Decoupling a novel Trichormus variabilis-Synechocystis sp. interaction to boost phycoremediation

To conserve freshwater resources, domestic and industrial wastewater is recycled. Algal systems have emerged as an efficient, low-cost option for treatment (phycoremediation) of nutrient-rich wastewater and environmental protection. However, industrial wastewater may contain growth inhibitory compounds precluding algal use in phycoremediation. Therefore, extremophyte strains, which thrive in hostile environments, are sought-after. Here, we isolated such an alga - a strain of Synechocystis sp. we found to be capable of switching from commensal exploitation of the nitrogen-fixing Trichormus variabilis, for survival in nitrogen-deficient environments, to free-living growth in nitrate abundance. In nitrogen depletion, the cells are tethered to polysaccharide capsules of T. variabilis using nanotubular structures, presumably for nitrate acquisition. The composite culture failed to establish in industrial/domestic waste effluent. However, gradual exposure to increasing wastewater strength over time untethered Synechocystis cells and killed off T. variabilis. This switched the culture to a stress-acclimated monoculture of Synechocystis sp., which rapidly grew and flourished in wastewater, with ammonium and phosphate removal efficiencies of 99.4% and 97.5%, respectively. Therefore, this strain of Synechocystis sp. shows great promise for use in phycoremediation, with potential to rapidly generate biomass that can find use as a green feedstock for valuable bio-products in industrial applications.


Results
Acclimation for growth in wastewater. Initial attempts to grow algal cells in the mixed domestic + industrial wastewater were futile. Algal cells failed to multiply and colonise the wastewater (Fig. 1). There was still no growth even after adjusting the wastewater pH from the pH 8-9 range down to the BG-11 growth medium pH 7.1. T. variabilis grows optimally in BG-11 medium. It was not clear whether failure to support algal growth was a result of lack of some essential nutrient(s) or the presence of growth inhibitory substances. We then made a 50% dilution of the wastewater with BG-11 growth medium as a way to supplement essential nutrients, which could be absent from the wastewater. Again there was no growth in this diluted wastewater (Fig. 1), suggesting that growth inhibition from components within the wastewater was the likely explanation of these results. Thus, we concluded that the wastewater contained growth inhibitory compounds affecting algal cell multiplication.
In order to survive in adverse environmental conditions, many prokaryotes and eukaryotes require an acclimation period during which the organism is first exposed to intermediate levels of the stress factor(s) to gradually change gene expression for growth under the extreme conditions 49,50 . We reasoned that exposure of T. variabilis cells to increasing concentrations of the wastewater could enable them to gradually acclimate and survive the "stress factors" present in the wastewater. Thus, BG-11 medium supplemented with 1% wastewater was inoculated www.nature.com/scientificreports www.nature.com/scientificreports/ with algal cells. The cells multiplied successfully and 7 days later, 1% wastewater culture was used to inoculate BG-11 medium spiked with 3% wastewater. Again there was successful growth of the cells in 3% wastewater. The sequential inoculation of progressively higher wastewater strength was repeated at 3 percentage point increases until we attained a stably growing algal culture in 50% wastewater over a 5 months period. Interestingly, attempts to increase wastewater strength by ≥5% at any point resulted in failure to grow, a strong indication that acclimation was vital for growth in increasing wastewater strength. Figure 1 shows acclimated algal cultures growing in wastewater.
But we noticed that algal cultures that had acclimated for growth in wastewater had a different appearance to original T. variabilis cultures growing in BG-11 0 or BG-11 media. BG-11 0 and BG-11 differ in that the latter contains the full complement of nutrients required for algal growth, while the former does not have nitrates. While control BG-11 cultures had thalloid or spherical clumps growth forms, acclimated cultures were homogenous cell suspensions (Fig. 1). When aeration is switched off from BG-11 0 and BG-11 cultures, the algal sheets immediately settle to the bottom of the flask, while acclimated cells remain in suspension for several hours before finally sedimenting to the bottom of the flask. When the acclimated cells were inoculated back into BG-11 medium, they maintained the same culture properties, suggesting that acclimation had triggered long-lasting changes.
Synechocystis sp. constitutes the acclimated cultures. To understand the basis for the change in appearance and physical properties of the acclimated cultures, we used microscopy to visualise the cells and make comparisons with the original T. variabilis cultures. Low power light microscopy of BG-11-grown cultures not exposed to wastewater revealed T. variabilis filaments with the typical vegetative cells, heterocysts and akinetes (Fig. 2a). In contrast, only single cells with no filaments were present in samples from acclimated cultures (Fig. 2d). This explains why the acclimated cultures were homogenous cell suspensions as opposed to the thalloid sheets in the non-acclimated cultures. The average vegetative cell diameter in filaments was 4.18 ± 0.54 μm (n = 10), while the diameter of acclimated single cells was 1.43 ± 0.53 μm (n = 10). Transmission electron microscopy (TEM) revealed detailed ultrastructural differences between vegetative cells of filaments from BG-11 cultures (Fig. 2b,c) and single cells of the acclimated cultures (Fig. 2e,f). Cells from acclimated cultures had pronounced folds of thylakoid membranes and many darkly-stained bodies. It is these single cells in the acclimated cultures that have resilience to as yet unidentified growth inhibitory compound(s) in wastewater, being able to rapidly multiply and attain very high biomass in a short space of time (~2.7 ± 0.1 mg.mL −1 .day −1 ) through cell division, (Fig. 2f), instead of germination known to occur in akinete cells.
Transition of the filamentous T. variabilis culture to a rapidly growing, stress-resilient (SR) unicellular culture after acclimation raised questions. Was the original algal culture obtained from CCAP not axenic? Did it contain 2 distinct algal species -one filamentous and one unicellular? We decided to work on the hypothesis that, in addition to having T. variabilis filaments, a second unicellular alga might have constituted a very minor component of the original culture. This can easily be resolved via sequencing of genes used in bacterial taxonomy and identification.
Therefore, we isolated algal DNA and sequenced the 16S ribosomal RNA (rRNA) from filamentous non-acclimated samples and from acclimated unicellular samples. The sequence obtained from the filamentous culture was used as a query in BLAST searches against the entire NCBI nucleotide database and the top 5 hits, all with sequence identities above 99%, are displayed in Table 1. The top hit is T. variabilis and the others are either T. variabilis or Anabaena sp. This confirms that the filamentous alga is indeed T. variabilis. However, the sequence obtained from the unicellular algal cells had high identity with several Synechocystis sp. The top 5 candidates, with sequence identity at 100%, are various strains of Synechocystis sp. (Table 2). More importantly, aligning the T. variabilis and Synechocystis sp. 16S rRNA sequences led to a low sequence identity of 88.5%, confirming that the multicellular filaments and the single cell cultures are from two distinct algae. Taken together, these results show that the acclimated cell culture is a strain of Synechocystis sp., whose precise identity will require future whole genome RNA sequencing.

Synechocystis sp. procures nitrogen via commensal interaction with T. variabilis.
Our initial microscopic examination of the original cultures obtained from CCAP (Edinburgh, UK) and subsequent subcultures derived from these had revealed the presence of only filamentous T. variabilis cells with no unicellular Synechocystis sp. (Fig. 2). Thus, we reasoned that Synechocystis sp. must have been present as a very minor component, which acclimates and starts to increase in abundance when T. variabilis is killed off by toxic compounds in the wastewater. This could explain how a stress-resilient pure culture of Synechocystis sp. arose. Finding the elusive cells of Synechocystis sp. in the pre-acclimation T. variabilis culture would support this line of reasoning. Therefore, we prepared numerous additional sections of algal samples taken from N-deficient medium-grown non-acclimated culture (dominated by T. variabilis filaments) for TEM analysis. After examining several dozens of sections, we finally identified rare occurrences of cells of Synechocystis sp. within the BG-11 0 cultures. They were invariably found in the immediate vicinity of T. variabilis vegetative cells with a very intriguing morphology. The cells had developed nanotubular outgrowths from the cell periphery, which tethered the cells to the polysaccharide capsule of neighbouring T. variabilis filaments (Fig. 3). This was very clear in instances where both the Synechocystis sp. and T. variabilis cells were in the same focal plane of the section.
Trichormus variabilis can switch between 2 modes of nitrate acquisition -uptake from the growth medium or fixing atmospheric nitrogen when growing in a medium bereft of nitrate. Acquisition of nitrogen is one likely reason why cells of Synechocystis sp. may attach to T. variabilis filaments in a commensal relationship, which does not harm the host. Therefore, we conducted experiments to investigate if Synechocystis sp. can grow in the absence of nitrate in the medium. Synechocystis sp. grew well in BG-11 medium (Fig. 4a), which has nitrates, but attempts to get growth in the nitrate-deficient BG-11 0 medium were unsuccessful (Fig. 4a). When biomass grown in the optimal BG-11 medium (with nitrate) was transferred to the nitrogen-deficient BG-11 0 medium, www.nature.com/scientificreports www.nature.com/scientificreports/ Synechocystis sp. cultures died after several days (Fig. 4b). T. variabilis cultures similarly grown in BG-11 and then transferred to BG-11 0 not only survived, but continued to grow under the nitrogen deficiency conditions (Fig. 4c). Taken together, this demonstrates that Synechocystis sp. does not have the ability to fix atmospheric nitrogen and so has to assimilate nitrate from an external source. We speculate that the nanotubular physical connections seen between cells of Synechocystis sp. and T. variabilis filaments may effect material exchange (particularly nitrate) between the two organisms. In support of such a hypothesis are the observations that (i) in pure culture, Synechocystis sp. fails to grow when inoculated into N-deficient BG-11 0 medium and (ii) Synechocystis sp. originally grown to high density in the N-containing BG-11 medium undergoes chlorosis and then dies if transferred to N-deficient BG-11 0 medium (Fig. 4). Thus, Synechocystis sp. survives in N-deficient BG-11 0 medium only when in association with T. variabilis filaments. However, its growth under such conditions is infinitesimally low, presumably as to be almost imperceptible. www.nature.com/scientificreports www.nature.com/scientificreports/ potential for application in phycoremediation. We investigated the effectiveness of the stress-resilient Synechocystis sp. strain in nutrient removal from wastewater. Mixed domestic and industrial wastewater was inoculated with 1 g/500 mL (FW/V ratio) of culture. The cells grew very rapidly, giving dark green algal cultures. A week after inoculation, samples were withdrawn for analysis of ammonium, phosphates, and chemical oxygen demand. Comparison of the background ammonium and phosphate content of untreated effluent with post-treatment levels revealed 99.4 and 97.5% removal of these nutrients by acclimated cultures, respectively ( Table 3). The effluent's oxygen demand (COD) was also dramatically decreased by more than 40% (Table 3). Reduction of COD together with ammonium and phosphate are key targets for effluent treatment. Taken together, the increased growth of acclimated Synechocystis sp. (Fig. 4a) and its efficient nutrient removal (Table 3) demonstrate how this alga could be a very effective tool for effluent treatment.

Discussion
An impediment to use of algal systems for wastewater treatment is the presence of phytotoxic or growth-inhibiting compounds within certain streams of industrial and/or domestic waste. We found that our starting algal material was a composite culture of T. variabilis and a strain of Synechocystis sp., with the latter being a minor component. Neither of these algae survived when directly inoculated into a mixed domestic and industrial effluent. Our objective was to acclimate the culture for growth in the wastewater. Gradual exposure of the culture to minute quantities of the wastewater and increasing the quantities over time appeared to recapitulate the process of acclimation seen in nature. This phenomenon is wide-spread, having been reported in many organisms, including plants 49 cyanobacteria 50 and insects 51,52 . In all of these instances, the acclimation process is driven by changes in gene expression enabling the organism to survive under what previously was a threat to the organism.
Microscopy revealed that our starting composite algal culture switched from T. variabilis dominance to a culture exclusively consisting of Synechocystis sp. It is clear that T. variabilis was eliminated from the culture by toxic compounds within the wastewater. However, the nature of these phytotoxic and growth-inhibitory compound(s) in the mixed effluent remains unknown. From examination of the cultures, we noticed that T. variabilis cells were purged early during acclimation, at ≤3% wastewater strength. However, Synechocystis sp. remaining as a monoculture still continued to require step-wise small increments in wastewater strength, with large increases ≥5% not being tolerated. This indicates that Synechocystis sp. required gradual acclimation to grow in the wastewater. At this point, the biochemical basis for Synechocystis sp. acclimation is unclear and the subject of ongoing research. We found this organism to already have a fascinating innovation for survival in natural environments with limited nitrogen. We demonstrated that Synechocystis sp. cannot fix nitrogen and, in axenic culture under laboratory conditions, fails to grow or survive in the absence of an external nitrogen source. To solve this serious problem, the organism forged a strategic association with T. variabilis in nature, as we found in the CCAP culture obtained from the wild.
The vegetative cells of filamentous T. variabilis can differentiate into 2 cell types (akinetes and heterocysts) under specific environmental conditions 53 . Akinetes are survival cells usually appearing when there is stress, such as nutrient limitation, low light, or low temperature 40,54 . Heterocysts differentiate in response to depletion of external sources of nitrogen. Differentiation of heterocysts is accompanied by changes in cell morphology and expression of genes for the oxygen-sensitive nitrogenase enzymes responsible for nitrogen fixation. Only a few cells along each filament differentiate into heterocysts, which exchange nitrate for sugar with neighbouring photosynthetic vegetative cells along the chain. In the composite culture grown in nitrogen-depleted BG-11 0 medium, we found cells of Synechocystis sp. nestling to T. variabilis vegetative cells. Since Synechocystis cells die when incubated in BG-11 0 medium, their perpetual presence in the composite culture means they somehow derive their nitrate requirements from the nitrogen-fixing T. variabilis. Development of nanotubular structures on Synechocystis cells, hooking them to the polysaccharide capsule of T. variabilis filaments is a possible route for nitrate acquisition in what appears to us a commensal relationship. We speculate that the very low cell number of www.nature.com/scientificreports www.nature.com/scientificreports/ Synechocystis sp. in the composite culture ensures that nutrient withdrawals are not too high to prevent debilitating T. variabilis growth.
The nanotubular structures were not present in Synechocystis cells growing in the nitrogen-rich BG-11 medium or wastewater. Therefore, their development could be cued by the environmental nitrogen limitation. Morphological change entailing development of extracellular appendages hitching together neighbouring cells is not unprecedented in the Synechocystis genus. For example, Synechocystis sp. PCC 6803 was reported to develop such appendages in response to limited oxygen gas 55 . In fact this phenomenon was observed in other algae, such as the appendages connecting co-cultured cells of Pelotomaculum thermopropionicum and Methanothermobacter thermoautotrophicus, which are thought to effect cross-species material transfer 55,56 .
Changes in cell morphology in response to environmental cues are also prevalent in microorganisms beyond algae. For example, the wheat pathogen Zymoseptoria tritici can grow as filamentous hyphae or a yeast-like culture of unicellular micropycnidiospores 57 as driven by external light cues. The micropycnidiospores are not dormant spores, but rather vegetative cells, which divide and grow in a yeast-like fashion. Darkness stimulates filamentous growth and light stimulates yeast-like growth 58 . While the change in morphotype of this fungus is cued by light, stress has previously been reported to serve as a stimulus triggering a change in bacterial morphology. For example, the loosely packed coccobacillus cells of the heterotrophic bacterium Acidocella sp. (strain GS19h) can switch to chains of coccidal lenticular cells or filaments when exposed to heavy metal stress 59 . In T. variabilis nitrogen depletion is a powerful cue for vegetative cell differentiation into heterocysts 53 .
The strain of Synechocystis sp. isolated here could be employed for commercial development of an algal wastewater treatment system. Rapid growth of acclimated Synechocystis sp. in wastewater led to efficient reduction of nutrients, such as ammonium and phosphate. The biomass can be used for fractionation of products for use in industrial applications. However, many aspects of algal growth and large-scale production require optimisation before developing this strain for industrial exploitation. Therefore, our study provides an avenue by which algae could be optimized for both wastewater treatment and as a resource for industrial feedstocks. www.nature.com/scientificreports www.nature.com/scientificreports/

Materials and Methods
Cultivation of algae. The original culture of Trichormus variabilis CCAP 1403/4B was obtained from the CCAP collection (Edinburgh, UK) and cultivated in BG-11 (Blue-Green) culture medium according to Stainer et al. 60  transmission electron microscopy. For electron microscopy, cells were cryo-fixed by high-pressure freezing with a high pressure freezer EM ICE (Leica Microsystems Inc, Buffalo Grove, US). Super quick-freeze substitution was performed using 1% OsO 4 , 0.5% uranyl acetate and 5% H 2 O in acetone followed by embedding into epoxy resin 62 . Ultrathin sections (50 nm) were prepared using ultramicrotome UC6 (Leica Microsystems Inc, Buffalo Grove, US) contrasted for 10 min each in 1% uranyl acetate in ethanol followed by Reynolds lead citrate. Sections were viewed at 100 kV with a Hitachi H-7600 transmission electron microscope (Hitachi, Tokyo, Japan).    Table 3. Wastewater parameters before and after treatment with acclimated Synechocystis sp.

Wastewater treatment and nutrient analysis.
www.nature.com/scientificreports www.nature.com/scientificreports/ Samples of this sludge liquor return were obtained and all endogenous organisms killed by autoclaving prior to use in experiments. The wastewater was sterilised by autoclaving prior to inoculation with T. variabilis. We acclimated the cultures for growth in wastewater via consecutive passaging through a serial dilution of increasing wastewater strength. Experiments were conducted using glass culture vessels incubated at 25 °C/65.5 μE.m −2 .s −1 and bubbling compressed air. Inoculums of 100 mg filamentous culture was added to 70 mL BG-11 containing -1% (v/v) wastewater and allowed to grow for a week. An aliquot from this culture was used to inoculate a 3% wastewater-containing BG-11 medium. This process was repeated successively, with 3% wastewater increments each week until growth was stabilised at 50% wastewater. The acclimated culture was then used for subsequent experiments.
To evaluate the efficacy of algal cultures in wastewater treatment, triplicates of 500 mL wastewater were inoculated with 1 g of the acclimated algal culture. A week later, samples were withdrawn for determination of ammonium and phosphate content, and the chemical oxygen demand (COD) were analysed using LCK test kits according to the manufacturer's instructions (Hach Lange, Manchester, UK).
Sequencing 16S rRNA gene. The original algal culture obtained from CCAP was grown in BG-11 0 medium and the biomass, which was dominated by T. variabilis filaments, harvested at the exponential growth stage for DNA extraction. An algal culture acclimated for growth in wastewater (as described above) was grown in BG-11 medium and harvested at the exponential growth phase for DNA extraction. This culture consisted entire of single cells with no filaments. DNA extraction and PCR analysis were performed as described previously 63 using the primers Cya359F "GGGGAATYTTCCGCAATGGG" and "ACGGGCGGTGTGTAC", where Y is any pyrimidine. The DNA samples were submitted to Sequetech (Mountain View, CA, USA) to perform the sequencing. To identify the species, the NCBI BLASTN 2.8.0+ program 64 was used to search the entire NCBI database (https://blast.ncbi.nlm.nih.gov/Blast.cgi) using the nucleotide sequences as the query. For comparison of the sequence obtained from the filamentous algal culture to the sequence from the unicellular algal culture, we used the William Pearson's LALIGN program, an online sequence alignment tool found on the EXPASY Bioinformatics Resource Portal (https://embnet.vital-it.ch/software/LALIGN_form.html), which utilizes an algorithm developed by Huang and Miller 65 .