Arbuscular mycorrhiza Symbiosis Induces a Major Transcriptional Reprogramming of the Potato SWEET Sugar Transporter Family

Abstract

Biotrophic microbes feeding on plants must obtain carbon from their hosts without killing the cells. The symbiotic Arbuscular mycorrhizal (AM) fungi colonizing plant roots do so by inducing major transcriptional changes in the host that ultimately also reprogram the whole carbon partitioning of the plant. AM fungi obtain carbohydrates from the root cortex apoplast, in particular from the periarbuscular space that surrounds arbuscules. However, the mechanisms by which cortical cells export sugars into the apoplast for fungal nutrition are unknown. Recently a novel type of sugar transporter, the SWEET, able to perform not only uptake but also efflux from cells was identified. Plant SWEETs have been shown to be involved in the feeding of pathogenic microbes and are, therefore, good candidates to play a similar role in symbiotic associations. Here we have carried out the first phylogenetic and expression analyses of the potato SWEET family and investigated its role during mycorrhiza symbiosis. The potato genome contains 35 SWEETs that cluster into the same four clades defined in Arabidopsis. Colonization of potato roots by the AM fungus Rhizophagus irregularis imposes major transcriptional rewiring of the SWEET family involving, only in roots, changes in 22 of the 35 members. None of the SWEETs showed mycorrhiza-exclusive induction and most of the 12 induced genes belong to the putative hexose transporters of clade I and II, while only two are putative sucrose transporters from clade III. In contrast, most of the repressed transcripts (10) corresponded to clade III SWEETs. Promoter-reporter assays for three of the induced genes, each from one cluster, showed re-localization of expression to arbuscule-containing cells, supporting a role for SWEETs in the supply of sugars at biotrophic interfaces. The complex transcriptional regulation of SWEETs in roots in response to AM fungal colonization supports a model in which symplastic sucrose in cortical cells could be cleaved in the cytoplasm by sucrose synthases or cytoplasmic invertases and effluxed as glucose, but also directly exported as sucrose and then converted into glucose and fructose by cell wall-bound invertases. Precise biochemical, physiological and molecular analyses are now required to profile the role of each potato SWEET in the arbuscular mycorrhizal symbiosis.

Keywords: Arbuscular mycorrhiza; SWEET transporters; plants; potato; root; sugar transport; symbiosis.

Figure 1

Phylogenetic tree of potato, tomato, rice, M. truncatula, and A. thaliana SWEETs. The unrooted tree was generated based on an amino acid alignment using Clustal Omega. The phylogenetic tree was constructed using the neighbor-joining method and the p-distance model using MEGA 6. Scale bar represents evolutionary distance in number of amino acid differences per site. Bootstrapping was performed with 1000 replicates and values are displayed on branches in %. Colored circles represent regulation by mycorrhization in the same color code as in Figure 5. Gene names and accession numbers can be found in Supplementary Table 1.

Figure 2

Synteny analyses of SWEET genes in potato and tomato. SWEET genes are present in almost all chromosomes with the exception of chromosome 10 in potato and 10 and 11 in tomato. The distribution of the genes in both genomes is almost identical, with only two inversions in chromosome 6, mirroring the general high degree of synteny between both genomes.

Figure 3

Expression analyses of potato SWEET genes. (A) In silico expression analysis of 26 of the 35 potato SWEET genes in several tissues, including flower, leaf, stem, root, stolon, and tuber from either RH89-039-16 (RH) or DM3-1 (DM), according to the RNA-seq data from the Spud DB expressed as FPKM values (fragments per kilobase of transcript per million mapped reads). Two independent root expression data sets are available for DM root (DM ROOT1, DM ROOT2). Since SWEET12e was not correctly annotated prior to this work, two transcripts can be found corresponding to this gene. The color code indicates level of expression in logarithmic scale from dark purple to bright yellow. (B) RT-PCR expression analyses in leaf, stem, root, and tuber for the nine potato SWEET genes not annotated in the SGN database. The house keeping gene elongation factor 1α (Stef1) was used as control. Forty cycles were used for the PCR.

Figure 4

Expression analysis of all potato SWEETs in roots in response to colonization by the Arbuscular mycorrhizal fungus R. irregularis. (A) The expression of the phosphate transporter PT4 and the cell wall-bound invertase InvCD141 from potato, as well as the fungal translation elongation factor 1a RiTEF and the monosaccharide transporter gene RiMST2 were analyzed at 4, 6, and 8 weeks post inoculation (wpi). (B) Expression of potato SWEETs was measured at 6 and 8wpi. Expression is shown as relative expression to potato elongation factor 1α (Stef1) or to RiTEF for RiMST2. Error bars represent standard error of the mean. Non colonized samples are indicated by minus (−) and colonized samples by plus (+), per treatment the average expression of three biological replicates is shown. Student's t-test was used to calculate significance of mycorrhized compared to non mycorrhized samples in the same time point (^**p < 0.01, ^*p < 0.05).

Figure 5

Promoter-reporter assays of three SWEET promoters during AM symbiosis. 2kb fragments upstream of the ATG of potato SWEETs 2c, 7a, and 12a were cloned in front of the GUS reporter gene and transformed in M. truncatula. Composite plants were inoculated with R. irregularis and harvested 5 wpi (weeks post inoculation). β-Glucuronidase staining and WGA-FITC (wheat germ agglutinin-fluorescein isothiocyanate) counterstaining of fungal cell walls was carried out in mycorrhizal and non-mycorrhizal control roots. (A) Empty vector control roots. (B) Non-mycorrhizal and mycorrhizal promoter-reporter roots. Scale bars represent 100 μm. DIC: differential interference contrast, WGA-FITC signal is shown in green.

Figure 6

Model of sugar partitioning during arbuscular mycorrhiza symbiosis. Sucrose (Suc) is delivered to arbuscule containing cells symplastically from the endodermis to overcome the casparian strip (in red). In the cytoplasm, Suc can be cleaved by sucrose synthase (SUSY) or cytoplasmic invertase (cINV) to glucose (Glc) and fructose (Fru). To maintain the favorable concentration gradient, hexoses could be translocated into the vacuole via tonoplast located SWEETs or other transporters. Alternatively, hexoses could be exported into the apoplast with the help of SWEET7a. Direct export of sucrose into the apoplast or the periarbuscular space could be achieved by sucrose efflux transporters from clade III such as SWEET12a. In the apoplast and periarbuscular space sucrose is cleaved by cell wall-bound invertase (CWI). The sugars in the apoplast are either taken up by the fungus via monosaccharide transporters such as RiMST2 or by the plant cell via monosaccharide transporters such as MST1 (shown in M. truncatula) and via sucrose transporters such as SUT2 (shown for S. lycopersicum). Neighbor cells might also contribute to the nutrition of the arbuscule containing cell by providing sugars symplastically. Black arrows on SWEETs: sugar transport direction as described above. Red arrows on SWEETs: alternative sugar transport direction.