Aquatic Ecology 36: 219–227, 2002. © 2002 Kluwer Academic Publishers. Printed in the Netherlands.

219

Heterotrophic capability of a metalimnetic plankton population in saline Lake Shira (Siberia, Khakasia) Antonio Quesada1 , Friedrich Jüttner2 , Tatiana Zotina3, Alexander P. Tolomeyev3 and Andrei G. Degermendzhy3 1 Departamento

de Biolog´ıa. Universidad Aut´onoma de Madrid, 28049 Madrid, Spain (Fax: 34-91-3978344; E-mail: [email protected]); 2 Institut für Pflanzenbiologie, Limnologische Station, Seestrasse 187, CH8802 Kilchberg, Switzerland; 3 Institute of Biophysics of SB RAS, Akademgorodok, Krasnoyarsk 660036, Russia Accepted 18 September 2001

Key words: cyanobacteria, glucose uptake, mixotrophy, photoheterotrophy

Abstract The heterotrophic potential of a deep (12 m) phytoplankton community layer in Lake Shira (Siberia), dominated by several taxa of cyanobacteria (Aphanocapsa, Lyngbya contorta, and other unidentified species) was investigated. The plankton community was fractionated by size, allowing separation between the bacterioplankton and the phytoplankton, and 13 C-labelled organic compounds were used as tracers. The uptake of 13 C-labelled glucose and of 13 C-labelled glycine was maximal in the bacterioplankton-enriched fraction (δ 13 C = 557 and 323, respectively), but was also high in the cyanobacterial fraction (δ 13 C = 138 and 80, respectively). An inverse relationship between the uptake of organic compounds and the light intensity when the whole community was exposed to different irradiances was also investigated. These results suggest that the photosynthetic microorganisms from the investigated community are able to assimilate organic compounds and thus supplement their carbon and energy requirements. This heterotrophic capability appears to be favoured by the high in situ concentrations of dissolved organic carbon (> 15 mg C l−1 ), and may offset the effects of severe light limitation on the phytoplankton in this deep, highly shaded environment.

Introduction All aquatic ecosystems are subjected to light attenuation caused by constituents of the water. The deeper parts of the water body may receive no light at all, and in the upper part a gradient of decreasing light intensity is formed with increasing depth. Typically, the depth at which 1% of the surface photosynthetic active radiation is reached is considered to be the limit for the net photosynthetic activity of the organisms, where these organisms produce more than they consume. However, in many aquatic ecosystems the highest concentrations of phytoplankton biomass have been recorded at or below this depth, which is often associated with a deep chlorophyll a maximum (e.g. Eguchi et al., 1996; Bright & Walsby, 2000). At these depths, light is likely to limit photosynthesis during a

considerable part of the day, especially under cloudy weather conditions. Algal and cyanobacterial taxa have been described in laboratory experiments as potential consumers of organic compounds. Photosynthetic organisms with an ability to take up dissolved organic carbon (DOC) may express four different kinds of energy metabolism: – photoautotrophic metabolism in which the microorganisms do not make use of the organic compounds in the solution of the aquatic environment; – chemoheterotrophic metabolism in which the microorganisms can grow in darkness using organic compounds; – photoheterotrophic metabolism in which the microorganisms take up organic carbon concomitant

220 with the inhibition of assimilating dissolved inorganic carbon (DIC); – mixotrophic metabolism in which the microorganisms take up and metabolise simultaneously both the organic and inorganic carbon (Nieva, 1994). These capabilities seem to be widespread among algae and cyanobacteria. For example, about 50% of the cyanobacterial strains tested were capable of growing under photoheterotrophic conditions, of which 15–20% were able to grow under chemoheterotrophic conditions (Tandeau de Marsac & Houmard, 1993). Lewitus & Kana (1994) found that the uptake of dissolved organic compounds by phytoplankton organisms is a common activity in estuarine ecosystems. McKnight et al. (2000) also suggest that heterotrophic capability in phytoplankton is one of the strategies for surviving through the Antarctic winter. Typically, photosynthetic microorganisms exhibit lower growth rates under chemoheterotrophic conditions than under photoautotrophic conditions (Peschek, 1987). The simultaneous uptake of organic and inorganic compounds has been less thoroughly investigated, but some strains yielded higher growth rates than those obtained under photoautotrophic conditions (Fernández-Valiente et al., 1992). However, distinguishing between photoheterotrophic and mixotrophic metabolisms in field experiments is very difficult. Thus, in this paper the term mixotrophy is used to refer to any uptake of dissolved organic C by photosynthetic organisms. In the present study, dim environments rich in DOC are places where the photosynthetic organisms with heterotrophic capabilities are better suited and they may use organic compounds as a supplement to acquire energy and reduced carbon for maintenance and growth (Carr, 1973). Nevertheless, in some cyanobacterial strains the uptake of organic compounds takes place at any irradiance (Prosperi et al., 1992). It is not clear to what extent the photosynthetic microorganisms make use of the supplementary energy and carbon sources under natural conditions. Most authors (e.g. Ellis & Stanford, 1982; Vincent & Goldman, 1980) believe that the uptake rates of sugar by the phytoplanktonic organisms in the natural environments are too low to be ecologically significant, although these workers assume the potential ecological advantages of having this capability. Heterotrophy in photosynthetic microorganisms in the laboratory has been typically determined using axenic cultures by measuring growth parameters or the uptake of radiolabelled organic substances, or both (Smith & More, 1988). Such experiments are not

easily applicable to the field situation, because the radiolabelled substances are also taken up by the nonphotosynthetic heterotrophic microorganisms. Some workers used autoradiographic techniques to identify the organisms taking up these radiolabelled organic compounds (Saunders, 1972). Most of the laboratory experiments were conducted with D-glucose, which is one of the most abundant free sugars in aquatic environments (Smith, 1982), although acetate has also been used in field experiments (e.g. Vincent & Goldman, 1980). We applied size-fractionation of the lake water to separate the microbial plankton community into fractions enriched in bacterioplankton and phytoplankton. This separation methodology, together with the utilisation of 13 C-labelled organic compounds, represents a suitable experimental design to identify in situ the heterotrophic potential of photosynthetic microorganisms. In this paper the results obtained in situ in Lake Shira, a mesotrophic saline lake are presented, in order to assess the heterotrophic potential of the phytoplankton community from a dim environment.

Material and methods Study area The experiments were conducted in Lake Shira, a Siberian salt lake situated in Khakassia (54◦30 N; 90◦ 10 E). The physico-chemical and biological features of this lake have been recently described (Zotina et al. 1999; Kalacheva et al., 2002). For the uptake experiments of organic compounds the lake water was sampled from 12 m, at the deepest area of the lake (Zmax = 26 m), on 17 August 2000. Chemical and light determinations Ammonium-N was determined using the Nessler reagent in presence of potassium sodium tartrate, immediately after collecting the sample. Dissolved inorganic carbon (DIC) was measured by titration with NaOH using phenolphthalein as the indicator. The other inorganic components were analysed according to Standard Methods (APHA, 1989). DOC concentration was measured in samples pre-filtered through hollow fibre capsules (pore size, 0.2 µm) by gravity (pressure of a 40 cm water column). The filtrate was acidified with 1% (final concentration) of 2 M HCl, and stored in Teflon bottles (pH 2–3). The samples

221 were analysed within one week after collection by total organic carbon (TOC) analyser (Shimadzu, TOC 5000), equipped with a high sensitivity platinum catalyst and a halogen scrubber. The sample was diluted 10 times with DOC-free water and sparkled for 15 min with high purity dioxygen. Light was measured in relative units with a submersible photometer (silicon phototransistor) equipped with a frosted hemisphere. The light attenuation coefficient was calculated and utilised for light estimations during the day. The photometer was intercalibrated with actinometric observations.

Glycine (labelled in all the carbon, 99% atom enrichment) and glucose (labelled in all the carbons, 98% atom enrichment) from Cambridge Isotope Lab. were added to give a 100 nM final concentration. The flasks were placed horizontally with the maximum surface exposed to the light and deployed for 1 h in triplicate at the 12 m depth in the fractionation experiment, and at the two metre or twelve metre depth in the whole community experiment, immediately after adding the tracer. The incubation was terminated by gentle filtration through pre-combusted, at 475 ◦ C for 1 h, glass fibre filters (GF/F, Whatman), collecting the labelled particles onto the filters.

Determination of chlorophyll a For chlorophyll a (Chl a) determination, lake water samples were filtered under gentle vacuum through Vladipor (Russia) membranes (pore size, 0.85– 0.95 µm) covered with a thin layer (1–2 mm) of MgCO3 powder. Chl a was extracted with 90% ethanol according to Nusch (1980) and determined spectrophotometrically. Fractionation experiments Fractionation experiments were conducted to identify the heterotrophic capabilities of different groups of planktonic microorganisms. The plankton community of the 12 m layer was separated into two different size fractions by filtration through cellulose acetate membranes (8 µm pore size, Schleicher & Schuel, Dassel, Germany) to obtain bacteria and phytoplankton-enriched fractions. The lake water was sampled using a light-protected sampler from 12 m depth, and filtration through the membrane filter was conducted immediately on board, applying gentle vacuum by a hand vacuum pump. The filtrate was used as the bacterial sample. Bacteria-free water was obtained by passage of lake water through a MediaKap-5 hollow fibre cartridge (0.2 µm pore size, Microgon, Inc. Laguna Hills, California, USA). The phytoplanktonenriched sample was obtained by resuspending the particles retained on the 8 µm membrane filter with the same volume of bacteria-free water. Both samples were stored (< 2 h) in darkened bottles until incubation was started. Incubation with 13 C-labelled compounds The lake water samples were incubated in full 277.5 ml transparent plastic culture flasks in the presence of uniformly 13 C-labelled organic compounds.

Determination of isotope ratios The GF/F filters containing the particulate fractions were wrapped in aluminium foil and stored frozen (−20 ◦ C) until the analysis. The isotope analyses were performed by EA-CF-IRMS (elemental analysercontinuous flow-isotopic ratio mass spectrometer). The filters were encapsulated in zinc capsules and combusted in an elemental analyser Carlo Erba 1108 CHNS, coupled in continuous flux to a mass spectrometer (Micromass Isochrom). Data are expressed as 13 δC values. Preservation of samples Samples for microscopic examination of the fractionated water were fixed with glutaraldehyde (50%) to a final concentration of 2% (v/v) Epifluorescence analysis Bacterial numbers were estimated using the DAPI (4 ,6-diamidino-2-phenylindol-dihydrochloride) staining method. The method consisted of diluting 1 ml of glutaraldehyde-fixed lake water sample to 10 ml with 0.2 µm-filtered water. One ml of a DAPI solution (10 µg ml−1 ) was added to each diluted sample. After an incubation of 20–30 min, the suspension was filtered through Nuclepore membranes (0.2 µm pore size) and examined under an epifluorescence microscope (Leitz DMRXE, Wetzlar, Germany) under x1000 magnification. DAPI stained bacteria were detected by UVA excitation (Filter A, excitation 340– 380 nm and cut filter LP 425). Cyanobacterial autofluorescence was obtained after excitation with green light (Filter Y3, excitation BP 535/550 nm and a cut filter LP 610/675), and identification was made

222 Table 1. Chlorophyll a content in the fractionation experiments. A plankton sample from 12 m depth was separated by filtration through a 8 µm membrane into a > 8 µm and < 8 µm fraction. Values are given in µg l−1 for the glucose and glycine uptake experiments

Figure 1. Temperature (triangles and dashed line) and Chl a concentration (solid circles and solid line) profiles in Lake Shira during the experimental period (August 2000).

following Komareck and Agnanostidis, 1999. Fluorescent particles under UVA excitation, which were not fluorescent under the green excitation, were counted as bacterioplankton. High numbers of bacterial-sized microorganisms did not show autofluorescence under the microscope, but emitted fluorescence at 840 nm (typical wavelength for bacteriochlorophyll a) though they were not DAPI stained. These particles were regarded as photosynthetic bacteria, although further investigation needs to be carried out.

Results During the study period Lake Shira was stratified and the metalimnion was located between 7 and 11 m depth. The thermal stratification was pronounced with an epilimnetic temperature above 20 ◦ C and hypolimnetic temperature below 2 ◦ C. Most of the photosynthetic organisms were stratified below the metalimnion, at about 12 m depth, with Chl a values as high as 25 µg Chl a l−1 (Figure 1). All the isotope

Size

µg Chl a l−1 Glucose Glycine

> 8 µm < 8 µm

16.9 1.2

16.1 1.2

incorporation experiments were carried out using the plankton community at the 12 m depth. Although the light sensor used was not sensitive enough to determine the light at this depth, an extrapolation from the closest reliable data (10.5 m) indicated that light at the 12 m depth might be ca. 0.6–0.7% of the surface light. The water at 12 m was hypoxic, with oxygen concentrations of about 1 mg O2 l−1 , which represents less than 10% oxygen saturation. The dissolved reactive phosphorus (DRP) concentration was 11 µg P l−1 and dissolved inorganic nitrogen (DIN) was dominated by ammonium, which was higher than 400 µg N−1 NH+ 4 l . The DIC concentration was extremely high (1900 mg C l−1 ), dominated by HCO− 3 because of the alkaline conditions (pH = 8.5). The H2 S concentration was not measured during the experiments. Previous (unpublished) data indicate that the H2 S concentration was below our detection limit at 12 m, as expected, although it was detectable at 13 m and at 14 m it was > 5 mg l−1 . Therefore, the community from 12 m depth resides in a boundary layer in terms of H2 S concentration. DOC concentrations were quite high in the whole water column, ranging from > 15 mg C l−1 at the surface, to > 20 mg C l−1 at the bottom of the metalimnion. The fractionation technique applied in this work is very efficient. Our data indicate that most of the Chl a was found in the > 8 µm fraction, with only 7–8% of total chlorophyll a in the < 8 µm fraction (Table 1). Microscopic observation of the samples showed that the < 8 µm fraction contained mostly bacteria, both heterotrophic and phototrophic. Picocyanobacteria and small eukaryotes could not be detected in this fraction by the technique used, although they were likely to be present according to the Chl a results. The photosynthetic bacteria had cell sizes of 0.5 × 2.5 µm but further identification was not pos-

223 sible. The > 8 µm fraction contained mainly colonyforming cyanobacteria of the genus Aphanocapsa and coiled filaments of Lyngbya contorta (Table 2). In this fraction, cyanobacterial cells represented 92% of the total cell count. Picocyanobacteria were also virtually absent from this fraction. All cyanobacteria exhibited the characteristic orange emission under green excitation, indicating the presence of phycoerythrin-like pigments. Chlorophyta and diatoms were too scarce to allow representative counts. The concentrations of heterotrophic bacteria (blue fluorescent particles after DAPI staining) attached to the phytoplankton particles were very low, at the most 2–3% of the total count obtained in the < 8 µm fraction. The < 8 µm fraction, consisting of heterotrophic and photosynthetic bacteria, was responsible for incorporating 4 times more glucose and glycine than the > 8 µm fraction. However, the larger fraction, dominated by cyanobacteria, also exhibited an important uptake of these organic substrates (Table 3). The natural abundance of 13 C in the plankton community was −26.6δ units, while the cyanobacterial (> 8 µm) and bacterial (< 8 µm) fraction supplemented with 13 C labelled glucose, showed average values of +138δ units, and +557δ units, respectively. For glycine the uptake was slightly lower, with δ + 80 and +323 respectively. If the δ 13 C values are expressed as total number of cells, the uptake values are only 1.6-fold higher in the < 8 µm fraction than in the larger fraction. Nevertheless, δ 13 C values referred to heterotrophic bacteria numbers were ca. 20 times larger in the > 8 µm fraction, indicating that bacteria are not the only organisms responsible for organic compounds uptake. When δ 13 C values are expressed as a function of biovolume, they are about 20% higher in the < 8 µm than in the > 8 µm fraction, indicating that the larger plankton is less efficient in the uptake of organic compounds than the bacterial fraction. The uptake of organic compounds under different irradiances is shown in Figure 2. To obtain the lowest light intensity, the samples were incubated at 12 m depth. The other three light intensities were obtained by incubating the population during the afternoon at 2 m depth, making use of the natural decline in the solar angle and thus in the irradiance, before sunset. The glucose uptakes seem to be a light related process, and evidence of a clear inverse relationship with irradiance. The δ 13 C value was almost three-fold higher under moderate irradiance (112 µE m−2 s−1 ) than under maximum irradiance conditions (890 µE m−2 s−1 ). Glycine uptake also showed a negative relationship

with irradiance, but the trend was less significant, with a reduction of 35% in the δ 13 C value from low to high light conditions. At very low light, unlike glucose which was much less incorporated than under moderate irradiance, glycine exhibited nearly the same uptake under both light conditions.

Discussion The phytoplankton community in Lake Shira was dominated by cyanobacteria, which were stratified at the lower boundary of the metalimnion. The habitat where cyanobacteria were found was characterised by low temperature (about 2 ◦ C), and low light conditions (estimated irradiance < 1% of the surface irradiance). The cyanobacterial groups identified were characteristic for this lake and were dominated by Lyngbya contorta and by several Aphanothece morphotypes. Unidentified colonies of coccoid cyanobacteria dominated the layer in terms of cell numbers. As a characteristic feature, without exception, the coccoid cyanobacteria were organized in colonies. In addition, the photosynthetic bacteria represented an important fraction of the planktonic community. The colonial growth habit of the cyanobacteria allowed us to separate them efficiently from bacterioplankton by filtration through 8 µm membrane filters. The metalimnion associated cyanobacterial community accounted for more than 60% of the total Chl a (on areal basis) in the water column, and was 3-fold larger than the total epilimnetic photoautotrophic community (on areal basis). The depth at which the major cyanobacterial population occurred was characterised by low light conditions, particularly rich in the DOC and nitrogen, as ammonium. Under these circumstances light limitation is very likely, and the uptake of DOC by the photosynthetic organisms might be one of the processes responsible for the growth of the community, as suggested by Lewitus & Kana (1994). Several algal and cyanobacterial species from different groups, which are known as photosynthetic organisms, are also capable of growing on organic compounds in laboratory experiments. Among them, the chlorophytes (e.g. Chlorella vulgaris) (Martínez & Orús, 1991) and cyanobacteria (Smith & More, 1988) are the most investigated. Laboratory data indicate that among cyanobacteria the capability of taking up organic compounds is widespread. Nostoc (Tredici et al., 1990), Synechocystis (Pierce et al., 1989), Syne-

224 Table 2. Cell counts in the 12 m lake water sample and in the > 8 and < 8 µm fractions. Cell counts expressed in 106 cells ml−1 106 cells ml−1 Total community > 8 µm

Organisms

< 8 µm

Heterotrophic bacterioplankton Photosynthetic bacteria Coccoid cyanobacteria (colonies) Aphanothece Lyngbya contorta

5 5 1.4 0.327 1

0.042 0.198 1.2 0.141 1.4

3.4 4.1 0 0 0

Total

12.727

2.981

7.5

Table 3. Glucose and glycine uptake experiment in size-fractionated plankton community. Biovolume expressed as 106 µm3 ml−1 . Standard deviation given in brackets Parameter

Glucose > 8 µm < 8 µm

Glycine > 8 µm < 8 µm

δ 13 C δ 13 C/number cells × 106 δ 13 C/number of bacteria × 106 δ 13 C/biovolume

138 (22) 46 3286 86

80 (33) 27 1905 50

chococcus (Willey & Waterbury, 1989), Anabaena (Fernández-Valiente et al., 1992), Calothrix (Prosperi et al., 1992), Plectonema (Raboy & Padan, 1978), Oscillatoria (Hashem et al., 1989) and Spirulina (Vonshak et al., 2000) show this activity. Primarily, these organisms make use of glucose and fructose in addition to several other organic substances (Smith, 1982). However, this type of metabolism was observed in laboratory experiments and it is not clear to what extent these organisms use it in the field, under natural environmental conditions. Furthermore, data from saline lakes are virtually non-existent. Saunders (1972) and Ellis & Stanford (1982) demonstrated by autoradiographic techniques that some cyanobacteria strains can take up sugars at natural concentrations. Mixotrophic capabilities may represent an adaptive advantage to those niches, since the simultaneous uptake of both organic and inorganic C may enable an significant increase in the growth rates, as has been demonstrated in laboratory experiments. Under mixotrophic conditions Anabaena variabilis exhibited a 3-fold higher growth rate than under photoautotrophic conditions (Fernández-Valiente et al., 1992). As expected, our results indicated that most of the uptake of dissolved organic compounds was carried out by the < 8 µm bacterial fraction, but the

557 (106) 74 163 105

323 (113) 43 95 61

uptake of both glucose and glycine was also important in the coarser fraction, dominated by cyanobacteria. The bacteria attached to the colonial organisms (> 8 µm) were too scarce (about 1% of the planktonic bacterial community) to be responsible for such an important uptake. In fact, Ellis & Stanford (1982) found that attached bacteria showed very low uptake of free glucose. The δ 13 C data, expressed as total number of cells and by number of heterotrophic bacteria, indicated that the cyanobacterial population takes up an important fraction of the glucose and glycine. The lower uptake in the > 8 µm fraction and the lower values of uptake expressed by biovolume, were expected, since the uptake rates of organic compounds in heterotrophic bacteria are typically higher (between 5 and 10 times) than those determined in photosynthetic organisms. Considering that all the heterotrophic bacteria have the same uptake rate we would obtain an increase of 0.000164 δ 13 C per bacterial cell. This uptake would represent an increase of 6.9 δ 13 C units in the > 8 µm fraction, which is about 5% of that observed. Subtracting the theoretical bacterial uptake values, the cyanobacterial cells would show a δ 13 C increase of 0.00048 units, which is 3.4 fold lower than the bacterial uptake, in agreement with the previously discussed results. The high DOC concentration

225

Figure 2. Light dependency of organic compounds uptake by the whole planktonic community. A) glucose uptake and B) glycine uptake. The low irradiance (3 µmol photon m−2 s−1 ) data were obtained by incubating at 12 m depth, the other irradiances (112, 660 and 890 µmol photon m−2 s−1 ) were obtained by incubating at 2 m depth.

found in Lake Shira would support the mixotrophic metabolism even with lower uptake rates than heterotrophic bacteria. The mixotrophic capabilities of the photosynthetic bacteria found in this community are unknown, although their high abundance may represent an important variation in previous calculations. The negative relationship between the uptake of organic compounds and irradiance may indicate that the photosynthetic organisms are involved in the uptake process. However, the uptake rate measured at the lowest irradiance is not directly comparable with that at higher light intensities because the former was incubated at 12 m depth and the latter at 2 m depth.

The water temperature in these strata differed greatly, > 20 ◦ C at 2 m and < 2 ◦ C at 12 m depth. The effect of the temperature difference between the community habitat and incubation depth could also be responsible for the variation. The differences in the data obtained could even be due to a physiological shift along the sequential incubations. Nevertheless, the trend of these field data agree with the results obtained in laboratory experiments with cyanobacterial strains, which show an inverse relationship between the irradiance and the sugar uptake rates (Carr, 1973). Other cyanobacterial strains, for example Anabaena variabilis, showed mixotrophy even under saturating light

226 conditions (Haury & Spiller, 1981). Ellis & Stanford (1982) found a direct light dependency of the glucose uptake by a natural community, although in this case the uptake was due mostly to bacteria, suggesting relationships between bacterial uptake and phytoplankton photosynthetic activity. The observed decrease in the uptake of organic compounds at maximum light conditions might be a result of the reduction of the uptake by the phytoplankton, because of the lower energetic demand of the phytoplankton cells under full light conditions. Considering that the bacterial uptake may be constant at any light intensity, the reduction in the uptake by the phytoplankton would result in the dilution of the 13 C label in the whole community. Our study contributes to the understanding of the ecological role and the mixotrophic capabilities of the phytoplankton in saline, aquatic environments. Although realistic sugar concentrations were used, more data are needed to demonstrate that organic compounds are taken up and metabolised by phytoplankton in the field at natural concentrations. In particular, future studies will require a budget approach to assess the relative magnitude of heterotrophic versus phototrophic carbon uptake in this population.

Acknowledgements These experiments were carried out with the help of all the participants of the INTAS project number 970519. This work was financed by INTAS (project number 97-0519). Mr. Ramón Redondo from SIDI (UAM) performed the stable isotopes analyses. We also especially thank Dr. Fernández-Valiente and two anonymous reviewers whose comments and suggestions improved the manuscript.

References APHA (1989) Standard Methods for the Examination of Water and Wastewater. 17th edition, Washington. Bright DI and Walsby AE (2000) The daily integral of growth by Planktothrix rubescens calculated from growth rate in culture and irradiance in Lake Zürich. New Phytol 146: 301–316 Carr NG (1973) Metabolic control and autotrophic physiology. In: Carr NG and Whitton BA (eds), The Biology of Blue-Green Algae. Blackwell Sci. Publ., Oxford, pp. 39–65 Eguchi M, Oketa T, Miyamoto N, Maeda H and Kawai A (1996) Occurrence of viable photoautotrophic picoplankton in the aphotic zone of Lake Biwa, Japan. J Plankton Res 18: 539–550 Ellis BK and Stanford JA (1982) Comparative photoheterotrophy, chemoheterotrophy, and photolithotrophy in a eutrophic reservoir and an oligotrophic lake. Limnol Oceanogr 27: 440–454

Fernández-Valiente E, Nieva M, Avendaño MC and Sánchez-Maeso E (1992) Uptake and utilization of fructose by Anabaena variabilis ATCC 29413. Effect on respiration and photosynthesis. Plant Cell Physiol 33: 307–313 Hashem MA, Gallon JR and Chaplin AE (1989) Nitrogen fixation by Oscillatoria sp. UCSB88 under alternating light and darkness. In: Rogers LJ and Gallon JR (eds), Biochemistry of the Algae and Cyanobacteria, Vol. 28. Oxford Science Publications, Oxford, p. 181. Haury JF and Spiller H (1981) Fructose uptake and influence on growth and on nitrogen fixation by Anabaena variabilis. J Bacteriol 147: 227–235 Kalacheva GS, Guvanov VG, Gribovskaya IV, Gladchenko IA, Zinenko GK and Savitsky SV (2002) Chemical analysis of Lake Shira water (1997–2000). Aquatic Ecol 36: 123-141 Komarek J and Anagnostidis K (1999) Cyanoprokaryota. 1. Teil: Chroococcales. G. Fisher, Jena Lewitus AJ and Kana TM (1994) Responses of estuarine phytoplankton to exogenous glucose: Stimulation versus inhibition of photosynthesis and respiration. Limnol Oceanogr 39: 182–189 Martínez F and Orús MI (1991) Interactions between glucose and inorganic carbon metabolism in Chlorella vulgaris strain UAM101. Plant Physiol 95: 1150–1155 McKnight DM, Howes BL, Taylor CD and Goehringer DD (2000) Phytoplankton dynamics in a stably stratified Antarctic lake during winter darkness. J Phycol 36: 852–861 Nieva MM (1994) Caracterización fisiológica del crecimiento en condiciones mixotróficas de la cianobacteria Anabaena variabilis ATCC 29413. Ediciones de la Universidad Autónoma de Madrid, Madrid Nusch EA (1980) Comparison of different methods for chlorophyll and phaeopigment determination. Arch Hydrobiol Beih Ergebn Limnol 14: 14–36 Peschek GA (1987) Respiratory electron transport. In: Fay P and Van Baalen C (eds), The Cyanobacteria. Elsevier Science Publishers, Amsterdam, pp. 119–161 Pierce J, Carlson TJ and Williams JGK (1989) A cyanobacterial mutant requiring the expression of ribulose bisphosphate carboxylase from a photosynthetic anaerobe. Proc Natl Acad Sci USA 86: 5753–5757 Prosperi C, Boluda L, Luna C and Fernández-Valiente E (1992) Environmental factors affecting in vitro nitrogenase activity of cyanobacteria isolated from rice fields. J Appl Phycol 4: 197–204 Raboy B and Padan E (1978) Active transport of glucose and αmethylglucoside in the cyanobacterium Plectonema boryanum. J Biol Chem 253: 3287–3291 Saunders GW (1972) Potential heterotrophy in a natural population of Oscillatoria agardhii var. isothrix Skuja. Limnol Oceanogr 17: 704–711 Smith AJ (1982) Modes of cyanobacterial carbon metabolism. In: Carr NG and Whitton BA (eds), The Biology of Cyanobacteria. Blackwell Scientific Publications, Oxford, pp. 47–85 Smith AJ and More JE (1988) Transport of carbohydrates into cyanobacteria. Methods in Enzymology 167: 551–556 Tandeau de Marsac N and Houmard J (1993) Adaptation of cyanobacteria to environmental stimuli: new steps towards molecular mechanisms. FEMS Microbiol Rev 104: 119–189 Tredici MR, Margheri MC, De Philippis R and Materassi R (1990) The role of hydrogen metabolism in photoheterotrophic cultures of the cyanobacterium Nostoc sp strain cc isolated from Cycas circinalis L . J Gen Microbiol 136: 1009–1015 Vincent WF and Goldman C (1980) Evidence for algal heterotrophy in Lake Tahoe, California-Nevada. Limnol Oceanogr 25: 89–99

227 Vonshak A, Cheung SM and Chen F (2000) Mixotrophic growth modifies the response of Spirulina (Arthrospira) platensis (Cyanobacteria) cells to light. J Phycol 36: 675–679 Willey JM and Waterbury JB (1989) Chemotaxis toward nitrogenous compounds by swimming strains of marine Synechococcus spp. Appl Environ Microbiol 55: 1888–1894

Zotina T, Tolomeyev AP and Degermendzhy NN (1999) Lake Shira, a Siberian salt lake: ecosystem structure and function.1. Major physico-chemical and biological features. Int J Salt Lakes Res 8: 211–232