We imagine you're dying to know how this intriguing molecule is structured...
The crystallography experiments carried out by Schmidt et al. in May 2012 enabled researchers to determine the structure of the dynein motor domain to a resolution of 3.3 Å.
The main core of the dynein motor is a ring of six AAA+ domains (AAA1-6), each of which can be divided into two subdomains (large and small). There are other structures relevant to the function of the motor: the microtubule binding domain (MTBD) and the linker, a motile element which is relocated in a process driven by ATP hydrolysis.
The MTDB is located at the end of a stalk which emerges from AAA4. The stalk also interacts with the buttress from AAA5. The linker is in contact with AAA1 and spans across the whole ring. The following cartoon representation (Fig. 2) illustrates the organization of the domains in the motor. The video (Fig. 3) shows the spatial folding of these domains. A top view of the motor domain is also provided (Fig. 4), with the same colour code as Fig. 2.
Figure 2: Spatial organization of the domains that
compose the dynein motor (Adapted from Figure 1a of Schmidt et al., 2012)
Figure 3: Animation showing the folding of the dynein motor domain
Figure 4: Top view of the dynein motor domain (N-tail not shown). The image has been obtained using a 2.8 Å crystal structure2 (see reference list). PDB ID: 3VKG
The linker deserves a section of its own!
It is divided into different subdomains, labelled in the following image (Fig. 5).
+Labelled.png)
+Labelled.png)
Figure 5: Structure of the linker that spans across the AAA+ ring (PDB ID: 3VKG)2
Subdomain 1 contains a hydrophobic pocket which accommodates a highly conserved Phe residue in AAA5L (Phe3466). This contact interaction is shown here (Fig. 6):
Figure 6: Contact interactions between AAA5L and the linker subdomain 1. The main residues that form the hydrophobic pocket in subdomain 1 are Val1366 and Met1415. These accommodate Phe3446 from AAA5L. There are also electrostatic interactions that stabilize the structure in this region (Arg3445 - Glu1411; Lys3438 - Glu1407)
(Schmidt et al., 2012). PDB ID: 4AKI
Subdomain 2 spans over the AAA+ ring, and subdomains 3 and 4 (C tail) interact with AAA1L through multiple conserved contact sites, as is shown in here (Fig. 7):
Figure 7: There is a large contact surface between the linker subdomains 3, 4 and the AAA1 subunit. (PDB ID: 3VKG)2
There is an additional conserved site in AAA2 (two β-hairpins). Although these do not contact the linker in the crystallized conformation of the protein, they are close to a cleft (also conserved) between subdomains 2 and 3 of the linker, shown in Fig. 8 and Fig. 9. This suggests that there may be interactions between these conserved regions in other conformations (at a different point of the mechanochemical cycle).
Figure 8: Cleft between the linker subdomains 2 and 3 (PDB ID: 3VKG)2
Figure 9: There are two conserved beta-hairpins close to the cleft (Schmidt et al., 2012) (PDB ID: 3VKG)2
Where does ATP bind?
The motor has conserved nucleotide binding sites between neighbouring AAA+ domians. There are several motifs that compose these binding sites:
- The Walker A motif emerges from the first large domain and binds phosphate groups.
- The Walker B motif, also from the first large domain, contains a magnesium-coordinating aspartate as well as the catalytic glutamate residue.
- Sensor 1, from the first large domain, has a hydrophilic residue that contributes to catalysis.
- The adenine base binding pocket is located between the large and small domains.
- Sensor 2 in the small domain contains conserved arginine residues that make further contacts to the ATP phosphate groups.
- Arginine finger in the neighbouring large domain also interacts with the phosphate groups in ATP.
Figure 10: A view of the ADP-bound nucleotide binding site between AAA1 and AAA2. This is the main ATP hydrolysis site and is in a low-nucleotide affinity conformation (Schmidt et al., 2012) (PDB ID: 3VKG)2
There are four nucleotide binding sites: AAA1-4. They are all similar but some differ in structural features that change their properties.
AAA1 has a more open conformation, and thus presents lower affinity for the nucleotide. This also makes the base more exposed to the solvent than in the other binding sites.
AAA2 binds Mg-ATP tightly. It lacks the catalytic glutamate in the Walker B motif. This position is occupied by a conserved arginine from AAA3 that interacts with an aspartate residue from the Walker B motif, holding the ATP molecule in place (See fig. 11)
Figure 11: The position of the catalytic glutamate in the Walker B motif of AAA2 is taken by Arg2549 from AAA3, which interacts with Asp2155 from the AAA2 Walker B motif, holding the phosphate groups from ATP in place. PDB ID: 4AKG (Schmidt et al., 2012)
The AAA3 binding sites presents all the residues involved in ATP binding in the expected orientation. The ATP molecules is more accessible to solvent in this site than in the AAA2 site (AAA3 has lower affinity than AAA2).
The AAA4 binding site shows unusual orientation of some residues such as the catalytic glutamate, which is angled away from the nucleotide by an alpha-helix (See fig. 12). This suggests that this site is unable to catalyse ATP hydrolysis. In this binding site the nucleotide base is buried but the ribose and the phosphate groups are exposed.
Figure 12: The catalytic glutamate (Glu2819) does not make contact with the nucleotide in AAA4, suggesting that this site cannot catalyse ATP hydrolysis. PDB ID: 4AI6 (Schmidt et al., 2012)
