Amyloid Precursor
Protein E2 domain in complex with Death Receptor 6 cysteine-rich domain (PDB
ID: 4YN0) from Mus musculus
Created by: William T Mills IV
In recent years the Amyloid Precursor Protein (APP) (PDB ID: 3UMH), an integral membrane protein, has garnered great interest from scientists for its links to Alzheimer’s Disease, a neurodegenerative disease marked by dementia and the accumulation of β-amyloid plaques in the brain. Studies have found that a crucial step in the development of Alzheimer’s Disease is proteolysis of APP which generates the neurotoxic β-amyloid peptide (1). Trophic factor deprivation triggers β-secretase to cleave off a fragment of APP (N-APP), which is then able to act as a ligand for membrane proteins. Studies have found that APP acts as a ligand for Death Receptor 6 (DR6) (PDB ID: 3QO4) to form a complex (MW: 47867.27 Da, pI: 7.61, PDB ID: 4YN0) that triggers axonal degradation via the caspase-6 pathway (2, 3). The process of axonal pruning and neuronal cell death is a natural and necessary process during the development of the central nervous system, but unfortunately the activation of these pathways in adulthood leads to neurodegenerative diseases like Alzheimer’s Disease and Amyotrophic lateral sclerosis (ALS) (4). The link between DR6 and ALS has been indicated by elevated levels of DR6 in post-mortem samples from the spinal cords of humans with ALS (4). Recent advances in the understanding of the role of DR6 and APP in the progression of neurodegenerative diseases have provided a new avenue for the development of a cure for these devastating diseases.
DR6, also known as tumor necrosis factor receptor super family 21 (TNFRSF21), is a single-pass transmembrane protein located in the mammalian central nervous system (5). TNF receptors are characterized by their influence on biological processes such as inflammation, cell survival and apoptosis (6). These receptors are single-pass transmembrane proteins and characteristically contain between one and six cysteine-rich domains (CRDs) (7). The ectodomain of DR6 is comprised of 4 CRDs in an extended arrangement with a kink between CRD2 and CRD3. The majority of the secondary structure of DR6 is in the form of random coils stabilized by disulfide bonds that give the CRDs their name. CRD3 of DR6 is further stabilized by the coordination of a magnesium ion with the side chain of Ser-117 and the main-chain carbonyls of Cys-150, Pro-151, Trp-154, and Val-179 (8). APP-E2’s secondary structure consists of five long alpha helices in a helix-turn-helix organization. The characteristics of the secondary structure of each protein that comprises the APP/DR6 complex contributes to various aspects of the unique interface between the two proteins.
The interface of the APP/DR6 complex exists mainly between the first and second helices (H1 and H2, respectively) of the E2 domain of APP and the CRD1 and CRD2 modules of DR6 and is relatively small (~680 Å2) (8). This major interface is made up of a series of hydrophobic and hydrophilic interactions. The first hydrophobic interaction, starting near the random coil connecting H1 and H2 and moving downward, is marked by the insertion of Leu-85DR6 into a pocket formed by three lysine residues on the surface of APP (Lys-350APP, Lys-353APP, Lys-354APP). Following this hydrophobic interaction, a hydrophilic interaction exists in the form of two salt bridges between Arg-86DR6 and Glu-342APP and between Asp-68DR6 and Lys-354APP. Next, there is a hydrophobic interaction in the form of a patch between Pro-71DR6 and Ala-72DR6 and Met-335APP and Trp-338APP. The major interface is further stabilized by the formation of two hydrogen bonds between Glu-100DR6 and Gln-361APP and between the main chain carbonyl of Gly-52DR6 and Gln-361APP. In addition to the major interface between the E2 domain of APP and the first and second CRDs of DR6, a minor interface exists near the N-terminus of the H1 helix of APP and the third CRD of DR6 in the form of a salt bridge between Arg-157DR6 and Glu-310APP. The significance of these residues in the formation of the APP/DR6 complex is supported by the conservation of these residues across different species (8).
The weak
forces that maintain the tertiary structure of DR6 and APP allow for increased
flexibility and conformational changes that help improve the interaction
between the two proteins and thus their ability to initiate downstream
signaling. Specifically, the helix-turn-helix arrangement of the E2 domain of
APP allows for conformational changes between the H1 and H2 helices to better
align with DR6. Additionally, the kink between CRD2 and CRD3 of DR6 allows for
some degree of movement that brings the third and fourth CRDs of DR6 closer to
APP to help form the minor interface between CRD3 of APP and the N-terminus of
the APP E2 domain (8).
A number of
site-directed mutagenesis experiments have been preformed to determine the functional
significance of several important residues along the interface of the APP/DR6
complex. One such experiment examined the differences between APP and its close
homolog APLP2, which is unable to bind DR6. Sequence alignment of APP and APLP2
revealed 2 nonconserved residues (Met-335APP and Lys-354APP)
in the E2 domain of these structures that may explain their different binding
affinities for DR6. It was found that the single mutation from Lys to Met at
residue 335 in APLP2 was able to confer DR6 binding in APLP2. Conversely, the
single mutation from Met to Lys at residue 335 in APP was able to eliminate
APP’s binding capabilities with DR6 (8). These results highlight the importance
of Met-335APP in the formation of the APP/DR6 complex and distinguish
it as a possible therapeutic target for drugs aimed at preventing the
development of Alzheimer’s Disease.
The Dali server is a
database-searching tool that allows scientists to compare the three-dimensional
structure of a protein of interest with the three-dimensional structure of all
of the proteins cataloged in the Protein Database (PDB). The Dali server uses
the weighted sum of the intermolecular distances between similar structures to
assign a z score to possible homologous structures; z scores above 2 indicate
significant similarity between two structures (9). This tool allows scientist
to explore structures that are found in multiple different proteins and
generate hypotheses about how these similar structures relate to protein
function. Because the Dali server is used for comparing the tertiary structure
of polypeptides it does not allow for the comparison of nucleic acids.
The Position-Specific Iterative
Basic Local Alignment Search Tool (PSI-BLAST) is a resource created by the
National Library of Medicine that allows for the comparison of the primary
structure of a protein of interest with a database of polypeptide sequences. PSI-BLAST
calculates the number of random alignments in the BLAST database that would
have a better sequence alignment than that between the protein of interest and
a possible homolog (10). PSI-BLAST assigns an e score to possible homologous
structures to represent the likelihood that the alignment occurred by chance; an
e score below 0.015 represents significant likelihood that the alignment
between the two proteins did not occur by chance (11). This tool allows
scientists to compare the primary structures of proteins to look for conserved
and non-conserved residues, which may indicate the importance of specific
residues in the function of a protein. The National Library of Medicine has
also created a resource for comparing the sequence of nucleic acids, n-BLAST, which
may help scientists determine evolutionary relationships between structures or
organisms based on their nucleic acid sequence.
Analysis of the APP/DR6 complex with the Dali Server and PSI-BLAST generated several proteins with similar three-dimensional structures and amino acid sequences. One of these proteins, the LIGHT/DcR3 complex (PDB ID: 4J6G), produced z and e scores of 16.7 and 1 x 10-24, respectively (12, 13). These values indicate extensive similarities between the LIGHT/DcR3 complex and the APP/DR6 complex with respect to three-dimensional structure and amino acid sequence. Like DR6, DcR3 (PDB ID: 3MHD) consists of 4 cysteine-rich domains, two of which create the interface of the LIGHT/DcR3 complex (CRD2 and CRD3). Several loops on the surface of LIGHT (PDB ID: 4EN0) are responsible for interacting with the CRDs of DcR3 through a series of hydrophobic and ionic interactions similar to those seen in the APP/DR6 complex. These diverse interactions allow for increased specificity of the substrate and greater binding affinity of the ligand to influence biological processes. Unlike the APP/DR6 complex, however, the protein containing the CRDs in the APP/DR6 complex (DR6) is a transmembrane protein while the protein containing the CRDs in the LIGHT/DcR3 complex (DcR3) is a soluble, secreted protein. The LIGHT/DcR3 complex, like the APP/DR6 complex, is involved in immune response. Unlike how binding of APP to DR6 initiates downstream signaling and generation of the immune response, DcR3 binds LIGHT to inhibit the generation of the immune response, seen in the attenuation of the T cell-mediated immune response, by competing with the binding of LIGHT to other receptors such as HVEM (PDB ID: 1JMA) and LTβR (PDB ID: 4MXV) that generate the immune response (7).