Coupling of Solar Energy to Hydrogen Peroxide Production in the ...

Vol. 55, No. 2

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Feb. 1989, p. 483-487

0099-2240/89/020483-05$02.OO/O Copyright © 1989, American Society for Microbiology

Coupling of Solar Energy to Hydrogen Peroxide Production in the Cyanobacterium Anacystis nidulans MERCEDES RONCEL, JOSE A. NAVARRO, AND MIGUEL A. DE LA ROSA* Instituto de Bioquimica Vegetal y Fotosintesis, Facultad de Biologia, Apartado 1113, Universidad de Sevilla Superior de Investigaciones Cientfficas, 41080 Seville, Spain

y

Consejo

Received 4 August 1988/Accepted 16 November 1988

Hydrogen peroxide production by blue-green algae (cyanobacteria) under photoautotrophic conditions is of as a model system for the bioconversion of solar energy. Our experimental system was based on the photosynthetic reduction of molecular oxygen with electrons from water by Anacystis nidulans 1402-1 as the biophotocatalyst and methyl viologen as a redox intermediate. It has been demonstrated that the metabolic conditions of the algae in their different growth stages strongly influence the capacity for hydrogen peroxide photoproduction, and so the initial formation rate and net peroxide yield became maximum in the mid-log phase of growth. The overall process can be optimized in the presence of certain metabolic inhibitors such as iodoacetamide and p-hydroxymercuribenzoate, as well as by permeabilization of the cellular membrane after drastic temperature changes and by immobilization of the cells in inert supports such as agar and alginate. great interest

The production of fuels and chemicals from water by sunlight is a particularly attractive means for the bioconversion of solar energy to a valuable renewable resource. In fact, biophotolysis of water by algal and plant systems to produce hydrogen gas on a continuous basis has become a challenging problem for those involved in biological energy production (12, 13). Notwithstanding, the biological production of other energy-rich compounds such as hydrogen peroxide-widely used in pharmacy and chemistry, and a powerful source of energy-is also worth consideration (4-6, 13, 19, 20). Once electrons become photoexcited at the end of the electron transport chain, they have enough energy to reduce not only protons to hydrogen but also oxygen to hydrogen peroxide (14). The classical experiments of Mehler in 1951 (18) first demonstrated the production of H202 by isolated chloroplasts in the light. In subcellular preparations the role of molecular oxygen as a Hill oxidant and production of H202 have been well established, even in the absence of autooxidizable additives (2, 10). In recent papers (4, 5, 19, 20), we have reported the photosynthetic production of H202 by isolated spinach thylakoids, either free or immobilized in an inert support, in a reaction catalyzed by riboflavin, which appears to accept electrons at or near the terminal acceptor of photosystem I. On the other hand, Patterson and Myers in 1973 (23) were the first to report the photosynthetic production of H202 by the cyanobacterium Anacystis nidulans TX-20, although it was noted that two other blue-green algae had failed to show such production of H202. Soon afterwards, Stevens et al. (28) carried out a more extensive survey of cyanobacteria isolated from a variety of sources which revealed that H202 production is widespread among these microorganisms, although not always with the same pattern (kinetics) of production seen in TX-20. This paper deals with H202 photoproduction by free and immobilized cells of Anacystis nidulans 1402-1 as an attractive means for the bioconversion of solar radiation into storable chemical energy. *

MATERIALS AND METHODS A. nidulans 1402-1 (Synechococcus leopoliensis; Gottingen University culture collection) was grown autotrophically under continuous fluorescent illumination (25 W/m2) at 40°C in a stream of 5% (vol/vol) CO2 in air on the synthetic medium of Guerrero et al. (11) supplemented with 12 mM NaHCO3. Cells grown in this manner were used routinely. For experiments, harvested cell suspensions at known chlorophyll concentrations were centrifuged at 3,300 x g for 10 min and suspended in medium C-m containing only the

major salts (23). Continuous recording of H202 photoproduction by free algal cells was performed by the procedure of Patterson and Myers (23), which is based on the H202-dependent, peroxidase-catalyzed quenching of scopoletin luminescence emission at 460 nm when excited at 388 nm (24). The basic reaction system consisted of 1 ml of a cell suspension in C-m medium (9 jig of total chlorophyll) and 2 ml of 50 mM phosphate buffer, pH 7.5, to which 20 ,ul of scopoletin (0.2 mg/ml) and 20 [lI of horseradish peroxidase (100 U/ml) solutions were added. After 15 min of 02 bubbling in the dark, the H202-forming photochemical reaction started after irradiation of the algal cells with orange light (X >520 nm) at 600 microeinsteins m-2 s-. The cuvette temperature did not rise significantly above room temperature (24 to 26°C). The experimental setup for H202 photoproduction by immobilized algal cells, on the basis of a continuous-flow system, was that described by De la Rosa et al. (5). The reservoir and the reaction cell were filled with 50 mM phosphate buffer, pH 7.5, supplemented with 1 mM methyl viologen. The reaction cell (total volume, 40 ml) also contained immobilized algal cells (total chlorophyll content, 120

,ug). The reservoir solution was bubbled with pure oxygen at a flow rate of 0.6 liter/min, while the reaction cell was irradiated with orange light at 600 microeinsteins m-2 S-1. The buffer flow was 22 ml/h. Quantitation of the H202 so formed by immobilized cells was based on H202-dependent NADH oxidation catalyzed by NADH peroxidase (5). Portions (0.1 to 1 ml) were withdrawn from the algal suspension and added to 0.25 mM NADH in 125 mM phosphate buffer, pH 7.5 (4 ml, final volume). The total contents of the test tubes were mixed well and transferred at equal volumes into

Corresponding author. 483

484

RONCEL ET AL.

APPL. ENVIRON. MICROBIOL.

TABLE 1. Characterization of the H202 photoproduction system'J H,02 produced

400

80

300

60

Component (concn)

(pLmol/mg of chlorophyll per h) Basic ......................................... 56 Basic minus cells ......................................... ND* Basic minus light ......................................... 1 Basic with air bubbling ........................................ 32 Basic with N2 bubbling ........................................ ND Basic plus DCMU' (10 p.M) ................................. ND Basic plus methyl viologen (100 FiM) ........... .......... 110 Basic plus iodoacetamide (1 mM) .......................... 100 Basic plus p-hydroxymercuribenzoate (0.5 mM) 80 Basic plus D,L-glyceraldehyde (1 mM) .................... 16

-

. cmi

0

0

E

100

20 E

2

4

6

8

of H202, the typical Mehler reaction. As can also be seen from Table 1, light-dependent H202 production was completely eliminated by the addition of 3-(3,4-dichlorophenyl)1,1-dimethylurea, indicating the requirement for photosystem II in this process. The effects of culture age were investigated (Fig. 1). Under continuous illumination with white light, A. nidulans grew logarithmically for about 30 to 40 h up to a concentration of 30 ,ug of chlorophyll per ml; thereafter, light-limited linear growth was observed. Algal cells were harvested at different phases of growth, and H202 photoproduction for each sample was monitored under 02-rich atmospheres and continuous irradiation with orange light. Algal cells collected in the mid-log phase were the most efficient in producing H202 (Fig. 1). These results confirm that the age of the algal culture is a critical factor affecting the Mehler reaction. For this reason, the following experiments were carried out with algae in the mid-log phase of growth. Figure 2 illustrates how the cell density in the reaction mixture was also a critical factor affecting the relative efficiency of H202 production by algal cells. Increasing the cell density up to 4 ,ug of chlorophyll per ml resulted in an

60

30

0 _

-0

0

300

c9

-

40 E N

>W10

CY

0

.e

0

0.

0Lcm::

100

0,

cm

20 E

0

a.

OL 30 10

10

sg Chlorophyll /ml FIG. 2. Effect of cell density on H202 photoproduction. The algal sample was removed from the culture when its chlorophyll content was 6 jig/ml, centrifuged, and suspended in C-m medium; the final chlorophyll (Chl) concentration in the reaction mixture was adjusted to the indicated values.

RESULTS As shown in Table 1, the H202 photoproduction process in whole cells of A. nidulans 1402-1 was strictly dependent on light and oxygen; in fact, peroxide formation was practically negligible in the dark as well as in the absence of oxygen. These results clearly demonstrate that algal cells are able to carry out 02 photoreduction with the concomitant formation

E

o.

0

two cuvettes, which were placed in the sample and reference compartments of a Pye Unicam SP8-150 spectrophotometer. Absorbance at 340 nm was measured after addition of 0.3 U of NADH peroxidase to the reference cuvette. The procedures used for the preparation of immobilized cells were those of De la Rosa et al. (5) and Gisby et al. (9). The rates of oxygen exchange were determined with a Clark-type 02 electrode. The reaction cell contained 1 ml of cell suspension in C-m medium (9 ,ug of total chlorophyll) and 2 ml of 50 mM phosphate buffer, pH 7.5, and was irradiated with orange light at 600 microeinsteins m-2 S-1. The cuvette temperature was kept constant at 30°C. Chlorophyll content was determined as described by Mackinney (15). Protein content was estimated by the procedure of Lowry et al. as modified by Markwell et al. (16) with ovalbumin as the standard.

0

200

I

DCMU, 3-(3,4-Dichlorophenyl)-1,1-dimethylurea.

.

E 40 ' cmJ

-

a The basic reaction system was as described in Materials and Methods. The reaction mixture was bubbled with a stream of air or N2 for 15 min before irradiation as indicated. b ND, Not detected.

1000

-0 =

20 TIME

30

0 40

(h)

FIG. 1. Initial rates of H202 photoproduction by algal cells as a function of culture age. Chl, Chlorophyll.

VOL. 55, 1989

PHOTOBIOLOGICAL PRODUCTION OF H,O

0.3 AmoI 02 MV

60

485

80

0 -

2 u

60 40

N,

E

4Q0

Light on 0 CY I

I

20

0

20 E

E 0

0 10 min

Light off

I

0

1

2

3

4

TIME (h)

FIG. 3. Time course of the oxygen exchange catalyzed by methyl viologen (MV) in algal cells. Where indicated, 0.1 mM methyl viologen and 100 U of catalase were added.

increase in the total amount of hydrogen peroxide formed. However, the rate of H202 formation relative to the amount of chlorophyll present in the reaction mixture reached its highest value at about 2 to 3 p.g of chlorophyll per ml, which might well be a consequence of mutual shading of cells. In fact, illumination conditions dramatically affected H202 production by algal cells, as both the initial formation rate and final hydrogen peroxide yield varied with light intensity; illumination intensities with orange light of up to 600 microeinsteins m-2 s-' gave linear increases in H202 formation, while light saturation required photon flux densities higher than 800 microeinsteins m-2 s-1 (data not shown). The effects of several inhibitors of both the photosynthetic carbon oxidation and photosynthetic carbon reduction cycles were investigated. In their presence, ferredoxin should be relatively reduced and can favorably compete as an electron acceptor (1, 8, 17). Table 1 shows that iodoacetamide and p-hydroxymercuribenzoate significantly stimulated H202 photoproduction, while glyceraldehyde inhibited the hydrogen peroxide-forming activity of A. nidulans. Similar to its mode of action in isolated chloroplasts, methyl viologen functions as a Hill oxidant in vivo, as it is reduced by photosystem I in the light (3). When studying the dependence of H202 photoproduction on the concentration of methyl viologen added to the cell suspension, the rate of H202 formation was observed to become constant at a methyl viologen concentration higher than 0.1 mM; this concentration was then selected as a saturating concentration and used throughout our work. For comparative purposes, the stimulating effect of methyl viologen at saturating concentration is shown in Table 1. The time course of the exchange catalyzed by methyl viologen in whole cells of A. nidulans is displayed in Fig. 3. After illumination of the cells in the absence of methyl viologen, there was rapid evolution, followed by likewise rapid uptake after addition of methyl viologen. The rate of such uptake quickly decreased to zero in the dark. Subsequent addition of catalase resulted in evolution of the 02 previously photoreduced because of dismutation of the H202 formed. In order to try to remove barriers to the accumulation of methyl viologen and H202 by intact cells without grossly disrupting their structure, several mild permeabilization treatments were assayed. Repeated freezing and thawing seemed to be one of the most effective procedures. Howevolution with CO2 as the electron ever, it inactivated 02

02

02

02

02

02

FIG. 4. H202 photoproduction by agar-immobilized algal cells in a continuous-flow system. At the indicated times, the hydrogen peroxide concentration in the reaction cell (0) and the amount of H202 flowing out of the cell (0) were determined.

acceptor and also the light-induced 02 uptake in the presence of methyl viologen, but it did not damage the 02 uptake

activity of photosystem I when supplied with an artificial electron donor such as ascorbate-reduced 2,6-dichlorophenolindophenol. The H202-forming activity of intact cells was significantly enhanced by only one freeze-thaw cycle because of permeabilization of the cell membrane. Further freeze-thaw treatments resulted in an important decrease in production of H202, probably because of significant damage to cellular structures (data not shown). When whole cells of A. nidulans were immobilized in inert supports such as agar and alginate, the time course of H202 production was quite similar to that obtained with free-living cells. However, the concentration of methyl viologen in the reaction mixture had to be somewhat higher with immobilized cells because of diminished diffusion rates of the methyl viologen in the supports. A continuous-flow system resembling that described by De la Rosa et al. (5) was then devised in which the hydrogen peroxide was removed as formed. Under such conditions, a suspension containing immobilized cells forming H202 could be sustained for several hours without a significant decrease in the production rate (Fig. 4). DISCUSSION An extensive study of H202 production by A. nidulans TX-20 led Patterson and Myers (23) to the conclusions that H202 production was due to the photosynthetic generation of excess reductant and that the site of production was probably near the reducing side of photosystem I. However, an origin of H202 not directly linked to photosynthesis was proposed by Stevens et al. (28) to explain the observed production of H202 in the dark by other blue-green algae. H202 production by A. nidulans 1402-1 as described herein resulted from a strictly light-dependent process which required the active participation of photosystem II, therefore involving water oxidation. The data reported herein also indicate considerable diversity in the efficiency of H202 formation depending on culture age, which unequivocally suggests that the metabolic conditions of the cells critically affect the Mehler reaction. This is in good agreement with Stevens et al. (28), who observed three different kinetic profiles for H202 production among a

486

RONCEL ET AL.

few species of blue-green algae examined; they proposed that such kinetic profiles reflect real metabolic differences among the cells. It is a well-known fact that methyl viologen is able to accept electrons at or near the terminal acceptor of photosystem I, reduced methyl viologen in turn reducing oxygen to superoxide radicals. The final stable product of such photosystem I-driven autooxidation of reduced low-potential dyes is hydrogen peroxide, apparently derived via the dismutation of superoxide radicals (3, 7). In this paper we show that illumination of algal cells in the presence of methyl viologen actually yields increased production of hydrogen peroxide and that under appropriate conditions it is possible to demonstrate light-dependent 02 uptake attributable to the methyl viologen-catalyzed Mehler reaction. It is important to note that the saturating methyl viologen concentration for the methyl viologen-dependent H202 production in whole cells of A. nidulans 1402-1 was several times higher than that for the same reaction in spinach thylakoids (6). In accordance with Chua (3), this discrepancy can be attributed to at least two factors. First, the methyl viologen has to be present at a high enough concentration so that it can compete with the endogenous ferredoxin for electrons from photosystem I. Second, in our systems a permeability barrier is imposed by the living cells. As is well recognized, the gram-negative envelope of blue-green algae, consisting of the cell wall (outer membrane, periplasm, and peptidoglycan layer) and the cell membrane, is not freely permeable to ions (22). Therefore, in order to expose the thylakoids of the cytoplasm to the electrolytes of the suspension medium, either the cells must be fractured or their envelope must be permeabilized. Cell permeabilization to ions, on the other hand, offers the advantage of better-preserved thylakoids (22). One freezethaw cycle of whole cells of A. nidulans 1402-1 was one of the most effective procedures for cell permeabilization, significantly enhancing H202 photoproduction. Smith et al. (27) have reported, however, that about six cycles of freezing and thawing were required for maximal rates of H2 photoproduction by thermophilic blue-green algae. In both cases, complete inactivation of the photosynthetic electron transport from water to photosystem I was observed, which included multiple effects (21, 25, 27). It is worth noting that H202 photoproduction by A. nidulans 1402-1 cannot be sustained long because of inactivation of photosynthetic electron transport. In fact, the well-known Warburg effect (02 inhibition of photosynthesis) in spinach chloroplasts has been reported to be due to the photosynthetic generation of H202 (26). For this reason, we devised a continuous-flow system for immobilized cells in which the hydrogen peroxide is removed as it is formed, avoiding prolonged contact between the algal cells and H202 (5). Under such conditions, a suspension of agar- or alginateimmobilized cells forming H202 can be maintained for several hours without a significant decrease in the production rate. These results represent an important improvement in H202-forming biological systems, especially in terms of operational stability. Immobilized catalytic systems as constituents of bioreactors actually offer many advantages over homogeneous systems, and immobilized photosynthetic systems, in particular, may play an important role in the future as components of light-catalyzed chemical energy conver-

sion devices.

APPL. ENVIRON. MICROBIOL. ACKNOWLEDGMENTS We thank INTEROX Quimica SA (c/o Solvay & Cie, Brussels, Belgium) for financial support. M.R. and J.A.N. are recipients of fellowships from the Ministry of Education and Science and from the National Research Council, respectively. LITERATURE CITED 1. Behrens, P. W., T. V. Marsho, and R. J. Radmer. 1982. Photosynthetic 02 exchange kinetics in isolated soybean cells. Plant Physiol. 70:179-185. 2. Brown, A. H., and N. Good. 1955. Photochemical reduction of oxygen in chloroplast preparations and in green plant cells. I. The study of oxygen exchanges in vitro and in vivo. Arch. Biochem. Biophys. 57:340-354. 3. Chua, N. H. 1971. The methyl viologen-catalyzed Mehler reaction and catalase activity in blue-green algae and Chlamydomonas reinhardi. Biochim. Biophys. Acta 245:277-287. 4. De la Rosa, M. A., P. F. Heelis, K. K. Rao, and D. 0. Hall. 1987. Flavin-mediated hydrogen peroxide production by biological and chemical photosystems, p. 597-600. In D. E. Edmonson and D. B. McCormick (ed.), Flavins and flavoproteins. Walter de Gruyter & Co., Berlin. 5. De la Rosa, M. A., K. K. Rao, and D. 0. Hall. 1986. Hydrogen peroxide photoproduction by free and immobilized spinach thylakoids. Photobiochem. Photobiophys. 11:173-187. 6. De la Rosa, M. A., M. Roncel, and J. A. Navarro. 1988. Hydrogen peroxide photoproduction by biological and chemical systems, p. 233-240. In D. 0. Hall and G. Grassi (ed.), Photocatalytic production of energy-rich compounds. Elsevier Applied Science Publishers, London. 7. Elstner, E. F., and D. Frommeyer. 1978. Production of hydrogen peroxide by photosystem II of spinach chloroplast lamellae. FEBS Lett. 86:143-146. 8. Furbank, R. T., M. R. Badger, and C. B. Osmond. 1982. Photosynthetic oxygen exchange in isolated chloroplasts and cells of C3 plants. Plant Physiol. 70:927-931. 9. Gisby, P. E., K. K. Rao, and D. 0. Hall. 1987. Entrapment techniques for chloroplasts, cyanobacteria and hydrogenases. Methods Enzymol. 135:440-454. 10. Good, N., and R. Hill. 1955. Photochemical reduction of oxygen in chloroplast preparations. II. Mechanism of the reaction with oxygen. Arch. Biochem. Biophys. 57:355-366. 11. Guerrero, M. G., C. Manzano, and M. Losada. 1974. Nitrite photoreduction by a cell-free preparation of Anacystis nidulans. Plant Sci. Lett. 3:273-278. 12. Hall, D. O., D. A. Affolter, M. Brouers, D. J. Shi, L. W. Yang, and K. K. Rao. 1985. Photobiological production of fuels and chemicals by immobilized algae. Annu. Proc. Phytochem. Soc. Eur. 26:161-185. 13. Hall, D. O., M. Brouers, H. de Jong, M. A. De la Rosa, K. K. Rao, D. J. Shi, and L. W. Yang. 1987. Immobilized photosynthetic systems for the production of fuels and chemicals. Photobiochem. Photobiophys. Suppl. 167-180. 14. Losada, M. 1979. Photoproduction of ammonia and hydrogen peroxide. Photobiochem. Photobiophys. 6:205-225. 15. Mackinney, G. 1941. Absorption of light by chlorophyll solutions. J. Biol. Chem. 140:315-322. 16. Markwell, M. A. K., S. M. Hass, L. L. Bieber, and N. E. Tolbery. 1978. A modification of the Lowry procedure to simplify protein determination in membrane and lipoprotein samples. Anal. Biochem. 87:206-210. 17. Marsho, T. V., P. W. Behrens, and R. J. Radmer. 1979. Photosynthetic oxygen reduction in isolated intact chloroplasts and cells from spinach. Plant Physiol. 64:656-659. 18. Mehler, A. H. 1951. Studies on reactions of illuminated chloroplasts. I. Mechanism of the reduction of oxygen and other Hill reagents. Arch. Biochem. Biophys. 33:65-77. 19. Navarro, J. A., M. Roncel, F. F. De la Rosa, and M. A. De la Rosa. 1987. Light-driven hydrogen peroxide production as a way to solar energy conversion. Bioelectrochem. Bioenerg. 18:71-78. 20. Navarro, J. A., M. Roncel, and M. A. De la Rosa. 1988. Biological and chemical photoproduction of hydrogen peroxide,

VOL. 55, 1989

21.

22.

23. 24.

p. 153-156. In G. Moreno, R. H. Pottier, and T. G. Truscott (ed.), Photosensitisation: molecular, cellular and medical aspects. North Atlantic Treaty Organization ASI Series, vol. H15. Springer-Verlag, Berlin. Ono, T. A., and N. Murata. 1981. Chilling susceptibility of the blue-green alga Anacystis nidulans. II. Stimulation of the passive permeability of cytoplasmic membrane at chilling temperatures. Plant Physiol. 67:182-187. Papageorgiu, G. C., and T. Lagoyanni. 1985. Photosynthetic properties of rapidly permeabilized cells of the cyanobacterium Anacystis nidulans. Biochim. Biophys. Acta 807:230-237. Patterson, C. 0. P., and J. Myers. 1973. Photosynthetic production of hydrogen peroxide by Anacystis nidulans. Plant Physiol. 51:104-109. Perschke, H., and E. Broda. 1961. Determination of very small amounts of hydrogen peroxide. Nature (London) 190:257-258.

PHOTOBIOLOGICAL PRODUCTION OF H202

487

25. Rao, V. S. K., J. J. Brand, and J. Myers. 1977. Cold shock syndrome in Anacystis nidulans. Plant Physiol. 59:965-969. 26. Robinson, J. M., M. G. Smith, and M. Gibbs. 1980. Influence of hydrogen peroxide upon carbon dioxide photoassimilation in the spinach chloroplasts. I. Hydrogen peroxide generated by broken chloroplasts in an "intact" chloroplast preparation is a causal agent of the Warburg effect. Plant Physiol. 65;755-759. 27. Smith, G. D., A Muallem, and D. 0. Hall. 1982. Hydrogenasecatalyzed photoproduction of hydrogen by photosystem I of the thermophilic blue-green algae Mastigocladus laminosus and Phormidium laminosum. Photobiochem. Photobiophys. 4:307319.

28. Stevens, S. E., Jr., C. 0. P. Patterson, and J. Myers. 1973. The production of.hydrogen peroxide by blue-green algae: a survey. J. Phycol. 9:427-430.