Two-Stage System for Hydrogen Production by Immobilized Cyanobacterium Gloeocapsa alpicola CALU 743

Biotechnol. Prog. 2007, 23, 1106−1110

1106

Two-Stage System for Hydrogen Production by Immobilized Cyanobacterium Gloeocapsa alpicola CALU 743 Larissa T. Serebryakova* and Anatoly A. Tsygankov Institute of Basic Biological Problems, RAS, Pushchino, Moscow Region, 142290, Russia

Previous studies showed that cell suspensions of unicellular nondiazotrophic cyanobacterium G. alpicola grown under nitrate-limiting conditions intensively produces H2 via fermentation of endogenous glycogen with hydrogen yield more then 90% of theoretical maximum (3.8 mol H2 per mol glucose). H2 production is realized by a Hox hydrogenase on the stages of NAD(P)H generation. Exploiting this property, the two-stage cyclic system for sustained hydrogen production was developed using a photobioreactor (PhBR) with G. alpicola immobilized on glass fiber TR-0.3. Immobilization of the cells on the matrix occurred during growth directly in PhBR operated in continuous mode; the density of culture immobilized achieved 37 g Chl R cm-2. The first stage of the cycle was the photosynthetic incubations of G. alpicola in the flow of the culture medium, which contained limiting concentrations of nitrate for efficient glycogen accumulation and activation of hydrogenase. The second stage was the fermentation of glycogen, with H2 production realized in darkness with continuous Ar sparging and without medium flow. Standardization of optimal parameters for both stages provided a stable cyclic regime of the system: photosynthesis (24 hours)-fermentation (24 hours). The total amount of H2 evolved in one cycle was 957.6 mL L-1matrix, and the overage rate of H2 production during the cycle (48 hours) was about 20 mL h-1 L-1matrix. Ten consequent cycles was carried out in this regime with reproducible H2 production, although PhBR with the same sample of immobilized culture was operated over a period of more then three months.

1. Introduction Global awareness of the growing energy demand depleting conventional sources and the increasing environmental pollution has resulted in the development of alternative energy sources. The concept of using biological hydrogen evolution for the production of fuel has received considerable attention and has become a subject of research for many scientists working in the field of hydrogen metabolism of various microorganisms. Phototrophic microorganisms, especially cyanobacteria and microalgae, are considered the most perspective candidates for future applications, because they use the sunlight as the only source of energy. Cyanobacteria, phototrophic O2-evolving prokaryotes, produce H2 in two physiological processes: in the nitrogen fixation catalyzed by nitrogenase (it is mainly concerned heterocystous cyanobacteria) and in the fermentation involving reversible hydrogenase (Hox hydrogenase (1, 2). Light-dependent nitrogenase-mediated H2 production is investigated in more details; conditions and factors affecting this process are well studied and optimized (3, 4). Maximum rates of H2 evolution were observed with mutant forms of cyanobacteria lacking uptake hydrogenase activity. It was demonstrated that under microaerobic conditions Anabaena Variabilis PK84 evolved H2 at the rate of 14 mL L-1 h-1 and was able to perform this process even in air (5). Higher rates of H2 production (about 22 mL L-1 h-1) was achieved in experiments with intensive cultures of Anabaena Variabilis PK84 and an efficiency of light energy utilization in this case reached 1.4% (6). In spite of the fact that H2 production by cyanobacteria via fermentation is a relatively low-yield process, it is also regarded * To whom correspondence should be addressed. Email: sereb@ issp.serpukhov.su. 10.1021/bp070168p CCC: $37.00

as one of the approaches for practical biohydrogen production (7). In particular, it concerns the cultures growing under conditions of some nutrient deficiency in the medium when photosynthesis is directed basically on an accumulation of storage compounds serving as substrates for the fermentation. Previously we have reported that unicellular nondiazotrophic cyanobacterium Gloeocapsa alpicola CALU 743 grown under nitrate limitation possessed high hydrogenase activity and intensively produced H2 during fermentation of endogenous glycogen (8-10). The content of stored glycogen under these conditions reached 40-50% of dry weight, and its degradation via the Empden-Meyerhof-Parnas pathway resulted in a hydrogen yield of 90% of theoretical maximum (1glycogenglucose f 4H2+ 2acetate + 2CO2). NAD(P)H are primary electron donors for Hox hydrogenase, and therefore, H2 production can be effective at low H2 pressure in the reaction zone. We managed to show H2 production by suspension of nitratelimited G. alpicola cells in the darkness at 30 °C in an open system (continuous bubbling with Ar at a rate of 2 L h-1 to maintain low partial pressure of H2) at a constant rate of 6.3 mL h-1 per L of suspensions over a period of 7-8 h. Then the rate of H2 production lowered. G. alpicola cells grown at sulfur deficiency in the medium were shown to evolve H2 in considerable amounts during fermentation as well (11). In general, H2 production via fermentation with photosynthetic stored glycogen as a substrate may be considered as a twostage process of the light energy transformation, and the cyclic functioning of this process should implement a sustained H2 production. To evaluate this possibility, it is necessary to determine conditions for (i) effective storage of glycogen in cyanobacterial culture in the light with simultaneous hydrogenase activation and (ii) production of hydrogen in the dark with

© 2007 American Chemical Society and American Institute of Chemical Engineers Published on Web 08/11/2007

Biotechnol. Prog., 2007, Vol. 23, No. 5

the subsequent transition of culture to the following cycle of glycogen accumulation in the light. In addition, an optimization of hydrogen production in the dark is also important. For this the concentration of cells within the system should be increased without incurring light limitation. It might be achieved by uniform immobilization of the cultures on a thin translucent matrix (12). Recently we succeeded in finding an inexpensive glass-fiber cloth with a high ratio of surface to volume and in demonstrating hydrogen photoproduction using this matrix with immobilized green microalga Chlamidomonas reinhardtii (13). The aim of this study is to develop a system for sustained H2 production by immobilized G. alpicola placed in a photobioreactor (PhBR) operated in a two-stage cyclic regime “photosynthesis-fermentation”. The work included (i) immobilization of cyanobacterium, (ii) optimization of light and dark stages, and (iii) determination of conditions for the stable cyclic regime.

2. Materials and Methods 2.1. Organism and Growth Conditions. The unicellular nonnitrogen-fixing cyanobacterium Gloeocapsa alpicola CALU 743 () Synechocystis PCC 6308) was obtained from the Alga Collection of St. Petersburg University (Russia). Liquid cultures were grown in BG11 (14) supplemented with 1.5 µM NiCl2. Cultivation was performed at 30 °C in 55 mL cylindrical glass flasks with 35-40 mL of the culture and exposed to light with an intensity of 165 µE m-2 s-1. Growing cultures were bubbled (300 mL min-1) with a gas mixture of air and 2% CO2. 2.2. Immobilization, Photobioreactor (PhBR), and Operation in Two-Stage Regime. As a matrix for immobilization of cyanobacterial cells, the glass fiber TR-0.3 (“STEKLOVOLOKNO”, Gus’-Chrustalniy, Russia) was used. To inoculate, a degreased autoclaved fiber (18.3 × 9.0 × 0.1 cm3) was incubated in the flask with growing culture over a period of 1-2 days and then was aseptically transferred inside the sterile PhBR. The PhBR (13) consists of two identical glass plates held together by binder clips and vacuum grease as a sealant. The apparatus was supplied with separate gas and liquid inlets and a common outlet; partitions made of silicone tubing were fixed to each of the inner glass surfaces to enlarge the path for gas bubbling and liquid movement inside of the PhBR. A total operating volume of the PhBR is equal to 160 mL (inside dimensions, 20.0 × 10.0 × 0.8 cm3). The PhBR with inoculated glass fiber placed between glass plates was filled with BG11 and subjected to continuous liquid flow with the same medium at 35 ( 5 mL h-1 and to continuous gas mixture (air + 2% CO2) flow at 600-700 mL h-1 and operated under two-sided fluorescent light illumination (72 µE m-2 s-1 × 2) at 28 °C over a period of 7-8 days. In this period, the fiber become uniformly covered with cells on both sides and their loading on the matrix achieved 37 ( 8 µg Chl R cm-2. 2.2.1. First StagesNitrate Limitation. Nitrate limitation of immobilized G. alpicola in the first cycle was initiated by changing to BG11o (no nitrate) medium supplemented with 1.5 µM NiCl2 and 10 mM HEPES (pH 7.5) without its flow. The degree of nitrogen depletion was evaluated by the decrease of photosynthetic O2 evolution and yellowing culture due to degradation of phycobiliproteins under nitrogen-deprived conditions. In this stage of the second and following cycles, double BG11o contained 1.5 µM NiCl2 and 10 mM HEPES (pH 7.5), and limiting nitrate content (3-5 mM) at flow rate of 15 mL h-1 was used. 2.2.2. Second StagesH2 Production During Fermentation. To start the fermentation, the PhBR with immobilized nitrate-

1107

limited cultures was darkened and bubbled continuously with argon, which was enriched with H2 produced by cyanobacterium. PhBR was operated without the medium flow and at the same temperature as the previous stage temperature (28 °C). It has been established earlier that optimal pH for H2 production during fermentation was in the range 6.8-8.3 (10). To maintain the pH in this range, the incoming culture medium during previous photosynthetic stage was supplied with 10 mM HEPES, pH 7.5. The optimal rate of argon flow and duration of first and second stages were determined experimentally. 2.3. Measurements. The total Chl R content of the cells absorbed on the matrix was estimated in methanol extracts according to Mackinney (15). Extraction was performed after experiments by overnight incubation of the fiber with immobilized cells in methanol in the darkness at room temperature. It was impossible to determine glycogen content and hydrogenase activity in immobilized cells during experiments. Therefore, these parameters were assayed in the liquid cultures incubated under identical to immobilized cultures conditions. For this purpose, PhBR was filled with the cell suspensions of the Chl R content equal to that of immobilized cells and operated in identical regimes. Glycogen content was determined by the phenol sulfuric acid method (16) after its extraction from the cells and hydrolysis (17). Hydrogenase activity was determined in the reaction of reduced methyl viologen (MV)-dependent H2 evolution by amperometric method (18) using O2/H2 electrode Hansatech DWI (Hansatech Ltd., King’s Lynn, Norfolk, UK). Activity was expressed in µmoles of H2 evolved per hour per mg of protein. Protein concentration was determined by the Lowry method (19). Concentrations of O2 and H2 evolved in the stages photosynthesis and fermentation, correspondingly, were measured in outlet gas by gas chromatography with an accuracy of 10%. The rates of H2 production were calculated on the basis of these measurements.

3. Results and Discussion 3.1. Immobilization of G. alpicola. In consequence of a high natural adhesion of G. alpicola to a glass fiber cyanobacterial cells were immobilized on the matrix during growth directly in PhBR operated in continuous mode. Moderate medium (35 ( 5 mL h-1) and gas (air + CO2) flow rates (600-700 mL h-1) did not induce active mixing and allowed the cells to cover fiber surface entirely and uniformly on both sides because of two-sided illumination (Figure 1). The total density of the culture immobilized after 7 day’s of incubation under these conditions achieved 12.6 ( 3 mg Chl R per the matrix. 3.2. Stage of Photosynthesis. By virtue to the fact that effective H2 production during fermentation of endogenous glycogen was observed in the suspensions of cells grown under nitrogen limiting condition (8-10) it was needed to obtain nitrate-limited immobilized culture. For this the PhBR was filled by the medium without nitrate and its flow was stopped. Incubation in such regime resulted in the reduction of O2 concentration in outlet air (Figure 2A), evidencing the decrease of photosynthetic O2 evolution in response to nitrate depletion. Measurements made using liquid cultures incubated identically showed the development of hydrogenase activity and effective glycogen synthesis in cyanobacterial cells during this period (Figure 2B). This stage was confined to 24 h because a longer period of incubation (>30 h) induced desorption of cells from the matrix and washout of them from PhBR (visible observation and microscopic analysis of the outlet medium; data not shown). Reduction of cell adhesion to glass fiber apparently was caused

1108

Figure 1. Photobioreactor with G. alpicola cells immobilized on a glass-fiber cloth TR-0.3.

Figure 2. O2 concentration in flow gas (air + 2% CO2) from PhBR, with immobilized culture (A) and hydrogenase activity (HA) (1) and glycogen content (2) in liquid culture (B) of G. alpicola during incubation under nitrate-limiting conditions. Error bars ( 1 SD, n ) 3.

by formation mucilaginous cover around the cells as an effect of deep nitrogen depletion. It should be noted that the levels of storage glycogen and hydrogenase activity in cells after 24 hours of incubation could be sufficient to implement H2 production under dark anaerobic conditions. In the second and following

Biotechnol. Prog., 2007, Vol. 23, No. 5

Figure 3. Effect of Ar flow rate on the rate of H2 production (A) and H2 production for 24 h (B) by PhBR with immobilized G. alpicola during fermentation. The rates of Ar flow: 1-400 mL h-1; 2-600 mL h-1; 3-800 mL h-1. Error bars ( 1 SD, n ) 3.

cycles, the growth of culture in the photosynthetic stage was provided with a continuous flow of double BG11o medium (15 mL h-1), contained limiting concentrations of nitrate (3-5 mM), which is aimed to maintain culture in a nitrogen-limited but not in a nitrogen-starved situation for glycogen storage and activation of hydrogenase. Additionally, the medium flow through the PhBR prevented accumulation of acetate produced during fermentative stages. 3.3. Stage of FermentationsH2 Production. To optimize H2 production, we focused on the rate of Ar flow through PhBR. Because hydrogenase catalyzes H2 evolution in NAD(P)Hdependent reaction (10, 20), which is not thermodynamically favorable, the pressure of H2 inside of PhBR should be quite low. It was achieved by a continuous sparging of the system by argon. Maximal rate of H2 production depended on the rate of Ar flow (Figure 3A): the increase of Ar flow rate from 400 mL h-1 up to 800 mL h-1 resulted in an increase of the maximal rate of H2 evolution to more then in 1.5 times, and what is more, the duration of H2 evolution at maximal rate reached 30 h. Hence, a total volume of H2 produced by PhBR in 24 h was the most at the rate of Ar flow of 800 mL h-1 and averaged 16.4 ( 2.5 mL (Figure 3B). The concentration of H2 inside PhBR measured in outlet Ar was 0.1 ( 0.05% and could not repress the process as it follows from our previous data (10). A further increase of Ar flow rate (>1 L h-1) triggered turbulence inside PhBR and partial desorption of cells from the matrix. Taking into account these data, each stage of fermentation in the long-term operation of PhBR in cyclic regime was performed during 24 h using an Ar flow equal to 800 mL h-1. It is necessary to note that CO2 produced via fermentation was removed from PhBR by the Ar flow, which was also important for sustained fermentation and consequently for sustained H2 production. 3.4. Cyclic Regime. The graphic diagram (Figure 4) reflects the dynamics of the key parameters in the system with immobilized nitrate-limited G. alpicola during two cycles.

Biotechnol. Prog., 2007, Vol. 23, No. 5

1109

Figure 4. H2 production by immobilized G. alpicola in a two-stage system (photosynthesis-fermentation) during two cycles. (1) Rate of H2 production; (2) O2 concentration in outlet gas; (3) glycogen content (liquid culture); and (4) hydrogenase activity (HA, liquid culture). Arrows indicate the start of each stage.

The duration of the photosynthetic stage and the stage of fermentation was 24 h, that is, each cycle duration was equal to 2 days. H2 production reached the maximal rate (730 ( 50 µL h-1) in the first 4 h from the start of the fermentation stage, both in the first and in the second cycles, and was kept at the same level. The total amount of H2 evolved in one cycle was 15.8 ( 3 mL. Taking into account the volume of the matrix (16.5 mL), this value for one cycle was 957.6 ( 200 mL L-1matrix, and the overage rate of H2 production during the cycle (48 h) was about 20 mL h-1 L-1matrix. These estimates far exceeded the values obtained with liquid nitrate-limited cultures of G. alpicola (rate of H2 production of 6.3 mL h-1 L-1 over a period of 7-8 h at 30 °C with Ar bubbling at the rate of 2 L h-1) (10). Besides, decline of expenditure for Ar from 2 L h-1 (cell suspension) up to 800 mL h-1 (immobilized culture) makes the process essentially cheaper. The PhBR with immobilized nitrate-limited G. alpicola was operated in a cyclic regime over a period of 20 days, that is, 10 cycles were performed. A total of 15 mL of H2 produced per one cycle and variation between cycles was inside our methods accuracy (Figure 5), evidencing stability of the system. It means that immobilized cyanobacterial cells were able to reaccumulate endogenous glycogen during photosynthetic stages and to maintain hydrogenase activity on a high level for effective H2 release during fermentation. H2 and other products of fermentation (acetate and CO2) were constantly eliminated from the system by the medium and Ar flows to prevent their negative influence on the process.

4. Conclusions This work is the first report on H2 production by immobilized cells of cyanobacterium during fermentation of an endogenous substrate. The use of the culture of phototrophic microorganisms immobilized on the translucent matrix has been reported (12, 21) to have some advantages for the increase of H2 production over identical use of the cell suspensions. First of all, it is stipulated by the increase of biomass in the unit of volume without light limitation. Additionally, in our experiments, the cyclic regime requires the obtaining of nitrate-limited cultures in the photosynthetic stage of the cycle and the removing of the liquid products of fermentation after the dark stage of the cycle using the medium flow through the PhBR. It is precisely the culture immobilized that allows this without loss of a biomass. We have shown a simple and effective method for immobilization of the unicellular cyanobacterium G. alpicola on

Figure 5. H2 production by immobilized G. alpicola cells in PhBR operated in a two-stage cyclic regime photosynthesis-fermentation. Error bars ( 1 SD, n ) 3.

cheap translucent matrix, which has been applied in a two-stage system for bioconversion of light energy into the fuel (H2). Operating this system in a long-term cyclic regime provided relatively stable hydrogen production over a period of not less then 20 days at the overage rates of about 20 mL h-1 L-1matrix, which are comparable with those measured for heterocystous cyanobacteria and green alga (3, 4, 13). Further efforts should be directed to the estimation of the efficiency of the process and to adaptation of the system for the ambient light-dark regime.

Acknowledgment This study was supported by the program No. 7 of Russian Academy of Sciences. The authors thank Laurinavichene T.V. for valuable comments during experiments and discussion during manuscript preparation.

References and Notes (1) Madamwar, D.; Garg, N.; Shah, V. Cyanobacterial hydrogen production. J. Microbiol. Biotechnol. 2000, 16, 257-267. (2) Schutz, K.; Happe, T.; Troshina, O.; Lindblad, P.; Leitao, E.; Oliveira, P.; Tamagnini, P. Cyanobacterial H2 productionsA comparative analysis. Planta 2004, 218, 350-359. (3) Dutta, D.; De, D.; Chaudhuri, S.; Bhattacharya, S. Hydrogen production by cyanobacteria. Microbiol. Cell Fact. 2005, 4, 3647. (4) Tsygankov, A. Nitrogen-fixing cyanobacteriasH2 producers. Appl. Biochem. Microbiol. (Moscow) 2007, 43, 279-288. (5) Tsygankov, A.; Fedorov, A.; Kosourov, S.; Rao, K. Hydrogen production by cyanobacteria in an automated outdoor photobioreactor under aerobic conditions. Biotechnol. Bioeng. 2002, 80, 777-783.

Biotechnol. Prog., 2007, Vol. 23, No. 5

1110 (6) Liu, J. G.; Bukatin, V.; Tsygankov, A. An efficient light energy conversion into H2 by dense-culture system of Anabaena Variabilis mutant PK84 exposed to high light intensity. Int. J. Hydrogen Energy 2006, 9, 1591-1596. (7) Hallenbeck, P. C. Fundamentals of the fermentative production of hydrogen. Water Sci. Technol. 2005, 52, 21-29. (8) Serebryakova L.; Sheremetieva M.; Tsygankov A. Reversible hydrogenase activity of Gloeocapsa alpicola in continuous culture. FEMS Microbiol. Lett. 1998, 166, 89-94. (9) Serebryakova L.; Sheremetieva M.; Lindblad, P. Hydrogenase activity of the unicellular cyanobacterium Gloeocapsa alpicola under conditions of nitrogen starvation. Microbiology (Moscow) 1999, 68, 293-298. (10) Troshina O.; Serebryakova L.; Sheremetieva M.; Lindblad P. Production of H2 by the unicellular cyanobacterium Gloeocapsa alpicola CALU 743 during fermentation. Int. J. Hydrogen Energy 2002, 27, 1283-1289. (11) Antal, T.; Lindblad, P. J. Production of H2 by sulphur-deprived cells of the unicellular cyanobacteria Gloeocapsa alpicola and Synechocystis sp. PCC 6803 during dark incubation with methane or at various extracellular pH. Appl. Microbiol. 2005, 98, 114-120. (12) Tsygankov, A. Hydrogen Production by Suspension and Immobilized Cultures of Phototrophic Microorganisms. Technological Aspects. In Biohydrogen lll. Renewable Energy System by Biological Solar Energy ConVersion; Miyake, J., Igarashi, Y., Rogner, M., Eds.; Elsevier Ltd.: Kidlington, U.K. 2004; pp 57-71. (13) Laurinavichene, T.; Fedorov, A.; Ghirardi, M.; Seibert, M.; Tsygankov, A. Demonstration of sustained hydrogen photoproduction by immobilized, sulfur-deprived Chlamydonas reinhardtii cells. Int. J. Hydrogen Energy 2006, 31, 659-667.

(14) Rippka, R.; Deruelles, J.; Waterbury, J. B.; Herdman, M.; Stanier, R. Y. Genetic assignments, strain histories, and properties of pure cultures of cyanobacteria. J. Gen. Microbiol. 1979, 111, 161. (15) Mackinney, G. Absorption of light by chlorophyll solutions. J. Biol. Chem. 1941, 140, 315-322. (16) Dubois, M.; Gille, K. F.; Hamilton, J. K.; Rebberrs, P. A.; Smith, F. Colorimetric method for determination of sugars and related substances. Anal. Chem. 1956, 28, 350-356. (17) Ernst, A; Kirschenlohr, H.; Diez, J.; Boger, P. Glycogen content and nitrogenase activity in Anabaena Variabilis. Arch. Microbiol. 1984, 140, 120-125. (18) Oxelfeld F.; Tamagnini P.; Salema R.; Lindblad P. Hydrogen uptake in Nostoc strain PCC 73102: Effects of nickel, hydrogen, carbon and nitrogen. Plant Physiol. Biochem. 1995, 33, 617-623. (19) Lowry, O. H.; Rosenbrough, W. J.; Far, A. L. Protein measurements with the folin phenol reagent. J. Biol. Chem. 1951, 193, 265275. (20) Serebryakova, L.; Sheremetieva, M. Characterization of catalytic properties of hydrogenase isolated from the unicellular cyanobacterium Gloeocapsa alpicola CALU 743. Biochemistry (Moscow) 2006, 71, 1683-1690. (21) Fukui, S.; Tanaka, A. Immobilized Microbial Cells. Ann. ReV. Microbiol. 1982, 36, 145-172. Received May 24, 2007. Accepted July 6, 2007. BP070168P