Mapping the amino acid properties of constituent nucleoporins onto the yeast nuclear pore complex

Abstract

Visualization of molecular structures aids in the understanding of structural and functional roles of biological macromolecules. Macromolecular transport between the cell nucleus and cytoplasm is facilitated by the nuclear pore complex (NPC). The ring structure of the NPC is large and contains several distinct proteins (nucleoporins) which function as a selective gate for the passage of certain molecules into and out of the nucleus. In this note we demonstrate the utility of a python code that allows direct mapping of the physiochemical properties of the constituent nucleoporins on the scaffold of the yeast NPC׳s cytoplasmic view. We expect this tool to be useful for researchers to visualize the NPC based on their physiochemical properties and how it alters when specific mutations are introduced in one or more of the nucleoporins. The code developed using Python is available freely from the authors.

Background

A unique feature of the eukaryotic cell compared to the prokaryotic cell is the compartmentalization of nucleic acid synthesis and processing within the nucleus, and their separation from protein synthesis in the cytoplasm. This separation is achieved by the double-membrane of the nuclear envelope and the selective transport channels of nuclear pore complexes (NPCs) embedded within this barrier [1–3]. NPCs are the sole gateways that facilitate macromolecular exchange across the nuclear envelope between the nucleoplasm and cytoplasm. NPCs are composed of individual proteins called nucleoporins (nups) and are characterized by a central framework with 8-fold rotational symmetry with additional filamentous structures extending from the nuclear and cytoplasmic facades [4– 7]. Intrinsically disordered /unstructured proteins (IDPs) exist without a stable threedimensional (3D) structure as highly flexible conformational ensembles [8, 9]. These proteins seem to be more abundant in both prokaryotic and eukaryotic proteomes than originally assumed [10– 12]. IDPs are multifunctional in nature and involved in various biological processes such as signaling and regulation [9, 13, 13]. Nucleoporins, the protein components of the central conduit play a predominant role in nuclear transport. The permeability barrier in the yeast NPC is formed by a number of conserved nucleoporins that are known to be intrinsically disordered. Being one of the largest and most complex macromolecular assemblies known in cells, the NPC presents a formidable challenge for structural determination. NPC structure has been studied for over five decades by various techniques such as transmission electron microscopy [15], cryoelectron tomography of whole NPCs and X-ray crystallography of individual nucleoporins and their complexes [16– 18]. Cytoplasmic cross-sectional view of the NPC is generally obtained by cryoelectron tomographic reconstruction [17]. Other structural views are also generated by a combination of experimental and computational data that are invaluable in understanding the architecture, topology and organization of a NPC [19]. A molecule transported through the NPC from the cytoplasmic side into the nucleus does not recognize a geometric representation of the NPC, but more from physiochemical representation of the proteins that constitute the NPC. In order to facilitate an improved visualization of the NPC in terms of the primary characteristics of the constituent nucleoporins, we have developed a visualization tool. A straightforward code written in Python using PyGame and BioPython libraries provides researchers a new tool to visualize the NPC, which is color-coded with respect to a specific property of choice (e.g., hydrophobicity or charge). This tool can also be used to see how mutations in one or more of the proteins can alter the overall organization

Methodology

The geometric organization of the nucleoporins in the NPC from the cytoplasmic view is obtained from the structural model developed by Hoelz et al [17] and the forest model by Yamada et al [19]. Amino acid sequences of the constituent nucleoporins are obtained from ( http://www.uniprot.org/). All the sequences are used in the standard FASTA format and are used without any additional changes. The size/shape of each protein is scaled proportionally to its expected hydrodynamic volume to fit the symmetry of the NPC. The primary sequence information of the amino acids was converted into physiochemical/biochemcial value or the linear combinations thereof to define a score for each of the 20 naturally occurring amino acids. To date, there are more than 500 different properties of amino acids, all of which have been investigated through a large number of experiments and theoretical studies ( http://www.genome.jp/aaindex/) [20]. This collection of AA indices has redundant information, such as the same physical/chemical variation might be represented by more than one AA index. To address this issue using multivariate statistics, Atchely et al [21] defined a linear combination of the complete set, to a set of 5 unique parameters. Of these five factors, we have used the ones that have shown significant differences between the nucleoporins: Factor1 (PAH): Polarity, Accessibility, Hydrophobicity; Factor2 (PSS): Propensity for Secondary Structure; and Factor5 (EC): Electrostatic Charge. In addition, we have also included the amphiphilicity index from one of the AA indices (MITS020101) [20]. Table 1 (see supplementary material) lists these indices. The code was written in Python (version 2.71) with PyGame (version 1.9.1) and BioPython (version 1.61) libraries on a Microsoft Windows 7 desktop computer. A copy of the code and the instructions to run on a windows system are available from the authors.

Results & Discussion

Figure 1 shows the mapping of the amino acid indices PAH (Polarity, Accessibility, Hydrophobicity), PSS (Propensity for Secondary Structure), EC (Electrostatic Charge) and AI (amphiphilicity index) in panels (a), (b), (c) and (d), respectively. PAH, PSS, and EC are cumulative indices derived from multiple physiochemical properties (Table 1). The color scales in Figure 1 were determined linearly from red (high) to blue (low) and represented using the RGB (Red:Green:Blue) color model (with Green set to zero). The mapping of PAH on the NPC (Figure 1a) emphasizes the maximum differentiation in PAH by scaling the colors between the proteins accordingly. The outermost nucleoporin (Nup159) is differentiated from the folded (red) and the charged/unstructured (blue) regions. Similarly, the innermost part of the NPC is also highly differentiated with respect to the PAH scale (classified as a tree in the bimodal distribution model [19]). According to the current mapping of propensity to form secondary structures, (PSS, Figure 1b) most of the proteins show low propensities except the C-terminal part of Nup116 at the center of the NPC. Panels (c) and (d) demonstrate the variation in the charge factor and the amphiphilicity index across the NPC. Though panels (c) and (d) do not show much differentiation between the nucleoporins as in panels (a) and (b), several features are noted including that of Nup159 at the outermost ring of the NPC. The actual structure of the NPC is an evolutionarily conserved set of ~30 different proteins and is organized into several subcomplexes, each of which occur in multiple copies, resulting in ~500−1,000 protein molecules in the fully assembled NPC. Our results only use a representative set of these proteins from the cytoplasmic view and a first demonstration suggests that direct mapping of the physiochemical properties of amino acids could prove to be a useful tool in understanding the complex structures. The combination of PyGame and other high-level graphics programming can be utilized to build a complex visualization tool for the NPC using the principles demonstrated in this note.

Figure 1.

Direct mapping of the selected set of AA indices on the cytoplasmic view of yeast NPC. PAH: Polarity, Accessibility, Hydrophobicity; PSS: Propensity for Secondary Structure EC: Electrostatic Charge and AI: Amphiphilicity index. The index values are listed in Table 1 (see supplementary material). The average values of each nucleoporin was calculated and transformed into a linear scale between red (high) and blue (low).

The NPC is flexible and highly dynamic as one would expect during cargo translocation. Any dynamic information determined experimentally can easily be mapped onto the NPC by defining an amino acid/protein specific scale. The intrinsically disordered nature of the nucleoporins is an additional advantage of this model: it is not necessary to identify an orientation of a specific surface of a nucleoporin with respect to the NPC due to its high flexibility. The permeability barrier of the NPC is formed by unstructured phenylalanine-glycine (FG)-repeat regions that also serve as docking sites for transport factor-cargo complexes. Therefore, mapping the average properties of the nucleoporins with respect to any amino acid scale may be considered justifiable. Several emerging techniques such as single molecule spectroscopic methods are now capable of studying a single NPC [22, 23]; yet determining the complete 3D-structure with atomic resolution is difficult. While the precise nature of the permeability barrier remains to be an area of heavy debate, we hope that new tools such as the one presented here may prove useful in understanding the functional architecture of NPC.