Method of culturing algae

Abstract

The present invention relates to a heterotrophic methods of culturing algae, particularly Galdieria species, to produce the valuable pigment phycocyanin. The methods rely on high oxygen saturation and controlled base dosing to provide improved phycocyanin production. The present invention further relates to algal biomass, compositions comprising said biomass or phycocyanin, uses thereof in various products, and a reactor for culturing the algae.

Claims

1. A heterotrophic method of culturing Galdleria sulphuroria algae, comprising: (a) Growing Galdleria sulphuroria algae in a medium comprising a carbon source and a nitrogen source, dissolving oxygen in the medium by aeration, and maintaining the oxygen saturation of the medium above 75%.

2. The method according to claim 1, wherein the carbon source comprises a sugar, carbohydrate or polyol, or any combination thereof.

3. The method according to claim 2, wherein the sugar is selected from a monosaccharide, disaccharide, oligosaccharide, polysaccharide or any combination thereof.

4. The method according to claim 2 wherein the polyol is selected from glycerol, erythritol, inositol, lactitol, mannitol, xorbitol, xylitol, or any combination thereof, and wherein the carbohydrate is selected from: isomalt, cellulose, hemicellulose, pectin, starch, glycogen, chitin, chitosan, guar gum, beta-glucan, alginate, acacia gum, beta mannan, inulin, tara gum, xanthan gum, carrageenan gum, polydextrose, glucomannan, or any combination thereof.

5. The method according to claim 1, wherein the carbon source comprises glucose, fructose, sucrose, or glycerol, or any combination thereof.

6. The method according to claim 1, wherein the medium is acidic.

7. The method according to claim 1, wherein the oxygen saturation of the medium is above 80%, above 85%, above 90%, above 95%.

8. The method according to claim 1, wherein at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95% of the carbon source is glycerol.

9. The method according to claim 8, wherein the glycerol is least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, least 95%, at least 99% pure.

10. The method according to claim 1 wherein the method is continuous.

11. The method according to claim 1 wherein the temperature of the medium is between 35 to 55 C.

12. The method according to claim 1, wherein the method further comprises a step of collecting algal biomass from the culture medium.

13. The method according to claim 1, wherein the method produces between 35 and 45 g.Math.L.sup.1.Math.day.sup.1 of algal biomass.

14. A method of producing phycocyanin comprising culturing Galdleria sulphuroria algae according to the method of claim 1.

15. The method according to claim 12, wherein the algal biomass has a mean intracellular concentration of phycocyanin of between 25-50 mg.Math.g.sup.1 dry cell weigh.

16. The method according to claim 14, wherein the method produces between 1-2.5 g.Math.L.sup.1.Math.day.sup.1 of phycocyanin.

17. Algal biomass produced by the method of claim 1.

18. The algal biomass according to claim 17 having a mean intracellular concentration of phycocyanin of at least 25 mg.Math.g.sup.1 dry cell weight and a mean intracellular concentration of allophycocyanin of less than 1.0 mg.Math.g.sup.1 dry cell weight.

Description

FIGURES

(1) The invention will now be described with reference to the following figures in which:

(2) FIG. 1 shows G. sulphuroria grown in continuous flow conditions at 0.6 D.sup.1 with 20 g.Math.L.sup.1 glycerol as the growth limiting substrate. Batch phase growth was supported by 10 g.Math.L.sup.1 glucose to provide a high starting point of intracellular phycocyanin (circles .circle-solid.) before initiation of continuous flow at day 4. Oxygen saturation of either 50 or 35% were automatically controlled by the reactors computer system (line). The dry cell weight remained stable throughout the culture during continuous flow at around 10 g.Math.L.sup.1 (squares .square-solid.).

(3) FIG. 2 (A) shows growth rate of G. sulphuroria decreases as impeller velocity increases, Impeller velocity (circles , Growth rate (day.sup.1), (squares.square-solid.biomass concentration (g.Math.L.sup.1) and (triangles .box-tangle-solidup.oxygen saturation (%)) and (B) shows reduced growth rate at impeller speeds of above 500 rpm; Growth rates (black circles .circle-solid.) against impeller velocity taken from several stirred tank bioreactor runs at 42 C. wherein glycerol concentration ranges from 5-15 g.Math.L.sup.1. Line represents the regression mean flanked by 95% confidence intervals.

(4) FIG. 3 shows G. sulphuroria cultivated in the 4 L airlift bioreactor in fed-batch with 20 g.Math.L.sup.1 glycerol as initial substrate and 500 g.Math.L.sup.1 glycerol feeding solution initiated at day 3.5. Oxygen (Measured diamonds .diamond-solid.; Trace information from bioreactor line ) was maintained above 85% throughout the study despite rapid growth and high carbon concentrations. A drop in O2 around day 6 was caused by a build-up of biofilm on the dissolved oxygen probe, which returned to >85% after cleaning. Dry cell weight (squares .square-solid.) increased exponentially except for a period of pump blockage at day 6. Intracellular phycocyanin (circles .circle-solid.) reached 42 mg.Math.g.sup.1 following the batch phase and 45 mg.Math.g.sup.1 by the end of cultivation. Glycerol (triangles .box-tangle-solidup.) was maintained close to zero during fed-batch phase.

(5) FIG. 4 shows Continuous cultivation on glycerol in the dark at various temperatures.

(6) FIG. 5A shows the reactor for culturing algae according to the invention in use during a culture.

(7) FIG. 5B shows the reactor design according to the invention, the reactor 100 comprises a container 101, with a lid 103, and two triangular profile baffles 102 disposed within the container 101, at the lower end of the container 101, the total volume is indicated by arrow A, and the working volume is indicated by arrow B, arrow C indicates the non-working volume or headspace of the container 101.

(8) FIG. 6 shows how increased hydrogen ions in culture medium correlate with intracellular phycocyanin concentration, (A) Delta [H*] after 167 hours flask cultivation, with varied starting culture pH. (B) intracellular phycocyanin concentration after 167 hours.

(9) FIG. 7 shows how hydrogen ion release correlates with uptake of NH.sub.4 from growth medium, (A) Growth curves of 20 g.Math.L.sup.1 glucose flask cultures grown at 42 C. in a shaking water bath. Cultures vary by initial pH. Circles .circle-solid. are pH 2.0 control medium, squares .square-solid. is control with glycyl-glycine buffer, triangles .box-tangle-solidup. is pH 2.6, inverted triangles .Math. is pH 3.1, orange diamonds .diamond-solid. is pH 3.6. (B) hydroxyl concentration in the medium as measured by pH probe, reductions infer decreased pH and increased H+ in growth medium. (C) Total concentration of N in the medium as determined by Berthelot reaction.

(10) FIG. 8 shows nitrogen uptake at the algal cell membrane and detection by the automatic pH control system; NH.sub.4.sup.+ is deprotonated at the extracellular side of the AMT transporter, allowing facilitated diffusion of NH3(g) into the cell. The extracellular proton concentration is detected by pH probe, leading to an automated addition of basic ammonium hydroxide/ammonia gas balancing NH4+ concentration in the medium.

(11) FIG. 9 shows ammonium hydroxide based semi continuous cultivation, (A) Cultivation of Galdieria over 6 days with 500 g.Math.L.sup.1 glycerol feeding addition from day 2.3 at 1.0 day.sup.1 intended growth rate in the MK 3 airlift bioreactor. Dry cell weight (g.Math.L.sup.1) .square-solid.; Glycerol concentration (g.Math.L.sup.1) .box-tangle-solidup.; Growth medium oxygen saturation (%) .diamond-solid.; Specific phycocyanin concentration (mg.Math.g.sup.1) .circle-solid., (B) Maintenance of medium nitrogen concentration x through pH .diamond-solid. control.

(12) Nitrogen maintained between 0.5-1.5 g.Math.L.sup.1 throughout the course of the study through the automated addition of 150 mL 10 M NH.sub.4OH .box-tangle-solidup.. The downwards trend of medium nitrogen may be explained by the difference between the feeding solution (pH 2.0) and reactor (pH 1.9).

(13) FIG. 10 shows an automatic pH controlled ammonia gas based system at a 3L scale comprising a reactor and an automatic pH control system having a pH probe, an ammonia gas reservoir, and a solenoid valve.

(14) FIG. 11 shows an automatic pH control system feedback loop; pH is continuously monitored by the system and recorded in 1 second intervals. A setpoint is selected, in this case pH 2.0, as well as lockouts either side as a safety feature. If pH is detected below the setpoint, but above the low lockout point, the valve opens and a 30 second timeout occurs in order to allow the pH to equilibrate. If pH is detected above or below the lockout values, a 60 minute lockout occurs. This is to prevent erroneous addition of NH.sub.3 in the case of valve failure, probe failure, or disconnect between the reactor and NH.sub.3 supply.

(15) FIG. 12 shows semi-continuous cultivation of Galdieria using automated NH.sub.3 gas as a nitrogen source, (A) Cultivation of Galdieria over 7 days with 500 g.Math.L.sup.1 glycerol feeding addition from day 3.5 at 1.0 day.sup.1 intended growth rate in the MK 4 airlift bioreactor. Dry cell weight (g.Math.L.sup.1) .square-solid.; Glycerol concentration (g.Math.L.sup.1) .box-tangle-solidup.; Growth medium oxygen saturation (%)+; Specific phycocyanin concentration (mg.Math.g.sup.1) .circle-solid.. Biofilm build-up on dissolved oxygen probe contributed to drop in measured oxygen saturation around day 6, (B) Maintenance of medium nitrogen concentration * through pH+control. Nitrogen maintained between 0.5-1.5 g.Math.L.sup.1 throughout the course of the study, despite a total of 43 g NH.sub.3 added. Volumetric productivity of phycocyanin (mg.Math.L.sup.1) .square-solid. increases exponentially with cell growth.

(16) FIG. 13 shows initial continuous glycerol culture with ammonium hydroxide automated nitrogen source; Continuous flow cultivation in the MK 2 airlift bioreactor at 0.5 dilutions per day and 75 g.Math.L.sup.1 glycerol preceded by a semi-continuous phase from day 3 to day 6. Nitrogen is added automatically to maintain a pH of 1.7 throughout the study .diamond-solid.. Dry cell weight (g.Math.L.sup.1) .square-solid.; Glycerol concentration (g.Math.L.sup.1) .box-tangle-solidup.; Growth medium oxygen saturation (%) .Math.; nitrogen concentration x (g.Math.L.sup.1).

(17) FIG. 14 shows continuous cultivation with nitrogen maintained by automated pH control in MK 4 airlift bioreactor; Continuous flow cultivation at 1.0 per day dilution rate in the MK 4 airlift bioreactor with 50 g.Math.L.sup.1 glycerol as the growth limiting substrate. Oxygen is maintained at .sup.100% from days 4 to 9, at which point it was reduced to 50% by lowering air flow rate. The reduced mixing and high dilution rate exerted a washout effect and glycerol concentration in the reactor increased dramatically. Dry cell weight (g.Math.L.sup.1) .square-solid.; Glycerol concentration (g.Math.L.sup.1) .box-tangle-solidup.; Growth medium oxygen saturation (%) .Math.; Specific phycocyanin concentration (mg.Math.g.sup.1) .circle-solid.; nitrogen concentration x (g.Math.L.sup.1), pH .diamond-solid..

(18) FIG. 15 shows the lack of allophycocyanin (APC) pigment from purified cell extract of Galdieria (B) grown heterotrophically on glycerol compared to autotrophic Spirulina (A).

(19) FIG. 16 is another figure showing the growth rate of G. sulphuroria decreases as impeller velocity increases. A fed batch cultivation in the stirred tank reactor was performed with 15 g.Math.L.sup.1 glycerol as the starting growth substrate and 300 g.Math.L.sup.1 glycerol feed. Impeller velocity (circles .circle-solid.) was set automatically by the system in order to maintain 100% oxygen saturation with 4.0 L.Math.min.sup.1 air. Natural logarithm of the biomass concentration was used to calculate growth rates in three sample intervals at increasing impeller rates (squares.square-solid. 400-500 rpm, triangles .box-tangle-solidup.500-700 rpm, diamonds .diamond-solid.700-800 rpm).

(20) FIG. 17 shows the same data as FIG. 4 in more detail. Refined glycerol (800 g.Math.L.sup.1 fed batch, 120 g.Math.L.sup.1 continuous feed) was used to cultivation Galdieria at 4.0 L scale in the MK 4 bioreactor at 42 C. Biomass (squares .square-solid.) was grown up to 45 g.Math.L.sup.1 in semi-continuous conditions up to day 8, at which point continuous flow of feed was started at a dilution rate of 0.7 day.sup.1 for a 12 day period. Automatic addition of 10 M NH.sub.4OH maintained the pH at 2.0 and formed the main nitrogen source for the culture. Phycocyanin (circles .circle-solid.) is maintained above 35 mg.Math.g.sup.1 throughout the continuous phase, accordingly with high oxygen saturation (inverted triangles .Math.) in the reactor throughout the same period.

(21) FIGS. 18 and 19 show further images of the reactor for culturing algae according to the invention in use during a culture. FIG. 19 shows the same reactor as FIG. 18 with the faceplate removed demonstrating the Galdieria culture inside.

(22) FIG. 20 shows D-glucose, fructose and sucrose (500 g.Math.L.sup.1 fed batch, 150-200 g.Math.L.sup.1 continuous feed) was used to cultivate Galdieria at 4.0 L scale in the MK 4 bioreactor at 42 C. Growth monitoring was initiated at semi-continuous phase (day 0) with 500 g.Math.L.sup.1 glucose, at a growth rate of 0.6 day.sup.1 Biomass (squares .square-solid.) was grown up to 25 g.Math.L.sup.1 in semi-continuous conditions up to day 6, at which point continuous flow of feed was started at a dilution rate of 0.6 day.sup.1 with 150 g.Math.L.sup.1 D-glucose. Once the culture reached a steady biomass concentration (.sup.65 g.Math.L.sup.1), the feed glucose concentration was increased to 200 g.Math.L.sup.1 on day 11. Stable production was maintained for 5 days at which point the carbon was changed to fructose. Some samples were lost representing the gap in the data. A pump blockage at day 19 prompted a pause in the continuous flow regime, restarted 2 days later with sucrose (200 g.Math.L.sup.1) as the carbon source. Steady state was reached for 5 more days until the experiment was halted. Phycocyanin (circles .circle-solid.) is maintained above 35 mg.Math.g.sup.1 throughout the continuous phase, accordingly with high oxygen saturation (inverted triangles .Math.), except for day 19 when the feed was changed to fructose. Dry cell weight measurements are in triplicate error bars only shown when larger than point (SEM).

(23) Certain aspects and embodiments of the invention will now be demonstrated by the following non-limiting examples.

Examples

Materials and Methods

(24) A. Medium

(25) Heterotrophic medium is based on optimisations from Minoda, with inorganic components from the below concentrations suitable for 10 g.Math.L.sup.1 growth limiting carbon substrate (Minoda et al., 2004).

(26) These elements are scaled proportionally depending on final carbon concentration of medium, with typical batch medium at 20 g.Math.L.sup.1 carbon, semi-continuous from 500-750 g.Math.L.sup.1 carbon, and continuous from 50-300 g.Math.L.sup.1. Flask studies described used glucose as the primary carbon substrate, whilst bioreactor scale studies used glycerol unless otherwise stated.

(27) Final inorganic components and concentrations are as follows for 10 g carbon (g.Math.L.sup.1): (NH.sub.4).sub.2SO.sub.4 2.62, KH.sub.2PO.sub.4 0.54, MgSO.sub.4.Math.7H.sub.2O 0.5, CaCl.sub.2.Math.2H.sub.2O 0.056, FeCl.sub.3 0.028, EDTA.Na.sub.2 0.016; with trace elements to a final concentration of (mg.Math.L.sup.1): H.sub.3BO.sub.3 5.72, MnCl.sub.2.Math.4H.sub.2O 3.64, ZnCl.sub.2 0.21, Na.sub.2MoO.sub.2.Math.2H.sub.2O 0.78, CoCl.sub.2.Math.6H.sub.2O 0.08, CuCl.sub.2 0.086. For plate preparation, double concentrated solutions of medium and agar were sterilised separately by autoclave and combined in a 1:1 ratio after brief cooling.

(28) pH of medium is adjusted to 2.0 with 5 M H.sub.2SO.sub.4 unless otherwise stated. For continuous cultivations utilising glycerol, (NH.sub.4).sub.2SO.sub.4 is replaced with H.sub.2SO.sub.4 in a 1:2 mole ratio, see Table. For feeding solutions of semi-continuous cultivations, (NH.sub.4).sub.2SO.sub.4 is not added, and pH is adjusted to 2.0.

(29) TABLE-US-00002 TABLE 1 Continuous nitrogen-free medium Mr Conc (g .Math. L.sup.1) M Glycerol 92.09 150.00 1.63 Carbon 11.00 53.75 4.89 N for 10:1 C:N ratio 7.00 3.42 0.49 Moles H+ needed 1.00 0.49 0.49 Minus pH 2.0 H+ 0.47 Required conc H.sub.2SO.sub.4 (mL .Math. L.sup.1) 25.47 Equivalent (NH.sub.4).sub.2SO.sub.4 mass 132.14 32.28 0.24

(30) TABLE-US-00003 TABLE 1 Semi-continuous nitrogen-free medium Mr Conc (g .Math. L) M Glycerol 92.09 500.00 5.43 Carbon 11.00 179.17 16.29 N for 10:1 C:N ratio 7.00 11.40 1.63 Moles H+ needed for pH 2.0 1.00 0.01 0.01 Required conc H.sub.2SO.sub.4 (mL .Math. L.sup.1) 0.54 Equivalent (NH.sub.4).sub.2SO.sub.4 mass 132.14 107.61 0.81
B. Cultivation Techniques
Flask

(31) Flask cultures are operated at 100 mL working volume in 250 mL Erlenmeyer flasks, maintained at 42 C. and 200 rpm agitation in the dark using an Incu-Shake MIDI incubating shaker (SciQuip, UK). Offline pH monitoring is performed using a benchtop pH-temperature probe (Eppendorf, UK).

(32) Stirred Tank Bioreactor

(33) Stirred tank bioreactor experiments used a Minifors 2 (Infors, CH) 1.0-2.0 L working volume glass reactor containing a 2 bladed Rushton turbine operated between 400-800 rpm depending on the study.

(34) Airlift Bioreactors

(35) A series of airlift bioreactors with working volumes ranging from 2.5-4.0 L were used in this study. All are made of acrylic sheet 0.6-0.8 mm thick for bespoke construction. Air flow rates used depend on the working volume, set between 0.5-1.5 VVM. Excessive foaming in the reactors was controlled by antifoaming agent (Sigma, UK).

(36) Bioreactor Control

(37) The Minifors control system was used in conjunction with all variant bioreactors, and connected to a computer via the OPC-UA standard to allow time dependent control of parameters. Exponential pump feeding profiles for semi-continuous cultivations were calculated using Microsoft Excel with intended resultant growth rates between 1.0-1.2 day.sup.1.

(38) A bespoke Arduino based pH controller was designed in order to control ammonia gas dosing via a solenoid valve 100T3MP12-32 (Biochemfluidics, UK).

(39) C. Monitoring

(40) Growth

(41) Growth was monitored in a time-dependent basis. Cell suspension absorbance was measured at 800 nm to prevent any pigment interactions (Gross et al., 1998) using an 5-200 spectrophotometer (Boeco, Germany) with a light path of 1 cm and a blank of deionised water. Where necessary, cells were diluted with deionised water to ensure absorbance measurements below 0.7 cm.sup.1.

(42) 1-10 mL of cells (or 5-50 mg dry cells) were taken in triplicate unless otherwise stated for determination of dry cell weight (DCW). Cells were centrifuged at 18,000g for 30 seconds, the supernatant discarded, repeated twice with a resuspension in deionised water to remove medium components. Washed cells were resuspended in 1 mL deionised water and frozen at 80 C. at least overnight for subsequent lyophilisation for 24 hours under vacuum.

(43) Phycocyanin

(44) 10-50 mg of lyophilised cells were resuspended in 1.5 mL of 100 mM potassium phosphate buffer at pH 7.2. Mechanical disruption was performed using cell disruption tubes (2 mL capacity, 0.5 mm Zirconium beads, Sigma-Aldrich, USA) and a BeadBug homogeniser (Benchmark Scientific Inc., USA) at a speed of 4000 rpm in 5-8 60 second cycles, with tubes cooled at 4 C. between cycles.

(45) Cell lysate was centrifuged at 18,000g for 60 minutes at 4 C., and 500 L of blue supernatant was removed and diluted as necessary for spectrophotometric analysis as described below.

(46) Phycocyanin extracts were transferred to a 1 cm light path cuvette to undergo analysis. Absorbance was read by spectrophotometer at 320, 562, 620 and 652 nm, and the C-PC content estimated as described by (Kursar and Alberte, 1983) demonstrated with equation 1 below.

(47) $\begin{array}{cc}\mathrm{Phycocyanin}\left(\mathrm{mg}.{\mathrm{ml}}^{-1}\right)=\frac{A_{620}-0.474\left(A_{652}\right)}{5.34}.& 1\end{array}$
Glycerol

(48) Glycerol concentration of the medium was monitored when glycerol was supplemented into the medium. Glycerol free reagent (Sigma-Aldrich) was used according to the manufacturer's instructions, which is an enzymatic assay leading to the production of a green quinoneimine from glycerol. Medium samples were diluted 100 and final absorbance read at 540 nm.

(49) Nitrogen

(50) Ammonium concentration of the growth medium was determined using a modified version of the Berlethot reaction (Rhine et al., 1998) wherein a solution containing ammonium is made basic, it produces ammonia, which reacts with 2-phenylphenol to produce a blue indophenol complex. Initial samples were diluted 250.sup.1000 and absorbance was read at 660 nm following reaction mixture incubation of 1 hour using a plate reader (Victor.sup.2 1420, Wallac).

Results

(51) Oxygen Saturation and Reactor Efficacy in Production of Phycocyanin from Galdieria sulphuroria

(52) a) Failure to promote high oxygen saturation in stirred tank reactors and reduction in phycocyanin during glycerol growth G. sulphuroria was grown in continuous flow conditions at 0.6 D.sup.1 with 20 g.Math.L.sup.1 glycerol as the growth limiting substrate at 2.0 L working volume in a standard Minifors 2 (Infors, CH) dual-Rushton turbine mixed glass bioreactor. Batch phase growth was supported by 10 g.Math.L.sup.1 glucose to provide a high starting point of intracellular phycocyanin of 32.80.8 mg.Math.g.sup.1 dry cell weight before initiation of continuous flow at day 4 (FIG. 1). At 50% oxygen saturation and glycerol as the carbon source, intracellular phycocyanin reduces to a low of around 10 mg.Math.g.sup.1 dry cell weight over the course of four days of continuous cultivation. The dry cell weight remained stable throughout the culture (black squares .square-solid.). At day 11, the maximum non-damaging settings for the stirred tank reactor of 500 rpm and 2.0 L.Math.min.sup.1 air flow were unable to reach close to 100% oxygen saturation, becoming stable at 70% despite the low glycerol substrate concentration of 20 g.Math.L.sup.1.
b) Reduction in growth rate at impeller speeds above 500 rpm

(53) FIG. 2B and FIG. 16 demonstrate reductions in growth rate of Galdieria sulphuroria under heterotrophic conditions as impeller velocity of a stirred tank bioreactor increases.

(54) 15 g.Math.L.sup.1 glycerol was used as the starting carbon substrate in order to prevent glycerol-mediated growth inhibition ensuring a high initial growth rate of 1.35 day.sup.1. Semi-continuous cultivation was initiated at day 4 preventing glycerol dropping below 5 g.Math.L.sup.1, and did not go above 12 g.Math.L.sup.1 for the remainder of the experiment. Impeller velocity was automatically controlled by the system to maintain >95% oxygen saturation, with a minimum setting of 400 rpm. Above 10 OD.sub.800 (3.25 g.Math.L.sup.1 dry cell weight), the system began increasing impeller velocity, leading to reductions in growth rate to 1.05 day.sup.1 for a 24 hour period between 500-700 rpm, and 0.54 day.sup.1 at 780-700 rpm at 4.5 to 6 days. A peak impeller velocity is observed at 4.5 days at which point the velocity begins to fall. This is likely due to the relationship between the density of cells taking up oxygen and those being damaged by the impeller, leading to an equilibrium in oxygen requirement.

(55) In a stirred tank reactor, shear is exerted into the medium in order to reduce the size of flowing air bubbles and reduce mixing times, improving mass transport in the system. Our observations determined that maintaining high growth rates as well as high oxygen saturation during heterotrophic growth, with cell densities above 10 g.Math.L.sup.1, required impeller velocities damaging to the organism. An alternative strategy was pursued, through the use of an airlift system.

(56) c) Improved Oxygen in airlift reactor and production of phycocyanin in the dark maintained.

(57) Initial attempts to increase oxygen concentration in traditional stirred tank bioreactors failed due to shear on the algal cells causing damage (see section (a) and (b)). It was not possible to maintain high oxygen concentration (by mixing) whilst also maintaining high productivity of biomass and phycocyanin. In order to solve this problem, we developed a low-shear airlift bioreactor design specifically adapted to achieve high oxygen concentrations for high density cultivation of g. sulphuroria.

(58) Cultivation was initiated on 10.5 g.Math.L.sup.1 glycerol at pH 2.0 and 42 C. with 6.0 L.Math.min.sup.1 air flow at 4.0 L working volume (FIGS. 4 and 17) in the airlift bioreactor (FIGS. 5A,B,C and FIGS. 18 and 19). Batch phase growth proceeded at 1.0 day.sup.1 and reached a biomass concentration of 5.6 g.Math.L.sup.1 by day 3. At this point, semi-continuous growth was initiated at a rate of 0.6 day.sup.1 with 800 g.Math.L.sup.1 glycerol feed. Nitrogen addition occurred automatically under pH feedback control with 10 M NH.sub.4OH used to maintain pH 1.9 in the reactor. 500 mL of semi-continuous feed was added until day 7, at a resultant growth rate of 0.54 day.sup.1 to a biomass concentration of 43.4 g.Math.L.sup.1. On day 8, continuous flow of 120 g.Math.L.sup.1 nitrogen free glycerol feed was started at a dilution rate of 0.6 day.sup.1. At day 15 the dilution rate and feed glycerol concentration was increased to 0.7 day.sup.1 and 130 g.Math.L.sup.1 respectively for the remainder of the experiment (FIGS. 4 and 17).

(59) During the continuous flow period at a dilution rate of 0.6 day.sup.1 a mean biomass productivity of 32.30.8 g.Math.L.sup.1.Math.day.sup.1 for a yield on substrate of 0.435, very close to previously reported data (Y.sub.x/s 0.43) (Graverholt and Eriksen, 2007). The culture produced a high phycocyanin productivity which was 115161 mg.Math.L.sup.1.Math.day.sup.1 compared to the literature (861 mg.Math.L.Math.day.sup.1) representing a yield of phycocyanin on glycerol of 15.5 mg.Math.g.sup.1. Increasing the dilution rate and feed glycerol increased biomass productivity to 40.50.6 g.Math.L.Math.day.sup.1, for a Y.sub.x/s of 0.45. During this period oxygen dropped slightly compared to at 0.6 day.sup.1 dilution rate (85-75% saturation), however, intracellular phycocyanin continued to increase up to a high of 41.0 mg.Math.g.sup.1 on day 20, producing a very high phycocyanin productivity over the last three days of 156057 mg.Math.L.Math.day.sup.1, almost double that previously reported in the literature (Graverholt and Eriksen, 2007).

(60) Table 3 demonstrates the high productivities obtained in our airlift reactor system in the dark. These data compare favourably to existing literature, with a mean intracellular phycocyanin content of 27.95 mg.Math.g.sup.1 dry cell weight over the course of 10 days at 42 C. During this period feed glycerol concentration varied from 90-230 g.Math.L.sup.1, and a high of 1.7 g.Math.L.sup.1.Math.day.sup.1 volumetric productivity was obtained for 2.5 days with lab grade glycerol at 100 g.Math.L.sup.1 concentration. These data are double the highest productivities observed in stirred tank reactors using glucose in the literature: Graverholt 2007 achieved an intracellular concentration of 15.6 mg.Math.g.sup.1 dry cell weight and volumetric productivity 0.86 g.Math.L.sup.1.Math.day.sup.1.

(61) Indeed the level of phycocyanin expressed is more comparable to data obtained with the addition of illumination of the cells using light emitting diode arrays to reach somewhere in the region of 35-40 mg.Math.g.sup.1 dcw (WO2017/050917). Additional observations by Sloth also demonstrated a greatly increased phycocyanin content of 20 mg.Math.g.sup.1 dcw when grown algae were grown on glycerol and in light (at 65 mol photons.Math.m.sup.2.Math.s.sup.1) (Sloth et al., 2006).

(62) TABLE-US-00004 TABLE 3 Productivity of continuous culture in novel reactor on different glycerol substrates Dry cell Interval weight P.sub.x Phycocyanin P.sub.phycocyanin time So D (day.sup.1) (g .Math. L.sup.1) (g .Math. L .Math. day.sup.1) Y.sub.x/s (g .Math. g.sup.1) (mg .Math. g.sup.1) (mg .Math. L .Math. day.sup.1) Condition (days) Substrate (g .Math. L.sup.1) Mean SEM Mean SEM Mean SEM Mean SEM Mean SEM Mean SEM Batch 1.82 Glucose 10 0 1.2 0.8 0 0 14.1 3.2 0 18 Fed-batch 2.94 Glucose 540 0 27.7 0.3 0 0 0.05 0 31.4 1.5 0 107 BDG 9% 3.05 Biodiesel 90 0.43 0 33.7 0.4 14.5 1.7 0.38 0 30.8 2.3 453 161 glycerol BDG 10% 1.98 Biodiesel 115 1.04 0.1 42.8 2.1 41.7 5.4 0.37 0 27.5 3.2 1126 165 glycerol BDG 20% 2.5 Biodiesel 230 0.7 0.1 53 2.6 36.9 4.8 0.2 0 23.6 1.2 1210 131 glycerol Gly 100 2.69 Lab 100 0.73 0 58.6 2.6 42.8 2.1 0.59 0 29.9 1.8 1703 204 g .Math. L 42 c. glycerol
d) Fed batch culture also shows improvement compared to the literature

(63) FIG. 3 adds further evidence supporting improvements in phycocyanin production and intracellular expression of phycocyanin when using a high oxygen saturation in our airlift system. Exponential growth on glycerol initiated at 20 g.Math.L.sup.1 batch substrate concentration was maintained over a period of 7 days averaging a growth rate of 0.81 D.sup.1 despite feed pump blockage towards the end of the study. Oxygen was maintained at a high saturation throughout the study, indeed correlating with a very high intracellular phycocyanin content reaching 37.72.4 mg.Math.g.sup.1 at the end of the batch phase at day 3.5, and 45.23.0 mg.Math.g.sup.1 by day 7 when feeding was stopped. This compares favourably with our own stirred tank data that reached 32.80.8 mg.Math.g.sup.1 on glucose in the batch phase, the difference suggested to be caused by a short period wherein the oxygen concentration dropped between days 2 and 3 (FIG. 1). It is believed that the reduction to 50% oxygen saturation around day 6 is an erroneous measurement, as upon inspection a large biofilm had clogged the dissolved oxygen probe. Indeed further supporting this conclusion is the fact that no effect on phycocyanin was observed. A total of 950 mL of 500 g.Math.L.sup.1 glycerol feed was added resulting in a final dry cell weight of 58 g.Math.L.sup.1 and a yield of 0.48, comparable to figures reported in the literature.

(64) As shown by FIG. 4 and table 3, phycocyanin is maintained at high intracellular levels (23-30 mg.Math.g.sup.1 dry cell weight) for an extended period of 10 days, whilst oxygen was maintained above 75% saturation within the reactor even with high glycerol concentrations between 90-230 g.Math.L.sup.1 using our airlift reactor system.

(65) 1. e) Continuous cultivation on other carbon sources

(66) FIG. 20 demonstrates a continuous study using mono- and dissacharides.

(67) Glucose, fructose and sucrose were sequentially used to cultivate Galdieria sulphuroria in continuous conditions in the airlift bioreactor, producing high phycocyanin productivities when oxygen was maintained at high saturations (FIG. 20, Table 4).

(68) Cultivation was initiated in semi-continuous mode for 6 days, wherein the biomass reached 25.0 (0.4) g.Math.L.sup.1. In this study computer control problems led to the pump rate not being variable, and instead the feed pump rate was manually changed according to the current biomass on a daily basis. As such, a smooth growth curve is not observed (FIG. 20). Continuous phase growth was started at 150 g.Math.L.sup.1 D-glucose and a dilution rate of 0.6 day.sup.1, leading to rapid growth for two days up to a stable concentration (e.g. days 7-10 64.7 (1.8) g.Math.L.sup.1). Oxygen reduced and became stable between 75-80% saturation, whilst intracellular phycocyanin was maintained accordingly at 44.2 (2.3) mg.Math.g.sup.1 between days 10-14, despite the glucose concentration being increased on day 11 to 200 g.Math.L.sup.1. Mean phycocyanin productivity was 1796 (151) mg.Math.L.Math.day.sup.1. Yield of biomass on substrate was 0.38 (0.03) g.Math.g.sup.1.

(69) The growth substrate was changed to fructose (200 g.Math.L.sup.1) on day 14, leading to a sharp drop in phycocyanin. This recovered after 24 hours, and is likely a consequence of a short lag phase wherein the organism is switching its metabolism to utilise the new substrate causing a rise in carbon in the reactor. Excessive carbon is well known to inhibit phycocyanin production (Sloth 2006). Some data was lost between days 16-19, however the continuous culture remained operational until a pump blockage caused a pause on day 19.

(70) On day 21, the carbon substrate was again changed to sucrose (200 g.Math.L.sup.1). Sucrose, as a disaccharide is presumed to require additional oxygen to overcome its higher energy density compared to say glucose (i.e. sucrose is more reduced). Accordingly, oxygen saturation in the reactor was stable at 65% between the steady state of days 25-27, compared to 75%+ on 200 g.Math.L.sup.1 glucose. Tests are necessary to find maximum growth rates for the given system, as phycocyanin concentration is reduced when oxygen is lower (33.92.04 mg.Math.g.sup.1). The higher biomass produced from sucrose (81.22.75 g.Math.L.sup.1, 48.7 g.Math.L.sup.1.Math.day.sup.1) offset reduced phycocyanin, for a productivity of 154095 mg.Math.L.sup.1.Math.day.sup.1.

(71) TABLE-US-00005 TABLE 4 Continuous cultivation on mono- and di-saccharides Dry cell P.sub.x P.sub.phycocyanin weight (g .Math. L .Math. Y.sub.X/S Phycocyanin (mg .Math. L .Math. D (day.sup.1) (g .Math. L.sup.1) day.sup.1) (g .Math. g.sup.1) (mg .Math. g.sup.1) day.sup.1) Substrate So Mean SEM Mean SEM Mean SEM Mean Mean SEM Mean SEM Glucose 200 0.54 0.0 75.9 5.0 40.8 4.1 0.38 44.2 2.3 1796 151 Fructose 200 0.79 0.0 91.1 3.4 72.4 0.8 0.46 24.4 8.3 1768 616 Sucrose 200 0.60 0.0 76.4 5.1 45.9 11.9 0.38 33.8 2.0 1540 95

(72) TABLE-US-00006 TABLE 5A Comparison of Phycocyanin Production Methods Table 2 Comparison of phycocyanin production Culture Organism/ Scale Trophic Carbon Scale period Light O.sub.2 S.sub.0 strain [type) state source (L) (Days) intensity saturation (g .Math. L.sup.1) (day.sup.1) A. platensis Open Pond, Auto- 135,000 365 Sunlight 0.05 continuous trophic G. sulphuraria STR, Auto- 3% CO2 1 0.7 11 100; RGB n.d. SAG 108.79 continuous trophic L .Math. min.sup.1 60:0:40% G. sulphuraria STR, Fed Hetero- D-Glucose 1.2-2.5 8 50; 500 <0.6 074G-G1 batch trophic G. sulphuraria STR, Hetero- D-Glucose 2.5 17 50; 50-150 0.6 074G-G2 Continuous trophic A. platensis Bioreactor, Mixo- D-Glucose 2.5 12.5 80-160 2; 100 n.d. Fed batch trophic G. sulphuraria Flask Mixo- Food waste 0.15 7 5.00 1.22 074G trophic hydroly- sates G. sulphuraria STR, Batch Mixo- Glucose, 2.5 10 30-50 5; 5-500 1.01-1.1 074G trophic fructose, sucrose G. sulphuraria Outdoor Mixo- Urban 100 60 Sunlight n.d. n.d. CCMEE 5587.1 bioreactor trophic wastewater 14:10 G. sulphuraria STR, Hetero- Glucose 2 17 50%; 35% 10; 20 0.6 (D.sup.1) SAG 108.79 continuous trophic batch; glycerol G. sulphuraria Airlift, fed- Hetero- Glycerol 4 7 86% 20; 500 0.81 SAG 108.79 batch trophic Glucose; glycerol; G. sulphuraria Airlift. Hetero- biodiesel 4 23 80% 10; 90; 0.6-1.0 SAG 108.79 continuous trophic glycerol 100; 230 (D 1) G. sulphuraria UTEX # 2919 Mixo- Glycerol 2 16 150-300 5-30% 30 0.6 trophic Organism/ P.sub.X X.sub.Final P.sub.c.Math.pc Y.sub.x/c.Math.pc strain (g .Math. L.sup.1 .Math. day.sup.1) (g .Math. L.sup.1) Y.sub.x/s (mg .Math. L.sup.1 .Math. day.sup.1) (mg .Math. g.sup.1) Reference A. platensis 2.94 58.8 (Jimenez et al., 2003) G. sulphuraria 0.262 n.d. 4.99 19 (Baer et al., 2016) SAG 108.79 G. sulphuraria 109 0.41 473 26.7 (Graverholt and 074G-G1 Eriksen, 2007) G. sulphuraria 24.4-83.3 0.43 163-861 15.6 (Graverholt and 074G-G2 Eriksen, 2007) A. platensis 10.24 n.d. 87 125 (Chen and Zhang, 1997) G. sulphuraria n.d. 2.75 0.55 n.d. 3 (Sloth et al., 2017) 074G G. sulphuraria n.d. 72-100 0.46-0.54 n.d. 3.4-4.3 (Schmidt et al., 2005) 074G G. sulphuraria 0.165 2.5 n.d. n.d. n.d. (Selvaratnam et al., CCMEE 5587.1 2014) G. sulphuraria 3.5 9.1 0.46 100 11 This study SAG 108.79 G. sulphuraria 22 58 0.48 1541 45.2 This study SAG 108.79 G. sulphuraria 42.8 58 0.52 1703 29.9 This study SAG 108.79 G. sulphuraria 24 0.5 696-2400 (calc) 29-100 WO 2017/050917 A1

(73) TABLE-US-00007 TABLE 5B Further Comparison of Phycocyanin Production Methods Culture O.sub.2 range Trophic Carbon period Light (mean) S.sub.0 Strain Reactor state source Scale (L) (Days) (mol .Math. m.sup.2 .Math. s.sup.1) (%) (g .Math. L.sup.1) G. sulphuraria STR; semi- Hetero- Glucose, 2.5 10 20-30 Not 5; 5-500 074G continuous trophic fructose, reported sucrose G. sulphuraria STR; Hetero- D-Glucose 2.5 17 Not 50; 50-150 074G-G2 continuous trophic reported G. sulphuraria Stirred tank Mixo- Glycerol 2 16 150-300 5-30% 30 UTEX # 2919 reactor trophic G. sulphuraria Stirred tank Hetero- Glycerol 2 17 100; 50; 35 20 ACUF141 Reactor trophic G. sulphuraria Airlift; semi- Hetero- Glycerol 4.0-5.0 7 100-50 20 ACUF141 continuous trophic (90) G. sulphuraria Airlift; Hetero- Glycerol 4 20 100-75 10.5-130 ACUF141 continuous trophic G. sulphuraria Airlift; Hetero- Glucose; 4 27 100-65 200; ACUF141 continuous trophic fructose; 200; 200 sucrose P.sub.X P.sub.c.Math.pc /D (g .Math. L.sup.1 .Math. X.sub.Final (mg .Math. L.sup.1 .Math. Y.sub.x/c.Math.pc Figure/ Strain (day.sup.1) day.sup.1) (g .Math. L.sup.1) Y.sub.x/s day.sup.1) (mg .Math. g.sup.1) reference G. sulphuraria 1.01-1.1 n.d. 72-100 0.46-0.54 n.d. 8.00-18.0 (Sloth et al., 2006) 074G G. sulphuraria 0.6 14.6-50 24.4-83.3 0.43 163-861 15.6 (Graverholt and 074G-G2 Eriksen, 2007) G. sulphuraria 0.6 24 0.5 696-2400 29-100 WO 2017/050917 A1 UTEX # 2919 (calc) G. sulphuraria 0.45 4.1 9.1 0.45 45 11.3 FIG. 1 ACUF141 G. sulphuraria 0.8 22.1 58 0.48 1188 13.8-45.2 FIG. 3 ACUF141 G. sulphuraria 0.6, 0.7 32.3-40.5 59.6 0.44-0.45 1150, 1560 31.9-41.0 FIG. 4 ACUF141 G. sulphuraria 0.54; 40.8; 75.9; 0.38; 1796; 44.2; FIG. 6 ACUF141 0.79; 0.6 72.4; 45.9 91.1; 76.4 0.46; 0.38 1767; 1540 24.4; 33.8

(74) As demonstrated by Table 5A and B, as well as the data presented thus far, our strategy for producing phycocyanin from Galdieria sulphuroria by ensuring high oxygen saturation, inaccessible in a stirred tank system, is able to generate far superior productivities in heterotrophic modes.

(75) Under continuous conditions we are able to generate over 1.7 g.Math.L.sup.1.Math.day.sup.1 using glycerol (FIG. 4 or 17), and over 1.75 g.Math.L.sup.1.Math.day.sup.1 using glucose (FIG. 20), compared to a literature high of 0.86 g.Math.L.sup.1.Math.day.sup.1 using glucose (Graverholt and Eriksen, 2007). Indeed, these data are at least comparable, if not superior to processes requiring additional illumination to a stirred tank reactor (WO 2017/050917 A1), which only achieved a biomass productivity of 24 g.Math.L.sup.1.Math.day.sup.1 compared to our 42 g.Math.L.sup.1.Math.day.sup.1, and an estimated phycocyanin productivity between 0.7-2.4 g.Math.L.sup.1.Math.day.sup.1.

(76) These data present a 567 times higher phycocyanin productivity compared to existing, industrial spirulina-based open pond technology (Jimenez et al., 2003), and at least 20 times improvement over even the most generous mixotrophic bioreactor based spirulina cultivation (Chen and Zhang, 1997).

(77) Ammonium and pH Control for Production of Phycocyanin from Galdieria

(78) e) pH Changes as Nitrogen Taken Up

(79) FIG. 6 and FIG. 7 demonstrate the relationships between growth, low pH and nitrogen uptake of Galdieria cultivations. Flask cultures were grown heterotrophically at 42 C. in the dark in a water bath with 20 g.Math.L.sup.1 glucose as the growth limiting substrate. pH and medium nitrogen was measured throughout the experiment, and cultures additionally contained glycyl-glycine as a pH buffer. In FIG. 6, the concentration of protons added to the medium during cultivation correlates with the intracellular phycocyanin concentration, suggesting a relationship between protein production and acidification of the medium. Replicate cultures were grown at different initial pH values, with the controls at pH 2.0. Higher initial pH reduced the overall acidification of the medium, correlating with reduced phycocyanin production, despite no significant effect of growth rate as observed in FIG. 7A.

(80) It is unclear why an increased pH would inhibit the uptake of nitrogen and subsequent release of protons into the medium. It may be that the acid in the medium acts with the uptake transporter to promote nitrogen uptake, although no mechanism for this is observed at this time. The implications of reducing phycocyanin expression lead to cultivation processes requiring more tightly controlled pH values.

(81) f) Large amounts of nitrogen needed to produce phycocyanin

(82) Studies into optimising carbonC:N ratios for Galdieria identified the greatest phycocyanin productivity occurring at a ratio between 10.sup.15:1 C:N, with impaired productivity at higher ratios (Sloth et al., 2006). As such, to support high density culture, extremely solute rich feeding solutions are requireda 500 g.Math.L.sup.1 glucose based feeding solution required 110 g.Math.L.sup.1 ammonium sulphate (Graverholt and Eriksen, 2007), as well as other inorganic components. These amounts are very close to solubility limits and require often overnight preparation with heating to fully dissolve.

(83) Turning to glycerol as the carbon source, highly concentrated solutions as above become impossible. Glycerol is hygroscopic, removing non-complexed waterfrom bulk that is required to solvate the large amounts of ammonium sulphate. Ammonium sulphate is not soluble in glycerol. For glycerol therefore, separating the nitrogen source is a requirement, posing a problem to the design of culture systems. In the present invention an automated control system was developed to solve this problem, as described below.

(84) g) Automated Control is Able to Maintain Nitrogen Levels

(85) Further to the above solubility issues in glycerol, Galdieria's cultivation on ammonium leads to significant drop in medium pH. Indeed a typical high density system easily reduces to well below pH 1.0 if uncontrolled, impacting growth rate. There is evidence to suggest that Galdieria expresses the AMT ammonium uptake transporter system, which results in a proton being left outside of the cell as the ammonium ion is deprotonated into ammonia, ammonia taken up as a gas through the facilitating transporter, and reprotonated on the inside of the cell (Lamoureux et al., 2010).

(86) By stochiometrically balancing either the amount of protons in a continuous medium, or ensuring an equal pH of feed in batch and fed-batch systems, we were able to maintain stable nitrogen concentrations in the reactor without intervention through an automated control system which automatically responds to pH drop (FIG. 8).

(87) 1. Semi-Continuous

(88) a) NH.sub.4OH

(89) FIG. 9 demonstrates a successful glycerol based cultivation wherein ammonium hydroxide is used to maintain nitrogen levels in the reactor during feeding. Balancing the pH of the feed and reactor is important, as shown by downward trending medium nitrogen throughout the study. It is thought that pH differences between feed (2.0) and bioreactor (1.9) contributed primarily to this reduction, as the feed would exert a diluting effect under semi-continuous conditions. Nevertheless, nitrogen was maintained between 1.4-0.6 g.Math.L.sup.1 in the reactor despite rapid growth to high cell density of 75 g.Math.L.sup.1 in 6 days and addition of 150 mL 10 M NH.sub.4OH during feeding. This is considerably faster than reported growth in the literature, the highest of which reached 109 g.Math.L.sup.1 dry cell weight after 16 days. These data also support the need for high oxygen to promote production of phycocyaninoxygen reached a low of 28% saturation at day 5.5, and co-ordinates with a reduction of intracellular phycocyanin towards the end of the cultivation to 23 mg.Math.g.sup.1. This is similar to reported productivities, the highest of which is 26 mg.Math.g.sup.1. The reduction of oxygen in this MK 3 airlift bioreactor revealed problems with the oxygen transport, spurring the construction of the MK 4 airlift reactor with an improved sparger (FIG. 12). That experiment uses NH.sub.3 gas, but importantly the oxygen saturation doesn't reduce below 80%, and produced an intracellular phycocyanin of 45 mg.Math.g.sup.1 at the final sample.

(90) b) NH.sub.3

(91) We found that use of ammonia gas identifies as a more reliable technique of adding basic nitrogen to the system without generating highpH localised zones due to inadequate mixing, which is particularly important for scalability. FIG. 10 outlines the system setup for this type of control. Instead of a peristaltic pump, a solenoid valve manages ammonia flow into the bioreactor, controlled by a pH sensing computer. FIG. 11 outlines safety characteristics of the automated system. By introducing lockouts for exceptionally high or low pH measurements, addition of ammonia gas into the system is prevented under failure conditions: say the reactor has failed and culture is leaking, at the point that the pH probe is no longer in contact with medium the pH measurement will rise, locking out the system.

(92) By using ammonia gas instead of ammonium hydroxide, two process benefits were observed. Volume inside the reactor is not readily changed by addition of gas, compared to hydroxide solution, simplifying process kinetics. Inadequate mixing is overcome, due to the much larger contact area of gas bubbles compared to a liquid hydroxide tube. As the airlift reactor of the invention has a slower mixing time than a stirred tank in order to prevent excessive shear being exerted on the cells, the high rate of gas exchange in the airlift design may be exploited.

(93) Indeed, the system performed as intended, being able to maintain a stable pH and medium nitrogen concentration between 0.5-1.5 g.Math.L.sup.1, despite a total of 43 g of ammonia gas being added over the course of 7 days (FIG. 12). pH is observed to drop at the start of the study until day 3, at which point the ammonia control system was activated, and pH is stabilised.

(94) 2. Continuous

(95) a) NH.sub.4OH

(96) In a continuous system, the relationship between pH and addition of basic nitrogen differs to the semicontinuous systems described above, as acidic cell suspension is leaving the system and must be accounted for. In this case, additional acid is required in the flowing continuous feed to replace that lost from the system. Acid more readily dissolves in glycerol solutions than ammonium sulphate, and has the added benefit of inhibiting contamination of the glycerol, simplifying capital equipment and ensuring sterility. During the experiments presented here, ammonium sulphate has been used to provide an initial nitrogen source upon inoculation. A more optimal strategy further minimising contamination risk, which is highest at the start of cultivation, would be to prepare starter batch medium with a very low pH free of nitrogen, and following inoculation raise the pH with basic nitrogen addition over the course of the first few days.

(97) Using an acid concentration proportional to that of the carbon in the continuous feed, as would be done for nitrogen, ensures the correct stochiometric addition of balance nitrogen to support growth and phycocyanin production in Galdieria (Table 1).

(98) FIG. 13 and FIG. 14 provide two examples where the acidic nitrogen-free continuous medium is able to maintained stable nitrogen concentrations in the bioreactor for the cultivation of Galdieria. In FIG. 13, nitrogen as well as pH in the reactor is maintained around 5.0 g.Math.L.sup.1, except at day 7, where problems with the MK 2 airlift reactor led to some culture loss. FIG. 14 demonstrates an even more stable nitrogen concentration, around 1.0 g.Math.L.sup.1, using the MK 4 bioreactor. In this experiment further evidence supporting the high oxygen requirement of phycocyanin is observed, intracellular concentration of which drops from days 9.sup.11 at 50% O.sub.2. As the air flow rate must be lowered to reduce oxygen saturation in the airlift bioreactor, the mixing has also been affected, leading to a positive feedback effect of excessive glycerol inhibiting cell growth towards the end of the experiment. The baseline glycerol concentration around 10 g.Math.L.sup.1 throughout the experiment also causes the overall reduced phycocyanin concentration at 25-30 mg.Math.g.sup.1 dry cell weight.

(99) Comparison of Pigments produced in Galdieria grown heterotrophically on glycerol compared to autotrophic culture of Spirulina

(100) FIG. 15 shows absorbance measurements from pigments extracted in the same way from Galdieria (B) grown heterotrophically on glycerol compared with Spirulina (A) grown in the traditional autotrophic manner. C-Phycocyanin (CPC) absorbs most strongly at 620 nm whilst the related protein allophycocyanin (APC) absorbs most strongly at 652 nm. When the two proteins are in a mixture, the characteristic shoulder to the absorbance curve is observed as in (A), demonstrating the resultant absorbance spectrum. Highly purified CPC can be extracted from heterotrophic cultures of Galdieria as shown in (B), where no such shoulder is present at 652 nm indicating a lack of allophycocyanin pigment impurity. The Spirulina sample contained 2.06 mg.Math.ml-10.024 phycocyanin and 0.037 mg.Math.mL10.001 allophycocyanin, whilst the heterotrophic Galdieria sample contained 1.72 mg.Math.mL.sup.1 phycocyanin and undetectable levels of allophycocyanin having undergone the same purification procedure CARBONE, D. A., OLIVIERI, G.g., POLLIO, A. & MELKONIAN, M. 2020. Biomass and phycobiliprotein production of Galdieria sulphuroria, immobilized on a twin-layer porous substrate photobioreactor. Applied Microbiology and Biotechnology, 104, 3109-3119. GRAVERHOLT, 0. S. & ERIKSEN, N. T. 2007. Heterotrophic high-cell-density fed-batch and continuous-flow cultures of Galdieria sulphuroria and production of phycocyanin. Applied Microbiology and Biotechnology, 77, 69-75. GRAZIANI, G.g., SCHIAVO, S., NICOLAI, M. A., BUONO, S., FOGLIANO, V., PINTO, G. & POLLIO, A. 2013. Microalgae as human food: chemical and nutritional characteristics of the thermo-acidophilic microalga Galdieria sulphuroria. Food Funct, 4, 144-52. GROSS, W., KVER, J., TISCHENDORF, G., BOUCHAALA, N. & BOSCH, W. 1998. Cryptoendolithic growth of the red alga Galdieria sulphuroria in volcanic areas. European Journal of Phycology, 33, 25-31.p KURSAR, T. A. & ALBERTE, R. S. 1983. Photosynthetic Unit Organization in a Red Alga: Relationships between Light-Harvesting Pigments and Reaction Centers. Plant Physiol, 72, 409-14. LAMOUREUX, G., JAVELLE, A., BADAY, S., WANG, S. & BERNECHE, S. 2010. Transport mechanisms in the ammonium transporter family. Transfus Clin Biol, 17, 168-75. MINODA, A., SAKAGAMI, R., YAGISAWA, F., KUROIWA, T. & TANAKA, K. 2004. Improvement of culture conditions and evidence for nuclear transformation by homologous recombination in a red alga, Cyanidioschyzon merolae 10D. Plant Cell Physiol, 45, 667-71. RHINE, E. D., MULVANEY, R. L., PRATT, E. J. & SIMS, G. K. 1998. Improving the Berthelot Reaction for Determining Ammonium in Soil Extracts and Water. Soil Science Society of America Journal, 62, 473-480. SLOTH, J. K., WIEBE, M. G. & ERIKSEN, N. T. 2006. Accumulation of phycocyanin in heterotrophic and mixotrophic cultures of the acidophilic red alga Galdieria sulphuroria. Enzyme and Microbial Technology, 38, 168-175. LEMASSON, C., TANDEAUD.N & COHENBAZ.G 1973. ROLE OF ALLOPHYCOCYANIN AS A LIGHT-HARVESTING PIGMENT IN CYANOBACTERIA. Proceedings of the National Academy of Sciences of the United States of America, 70, 3130-3133.