perturbations that can be disregarded and that static models are
sufficient for drug discovery efforts [20]. However, based on what
we now know about the role of protein dynamics in catalysis and
inhibitor binding, assuming that protein motions can be ignored
may be a mistake. Mauldin et al. used
15
N and
2 H NMR spin
relaxation experiments to investigate how the functional motions
of DHFR respond to MTX and TMP. They profiled the motions of
ecDHFR bound to NADPH in the presence and absence of the two
inhibitors and found that drug binding at the substrate binding site
breaks up the usual μs–ms loop motions of the holoenzyme into
smaller, unproductive clusters of local motion, which they refer to
as “dynamic dysfunction” [102]. Interestingly, they found that
both MTX and TMP cause the same “dysfunction,” which suggests
that these dynamic changes may be important to inhibitory activity.
These results demonstrate that MTX and TMP do not just block
substrate binding. They also cause a breakdown of communication
within the network of collective motions required for DHFR catalysis [20, 102]. The results of this study are supported by the results
of a more recent investigation of TMP binding by Abdizadeh et al.
They also found that TMP effectively “locks” DHFR in the closed
conformation where the closed conformation dominates in the
ensemble. They propose that binding of TMP sends a “signal” for
conformational change on the μs–ms timescale for the drugged
complex to remain locked in the closed conformation. If this is
true for DHFR inhibitors, then it is reasonable to assume that other
drugs may act in a similar fashion.
The study conducted by Mauldin et al. highlights several points
relevant for drug discovery. First, flexibility-function studies can
indicate new modes of drug action that would not be observed
using traditional, static model drug discovery and design strategies.
Second, drug action is most likely broader than we originally
thought. Competitive inhibitors may also inhibit functional protein
dynamics in addition to preventing substrate from binding to the
active site. Third, protein functional motions can be distributed
among networks of amino acid residues throughout the protein.
Thus, motions at one site can be inhibited by binding at a distant
site. Although this makes the drug discovery and design process
much more complex, it also widens the range of potential inhibitor
binding sites, facilitating the design of more specific inhibitors.
Finally, protein dynamics studies should be complemented by
studying the dynamics of the ligand. Since drug design entails
modifying the ligand, not the protein target, the conformation of
the ligand when bound to the protein can provide added
insight [20].
In addition, as described earlier, inhibitor binding modulates
the conformational ensemble of a protein by shifting the conformational equilibrium. By studying the conformational motions of a
protein or enzyme, we may be able to identify the ligands preferred
Distal Regions Regulate Dihydrofolate Reductase-Ligand Interactions
213
sufficient for drug discovery efforts [20]. However, based on what
we now know about the role of protein dynamics in catalysis and
inhibitor binding, assuming that protein motions can be ignored
may be a mistake. Mauldin et al. used
15
N and
2 H NMR spin
relaxation experiments to investigate how the functional motions
of DHFR respond to MTX and TMP. They profiled the motions of
ecDHFR bound to NADPH in the presence and absence of the two
inhibitors and found that drug binding at the substrate binding site
breaks up the usual μs–ms loop motions of the holoenzyme into
smaller, unproductive clusters of local motion, which they refer to
as “dynamic dysfunction” [102]. Interestingly, they found that
both MTX and TMP cause the same “dysfunction,” which suggests
that these dynamic changes may be important to inhibitory activity.
These results demonstrate that MTX and TMP do not just block
substrate binding. They also cause a breakdown of communication
within the network of collective motions required for DHFR catalysis [20, 102]. The results of this study are supported by the results
of a more recent investigation of TMP binding by Abdizadeh et al.
They also found that TMP effectively “locks” DHFR in the closed
conformation where the closed conformation dominates in the
ensemble. They propose that binding of TMP sends a “signal” for
conformational change on the μs–ms timescale for the drugged
complex to remain locked in the closed conformation. If this is
true for DHFR inhibitors, then it is reasonable to assume that other
drugs may act in a similar fashion.
The study conducted by Mauldin et al. highlights several points
relevant for drug discovery. First, flexibility-function studies can
indicate new modes of drug action that would not be observed
using traditional, static model drug discovery and design strategies.
Second, drug action is most likely broader than we originally
thought. Competitive inhibitors may also inhibit functional protein
dynamics in addition to preventing substrate from binding to the
active site. Third, protein functional motions can be distributed
among networks of amino acid residues throughout the protein.
Thus, motions at one site can be inhibited by binding at a distant
site. Although this makes the drug discovery and design process
much more complex, it also widens the range of potential inhibitor
binding sites, facilitating the design of more specific inhibitors.
Finally, protein dynamics studies should be complemented by
studying the dynamics of the ligand. Since drug design entails
modifying the ligand, not the protein target, the conformation of
the ligand when bound to the protein can provide added
insight [20].
In addition, as described earlier, inhibitor binding modulates
the conformational ensemble of a protein by shifting the conformational equilibrium. By studying the conformational motions of a
protein or enzyme, we may be able to identify the ligands preferred
Distal Regions Regulate Dihydrofolate Reductase-Ligand Interactions
213