Protozoologica. How Might Mixing Bias Protozoan-experiments that Use the Common Micro-alga Isochrysis galbana? Acta Protozool. (2008) 47:

Transcription

1 Acta Protozoologica Acta Protozool. (2008) 47: How Might Mixing Bias Protozoan-experiments that Use the Common Micro-alga Isochrysis galbana? Naomi Downes-Tettmar and David J.S. Montagnes School of Biological Sciences, University of Liverpool, Biosciences Building, Crown Street Liverpool, UK Summary. Although there are strong recommendations to mix micoalgal cultures, to improve productivity, there is only anecdotal evidence to encourage protozoologists to mix cultures, to maintain constant quality of microalgae. Isochrysis galbana is extensively used in laboratory experiments associated with applied and pure protozoan studies; thus, there is a need to assess how routine mixing may bias experiments. Although, mixing of cultures harms some microalgae, there is no indication of how mixing affects I. galbana. We address this problem by assessing mixing effects on key experimental variables: specific growth rate, cell volume, nutritional quality (as carbon and nitrogen content and ratio), and production (the product of growth rate and carbon or nitrogen content). Treatments, quantified as dissipation (ε) m 3 s -1 were: mixed once per day (ε = 0), rotated on a plankton wheel (ε = ), and shaken on agitating tables (ε = , ). Mixing levels regularly used in experimental work (no-mixing, plankton wheel, and low agitation) did not alter I. galbana specific growth rate and thus need not be carefully controlled for, but exceptionally high-mixing did significantly reduce growth rate. Cell volume decreased at exceptionally high-mixing, but otherwise there was no size-bias caused by mixing. Carbon and nitrogen levels were highest at low-mixing but were otherwise generally unaffected by mixing. The nutritional quality of I. galbana (i.e. carbon:nitrogen) remained unaltered by mixing. Production was doubled at low-mixing, compared to all other treatments (which did not differ). Overall, I. galbana is a relatively robust species that withstands mixing levels well within those experienced in the laboratory and still grows at exceptionally high-mixing levels. Thus, within reason, protozoologists using I. galbana do not need to be rigorous in their monitor of mixing. Furthermore, meta-analyses that examine this species need not be too concerned with variation in mixing methods as a confounding factor. However, the data suggest that gentle mixing, using a plankton wheel, is preferable to no-mixing. Key words: Biochemical composition, carbon, experimental design, growth, nitrogen, phytoplankton, production, nutritional quality. INTRODUCTION Although there are strong recommendations to mix micoalgal prey cultures to maintain their constant quality (Stein 1973), there is only anecdotal evidence to encourage protozoologists to mix their prey. In fact, key texts that outline routine maintenance methods Address for correspondence: D. Montagnes, School of Biological Sciences, University of Liverpool, Biosciences Building, Crown Street Liverpool, UK, L69 7ZB Telephone number: (44) ; E -mail: dmontag@liv.ac.uk. (e.g. Fogg 1965, Stein 1973, Boney 1975, Becker 1994, Andersen 2005) fail to provide guidance regarding the application, extent, or consequences of mixing. There is, admittedly, a belief by many protozoologists that mixing of cultures is important, but to what extent? Here we address this question, focusing on an extensively used experimental species in marine protozoology, to provide guidance for future work. Since its description ~60 years ago (Parke 1949), Isochrysis galbana has been extensively and routinely used in laboratory experiments associated with applied and pure marine sciences; every year 100s of studies

2 288 N. Downes-Tettmar and D.J.S. Montagnes examine this species. In these works, I. galbana is often employed as a model to investigate various algal processes (e.g. light responses, Thompson et al. 1990; temperature responses, Montagnes and Franklin 2001; clonal differences Sayegh et al. 2007). Furthermore, for the last 35 years I. galbana has routinely been used as a prey for a wide range of marine protozoa (e.g. Gold 1973, Stoecker et al. 1988, Bernard and Rassoulzadegan 1990, Verity 1991, Christaki et al. 1996, Montagnes 1996, Dolan and Šimek 1997, Jakobsen and Hansen 1997, Strom and Morello 1998, Klein Breteler et al. 1999, Johansson and Coats 2002, Kimmance 2006, Guermazi et al. 2008). The quality of I. galbana grown in experimental culture is thus a focus of concern, and studies have examined the impact of several key environmental variables (e.g. light, temperature, salinity, nutrients) on a range of attributes, including specific growth rate, composition (e.g. carbon and nitrogen content), predator-prey dynamics, and prey choice (e.g. Thompson et al. 1990, Montagnes and Franklin 2001, Montagnes et al. 2001). Here we test the hypothesis that mixing will affect the outcome of experimental studies that use I. galbana, as mixing microalgal cultures, in general, can alter their quality by reducing nutritional gradients, improving gas transfer, preventing sedimentation, and ensuring equal light exposure (Jiménez et al. 1996), but excessive mixing can be detrimental to some microalgae, disrupting nutrient uptake, causing cell wall damage, inducing filament loss, and increasing mortality (Thomas and Gibson 1990, Borowitzka 1997). If I. galbana responds like other microalgae, then this is a concern for experimental protozoologists, and if it is not affected, then this observation will provide extremely useful time-saving guidance for researchers. Surprisingly, virtually nothing is known regarding mixing effects on I. galbana, except that it is not affected by low-level turbulence (i.e. a dissipation rate of 10-4 m 2 s - 3 ; Havskum 2003). A review of the literature also reveals that a wide range of mixing levels are used in experimental work, and these are rarely quantified: some studies mix cultures once a day, others continuously mix them, and in many studies mixing it is not mentioned (e.g. Davidson et al. 1992, Kleppel and Burkart 1998, Martel 2006). Thus, we test if variation in mixing could alter the outcome of experiments that use I. galbana, and determine if future experiments should carefully consider mixing. To this end, we have assessed the impact of mixing I. galbana cultures over a range of levels that cover, and extend beyond, the spectrum potentially used in laboratory incubations. We have focused on examining mixing effects on key experimental variables examined by others and applicable to protozoan-related experiments: growth rate, cell volume, carbon and nitrogen content of cells (as proxies of nutritional quality), and production (the product of growth rate and carbon or nitrogen content). Methods Isochrysis galbana Parke 1949, CCAP 927/1 was maintained in artificial seawater, enriched with f/2 (Guillard 1972) at 16 ± 1 C under a constant irradiance of 85 µmol photons m -2 s -1. Cultures (sample volume 175ml) were grown in 250 ml plastic, rectangular tissue culture flasks. Four replicated treatments were applied (n = 3): non-mixing, where flasks were only mixed (by gentle rotation) once daily when sampled; low-mixing, where flasks were rotated on a vertically oriented plankton wheel ; medium and heavy mixing, where flasks were placed on horizontal agitating tables that moved back and forth, rather than rotating (Table 1). For the agitated samples, mixing was characterised as dissipation (ε, m 2 s -3 ); ε = U 3 /L, where U, the velocity (m s -1 ), was determined from the speed of agitation (Table 1), and L, length (m), was determined as the cube-root of the sample volume (0.056 m). For the rotated samples (low-mixing), turbulence was also assessed as dissipation, but this was a more complex calculation. In this case U was determined from the kinetic energy of rotation of the fluid and the bottle together, in the first instance assuming it was a ridged body; then assuming that the fluid rotates with the bottle, kinetic energy was converted to velocity by U = (0.5 Iω/M w ) 0.5, where M w is the mass of the water, ω is the angular rotation (2π/rotation time), and I is the moment of inertia, which is (M w + M f )r 2 (where r is the radius of rotation and M f is the mass of the flask), but assuming M w >> M f this approximates to M w r 2 ; U then simplifies to (0.5r 2 ω) 0.5. Furthermore, mixed samples are likely to experience shearing as well as a dissipation effect (e.g. Havskum 2003); thus we have made estimates Table 1. Characteristics of the four mixing regimes that Isochrysis galbana was exposed to. Calculations are described in the text. Regime Method Displacement (m) Cycles Velocity (U, m s 1 ) Dissipation (ε, m 2 s 3 ) Shear (S, s 1 ) None Mixed daily ~0 ~0 Low Rotated on wheel Wheel radius = Rotation time 202 s Medium Shaker table Shaking distance = cycles min High Shaker table Shaking distance = cycles min

3 Mixing and Isochrysis galbana 289 of shear rate (S, s -1 ), using the above parameters (i.e. S = U/L, see Table 1). However, for the purpose of this study we have focused on dissipation (ε) as a parameter to quantify mixing. See Tennekes and Lumley (1972) for a more detailed description of these estimates. Specific growth rates (µ, d -1 ) were determined by regressing ln cell numbers vs. time, over the exponential growth phase. Cell volumes were determined from linear dimensions of >30 live cells, harvested from exponentially growing cultures, using an image analysis system (Scion image for Windows, Scion Corp., MD, USA); cells were assumed to be prolate spheroids. Carbon and nitrogen content (pg cell -1 ) was determined from duplicate samples, from each replicate, by elemental analysis (NC 2500, CE Instruments). The C:N ratio, C and N per unit volume, and C and N production (i.e. the product of specific growth rate and biomass) were then determined. Treatment effects on the above measurements were compared using one way ANOVA, followed by pairwise comparisons (Holmsidak method, α = 0.05, n = 3). All datasets were normally distributed and homoscadastic; Kolmogorov-Smirnov test). We have addressed this issue by assessing growth rate, volume, nutritional quality, and production at a range of mixing levels that cover and extend above those typically used in experimental laboratory practices (Table 1, Richmond 1987); each of these is, in turn, addressed below. Growth rate is undoubtedly the most measured parameter associated with examining microalgae; for I. galbana it is used as an index of physiological response (e.g. to temperature Montagnes and Franklin 2001), as a component of population dynamic studies and ecosystem models (e.g. Kimmance et al. 2006), and as an integral part of other measurements (e.g. determination of grazing on microalgae, Kimmance et al. 2006). We found that I. galbana growth rate decreased due to heavy mixing (a level far beyond that typically applied to laboratory cultures) but remained unaffected by the Results There was a significant decrease in specific growth rate of I. galbana between the lowest and highest mixing levels (Fig. 1a), but non-mixing was not significantly different from low-mixing or medium-mixing. Cell volume was significantly higher in low-mixing compared with high-mixing, and there was an apparent decrease in volume with mixing, although there was no significant difference between all the other treatments (Fig. 1b). The carbon and nitrogen analysis demonstrated that at low-mixing C and N cell quota was significantly higher than all other treatments, and there were no significant differences across the other mixing treatments (Fig. 1c). There was no significant effect of mixing on C:N ratio (Fig. 1d). The estimates of C and N per unit volume were significantly higher at high-mixing, and there were no significant differences between the other mixing treatments (Fig. 1e). Production (in terms of both C and N) was significantly higher at low-mixing compared with the other mixing treatments, and there were no differences between the other mixing treatments with the exception of N production where there was a small but significant difference between medium-mixing and high-mixing (Fig. 1f). Discussion Given the extensive use of Isochrysis galbana in experimental protozoological works (see Introduction), there is a need to assess if routine maintenance will alter culture parameters and potentially bias experiments. Fig. 1. A comparison of Isochrysis galbana responses to mixing. a comparison of the specific growth rates (µ, d -1 ); b comparison of cell volume (µm 3 ); c comparison of the total carbon and nitrogen contents (pg cell -1 ); d comparison of C:N ratios; e comparison of the carbon and nitrogen content per unit volume (C, N pg µm -3 ); f comparison of production C, N pg d -1 ). Error bars are one standard error. See Table 1 and Methods for a detailed explanation of treatments. In Figs 1c, e, and f, carbon is represented by ( ), and nitrogen is represented by ( ).

4 290 N. Downes-Tettmar and D.J.S. Montagnes other mixing treatments. This decrease in growth rate at an excessively high level could result from increased cell lyses and damage at the air surface interface (Havskum 2003) or hydrodynamic stress acting on the cells causing a physiological disruption (Thomas and Gibson 1990). This might be relevant if studies were interested in the impact of breaking waves on shallow-water, coastal populations, but otherwise mixing would have little ecological or laboratory significance. Clearly, I. galbana is a robust species, compared to other more sensitive microalgae (Thomas and Gibson 1990, Borowitzka 1997). Small changes in prey cell size will also alter experiments with protozoa; e.g. prey choice and grazing experiments may be influenced by prey size (e.g. Jonsson 1986) and changes in volume may alter the cell nutritional quality (see below). However, I. galbana cell volume remains unaffected by lower mixing levels and only decreases with the influence of heavy mixing. Possibly heavy mixing inhibits nutrient uptake at the cell surface thus selecting for smaller cells, or possibly the small cell size is an adaptive mechanism to reduce cell lyses (see above). Regardless, of the cause, there does appear to be a potential size-bias caused by mixing. Protozoologists must also recognise the potential impact of mixing, on the nutritional quality of I. galbana; this is often determined by examining total carbon (C) and nitrogen (N) cell content (e.g. Mitra and Flynn 2005). We found that C and N quotas per cell reflect the overall trend observed in cell volume, with large cells at low-mixing having the highest C and N levels and small cells at high-mixing having the lowest C and N quotas. However, in terms of C and N per unit volume the relative densities of C and N were the same between non- and medium mixed treatments, but at high-mixing the smaller cells were more densely packed with C and N. We also found that there was no significant stoichiometric change in C:N ratio of cells; this ratio can be used as an index of prey quality, and our data thus suggests that, in this respect, mixing will have little influence on predator responses to I. galbana. Consequently, while mixing may affect growth rate and cell volume at high-mixing, the overall nutritional quality of I. galbana remains unaltered. Finally, many experimental measurements (e.g. grazing rates, population dynamics) require estimates of production, rather than simply growth rate, as external stimuli may influence both growth rate and nutritional quality of cells (e.g. Montagnes and Franklin 2001). Furthermore, there are many cases where production needs to be maximised, such as when rearing batch cultures of algae as food in grazing and prey choice experiments (Montagnes et al. 2001). We found a clear trend across mixing treatments: as mixing is increased there is a decrease in production, although unmixed cultures had a significantly lower production than gently mixed ones. Note that production, being the derived product of growth rate and carbon or nitrogen content of cells accentuates the trends exhibited by these independently. Our results illustrate that low-mixing approximately doubles the productivity of cultures, supporting the case to use plankton wheels to optimise production. In conclusion, mixing effects on I. galbana quality are contrary to those expected from past studies, which indicate that microalgae can be highly sensitive to mixing (see Introduction). In contrast, I. galbana is a robust species that withstands low to medium-mixing and is not prevented from growth even at unrealistically highmixing levels. This suggests that within reason, protozoologists, and other researchers, using I. galbana, do not require rigorous monitoring of mixing and that meta-analyses that compare the many studies on this species (e.g. Sayegh et al. 2007) need not be concerned with variation in mixing methods. Even though we suggest mixing variations are not important, we still found effects of extreme mixing levels and, therefore, can provide some guidance. For instance, I. galbana was deleteriously affected by our high-mixing treatment; fortunately, this level is rarely, if ever, used in the laboratory and would even be beyond levels experienced when samples are shipped in the post. Furthermore, we found that low-mixing, which reflects most laboratory conditions, is the most favourable treatment in terms of maximising cell size, production, and C and N content. Finally, there was a clear indications that continuous low-mixing, provided more productive cultures with cells containing more carbon and nitrogen than non-mixed (shaken once per day) cultures. Many researchers do not continuously mix their cultures, and this may result in some bias. Thus, we recommend that although I. galbana is relatively insensitive to mixing, experimental work should probably be conducted on gently mixed cultures, using a plankton wheel (e.g. Crocker and Gotschalk 1997), which provides homogeneous mixing at extremely low levels; i.e. lower than those achievable on a horizontal mixing plate. Acknowledgements. We thank Jason Dzikowski, Jan Bissinger, Sam Fielding, David Lewis, and Fabienne Marret for comments and assistance. We also thank the Esmée Fairbairn Foundation, the British Ecological Society, and the University of Liverpool Resarch Development Fund for financial assistance.

5 Mixing and Isochrysis galbana 291 References Andersen R. A. (2005) Algal Culturing Techniques, Elsevier Academic Press, London, 160 pp. Becker E. W. (1994) Microalgae Biotechnology and Microbiology, Cambridge University, Cambridge, Bernard C., Rassoulzadegan F. (1990) Bacteria or microflagellates as a major food source for marine ciliates: possible implications for the microzooplankton. Mar. Ecol. Prog. Ser. 64: Boney A. D. (1975) Phytoplankton. Edward Arnold, London, Borowitzka M. A. (1997) Microalgae for aquaculture: Opportunities and constraints. J. App. Phycol. 9: Christaki U., Belviso S., Dolan J. R., Corn M. (1996) Assessment of the role of copepods and ciliates in the release to solution of particulate DMSP. Mar. Ecol. Prog. Ser. 141: Crocker K. M., Gotschalk C. (1997) A simple seawater-powered plankton wheel. J. Plankton Res. 19: Davidson K., Flynn K. J., Cunningham A. (1992) Non-steady state ammonium-limited growth of the marine phytoflagellate, Isochrysis galbana Parke. New Phytol. 122: Dolan J. R., Šimek K. (1997) Processing of ingested matter in Strombidium sulcatum, a marine ciliate (Oligotrichida). Limnol. Oceanogr. 42: Fogg G. E. (1965) Algal Cultures and Phytoplankton Ecology. University of Wisconsin Press, Madison, pp Gold K. (1973) Methods for growing Tintinnida in continuous culture, Amer. Zool. 13: Guermazi W., Elloumi J., Ayadi H., Bouain A., Aleya L. (2008) Rearing of Fabrea salina Henneguy (Ciliophora, Heterotrichida) with three unicellular feeds. C. R. Biologies 331: Guillard R. R. L. (1972) Culture of Phytoplankton for Feeding Marine Invertebrates. In: Culture of marine invertebrate animals, (Eds. W. L. Smith, M. H. Chanley). Plenum Press, New York, Havskum H. (2003) Effects of small-scale turbulence on the interactions between the heterotrophic dinoflagellate Oxyrrhis marina and its prey Isochrysis sp. Ophelia 57: Jakobsen H. H., Hansen P. J. (1997) Prey size selection, grazing and growth response of the small heterotrophic dinoflagellate Gymnodinium sp. and the ciliate Balanion comatum a comparative study. Mar. Ecol. Prog. Ser. 158: Jiménez C., Mercado J. M., Aguilera J., Rodríguez-Maroto J. M., Niell F. X. (1996) Effects of turbulence and inorganic carbon supply on growth of Dunaliella viridis Teodoresco. Int. J. Salt Lake Res. 4: Johansson M., Coats D. W. (2002) Ciliate grazing on the parasite Amoebophrya sp. decreases infection of the red-tide dinoflagellate Akashiwo sanguine. Aquat. Microb. Ecol. 28: Jonsson P. R. (1986) Particle size selection, feeding rates and growth dynamics of marine planktonic oligotrichous ciliates (Ciliophora: Oligotrichina). Mar. Ecol. Prog. Ser. 33: Kimmance S. A., Atkinson D., Montagnes D. J. S. (2006) Do temperature-food interactions matter? Responses of production and its components in the model heterotrophic flagellate Oxyrrhis marina. Aquat. Microb. Ecol. 42: Klein Breteler W. C. M., Schogt N., Baas M. (1999) Trophic upgrading of food quality by protozoans enhancing copepod growth: role of essential lipids. Mar. Biol. 135: Kleppel G. S., Burkart C. A., Houchin L. (1998) Nutrition and the regulation of egg production in the calanoid copepod Acartia tonsa. Limnol. Oceanogr. 43: Martel C. M. (2006) Prey location, recognition and ingestion by the phagotrophic marine dinoflagellate Oxyrrhis marina. J. Exp. Mar. Biol. Ecol. 335: Mitra A., Flynn K. J. (2005) Predator-prey interactions: Is ecological stoichiometry sufficient when good food goes bad? J. Plankton Res. 27: Montagnes D. J. S. (1996) Growth responses of planktonic ciliates in the genera Strobilidium and Strombidium. Mar. Ecol. Prog. Ser. 130: Montagnes D. J. S., Franklin D. J. (2001) Effect of temperature on diatom volume, growth rate, and carbon and nitrogen content: Reconsidering some paradigms. Limnol. Oceanogr. 46: Montagnes D. J. S., Kimmance S. A., Tsounis G., Gumbs J. (2001) Combined effect of temperature and food concentration on the grazing rate of the rotifer Brachionus plicatilis. Mar. Biol. 139: Parke M. W. (1949) Studies on marine flagellates. J. Mar. Biol. Ass. U.K. 28: Richmond A. (1987) The challenge confronting industrial microagriculture: High photosynthetic efficiency in large scale reactors. Hydrobiologia 151/152: Sayegh F. A. Q., Radi N., Montagnes D. J. S. (2007) Do strain differences in microalgae alter their relative quality as a food for the rotifer Brachionus plicatilis? Aquaculture 273: Strom S. L., Morello T. A. (1998) Comparative growth rates and yields of ciliates and heterotrophic dinoflagellates. J. Plank. Res. 20: Stein J. R. (ed.) (1973) Handbook of Phycological Methods: Culture Methods & Growth Measurements. Cambridge Univ. Press, Cambridge, Stoecker D. K., Silver M. W., Michaels A. E., Davis L. H. (1988) Obligate mixotrophy in Laboea strobila, a ciliate which retains chloroplasts. Mar. Biol. 99: Tennekes H., Lumley J. L. (1972) A first course in turbulence. MIT Press, Cambridge, 300 pp. Thomas W. H., Gibson C. H. (1990) Effects of small-scale turbulence on microalgae. J. Appl. Phycol. 2: Thompson P. A., Harrison P. J., Whyte J. N. C. (1990) Influence of irradiance on the fatty acid composition of phytoplankton. J. Phycol. 26: Verity P. G. (1991) Measurement and simulation of prey uptake by marine planktonic ciliates fed plastidic and aplastidic nanoplankton. Limnol. Oceanogr. 36: Received on 20 th June, 2008; revised version on 23 rd September, 2008; accepted on 25 th September, 2008