Efforts to scale up the capacity of green algae and cyanobacteria to use sunlight to convert water into hydrogen gas for energy use
Pin-Ching Maness, Jianping Yu, Carrie Eckert, and Maria L. Ghirardi
Pin-Ching Maness and Maria L. Ghirardi are Principal Scientists, Jianping Yu is a Staff Scientist, and Carrie Eckert is a Postdoctoral Research Fellow at the Chemical and Biosciences Center, National Renewable Energy Laboratory, Golden, Colo.
*Cyanobacteria and green algae produce hydrogen gas in the dark fermentatively and in the light under photosynthetic conditions.
*Hydrogenase enzymes catalyze hydrogen production in these phototrophs; the algal hydrogenase has an iron-iron cluster at its catalytic site, while the cyanobacteria enzyme contains a nickel-iron cluster.
*Key technical challenges include overcoming oxygen sensitivity of hydrogenase enzymes, outcompeting other metabolic pathways for photosynthetic reductants, dissipating the proton gradients across the photosynthetic membrane, and ensuring adequate efficiency when capturing and converting solar energy.
*Despite technical challenges, abundant solar energy and water provide incentives for developing large-scale means to produce hydrogen photobiologically.
Global energy demand is projected to increase 50% or more by the year 2030, with coal remaining the primary fuel for developing countries. Because of the deleterious effects of fossil fuels and the uncertainties of future sources, the development of alternative renewable energy supplies such as hydrogen is critical. Hydrogen, the simplest and most abundant element, can be produced as a gas (H2) from various resources and, when combusted, generates water as the only byproduct. Its enabling technology, the H2-based fuel cell, is twice as efficient as the traditional internal combustion engine currently used in automobiles. However, nearly all hydrogen gas being produced industrially comes from steam-reforming of natural gas, a fossil fuel. Thus the development of alternative and renewable pathways for producing hydrogen fuels is of utmost importance.
Green algae and cyanobacteria can use solar energy to convert water into hydrogen gas, an energy carrier whose use does not emit greenhouse gases. Barriers to microbially based, large-scale production of hydrogen include (a) inherent properties of the microbes that preclude continuity and efficiency of H2 production; (b) underlying limitations of photosynthetic efficiency; (c) limitations of the hydrogenase catalyic function; and (d) engineering challenges.
Photosynthetic Hydrogen Production
A promising and renewable route to producing hydrogen gas entails harnessing microbial metabolic systems, specifically hydrogenase enzymes that convert protons and electrons into hydrogen gas:
2H+ + 2e- ↔ H2 .
Hydrogen is a key intermediate metabolite in ecosystems inhabited by a diverse group of photosynthetic and nonphotosynthetic microbes. The nonphotosynthetic microbes ferment sugars, releasing excess electrons to protons and generating hydrogen gas. This process regenerates nicotinamide dinucleotide (phosphate), NAD(P)+, which cells need to metabolize sugars for growth. Although this process for producing hydrogen is fast, it depends on readily fermentable sugars and thus would be costly to scale up.
Photosynthetic green algae and cyanobacteria provide a more promising pathway for generating hydrogen on a large scale. Hydrogen production by these microorganisms depends on the availability of plentiful resources, namely water as a substrate and solar energy as the energy source. Moreover, the oxygen and hydrogen that such cells produce could be used in a fuel cell to generate electricity (Fig. 1).
Green algae and cyanobacteria absorb light through pigments that are associated with two photosystems, photosystem I (PSI) and photosystem II (PSII) (Fig. 2). The absorbed light energy is transferred from the antenna pigments to chlorophyll reaction center molecules where charge separation occurs, yielding oxidants and reductants. The strong oxidant generated by PSII extracts electrons from water while releasing oxygen and protons as byproducts (Fig. 2).
The generated electrons reduce a series of membrane- bound and soluble carriers, ultimately reducing oxidant generated by PSI.
PSI generates a reductant that eventually reduces the iron-sulfur protein ferredoxin, which plays several roles. Its main function is to provide electrons to generate NADPH via ferredoxin- NADP oxidoreductase (FNR).
NADPH, along with ATP, is needed for fixing carbon dioxide via the Calvin- Benson-Bassham cycle and for producing carbohydrates. However, in the absence of carbon dioxide and under anaerobic conditions, reduced ferredoxin or NADPH reduces protons to yield hydrogen gas, a reaction catalyzed by hydrogenase. Ferredoxin links photosynthetic electron transport directly to hydrogen production in green algae, whereas NAD(P)H is the likely electron donor to hydrogenease in cyanobacteria (Fig. 2).
History and New Players
The transient activity of the algal and cyanobacterial hydrogen-producing enzymes delayed their discovery. In 1942 Hans Gaffron and Jack Rubin at the University of Chicago observed that the green alga Scenedesmus obliquus, when subject to anaerobic conditions, produces hydrogen in the absence of carbon dioxide, or uses hydrogen as the electron donor to fix carbon dioxide when the latter is present. Other green algae, including Chlorella fusca and Chlamydomonas reinhardtii, behave similarly when incubated anaerobically.
During the 1980s, Paul Roessler and Stephen Lien at the Solar Energy Research Institute, now the National Renewable Energy Laboratory (NREL), purified a hydrogenase from C. reinhardtii and determined that ferredoxin is its direct electron donor. By 2000, Michael Seibert and Maria Ghirardi at NREL and Anastasios Melis at the University of California, Berkeley, reported that, by depleting sulfate from sealed cultures of C. reinhardtii, the cells will produce substantial amounts of hydrogen for a prolonged period. This is achieved presumably by down-regulating photosynthesis, thereby producing less oxygen which is then consumed by respiration, thus establishing, at least in part, anaerobic conditions. This phenomenon helped to renew interest in algal hydrogen production.
In 1949 Howard Gest and Martin Kamen at Washington University in St. Louis, Mo., concluded that the photosynthetic bacterium Rhodospirillum rubrum produced hydrogen through a side reaction of nitrogenase, the enzyme responsible for converting nitrogen into ammonia. Nitrogen-fixing cyanobacteria contain nitrogenase in specialized cells called heterocysts; in sunlight, several Anabaena species produce hydrogen via nitrogenase in heterocysts. Because nitrogenase requires energy from adenosine triphosphate (ATP) molecules, this metabolic route to hydrogen production is less efficient than that for hydrogenase.
In 1981 Jeffrey Houchins and Robert Burris at the University of Wisconsin-Madison reported the discovery of a reversible, or bidirectional, hydrogenase in the cyanobacterium Anabaena 7120 (ATCC Nostoc muscorum) that produced hydrogen when ammonia repressed the nitrogenase. Although not universal, bidirectional hydrogenases are found in many cyanobacteria, including various species of Nostoc, Anabaena, Synechocystis, and Synechococcus.
The reaction direction of the bidirectional hydrogenase is dictated by the partial pressure of hydrogen and the presence of alternative electron sinks such as carbon dioxide. Hydrogen production increases considerably under either anaerobic or microaerobic conditions, as is the case with green algal hydrogenases. The physiological function of the bidirectional hydrogenase is still under debate, with proposed roles in fermentation and in serving as an electron valve during photosynthesis. The bidirectional hydrogenase in the genetically manipulatable Synechocystis sp. PCC 6803 is constitutively transcribed, making this strain attractive for further study of photobiological hydrogen production.
Green algal hydrogenases belong to the class of FeFe-hydrogenases, which are also found in strict anaerobes, fungi, and protists. So far hydrogenase genes have been characterized in diverse green algal species including Scenedesmus obliquus, Chlamydomonas reinhardtii, Chlorella fusca, and Chlamydomonas moewusii. Each of these genes encodes a protein of about 48 kDa, with about 50% sequence similarities. The monomeric hydrogenase protein harbors a metallo-catalytic site, the H-cluster. The FeFehydrogenases contain only iron as a metal within their catalytic sites. The H-cluster consists of a [4Fe-4S] cubane linked through a protein cysteine residue to a 2Fe subcluster (Fig. 3A). The iron atoms of the [4Fe-4S] center bind to the protein structure by three additional cysteine residues. Except for the bridging cysteine, the iron atoms of the 2Fe center are coordinated to carbon monoxide (CO) and cyanide (CN) ligands.
Expression of the hydrogenase gene is tightly regulated in green algae. In C. reinhardtii, for example, anaerobiosis induces transcription of the two hydrogenase structural genes, as well as two maturation genes involved in the biosynthesis and assembly of the H-cluster. The FeFehydrogenases link to ferredoxin as an electron donor, and typically produce hydrogen. Oxygen irreversibly inactivates the algal hydrogenase Hcluster, with a half-life of a few seconds. The chemical nature of the species bound to the H-cluster after exposure to oxygen is not known. However, the distal Fe appears to be oxidized when the enzyme is exposed to oxygen.
Cyanobacterial NiFe-Bidirectional Hydrogenases
Cyanobacteria contain NiFe-hydrogenases, which are phylogenetically unrelated to the FeFe-hydrogenases. The NiFe-hydrogenases are more widespread than are the FeFe-hydrogenases, and are found throughout Archaea and Bacteria.
The simplest NiFe-hydrogenases are heterodimeric, but some are more complex, notably the bidirectional hydrogenase of cyanobacteria that consists of five subunits (Fig. 3B). The catalytic center of the pentameric hydrogenase is bound to the large subunit, HoxH, and contains Fe and Ni atoms linked to CN and CO ligands and sulfur atoms from cysteine residues within the surrounding protein. The hydrogenase small subunit, HoxY, contains a [4Fe-4S] cluster that is necessary for electron transfer to the large catalytic subunit. The remaining three subunits, HoxF, HoxU, and HoxE, together form the diaphorase moiety of the complex, which likely channels electrons between NAD(P)H and the hydrogenase active site (Fig. 3). In the purified complex from the cyanobacterium Synechocystis sp. PCC 6803, the preferred electron donor is NAD(P)H and not the energetically more favorable ferredoxin used by algal FeFe-hydrogenases, according to Thomas Happe of Ruhr-Universität-Bochum in Bochum, Germany.
Unlike the FeFe-hydrogenase from C. reinhardtii, the NiFe-bidirectional hydrogenase from Synechocystis sp. PCC 6803 is expressed under oxygenic conditions. Shortly after being transferred from the dark into the light, photosynthesis in Synechocystis resumes, as does electron transport, and hydrogen is produced. However, as soon as oxygen accumulates in the light, hydrogen production is inhibited. When oxygen is removed, the bidirectional NiFe-hydrogenase is reactivated and photoproduces hydrogen once again. This reversible phenomenon is consistent with what is seen for the model NiFehydrogenase from the sulfate-reducing bacterium Desulfovibrio gigas. In that case, oxygen forms an oxo- or a hydroxo-bridge that chelates nickel and iron, and this complex is chemically reduced when the cells are returned to reducing conditions.
Approaches to Overcoming Practical Barriers
To realize the economic potential of producing hydrogen photobiologically, several research challenges must first be faced, including overcoming the oxygen sensitivity of the hydrogenase enzymes, addressing the competition between hydrogenases and other enzymes for photosynthetic reductants, preventing the down-regulation of photosynthesis caused by the nondissipation of the proton gradient across the photosynthetic membrane, and ensuring adequate efficiency when capturing and converting solar energy.
Because both algal and cyanobacterial hydrogenases are sensitive to oxygen, an essential product of photosynthesis, hydrogen is produced only transiently when such cells are exposed to light. There are several strategies for extending the catalytic lifetime of such hydrogenases, including: (i) computational simulation of pathways for oxygen gas diffusion into the catalytic site of hydrogenases and molecular engineering of these pathways, perhaps by narrowing the gas channels, to block O2 from reaching the catalytic site; (ii) mutagenizing the hydrogenase gene and then screening for an oxygen-tolerant version of this enzyme; (iii) employing a metabolic switch such as sulfate deprivation to down-regulate PSII-catalyzed oxygen evolution, inducing an anaerobic environment to sustain hydrogenase activity in the light; and (iv) searching for more oxygen-tolerant hydrogenases from nature and then transferring such genes into green algae and cyanobacteria. Except for the case in which oxygen production would be suppressed, it will be necessary to separate the hydrogen and oxygen being produced to avoid accumulating flammable or explosive mixtures.
In cells, the primary reductants for producing hydrogen, ferredoxin and NAD(P)H, act as key electron donors in other biochemical reactions, including those catalyzed by enzymes such as nitrate reductase and glutamate synthase, which have higher affinity for ferredoxin than does hydrogenase. Culturing cells without nitrate but supplemented with ammonia could suppress nitrate reductase activity. Further, depleting carbon dioxide would also leave more reduced ferredoxin available for producing hydrogen. Another strategy for directing more reductant to hydrogenase is to prevent electron transfer around PSI to maintain a high ratio of NADPH to ATP required for producing hydrogen. For example, a C. reinhardtii mutant with low cyclic photophosphorylation produces hydrogen more effectively than do ordinary cells, according to Ben Hankamer of the University of Queensland in Brisbane, Australia.
In yet another approach, John Golbeck and his collaborators at Pennsylvania State University in University Park plan to link hydrogenase chemically to the reducing side of PSI, bypassing competing pathways. In addition, other strategies to bypass competing pathways are also being investigated. For example, in cyanobacteria, the bidirectional hydrogenase is linked to NAD(P)H, which is also a substrate for respiratory complex I. One cyanobacterial complex I mutant produces higher levels of hydrogen, presumably because more NAD(P)H is diverted into that pathway. In addition, deleting genes encoding components in competing electrontransfer pathways that are involved in respiration and nitrate assimilation in Synechocystis sp. PCC 6803 also improves hydrogen production, according to Jens Appel and collaborators at Christian-Albrechts-Universität in Kiel, Germany.
Photosynthetic electron transport is coupled to the buildup of a proton gradient across the photosynthetic membrane. ATP synthase dissipates this gradient while generating ATP, which is used to fix carbon dioxide during photosynthesis (Fig. 2). Under anaerobic, hydrogenproducing conditions, the rates of carbon dioxide fixation in green algae decrease, ATP is not consumed, and the proton gradient is not effectively dissipated. These conditions lead to down-regulation of photosynthetic electron transport, resulting in further decreases in the rates of hydrogen production by the hydrogenase. This effect has been confirmed by the observation that addition of chemical uncouplers to hydrogen-producing C. reinhardtii restores the rates of hydrogen photoproduction. A similar effect might arise from inserting an inducible proton channel to disrupt that gradient only while hydrogen is being produced, suggest Eli Greenbaum and James Lee at the Oak Ridge National Laboratory in Oak Ridge, Tenn.
When grown on a mass scale, green algae and cyanobacteria are inefficient in their use of high-intensity light because their large lightabsorbing antennae cause a saturation of electron transport at less than 10% full sunlight (light absorption is faster than the rate of electron transport at sunlight intensities above 10%). The excess absorbed light is therefore dissipated as heat or fluorescence. Moreover, since the top layer of cells in the reactor capture most of the incident photons, the remaining cells will be shaded and will not contribute to hydrogen photoproduction, which will result in overall low light conversion efficiencies per reactor.
One promising approach to address this challenge is to genetically truncate the pigment-containing antennae in green algae and cyanobacteria. For example, several mutants with reduced antenna size in PSII, PSI, or both saturate at much higher light intensity, which is promising, according to Anastasios Melis at the University of California, Berkeley and Ben Hankamer in Australia and their respective collaborators. However, whether these mutants produce higher amounts of hydrogen than do their parent cell lines is not yet known. Nevertheless, this finding is a step toward regulating antenna size of algae and cyanobacteria.
We thank the DOE Hydrogen, Fuel Cells, and Infrastructure Technologies Program and the NREL Laboratory Directed Research and Development Program for financial support. We thank Dr. Michael Seibert and George Sverdrup of NREL for commenting on the manuscript.
Ghirardi, M. L., A. Dubini, J. Yu, and P. C. Maness. 2009. Photobiological hydrogen-producing systems. Chem. Soc. Rev. 38: 52-61.
Ghirardi, M. L., M. C. Posewitz, P. C. Maness, A. Dubini, J. Yu, and M. Seibert. 2007. Hydrogenases and hydrogen photoproduction in oxygenic photosynthetic organisms. Annu. Rev. Plant Biol. 58:71-91.
Gutthann, F., M. Egert, A. Marques, and J. Appel. 2007. Inhibition of respiration and nitrate assimilation enhances photohydrogen evolution under low oxygen concentrations in Synechocystis sp. PCC 6803. Biochim. Biophys. Acta 1767:161-169.
Melis, A., L. Zhang, M. Forestier, M. L. Ghirardi, and M. Seibert. 2000. Sustained photobiological hydrogen gas production upon reversible inactivation of oxygen evolution in the green alga Chlamydomonas reinhardtii. Plant Physiol. 122:127-135.
Melis, A. 2002. Green algal hydrogen production: progress, challenges and prospects. Int. J. Hydrogen Energy 27:1217-1228.
Peters, J. W., W. N. Lanzilotta, B. Lemon, and L. Seefeldt. 1998. X-ray crystal structure of the Fe-only hydrogenase (CpI) from Clostridium pasteurianum to 1.8 angstrom resolution. Science 282:1853-1858.
Seibert, M., P. W. King, M. C. Posewitz, A. Melis, and M. L. Ghirardi. 2008. Photosynthetic water-splitting for hydrogen production. In J. Wall et al. (ed.), Bioenergy. ASM Press, Washington, D.C.
Tamagnini, P., R. Axelsson, P. Lindberg, F. Oxelfelt, R. Wünschiers, and P. Lindblad. 2002. Hydrogenases and hydrogen metabolism of cyanobacteria. Microbiol. Mol. Biol. Rev. 66:1-20.
Vignais, P. M., B. Billoud, and J. Meyer. 2001. Classification and phylogeny of hydrogenases. FEMS Microbiol. 25:455-501.
Volbeda, A., M. H. Charon, C. Piras, E. C. Hatchikian, M. Frey, and J. C. Fontecilla-Camps. 1995. Crystal structure of the nickel-iron hydrogenase from Desulfovibrio gigas. Nature 373:580-587.