Morpho-physiological responses of a subtropical strain of Cylindrospermopsis raciborskii ( Cyanobacteria ) to different light intensities

Th e toxigenic cyanobacterium Cylindrospermopsis raciborskii previously restricted to tropical latitudes, has been increasingly reported in temperate lakes in recent decades. Th e causes of its biogeographical expansion are under investigation, but effi cient physiological adaptation to changes in temperature and light regimes are likely to be involved. Th e present study evaluated the morpho-physiological responses of a strain of C. raciborskii from southern Brazil to nine light intensities, from 9 to 250 μmol photons m-2 s-1. Blooms of this cyanobacterium are regularly recorded in the region. Morpho-physiological responses were measured based on growth rate and trichome length. Cylindrospermopsis raciborskii showed slow growth at low light intensities, 9 and 20 μmol photons m-2 s-1, and responded morphologically by increasing the length of trichomes. In turn, the strain displayed constant maximum growth rates at light intensities higher than 50 μmol photons m-2 s-1. Th ese results support the hypothesis that C. raciborskii can survive under low light conditions and continue to produce viable trichomes. Moreover, the strain achieved high growth rates under a relatively wide range of light intensities, a physiological adaptation that can potentially be a competitive advantage in the phytoplankton community.


Introduction
Cylindrospermopsis raciborskii is a toxigenic cyanobacterium initially ascribed as a tropical to subtropical species (Padisák 1997).However, blooms of this species have increased over the past two decades in many lakes and reservoirs around the world, including temperate latitudes, leading researchers to reclassify the species as cosmopolite (Briand et al. 2004).Th e increasing number of reports of C. raciborskii and its expanding geographical range has been correlated to the global climatic change and to the eutrophication induced by human activities (O'Neil et al. 2012;Sinha et al. 2012).
Th ere is no unanimity regarding the main environmental mechanisms that have permitted the expansion of C. raciborskii into temperate regions.However, achievements have been done to understand the factors that promote the success of this algae worldwide (Piccini et al. 2011;Bonilla et al. 2012).
Among them, wide ranges of environmental preferences and tolerance to variable light intensity have been referred to as factors that promote the expansion of C. raciborskii (Briand et al. 2004;Piccini et al. 2011).
Light is a significant factor regulating the growth and bloom development of C. raciborskii as this species apparently has a peculiar response regarding this abiotic parameter.Historically, Padisák & Reynolds (1998) considered C. raciborskii a cyanobacterium adapted to grow under low light intensities.Later, more research in the field and using strains from different water bodies has pointed out C. raciborskii grows in a wide range of light intensity (Saker et al. 1999;O'brien et al. 2009;Bittencourt-Oliveira et al. 2011;Pierangelini et al. 2014), while the light requirements for growth were low (I k near to 20 μmol photons m -2 s -1 ; Briand et al. 2004;Wu et al. 2009;Bonilla et al. 2012;Gomes et al. 2013).These results indicate that C. raciborskii is tolerant to a variety of light conditions and, together with its eurithermy, are the main promoters of the current expansion of this cyanobacterium to temperate latitudes.
Several authors have been investigating the effects of light in the ecophysiology of C. raciborskii (e.g.Dyble et al. 2006;Carneiro et al. 2009;Mehnert et al. 2010;Marinho et al. 2013); however, many aspects of this interaction have yet to be further elucidated.For instance, few investigations explored the isolated effects of light on the growth of strains from different latitudes, especially in subtropical and tropical environments (Briand et al. 2004;Dyble et al. 2006;Carneiro et al. 2009;Piccini et al. 2011;Bonilla et al. 2012).This cyanobacterium is widespread in Brazilian freshwaters (Bouvy et al. 2000;Sant'Anna & Azevedo 2000;Tonetta et al. 2015).However, only a few Brazilian strains from Pernambuco state (Briand et al. 2004;Marinho et al. 2013), São Paulo (Carneiro et al. 2009) and Minas Gerais (Marinho et al. 2013) were used for light experiments.
Here, we aimed to investigate the morpho-physiological responses of C. raciborskii to different levels of light intensity, after cultivating a strain isolated from a subtropical reservoir used for water supply.We hypothesized that; first, high light intensities could enhance the potential growth of C. raciborskii and; second, C. raciborskii net growth could be sustained even under low light conditions.

Environmental settings
The Alagados reservoir (25°01'09"S and 50°03'43"W), located in the Paraná state, South Brazil, is used for public water supply and hydroelectric power generation.This reservoir is classified as eutrophic to hypereutrophic (IAP 2004;2009), with 8.1 meters average depth and area of 7.2 km 2 (Rodrigues et al. 2005).Its lentic region is polymictic, with water temperature varying from 12°C to 27°C (IAP 2009).Z eu /Z mix is less than 1 in most of the year (on average 0.7), thus, the Alagados reservoir can be considered a turbid environment.Seasonal blooms of Cylindrospermopsis raciborskii (Woloszynska) Seenayya et Subba Raju have regularly been recorded since 2001 (IAP 2009), following a characteristic pattern; increased cell division in late spring, reaching the highest cell densities from January to May.Such a trend coincides with the seasonality of light intensity (and temperature) in South Brazil (Fernandes et al. 2005a).Regular monitoring since 2001 carried out by the state Water Treatment Company (SANEPAR) recorded cell densities up to 4 x 10 5 cell mL -1 ; historically, the highest abundance reached 4.25 x 10 6 cell mL -1 in August 2006 (IAP 2009).Saxitoxins are commonly recorded in the reservoir associated with C. raciborskii blooms.Normally, toxin levels range from 14 μg L -1 down to undetectable; a maximum of 644.9 μg L -1 was reported in July 2002 (Fernandes et al. 2005b).High cell abundances and presence of saxitoxins were already responsible for sporadic interruptions of supplying drinking water in the region.
During winter, average incident daily solar radiation in the region is around 900 W m -2 , comparatively lower than the maximum daily solar radiation during summer (1300 W m -2 ).Day lengths of 11-12 hours in winter are shorter than in summer days, when longer days range from 13 to 14 hours.

Sampling, isolation and culture conditions
The strain of C. raciborskii was isolated from samples obtained with a plankton net 20 μm mesh aperture, during a bloom in the Alagados reservoir in May 2011.Pre-monocultures of C. raciborskii were kept with ASM-1 medium (Gorham et al. 1964), but modified to use reservoir local water instead distilled water and pH adjusted from 7.0-7.5 to 7.8.Pre-cultures were kept in incubators under 100 μmol photons m -2 s -1 light provided by daylight fluorescent bulbs (GE 20 Watts) at 20 ± 1ºC, with 12:12 hours light:dark photoperiod.

Experimental design
Prior to the experiment, two methodologies for monitoring the cell growth were tested: cell abundance (cell mL -1 ), estimated from counting in inverted microscope, and optical density (OD) at 750 nm absorbance through readings in spectrophotometer.As cell abundance was strongly correlated with optical density (r = 0.929, n = 22, P < 0.01), we performed our experiments employing the latter method for practical reasons.
Inoculates of the pre-culture previously acclimated (one life cycle, 10 days) in each light intensity were used to set up cultures in 300 mL Erlenmeyer flasks filled with 200 mL of ASM-1 medium.Cell abundance in each initial inoculate was about 2 x 10 6 cell mL -1 , collected from the pre-cultures in exponential growth phase.To investigate the effects of light in the growth of C. raciborskii, nine light intensities (9,20,50,80,100,125,150,200 and 250 μmol photons m -2 s -1 ) were tested at 25°C and 12:12h light:dark cycle, manually shaken every day.Proper illumination was ensured by placing the flasks at specific distances from the light source and by using neutral screening.Light intensities were measured with a LI-COR model LI-193 underwater spherical quantum sensor immersed in distilled water.Experiment was carried out in triplicates.Samples were taken under sterile conditions from each of the treatments every 24 hours at the same hour over 20 days.Optical density at 750 nm absorbance of each of the samples was measured with a Hitachi U-2910 spectrophotometer.Cell abundances (cell mL -1 ) for three treatments (9, 100 and 250 μmol photons m -2 s -1 ) were estimated from counting trichomes and cells in Sedgewick-Rafter chambers.A minimum number of 1000 trichomes were counted.Fifty trichomes of each treatment were measured in inverted microscope (Olympus IX70, equipped with phase contrast) at 2-days intervals (n = 5400 trichomes).

Estimation of growth and statistics
Growth rate (μ day -1 ) of each treatment was calculated daily, after Andersen (2005).Maximum growth rate μ max , initial slope of the light vs. growth rate relationship (α), and the light intensity approaching the growth saturation I k (I k = μ max α -1 ) were obtained from the adjusted model of Jassby & Platt (1976).One-way analysis of variance (ANOVA) was performed to verify significant differences between the average growth rates in replicates cultures incubated at different light intensities, as well as differences between trichome lengths in the three specific light intensities selected.A Tukey HSD multiple comparison analysis was made a posteriori to discriminate the treatments showing significant differences (P < 0.05).Analyses were performed in R software 3.1.3(R Core Team 2015) using package stats (Chambers et al. 1992).
Cultures grown at light intensities higher than 50 μmol photons m -2 s -1 reached optical densities as high as 0.250 and 0.350; at least 20 times higher than observed for 9 and 20 μmol photons m -2 s -1 treatments (Fig. 1).These cultures entered the log-growth phase at the third or fourth day incubation except the culture at 125 μmol photons m -2 s -1 , which started the log growth earlier in the second day.Log phase lasted 8 to 10 days in all the treatments ranging from 50 and 250 μmol photons m -2 s -1 .There were no significant differences in growth among the cultures submitted to higher light intensities (50 -250 μmol photons m -2 s -1 ).

Morphology of trichomes
Average length of trichomes in cultures grown at 9 μmol photons m -2 s -1 was significantly larger than in the treatments at 100 and 250 μmol photons m -2 s -1 (F 2.24 = 13.103,P < 0.001) (Fig. 3).At low light intensity (9 μmol Morpho-physiological responses of a subtropical strain of Cylindrospermopsis raciborskii (Cyanobacteria) to different light intensities photons m -2 s -1 ) trichomes were longer, ranging from 313 to 790 μm length, while cultures at higher light presented shorter trichomes, 80 to 320 μm.In all three treatments monitored, trichomes remained straight throughout the experiment.

Growth responses
In this experiment, increasing light intensities promoted the growth enhancing of the C. raciborskii strain isolated from the Alagados reservoir.The growth rate doubled when the strain was cultivated at 50 μmol photons m -2 s -1 or higher compared to 9 μmol photons m -2 s -1 .Moreover, the strain attained maximum growth over a wide range of light intensity (50 -250 μmol photons m -2 s -1 ).This finding differs from other results, which reported a more narrow range of optimum light (75 to 125 μmol photons m -2 s -1 ) for strains isolated from a variety of sites (Dyble et al. 2006;Bonilla et al. 2012;Briand et al. 2004).The ability of C. raciborskii to grow faster than other diazotrophic species under high light intensities is an important physiological feature (Fabbro & Duivenvoorden 1996;Briand et al. 2004) that enhances their potential threat for water use, particularly in tropical regions.
During our experiment, the C. raciborskii strain showed no evidence of photoinhibition when submitted to light levels as high as 250 μmol photons m -2 s -1 , considering that maximum optical density and pattern of growth curves were statistically similar to the treatments at 50 to 200 μmol photons m -2 s -1 .Indeed, strains from different latitudes are capable of fast growth rate (> 0.5 day -1 ) under light levels as high as 500 μmol photons m -2 s -1 in laboratory (Briand et al. 2004).It might be that C. raciborskii is adapted to grow (up to bloom densities) at even higher irradiances in natural conditions than those tested in laboratory, especially in tropical regions subjected to year-round high solar irradiances (e.g.Dokulil & Mayer 1996;Fabbro & Duivenvoorden 1996).Therefore, our data corroborate previous laboratory evidence suggesting that C. raciborskii is able to explore a wide range of light intensities, spanning from 20 to 500 μmol photons m -2 s -1 at least (Bouvy et al. 1999;2000;Briand et al. 2004;our results).
Cultures incubated at low light intensities (9 and 20 μmol photons m -2 s -1 ) were unable to attain intensive cell division, showing significantly lower growth rate (0.14 and 0.24 day -1 , respectively) when compared to the other higher light treatments tested.Nevertheless, the strain sustained net increase even under low light conditions, with potential implications to the success of C. raciborskii.This is particularly relevant in Alagados reservoir, where light is a regulating factor of phytoplankton growth during late fall and winter, due to both the seasonal declining of atmospheric irradiance coupled with shortening of day length in fall and winter in South Brazil (Fernandes et al. 2005a), and the mixing zone usually greater than the photic zone (unpublished data).Large overwintering populations are one reason why cyanobacteria with low specific growth rates become dominant in summer  phytoplankton communities (Reynolds 1994).As reported by Dokulil (2015), the C. raciborskii population of Lake Alte Donau survived adverse periods, being able to inoculate the phytoplankton assemblage in the following spring.The same was reported by other authors in other regions of the globe (Everson et al. 2011;Wood et al. 2014), including the temperate zone, which most authors suggest that the species survive in winter in the form of akinetes (Mehnert et al. 2010).
The light saturation parameter (I k ) is a reliable indicator to evaluate light requirements of a certain species and to allow for comparisons between species (Briand et al. 2004).The calculated I k for the strain tested here is 19 μmol photons m -2 s -1 (α = 0.0361), which is similar to most other values for C. raciborskii from different regions of the world, ranging from 15 to 26 μmol photons m -2 s -1 (Shafik et al. 2001;Briand et al. 2004;Dyble et al. 2006).Briand et al. (2004) included four tropical strains (two from Brazil), but no correlationship was found between latitude and I k , suggesting there is no separation among the clones tested.Recently, Bonilla et al. (2012) recorded I k as low as 8 μmol photons m -2 s -1 for Uruguayan strains isolated from shallow lakes.In our case, the strain can be considered highly adapted to low light due to its low I k and adequate growth rates (μ max = 0.26 at 50 μmol photons m -2 s -1 ), being potentially able to saturate photosynthesis even at low irradiances in the field.Among diazotrophic genera, relatively low I k are commonly found in bloom-forming species adapted to low luminosity like Planktothrix agardhii, Planktothrix rubescens and Limnothrix redekei (Padisák & Reynolds 1998;Reynolds et al. 2002;Bonilla et al. 2012).These physiological traits led Reynolds et al. (2002) to classify those cyanobacteria and C. raciborskii in the functional groups S and R (tolerant to low light conditions).The capacity of undergoing growth at low irradiance also enables C. raciborskii to thrive under the light limiting conditions imposed by periods of intensive phytoplankton growth (Padisák 1997;Padisák & Reynolds 1998;Briand et al. 2002;Havens et al. 2003).The few previous investigations in Brazil recording light intensity in the field found elevated abundances of C. raciborskii at irradiances as low as 15 μmol photons m -2 s -1 (Bouvy et al. 1999;2000) in shallow eutrophic lakes.Tolerance to shading, usually prevalent in turbid environments, adds to the factors invoked to explain the dominance of C. raciborskii in Brazilian lakes.Therefore, in polimyctic reservoirs with active vertical mixing like Alagados and the ability of sustaining growth under variable light confers a relevant ecological advantage to C. raciborskii, particularly since this species can compensate lower light availability by saturating its photosynthetic rate.

Morphological responses
The C. raciborskii strain also showed an interesting morphological response to the different light regimes tested in this work.Significant increase in length, up to 790 μm, was observed in trichomes incubated at low light intensity (9 μmol photons m -2 s -1 ) in comparison to the shorter trichome length achieved at higher intensities.
The relevance of investigating the morphology of trichomes and cells of cyanobacteria are two-fold.First, information about length, width, presence or absence of heterocytes and akinetes among others furnishes support to taxonomic studies.Second, alteration in morphology can reflect changing in environmental parameters.Cylindrospermopsis raciborskii has been found highly variable morphologically, making its identification somewhat difficult in some situations (Komárková et al. 1999).Most of the studies investigating the factors responsible for changes in morphology of C. raciborskii tested temperature (Chonudomkul et al. 2004) or nutrients (Saker & Neilan 2001;Shafik et al. 2003).On the other hand, only a few papers aimed to verify the influence of light intensity on trichomes (Bittencourt-Oliveira et al. 2012;Bonilla et al. 2012;Beamud et al. 2016).
It is difficult to discuss the factors underlying the observed changes in C. raciborskii morphology.We preliminarily attribute the enlargement of trichomes in our experiment as an adaptation to optimize light absorption under limiting conditions.However, contrary to our results, Bonilla et al. (2012) observed longer and more voluminous individuals when cultivated at higher light intensities (100 μmol photons m -2 s -1 ).In the field, these authors found no correlation between biovolume of C. raciborskii and ambient light availability in the water column.Moreover, other authors found a positive relationship between length of trichomes and input of nutrients either in laboratory or environmental conditions, especially phosphorus and nitrogen (Komárková et al. 1999;Saker & Neilan 2001;Shafik et al. 2003).

Concluding remarks
Laboratory experiments are one of many important steps to understand the adaptiveness and success of toxigenic species in various freshwater systems across latitudes as well as having specific water circulation patterns and trophic status.We recognize that a number of genetically diverse strains should make up the population of C. raciborskii in reservoirs and the response of individual strains to light may be specific.Therefore, the physiological responses of one or two strains from a given reservoir do not necessarily reflect what happens with the population.Nonetheless, our laboratory results suggest that C. raciborskii can survive under low light conditions, still producing viable trichomes.Additionally, our results also give additional support to the hypothesis that C. raciborskii is a tolerant cyanobacterium adapted to thrive in distinct light conditions irrespective of latitudinal variation (Briand et al. 2004).Hence, the species can dominate the phytoplankton in Morpho-physiological responses of a subtropical strain of Cylindrospermopsis raciborskii (Cyanobacteria) to different light intensities tropical, subtropical or temperate freshwater systems; such a geographic spreading should have also been favored by the observed rising temperatures of lakes around the world in the last three decades (Bonilla et al. 2012).Nonetheless, further investigations using a greater number of strains and from different latitudes are recommended to better elucidate the role of light on the growth, physiology and phenotypic plasticity of C. raciborskii.Moreover, laboratory experiments coupled with field studies will allow for a better understanding of the causes underlining the seasonal blooming of this harmful species in subtropical regions.

Figure 1 .
Figure 1.Growth curves of C. raciborskii strain from Alagados reservoir at nine different light intensities, based on readings at optical density of 750 nm.Vertical bars indicate the range of values for three replicates.Codes are as follows: L9 = light intensity at 9 μmol photons m -2 s -1 and so on.

Figure 2 .
Figure 2. Growth rates (μ day -1 ) of C. raciborskii isolated from Alagados reservoir at different light intensities, with the indication of μ max (0.3 μ day -1 ).Vertical bars indicate the range of values for three replicates.