† Corresponding author. E-mail:
Docking of the kinesin’s neck linker (NL) to the motor domain is the key force-generation process of the kinesin. In this process, NL’s β10 portion forms four backbone hydrogen bonds (HBs) with the motor domain. These backbone hydrogen bonds show big differences in their effective strength. The origins of these strength differences are still unclear. Using molecular dynamics method, we investigate the stability of the backbone HBs in explicit water environment. We find that the strength differences of these backbone HBs mainly arise from their relationships with water molecules which are controlled by arranging the surrounding residue sidechains. The arrangement of the residues in the C-terminal part of β10 results in the existence of the water-attack channels around the backbone HBs in this region. Along these channels the water molecules can directly attack the backbone HBs and make these HBs relatively weak. In contrast, the backbone HB at the N-terminus of β10 is protected by the surrounding hydrophobic and hydrophilic residues which cooperate positively with the central backbone HB and make this HB highly strong. The intimate relationship between the effective strength of protein backbone HB and water revealed here should be considered when performing mechanical analysis for protein conformational changes.
Conventional kinesin[1] (kinesin-1, here is referred to as kinesin) is a highly processive motor protein, which effectively converts the chemical energy carried by adenosine triphosphate (ATP) into mechanical force and walks hundreds of steps along a microtubule with cargos.[2–5] Kinesin’s neck linker (NL) docking to the motor domain is the key force generation step of kinesin.[6] The NL consists of
In kinesin walking cycle, NL undergoes a large conformational change from the undocked state to the docked state repeatedly (Fig.
The difference between the unbinding forces in the above two studies with implicit and explicit water models may be attributed to the difference in simulation times. However, another reason for the difference might be due to the way the role of water is described. In this work, we set up an MD model with kinesin motor head surrounded by explicit water molecules and investigate the origin of the strength difference of β10’s backbone HBs. We find that the effective strength of a backbone HB is mainly determined by whether its hydrogen bonding sites could be attacked by water molecules. The high strength of the N-terminal ASN latch is achieved due to the perfect protection against water attack from both sides of the β-sheet and the C-terminal backbone HB has two water attack channels that make it rather weak. The water-dependence of the effective strength of protein backbone HBs must be taken into account when performing mechanical analysis for proteins.
The MD simulations were performed by using NAMD (version 2.9)[14] with force field CHARMM.[15] Modeling and data analysis were performed with VMD (version 1.9.2).[16] The non-bonded Coulomb and van der Waals interactions were calculated with a cutoff using a switching function starting at a distance of 20 Å and reaching zero at 22 Å. The structural model for simulation is based on a crystal structure of rat kinesin (PDB ID: 2KIN).[10] The 2KIN’s β7 (the residues after Ala339), ADP and SO4 were deleted since they had little relation with interactions between β10 and motor domain. Because L11 loop could not affect β10-motor domain interaction, the missing part of L11 loop was ignored. The Val239 and Asn256 were directly connected in the modeling. The motor domain was surrounded by explicit water molecules and the spherical boundary condition was used. The water model is TIP3P.[17] To mimic the microtubule-bound state, we fixed the Cα atoms of Asn264, Ser267, Lys315, and Met319 in the simulation. The system was minimized to 30000 steps and the MD simulation took 40 ns. The system was simulated 11 times to perform statistical analysis. Molecular drawing was produced by using Discovery studio 3.5 visualizer.
A backbone HB in protein is formed between the backbone carbonyl oxygen and the backbone amide hydrogen. In the docked state, the β10 portion of NL forms four backbone HBs with the motor domain, including HB1 (Glu336:O–Lys223:H), HB2 (Glu336:H–Lys223:O), HB3 (Asn334:O–Ser225:H), and HB4 (Asn334:H–Gly77:O) (Fig.
A detailed analysis of the trajectories shows that the instability of HB1 arises from water attacks. The HB1 is located at the C-terminus of β10. In Fig.
The possible water attack region around the four HBs is divided into three sides, A, B, and C, respectively (as shown in Fig.
Glu336 and Lys223 form two backbone HBs, i.e., HB1 and HB2. The bonding ratio of HB2 is 80%, indicating that it is an effective HB. Structural analysis shows that there is a water channel formed by Lys223, Ser225, Asn334, and Glu336 at the B side of HB2 that allows direct water attack on this backbone HB. However, the water attack along this channel is barely effective due to the constraint from two adjacent backbone HBs (HB1 and HB3). The breaking of HB2 has a correlation with the status of HB1. As seen from the distance–time curves of
At the N-terminus of β10, Asn334 forms two backbone HBs with Ser225 (HB3) and Gly77 (HB4). The bonding ratio of HB4 is 97%, meaning that it is highly stable. However, the bonding ratio of HB3 is 78%, even lower than that of HB2 (80%). Figure
The HB4 is the central HB of the ASN latch which has a high effective strength. In the unbinding case,[11,12] ASN latch shows strong resistance to the unbinding force. The high stability of ASN latch is achieved through the cooperation of the residues arranged exquisitely (Fig.
Kinesin is a molecular walking device working in water environment. The movement of kinesin is achieved through a series of conformational changes, in which many nonbonding interactions, especially backbone HBs, are formed or broken. The result of this work shows that water is highly important in controlling the effective strength of backbone HBs. This knowledge must be taken into account in understanding the design principle of kinesin.
The average bond energy of a protein backbone HB is ∼25 kJ/mol. Taking the force-range of a HB to be ∼1 Å, the average force of a backbone HB is ∼410 pN. In this calculation, the water influence is not taken into account, i.e., the direct breaking of an HB in the absence of water needs a force
The backbone HBs play a major role in the mechanical behavior of proteins. Since the effective strength of backbone HBs is highly water-dependent, this water dependence must be taken into account in the analysis of the mechanical behavior of proteins. Then, how to quantitatively describe the water dependence of backbone HBs becomes an unavoidable question. Due to the complicity concerning the properties of the sidechains of the related residues and their relationship, this question is still a challenging open question.
The NL docking is the key force-generation process of kinesin, which is accomplished in three steps including the initiation step of the first three residues forming an extra turn, the docking of β9 via the CNB mechanism, and the docking of β10. Unlike the first two steps, the docking of β10 takes a zipper mechanism, i.e., the four backbone HBs form one after another automatically. As shown in the above section, neighboring backbone HBs have close correlations. Formation of one backbone HB will largely promote the formation of the neighboring one. Hydrogen bonds are of short-range interactions with force-range within 3 Å. Once a backbone HB is formed, the donor and acceptor atoms of the neighboring HB are restricted in a small space so that the formation probability of the neighboring HB is greatly increased. A backbone HB might be attacked by water molecules from all possible channels. Formation of one backbone HB will lessen the water attack on the neighboring HB. Evidently, to ensure efficient docking of NL, the formation of the initial backbone HB of the zipper is crucial. This HB should be strong and stable so that zippering up process could go on efficiently. In the design of kinesin, the initial backbone HB of the β10 zipper is HB4. As shown above, HB4 is the strongest and stablest backbone HB of the β10 zipper. Once HB4 is formed, it is hard to break. In one of our previous works,[8] we find that more than one-third of the work done in unbinding the entire NL is for deposition in the process of opening the ASN latch. The design with HB4 serving as the first HB of the β10 zipper ensures the efficient docking of NL.
It should be pointed out that the high strength of HB4 does not mean that its spontaneous formation could take place easily. The simulation work with implicit water by Hwang et al.[12] shows that CNB bends toward the binding pocket nearly deterministically, but the docking of β10 from an undocked state, especially the latching of the ASN latch, was not observed within the simulation time. In our simulation with explicit water, we capture the spontaneous docking of β10 starting from a CNB docked and β10 undocked structure. Figure
Life arises from water. The intimate relationship between protein backbone hydrogen bonds and water as revealed here shows that the protein properties are determined together with water. Therefore, to obtain a real understanding of protein behavior, one must figure out those protein–water interaction details.
The authors would like to thank the Computer Network Information Center of Chinese Academy of Sciences and the Shanghai Supercomputer Center of China for computing service.
Distance–time curves of
[1] | |
[2] | |
[3] | |
[4] | |
[5] | |
[6] | |
[7] | |
[8] | |
[9] | |
[10] | |
[11] | |
[12] | |
[13] | |
[14] | |
[15] | |
[16] | |
[17] | |
[18] |