Effect of charge, hydrophobicity, and sequence of nucleoporins on the translocation of model particles through the nuclear pore complex

Mario Tagliazucchi, Orit Peleg, Martin Kröger, Yitzhak Rabin, Igal Szleifer
2013 Proceedings of the National Academy of Sciences of the United States of America  
The molecular structure of the yeast nuclear pore complex (NPC) and the translocation of model particles have been studied with a molecular theory that accounts for the geometry of the pore and the sequence and anchoring position of the unfolded domains of the nucleoporin proteins (the FG-Nups), which control selective transport through the pore. The theory explicitly models the electrostatic, hydrophobic, steric, conformational, and acid-base properties of the FG-Nups. The electrostatic
more » ... al within the pore, which arises from the specific charge distribution of the FG-Nups, is predicted to be negative close to pore walls and positive along the pore axis. The positive electrostatic potential facilitates the translocation of negatively charged particles, and the free energy barrier for translocation decreases for increasing particle hydrophobicity. These results agree with the experimental observation that transport receptors that form complexes with hydrophilic/ neutral or positively charged proteins to transport them through the NPC are both hydrophobic and strongly negatively charged. The molecular theory shows that the effects of electrostatic and hydrophobic interactions on the translocating potential are cooperative and nonequivalent due to the interaction-dependent reorganization of the FG-Nups in the presence of the translocating particle. The combination of electrostatic and hydrophobic interactions can give rise to complex translocation potentials displaying a combination of wells and barriers, in contrast to the simple barrier potential observed for a hydrophilic/neutral translocating particle. This work demonstrates the importance of explicitly considering the amino acid sequence and hydrophobic, electrostatic, and steric interactions in understanding the translocation through the NPC. non-additivity | nuclear transport | disordered protein | coarse grain model | mean force potential N ucleocytoplasmic transport occurs exclusively through protein pores that perforate the nuclear envelope, the nuclear pore complexes (NPCs) (1). Whereas the NPC is permeable to small molecules (e.g., water, ions) that can diffuse freely through it, bigger cargoes, such as proteins and mRNA, require the assistance of transport receptors (known as karyopherins or "kaps") to be effectively transported between the cytoplasm and the nucleus. It is challenging to understand how a cargo that is not able to pass through the pore by itself can successfully traverse the pore on forming a substantially larger kap-cargo complex. Because of its importance to the functioning of eukaryotic cells, this apparent paradox has been the focus of attention of numerous studies throughout the past decade (reviewed in refs. 1-9 ). There is no universally agreed picture of the detailed mechanism of selective transport through the NPC, although there is broad agreement that a family of proteins called nucleoporins (Nups) is essential for selective transport through the pore (10-14) . The folded domains of the Nups form the outer envelope of the NPC (in contact with the nuclear scaffold), and their intrinsically disordered domains protrude into the inner space of the pore. These disordered domains, known as FG-Nups due to their high content of phenylalanine-glycine residues (FG-repeats), interact with the translocating particles to set up the permeability barrier that controls selective translocation through the NPC. A definitive transport mechanism remains elusive because directly visualizing FG-Nups and the kap-cargo complex within individual NPCs is at the limits of current single-molecule tracking technology (15-17); therefore, theory (18, 19) and computer simulations (20-22) have been used in an attempt to elucidate the essential features of the translocation process. In a recent coarsegrained molecular dynamics (MD) simulation, the kap-FG interaction was found to be highly dynamic and the FG-Nups formed a layer on the pore walls (20). The kap-cargo complex particle interacts with the FG residues in this layer as it diffuses through the channel. Another simulation study suggested that the translocating particle remains bound to the same Nup for its entire trajectory through the NPC (21). The differences between these works arise due to the choice of the molecular model, which, in neither case, considered the specific sequence and length of the FG-Nups and the particular properties of each amino acid in the sequence (e.g., hydrophobicity, charge). Until recently, it was believed that hydrophobic interactions were solely responsible for the selectivity of the translocation process (11, 14, 20, 23, 24) . According to this view, water-soluble proteins generally present a hydrophilic surface and are repelled by the hydrophobic domains of the FG-Nups, but hydrophobic patches on the surface of kaps interact attractively with the FG-Nups. It was assumed that the main role of charged amino acids in FG-repeats (about 15%) is to stabilize the hydrophobic sequences against self-aggregation and collapse. Although this argument suggests that the sign and magnitude of the charges do not play important roles, a recent analysis has shown that kaps and kap-cargo complexes are hydrophobic and highly negatively charged, whereas the unfolded Nup domains have a small net positive charge (25), suggesting that electrostatics may be essential for the selective filtering mechanism. The effect of sequencedependent electrostatic interactions has not been considered in previous simulations and theories; therefore, its contribution to the overall transport process remains unclear. The goal of the present work is to address the structure of the FG-Nups within the NPC and the molecular factors that Author contributions: M.T., O.P., M.K., Y.R., and I.S. designed research, performed research, analyzed data, and wrote the paper. The authors declare no conflict of interest. This article is a PNAS Direct Submission. Freely available online through the PNAS open access option. 1 M.T. and O.P. contributed equally to this work.
doi:10.1073/pnas.1212909110 pmid:23404701 pmcid:PMC3587244 fatcat:d3kjnie6lnf3jn2vg55jhccpou