(604i) The Influence of Phosphorylation on the Kinesin-5 Tail in Regulating Tail-Motor Interaction for Chromosome Segregation

Introduction: Kinesin-5 is an essential motor protein involved in mitotic spindle assembly, playing a significant role in cell division by ensuring proper chromosome segregation. It is composed of an N-terminal motor domain, a stalk domain that forms a coiled coil, and a C-terminal tail domain. While the motor domains exhibit a high binding affinity for microtubules (MTs), they show a lower affinity for the tail domain. Kinesin-5 motors move toward the microtubule plus ends. A key aspect of their function is mechanochemical gating through interactions between the C-terminal tail domains and N-terminal motor domains, which are crucial for regulating their activity. Tail-mediated kinesin-5 clustering is critical for organizing the forces generated by multiple kinesin-5 motors, resulting in MT crosslinking, alignment, and sliding. However, the regulatory mechanisms behind tail-to-motor clustering remain unclear. A molecular investigation of this interaction could uncover the underlying mechanisms of this vital process. Additionally, tail phosphorylation has been proposed as a potential regulator of tail-motor interactions and kinesin-5 activity. A deeper understanding of tail-motor dynamics following phosphorylation could enhance therapeutic strategies targeting this motor protein. This project aims to computationally predict the full structure of the Kinesin-5/Cut7 assembly and investigate the influence of tail phosphorylation on MT sliding.

Materials and Methods: In this study, we focus on the Schizosaccharomyces pombe kinesin-5/Cut7 sequence, which has been retrieved in FASTA format from the PomBase database. Sequence analysis has been conducted using BLAST to identify homologous proteins as potential templates. To achieve a more accurate structural prediction, a hybrid strategy has been employed, combining homology modeling with Modeller and deep learning-based prediction via AlphaFold3. Subsequently, full-atomistic molecular dynamics (MD) simulations have been performed to elucidate the secondary structure of both the wild-type and tail-phosphorylated assemblies, with the aim of obtaining a reliably equilibrated structure under physiological conditions using the CHARMM36 force field. Given the crucial role of the interaction between microtubules and the kinesin-5 tail in the mechanochemical cycle of kinesin-5 motors, microtubules have been incorporated into the model to accurately represent this interaction. However, all-atom simulations of such large systems are, in practice, computationally expensive and time-consuming. Therefore, we have adopted an alternative approach involving coarse-grained (CG) molecular dynamics simulations combined with Elastic Network (EN) models to accelerate the MD simulation process. Since MTs sliding is driven by the movement of the kinesin-5 motor domain toward the plus end of microtubules, employing non-equilibrium molecular dynamics techniques to investigate this process is essential. To examine the binding strength of tail-motor interactions during this movement, steered molecular dynamics (SMD) has been utilized. By applying an external pulling force, SMD mimics the mechanical forces involved in kinesin-5-mediated MT sliding, allowing for quantification of the interaction strength governing motor-driven MT dynamics. To gain a comprehensive understanding of the impact of phosphorylation on MT sliding, comparing the interactions between wild-type and phosphorylated tail-motor complexes is crucial. Therefore, a full atomistic phosphorylated model of Cut7 has been modeled using CHARMM-GUI. Phosphorylation parameters for the studied phosphorylated amino acids, serine and threonine, are available in the CHARMM36 force field. Since the Martini3 force field does not natively include phosphorylated amino acids, mapping and parameterization of phosphorylated serine and threonine residues have been conducted, with the phosphate group (PO₄⁻) represented as a separate, negatively charged bead. All simulations and corresponding analyses have been performed using the GROMACS software package.

Results and Discussion: BLAST similarity searches against the Protein Data Bank have revealed structurally homologous proteins with perfect alignment and 100% sequence identity, but only for the motor domain. After identifying homologous proteins as potential templates through BLAST, homology modeling using Modeller has provided a reliable motor domain model. However, Modeller was unsuccessful in finding any homologs for the rest of the sequence. Given the availability of crystallographic conformers of tetrameric bipolar assembly (BASS) domains with different amino acid sequences, the stalk domain of the full Cut7 sequence has been replaced with an existing BASS domain that possesses the highest-resolution crystallographically available structure in the PDB (PDB ID: 7S5U), while retaining the Cut7 tail sequence. Subsequently, using AlphaFold3, the full assembly of the Kinesin-5 polypeptide has successfully captured the dimeric motor domains and the homotetrameric BASS domain, with a length of ~31 nm. Due to the insufficient resolution of the available Cryo-EM data, the experimental secondary structure of the tail remains unresolved, and none of the preprocessing steps can provide information on the tail's secondary structure. Therefore, MD simulations have been performed to reveal the folded tail conformations within the wild-type and phosphorylated assemblies. The wild-type assembly of the Kinesin-5 polypeptide contains 21,192 atoms and a total of 2,708 residues. The simulations have confirmed that the tail region exhibited the most significant changes, as it was the most fluctuating area based on our analysis. After embedding the obtained assembly onto the microtubules, SMD simulations on the CG models have been performed, confirming that introducing negative charges from phosphorylated residues may increase the affinity for the positively charged motor domain. This enhancement of the tail-motor interaction is essential for the kinesin-5 motor's ability to crosslink and slide microtubules.

Conclusions: Our investigation uncovers the secondary structure of the kinesin-5/Cut7 tail domain and its interaction with the motor domain as it moves toward the plus end of microtubules. Functional studies show that full-length kinesin-5 motors exhibit slow motility and cluster along microtubules, whereas tail-deleted motors move rapidly without clustering, highlighting the tail’s crucial role in microtubule sliding. These findings suggest a regulatory mechanism in which the tail domain directly interacts with the motor domain, ultimately increasing the motor's affinity for microtubules after phosphorylation of the kinesin-5 tail.