Cyanobacterial Growth on Municipal Wastewater Requires Low Temperatures

Side-streams in wastewater treatment plants can serve as concentrated sources of nutrients (i.e. nitrogen and phosphorus) to support the growth of photosynthetic organisms that ultimately serve as feedstock for production of fuels and chemicals. However, other chemical characteristics of these streams may inhibit growth in unanticipated ways. Here, we evaluated the use of liquid recovered from municipal anaerobic digesters via gravity belt filtration as a nutrient source for growing the cyanobacterium Synechococcus sp. strain PCC 7002. The gravity belt filtrate (GBF) contained high levels of complex dissolved organic matter (DOM), which seemed to negatively influence cells. We investigated the impact of GBF on physiological parameters such as growth rate, membrane integrity, membrane composition, photosystem composition, and oxygen evolution from photosystem II. At 37°C, we observed an inverse correlation between GBF concentration and membrane integrity. Radical production was also detected upon exposure to GBF at 37°C. However, at 27°C the dose dependent relationship between GBF concentration and lack of membrane integrity was abolished. Immediate resuspension of strains in high doses of GBF showed markedly reduced oxygen evolution rates relative to the control. Together, this suggests that one mechanism responsible for GBF toxicity to Synechococcus is the interruption of photosynthetic electron flow and subsequent phenomena. We hypothesize this is likely due to the presence of phenolic compounds within the DOM. IMPORTANCE Cyanobacteria are viewed as promising platforms to produce fuels and/or high-value chemicals as part of so-called “bio-refineries”. Their integration into wastewater treatment systems is particularly interesting because removal of the nitrogen and phosphorus in many wastewater streams is an expensive but necessary part of wastewater treatment. In this study, we evaluated strategies for cultivating Synechococcus strain PCC 7002 on media comprised of two wastewater streams; treated secondary effluent supplemented with the liquid fraction extracted from sludge following anaerobic digestion. This strain is commonly used for metabolic engineering to produce a variety of valuable chemical products and product precursors (e.g. lactate). However, initial attempts to grow PCC 7002 under otherwise standard conditions of light and temperature failed. We thus systematically evaluated alternative cultivation conditions and then used multiple methods to dissect the apparent toxicity of the media under standard cultivation conditions.

3 streams is an expensive but necessary part of wastewater treatment. In this study, we evaluated 51 strategies for cultivating Synechococcus strain PCC 7002 on media comprised of two wastewater 52 streams; treated secondary effluent supplemented with the liquid fraction extracted from sludge 53 following anaerobic digestion. This strain is commonly used for metabolic engineering to produce 54 a variety of valuable chemical products and product precursors (e.g. lactate). However, initial 55 attempts to grow PCC 7002 under otherwise standard conditions of light and temperature failed. 56 We thus systematically evaluated alternative cultivation conditions and then used multiple 57 methods to dissect the apparent toxicity of the media under standard cultivation conditions. 58 59 60

INTRODUCTION 61
The need to develop non-petroleum based platforms for fuel and chemical production is   Figure 1. In order to calculate nutrient 204 stoichiometries, we took the sum of NH 3 -N and NO 3 -N to be the bioavailable N. Nutrient levels 205 in the GBF were markedly more variable than in the secondary effluent. The average molar ratio 206 of bioavailable N to SRP was 35 ± 7 in GBF (12.5% v/v) diluted with secondary effluent, as 207 compared to 32 in Medium A + . 208 We also measured DOM quality in the secondary effluent and GBF using 209 excitation−emission matrix (EEM) fluorescence spectroscopy (37) because we hypothesized that 210 DOM was linked to toxicity during cultivation, as has been shown in prior studies (16,17,38). 211 EEM fluorescence spectroscopy revealed that secondary effluent contained diffuse constituents, 212 including humic [excitation wavelengths (>280 nm) and emission wavelengths (>380 nm)] and 213 fulvic acid-like [excitation wavelengths (<250 nm) and emission wavelengths (>350 nm)] spectra 214 relative to Medium A + (Figure 2). The distinction between the two substances is historically based 215 to the presence of soluble microbial products, which include aromatic amino acids, carbohydrates, 220 or phenols (37, 41). Prior work has shown that the GBF stream of the Nine Springs Treatment 221 Plant contained high molecular weight DOM, which was dissimilar in composition to other 222 internal streams (42) 223 Dose-dependent tolerance to GBF is a function of growth temperature 224 To evaluate the effects of GBF dosage on PCC 7002 physiology, we measured biomass 225 accumulation ( Figure 3) and growth rates ( We wondered whether the decreased growth rates in GBF at 37°C was a result of 239 membrane permeability, given the known effect of humic acids on membrane integrity (16, 17). 12 To track the dynamics of GBF induced membrane permeability, we employed forward scatter flow 241 cytometry using SYTO 59 as a counterstain to identify cells, which were subsequently visualized 242 for membrane permeability using Sytox Green. As Sytox Green is a membrane impermeable dye 243 that fluoresces when binding to nucleic acids (43), fluorescence would indicate compromised outer 244 membrane structures. Two distinct phases were identified upon exposure to GBF, which we 245 interpreted as initial and chronic membrane permeability ( Figure 4). Initial membrane permeability 246 was defined as the Sytox Green positive events for samples analyzed within the first 10 hours of 247 growth, while chronic membrane permeability accounted for the Sytox Green positive events 248 during subsequent time points. As was the case for growth rate, a relationship between both initial 249 and chronic membrane permeability and increasing GBF concentrations was found at 37°C ( Figure  250 4a). While we still detected considerable initial membrane permeability with GBF exposure at 251 27°C, chronic membrane permeability decreased over time, mostly likely to the increase in 252 biomass ( Figure 3). Altogether, this suggested that there was a temperature dependent adaptation 253 that ameliorated the susceptibility of cultures to GBF induced membrane permeability. 254 Exposure to GBF at high temperatures generates radicals and destroys the photosynthetic 255 pigments 256 We directly measured the ROS content and membrane permeability in response to 257 overnight GBF exposure at 37°C using the fluorophores Sytox Green and CellROX Orange (

Exposure to GBF retards oxygen evolution 275
To better delineate the cause of initial toxicity associated with high GBF concentrations, 276 we measured maximal oxygen evolution rates for strains briefly exposed to media with 12. Next, we examined the effect of temperature in a similar experiment. Assays carried out in 289 Medium A + after growth in Medium A + at 27°C showed much higher maximal O 2 evolution rates 290 at all tested light intensities than with cells grown at 37°C (Figure 7b), peaking at a rate 556 ± 102 291 µmol O 2 (mg Chl a) -1 h -1 ). This was expected because elevated O 2 evolution rates in low 292 temperature grown cells have been previously reported and were attributed to a substantial change 293 in photosystem stoichiometry (47). The O 2 evolution rates of cells grown in Medium A + at 27°C 294 and then resuspended in 12.5% GBF stayed relatively constant at all of the tested intensities and 295 were roughly 5-fold lower than in controls, with a maximal rate of 109 ± 22 µmol O 2 (mg Chl a) -296 1 h -1 at an intensity of 75 µmol photons m − 2 s − 1 . We compared the above rates to those from 297 cultures grown in 12.5% GBF at 27°C to test if adaptation to GBF was met with changes in 298 photosynthetic activity. At the examined light intensities, O 2 evolution rates with 27°C GBF-299 adapted cultures were not statistically different from those measured in cultures grown at the same 300 temperature in Medium A + . Finally, we conducted the inverse experiment, using cultures grown 301 in GBF media at 27°C but assayed in Medium A + under saturating light. Interestingly, they 302 displayed the highest evolution rate of any tested condition, at 944 ± 96 µmol O 2 (mg Chl a) -1 h -1 303 ( Figure 7b). This suggested that there is a period of dynamic photosynthetic adaptation to 304 overcome the stress of GBF, and that when the stress is removed, the cells have an enhanced 305 capacity for photosynthetic activity. 306 Additional assays were performed at light saturation (2700 µmol m − 2 s − 1 ) with cultures 307 adapted to 37°C ( Figure 8) to further examine the effects of media conditions on oxygen evolution 308 rates. Again, oxygen evolution rates in GBF (73 ± 8 µmol O 2 (mg Chl a) -1 h -1 ) were much lower in overall ionic strength, we increased the osmolarity of the GBF medium by adding NaCl to match 311 the levels found in Medium A + , and found no statistical difference in oxygen evolution as 312 compared to experiments without added NaCl. Attempts to eliminate the inhibitory effect of GBF 313 on oxygen evolution by gravity filtration through powdered activated carbon were unsuccessful. 314 We also measured oxygen evolution rates in control experiments designed to inhibit 315 photosynthetic electron transfer through 10 µM DCMU addition to Medium A + (48), and to prevent 316 de-novo protein synthesis by pre-treating cultures with 800 µg ml -1 Lincomycin in Medium A + 317 (49), prior to incubation at light saturation for 1 hour. However, evolution rates with these controls 318 were higher than those treated with GBF. Therefore, no definitive conclusion about the molecular 319 mechanism responsible for reduced oxygen evolution rates in the presence of GBF could be made. 320

Acclimation to GBF changes lipid content and composition 321
Based on the results described above, we hypothesized that the temperature dependent 322 adaptation may be related to changes in membrane content and composition. We extracted total 323 fatty acids of cultures grown at 27°C for 72 hours in 12.5% GBF or Medium A + , and analyzed the 324 content after derivatization. We could detect and resolve all major saturated and unsaturated fatty 325 acid species (Table 4). Cultures grown in GBF had greater totals of assayed fatty acid species (27 326 ± 7 mg FAME gDCW -1 ) when compared to cells grown in Medium A + (15 ± 1 mg FAME gDCW -327 1 ) ( Table 4). The most abundant fatty acid in all samples was 16:0 (42-45% of the total fatty acids). 328 C18:2 ∆9,12 fatty acids comprised a significant fraction of Medium A + grown cultures with 22% 329 of the total fatty acid species. However, in GBF-grown cells C18:2 ∆9,12 fatty acids were only 330 15% of the total fatty acid pool, while C18:3 ∆9,12,15 fatty acids were twice as high in GBF grown 331 cells (16%) than in Medium A + grown cells (9%) ( Table 4). These results suggest that cells were 16 altering their membrane homeostasis when grown in GBF, as compared to standard growth in 333 Medium A + . While we did not differentiate the fatty acid content and composition between the 334 outer, plasma, or thylakoid membranes in our study, prior work done with the cyanobacterium 335 Synechocystis PCC 6803 found similar fatty acid composition between thylakoid and plasma 336 membranes (50). 337

DISCUSSION 338
The amount and distribution of arable land and potable water are projected to change over 339 the next several decades due to climate change, while the rise in global population and standard of 340 living are expected to increase demand for these resources (51). Microalgal cultivation integrated 341 with industrial and municipal wastewater treatment circumvents many of the resource concerns 342 raised over biofuel production (2), while simultaneously removing additional nutrients and 343 pollutants present in the wastewater (52). Synechococcus PCC 7002 is an intriguing possible 344 platform for this purpose, since it can be readily engineered to produce high-value chemicals (53). 345 However, under standard environmental conditions we were unable to obtain robust growth of 346 PCC 7002 using a diluted municipal side stream (GBF) as a nutrient source. We hypothesized that 347 this effect may be due to the presence of DOM, which has been demonstrated to cause a decrease 348 in photosynthetic performance in various cyanobacteria strains (54-56). We investigated the 349 effects of light intensity and temperature on the physiology of cultures grown in GBF, in an effort 350 to find conditions conducive to high growth rates and biomass generation, and to better understand 351 the mechanisms of GBF toxicity. 352 We propose that the herbicidal effect of GBF is primarily due to PSII inhibition as shown     hours. The spectra were recorded in dilute cell suspensions and normalized to an OD 730 . The peak at 438 nm is due to carotenoids, the peak at 637 nm is due to phycobilisomes, and the peak at 683 nm is due to chlorophyll a.        hours. The spectra were recorded in dilute cell suspensions and normalized to an OD 730 . The peak 5 at 438 nm is due to carotenoids, the peak at 637 nm is due to phycobilisomes, and the peak at 683 nm is due to chlorophyll a. 1 % CO 2 at 37°C). Cells were pelleted and resuspended in Medium A + , 12.5% GBF, 12.5% GBF + 310 mM NaCl (control for osmolarity), 12.5% GBF prefiltered with activated carbon, Medium 5 A + + 10 µM DCMU (control for inhibition for electron transport), or pre-treated with 800 µg ml -1 Lincomycin for 1 hour (control for the inhibition of protein synthesis.) Maximal oxygen evolution rates were measured with 10 mM HCO 3 as an electron acceptor at 2700 µmol photons m -2 s -1 . The values represent the mean ± SE of biological duplicates.