Nitrogen is essential for the biosynthesis of a broad range of organic molecules, including proteins, DNA and many types of sugars, without which living things would not be able to grow. Plants are unable to use the relatively stable atmospheric nitrogen, as the molecule’s triple bond is very hard to crack. Almost all plants are therefore dependent on the supply of reactive nitrogen compounds (nitrate or ammonia) through their roots, which they then incorporate into their amino acids. “In biochemical terms, humans are not even able to do this,” says Prof. Dr. Oliver Einsle from the Institute of Biochemistry at the University of Freiburg. “Humans have to eat plants to get to the nitrogen they need.” Einsle and his team are studying the enzyme that can convert atmospheric nitrogen into a form that humans can use. The enzyme is called nitrogenase and the process nitrogen fixation. The nitrogenase enzyme is a bacterial protein and is extremely sensitive to oxygen; it only works effectively inside a bacterial cell that protects it against exposure to oxygen. Some species of bacteria need to live in symbiotic relationships with other organisms in order to be able to convert atmospheric nitrogen (ed. note: Plants that live in symbiosis with bacteria synthesise leghaemoglobin, a protein that has chemical and structural similarities to haemoglobin. It has a high affinity to oxygen, ensuring that the oxygen concentration is kept so low that the nitrogenase enzyme functions, but that is high enough that the bacteria have enough oxygen for respiration.).

Structure of the nitrogenase enzyme

Einsle wants to understand how the nitrogenase enzyme works. It is still not known where and how nitrogen is bound and converted inside the enzyme. What we know is that nitrogenases consist of two metalloproteins, the iron (Fe) protein on the outside and the molybdenum-iron (MoFe) protein with a metal centre and a unique iron-molybdenum cofactor (FeMoCo). It is difficult to find out where exactly the nitrogen binds. “The reaction is induced as soon as the nitrogen molecule has bound to the metal centre. But we have not yet found a method that would enable us to keep the enzyme in this particular state and look at it in greater detail,” says Einsle.

In addition, the process is very complex because the entire enzyme complex needs to be repeatedly rearranged during the reaction. “This cannot be represented in crystals,” explains the biochemist. X-ray crystallography reveals that the state of the symmetric metal centre with the carbon in the middle and a nearby molybdenum molecule is extraordinarily stable. A ligand needs to be removed and a binding site exposed so that the enzyme can bind nitrogen. 

In order to study the enzyme in greater detail, Einsle and his colleagues eventually managed to slow down the chemical processes by cooling the crystals. They also used toxic compounds to inhibit the metal centre. “We had to resort to a nasty trick,” says Einsle. “We gave the enzyme everything it needs to work properly and then poisoned it with carbon monoxide.” Carbon monoxide is actually an inhibitor which blocks the enzyme at a specific site, preventing nitrogen from binding. A few years ago, researchers discovered that the enzyme could also convert the substance that had the capacity to inhibit it, i.e. carbon monoxide, into hydrocarbons, albeit at small quantities. However, the reaction is so slow that Einsle and his team can still see the carbon monoxide being bound before it is converted. The researchers were surprised, however, to see that it bound at a completely unexpected site.

It was clear that a bond had to be cracked in order to make room for a new molecule. However, instead of displacing the central carbon atom, another atom disappeared from the FeMo cofactor’s sphere of influence. “We thought that the metal centre was very stable, and did not expect it to be so flexible that it enables the enzyme to remove a sulphur atom and incorporate it at a different site,” says Einsle. “Such chemical rearrangements have not previously been observed in biological systems.” The carbon monoxide then ends up between two iron atoms. In chemical terms, this makes perfect sense. “Carbon monoxide and metals quite like each other,” adds Einsle going on to explain that “once the partners have reacted with each other, carbon monoxide disappears and the sulphur atom returns to its original site in what appears to be a rather complex process. Removing an atom from a complex structure, putting something else in, and then changing everything back again is quite sophisticated. This cannot be achieved in the laboratory.” 

Whether the same happens when the enzyme converts elementary nitrogen remains uncertain. In any case, the nitrogenase enzyme is able to convert all kinds of compounds that have a triple bond. Einsle assumes that different mechanisms are used, depending on the molecule being converted. “The exciting thing about carbon monoxide is that the reaction leads to hydrocarbons,” Einsle says. However, the hydrocarbon yield is fairly small. If the resulting butene or propene were to be used for the biotechnological production of fuel, the procedure would still have to be optimised.

However, Einsle was once again surprised to find that instead of converting carbon monoxide into methane, the nitrogenase enzyme converts it into far more complex and longer carbon chains. He is already trying to elucidate the mechanism and, as he says: “The nitrogenase enzyme works like a ribosome that assembles amino acids into a protein.”

Carbons have properties that make joining them difficult in the laboratory. However, although the nitrogenase enzyme is not optimised for the purpose, it is still able to do so. It is worth pointing out that the Azotobacter bacteria that Einsle is working with are able to produce nitrogenases that have different abilities, depending on the substrate available. The enzymes prefer molybdenum atoms in their metal centre, but sometimes they seem to be happy with vanadium. The vanadium-containing enzyme is more effective in converting carbon monoxide into hydrocarbons than the molybdenum variant, which in turn is far better at converting atmospheric nitrogen.

What can all this knowledge be used for? The goal is not to optimise the Haber-Bosch method as this cannot be improved further. Nor do we need to refine our fuels. Einsle is more focused on natural aspects such as the fixation of carbon monoxide in order to slow down the greenhouse effect. “Perhaps better catalysts could help close the natural carbon cycle that is out of balance,” says Einsle. “The ideal would be to be able to teach organisms how to make waste products that we could use as fuels.”