Swept Away: Protein Mobility in the Phloem.

Transport of macromolecules through the phloemhas received increasing interest since thediscovery thatFLOWERINGLOCUST(FT) moves from leaves to the shoot apical meristem, where it induces flowering (Corbesier et al., 2007), indicating thatmacromolecules such as proteins andRNAcan act asmobile regulators of gene expression in distant tissues. Indeed,Molnaretal. (2010)usedgrafting between Arabidopsis thaliana ecotypes and mutants to demonstrate that small interfering RNAs are mobile and can induce epigenetic modifications in distant cells. Thieme et al. (2015) used a similar grafting approach to identify more than 2000mobile mRNAs that likely move through the phloem. However, how macromolecules get into the phloem, particularly whether theirmobility occurs specifically or as a result of their abundanceandpresence inthephloemcompanion cells, remains a subject of active research. For example, a recent Breakthrough Report showed thatmodeling ofmRNA abundance and half-life could explain the presence of most of themRNAs identified in phloem sap (Calderwood et al., 2016). For somemRNAs that do not fit the abundance model, Zhang et al. (2016) showed that tRNA-like structures in somemRNAscan triggermovement through the phloem. Thus,most RNAs likely enter thephloemsapnonspecifically,although some do show sequence-specific transport. In a recent Breakthrough Report, Paultre et al. (2016) address the movement of proteins through thephloem. They usedagrafting system (see figure) where the protein or domainof interest isexpressed fromthe35S promoter as a fusion to GFP in the scion, which is then grafted onto a nontransgenic, wild-type rootstock. The authors then used confocal microscopy to examine the root tissues for fluorescence at ten days after grafting. A fusion of the chloroplast transit peptide of ferredoxin-NADP1 oxidoreductase to GFP was translocated across the graft junction and localized to plastids in the root stele. However, this fusion did not enter theendodermisorcortex,potentially indicating a size-based restriction for movement into these tissues. Fusions of GFP to other chloroplast transit peptides gave similar results, except for the transit peptide of RECA HOMOLOG1, which failed to translocate. The authors also observed movement and stele-specific unloading of GFP fused to signals targeting peroxisomes, the nucleus, and the actincytoskeleton.Their corresponding mRNAs are all translated by cytoplasmic ribosomes; by contrast, fusions targeted to the endoplasmic reticulum lumen, membrane, or Golgi apparatus and translated by endoplasmic reticulum ribosomes failed to translocate. Expression from the 35S and the companion cell-specific SUC2 promoters gave similar results, indicating that the fusions likely entered the phloemstream from the companion cells. To addresswhether the fusionswere translocated as proteins, or as mRNAs that were subsequently translated in the root cells, the authors used RT-PCR to look for the fusion mRNAs in the root tissues at 5 weeks after grafting. Although the fluorescent protein signalwasstrong in theroots, theywereunable to detect the mRNA, indicating that the fusions were translocated as proteins. The fusion proteins were expressed from strong promoters, suggesting that protein abundance might affect whether proteins get into the phloem sap, similar to mRNAs. To examine this possibility, the authors next used a bioinformatic analysis to tease apart the relationships among protein abundance (approximated as transcript abundance in phloem companion cells), protein size, and likely presence in the translocation stream. This analysis showed that, for proteins under 70 kD, as protein abundance increased the probability that the protein would move into thephloemstream increased.However, larger proteins were much less likely to move. Therefore, this report shows that many proteins in the shoot companion cells can get swept awayby the translocationstream and land in the root stele. This indicates that not every protein found in phloem exudate has a signaling function; indeed, FT is likely to be the exception rather than the rule. Similar tomRNAs, are someproteins transported specifically? Does this flow of proteins have some adaptive advantage for the plant? Much as we once labeled intergenic regions as junk DNA, do these proteins adrift in the translocation streamhave a utility that remains to be discovered? We eagerly await the answers in future Breakthrough Reports. Grafting system for examination of protein translocation. The scion expresses a protein (or domain) of interest fused to a fluorescent protein (FP) tag. This scion is grafted to a wild-type rootstock and the root tip is examined for translocation of the protein. (Reprinted from Paultre et al. [2016], Figure 1.)