The cyanobacterium Nostoc minutum [13] was used in this study. Stock cyanobacterial cultures were maintained in the BW₃ medium with the following composition (g/L): K₂HPO₄ 0.25; NaHCO₃ 0.5; MgSO₄ 0.7 H₂O 0.375; CaCl₂ 0.4; NaCl 4; NaNO₃ 0.5; FeCl₃∙6H₂O 0.024; EDTA.2H₂O 0.0625; micronutrients 6.25 ml [14] and BG11 medium with the following composition (g/L): NaNO₃ 1.5; K₂HPO₄∙3H₂O 0.04; MgSO₄∙7H₂O 0.075; CaCl₂∙2H₂O 0.036; C₆H₈O₇ 0.06; C₆H_5+4yFe_xN_yO₇ 0.06; Na₂EDTA 0.0005; Na₂CO₃∙(H₂O) 0.010; micronutrients 1 ml [15] . All strains were unicyanobacterial and non-axenic, but the number of observed associated bacteria was very low. All reagents used were of analytical grade, obtained from Merck.

All culturing procedures were performed aseptically. Stockcultures, 10 mL (0.15 g/L), were used to inoculate either erlenmeyers flasks (still cultures) or glass columns (stirred cultures). Cultivation was carried out in sterilised photobiorectors consisting of 250 mL Erlenmeyer flasks equipped with a device for aseptic removal of samples (still cultures) or 250 ml glass columns 37 mm i.d., containing 200 ml of BW3 or BG11 medium with or without the addition of NaNO₃ and mixed through air injection with an aeration flow of 6.13 mL/seg (stirred cultures). The aeration was performed with filter-sterilized air. The cultures were run for 2 weeks (14 days) at 30˚C under permanent lighting of 4530 (low intensity), 7300 luxes (intermediate intensity) and 9860 luxes (high intensity).

Cyanobacterial growth was estimated by optical density (OD) at 580 nm every 48 h. EPS production was determinated after cultures reached the stationary phase, qualitatively by Alcian blue tintion and India ink, and quantitatively by dry weight determinations. The experiment was conducted in triplicates and values were expressed as their mean.

EPS was quantified by the method of Mondal et al. modified [16] , culture broth were decanted into 50 ml centrifuge tubes, vortexed for 5 min and centrifuged at 10,000 rpm for 30 min at 4˚C to remove cells. The supernatants were collected and 2.0 volume of acetone 80% was added and kept overnight at 4˚C for precipitation. The pellets, collected by centrifugation at 12,000 rpm for 10 min, were dissolved in deionized distilled water and dialyzed overnight at 4˚C against deionized distilled water. Dialyzed materials were lyophilized and weighed (AdventurerTM OHAUS Corp. USA). Experiments were performed three times to ensure reproducibility.

For physical characterization of the EPS, it was determined the electric charge by precipitation with cetylpe- ridium chloride (CPC) [17] . The dried EPS (5 mg) was dissolved in 5 ml of 0.05 M NaCl; thereafter, 10% CPC solution was added into it until no more precipitate of EPS-CPC complex was formed. The rheological behaviour was studied using a Brookfield programmable rheometer at 25˚C following a standard method with some modification [18] . The EPS solution (2.0%) was prepared in deionized double distilled water. The pH of the EPS solution was adjusted to 7.0 using 1N HCl and 1M NaOH.

An increase in biomass production and specific growth rate with the increment of light intensity were observed. However, for still cultures the light intensity showed a slight incidence. Different authors reported the impor- tance to consider this parameter because it is often found as a limiting factor in culture systems [19] [20] . The efficient air-lift stirring of the glass columns plays a crucial rol for reducing shading between cells, thus allowing to obtain higher values of biomass than in still cultures. BW3 medium was better than BG11 for N. minutum growth, with maximal values of biomass concentration (4.98 DO) and the highest specific growth rate (0.019 h⁻¹) at 9860 lux of light intensity. The BW₃ medium have 50 times NaHCO₃ concentration and 5 times MgSO₄∙7H₂O and K₂HPO₄ than BG11, composition that was optimal for N. minutum growth. The presence of combined nitrogen also had a positive effect on growth and specific growth rate (µ) under all the conditions assayed. The lowest values of biomass and specific growth rate were obtained for diazotrophic conditions in BG11⁰ medium. In fact, biomass production was more affected by the availability of combined nitrogen than by light intensity (Table 1).

N. minutum cells are enclosed in a fibrous matrix at the cell wall surface that has a structural coherence suffi- cient to exclude particles (e.g. India ink, see Figure 1(A)). The observation of the preparations stained with Al- cian Blue and India ink under the light microscope, showed an EPS production according to that determined by dry weight technique. For non diazotrophic cultures, it was observed a gradual increase in EPS production with the days that was coinciding with a remarkable increase in the culture viscosity. The viscosity of the cultures was maximal at 14 days in BG11 medium (Figure 1(B) and Figure 1(C)). Besides, the Alcian Blue stain is

BG11⁰, BW3⁰: culture media without NaNO₃ added. Mean values ±SD are shown. Máximum values are in bold letter. nd = not determined.

specific to acid mucopolysaccharide [21] . The EPS produced by N. minutum was stainable with Alcian Blue (Figure 1(B)), which indicate that it is mainly composed of polysaccharides.

Figure 2 shows EPS production by N. minutum. Both the still and stirred cultures followed a similar pattern, with best yields of EPS for non diazotrophic BG11 and diazotrophic BW₃ cultures. The maximal concentration of EPS (2.87 g/L) was obtained with BG11 medium under stirred conditions and high irradiance (Figure 2(B)).

There is a marked difference between the main components of the two culture media used. While, both BG11⁰ and BW3⁰ have no NaNO₃ added, BG11 have three times NaNO₃ concentration than BW₃. The absence of NaNO₃ could be a stress factor favourable for EPS synthesis. In fact, a number of diazotrophic strains of Nostoc have shown to produce EPS under N2-fixing conditions, with reduce production when grown on an exogenous N source [22] . However, the EPS production by N. minutum was maximal when the NaNO₃ concentration was tripled. Similar results were informed by Nicolaus et al. for Anabaena WSAF with 56 mg/L in BG11 and 22 mg/L in BG11⁰ media [23] . The yields of EPS were in agreement with other values reported for cyanobacteria by Otero and Vincenzini [22] with 3.5 g/L of total carbohidrates and 1.8 g/L for soluble ones. On the other hand, the EPS production by N. minutum growing in BW₃ medium was maximal in diazotrophic conditions, both for still (1.66 g/L) and stirred (2.56 g/L) cultures showing that EPS production varies with nutrient availability [24] and environmental conditions [25] . The progressive increase in culture viscosity of N. minutum stirred cultures observed when growing non diazotrophically in BG11 medium, may be due to the release of large amounts of EPS. On the contrary, this phenomenon was reported for the cyanobacteria Cyanospira capsulata growing diazotrophically [26] [27] . N. minutum released the EPS while the capsule that surrounds the cells became minimal, in contrast, the unicellular cyanobacterium Cyanothece sp showed the solubilization of the external part while the thickness of the capsule remained almost constant [11] . EPS produced by N. minutum showed precipitation with CPC, indicating the presence of acidic groups. The EPS include different classes of macro- molecules such as proteins, polysaccharides, uronic acids and other compounds not yet fully characterised [28] . The precipitation of EPS with CPC allow to confirm the anionic nature of the EPS due to interaction of qua- ternary ammonium ions of the CPC with the acidic groups to form a polysaccharide-CPC complex [17] , showing interesting properties for metal uptake in bioremediation field [29] . The viscosity of N. minutum EPS showed a decrease with an increase in the shear rate. Such rheological behaviour is the characteristic of a typical pseu- doplastic non-Newtonian fluid [30] and agrees with results reported by Khattar et al. [31] .