Role of Alternative Elicitor Transporters in the Onset of Plant Host Colonization by Streptomyces scabiei 87-22

Abstract

Plant colonization by Streptomyces scabiei, the main cause of common scab disease on root and tuber crops, is triggered by cello-oligosaccharides, cellotriose being the most efficient elicitor. The import of cello-oligosaccharides via the ATP-binding cassette (ABC) transporter CebEFG-MsiK induces the production of thaxtomin phytotoxins, the central virulence determinants of this species, as well as many other metabolites that compose the 'virulome' of S. scabiei. Homology searches revealed paralogues of the CebEFG proteins, encoded by the cebEFG2 cluster, while another ABC-type transporter, PitEFG, is encoded on the pathogenicity island (PAI). We investigated the gene expression of these candidate alternative elicitor importers in S. scabiei 87-22 upon cello-oligosaccharide supply by transcriptomic analysis, which revealed that cebEFG2 expression is highly activated by both cellobiose and cellotriose, while pitEFG expression was barely induced. Accordingly, deletion of pitE had no impact on virulence and thaxtomin production under the conditions tested, while the deletion of cebEFG2 reduced virulence and thaxtomin production, though not as strong as the mutants of the main cello-oligosaccharide transporter cebEFG1. Our results thus suggest that both ceb clusters participate, at different levels, in importing the virulence elicitors, while PitEFG plays no role in this process under the conditions tested. Interestingly, under more complex culture conditions, the addition of cellobiose restored thaxtomin production when both ceb clusters were disabled, suggesting the existence of an additional mechanism that is involved in sensing or importing the elicitor of the onset of the pathogenic lifestyle of S. scabiei.

Keywords: Streptomyces scabiei; cello-oligosaccharide; cellulose utilization; elicitor sensing; phytotoxin; plant pathogen; secondary metabolism; sugar transport; virulence; virulome.

Figure 1

Chromosome position, gene organization, and identity levels of the three sugar ABC-type transporters presumed to interact with cello-oligosaccharide elicitors in S. scabiei 87-22. Numbers between ORFs are the percentage of amino acid identity between the corresponding homologous proteins.

Figure 2

Phylogenetic analysis of CebE and PitE homologues. (a) Phylogram with CebE1 and CebE2 homologues. Black dots indicate the strains that possess both CebE1 and CebE2 proteins. Except for S. scabiei strains, only one representative of proteins with identical accession numbers was used to generate the tree; (b) Phylogram of PitE homologues. Bootstrap values of main nodes are indicated. The scale bar indicates the number of amino acid sequence substitutions per site. All accession numbers of proteins used to generate the phylograms are listed in Table A2 in Appendix A.

Figure 3

Transcriptional response of known and putative elicitor transporter clusters upon cellobiose and cellotriose supply. Values in the heatmap are the mean of Log2 fold-change. The expression changes of housekeeping genes gyrA, recA, atpA, and of bxlE and dasA encoding the xylobiose and N-N’-diacetylchitobiose ABC-type sugar-binding proteins, respectively, were used as negative controls (for which the expression is known to be independent of either cellobiose or cellotriose). The expression changes of five genes of the thaxtomin biosynthetic gene cluster (txt) were used as positive controls. abbreviations: neg ctrl, negative controls; pos ctrl, positive controls; hpi, hour(s) post induction by either cellobiose or cellotriose.

Figure 4

Growth of wild type (87-22) and transporter deletion mutant strains on minimal medium supplemented with cellobiose as the sole carbon source at 3 and 7 dpi. Note that the growth delay was only observed when genes of the ceb1 cluster are deleted (mutants ΔcebEFG1 and ΔcebE1-2).

Figure 5

(a) Radish seedling assays showing the reduced virulence of the transporter deletion mutant strains compared to the wild type (87-22) at 7 dpi; (b) Thaxtomin levels as extracted from the radish seedling assay at 7 dpi and analyzed by HPLC. Thaxtomin levels produced by the wild type were set as 100%.

Figure 6

Thaxtomin production levels of the mutant strains compared to the wild type (87-22) on a plant-based medium without (OBA) and with (OBAc) the addition of cellobiose. Thaxtomin was extracted from agar plates at 3 and 7 dpi and analyzed by HPLC. Thaxtomin levels produced by the wild type were set as 100%. For the mutant strains, the means (± standard deviation) of at least three independently generated mutant isolates plated in duplicate are shown. Mutants that produced statistically different thaxtomin levels compared to the wild type grown under the same conditions are indicated by * if p < 0.05 and ** if p < 0.025.

Figure 7

3D model of CebE2. (a) Cartoon representing the CebE1 (Domain 1 in blue and Domain 2 in green) structure in complex with cellotriose (grey sticks) with the two domains of CebE2 superimposed separately (Domain 1 in yellow and Domain 2 in orange). (b) CebE1 (same coloring scheme as in (a)) and CebE2 (yellow) with their Domain 1 superimposed, highlighting the 45° rotation observed between Domain 2 of both proteins. (c) Interactions (black dashed lines) between cellotriose (grey sticks) and Domain 1 of CebE1 (blue sticks), including through water molecules (red spheres). Equivalent residues in CebE2 are displayed as yellow sticks. (d) Same as (c) for Domain 2 with CebE1 in green and CebE2 in orange. The figure was prepared using PyMOL (The PyMOL Molecular Graphics System v2.5.2 Enhanced for Mac OS X, Schrödinger, LLC).

Figure 8

Model of cellobiose and cellotriose transport and thaxtomin production for the onset of pathogenicity in S. scabiei 87-22. Red lines, transcription repression; green lines, transcription activation.