Identification of nuclear pore proteins at plasmodesmata

Plasmodesmata (PD) exchange small molecules, RNAs and proteins between plant cells with an exclusion limit for passive, non-specific transport that varies with cell type, developmental and physiological states of the tissue. Moreover, PD facilitate the transport of specific cargo that may require chaperones or carriers to be transported. The precise mechanisms underlying PD transport are not entirely understood. Parallels between PD and the nuclear pore complex (NPC), which controls transport between nucleoplasm and cytoplasm, have been discussed previously. PD transport shares similar features with the phase separation dependent NPC transport. Here, bioinformatics, proteomics and imaging were used to identify proteins with similarities to phase separating nucleoporins (NUP). We identified 21 NUPs in PD fractions, and validated dual localization to NPC and PD for 6 NUPs. The transmembrane anchor NUP CPR5 localized to PD orifices as detected by high-resolution imaging. cpr5 mutants had reduced intercellular transport activity for cargo larger than 54 kDa, but not for smaller cargo around 376 Da. The data indicates a role for NUP CPR5 in macromolecular transport through PD. Together this is consistent with the recruitment of NUPs in the green lineage to form a PD pore gating complex and is in line with the idea of a phase separation as diffusion barrier at PD.

INTRODUCTION

All kingdoms of life developed unique machineries for the direct intercellular exchange of small molecules, genetic information, and polypeptides: bacteria use pili, fungi use septa, metazoans developed gap junctions, and plants evolved the complex multimembrane plasmodesmata (PD) ( 1–3 ). PD are thought to mediate cell-to-cell sugar transport from source cells to the phloem, exchange metabolites between mesophyll and bundle sheath tissues in species that use C4 photosynthesis, translocate florigen from leaves to meristems, and perform many other functions ( 4–6 ). Despite their important role, their composition, structure, transport mechanism and regulation remain poorly understood.

RATIONALE

Some aspects of the transport of cargo through nuclear pore complexes (NPC) and PD, such as the types of cargo and the facilitation of transport for specific cargo, appear strikingly similar. We therefore hypothesized that PD may contain proteins with similar features as the ones constituting phase separation in the molecular barrier of the NPC [i.e., the phenylalanine–glycine nucleoporins (FG-NUPs)]. Such proteins may have arisen independently or may have evolved from the FG-NUPs. We used bioinformatics, proteomics and confocal imaging to identify potential candidate proteins for the transport channel of PD.

RESULTS

Nuclear pore proteins, in particular FG-NUPs as well as scaffold and transmembrane NUPs, constitute the functional gatekeepers of the NPC. FG-NUPs, scaffold NUPs and transmembrane NUPs were present in PD-enriched fractions and NUP fusions localized with PD markers. Topological studies revealed that the green lineage-specific membrane anchor NUP CPR5 is embedded into the ER membrane approaching the PD. High-resolution imaging showed that CPR5 localizes close to PD orifices. Analyses of passive passage of the heterologous green fluorescent protein (GFP) and of facilitated transport of the transcription factor SHORT-ROOT (SHR) indicated that intercellular exchange of polypeptides is partially impaired in cpr5 mutants. The transport changes were specific for macromolecules, as no changes were detectable for the diffusion of a small molecule dye; indicating an involvement of NUPs in macromolecular cell-to-cell trafficking, similar to their role in nuclear-cytoplasmic transport.

CONCLUSION

Our results indicate the possibility that the green lineage recruited at least parts of the NPC to PD to generate a plasmodesmatal pore complex (PDPC). The presence of FG-NUPs at PD is in line with the idea of a phase separation as the permeability barrier in the PDPC. The overall PDPC structure must differ from the NPC due to the presence of a central desmotubule. It is well established that the NPC is highly flexible and can likely adopt diverse conformations due to the multivalent interaction network. Hence, deletion of the membrane anchor CPR5 could affect conformation of the PDPC and its associated permeability barrier and thereby restrict macromolecular trafficking. This work may open new ways to understand the structure, function, transport mechanism and regulation of PD and how to engineer the permeability barrier between cells to alter plant properties.

Graphical Abstract

Left: nucleus with nuclear pore complexes (yellow) containing phase-separated domains generated by FG-NUPs (phenylalanine glycine-nuclear pore proteins). Middle: magnified nuclear pore complex. Right: Plasmodesmos embedded into the plant cell wall between two cells, lined by continguous plasma membrane and containing a central ER strand, and containing FG-NUP-based phase separation domains (position tentative).