The second environmental factor to address is solvent. Water
represents the bulk solvent media in most samples of interest,
particularly those of biological origin. Classical simulation method-
ology can represent solvent either explicitly or implicitly. Explicit
solvent uses many copies of discrete water molecules, empirically
parameterized to reproduce experimental properties of the pure
liquid. Although numerous water models are available, the most
widely used is TIP3P [26], which reasonably balances the accurate
reproduction of bulk solvent behavior with computational expense.
Alternatively, implicit solvent uses a mathematical function to pro-
duce a continuum around the solute with the dielectric and hydro-
phobic properties of water [27]. Implicit solvent is generally more
computationally efficient than explicit solvent; however, it is often
not the best approach for the study of protein–drug systems. Many
drugs interact with their protein receptors through water-mediated
hydrogen bonds, an important aspect of binding that only explicit
solvent can capture (seeNote 7). Notably, the crystallographic
structure for the example system in this chapter exhibits two
water-mediated hydrogen bonds, which bridge drug binding with
residues Tyr136 and Thr330. To ensure that these and other highly
conserved water interactions are in place going into the simulation
phase of the project, all crystallographically resolved water mole-
cules should be retained in the initial model (seeNote 8).
With respect to bulk solvent, MD setup packages offer standard
routines to immerse the biomolecular system of interest in a box of
explicit water molecules, but the user must take care to abide by the
minimum image convention and ensure that the box dimensions
are large enough to accommodate the system following equilibra-
tion of water density. A distance of at least 15 A ̊ between the
protein–drug complex and the box edges is recommended (see
Note 9). If the protein–drug complex is likely to tumble during
the planned timescale of the simulation, the generated water box
should be cubic (seeNote 10).
The third environmental factor to address is salt concentration.
Studies intending to investigate systems within a biological context,
i.e., interactions of drugs with their receptors within the human
body, should use physiological salt concentration of 150 mM NaCl.
Studies seeking to investigate systems with the context of an exper-
imental setup, mimicking the conditions of a drug assay or biosen-
sing reaction, should employ a salt concentration that reflects that
setup (seeNote 11). For the example system in this chapter, which
aims to investigate the interaction of an aryl acylamidase with the
drug acetaminophen within a biological sample, such as a patient’s
blood, physiological salt concentration is used.
MD setup packages offer standard routines to introduce a
desired salt concentration, either via a mathematical constant in
implicit solvent, or by substituting water molecules for salt ions in
explicit solvent. To reduce the equilibration time required for250 Jodi A. Hadden and Juan R. Perilla