Mesoscopic models of lipid bilayers and bilayers with embedded ...

12 Simulation method for coarse-grained lipids
a timestep at least three orders of magnitude larger than in atomistic molecular dynamics
simulations, where a timestep of the order of few fs is typically used.
In the next section we describe the coarse-grained model we use to represent lipid
molecules and we show that these model lipids, simulated with DPD, spontaneously
self-assemble into micelles and bilayers.
2.2 Coarse-grained model of lipids
Within the mesoscopic approach, each molecule of the system (or groups of molecules)
is coarse-grained by a set of beads. We consider three types of beads: a water-like
bead (denoted as ’w’), which models the solvent, an hydrophilic bead (denoted as ’h’)
which models a section of the lipid headgroup, and an hydrophobic bead (denoted
as ’t’) which models a segment of the lipid hydrocarbon tail. A lipid is constructed
by connecting head- and tail-beads with springs. The simplest lipid consists of a linear
chain of one hydrophilic head-bead and one tail of hydrophobic beads. A more
realistic model of a phospholipid can be constructed by connecting two hydrophobic
tails to an headgroup consisting of one or more head-beads. In both single- and
double-tail lipids the tail(s) can have different length. We denote a single-tail lipid
with one head-bead and n tail beads as htn, and a double-tail lipid with m headbeads
and n tail-beads as hm(tn)2 (see figure 2.1).
Figure 2.1: Schematic drawing of the single- and double-tail model lipids described in the text
and their nomenclature. The black particles represent the hydrophilic head-beads and the
white particles the hydrophobic tail-beads.
A mapping of coarse-grained lipids onto real phospholipids can be established
through the factor Nm. We have chosen a mapping factor of Nm = 3, corresponding
to three water molecules represented by one DPD-bead of volume of 90 ˚A 3 . In terms