4.3 Structural Insights into the Mechanism of N 2 Reduction
In parallel to the biochemical and spectroscopic endeavors described, a series of
structural studies also shed valuable insights on the workings of NifDK. It was
traditionally recognized that structural characterization of any reaction intermediates is exceedingly difficult because it would require the isolation of the reactive
NifDK away from the NifH/MgATP-containing reaction mixture, as well as a swift
crystallization procedure to capture the reactive conformation before it relaxes back
to the resting state [4, 5]. With this in mind, seminal work from the Rees group
generated a CO-bound NifDK and solved its crystal structure (Fig. 7a) [106]. CO is
isoelectric to N 2 , and a well-known inhibitor of nitrogenase, thus it has a much
higher chance to be accumulated on the M-cluster. The crystallization is also made
possible by an innovated protocol developed by the team, in which substrate
turnover was first initiated under an atmosphere of CO, followed by quick
re-isolation from the reaction mixture by filtration. The protein was immediately
crystallized with crystal seeds, and CO-bound NifDK crystals were reported to form
within a matter of hours. The resulting structure had a resolution of 1.50 Å, and
featured a l 2 -bridging CO, bound to Fe 2 and Fe 6 of the M-cluster (Fig. 7a). Surprisingly, analysis of the native and the sulfur anomalous data sets revealed that CO
completely displaced the bridging sulfur atom (designated S2B) that was originally
in the same position. This replacement of Fe ligations also results in a slight
distortion of the core geometry, in which that the Mo, the interstitial carbide, and
Fe1 atoms no longer align along the threefold symmetry axis, and an overall
contraction of the M-cluster around the CO ligand (Fig. 7a). Significantly, the loss
of the belt sulfur was demonstrated to be reversible, as recovery of S2B was
demonstrated by re-dissolving the CO-bound crystalline NifDK and subjecting
them again under turnover with C 2 H 2 . This finding, on the one hand, substantiates
the idea that the loss a belt sulfur is a catalytically relevant step, but on the other
hand, raises the question whether there is a nearby site that temporarily harbors the
sulfur during catalysis. Inspection of the structure identified a potential sulfur
binding site, albeit *22 Å away from the S2B position, and it is unclear how the
S2B sulfur should migrate back and forth across such a long distance within the
time frame of substrate turnover (Fig. 7a) [106].
Despite the unanswered questions, the structure nevertheless provided the field
with a novel direction of thinking about the mechanism of nitrogenase: the removal
of a sulfur ligand that opens up an active Fe site for substrate activation. More
recently, another structural report by the group of Einsle also revealed the presence
of a light atom ligand, deemed a putative reaction intermediate, in VnfDKG of the
A. vinelandii V-nitrogenase (Fig. 7b) [107]. In this work, the sample was directly
isolated from the cells using a lower dithionite concentration than normal, followed
by immediate crystallization. Throughout the process, the sample was kept under a
N 2 atmosphere and the structure were resolved to 1.20 Å. Consistent with the
CO-bound NifDK structure, the light atom ligand in the VnfDGK structure was also
shown to displace the S2B atom of V-cluster (the V-containing analog of
M-cluster). Furthermore, in this structure electron density consistent with a sulfur
176
C.-C. Lee et al.
In parallel to the biochemical and spectroscopic endeavors described, a series of
structural studies also shed valuable insights on the workings of NifDK. It was
traditionally recognized that structural characterization of any reaction intermediates is exceedingly difficult because it would require the isolation of the reactive
NifDK away from the NifH/MgATP-containing reaction mixture, as well as a swift
crystallization procedure to capture the reactive conformation before it relaxes back
to the resting state [4, 5]. With this in mind, seminal work from the Rees group
generated a CO-bound NifDK and solved its crystal structure (Fig. 7a) [106]. CO is
isoelectric to N 2 , and a well-known inhibitor of nitrogenase, thus it has a much
higher chance to be accumulated on the M-cluster. The crystallization is also made
possible by an innovated protocol developed by the team, in which substrate
turnover was first initiated under an atmosphere of CO, followed by quick
re-isolation from the reaction mixture by filtration. The protein was immediately
crystallized with crystal seeds, and CO-bound NifDK crystals were reported to form
within a matter of hours. The resulting structure had a resolution of 1.50 Å, and
featured a l 2 -bridging CO, bound to Fe 2 and Fe 6 of the M-cluster (Fig. 7a). Surprisingly, analysis of the native and the sulfur anomalous data sets revealed that CO
completely displaced the bridging sulfur atom (designated S2B) that was originally
in the same position. This replacement of Fe ligations also results in a slight
distortion of the core geometry, in which that the Mo, the interstitial carbide, and
Fe1 atoms no longer align along the threefold symmetry axis, and an overall
contraction of the M-cluster around the CO ligand (Fig. 7a). Significantly, the loss
of the belt sulfur was demonstrated to be reversible, as recovery of S2B was
demonstrated by re-dissolving the CO-bound crystalline NifDK and subjecting
them again under turnover with C 2 H 2 . This finding, on the one hand, substantiates
the idea that the loss a belt sulfur is a catalytically relevant step, but on the other
hand, raises the question whether there is a nearby site that temporarily harbors the
sulfur during catalysis. Inspection of the structure identified a potential sulfur
binding site, albeit *22 Å away from the S2B position, and it is unclear how the
S2B sulfur should migrate back and forth across such a long distance within the
time frame of substrate turnover (Fig. 7a) [106].
Despite the unanswered questions, the structure nevertheless provided the field
with a novel direction of thinking about the mechanism of nitrogenase: the removal
of a sulfur ligand that opens up an active Fe site for substrate activation. More
recently, another structural report by the group of Einsle also revealed the presence
of a light atom ligand, deemed a putative reaction intermediate, in VnfDKG of the
A. vinelandii V-nitrogenase (Fig. 7b) [107]. In this work, the sample was directly
isolated from the cells using a lower dithionite concentration than normal, followed
by immediate crystallization. Throughout the process, the sample was kept under a
N 2 atmosphere and the structure were resolved to 1.20 Å. Consistent with the
CO-bound NifDK structure, the light atom ligand in the VnfDGK structure was also
shown to displace the S2B atom of V-cluster (the V-containing analog of
M-cluster). Furthermore, in this structure electron density consistent with a sulfur
176
C.-C. Lee et al.