(designated NifEN
L
) indicated that the L-cluster is located in the analogous
M-cluster-binding site as in NifDK, and yet more exposed to the surface and
thereby accessible to modification [70]. Since the absence of NifH is a prerequisite
for the accumulation of L-cluster on NifEN, NifH is likely involved in the conversion of L- to M cluster. This is demonstrated by a series of biochemical and
spectroscopic studies which defined the exact role of NifH in the process, which is
to mobilize molybdate (MoO 4
2− ) and homocitrate at the expense of ATP hydrolysis
and subsequently carries these components to NifEN for their incorporation into the
L-cluster (Fig. 5a) [71–73]. This is confirmed by Mo K-edge XAS data, which
points to a change in the oxidation state and/or ligation pattern of the Mo atom upon
binding to NifH. EPR analysis revealed that the loading of Mo/homocitrate induces
changes to NifH’s Fe 4 S 4 cluster that is similar to those induced by the binding of
ATP or ADP [71]. For NifEN, upon incubation with dithionite, NifH, MgATP,
molybdate, and homocitrate, Mo K-edge XAS/EXAFS analyses of the matured
NifEN (designated as NifEN
M ) revealed to spectral features indistinguishable from
NifDK [72, 73]. In addition, the M-cluster that was matured on NifEN can be
chemically extracted into solvent just like in the case of NifDK [74, 75]. Both
versions of M-clusters, whether extracted from NifEN or NifDK, exhibit identical
spectroscopic properties as well as the same ability to activate apo-NifDK, which
proves conclusively that the M-cluster is formed on NifEN [74, 75].
After these events, NifEN
M undergoes a conformational change that renders
M-cluster less solvent-exposed compared to NifEN
L (Fig. 5a) [75]. This change
was demonstrated by metal-content analysis and chelation studies of NifEN
L and
NifEN
M . The results show that the Fe atoms in NifEN
L can be completely chelated,
which is supported by the solvent-exposed nature of the L-cluster observed in the
crystal structure. In stark contrast, the majority of Fe atoms in NifEN
M , that are
likely associated with the M-cluster, cannot be removed by a chelator [75]. Thus,
upon maturation to the M-cluster, a conformational change must occur that embeds
the cofactor within a less solvent-accessible location of the protein. This change is
likely important for the subsequent events where NifEN
M directly donates its fully
assembled M-cluster to NifDK (Fig. 5a).
3.3.3 Insertion of M-Cluster into the NifDK Scaffold
After formation of the M-cluster, NifEN
M was shown to fully activate apo-NifDK
without the need for accessory chaperons or carriers (Fig. 5) [36, 37, 67, 71–74].
Moreover, results from native PAGE experiments revealed that NifEN
M can form a
direct protein–protein complex with apo-NifDK, in which the M-cluster can
directly be transferred to NifDK [76]. Sequence analysis revealed a number of key
residues in NifDK that either provide a covalent linkage or tightly pack the
M-cluster within the binding pocket and are missing in NifEN. These residues
include Lys426 (which anchors the homocitrate), His442 (which coordinates Mo),
and Trp444 (which packs the M-cluster into the pocket with its bulky side chain)
[36, 37]. Their exclusive presence on NifDK, but not on NifEN, was believed to
create a high-affinity M-cluster site on NifDK and the low-affinity M-cluster site in
NifEN, wherein the difference in M-cluster affinity between these two sites provide
168
C.-C. Lee et al.
L
) indicated that the L-cluster is located in the analogous
M-cluster-binding site as in NifDK, and yet more exposed to the surface and
thereby accessible to modification [70]. Since the absence of NifH is a prerequisite
for the accumulation of L-cluster on NifEN, NifH is likely involved in the conversion of L- to M cluster. This is demonstrated by a series of biochemical and
spectroscopic studies which defined the exact role of NifH in the process, which is
to mobilize molybdate (MoO 4
2− ) and homocitrate at the expense of ATP hydrolysis
and subsequently carries these components to NifEN for their incorporation into the
L-cluster (Fig. 5a) [71–73]. This is confirmed by Mo K-edge XAS data, which
points to a change in the oxidation state and/or ligation pattern of the Mo atom upon
binding to NifH. EPR analysis revealed that the loading of Mo/homocitrate induces
changes to NifH’s Fe 4 S 4 cluster that is similar to those induced by the binding of
ATP or ADP [71]. For NifEN, upon incubation with dithionite, NifH, MgATP,
molybdate, and homocitrate, Mo K-edge XAS/EXAFS analyses of the matured
NifEN (designated as NifEN
M ) revealed to spectral features indistinguishable from
NifDK [72, 73]. In addition, the M-cluster that was matured on NifEN can be
chemically extracted into solvent just like in the case of NifDK [74, 75]. Both
versions of M-clusters, whether extracted from NifEN or NifDK, exhibit identical
spectroscopic properties as well as the same ability to activate apo-NifDK, which
proves conclusively that the M-cluster is formed on NifEN [74, 75].
After these events, NifEN
M undergoes a conformational change that renders
M-cluster less solvent-exposed compared to NifEN
L (Fig. 5a) [75]. This change
was demonstrated by metal-content analysis and chelation studies of NifEN
L and
NifEN
M . The results show that the Fe atoms in NifEN
L can be completely chelated,
which is supported by the solvent-exposed nature of the L-cluster observed in the
crystal structure. In stark contrast, the majority of Fe atoms in NifEN
M , that are
likely associated with the M-cluster, cannot be removed by a chelator [75]. Thus,
upon maturation to the M-cluster, a conformational change must occur that embeds
the cofactor within a less solvent-accessible location of the protein. This change is
likely important for the subsequent events where NifEN
M directly donates its fully
assembled M-cluster to NifDK (Fig. 5a).
3.3.3 Insertion of M-Cluster into the NifDK Scaffold
After formation of the M-cluster, NifEN
M was shown to fully activate apo-NifDK
without the need for accessory chaperons or carriers (Fig. 5) [36, 37, 67, 71–74].
Moreover, results from native PAGE experiments revealed that NifEN
M can form a
direct protein–protein complex with apo-NifDK, in which the M-cluster can
directly be transferred to NifDK [76]. Sequence analysis revealed a number of key
residues in NifDK that either provide a covalent linkage or tightly pack the
M-cluster within the binding pocket and are missing in NifEN. These residues
include Lys426 (which anchors the homocitrate), His442 (which coordinates Mo),
and Trp444 (which packs the M-cluster into the pocket with its bulky side chain)
[36, 37]. Their exclusive presence on NifDK, but not on NifEN, was believed to
create a high-affinity M-cluster site on NifDK and the low-affinity M-cluster site in
NifEN, wherein the difference in M-cluster affinity between these two sites provide
168
C.-C. Lee et al.