21
Figure 10:
Three-dimensional structure of the androgen receptor ligand binding domain (LBD)
and DNA binding domain (DBD). The arrow in LBD denotes the natural ligand
dihydrotestosterone (DHT). The DNA is indicated below the DBD.
Possible molecular interactions between ca27 and its analogs with the AR will include geometric
(structural) and chemical strain (described in depth below), as well as favorability parameters.
These interactions will be explored in respect to both the LBD and DBD. In regard to the LBD,
the investigations will be performed with and without a focus on the ligand-binding pocket (LPC).
In particular, MolSoft ICM determines possible bindings by assigning a numerical score, which
represents the geometric and chemical strain from the free state of
the ligand in the binding
pocket
32
. The ICM scoring function is weighted according to the following parameters (i) internal
force-field energy of the ligand, (ii) entropy loss of the ligand between bound and unbound states,
(iii) ligand-receptor hydrogen bond interactions, (iv) polar and
non-polar solvation energy
differences between bound and unbound states, (v) electrostatic energy, (vi) hydrophobic energy,
and (vii) hydrogen bond donor or acceptor de-solvation. The lower the ICM score, the higher the
chance the ligand will bind
32
.
22
The geometric strain from the free state represents the configuration of the ligand-protein structural
complex, where a favorable (or more negative) results indicates the receptor is actually more
relaxed with the ligand and an unfavorable interaction means there is more strain structural than
the free state. The chemical strain represents favorable or unfavorable interactions between the
amino acids of both the ligand and the receptor (meaning chemical compatibility) where, again,
more negative values represents more favorable interactions than the free state. MolSoft outputs
two major values to be considered, Score and VLS Score. The Score is the strain from the free
state for the protein alone whereas the VLS Score is the Score with the addition of the strain on
the ligand as well. For the purpose of our analysis, we focused on the VLS Score as it is a more
comprehensive model of the results as the strain on the ligand is important to docking. AutoDock
takes a similar approach utilizing a suite of automated docking tools used to predict how small
molecules bind to a receptor, given a three-dimensional structure
33
. Positive and negative controls
will be used to set relative scoring values in perspective to the determined binding affinity of ca27
and its analogs. For the LBD, the positive control is given by DHT, the natural ligand of the AR.
In addition, clinically used AR blockers with proven affinity for the AR LBD will also be subjected
to computational docking. These include bicalutamide and enzalutamide
30
. Because of the
transcriptional
function of the DBD, it has no natural ligand. However, the anti-helminthic
compound pyrvinium pamoate (PP) has recently been discovered to exert high binding affinity to
the AR DBD
34
and will thus be used as a positive control. Negative controls for both the LBD and
the DBD will be compounds that are not linked to the androgen signaling axis, including organic
small molecules from the taxol family of
compounds, such as docetaxel.
Finally, using tools primarily in AutoDock (however the results shall be corroborated with the
MolSoft software), I attempted to analyze the results of all the different analogs bound to the ligand