Functional Cross-Talk of MbtH-Like Proteins During Thaxtomin Biosynthesis in the Potato Common Scab Pathogen Streptomyces scabiei

Abstract

Thaxtomin A is a potent phytotoxin that serves as the principle pathogenicity determinant of the common scab pathogen, Streptomyces scabiei, and is also a promising natural herbicide for agricultural applications. The biosynthesis of thaxtomin A involves the non-ribosomal peptide synthetases (NRPSs) TxtA and TxtB, and an MbtH-like protein (MLP), TxtH, which may function as a chaperone by promoting the proper folding of the two NRPS enzymes in S. scabiei. MLPs are required for the proper function of many NRPS enzymes in bacteria, and they are often capable of interacting with NRPSs from different biosynthetic pathways, though the mechanism by which this occurs is still poorly understood. To gain additional insights into MLP functional cross-talk, we conducted a broad survey of MLPs from diverse phylogenetic lineages to determine if they could functionally replace TxtH. The MLPs were assessed using a protein solubility assay to determine whether they could promote the soluble expression of the TxtA and TxtB adenylation domains. In addition, the MLPs were tested for their ability to restore thaxtomin production in a S. scabiei mutant that lacked TxtH and other endogenous MLPs. Our results showed that the MLPs investigated vary in their ability to exhibit functional cross-talk with TxtH, with two of the MLPs being unable to compensate for the loss of TxtH in the assays performed. The ability of an MLP to serve as a functional partner for the thaxtomin NRPS was not correlated with its overall amino acid similarity with TxtH, but instead with the presence of highly conserved residues. In silico structural analysis of TxtH in association with the TxtA and TxtB adenylation domains revealed that several such residues are situated at the predicted interaction interface, suggesting that they might be critical for promoting functional interactions between MLPs and the thaxtomin NRPS enzymes. Overall, our study provides additional insights into the mechanism of MLP cross-talk, and it enhances our understanding of the thaxtomin biosynthetic machinery. It is anticipated that our findings will have useful applications for both the control of common scab disease and the commercial production of thaxtomin A for agricultural use.

Keywords: Streptomyces; non-ribosomal peptides; phytotoxin; plant pathogen; specialized metabolism; thaxtomin.

FIGURE 1

Phylogenetic analysis of MLPs. The phylogeny was generated from the amino acid sequences of 133 MLPs from the database. The MLPs originate from the phyla Actinobacteria (yellow), Acidobacteria (fuchsia), Proteobacteria (purple), Firmicutes (green), and Cyanobacteria (orange). TxtH from S. scabiei is highlighted in red, and the two other MLPs encoded in the S. scabiei genome (MLP_lipo and MLP_scab) are indicated in bold. MLPs that were chosen for functional analysis in this study are labeled in blue. Diverse lineages are shown in different colors, and the scale bar indicates the number of amino acid substitutions per site. Information about each MLP is provided in Supplementary Table 2. Ssc, Streptomyces scabiei; Sac, Streptomyces acidiscabies; Stu, Streptomyces turgidiscabies; and Seu, Streptomyces europaeiscabiei.

FIGURE 2

(A) Western blot analysis of soluble HIS₆-TxtA^A and HIS₆-TxtB^A proteins expressed in the presence and absence (–) of different HIS₆-tagged MLPs in E. coli BL21(DE3)ybdZ:aac(3)IV. The analysis was conducted three times, and one representative set of blots is shown. (B) Quantification of the HIS₆-TxtA^A (left) and HIS₆-TxtB^A (right) protein band intensities following co-expression with different HIS₆-tagged MLPs. The bars represent the mean percent band intensity from triplicate western blots relative to the control (co-expression with HIS₆-TxtH; set to 100%) and was determined using ImageJ. Error bars represent the standard deviation from the mean. Means with different letters (a–d) were determined to be significantly different (P ≤ 0.05).

FIGURE 3

HPLC analysis of culture extracts from wild-type S. scabiei 87.22 (A), the S. scabiei triple MLP deletion mutant (Δmlp_lipo/ΔtxtH/Δmlp_scab) containing the txtH expression vector (B) and the triple MLP deletion mutant containing the MXAN_3118 expression vector (C). The peak corresponding to thaxtomin A (retention time = 1.65 min) in each chromatogram is indicated with the red asterisks, and the peaks corresponding to thaxtomin B (retention time = 3.81 min) and thaxtomin D (retention time = 4.61 min) are indicated with ▼and Δ, respectively. The chemical structures of thaxtomin A, B, and D are also shown next to the corresponding peaks, and the hydroxyl groups of thaxtomin A and B are highlighted.

FIGURE 4

Relative quantification of thaxtomin production in the S. scabiei MLP triple mutant (Δmlp_lipo/ΔtxtH/Δmlp_scab; ΔΔΔ) expressing different non-cognate MLPs. The production levels are represented as the average% production of thaxtomins (thaxtomin A, thaxtomin B, and thaxtomin D) relative to wild-type S. scabiei 87.22 (±SD). n = 3 biological replicates for 87.22, ΔΔΔ, ΔΔΔ/VC (vector control); n = 5 biological replicates for SCLAV_p1293; and n = 6 biological replicates for all other strains. Means with different letters (a–g) were determined to be significantly different (P ≤ 0.05).

FIGURE 5

Summary of the results of the different assays examining the interaction between the thaxtomin NRPSs and the non-cognate MLPs. The heat map illustrates the relative amount of soluble HIS₆-TxtA^A and HIS₆-TxtB^A proteins produced in the presence of the different MLPs as compared to TxtH (set to 100%), as well as the relative thaxtomin production levels in the presence of different MLPs as compared to TxtH (set to 100%).

FIGURE 6

(A) Predicted interaction interface between the S. scabiei Txt A-domains and TxtH. TxtA^A is shown in green, TxtB^A is shown in yellow, and TxtH is shown in orange. The strictly conserved serine and leucine residues (red) of TxtH (S23 and L24) and two possible interacting alanine residues (blue) of TxtA^A (A378 and A383) are highlighted. The corresponding alanine residues of TxtB^A (A405 and A410) are not labeled. The residues that are associated with interaction interface are shown as sticks. (B) Partial amino acid sequence alignment of the S. scabiei TxtA^A and TxtB^A. The residues involved in the formation of α-helix (box) and β-sheets (arrows) within the predicted structures are indicated above the alignment in green (for TxtA^A) and yellow (for TxtB^A). Residues in TxtA^A and TxtB^A that fall within the interaction interface between the TioS/FscH NRPSs and their cognate MLPs (TioT/FscK) are indicated by the black lines above the amino acid alignment. The two alanine residues of TxtA^A and TxtB^A that are predicted to interact with S23 and L24 of TxtH are indicated by the astericks. (C) Amino acid sequence alignment of TxtH from S. scabiei and the non-cognate MLPs from other bacteria that were analyzed in this study. Residues within the non-cognate MLPs that match the amino acid residue in TxtH at the same position are colored. The consensus sequence (VxxNxExQxSLWP-x5-PxGW-x12-L-x6-WTDxRPxSL) appearing in more than 85% of the 133 MLPs used in the phylogenetic analysis are indicated above the alignment. The residues shown to be important for the soluble production of TxtA^A and/or TxtB^A by TxtH (Li et al., 2019a) are indicated by the red circles. Variant residues in the non-cognate MLPs that may have a negative impact on the interaction with the thaxtomin NRPSs are highlighted in black boxes. Extracted secondary structures for TxtH are shown using orange boxes (helixes) and arrows (β-sheets) above the alignment. Residues in TxtH that fall within the interaction interface between TioT/FscK and their corresponding NRPSs (TioS/FscH) are indicated by the black lines above the amino acid alignment.