of iron per 40 kDa subunit, the iron was initially thought to be a contaminant [50,
53]. The enzyme is not redox active, as it does not oxidize H 2 or reduce protons;
rather, this third hydrogenase heterolytically cleaves H 2 using its substrate
methenyl-tetrahydromethanopterin as a stereo-selective hydride acceptor [50, 54–
58]. For this reason, the enzyme was called the H 2 -forming methylenetetrahydromethanopterin dehydrogenase, Hmd; it is better considered as a hydrogenation
agent or catalyst. Methanogen expert Rolf Thauer and colleague Seigo Shima at the
MPI Marburg are chiefly responsible for unraveling the complexities of this perplexing enzyme; however a community of spectroscopists such as
Simon P. J. Albracht (Amsterdam), Albrecht Berkessel (Köln), Eckhard Bill
(Mülheim), Ulrich Ermler (Frankfurt), Victor Fernández (Madrid), Christian Griesinger (Göttingen), and Wolfram Meyer‐Klaucke (Hamburg) contributed much to
the understanding [54, 56–72].
In 2000, recombinant over-expression of the enzyme in Escherichia coli led to
the discovery of a cofactor that is necessary for its catalytic activity [73]. This
cofactor, which could be extracted from the holoenzyme, readily recombines with
apoenzyme to restore the enzyme’s activity. Still, there was no indication at that
time that the enzyme required a transition metal for its activity. For example,
enzyme kinetics suggested that a ternary mechanism is employed for H 2 splitting, as
opposed to a binary mechanism that would be indicative of the involvement of
transition metals [53, 74]. Furthermore, the nearly colorless enzyme had activity
that was not affected by the presence of 50% CO or acetylene in the gas phase [50].
A major clue to solve the conundrum of whether there was a metal in this
hydrogenase was the finding that both the enzyme and the cofactor isolated from it
were sensitive to light. Ultraviolet-A/blue-light inactivated and bleached the
enzyme, accompanied by release of up to one mole of iron per subunit [75].
Furthermore, the enzyme was stabilized by the presence of CO in the gas phase.
With 100% CO in the gas phase, the enzyme was reversibly inhibited by 50%. The
final acceptance that CO was present in the active site, as light-sensitive, labile
ligands to iron, resolved conflicts in the required vs. inhibitory CO at the active site
[59]. At this point it was clear that the common name for Hmd, ‘metal-free
hydrogenase’, was no longer accurate [75]. For this reason, Hmd is now referred to
as the iron-sulfur-cluster-free hydrogenase or simply [Fe]-H 2 ase, since it contains a
singular iron center in its active site.
4.2 Determination of Active Site/Cofactor Structure
Removal of iron from the isolated cofactor, by light or heat inactivation, revealed
the presence of a unique pyridone, (6-carboxymethyl-3,5-dimethyl-2-pyridone-4yl)-(5′-guanosyl)phosphate [66]. As shown in Fig. 7 X-ray crystallography eventually showed this pyridone is coordinated to the iron through N as well as an
Fe-acyl group involving an unusually short Fe–C bond [65, 68, 69]. It demonstrates
yet another type of metal–carbon bond in biology. Albeit much more elaborate in its
composition, the pyridone illustrates a non-protein-bound cofactor that orients its
284
M. Y. Darensbourg et al.
53]. The enzyme is not redox active, as it does not oxidize H 2 or reduce protons;
rather, this third hydrogenase heterolytically cleaves H 2 using its substrate
methenyl-tetrahydromethanopterin as a stereo-selective hydride acceptor [50, 54–
58]. For this reason, the enzyme was called the H 2 -forming methylenetetrahydromethanopterin dehydrogenase, Hmd; it is better considered as a hydrogenation
agent or catalyst. Methanogen expert Rolf Thauer and colleague Seigo Shima at the
MPI Marburg are chiefly responsible for unraveling the complexities of this perplexing enzyme; however a community of spectroscopists such as
Simon P. J. Albracht (Amsterdam), Albrecht Berkessel (Köln), Eckhard Bill
(Mülheim), Ulrich Ermler (Frankfurt), Victor Fernández (Madrid), Christian Griesinger (Göttingen), and Wolfram Meyer‐Klaucke (Hamburg) contributed much to
the understanding [54, 56–72].
In 2000, recombinant over-expression of the enzyme in Escherichia coli led to
the discovery of a cofactor that is necessary for its catalytic activity [73]. This
cofactor, which could be extracted from the holoenzyme, readily recombines with
apoenzyme to restore the enzyme’s activity. Still, there was no indication at that
time that the enzyme required a transition metal for its activity. For example,
enzyme kinetics suggested that a ternary mechanism is employed for H 2 splitting, as
opposed to a binary mechanism that would be indicative of the involvement of
transition metals [53, 74]. Furthermore, the nearly colorless enzyme had activity
that was not affected by the presence of 50% CO or acetylene in the gas phase [50].
A major clue to solve the conundrum of whether there was a metal in this
hydrogenase was the finding that both the enzyme and the cofactor isolated from it
were sensitive to light. Ultraviolet-A/blue-light inactivated and bleached the
enzyme, accompanied by release of up to one mole of iron per subunit [75].
Furthermore, the enzyme was stabilized by the presence of CO in the gas phase.
With 100% CO in the gas phase, the enzyme was reversibly inhibited by 50%. The
final acceptance that CO was present in the active site, as light-sensitive, labile
ligands to iron, resolved conflicts in the required vs. inhibitory CO at the active site
[59]. At this point it was clear that the common name for Hmd, ‘metal-free
hydrogenase’, was no longer accurate [75]. For this reason, Hmd is now referred to
as the iron-sulfur-cluster-free hydrogenase or simply [Fe]-H 2 ase, since it contains a
singular iron center in its active site.
4.2 Determination of Active Site/Cofactor Structure
Removal of iron from the isolated cofactor, by light or heat inactivation, revealed
the presence of a unique pyridone, (6-carboxymethyl-3,5-dimethyl-2-pyridone-4yl)-(5′-guanosyl)phosphate [66]. As shown in Fig. 7 X-ray crystallography eventually showed this pyridone is coordinated to the iron through N as well as an
Fe-acyl group involving an unusually short Fe–C bond [65, 68, 69]. It demonstrates
yet another type of metal–carbon bond in biology. Albeit much more elaborate in its
composition, the pyridone illustrates a non-protein-bound cofactor that orients its
284
M. Y. Darensbourg et al.