Even though Sugars Will Eventually be Exported Transporters (SWEETs) have been found in every sequenced plant genome, a comprehensive understanding of their functionality is lacking. In this study, we focused on the SWEET family of barley (Hordeum vulgare). A radiotracer assay revealed that expressing HvSWEET11b in African clawed frog (Xenopus laevis) oocytes facilitated the bidirectional transfer of not only just sucrose and glucose, but also cytokinin. Barley plants harboring a loss-of-function mutation of HvSWEET11b could not set viable grains, while the distribution of sucrose and cytokinin was altered in developing grains of plants in which the gene was knocked down. Sucrose allocation within transgenic grains was disrupted, which is consistent with the changes to the cytokinin gradient across grains, as visualized by magnetic resonance imaging and Fourier transform infrared spectroscopy microimaging. Decreasing HvSWEET11b expression in developing grains reduced overall grain size, sink strength, the number of endopolyploid endosperm cells, and the contents of starch and protein. The control exerted by HvSWEET11b over sugars and cytokinins likely predetermines their synergy, resulting in adjustments to the grain's biochemistry and transcriptome.

Based on recent in vitro data, a relatively large number of the plant Nitrate transporter 1/Peptide transporter Family (NPF) proteins has been suggested to function as gibberellic acid (GA) transporters. Most GA transporting NPF proteins also appear to transport other structurally unrelated phytohormones or metabolites. Several of the GAs used in previous in vitro assays are membrane permeable weak organic acids whose movement across membranes are influenced by the pH-sensitive ion-trap mechanism. Moreover, a large proportion of in vitro GA transport activities have been demonstrated indirectly via long-term yeast-based GA-dependent growth assays that are limited to detecting transport of bioactive GAs. Thus, there is a need for an optimized transport assay for identifying and characterizing GA transport. Here, we develop an improved transport assay in Xenopus laevis oocytes wherein we directly measure movement of six different GAs across oocyte membranes over short time. We show that membrane permeability of GAs in oocytes can be predicted based on number of oxygen atoms and that several GAs do not diffuse over membranes regardless of changes in pH values. In addition, we show that small changes in internal cellular pH can result in strongly altered distribution of membrane permeable phytohormones. This prompts caution when interpreting heterologous transport activities. We use our transport assay to screen all Arabidopsis thaliana NPF proteins for transport activity towards six GAs (two membrane permeable and four non-permeable). The results presented here, significantly reduce the number of bona fide NPF GA transporters in Arabidopsis and narrow the activity to fewer subclades within the family. Furthermore, to gain first insight into the molecular determinants of substrate specificities towards organic molecules transported in the NPF, we charted all surface exposed amino acid residues in the substrate-binding cavity and correlated them to GA transport. This analysis identified distinct residues within the substrate-binding cavity that are shared between GA transporting NPF proteins; the potential roles of these residues in determining substrate specificity are discussed.

Bioengineering aimed at producing complex and valuable plant specialized metabolites in microbial hosts requires efficient uptake of precursor molecules and export of final products to alleviate toxicity and feedback inhibition. Plant genomes encode a vast repository of transporters of specialized metabolites that— due to lack of molecular knowledge—remains largely unexplored in bioengineering. Using phlorizin as a case study—an anti-diabetic and anti-cancerous flavonoid from apple—we demonstrate that brute-force functional screening of plant transporter libraries in Xenopus oocytes is a viable approach to identify transporters for bioengineering. By screening 600 Arabidopsis transporters, we identified and characterized pu rine p ermease 8 (AtPUP8) as a bidirectional phlorizin transporter. Functional expression in the plasma membrane of a phlorizin-producing yeast strain increased phlorizin titer by more than 80 %. This study provides a generic approach for identifying plant exporters of specialized metabolites and demonstrates the potential of transport-engineering for improving yield in bioengineering approaches.