3D Structure of Antimicrobial Peptides
1. Structure annotation
The information on the three-dimensional (3D) structure of antimicrobial peptides was annotated in this database from the first version of the APD (Wang & Wang, 2004) based on the Pub-Med with a link to the Protein Data Bank (PDB). It includes unique structures of PDB-deposited and non-deposited AMPs free in water or in complex with its targets, usually membranes. The annotated information ranges from methods for structural determination, structural class and structural regions and critical residues to the type of membrane-mimetic environments, such as micelles and organic solvents. The structure can be viewed via the link to the Protein Data Bank (PDB) for each peptide entry when the coordinates are deposited. When there are multiple structures, the link is usually given to the structure with the highest resolution. Other structures solved for the same peptide under different conditions can be viewed via the PDB links. To study the structure in detail, you can download the coordinates and view the structure in your computer.
2. Structural determination of antimicrobial peptides
There are two methods for determining the 3D structures of antimicrobial peptides to the atomic resolution: X-ray crystallography and nuclear magnetic resonance (NMR) spectroscopy. According to the statistics of the APD database, most of the 3D structures of AMPs were determined by 2D solution NMR spectroscopy (Wuthrich K, 1986). For small peptides, this method is usually sufficient. It is also helpful to extend the observation to heteronuclei 15N and 13C without isotope labeling. The improved 2D NMR method can be essential to avoid misleading structure or dynamics for some peptides (Reviewed by Wang, 2013). This improved 2D NMR method requires the collection of additional NMR spectra that measure 13C and 15N chemical shifts at natural abundance. These heteronuclear 13C and 15N NMR chemical shifts are then used to validate proton assignments and refine the 3D structure. For more complex molecules, 3D NMR methods are needed using isotope-labeled peptides. For example, see the structure determination of human LL-37.
3. Structural classification of antimicrobial peptides
Host defense antimicrobial peptides are classified into four families (Wang, 2017): alpha (α), beta (β), alphabeta (αβ), and non-alphabeta (non-αβ) based on the types of secondary structures. The α family consists of AMPs with helical structures (e.g. magainins and LL-37). The β family is composed of AMPs with beta-strands (e.g. human α-defensins). While the αβ family comprises both α-helical and β-strands in the 3D structure (e.g. β-defensins), the non-αβ family contains neither α-helical nor β-strands (e.g. indolicidin). These four families of AMP structures are represented in the main page of this database. Additional examples are provided below to illustrate structural diversity.
4. Structural viewing of antimicrobial peptidesYou can rotate and view the 3D structure of each antimicrobial peptide from the APD if the coordinates are deposited in the PDB. For example, you can view the LL-37 structure here.
5. Structural statistics of antimicrobial peptidesA list of antimicrobial peptdes with a defined 3D structure can be obtained by searching the APD database. The statistics of a variety of structures is also provided in this database. Go to the Statistic Interface.
6. Structural citationIt took a lot effort to solve a structure. As a consequence, we suggest the citation of the original article that reports the 3D structure (for examples, see below). The reference can be obtained from the same PDB link. It is also proper for you to cite the APD and PDB if you obtain multiple entries from these resources. The sufficient citations may include both the database IDs and references that generate the structures.
7. Structural diversity of antimicrobial peptides
Here are the PDB IDs and references for the structures above:
(A) Micelle-bound aurein 1.2 from Australian Bell frogs (PDB ID 1VM5; Pub-Med Wang et al., 2005);
(B) Fish pardaxin 4 (PDB ID 1XC0; Pub-Med Raimondo et al., 2005);
(C) Micelle-bound human cathelicidin LL-37 (PDB ID 2K6O; Pub-Med Wang G, 2008);
(D) Amphibian distinctin in water (PDB ID 1XKM; Pub-Med Raimondo et al., 2005);
(E) Caenorhabditis elegans caenopore-5 (PDB ID 2JS9; Pub-Med Mysliwy et al., 2010);
(F) Micelle-bound lactoferrin B (PDB ID 1LFC; Pub-Med Hwang et al., 1998);
(G) Human alpha defensin 1 HNP-1 (PDB ID 3GNY; Pub-Med Wei et al., 2009);
(H) Human beta defensin 1 HBD-1 (PDB ID 1E4S; Pub-Med Bauer et al., 2001);
(I) Bacterial lasso peptide (PDB ID 3NJW ; Pub-Med Nar et al., 2010);
(J) Plant cyclotide kalata B1 (PDB ID 1KAL; Pub-Med Saether et al., 1995);
(K) Rhesus theta defensin 1 RTD-1 (PDB ID 2LYF; Pub-Med Conibear et al., 2012);
(L) Bovine cathelicidin indolicidin (PDB ID 1G89; Pub-Med Rozek et al., 2000).
Wang, G.* (2013) Database-guided discovery of potent peptides to combat HIV-1 or superbugs. Pharmaceuticals 6, 728-758. View the colored figure in the review article.