Cell culture

Lab-grown cultures of Microcystis aeruginosa, a species commonly found in cyanobacterial blooms, were used to test the effects of H₂O₂ on cyanobacteria fluorescence. M. aeruginosa are round cells ranging 4–5 μm in size. This cyanobacteria species produces multiple toxins including microcystin [2]. Cells were grown in BG-11 media in 15 mL polystyrene conical tubes under a 300–350 LUX grow light (T5 Grow Light Bulbs 2 ft, 24W 6500K).

Flow cytometry

A Cytek Aurora flow cytometer was used to obtain fluorescent data from M. aeruginosa. The system contains five lasers (355 nm UV, 405 nm violet, 488 nm blue, 561 nm yellow-green, and 640 nm red) and 64 detectors. The order, wavelength, and power of the five lasers are as follows: yellow-green (561 nm, 50 mW), violet (405 nm, 100 mW), blue (488 nm, 50 mW), red (638 nm, 80 mW), and UV (355 nm, 20 mW). The laser excitation of the cells is separated by 20 μs. The emission spectrum from each laser was collected with 8–16 detectors, each consisting of 17–25 nm in size (Cytek Biosciences Inc., Fremont, CA). The machine also measures forward scatter (FSC) with the blue laser and measures two side scatter (SSC) channels with the blue and violet lasers. FSC refers to the scatter of light around the cell and is often used as a proxy measurement for cell size. SSC is a measure of cell granularity, determined by the reflection of light off structures inside the cell [32].

Twenty-thousand cells for each sample were counted at a slow flow rate (10–20 μL/min) in a defined region of interest (ROI) in a FSC and SSC-violet cytogram. Counts were recorded within the ROI. A dual FSC and SSC threshold was used to remove the noise in the system. The flow cytometer underwent a quality assurance procedure with Spectro Flo 2000 series (lot 2003) 3 μm QC beads to check for acceptable CV values daily prior to use. The CVs of the beads measured in the majority of the 64 detectors were usually around 2–3%.

H₂O₂ and chemical treatments

Five mL of a growing M. aeruginosa culture was diluted around 3x (to 16 mL) with BG-11 growth media (Sigma, St. Louis, MO). Ten doses of 3% commercial H₂O₂ were added to 1 mL M. aeruginosa samples at the following concentrations: 3000, 1500, 750, 300, 150, 75, 30, 15, 7.5, and 3 μg/mL. Two 1 mL control samples with no chemicals added were also included.

Samples were measured on the flow cytometer at 24 hours, 48 hours, 72 hours, and 8 days to observe how M. aeruginosa cell fluorescence and morphology changed over time in response to different H₂O₂ concentrations. 200 uL of each sample was measured between day 1 and day 3. At day 8, 200 uL of DI water was added to reduce the cell count to less than 1,500 per second, which also decreased the abort rate.

To determine the growth of samples treated with different doses (75–750 μg/mL) of H₂O₂, cells were counted with the flow cytometer at 0, 5, 8, 17 and 24 days after the addition of H₂O₂. The growth curves of control and treated cells in addition to the spectral charts and cytograms are displayed in the results.

Acetylacetone (product number: P7754), diuron (DCMU, product number: D2425), peracetic acid (product number: 269339), and tryptoline (product number: 300764) were obtained from Sigma-Aldrich (St. Louis, MO) added to M. aeruginosa samples to compare their effect on fluorescence to that of H₂O₂. The chemicals were added to 1mL M. aeruginosa samples at the following concentrations: 740, 74, 7.4 μg/mL acetylacetone (diluted to 2.97% in DI water); 0.5, 0.2, 0.1 μg/mL DCMU (diluted to 1% in ethanol); 800, 80, 8, 0.8 peracetic acid (dilute to 3.2% in DI water); and 75, 7.5, 0.75 μg/mL tryptoline (diluted to 1% in DMSO). Samples were measured on the flow cytometer at 48 and 72 hours. The different chemical treatments added to M. aeruginosa are outlined in Table 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. M. aeruginosa treatment chemicals and doses.

Table shows chemicals and doses used to treat M. aeruginosa culture cells and concentration of chemicals.

https://doi.org/10.1371/journal.pwat.0000177.t001

Light stress

To assess how H₂O₂ and light exposure interact to affect M. aeruginosa fluorescence, control and H₂O₂-treated M. aeruginosa cells were placed under blue and green NIGHTSEA lights (Electron Microscopy Sciences, Hatfield, PA) and 1-foot-long UV LED lamps. The cells were exposed to blue (440–460 nm) light at intensities of 14 and 7 mW, green (510–540 nm) light at intensities of 15 and 6 mW, and UV-A (395 nm) light at 1 and 0.5 mW. Different light intensities were achieved by placing the sample tubes at different distances from the light source. Light intensity was measured using a Laser Mate-Q detector (Coherent, Santa Clara, California) with either a visible or UV probe dependent on the wavelength measured. Each light treatment included control M. aeruginosa cells and H₂O₂-treated samples at concentrations of 750, 75, and 7.5 μg/mL. Cell fluorescence was measured on the flow cytometer after an initial 2.5 hours of continuous light exposure, an overnight recovery period under a white grow light at 350 LUX, and a second 2.5-hour light exposure, and then at 24 hours, 48 hours, 72 hours, and 8 days after initial light and H₂O₂ exposure. When the cells were not exposed to green, blue, or UV light, they were incubated under a standard grow light with a 12-hour light-dark cycle.

Data analysis

Data was acquired in the Cytek Aurora flow cytometer using the Spectro Flo operating system software. The data was exported into FCS 3.0 files to enable the utilization of FCS Express (version 7.16, De Novo Software, Pasadena, CA) software that can be used to analyze the data. The samples were initially gated using a region defined by forward scatter area (blue laser) and side scatter area (violet laser). These cells from the scatter gate were then observed in a cytogram defined by the fluorescence channels YG4 (652–669 nm) on the y-axis and V7 (533–550 nm) on the x-axis to study the population of cyanobacteria. Mean light scatter and mean fluorescent data were exported into Microsoft Excel for analysis.

M. aeruginosa fluorescence response to H₂O₂

M. aeruginosa fluorescence responded in a dose- and time-dependent manner to H₂O₂. Green fluorescence from excitation with the violet laser increased over 3 days with exposure to H₂O₂ before declining on day 8 in samples with a dose of 750 μg/mL or higher (Table 2). Red fluorescence derived from yellow-green laser excitation was greater than the control cells in all treated samples until 72 hours of exposure. At 8 days, the red fluorescence of all H₂O₂ doses was lower than the control M. aeruginosa.

The observed increase in red fluorescence 24–48 hours after H₂O₂ incubation indicates a release of more photons into the flow cytometer detectors from M. aeruginosa rather than a utilization of those photons in the production of energy, oxygen, and glucose by the cyanobacteria (Figs 1 and 3). While we did not expect this increase in fluorescence, a few other studies exposing cyanobacteria and microalgae to various treatments, including chemical exposure, high light illumination, extreme heat, and high salinity, have observed an initial increase in fluorescence [6, 14, 22, 26, 27, 33, 34].

Similar to our finding of increased red fluorescence in the 660 nm range, Liu and others showed that strong green light can inhibit energy transfer between PBS pigments in the red alga Porphyridium cruentum, resulting in increased emissions between 550–600 nm from PE and decreased emissions in the 660–700 nm range [26]. The authors attributed these effects to photoprotection by preventing the absorption of excessive light, and thus further damage to the cell [26]. Tamary and others also show greater red fluorescence after exposure to high light in Synechocystis. They suggest that this occurs due to a decoupling of PBSs from reaction centers which prevents light from entering the photosystems, and thus inhibits photosynthetic electron transport [6]. The result is an increase in 663 nm fluorescence from 580 nm light excitation. This increase in red fluorescence may have occurred due to the detachment or uncoupling of PBSs from the PSII reaction center in response to various stressors, including high-intensity light exposure [5, 6, 26]. In this case, it appears that photons are utilized inefficiently, being absorbed and transmitted through the PBS pigments, but not transferred to PSII and thus emitted as fluorescence to our detectors [6]. Luimstra and others also state that 650–665 nm fluorescence is indicative of decoupled PBSs or PC [8]. We observe a similar effect of increased red fluorescence with FCM. The FCM data shows a fluorescence peak from the treated cells at 660 nm after excitation with the UV, violet, blue, and yellow-green lasers (Figs 4, 6, 8 and 10), which may indicate an inhibition at the PSII reaction center where light is processed. Thus, the 660 nm peak observed in the current study suggests that PBSs may have been decoupled from PSII. The yellow-green laser also shows a peak at 697 nm suggestive of inhibition after absorption of light by chlorophyll.

Backward energy transfer from chlorophyll in the PSII reaction center to the PBS core is one proposed mechanism for PBS decoupling [6]. Backward energy transfer occurs when PSII centers are closed due to a blockage in electron transport between primary electron acceptor Q_A and secondary electron acceptor Q_B that prevents the oxidation of Q_A [12, 35, 36]. The chlorophyll transfers the energy back to the PBSs which then may lead to energetic decoupling of the PBSs [6]. These mechanisms would inhibit light utilization by PSII, resulting in increased fluorescence emissions at wavelengths below 660 nm and a sharp 660 nm peak in the violet, blue, and yellow-green laser channels (Figs 6 and 8).

In the current study, we have shown that H₂O₂ exposure results in both an initial increase in red fluorescence as well as decreased red fluorescence at 72 hours and 8 days, an effect potentially linked to photosynthesis inhibition. It is conceivable that PBSs are absorbing photons, but the reaction centers are not fully utilizing them for photosynthesis. M. aeruginosa samples treated with acetylacetone, DCMU, peracetic acid, and tryptoline exhibited similar fluorescence intensity patterns as those treated with H₂O₂, causing a sharp fluorescent peak at 660 nm with UV, violet, blue, and yellow-green laser excitation and increased green fluorescence emissions (Fig 8). Previous research has indicated that acetylacetone, DCMU, and tryptoline cause an inhibition of electron transfer between the primary electron acceptor Q_A and the secondary acceptor Q_B, which blocks PSII [18, 19, 37], while H₂O₂ is shown to act on the D1 protein. These results indicate that chemicals with different modes of action may elicit a similar fluorescence response in M. aeruginosa.

An increase in green fluorescence was observed in M. aeruginosa samples with almost all H₂O₂ doses between 7.5 μg/mL and 3000 μg/mL (Table 2). This increase was not in a specific detector but spread over the whole blue and green spectral ranges detected from UV, violet, and blue laser excitation (Figs 4, 6, 8 and 10). Other studies have found a similar increase in green autofluorescence of perturbed or nonviable cyanobacteria [38–41], but the source of these fluorescence emissions below the 600 nm range remains unclear.

This increase in green fluorescence emissions may be due to upregulation of a green fluorescent molecule in response to H₂O₂ exposure. Several studies have investigated autofluorescence in the green spectral range that relates to metabolism and physiology in mammalian cells [42–44] plant cells [45–47], and bacteria [48]. Green and blue fluorescence emissions from mammalian cells is largely attributed to flavins and NADPH [42]. While NADPH and FAD fluorescence have also been studied in cyanobacteria [49, 50], further research is required to determine whether these compounds are contributing to the increased M. aeruginosa green fluorescence detected from UV, violet, and blue laser excitation recorded in this study.

While the pigments in cyanobacteria absorb yellow-green (561 nm) and red (640 nm) laser light most efficiently, cyanobacteria can also absorb small amounts of lower wavelengths of light from the other three lasers (355 nm, 405 nm, and 488 nm). Due to the short time interval (20 μs) between lasers in the flow cell, pigments excited with one laser may emit fluorescence in another laser path (i.e., emissions from yellow-green laser excitation may be detected in the violet path which is next to the yellow green laser), and this may be the cause the increased red fluorescence intensity and 660 nm peak observed in the UV, violet, and blue laser detectors. Cyanobacteria may be unique in that emission can occur over a longer period of time after the pigments are excited. Generally, the emission from the red laser (80 mW) and yellow green (50 mW) laser are considered very bright as cyanobacteria have pigments that absorb these wavelengths of light. Additionally, cyanobacteria contain photosynthetic pigments and internal structures that result in a different fluorescence and scatter response to bright laser light compared to mammalian cells. It is possible that the biological excitation of the cyanobacteria by one laser may affect the way the cyanobacteria respond to illumination with adjacent lasers. It remains unclear why emissions profiles are similar between UV, violet, and blue laser excitation. Thus, this observation requires further investigation.

If fluorescence emissions from cyanobacteria occur over a second, as has been reported with the bright light illumination of PAM (pulse amplitude modulated) fluorometry and FRRF (fast repetition rate fluorometry), then it is conceivable that this bright light illumination in a flow cytometer may cause pigments to emit light in an adjacent light path to the exciting laser. For example, excitation from the red laser may emit in the UV light path and excitation from the yellow-green laser may emit in the violet and blue light paths as the transit time between light paths is only around 20 μs. However, to examine the possibility of fluorescence spillover, M. aeruginosa cells or 3 μm Spherotech rainbow beads (excited with all lasers) were run with individual lasers covered to prevent excitation with that wavelength. Covering individual lasers did not appear to significantly affect the fluorescence emissions detected in other laser channels, suggesting that the fluorescence effect observed in this study may not be due to laser emission spillover.

Multiple studies have reported that energy transfer between PBS pigments and chlorophyll in the PSII reaction center occur within picoseconds-nanoseconds [12, 51]. However, PAM fluorometry and FRRF use bright light stimulation which results in a very fast emission of fluorescence followed by a slow (second-long) emission of fluorescence [52, 53]. It is also possible that energy transfer between pigments in cyanobacteria is affected due to the physiological state of the organism [54]. Illumination of cyanobacteria with one laser may affect the way the cells respond to adjacent laser excitation. It remains unclear whether the H₂O₂ and light treatments used in this study would cause a delay in fluorescence decay to an extent that fluorescence from one laser is detected in other lasers’ light path detectors or where the H₂O₂ would affect the interaction of adjacent lasers with the cyanobacteria resulting in an increase in green fluorescence.

Synergistic effect of light stress and H₂O₂ on M. aeruginosa fluorescence

In line with previous studies [17, 20, 22, 30, 31], we found a synergistic effect of light and H₂O₂ on cyanobacteria fluorescence. The addition of blue and UV-A light to H₂O₂ treatments had a greater effect on M. aeruginosa fluorescence than light or H₂O₂ alone, especially at higher light intensities.

Cyanobacteria exhibit photoprotective mechanisms under excessive light exposure by closing their reaction centers and decreasing light utilization [55]. The addition of H₂O₂, which damages the D1 protein, inhibits the ability of cells to recover from intense light exposure [17, 20]. For this reason, it appears that the addition of chemicals and light have a synergistic effect on M. aeruginosa, which is greater than the additive effect of both stimuli. This may be the cause of the synergistic decrease in red fluorescence and increase in green fluorescence observed in response to the UV-A light + H₂O₂ treatments (Figs 9 and 10).

Because the fluorescent intensity patterns of the light + H₂O₂ samples in Fig 10 were similar to H₂O₂ alone (i.e., containing 660 nm peaks), the addition of intense light exposure does not seem to change the mode of damage to the photosynthetic apparatus, but rather intensify the effects of H₂O₂ in a non-linear manner.

Our results demonstrate that the effect of light on H₂O₂-treated M. aeruginosa further varies by wavelength. UV-A (395 nm) light had the greatest additive effect on M. aeruginosa exposed to H₂O₂ followed by blue (440–460 nm) light while green light did not affect the cells to the same degree under similar exposure conditions.

UV-A (320–400 nm) light was used in this study as it is more common in the environment compared to UV-B (280–320 nm) and UV-C light (200–280 nm) which are often used in laboratory research studies [56, 57]. While the lower wavelengths UV-B and UV-C are more destructive to mammalian cells and cyanobacteria than UV-A [27, 29], UV-A light is not without its own destructive effects as it has also been shown to cause fluorescence bleaching, photosynthesis, and growth inhibition, D1 protein damage, disruption of electron transfer, and ROS formation [58, 59].

Previous studies have also demonstrated an adverse effect of blue light on cyanobacteria. Luimstra and others reported lower growth rates of Synechocystis under blue light compared to orange and red light [8]. Sinha and others suggest that blue light exposure of cyanobacteria caused phycobiliproteins to decouple from PSII. Bleaching of phycobiliproteins under blue light also occurred more quickly than under green light [27]. We hypothesize that this effect under blue light may be due to an imbalance between PSI, which absorbs blue light most efficiently, and PSII, which absorbs green and orange light most efficiently and inhibits photosynthetic electron transport [8, 20, 60–62].

However, Piel and others demonstrated that orange, red, and green light caused a greater decline in photosynthetic vitality of H₂O₂-treated M. aeruginosa cells than blue light. Because PSII absorbs orange light more efficiently than blue light, exposure to orange light is more likely to cause light stress to PSII. This effect combined with damage to the D1 protein and inhibition of PSII repair caused by H₂O₂ led to lower photosynthetic vitality [20]. By this logic, in the current study we would expect green light to cause greater changes in red fluorescence relative to the control and blue light, but we did not observe this effect. It is possible that inhibition of photosynthetic electron transport by both blue light illumination [due to a PSI:PSII imbalance; 8] and H₂O₂ affected M. aeruginosa fluorescence more greatly. However, further research should examine the combined effects of blue and green light with H₂O₂ on cyanobacteria to better understand this relationship.

Morphological response (cell size) of M. aeruginosa to H₂O₂ and light stress

FSC has been used by a number of investigators to measure the size of particles as there appears to be a rough correlation between the size of the particle and the amount of forward or small angle scatter [32]. However, there are limitations to using these scatter measurements to quantify the size of particles or organisms like cyanobacteria. There are a number of factors that affect scatter measurements making this parameter both interesting and difficult to study in relation to cellular physiology. The presence of strongly absorbing material like cyanobacteria pigments or textured surfaces will decrease forward scatter. Cells that are dead will have lower refractive indices and thus scatter less light and show a decreased size. The size estimate with cyanobacteria is greatly affected by membrane structures, pigments inside the cell, thylakoids attached to interior membranes, highly textured surfaces, and vacuoles that are used for cellular buoyancy. It was observed that the cyanobacteria responded to H₂O₂ by sinking to the bottom of the test tube, demonstrating that H₂O₂ affected their buoyancy. This change in vacuoles will greatly affect light scatter signals in different ways dependent on the degree of change within the cells [63, 64]. These parameters decrease the intensity of the FSC signal, resulting in an apparent decrease in cell size. FSC amplitude will not be a monotonic function of particle size, but will be influenced by a number of factors, and thus should only be considered a relative measure.

SSC involves the internal structure of the organism and can be greatly affected by the refractive index and other factors mentioned above for FSC. One of the best examples of this is the difference in SSC of different white blood cells (i.e., granulocytes vs. monocytes/lymphocytes) which varies based on the internal structures contained within the cell [32]. Many of the factors affecting FSC will also affect SSC signals, resulting in measuring a relative size and not an absolute size of the cyanobacteria. The FSC and SSC signals collected in this study reflect changes that are not totally related to the actual physical size of the organism after exposure to H₂O₂.

In agreement with previous studies [65–67], our results indicate a decrease in relative M. aeruginosa cell size over time of H₂O₂ exposure based on FSC measurements (S1 Table). The addition of high intensity blue and UV-A light to H₂O₂ treatments resulted in a general decrease in FSC (S2 Table). It has been suggested that H₂O₂ induces oxidative stress and apoptosis [66, 67], and has a similar effect in some mammalian systems [68, 69]. It is clear that H₂O₂ caused changes in the cyanobacteria internal structure which may have affected light scatter. Further research is necessary to identify the causes for this decrease in relative size of the organism, but it appears that both membrane alterations and vacuole changes can result in large changes in scatter measurements.

Similar to FSC, SSC also generally decreased with H₂O₂ dose and over time (S1 Table). This effect did not appear to be compounded by the light + H₂O₂ treatments. Other studies have observed a decrease in SSC in response to H₂O₂ exposure, indicating decreased intracellular complexity and granularity [65, 70]. This effect may be due to detachment or degradation of cellular components that contribute to internal complexity, like gas vesicles, [70, 71]. Because H₂O₂ targets membranes, it is possible that the decrease in SSC observed in this study is due to damage to the thylakoid membrane.

While most 750 μg/mL H₂O₂ + light treatments had lower FSC and SSC than the control, the 750 μg/mL H₂O₂ + 7 mW blue light sample exhibited consistently higher FSC values than the control and much lower SSC values compared to other treated samples (S2 Table). This effect may be explained by swelling of cells to a larger volume, which is known to decrease cells’ refractive index and lead to lower SSC measurements [72, 73]. As H₂O₂ interacts with the cyanobacteria, the internal structure of the organism will be affected which then affects scatter measurements [63]. However, it is unclear why the H₂O₂ + 7 mW blue light sample in particular had highly increased FSC and decreased SSC values relative to the control.

Lastly, light scatter and fluorescence are often positively correlated as larger particles tend to emit higher fluorescence [32]. However, Shapiro warns against using a scatter parameter is not absolute but relative. This makes it potentially laden with variations due to the changing internal parts. We observed higher red fluorescence in the first 48 hours and consistently higher green fluorescent intensity in treated samples that had lower FSC and SSC. The green and red fluorescence intensity were divided by the relative forward light scatter to determine how this indirect size measurement may influence fluorescence. This data suggest that the fluorescent changes in response to H₂O₂ were due to physiological or metabolic changes rather than a variation in FSC and SSC measurements.

Water treatment implications

Despite uncertainty regarding the mechanisms behind these changes, our results demonstrate that examining the fluorescent characteristics of M. aeruginosa cells may provide information regarding cell physiology and the effects of water treatment chemicals. Fluorescence changes can be observed in this cell culture model system to rapidly identify patterns of cell perturbance, thus serving as an efficient screening tool to determine which treatment chemicals merit deeper investigation for use in managing a given bloom situation.

Our results demonstrate that a flow cytometer can detect fluorescent and morphological differences between control and H₂O₂-treated M. aeruginosa cells in culture. Thus, this machine may be used to assess the efficacy of water treatment chemicals at various doses on a cell culture model. Cyanobacteria samples from lakes cannot usually be run on flow cytometers due to the potential risk of blocking the flow cell. Samples of cyanobacteria from the environment can be measured if the water is filtered through a 37-um nylon mesh filter. However, filtering the water to eliminate large clumps removes a substantial number of organisms and may produce a misrepresentative sample. The flow cytometer can be used to provide foundational data for further research on the efficacy of water treatment chemicals to control HABs in the environment.

Based on red and green fluorescence relative to the control, the three highest H₂O₂ doses in this experiment (3000, 1500, 750 μg/mL) were not very different; each exhibited lower red fluorescence than the control after 72 hours of exposure and a decline in green fluorescence at 8 days. Furthermore, by day 8 the ratio of red fluorescence in all treated samples was lower than that of the control, a potential indicator of photosynthesis inhibition. Understanding the effects each dose has on M. aeruginosa can aid in determining the lowest possible dose to use for treatment as to avoid adverse effects on non-target species.

The results of the current study suggest a synergistic effect of high-intensity light and H₂O₂ exposure on M. aeruginosa which has been reported in other studies [17, 20, 22, 30, 31]. This finding suggests that implementing H₂O₂ treatment in water systems may be more effective under exposure to sunlight [30, 31]. Our data also demonstrate that certain wavelengths of light (UV-A and blue) may have a greater effect on M. aeruginosa than others (green). Some drinking water treatment plants have already utilized UV light to remove other contaminants [74, 75]. Our results suggest that the use of H₂O₂ with UV light technology may be an effective method for targeting cyanobacteria in water treatment.

Potential viability test

The change in green fluorescence from UV, violet, and to a lesser extent blue laser excitation on the flow cytometer can be used to demonstrate cyanobacteria perturbation by different chemical and light conditions. Preliminary data suggest that this increase in green fluorescence is not reversible. Schulze and others [40] found that live Synechocystis cells exhibited brighter red fluorescence and nonviable cells emitted green fluorescence using imaging software. The authors suggest that red and green cyanobacteria autofluorescence may be viability indicators. In agreement with Cheloni & Slaveykova, the current study demonstrates that these cyanobacteria autofluorescence parameters can also be detected by FCM [38].

Previous FCM studies on cyanobacteria physiology and viability have measured different parameters (like chlorophyll fluorescence rather than PBS fluorescence) and utilized fluorescent staining [76, 77]. However, we found that M. aeruginosa autofluorescence was sensitive to chemical and light exposure and may indicate more sensitive and earlier changes in physiology and viability than fluorescent exclusion dyes like SYTOX green, SYTOX orange, SYTO 9, propidium iodide (PI), or DAPI, which only stain membrane-permeable cells. Permeability dyes are a good estimate for dead cells after the membrane has been altered, but these dyes will not enter cells in the initial phase of an interaction with a toxic chemical. We found that PI or SYTOX orange will stain dead cyanobacteria, causing an increase in fluorescence intensity which can be detected on a spectral system. The other permeability dyes that were tried (SYTOX green, DAPI, HO33324, HO33258, and SYTO 9) stained live cells as well as dead cells and were found unsuitable to detect cell death on a flow cytometer under the exposure conditions used. We believe the initial increase in red and green fluorescence occurs before membrane permeability changes allow viability dyes like PI to penetrate into damaged cells. Therefore, this dye permeability assay is not representative of all the dead cells in a population. Additionally, cyanobacteria emit fluorescence in multiple spectral regions, so using fluorescent dyes may obscure the fluorescent patterns recognized in this study and the biophysical factors associated with them. Thus, our findings suggest that it is better not to add a fluorescent dye to show viability change.

The relationship between cyanobacteria growth and changing fluorescence patterns was investigated by measuring cell counts (Fig 11) and spectral fluorescence changes over time (Fig 12). It was found that the samples treated with H₂O₂ did not proliferate, while the control samples continued to exhibit increased cell numbers (Fig 11). As the control cells proliferate, their fluorescent characteristics did not change. In contrast, cyanobacteria treated with H₂O_2, showed a sequential decrease in red fluorescence and maintain high green fluorescence over time (Fig 12). H₂O₂ treated cyanobacteria cells with decreased red fluorescence and high green fluorescence can be stained with red fluorescing dyes i.e., propidium iodide (PI) or Sytox orange, suggesting they had damaged membranes. These treated cells with low red fluorescence that had permeable membranes were associated with the late stage of cell death. Although the observed early changes consisting of higher red and green fluorescence (Figs 3–6) resulting from H₂O₂ treatment cannot be considered a direct assessment of cyanobacteria viability, because the fluorescence change was a consistent toxicological response to H₂O₂. It is conceivable that the change in fluorescence may be an early indication of cell death. Further experiments will be necessary to conclusively show these higher fluorescence changes are irreversible and related to viability.

A similar process of increased red fluorescence followed by a decline has been observed with PAM fluorometry and FRRF, which measure fluorescence with a pulse of light after cells have been inoculated in the dark. PAM and FRRF have been used to analyze photosynthetic potential and potential cell damage to algae, cyanobacteria, and plants [52, 53, 61, 78, 79]. H₂O₂ and other ROS can cause oxidative damage to cells which can exacerbate photoinhibition likely by targeting PSII repair [80].

A number of researchers have used PAM fluorometry and FRRF to examine fluorescence characteristics of cyanobacteria [14, 20, 22, 25, 52, 60–62, 78, 81, 82]. PAM fluorometry can be used to detect deficiencies in light processing and obtain values like minimal, maximal, and variable fluorescence and approximate other parameters like biomass and photosynthetic vitality [14, 20, 22]). The limitations of this technique, especially in differentiating between PBS and chlorophyll fluorescence, have been described [61, 83, 84]. The purpose of the current study is not to quantify photosynthetic vitality or efficiency but use FCM to detect changes in fluorescence parameters to possibly assess cyanobacteria physiology.

This research may be connected to studies using PAM and FRRF fluorescence equipment. In our study, there was an increase in fluorescence in the red and green ranges when DCMU was added to cyanobacteria (Fig 8), which was similar to the fluorescence observed when H₂O₂ was added to the cells for 24–72 hours. This increase in fluorescence observed on the flow cytometer was similar to an observation in a study from Stirbet [52], in which adding DCMU to dark-acclimated cells in the PAM assay showed an increase in fluorescence after excitation with bright light. This FCM technique may be a useful supplement the type of data collected by PAM fluorometry and FRRF to describe cyanobacteria physiology.