Materials and methods

Feed composition

Different ecological water-bodies of Netherlands ponds were inoculated in chemostat feeding medium with composition as follows (concentration in mg.l-1): MgSO4.7H2O (394.4); KCl (18.2); K2HPO4 (156.8); CaCl2.2H2O (22.0); FeCl3.6H2O (38.0); NaEDTA.2H2O (100.5); H3BO3 (46.4); NaSiO3.9H2O (85.3) and 1 ml/L trace element solution. Except for the nitrogen source, all other nutrients concentrations were adjusted in a way to ensure the nitrogen limitation was the only possible nutrient limitation occurred in all experiments. For different NLRs, the concentration of NaNO3 is shown in table 1.
The final trace elements composition in medium for all different nitrogen feeding regimes was as follows (concentration on µmol/L): ZnSO4.7H2 O (1.7); MnCl2.4H2O (19.3); CuSO4.5H2O (0.1); CoCl2.6H2 O (1.1); NaMoO4.2H2O (2.0); Na2SeO4 (0.3); Na3VO4 (0.02).
Table 1
A non-sterile 2 L bioreactor (Applikon, Schiedam, the Netherlands) was used in a continuous mode for three months with average light intensity at photobioreactor’s inner surface 375 µmol.m-2.s-1 provided by an LED lamp. The working volume of 1.5 L was used for all nitrogen loading rates. The pH of the culture was maintained at 7.5 using 1 M HCl and 1 M NaOH. The medium was aerated with enriched CO2 (5%, v/v in N2) at a flow rate of 46 ml.min-1controlled by a mass flow controller (Brooks Instruments, Ede, the Netherlands). The outflow gas stream was cooled to 4 °C and sent to the Rosemount NGA off-gas analyzer (Emerson, USA). The speed of stirrer and temperature were maintained at 350 rpm and 26 °C, respectively. Masterflex pump (Cole-Parmer, Vernon Hills, IL, USA) with two head was used at inflow and outflow rate of 0.52 ml.min-1 to provide the constant dilution rate of 0.5 d-1 and HRT of 2 d. A Biocontroller ADI 1030 (Applikon, Schiedam, The Netherlands) continuously measured pH and DOT (Dissolved Oxygen Tension). All data were stored by PC with MFCS_win software (Sartorius Stedim Systems, Goettingen, Germany). A schematic overview of the experimental setup is shown in Figure 1.
Figure 1

Photobioreactor operation

First, the fresh medium of F1 was inoculated as described in the Materials and methods section. The initial OD was 0.05 and the photobioreactor was run in batch mode for two days to reach the OD of 1.8. Then the fresh medium with the composition of F1 was supplied to the reactor continuously to reach the steady-state point. Off-gas compositions and the biomass concentration were used for monitoring the steady-state condition, in which the variation of these parameters should be less than 10% for three hydraulic retention times. The different nitrogen loading rates were applied by supplying different nitrate concentrations in the fresh medium, F1, F2, F3, F4, and F5.

Analytical methods

Samples were collected from the photobioreactor every day. All experiments were conducted in triplicate. Data were analyzed by an analysis of variance (P<0.05) and results were processed by a computer program: Excel software.

Nitrate measurement

The content of nitrate in the reactor was measured spectrophotometrically using Dr. Lange LCK 339 Nitrate cuvette tests (Hach Lange, Dusseldorf, Germany) .

Biomass measurements

The effluent from the reactor was collected in an ice bath for measuring biomass dry weight (DW). 300 ml of microalgae suspension was centrifuged, the supernatant was removed and the pellet was dried overnight in pre-weighed aluminum foil container at 104 °C. The optical density was measured at the wavelength of 680 nm every day. Biomass productivity was calculated by the following equation:
Biomass Productivity (mg.l-1.d-1) = DW (mg.l-1) ×D (d-1) (1)
where DW and D are dry weight and dilution rate, respectively.

Lipids and starch analysis

The content of Lipids was measured following the PHB extraction protocol described by Johnson et al. with minor modification. Formaldehyde was not added to the samples and Myristic, palmitic, palmitoleic, stearic, oleic, linoleic and linolenic acid were applied as standards. Into digestion tubes, 50 µl of internal standard benzoic acid in 1-propanol (1 g in 50 mL), 1.5mL of 1-propanol: hydrochloric acid (4:1 v/v) and 1.5 ml of 1,2-dichloroethane were added to weighed freeze-dried biomass. The tubes were placed in a digester block at 100 °C for 2 h. Gas chromatography (model 6890N, Agilent, USA) equipped with a flame ionization detector on an HP Innowax column was used for extracted lipids analysis. For starch measurement, approximately 4-mg freeze-dried biomass heated with 0.6 M HCl for 3 h at 100 °C. After centrifugation and filtration with a 0.45-μm pore size filter (PVDF Membrane, Millipore, Ireland), the poly glucose concentration was measured by high-performance liquid chromatography with the specification of an Aminex HPX-87H column from Bio-Rad (USA) (t=60 °C) coupled to an ultraviolet and a refractive index detector . The productivity of starch and lipids are calculated according to the following equations:
Starch Productivity (mg.l-1.d-1) = Starch Content(mg.l-1)×D(d-1) (2)
Lipid Productivity (mg.l-1.d-1) = Lipid Content (mg.l-1)×D(d-1) (3)

Staining of microalgal cells

4 μl BODIPY 505/515 (4,4-difluoro-1,3,5,7-tetramethyl-4- bora-3a,4a-diazasindacene) in anhydrous dimethyl sulfoxide (DMSO; 1 mg.ml-1) was added to 0.2 mL of algal culture for visualizing the lipids droplets in the cells by Leica DM500B light microscope (Leica Microsystems, Wetzlar, Germany) equipped with fluorescence filtercube A. The excitation and emission wavelengths for monitoring BODIPY 505/515 fluorescence were 488 and 515 nm, respectively as described by Klock et al. (2016). Starch staining was conducted by adding 4µl of Lugol’s solution to 0.2 ml of microalgal suspension . Different species present in the five NLR cultures were categorized based on the morphological microscopic cellular appearance by taking pictures with an above-mentioned microscope using 100 times magnification. Identification was based on the morphology of the individual cells following microscopic examination. The objects which were not morphologically consistent with algae, such as debris, bacteria, and particles were discarded.

PCR-DGGE Analysis

The microbial composition of chemostat enriched culture at different nitrogen loading rates was analyzed using the PCR-DGGE technique as described by Danesh et al. .