before changing the medium. The samples were either filtered through a 0.45 µm syringe filter (Millex-HV PVDF, Sigma Aldrich, Netherlands) or centrifuged to extract the supernatant (13,000 rpm, 5 min, 4 °C). The samples from day 2 were analyzed immediately using ammonia test kits (LCK 303, Hach Lange, Germany) and thereafter frozen at −20 °C. All the samples from days 3 and 5 were frozen. To measure the nitrite and nitrate concentration, the samples were later thawed and analyzed using Thermo Scientific™ Gallery TM Discrete Analyzer (Thermo Fisher Scientific TM, Waltham, USA). Seawater samples with known concentration were used for calibration. From these, correction factors were calculated and applied to the seawater samples to adjust for the salinity interference. Temperature, pH, and dissolved oxygen were measured using handheld sensors (AppliSens®, Netherland) with a controller (Applikon®, Netherlands). Conductivity was measured using a sensor with a multiparameter analyzer (Consort C3010, Belgium).

Data analysis and statistics

The ammonia oxidation rate on day 2 was calculated from the regression line for the ammonia concentration vs. time (Supplementary Figure A.1). The rate was normalized to the total surface area of biofilm carriers to calculate the specific rate of ammonia oxidation (AOR). The specific production rates of nitrite (NO2_PR) and nitrate (NO3_PR) were calculated similarly from the regression lines of nitrite or nitrate vs. time (Supplementary Figures A.2, A.3). Analysis of covariance (ANCOVA) to test the hypothesis of differences between treatments replicates and between treatments (Fox and Weisberg, 2018; Navada et al., 2019). As there was no significant difference between treatment replicates in most of the tests (p < 0.05), the combined data from all three replicates was used for the regression and statistical analysis on each day. The data analyses and visualization were performed in R (version 4.0) using packages car, reshape, and ggplot2.

In the freshwater phase (day 2), there was no significant difference in any of the nitrification performance indicators (AOR, NO2_PR and NO3_PR) between treatments (Figure 1). The average ammonia oxidation rate was 0.76 ± 0.05 g m -2 d -1, and considerable nitrite accumulation was observed (NO2_PR = 0.24 ± 0.03 g m -2 d -1). This is likely because the refrigerated carriers did not get sufficient time to revive and the nitrite oxidizers could not catch up with the rapid increase in ammonia oxidation rate. Future studies should measure both ammonia and nitrite oxidation rates during the revival period, as nitrite oxidizers can take longer to re-establish. The reactors were well-aerated with dissolved oxygen of 80%–90% and the total nitrogen concentration was stable over the duration of the capacity test (Supplementary Figure A.4). This indicates that the ammonia-N was mainly converted to nitrite-N and nitrate-N, and the contribution of other nitrogen removal processes (such as denitrification) was negligible in this study.

On day 3 (the first day after seawater transfer), the seawater treatments (S and O) had a ∼50 and 94% reduction in the NO2_PR and NO3_PR, respectively. There was no significant difference between the seawater treatments with or without osmolytes for NO2_PR (p = 0.74) and NO3_PR (p = 0.20). This indicates that the exogenous addition of the osmolyte cocktail did not significantly improve the performance in nitrifying biofilms immediately after the salinity increase. There were some problems with ammonia measurements on day 3 and day 5. However, as presented above, nitrification was the major nitrogen conversion process in the reactors, and as such, the nitrite and nitrate production rates can be considered as suitable indicators of nitrification activity.

Surprisingly, 2 days after the salinity increase (day 5), nitrification was completely inhibited in the osmolyte treatment, indicated by the negligible nitrite and nitrate production rates (Figure 1). The NO2_PR in the O treatment was −0.03 g m -2 d -1, as the residual nitrite concentration was consumed within the first 8 h (from ∼1 to <0.5 mgN L -1). This was significantly different (p < 0.001) from the positive NO2_PR (0.18 g m -2 d -1) in the S treatment. The nitrate concentration in both the seawater treatments was negligible during the first 8 h after exchange of medium, indicating a complete inhibition of nitrite oxidation. It is remarkable that contrary to our hypothesis, not only did the addition of osmolytes not improve salinity acclimation, it actually resulted in a lower nitrification activity compared to the treatment without osmolytes. This was likely due to the complex effects of osmolytes on the microbial community in the biofilms, as described further below.

We also observed a higher turbidity in the osmolyte treatment than in the other treatments on day 5. On measuring, the medium in the osmolyte treatment had an optical density (OD600) that was an order of magnitude higher than that of the other two treatments (0.050 compared to 0.007). This suggests a higher growth of free-living/pelagic bacteria in O compared to the other treatments. This was confirmed by light microscopy that showed a significantly higher abundance of bacterial cells (putative heterotrophs) in O. The increase in OD may also be attributed to the increased growth of heterotrophic microorganisms from the biofilm that were suddenly exposed to a carbon-rich medium. As the osmolytes are easily degradable organic compounds, they could have been consumed as substrates by the heterotrophs. The ratio of heterotrophs to nitrifiers can be as high as 4:1 even in nitrifying biofilms grown without an organic carbon source (Navada et al., 2019; Navada et al., 2020b). Thus, the osmolytes could have been taken up by both the planktonic and biofilm-attached heterotrophs. Although (Vyrides and Stuckey, 2017) argue that biodegradation of compatible solutes is less pronounced at higher salinities, we observed significant growth of planktonic heterotrophs in our study that can only be attributed to the uptake of osmolytes as substrate. As a majority of osmolyte studies have been on model heterotrophic bacteria like E. Coli (Rojas et al., 2014; Wood, 2015), the heterotrophs could also have acted as superior competitors relative to the nitrifiers for the uptake of osmolytes for osmoregulation. Both scenarios would have promoted the growth and activity of heterotrophs and thereby suppressed nitrification through microbial competition for resources, such as oxygen (Rittma et al., 2001). Thus, the osmoregulatory impact of the osmolytes on nitrification activity was confounded with the effect of the increased competition between the heterotrophs and nitrifiers.

As an osmolyte cocktail was used, we cannot determine which of the osmolytes contributed the most to the suppression in nitrification activity. The response of the microbial community would have been influenced by the type and concentration of osmolyte. It is possible that the growth of the heterotrophs could have been avoided by reducing the concentration of the highly degradable osmolytes (such as trehalose and sucrose) or by providing only the amine-based osmolytes. Future studies should investigate if there exist unique osmoprotectants that can selectively aid osmoregulation in nitrifiers, while not promoting the growth of heterotrophs. Further, some studies suggest that glycine betaine may be inhibited by other osmolytes (Feeney et al., 2014; Mendum and Smith, 2002). Thus, future studies should investigate different osmolytes separately under different ranges of concentrations, possibly at different levels of nitrogen loading or salinity stress.

Investigating the genomes of nitrifying species may provide an insight into the potential pathways for salinity acclimation in these microorganisms. A recent study investigated transporter genes in the metagenome-assembled genomes (MAG) of nitrifying bacteria subjected to salinity stress (Wu et al., 2024). The authors found nitrifying MAGs with genes encoding proline/betaine/ectoine transport (ProP), ectoine transport (EhuA) and glycine betaine transport (OpuA). Glutamate transporter genes have been reported in Nitrosomonas (Wu et al., 2024) and Nitrosopumilus (Kamanda Ngugi et al., 2015), suggesting that it could be a feasible osmoprotectant. Glutamate was not included in this study, and should be investigated. In future studies, metagenomic analysis could be performed first on the biofilm to identify the nitrifying microorganisms and if they encode any transporter genes for osmolytes. Based on this information, the respective osmolytes may be added during salinity stress to study if they are taken up during salinity acclimation. Metatranscriptomic and metaproteomic studies may help identify the genes and metabolites involved during salinity changes, thus giving a better understanding of the salinity acclimation mechanisms in nitrifying microorganisms.

Finally, the effect of osmotic stress on microorganisms growing in a biofilm can be more complex than for single free-living cells. Biofilms can be highly resistant to many different stresses, and biofilm formation is a protection strategy against stresses (Rode et al., 2020). Extracellular polymeric substances (EPS) in the biofilm matrix can protect against several types of stresses. Studies show that the EPS in the biofilm matrix near a bacterial cell may act as osmoprotectants, thus protecting against salinity stress (Seminara et al., 2012; Yan et al., 2017). Further, cells within a biofilm may respond to different stressors individually or collectively (Rode et al., 2020), leading to a higher versatility in response mechanisms. The collective response of microorganisms in biofilms to osmotic pressure changes needs to be better understood (Yan et al., 2017). Future studies should investigate the role of osmotic stress on the communication and response of the bacteria, along with the effect on matrix structure and function.

This study tested the exogenous addition of a cocktail of osmolytes as a potential sustainable strategy for salinity acclimation in nitrifying biofilms. The addition of an osmolyte cocktail did not improve salinity acclimation in nitrifying biofilms transferred from freshwater to seawater. On the contrary, after 48 h of exposure to seawater, the presence of osmoprotectants significantly decreased the nitrification activity. This was likely due to the uptake of osmolytes by the heterotrophs (as a substrate or for osmoregulation), which consequently led to a reduction of nitrification activity due to the competition between the heterotrophs and nitrifiers for oxygen and/or osmolytes. Further research is required on the metagenomic and metaproteomic level to investigate the physiological response of biofilm microorganisms to a salinity increase. Future studies should investigate the impact of individual osmoprotectants at different concentrations to identify osmolytes that are preferentially taken up by nitrifying microorganisms for salinity acclimation. The role of the biofilm matrix on salinity acclimation in nitrifying biofilms also needs further research.

In conclusion, the addition of an osmoprotectant cocktail did not succeed in improving the salinity tolerance of nitrifying biofilms. Nonetheless, our study highlights the complex contradicting effects of osmoprotectants on multispecies biofilms for water treatment, and opens the door for future studies on salinity acclimation in biofilms using osmoprotectants.