1.1.2 Peptides
A peptide is composed of amino acids. In general, there are 20 different amino acids
commonly found in peptides and proteins. Each of them is formed by an amino group,
a carboxyl group, a central CH group (the carbon of this central CH group is usually
called α-carbon or Cα) and a specific side chain (Figure A.1, Appendix A). The
sequence of Cα atoms connected through covalent peptide bonds, including the Nterminus
(free NH2 initial group) and C-terminus (free COOH final group) is called
the peptide backbone. It is the main structural part of the peptide that determines
its overall geometric properties. The side-chains of a peptide define its physical and
chemical properties.
The structure of a peptide or a protein can be described at different levels (Figure A.2,
Appendix A). The primary structure of the peptide describes the actual sequence of
amino acids, or residues, within the peptide. The term secondary structure refers to the
geometry or conformational behaviour of this primary sequence. A disordered peptide
chain is often called a random coil. However, many peptides have well-defined
three-dimensional secondary structure. Three of the most frequently occurring structures
are the α-helix, the β-sheet and β-turns. In a larger protein, the three dimensional
arrangement, or packing of secondary units, is characterized by the tertiary structure,
whereas assemblies of several proteins are classified as quaternary structures.
A common secondary structural motif in biologically active peptides is the amphipathic
α-helix. An α-helix is formed when a chain of amino acids twists around itself
in a well-ordered way (Figure 1.3(a)). This helical structure is stabilized by a network
of backbone hydrogen bonds between the backbone carbonyl oxygen of residue i and
the amide proton of residue i+4, with the side groups of the amino acid residues protruding
outward from the helical backbone. The rise along the helical axis for every
two successive α-carbons is 1.5 A and the respective rotation is about 100 degrees. ˚
Moreover, each helical turn extends for about 5.4 A along the long axis of the helix, ˚
resulting in 3.6 residues per turn (Figure 1.3(a)). Another important feature of an α-
helical peptide is the inherent net dipole that exists along its axis due to the synergy
of each of the small dipoles that exist in each peptide bond (Figure 1.3(b)). The helical
dipole plays an important role in pore formation and stabilization and ion transport
across membranes.
An α-helix is called amphipathic when it has both hydrophobic and hydrophilic residues
positioned along its axis. This distribution of hydrophobicity has been shown to play
an important role in the way with which the amphipathic α-helical peptides interact
with the biological membranes [6]. Amphipathic α-helices are the peptides of interest
in this study and more details about their function will be given in the next chapters.
Figure 1.3: The α-helix. (a) Ball-and-stick representation of an α-helix, showing the
hydrogen bonds between the ith and ith + 4 residues. (b) A helical dipole is created by the
transmission of the electric dipole of the peptide bonds along the helical axis. The figure
has been adapted from [5].
1.2 Peptide-membrane interactions
Peptide-membrane interactions are at the heart of a number of important biological
processes. For example, antimicrobial peptides are a family of peptides with a particular
propensity to recognize and disintegrate bacterial pathogens. A number of these
peptides have been identified as key components of the natural immune defence system
[7]. A related family of peptides is the so-called cell-penetrating peptides (CPPs)
capable of efficient translocation through the cell membrane, either by themselves or
together with a molecular cargo [8]. These peptides are being explored as potential
programmable drug delivery vectors. As a part of larger proteins, ion-conducting
channel peptides form well-organized transmembrane bundles capable of selective
transport of ions. Other peptides are believed to play a key role in mediation of various
complex cellular processes, such as membrane fusion. It is clear that a better understanding
of peptide-membrane interactions on molecular level not only is important
in the elucidation of various biological processes, but also could be instrumental in
designing peptides with tailored functionalities, for example, for antibiotic and drug
delivery applications.
Peptide-membrane interactions are complex and beautifully diverse phenomena. Depending
on their composition, charge, and structure different peptides evoke different
interaction mechanisms with the membrane. Here, I present some of the most commonly
seen scenarios in the studies of peptide-membrane interactions. In this analysis,
I exclusively focus on α-helical peptides. There are two reasons for this. First, α-helical
secondary structure is abundant among membrane active peptides. Second, development
of a fully comprehensive description of peptide-membrane interactions is a challenging
task. Having the peptide in a well-defined structure eliminates at least one
additional degree of complexity associated with the conformational behaviour of the
peptide itself. Thus, α-helical peptides are a natural starting point in the construction
of this description.
Let us first consider different peptide internalization mechanisms. These mechanisms
can be categorized into endocytosis mediated entry and direct penetration in the membrane.
Endocytosis is an important biological process, used by the cell for transport of
various molecular species across the cell membrane. In the case of peptide transport,
its mechanism can be described as follows; first, several peptides form an aggregate in
the aqueous phase, then the cell absorbs the aggregate from the outside environment
by engulfing it with its cell membrane and finally, a vesicle (endosome) is formed
and released on the inner side of the membrane (Figure 1.4(a)). In principle, spontaneous
formation of an endosome is possible as a result of membrane fluctuation and
budding. However, more commonly, endocytosis is a receptor mediated and energy
dependent process. Several classes of cell-penetrating peptides are believed to induce
this mechanism.
Direct penetration mechanisms, on the other hand, are receptor and energy independent,
and may also be classified in several distinct scenarios. One of these mechanisms
is the sinking raft model. In this model, the peptides form aggregates of limited size
and associate with one of the faces of the membrane. The mass imbalance of the lipid
bilayer due to this association induces curvature that provides the driving force for
the translocation of peptides across the bilayer [14] (Figure 1.4(b)). This mechanism
has been proposed for several antimicrobial peptides, for example delta-Lysin [10] .
Another scenario of direct penetration is the formation of an inverted micelle. In this
case, a peptide interacts with the negatively charged phospholipids, inducing the formation
of an inverted micelle inside the lipid bilayer (Figure 1.4(c)). Then, either the
peptide is entrapped within the micelle and then released into the cell, or the formation
of the micelle perturbs locally the membrane and induces a new peptide insertion
event.
The formation of transmembrane pores is another way of interaction between α-helical
peptides and membranes. Three different pore structures, the barrel-stave, the carpet
and the toroidal pore model, have been proposed and investigated. The main differences
between these models lie in the lipid structure around the pores and the pore
stability. In the barrel-stave model, the lipids are parallel to each other and the peptides
form a well-defined, very stable bundle, which, if it is of a sufficient diameter, can
serve as a pore. This is believed to be the structure of the peptides in ion-conducting
channels, either as a part of a larger protein, or formed through a self-assembly process.
In the case of the toroidal model, the lipids create a toroidal-shaped (or donutshaped)
opening covered with the peptides in different orientations. Toroidal pores
are generally less stable (i.e. they are transient) than the barrel-stave pores. Some studies
suggest that this mechanism is involved in membrane disruption action of some
antimicrobial peptides, leading to cell leaking out its contents [15].
In the carpet model, peptides accumulate on the membrane until its integrity is breached
and transient holes are formed. These holes, when the peptides are in high concentrations,
may result into the complete collapse of the membrane. Again, this mechanism
has been proposed, among others, as permeabilization mechanism of α-helical antimicrobial
peptides (AMPs) (Figure 1.4(d)).
Several peptide-membrane interaction mechanisms involve a peptide inserted in the
membrane. In membrane fusion, the fusion peptides, short hydrophobic parts of fusion
proteins, destabilize the lipid bilayer structure by adopting an oblique orientation
within the membrane [6]. This orientation has been linked to the gradient of
hydrophobicity along the helical axis of the peptides. In Figure 1.4(e), I show one of
the proposed scenarios, as presented in [12]. Other peptides, with different distribution
of hydrophobicity, adopt interfacial or transmembrane orientations relative to the
membrane.
Many mechanisms include several stages of interaction between peptides and membrane
or between different peptides. For example, in the case of Pep-1, a cell-penetrating
peptide with high efficiency as drug-delivery vector, one of the proposed internalization
mechanisms consists of four steps: a number of Pep-1 helices form a complex
with the cargo, other helices form a pore, the cargo-peptide complex passes through
the pore and finally it is released into the cytoplasmic side [13] (Figure 1.4(f)).
The mechanisms mentioned above have been validated for som