Table 1.

Experimentally determined inhibition constants and standard free energies compared to standard free energies calculated from simulation.

Terpene	${\mathit{IC}}_{50}$ ($\mathsf{\mu }$M)	$\mathbf{\Delta }{\mathit{G}}^{\circ }$ Experimental (kcal/mol)	$\mathbf{\Delta }{\mathit{G}}^{\circ }$ Theoretical (kcal/mol)
Betulinic acid	$115\pm 16$	$-4.5\pm 0.1$	$-11.0\pm 1.4$
Ursolic acid	$220\pm 40$	$-3.5\pm 0.1$	$-4.3\pm 0.8$
Oleanolic acid	$360\pm 70$	$-3.7\pm 0.1$	$-6.1\pm 0.6$
Madecassic acid	$540\pm 80$	ND	$-5.8\pm 0.5$
Betulin	$1470\pm 130$	ND	$-1.12\pm 0.01$
$\beta$-Boswellic acid	ND	ND	$-6.6\pm 0.5$

The $I{C}_{50}$ value of $\beta$-boswellic acid could not be determined because, at concentration of 2000 $\mathsf{\mu }$M, 70% of Batx-I proteolytic activity was still observed. $\Delta G^{\circ }$ experimental was estimated from measurements of enzyme activity via the Cheng–Prusoff equation (Equation (1)). It should be noted that quantitative agreement between the experimental and theoretical $\Delta G^{\circ }$ values is not expected because the simulations suggest significant differences in the bound poses of the compounds, which are not captured by the Cheng–Prusoff equation, and, furthermore, the experiments were performed with Batx-I, for which no atomic structure is available; hence, the simulations used the homologous protein BaP1 [40].