Algal Research 5 (2014) 103–111 Contents lists available at ScienceDirect Algal Research journal homepage: www.elsevier.com/locate/algal Effects of light and temperature on the photoautotrophic growth and photoinhibition of nitrogen-fixing cyanobacterium Cyanothece sp. ATCC 51142 Pongsathorn Dechatiwongse, Suna Srisamai, Geoffrey Maitland, Klaus Hellgardt ⁎ Department of Chemical Engineering, Imperial College London, South Kensington Campus, London SW7 2AZ, UK a r t i c l e i n f o Article history: Received 13 January 2014 Received in revised form 4 June 2014 Accepted 9 June 2014 Available online xxxx Keywords: Biomass Light intensity Photoinhibition Temperature Unicellular cyanobacterium Cyanothece sp. ATCC 51142 a b s t r a c t The unicellular, nitrogen-fixing cyanobacterium Cyanothece sp. ATCC 51142 is a promising strain with a remarkable capability of producing large quantities of hydrogen, an energy carrier long being promoted as an ideal fuel. Under extreme environmental conditions, significant reduction of cellular photosynthetic capability is commonly observed in algae and cyanobacteria. Even less severe conditions can induce photo-inhibitive growth dynamics, which in turn result in a marked decrease in biomass and gas productivities. In this study a detailed analysis of the effect of two extrinsic parameters, namely light intensity and temperature, on the photoautotrophic growth of Cyanothece was performed in order to reveal critical conditions that would lead to undesired photoinhibition. A high degree of coherence between cyanobacterial growth and nutrient uptake kinetics was observed, as well as a strong dependence on the change of the two parameters. Nitrogen depletion was confirmed as a trigger, which transforms an exponential into a stationary growth phase. A non-linear relationship between the maximum specific growth rate and the irradiance up to 320 μE m−2 s−1 was identified and found to be dominated by light saturation rather than photoinhibition. The relationship between the specific growth rate and the temperature was found to be linear until a remarkable drop in the final biomass productivity and cyanobacterial photosynthetic capability was observed at 40 °C. The cause of this is a high temperatureinduced photoinhibition effect. © 2014 Elsevier B.V. All rights reserved. 1. Introduction Cyanothece sp. ATCC 51142 (referred to as Cyanothece 51142) is a unicellular diazotrophic marine cyanobacterium [1], which deserves attention due to its remarkable rate of gaseous hydrogen (H2) production [2,3]. This gas has long been considered as a possible alternative energy carrier to provide clean and renewable power for transportation, electricity generation and heating. Like many oxygenic cyanobacteria, there are two distinct enzymes, which are directly involved with the biological H2 formation reaction of this strain: bidirectional hydrogenase and nitrogenase [4–6]. The former enzyme is responsible for catalysing the reversible recombination reaction between protons and electrons, which are derived from photosynthetic water splitting. In the case of nitrogenase, the enzyme catalyses an irreversible nitrogen fixation reaction, during which molecular atmospheric nitrogen (N2) is biologically fixed into ammonia (NH3), concomitantly releasing H2 as by-product. As the catalytic activity of nitrogenase is reported to be predominant for this species [3], our discussion here will only relate to the gas production by this particular enzyme. Under nitrogen-limiting conditions, this microorganism employs tight metabolic regulation by ⁎ Corresponding author. Tel.: +44 20 7594 5577. E-mail address: k.hellgardt@imperial.ac.uk (K. Hellgardt). http://dx.doi.org/10.1016/j.algal.2014.06.004 2211-9264/© 2014 Elsevier B.V. All rights reserved. an internal circadian clock to temporally separate oxygenic photosynthesis and O2-sensitive nitrogen-fixation [7]. Exceptionally, this strain is also capable of performing N2-fixation regardless of illumination conditions, such as 6 h–6 h light/dark cycles and continuous light [8], or even continuous dark [9]. In addition to standard photoautotrophic conditions, Cyanothece 51142 has been reported to also grow under photo/chemoheterotrophic and mixotrophic conditions [10–12]. As a consequence, this cyanobacterium offers the potential to biochemically convert carbon dioxide (CO2) and/or industrial waste such as glycerol into H2 and biomass. Despite promising biofuel applications offered by Cyanothece 51142, the most significant drawback of cyanobacterial H2 production is the generally low conversion efficiency from solar energy to H2 of less than 2% [13]. Under non-optimal conditions, 273 ml H2 was produced by our batch culture of Cyanothece 51142, corresponding to an efficiency of 0.91% (see supplementary sheet for experimental results and calculation). Previously reported values include 0.97% for the sulphurdeprived green alga Chlamydomonas reinhardtii [14] and 1.4% for the mutant cyanobacterium Anabaena variabilis PK84 [15]. Theoretically, this efficiency could be as high as 12% [16] and its significant reduction is a result of a number of controlling factors, both extrinsic (light, temperature, substrate level, etc.) and intrinsic (intracellular components and H2-mediated enzymes and genes) [17–20]. 104 P. Dechatiwongse et al. / Algal Research 5 (2014) 103–111 Nomenclature ASP2 Artificial seawater medium Cyanothece 51142 Unicellular cyanobacterium Cyanothece sp. ATCC 51142 Chl Chlorophyll Concentration of nutrient “i” at any specific time “t” [Ci (t)] [Ctotal, i] Total consumption of nutrient “i” DBC Dry biomass concentration Molecular hydrogen H2 I Light intensity IC Ion chromatography Light saturation term ks OD Optical density PBR Photobioreactor Dissolved oxygen concentration pO2 Phosphate anion PO3− 4 PSII Photosystem II reaction centre Nitrogen N2 Ammonia NH3 Nitrate anion NO− 3 Sulphate anion SO2− 4 μmax Maximum specific growth rate Maximum specific growth rate under the lightμmax,s saturated condition Maximum specific uptake rate under the lightrmax,s saturated condition I2 Photoinhibition term ki With the aim of improving H2 productivity (as well as the corresponding conversion efficiency), a two-stage strategy, which comprises two sequential stages – growth and gas production – can be adopted [2]. Systematically, during the primary stage, the optimal growth conditions are applied in order to maximise the cyanobacterial biomass productivity. Upon achieving the highest biomass density, various means such as nutrient stress or sparging with inert gas are successively employed to induce anaerobic conditions, necessary for H2 production [3,17]. Light is undoubtedly one of most important parameters, affecting the conversion of solar to H2 efficiency [14], as it provides the fundamental energy to drive a variety of physiological activities and metabolic processes. The relationship between the net photosynthetic rate – the sum of O2 evolution and respiration rates – and irradiance can be divided into three stages: (i) light limitation, (ii) light saturation, and (iii) light inhibition [21,22]. During the first stage, provided other extrinsic parameters are in excess without being inhibitive, the intensity of illumination becomes the limiting factor and the net rate of photosynthesis increases linearly with irradiance. However, eventually, the rate declines under higher light intensity. This is due to the finite capacity of light absorption by the photosynthetic apparatus in photosystem II (PSII). The specific intensity, at which the photosynthetic activity reaches its maximum value (Pmax), is known as the “light-saturated intensity.” Beyond this critical point, the light-induced breakdown of the protein D1, located at the reaction centre of PSII, occurs and results in the reduced activity of photosynthesis [23]; this phenomenon is extensively referred as “photoinhibition.” Under such conditions microbial growth is much reduced [24–26]. This means that the surplus amount of solar energy, beyond cellular absorption capacity, is wasted, thereby reducing the overall solar conversion efficiency [16]. If light levels are maintained below the critical value, the degree of microbial susceptibility to photoinhibition is influenced by other environmental factors [27]. Temperature is another key extrinsic parameter, which has long been known to have a strong influence on the photosynthetic capability of plants and microalgae. Similar to light, an increase in temperature generally enhances microbial photosynthesis and subsequent growth rates up to a critical temperature, above which these activities start to decline [28–31]. At temperatures below 20 °C, the recovery of damaged D1 protein has been shown to be suppressed, which leads to the reduction of cellular photosynthetic activities. However, these activities can be recovered and subsequently return to their original functionality by exposing cells to a higher temperature [32,33]. Conversely, under extreme heat stress, the detachment of the Mn-stabilising extrinsic 33 kDa protein from the PS II core complex has been reported as the potential cause of high temperature-induced inactivation of the photosynthetic capability [34]. In open raceways, where the control of temperature is impossible without using external heating or cooling devices, a marked decrease in biomass productivity was previously observed as the consequence of photoinhibition [35,36]. In this paper, we present the first precise study of the effect of light intensity and temperature on the photoautotrophic growth of Cyanothece 51142, and subsequently determine the optimal condition for each of these parameters. To this effect we employed a tubular photobioreactor [37] from Sartorius Ltd. to monitor the growth dynamics in real time under a variety of culture conditions. 2. Materials and methods 2.1. Culture of Cyanothece 51142 and medium The culture of Cyanothece 51142, purchased from American Type Culture Collection (ATCC), was grown in an artificial seawater medium – ASP2 [38] – with the addition of 1.5 g NaNO3 L−1 and 10 mM of glycerol (99% purity) from Sigma-Aldrich in 250 ml sterile Corning flasks at 30 °C, pH of 7.4, under continuous cool-white fluorescent illumination of 46 μE m−2 s−1. After 5 days of cultivation, cells achieved their maximum concentration and were then inoculated in a Sartorius tubular PBR, using ASP2 medium, but without glycerol. 2.2. Sartorius tubular PBR Our reactor choice was a tubular flow Biostat PBR 2S (see Fig. 1), which was purchased from Sartorius Stedim Biotech GmbH, Göttingen, Germany. The PBR consists of two major compartments, a 1-l central vessel, surrounded by a 2-l helix. The central vessel houses all measuring instruments as well as sampling and aerating equipment, whereas the helix module has been designed to provide optimum illumination efficiency per surface area. The temperature of the aqueous culture is measured by a Pt100 Type 25-3 thermocouple and is maintained by means of a water bath in the double-layered central vessel. pH and dissolved oxygen concentration (pO2) are measured by probes from Hamilton – Easyferm : Plus K8 160 pH electrode and Oxyferm FDA 160 oxygen tension probe, respectively. The growth of the cyanobacteria is monitored by the Sartorius Fundalux II OD probe. The circulation of cells through the reactor's body is performed by an Ismatec peristaltic pump at a recirculation rate of 2200 ml min−1. Variables such as light intensity and temperature are controlled using a touch-screen Micro DCU-System display on the control tower, which also performs datalogging, using specialised software. The tower and software were also supplied by Sartorius Stedim Biotech GmbH. 2.3. Determination of total chlorophyll concentration The total chlorophyll (Chl) concentration of cells was determined using the following equation: [mg Chl L−1] = 14.97 (A678 − A750) − 0.615 (A620 − A750), where An is an absorbance spectrophotometrically measured at the specific wavelengths of 620, 678 and 750 nm [3]. 2.4. Determination of total dry biomass concentration A 100 ml sample of algae was collected from the PBR and centrifuged at 13,000 rpm for 10 min. The resulting pellet was washed at least twice P. Dechatiwongse et al. / Algal Research 5 (2014) 103–111 105 A Gas Sample Liquid sample Gas purge pO 2 Temperature Touch screen OD probe pH Control tower B Central vessel Light source Helix module Fig. 1. Schematic diagram of Sartorius tubular PBR. (A) Side view: a schematic diagram showing the 3-L tubular PBR with its control tower, peristaltic pump and measuring and sampling instruments. (B) Top view (without the top cover, which hosts all probes and sampling locations). An outer green ring represents the 2-L helix module, whereas an inner green circle represents the 1-L central vessel. with deionised water, followed by re-suspension in a pre-weighed glass bottle. The total dry weight (biomass + bottle) was determined gravimetrically. The dry biomass concentration (DBC) was thus calculated from the following equation: Dry biomass concentration ðDBCÞ ¼ Total dry weight−Weight of bottle Volume of sample 2.5. Calibration The pH probe was calibrated using two buffer solutions, with known pH values of 7 and 10. The pO2 probe was calibrated to 0% (in argon-saturated water) and 21% (in air-saturated water). The profile of total DBC was determined from the corresponding profile of total Chl concentration using our developed linear regression equation: DBC (g L− 1) = 0.0764*OD (mg Chl L − 1 ) + 0.0084, which was based on individual 32 experimental data points, R2 = 0.9617 (see the supplementary sheet). 2.6. Experimental conditions 2.6.1. Study of light effect The cyanobacterial culture was cultivated under continuous illumination with the light intensity set at 23, 46, 92, 138, 207, 275 and 320 μE m−2 s− 1. The medium was ASP2, with an addition of 1.5 g NaNO3 L−1. Temperature was maintained at 35 °C; 20 ml min− 1 of sterile 10% volume CO2 volume air−1 was sparged through the aqueous culture to provide photoautotrophic conditions. 106 P. Dechatiwongse et al. / Algal Research 5 (2014) 103–111 2.6.2. Study of temperature effect The cyanobacterial culture was cultivated under continuous illumination with a fixed intensity of 69 μE m−2 s−1. The medium was ASP2, with an addition of 1.5 g NaNO3 L−1. The investigated temperatures were 25, 30, 32, 35, 37 and 40 °C; 20 ml min− 1 of sterile 10% volume CO2 volume air−1 was sparged through the aqueous culture to provide photoautotrophic conditions. 2.7. Analytical techniques 2.7.1. UV Vis-spectrophotometry Spectrophotometry was used to monitor the growth of the cyanobacterial culture by light absorption. A Lambda 40 UV/Vis Spectrometer from Perkin-Elmer Instruments running UV WinLab software was used to measure OD at the specific wavelengths of 620, 678 and 750 nm [3]. The OD readings at 620 and 678 nm represent absorption by phycocyanin and chlorophyll light responsive units of the cyanobacterium [1]. Cyanobacterial samples were analysed in a 1.5 ml cuvette and the instrument was autozeroed with the light absorbed measurement from a cuvette filled with ASP2 medium. 2.7.2. Ion chromatography Ion chromatography (IC) was used to quantitatively determine the 2− change of key nutrients – nitrate (NO− 3 ), sulphate (SO4 ) and phosphate (PO34 −) concentrations in the aqueous culture – and thus the associated cyanobacterial consumption rates. This task was carried out using an 882 Compact IC Plus with 863 Compact Autosampler from Metrohm running MagIC Net software. The separation column was a Metrosep A Supp 5, with the dimensions of 150 mm L × 4.0 mm ID. The concentrations of compounds eluting from the column were measured by means of a conductivity detector. The sodium-based anion solution (liquid phase) consisted of 3.2 mM Na2CO3 and 1.0 mM NaHCO3. Deionised water was used as the rinse and 0.1 M sulphuric acid as the regenerator. The liquid phase was pumped through the instrument using a peristaltic pump at a flow rate of 0.7 ml min−1. The concentrations were calibrated using CertiPUR Anion Multi-Element Standards from Merck at concentrations of 1, 10, 50, 100 and 150 ppm (1 ppm = 1 mg L− 1). The anion concentrations were determined from the areas of the respective peaks. Retention times, when the anions were eluted from the column, were 10.3, 13.8 and 16.2 min for 2− and PO3− NO− 3 , SO4 4 , respectively. For sample preparation, cyanobacterial samples were spun down in a mini-centrifuge at 13,000 rpm for 10 min; only the supernatant liquid was used for IC investigation. The purified liquid was diluted by a factor of 10 to avoid the column being overloaded. However, it was subsequently found that the PO34 − concentrations present in the medium were not sufficiently high to be tracked by this technique. 2.8. Modelling 2.8.1. Logistic model A logistic function, Eq. (1), has been frequently and effectively used to fit kinetic parameters to algal growth and nutrient uptake data [39,40] ODðt Þ ¼ ODmax 1 þ e−μ max
ðt−t 0 Þ ð1Þ For the modelling of growth, ODmax represents the maximum cell density obtained in a particular experiment. μ max is the maximum specific growth rate, which describes the number of cell divisions per unit time under the respective condition. t0 is the inflexion point of the function. In order to fit the logistic curve to discrete experimental data, the total least squares error between modelled and experimental ODs is minimised. By assigning values for the parameters of Eq. (1) in this way, OD(t) at any specific time can be consequently determined. In the case of nutrient uptake, an inverted form of the logistic function, as expressed in Eq. (2), was utilised, as it is able to reflect the observed decrease in nutrient concentration. ½C ðt ފ ¼ ½C total Š 1 þ ermax
ðt−t 0 Þ ð2Þ In this case, [Ctotal] represents the total consumption of a particular nutrient and rmax becomes the corresponding maximum specific uptake rate. 2.8.2. Aiba model The algal maximum specific growth rate increases with increasing light intensity up to the point of light saturation, at which the highest possible rate, μmax,s, is attained. Beyond the saturation intensity, the microbes start to experience photoinhibition effects and the subsequent growth rate will decline [41]. The Aiba model [42], Eq. (3), has been developed to describe this behaviour. For this model, μmax is the maximum specific growth rate attained by cells under respective light intensity (I). The parameters, including ks (light saturation term) and ki (photoinhibition term), of the model were determined in a similar way to those for the logistic model. μ max ðIÞ ¼ μ max;s
I ks þ I þ I2 ki ð3Þ 3. Results and discussion 3.1. Growth/nutrient kinetics and logistic model verification The microbial growth is commonly characterised by five distinct phases: (i) lag (adaptation), (ii) exponential (light-saturation), (iii) linear (light-limited), (iv) stationary (nutrient-limited) and (v) cell death [43]. Fig. 2A represents all of growth phases, except the death phase, of Cyanothece 51142 culture cultivated photoautotrophically under illumination of 207 μE m−2 s−1. During the lag phase, cells adapt to the new environment after inoculation; consequently, the change in cell density is minimal during the first 20 h. An exponential growth phase is then observed between 20 and 70 h, during which simultaneous increases in cell density and nitrate consumption are clearly seen. During this period, the rate of cellular reproduction proceeds as an exponential function of time, as long as both light and nutrients are sufficient. The positive gradient between the natural log of biomass concentration and time is commonly used to estimate “maximum specific growth rate,” denoted ‘μmax’ (black dashed line) of any microbial batch growth culture. In the case of the nitrate uptake, the negative gradient between the natural log of substrate concentration and time is known as the “maximum specific uptake rate,” denoted ‘rmax’ (red dashed line). The determination of growth parameters (μ max and maximum DBC) was performed iteratively by initially estimated the values of both parameters. As a result, the biomass concentration at each specific time can be calculated using the logistic model. Eventually, the sum of root squared deviations between the experimental and logistic biomass concentration data was minimised in order to precisely determine actual values for both parameters. In the case of nitrate uptake parameters, the same procedure was carried out to work out rmax and total nitrate uptake. The fitted functions are presented as black dotted lines in Fig. 2B and C, where the magnitudes of μ max and rmax under this particular light intensity are 0.073 and 0.068 h−1, respectively. Plotting one specific rate against another for all of our studied conditions yields a linear regression with a gradient close to unity (see supplementary sheet), which implies that cyanobacterial growth is directly proportional to nitrate consumption. As the cell concentration exponentially increases, a higher amount of photosynthetically available photons is absorbed by the culture. When P. Dechatiwongse et al. / Algal Research 5 (2014) 103–111 107 Fig. 2. Growth and nutrient kinetics of Cyanothece 51142, cultivated at 207 μE m−2 s−1, 35 °C, 20 ml min−1 of sterile 10% volume CO2 volume air−1 and 1.5 g NaNO3 L−1 in ASP2 medium. (A) Four common growth phases of the cyanobacterial culture. μmax was estimated from the natural log of the growth profile over time during an exponential phase. An increase in biomass concentration was observed in parallel with the decrease in NO−1 3 substrate. (B) A decreasing profile of dissolved O2 was due to an increasing culture density, which implies a greater rate of cellular respiration, but with the same rate of photosynthetic O2 production (under a fixed irradiance). (C) An alternating pH profile confirms the circadian behaviour of Cyanothece 51142, which periodically performs photosynthesis and N2-fixation in approximately 12 h light–12 h dark cycles, respectively. all photons are captured, light becomes the limiting factor and the rate of cyanobacterial cell divisions is observed at constant value, which is lower than μmax. This light-limited growth condition is known as the linear growth phase. Fig. 2A illustrates nitrogen deficiency acting as an inhibitive trigger on cyanobacterial growth, similar behaviour being observed under other light condititons studied (apart from the low-light condition of 23 μE m−2 s−1, where photon flux becomes the limiting factor). This observation is in agreement with many previous studies reported for algae and cyanobacteria [44–46], which show the subsequent cessation of cell growth after the nitrogen deprivation. During the stationary phase, μ (t) = 0, cells generally start to utilise their intracellular accumulated metabolites [47], in this case proteins, which were synthesised from extracellular nitrate. As soon as these metabolites are exhausted, the growth phase is transformed into an eventual death phase, where the rate of cell death is greater than the rate of cell reproduction, μ (t) b 0. In addition to nitrogen, there are three other essential macronutrients – sulphur, carbon and phosphorous – required by microorganisms. Our results obtained from ion chromatography show an excess sulphate L−1 in the medium concentration with a mean value of 2160 mg SO−2 4 (data not shown), whereas CO2 is continuously sparged through the aqueous phase of the PBR to ensure sufficient concentration of dissolved carbon. In the case of phosphorus, as mentioned in Section 2.7.2, the concentrations present in the prepared sample were extremely PO3− 4 low and could not be tracked by this technique. However, the study conducted by Sinetova et al. [46] showed that there was no significant impact on this cyanobacterial growth caused by the phosphorus starvation. 108 P. Dechatiwongse et al. / Algal Research 5 (2014) 103–111 In our study, the concentration of dissolved O2, which represents the sum of total photosynthetic rate (=the rate of photosynthetic O2 evolution − the rate of O2 consumption by cellular respiration) and rate of O2 dissolution from sparging air into liquid culture, can be seen to decrease sharply after the lag phase and the culture entered the exponential phase (Fig. 2B). This is due to the fact that as the culture matures, the rate of cellular respiration increases, whilst the photosynthetic oxygen evolution is limited by the decreased degree of light penetration. The onset of approximately 24 h periodic oscillations of pH was clearly observed at a nitrate concentration of 400 mg L−1, see Fig. 2C. The magnitude of the variation between upper and lower pH for each cycle (labelled on Fig. 2C) increased as the nitrate concentration decreased. This phenomenon occurs due to the circadian behaviour of Cyanothece 51142, which periodically performs photosynthesis and N2-fixation in approximately 12 h light–12 h dark cycles, respectively [7]. In the presence of nitrate, between time = 60–100 h, the nitrogenase enzyme, which catalyses the N2-fixing reaction, is inactivated [17]; consequently, the photosynthetic CO2-fixation can be the sole cause of the pH variation during this time period. During the subjective light period, CO2 is photosynthetically fixed and lower dissolved gas concentration causes the pH to increase. Conversely, during the subjective dark cycle, the carbon fixation process temporally ceases thereby increasing the pH. After time = 100 h, when nitrate becomes depleted, the variation of pH is partially affected by the formation of NH3, which acts as a weak base, during the subjective dark period. 3.2. Light intensity effect To identify optimal illumination conditions to attain maximum Cyanothece 51142 growth rate, the impact of different light intensities (23, 46, 92, 138, 207, 275 and 320 μE m−2 s−1) on the cyanobacterial growth and nutrient uptake were investigated. Both growth and uptake kinetic parameters for different light intensities were determined in the same way as previously described for the case of 207 μE m−2 s−1 (see supplementary data). Fitted model data obtained from both kinetic models are shown in Table 1 and Fig. 3A and B. Our results show that the duration of the lag phase appears to be the same of approximately 24 h, regardless of the illumination condition, whereas the effect of light intensity on the growth phases can be clearly seen. The most significant impact was observed when the light intensity was increased from 23 to 46 μE m− 2 s− 1, as the maximum specific growth rate was increased by a factor of 7.25. At 23 μE m−2 s−1, virtually no growth was observed due to the cyanobacteria not collecting sufficient solar energy to photosynthetically produce chemical energy, which is required to drive internal cellular processes, including the energy-dependent nitrate uptake [48]. It can be seen from Fig. 3C that as light intensity increases, the growth rate of the cyanobacterial culture increases. However, the relationship between these two parameters appears to be non-linear. Above 138 μE m−2 s−1, the maximum attained DBC of 3.7 g L−1 (red squares), during the stationary phase, started to decline, whereas the specific growth (green crosses), during the exponential phase, continued to rise with no sign of reaching a plateau. This non-linear light intensity-specific growth rate relationship was Table 1 Growth and uptake kinetics data under different studied light intensities. Light intensity (μE m−2 s−1) μmax (h−1) Maximum DBC (g L−1) rmax(h−1) Total nitrate uptake L−1) (mg NO−1 3 23 46 92 138 207 275 320 ~0.004 0.029 0.039 0.050 0.073 0.080 0.088 0.2 2.2 3.1 3.7 3.2 3 2.9 0.001 0.017 0.040 0.047 0.068 0.079 0.084 146 954 1046 1099 1276 1200 1087 found to be well-described by the Aiba model (black dots on Fig. 3C) with an estimated maximum specific growth, μmax,s, attained by the cyanobacterium under light saturated conditions, of 0.186 h− 1. This rate should not be confused with μmax, which represents the maximum specific growth rate under the respective irradiance. In the case of the nitrate consumption kinetics, a similar non-linear behaviour was observed (Fig. 3D), with a rmax,s value of 0.181 h−1, respectively. Similar to the logistic function, the two Aiba model constants – ks (light saturation term) and ki (photoinhibition term) – were determined by minimising the sum of root squared deviations between experimental data and model, and were subsequently found to be 347 and 10,068 μE m− 2 s−1, respectively. Even at 320 μE m−2 s− 1, the value of the photoinhibition term (I2/ki ~ 10 μE m−2 s−1) is still insignificant compared to the light saturation term. This suggests that the reduced DBC is a consequence of poor light distribution throughout the liquid culture due to mutual shading between cells [49], rather than an intensive light-induced photoinhibition. Our results are consistent with previous findings [50,51] which show that the effect of photoinhibition on green algae and cyanobacteria is generally seen under intensities of the order of 500–700 μE m−2 s−1. However, a recent study [46], which also investigated the batch growth dynamic of Cyanothece 51142 using red coloured light, λmax ~ 627 nm (in our study, artificial white light, whose wavelength range is 390–700 nm, was used), presents a parabolic profile of the cyanobacterial maximum specific growth rate, with the plateau of 1.5 day−1 (0.063 h−1), being observed under 200 μE m−2 s−1. Sinetova et al. [46] propose an intensive light-induced shift of the surplus photosynthetic electron flow towards the cellular maintenance and repair processes instead of growth metabolism, as the potential cause for an eventual decline in the specific growth rate. 3.3. Temperature effect In a similar approach to Section 3.2, the growth behaviour of Cyanothece 51142 was studied at different temperatures (25, 30, 32, 35, 37 and 40 °C). All parameters of the growth kinetics were determined using the logistic models fitted to the experimental data (Table 2); the good quality of fit of our experimental raw data to the models can be viewed in the supplementary sheet. By plotting all growth profiles in the same figure (Fig. 4A), the effect of temperature on cyanobacterial growth can be clearly illustrated, with the lowest final DBC and the longest exponential phase attained at 40 and 25 °C, respectively. From Fig. 4B, the relationship between the maximum specific growth rate and the temperature can be seen to be virtually linear with every 10 °C rise in temperature resulting in an approximately doubled specific growth rate (this is in fact a typical Arrhenius type correlation found for many chemical and biochemical reactions), whilst that of the DBC displayed a strong parabolic profile with a plateau close to the value of 2.5 g L−1 within the range of 28–35 °C. As previous experimental work on the continuously-illuminated batch growth of Cyanothece 51142 normally used a fixed temperature of 30 °C [46,49], whilst varying light intensity and quality, we experienced difficulty in comparing our results against any benchmark. Nevertheless, our observed maximum specific growth rate at 30 °C is 0.032 h−1, whereas the rates from Feng and Sinetova's publications [10,46] are 0.025 and 0.052 h−1, respectively. We reason that our rate is slightly higher than Feng's due to our use of marginally higher irradiance (69 μE m−2 s−1 versus 46 μE m−2 s− 1). Nevertheless, it is lower than Sinetova's, which is probably due to their use of red (λmax ~ 627 nm) light emitting diodes, whose wavelength overlaps with absorption wavelengths of chlorophyll a (650–700 nm) and phycocyanin (620 nm), the two light harvesting pigments within cyanobacteria. In order to determine the optimum temperature, the maximum biomass productivity, which is defined as the product of maximum specific growth rate and maximum DBC, was utilised as an indicator to locate P. Dechatiwongse et al. / Algal Research 5 (2014) 103–111 109 Fig. 3. Growth and nutrient kinetics of Cyanothece 51142, cultivated at different irradiances (23, 46, 92, 138, 207, 275 and 320 μE m−2 s−1), 35 °C, 20 ml min−1 of sterile 10% volume CO2 volume air−1 and 1.5 g NaNO3 L−1 in ASP2 medium. (A) Light intensity effect – growth profile: maximum DBC and μ max are enhanced with an increasing irradiance up to 138 μ E m−2 s−1, where the light-saturation effect starts to take place. (B) Light intensity effect – rmax: as the culture concentration increases, there is a greater demand of the NO−1 3 substrate. As a result, the nutrient profile shows a similar, but inverse form to the growth profile. (C) Aiba model – μ max: a non-linear relationship between μ max and irradiance was well-described by an Aiba model, with an estimated μ max,s of 0.186 h−1. (D) Aiba Model – rmax: the light-saturated value of rmax was estimated to be 0.181 h−1. The effect of photoinhibition was also found to be insignificant up to 320 μE m−2 s−1. the best conditions. By plotting the calculated productivity against the temperature (Fig. 4C), the parabolic relationship between these two parameters is illustrated, with the maximum value close to 0.1 g L−1 h−1 or 2.4 g L−1 day− 1 within the temperature range of 31–35 °C. The lowest productivities (0.04 g L−1 h−1) were equally observed at both the lower and upper limit of our studied temperature range i.e. 25 and 40 °C. Despite resulting in almost the same productivities, the major difference here is that the growth of the cyanobacterial culture cultivated Table 2 Growth and uptake kinetics data under different studied temperatures. Temperature (°C) μmax (h−1) Maximum DBC (g L−1) Maximum productivity (g L−1 h−1) 25 30 32 35 37 40 0.023 0.032 0.037 0.036 0.047 0.065 1.8 2.6 2.6 2.5 2.0 0.6 0.042 0.085 0.095 0.100 0.092 0.041 at 25 °C experienced a significantly longer lag phase and subsequently was still progressing in the exponential phase with no sign of entering the stationary phase, whereas the stationary phase with a final DBC of 0.5 g L− 1 was attained at 40 °C much earlier – only 75 h after inoculation – thereby displaying the fastest specific growth rate. Regarding the low biomass productivity observed at 25 °C (50% below peak value), there could be two explanations: (i) suppressed repair cycle of protein D1, which can lead to a lower PSII activity and subsequently a photoinhibition scenario [32,33], and (ii) slow rate of cellular nutrient uptake [31,51]. However, since we observed a steady pO2 profile, with eventual saturation at 34%O2 saturation (Fig. 4D), this implies an insignificant photoinhibition effect and thus the former possibility can be disregarded. On the other hand, the O2 level of the culture cultivated at 40 °C was seen to sharply decrease by more than 50% of its initial value within 80 h after the inoculation (Fig. 4D), dissimilar to that obtained at 37 °C, which eventually levelled off at 27% O2 saturation with again an expected circadian-periodic character [7]. Fig. 4D also displays a high degree of coherence between our pO2 and pH experimental results, as the 110 P. Dechatiwongse et al. / Algal Research 5 (2014) 103–111 Fig. 4. Growth and nutrient kinetics of Cyanothece 51142, cultivated at different temperature (25, 30, 32, 35, 37 and 40 °C), 69 μE m−2 s−1, 20 ml min−1 of sterile 10% volume CO2 volume air−1 and 1.5 g NaNO3 L−1 in ASP2 medium. (A) Temperature effect – maximum DBC: the final biomass concentration is enhanced with an increasing temperature up to 40 °C, where the temperature-induced photoinhibition effect starts to take place. (B) Temperature effect – μ max: as in a typical Arrhenius type correlation, every 10 °C rise in temperature resulting in an approximately doubled μ max. (C) Temperature effect – maximum productivity: a temperature range of 31–35 °C was found to be an optimal condition to cultivate the cyanobacterium Cyanothece 51142. (D) Temperature effect – pO2 and pH: at 40 °C a temperature-induced photoinhibitive is confirmed by a significant reduction in photosynthetic carbon fixation activity of the cyanobacterium, decreasing profiles of pH and pO2. decreasing photosynthetic activity commonly leads to an increasing level of unfixed dissolved CO2, which in turn causes the culture to become more acidic. The higher acidity (lower pH) can be clearly seen for the culture cultivated at 40 °C. The significant reduction of photosynthetic activities at 40 °C is most likely due to the high temperature-induced detachment of the Mn-stabilising extrinsic 33 kDa protein from the PS II core complex [34]. A similar reduction in photosynthetic activity induced by thermal stress has also been previously reported in a mesophilic cyanobacterium, Synechocystis sp. PCC6803 [31,52]. With greater degree of damage to the photosynthetic apparatus, the cyanobacterium is incapable of utilising the same amount of solar energy, thus becoming more susceptible to undergoing photoinhibition at a much lower light intensity. Our light study (Section 3.2) clearly demonstrates that there is no photoinhibition up to 320 μE m− 2 s− 1 at an optimal cultivation temperature of 35 °C, whereas photoinhibition is prominent at 40 °C for illumination as low as 69 μE m−2 s−1. conditions of light intensity and temperature provides a greater understanding of the correlation between these extrinsic parameters and the corresponding growth dynamics. By employing a real-time systemmonitoring bioreactor together with various means of quantitative analysis, we were able to locate the optimal conditions for maximum growth/productivity and minimal photoinhibition. Specifically, no photoinhibition was observed up to the light intensity of 320 μE m−2 s−1. Temperature-induced photoinhibition was observed at a cultivation temperature of 40 °C, which causes severe damage to the photosynthetic apparatus and, consequently lowers the light intensity threshold for photoinhibition to occur. Maximum biomass productivity of approximately 0.1 g L−1 h−1 was observed when the culture temperature was maintained between 31 and 35 °C. It is anticipated that this productivity can be increased further with increasing light intensity beyond 320 μE m−2 s−1. Acknowledgement 4. Conclusions An improvement in the cyanobacterial solar-to-H2 conversion efficiency would play a key role in transforming this sustainable gas production process from laboratory to commercial scale. As photoinhibition has been reported to be the potential cause for a marked decrease in biomass and subsequent H2 productivity, our study of the photoautotrophic growth of cyanobacterium Cyanothece sp. ATCC 51142 under different Pongsathorn Dechatiwongse is supported by a scholarship from the Royal Thai Government, Thailand. Solar Hydrogen Project was funded by the UK Engineering and Physical Sciences Research Council (EPSRC), project reference EP/F00270X/1. We would like to thank Dr. Fessehaye W. Zemichael for his technical support in the design of experiments and Chonlathep Usaku for his valuable discussions. P. Dechatiwongse et al. / Algal Research 5 (2014) 103–111 Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.algal.2014.06.004. References [1] K.J. Reddy, et al., Unicellular, aerobic nitrogen-fixing cyanobacteria of the genus cyanothece, J. Bacteriol. 175 (5) (1993) 1284–1292. [2] A. Bandyopadhyay, et al., High rates of photobiological H-2 production by a cyanobacterium under aerobic conditions, Nat. Commun. 1 (2010). [3] H. Min, L.A. Sherman, Hydrogen production by the unicellular, diazotrophic cyanobacterium Cyanothece sp strain ATCC 51142 under conditions of continuous light, Appl. Environ. Microbiol. 76 (13) (2010) 4293–4301. [4] J. Rupprecht, et al., Perspectives and advances of biological H-2 production in microorganisms, Appl. Microbiol. Biotechnol. 72 (3) (2006) 442–449. [5] P. Tamagnini, et al., Cyanobacterial hydrogenases: diversity, regulation and applications, FEMS Microbiol. Rev. 31 (6) (2007) 692–720. [6] N.J. Skizim, et al., Metabolic pathways for photobiological hydrogen production by nitrogenase- and hydrogenase-containing unicellular cyanobacteria Cyanothece, J. Biol. Chem. 287 (4) (2012) 2777–2786. [7] M.A. Schneegurt, D.M. Sherman, S. Nayar, L.A. Sherman, Oscillating behavior of carbohydrate granule formation and dinitrogen fixation in the cyanobacterium Cyanothece sp. strain ATCC 51142, J. Bacteriol. 176 (6) (1994) 1586–1597. [8] J.r Toepel, et al., Transcriptional analysis of the unicellular, diazotrophic cyanobacterium Cyanothece sp. ATCC 51142 grown under short day/night cycles, J. Phycol. 45 (3) (2009) 610–620. [9] M.A. Schneegurt, et al., Metabolic rhythms of a diazotrophic cyanobacterium, Cyanothece sp strain ATCC 51142, heterotrophically grown in continuous dark, J. Phycol. 36 (1) (2000) 107–117. [10] X.Y. Feng, et al., Mixotrophic and photoheterotrophic metabolism in Cyanothece sp ATCC 51142 under continuous light, Microbiology 156 (2010) 2566–2574. [11] M.A. Schneegurt, D.M. Sherman, L.A. Sherman, Growth, physiology, and ultrastructure of a diazotrophic cyanobacterium, Cyanothece sp. strain ATCC 51142, in mixotrophic and chemoheterotrophic cultures, J. Phycol. 33 (4) (1997) 632–642. [12] M.A. Schneegurt, et al., Metabolic rhythms of a diazotrophic cyanobacterium, Cyanothece sp strain ATCC 51142, heterotrophically grown in continuous dark, J. Phycol. 36 (1) (2000) 107–117. [13] H. Masukawa, et al., Genetic engineering of cyanobacteria to enhance biohydrogen production from sunlight and water, Ambio 41 (2012) 169–173. [14] L. Giannelli, A. Scoma, G. Torzillo, Interplay between light intensity, chlorophyll concentration and culture mixing on the hydrogen production in sulfur-deprived Chlamydomonas reinhardtii cultures grown in laboratory photobioreactors, Biotechnol. Bioeng. 104 (1) (2009) 76–90. [15] S.A. Markov, P.F. Weaver, M. Seibert, Spiral tubular bioreactors for hydrogen production by photosynthetic microorganisms - Design and operation, Appl. Biochem. Biotechnol. 63–5 (1997) 577–584. [16] A. Melis, Solar energy conversion efficiencies in photosynthesis: minimizing the chlorophyll antennae to maximize efficiency, Plant Sci. 177 (4) (2009) 272–280. [17] D. Dutta, et al., Hydrogen production by Cyanobacteria, Microb. Cell Factories 4 (2005). [18] P.C. Hallenbeck, M. Abo-Hashesh, D. Ghosh, Strategies for improving biological hydrogen production, Bioresour. Technol. 110 (2012) 1–9. [19] K. Srirangan, M.E. Pyne, C.P. Chou, Biochemical and genetic engineering strategies to enhance hydrogen production in photosynthetic algae and cyanobacteria, Bioresour. Technol. 102 (18) (2011) 8589–8604. [20] E. Eroglu, A. Melis, Photobiological hydrogen production: recent advances and state of the art, Bioresour. Technol. 102 (18) (2011) 8403–8413. [21] F.S. Chapin, P.A. Matson, P.M. Vitousek, Principles of Terrestrial Ecosystem Ecology, Second ed. Springer Science+Business Media, LLC, New York, USA, 2011. [22] U.G. Johan, Mass Production of Microalgae at Optimal Photosynthetic Rates, in: Z. Dubinsky (Ed.), Photosynthesis, In Tech, 2013. [23] B. Andersson, S. Styring, Photosystem-II-Molecular-organization, function, and acclimation, Curr. Top. Bioenerg. 16 (1991) 1–81. [24] J.H. Kim, J.A. Nemson, A. Melis, Photosystem-II reaction-center damage and repair in dunaliella-salina (green-alga) - analysis under physiological and irradiance-stress conditions, Plant Physiol. 103 (1) (1993) 181–189. [25] I. Baroli, A. Melis, Photoinhibition and repair in Dunaliella salina acclimated to different growth irradiances, Planta 198 (4) (1996) 640–646. 111 [26] L. Martinez, A. Moran, A. Isabel Garcia, Effect of light on Synechocystis sp and modelling of its growth rate as a response to average irradiance, J. Appl. Phycol. 24 (1) (2012) 125–134. [27] S.B. Powles, Photoinhibition of photosynthesis induced by visible light, Annu. Rev. Plant Physiol. Plant Mol. Biol. 35 (1984) 15–44. [28] A. Konopka, T.D. Brock, Effect of temperature on blue-green-algae (cyanobacteria) in Lake Mendota, Appl. Environ. Microbiol. 36 (4) (1978) 572–576. [29] W.W. Lester, M.S. Adams, A.M. Farmer, Effects of light and temperature on photosynthesis of the nuisance alga Cladophora glomerata (L.) kutz from Green Bay, Lake Michigan, New Phytol. 109 (1) (1988) 53–58. [30] G. Torzillo, A. Vonshak, Effect of light and temperature on the photosynthetic activity of the cyanobacterium Spirulina platensis, Biomass Bioenergy 6 (5) (1994) 399–403. [31] N. Inoue, et al., Acclimation to the growth temperature and the high-temperature effects on photosystem II and plasma membranes in a mesophilic cyanobacterium, Synechocystis sp PCC6803, Plant Cell Physiol. 42 (10) (2001) 1140–1148. [32] Z. Gombos, H. Wada, N. Murata, The recovery of photosynthesis from lowtemperature photoinhibition is accelerated by the unsaturation of membrane lipids: a mechanism of chilling tolerance, Proc. Natl. Acad. Sci. U. S. A. 91 (19) (1994) 8787–8791. [33] E. Kanervo, et al., Membrane lipid unsaturation modulates processing of the photosystem II reaction-center protein D1 at low temperatures, Plant Physiol. 114 (3) (1997) 841–849. [34] Y. Yamane, et al., Effects of high temperatures on the photosynthetic systems in spinach: oxygen-evolving activities, fluorescence characteristics and the denaturation process, Photosynth. Res. 57 (1) (1998) 51–59. [35] A. Richmond, et al., Quantitative assessment of the major limitations on productivity ofSpirulina platensis in open raceways, J. Appl. Phycol. 2 (3) (1990) 195–206. [36] A. Vonshak, R. Guy, Photoadaptation, photoinhibition and productivity in the bluegreen alga, Spirulina platensis grown outdoors, Plant Cell Environ. 15 (5) (1992) 613–616. [37] Sartorius-PBR, Available from: http://www.sartorius.com/en/products/bioreactorsfermentors/autoclavable-benchtop/biostat-pbr-2s/ . [38] L. Provasoli, J.J.A. McLaughlin, M.R. Droop, The development of artificial media for marine algae, Arch. Mikrobiol. 25 (4) (1957) 392–428. [39] V. Lemesle, L. Mailleret, A mechanistic investigation of the algae growth "Droop" model, Acta Biotheor. 56 (1–2) (2008) 87–102. [40] B. Tamburic, et al., A novel nutrient control method to deprive green algae of sulphur and initiate spontaneous hydrogen production, Int. J. Hydrog. Energy 37 (11) (2012) 8988–9001. [41] E.M. Grima, et al., Photobioreactors: light regime, mass transfer, and scaleup, J. Biotechnol. 70 (1–3) (1999) 231–247. [42] S. Aiba, Growth kinetics of photosynthetic microorganisms, Microbial Reactions, Springer Berlin Heidelberg, 1982, pp. 85–156. [43] L. Tomaselli, The Microalgal Cell, in: A. Richmond (Ed.), Handbook of Microalgal Culture: Biotechnology and Applied Phycology, Blackwell Publishing, Oxford, UK, 2004. [44] T. Cakmak, et al., Differential effects of nitrogen and sulfur deprivation on growth and biodiesel feedstock production of Chlamydomonas reinhardtii, Biotechnol. Bioeng. 109 (8) (2012) 1947–1957. [45] S. Nigam, M.P. Rai, R. Sharma, Effect of nitrogen on growth and lipid content of Chlorella pyrenoidosa, Am. J. Biochem. Biotechnol. 7 (3) (2011) 124–129. [46] M.A. Sinetova, et al., On the dynamics and constraints of batch culture growth of the cyanobacterium Cyanothece sp. ATCC 51142, J. Biotechnol. 162 (1) (2012) 148–155. [47] M.E. Clark, Q. He, Z. He, K.H. Huang, E.J. Alm, X.F. Wan, T.C. Hazen, A.P. Arkin, J.D. Wall, J.Z. Zhou, M.W. Fields, Temporal transcriptomic analysis as Desulfovibrio vulgari Hildenborough transitions into stationary phase during electron donor depletion, Appl. Environ. Microbiol. 72 (2006) 5578–5588. [48] G. Markou, D. Georgakakis, Cultivation of filamentous cyanobacteria (blue-green algae) in agro-industrial wastes and wastewaters: a review, Appl. Energy 88 (10) (2011) 3389–3401. [49] C. Posten, Design principles of photo-bioreactors for cultivation of microalgae, Eng. Life Sci. 9 (3) (2009) 165–177. [50] M. Janssen, Cultivation of microalgae: effect of light/dark cycles on biomass yield, Wageningen University, Wageningen, The Netherlands, 2002. [51] D.S. Reay, et al., Temperature dependence of inorganic nitrogen uptake: reduced affinity for nitrate at suboptimal temperatures in both algae and bacteria, Appl. Environ. Microbiol. 65 (6) (1999) 2577–2584. [52] R. Aminaka, et al., Acclimation to the growth temperature and thermosensitivity of photosystem II in a mesophilic cyanobacterium, Synechocystis sp PCC6803, Plant Cell Physiol. 47 (12) (2006) 1612–1621.