1. Introduction
Streptomyces are well-known soil saprophytes that actively participate in the recycling of nutrients in the environment, produce a great variety of secondary metabolites, and induce beneficial activities in plants, such as growth promotion and disease resistance [
1]. However, in the last decades, a particular group of
Streptomyces species was found to harbor the ability to infect important commercial root and tuber crops, including potatoes, beets, radishes, sweet potatoes, and peanuts [
2]. It has been confirmed that pathogenic
Streptomyces, including
Streptomyces scabiei,
S. acidiscabies, and
S. turgidiscabies, can cause raised, pitted, or superficial scab lesions on the potato surface [
3,
4]. Such potato common scab significantly decreases crop quality and leads to severe economic losses [
5].
Thaxtomin is a family of phytotoxins produced by pathogenic
Streptomyces and is known as the major causative agent of potato common scab [
6,
7]. These compounds can induce plant cell hypertrophy, root growth retardation, and tissue necrosis at a nanomolar concentration by inhibiting the cellulose synthase complex [
8,
9,
10]. Notably, almost all pathogenic
Streptomyces have a pathogenicity island (PAI) composed of two regions. The toxin region (TR) harbors the thaxtomin biosynthetic gene cluster, and the colonization region (CR) contains two genes of
nec1 and
tomA, which encode secreted proteins known or suspected to act as virulence factors [
11,
12,
13]. A previous study has experimentally confirmed the transferability of the PAI among
Streptomyces species, which might account for the emergence of new pathogenic
Streptomyces and the spread of potato common scab [
14].
Cello-oligosaccharide is the main product of cellulose degradation and is abundant in nature. Some saprophytic
Streptomyces species specifically utilize cellulose as their carbon source, and employ extracellular cellulolytic enzymes and an inducible ATP-dependent uptake system specific to cello-oligosaccharide [
15]. Interestingly, cellobiose and cellotriose were found to possess a function other than providing energy in
S. scabiei strains. As a major molecule released in the expansion period of plant tissues, cellotriose is proposed as the best inducer and trigger of the synthesis of virulome, including thaxtomin [
16,
17]. It has been authenticated that the cello-oligosaccharide transport system CebEFG-MsiK is essential for the induction of thaxtomin in pathogenic
Streptomyces strains [
4,
18,
19]. As the core component of this transporter system, the solute-binding protein CebE is responsible for recognizing and binding the cello-oligosaccharide substrates [
20,
21]. In contrast to the non-pathogenic
Streptomyces strains, the transcriptional repressor CebR not only regulates cello-oligosaccharides uptake by binding with cellobiose or cellotriose, but is also involved in unlocking thaxtomin synthesis when binding with the upstream sites within the thaxtomin biosynthetic cluster. CebR has two binding sites (
cbs) within the thaxtomin gene cluster; one located upstream of
txtR, the cluster-situated regulator, and another within
txtB, the peptide synthetase gene (
tha
xtomin synthetase). As the ligands of CebR, the binding of cellobiose and cellotriose can inhibit its DNA-binding activity, thereby allowing the transcription of
txtR and inducing thaxtomin biosynthesis [
19,
22,
23,
24].
The special regulation mode of thaxtomin biosynthesis confirms the significant contributions of cello-oligosaccharides in
Streptomyces pathogenicity; CebE was named the “doorway” in pathogenic
Streptomyces due to its extracellular binding of cello-oligosaccharides and transmission of the signal to the cells, and consequently triggering the synthesis of arsenal thaxtomins [
25]. The investigation of CebE can enhance our understanding of the pathogenic mechanism and elucidate how pathogenic
Streptomyces establishes a connection between sensing extracellular plant material and initiating its virulent behavior [
25], thereby establishing a solid foundation for future biological control. In accordance with some pathogenic
Streptomyces, the common scab representative species
S. scabiei 87–22 possesses three homologs of CebE, among which
SCAB57751 was initially identified as the major contributor to potato scab, exhibiting a nanomolar range
KD value towards cellotriose and cellobiose [
25]. Recently, the physiological functions of the other two CebE proteins,
SCAB2421 and
SCAB77271, were also investigated. The deletion of
SCAB2421 caused a moderately attenuated virulence phenotype, while the loss of the PAI-located CebE (
SCAB77271) had no significant influence on strain growth or virulence capacity [
26].
To date, many pathogenic
Streptomyces species have been isolated, and diverse sets of CebEFG-Msik ABC transporters have been annotated in different pathogenic
Streptomyces strains. Whether these CebEFG-MsiK ABC transporters play a role in virulence is of great interest and remains to be elucidated. However, the primary research focus on CebE primarily lies within strain
S. scabiei 87–22, with a scarcity of related studies concerning other strains. Therefore, it is interesting to explore whether other mechanisms exist in other pathogenic
Streptomyces strains. The pathogenic
Streptomyces sp. AMCC400023 was previously isolated from potato common scab in Hebei province, China [
27]. By possessing the well-known PAI component in its genome, all evidence from radish seedling assay, potted back experiment, and thaxtomin production confirmed
Streptomyces sp. AMCC400023 is a virulent pathogen causing potato scab. As revealed by genome-based phylogenetic analysis, average nucleotide identity (ANI value is 88%), and in silico DNA-DNA hybridization (isDDH), strain AMCC400023 has a relatively close relationship with
S. scabiei at the species level. However, comparative genomic analysis in the Virulence Factors of Pathogenic Bacteria Database identified 60 unique virulence-associated genes in the genome of
Streptomyces sp. AMCC400023 when compared to
S. scabiei 87–22. Also, unlike the three functional clusters for cello-oligosaccharides uptake in
S. scabiei 87–22, two CebEFG-MsiK clusters were identified in the genome of AMCC400023. In the present study, we found that these two CebEs perform diverse physiological roles in strain growth and pathogenesis by comparing their binding affinities with cello-oligosaccharide substrates and pathogenesis of CebE-deficient strains, as well as transcriptomic responses in the presence of cellobiose.
2. Materials and Methods
2.1. Bacterial Strains, Plasmids, and Growth Conditions
Bacterial strains and plasmids used in this study are described in
Table 1.
Streptomyces strains were grown in Gause’s No. 1 medium [
28], oat bran broth (OBB) medium [
29], thaxtomin-defined medium (TDM) that was supplemented with 7 g L
−1 cellobiose (TDMc) medium [
16], or mannitol soya flour (MS) medium [
30] at 28 °C.
E. coli strains were grown in Luria-Bertani (LB) medium at 37 °C. When required, the medium was supplemented with ampicillin (100 μg mL
−1), kanamycin (50 μg mL
−1), chloramphenicol (30 μg mL
−1), nalidixic acid (25 μg mL
−1), or thiostrepton (20 μg mL
−1). Cellobiose and cellotriose were biotech grade and purchased from MACKLIN.
2.2. Bioinformatic Analysis
MAFFT was used for multiple sequence alignment and conserved sites were selected to construct the NJ phylogenetic tree using the JTT substitution model, with a bootstrap value of 1000. The CebR binding sites were predicted using the PRODORIC tool.
2.3. Protein Expression and Purification
Genomic DNA was isolated by the Tiangen DNA extraction kit (Tiangen Biotech Co., Ltd., Beijing, China). The hydrophobic region of the sequence was analyzed using SignalIP 6.0 and DeepTMHMM v0.0.10. The open reading frame encoding GEO1108/GEO5601/GEO7671 without the hydrophobic region was amplified by PCR using the genomic DNA of
Streptomyces sp. AMCC400023 as the template and primers listed in
Table S3. The amplified fragment was ligated with pEASY-Blunt and transformed into
E. coli DH5α for DNA sequencing. Then, the gene fragment was ligated with pET28a and transformed into
E. coli BL21(C43) competent cells. For protein expression, the
E. coli transformants were cultured in the LB medium at 37 °C containing 50 μg mL
−1 of kanamycin until the growth reached an absorbance at 600 nm (OD
600nm) of 0.6. Then, the His-tagged CebE was induced overnight (12 h) at 16 °C by a final concentration of 0.5 mM isopropyl-β-D-thiogalactopyranoside (IPTG). Cells were collected by centrifugation, resuspended in the lysis buffer (25 mM Tris-HCl, 150 mM NaCl, pH 8.0), and lysed using a high-pressure cell crusher. After centrifugation, the soluble proteins were loaded onto the pre-equilibrated Ni-NTA column and eluted with 250 mM imidazole. The cell lysis and purified proteins were verified by SDS-PAGE.
2.4. MST Assays
Microscale thermophoresis (MST) assays measurements were performed in triplicate at 25 °C, with 90% excitation power and medium MST power, on the Monolith NT.115 system. The protein was labeled with a red fluorescent dye and applied at a final concentration of 800 nM. The cellobiose and cellotriose solutions were two-fold diluted 16 times, with an initial concentration of 100 μΜ for CebE5601 and 270 mM for GebE7671. The labeled protein was then added to a dilution solution of the cello-oligosaccharides. Samples were filled into standard-treated capillaries for measurement. All the data were analyzed with MO.Affinity Analysis v2.3 software provided by the manufacturer.
2.5. Construction of CebE-Deficient Mutants
The in-frame deletion
Streptomyces mutants were constructed by homologous recombination technology. DNA fragments about 1 kb upstream and downstream of
GEO5601 or
GEO7671 were amplified, using primers listed in
Table S3. The PCR products were purified and ligated into the shuttle plasmid pJTU1278 to generate the pJTU1278-5601 and pJTU1278-7671 plasmids. Then, the constructed plasmids were introduced into
E. coli ET12567 by transformation. After sequence verification,
E. coli ET12567 strains carrying pJTU1278 plasmids (
E. coli ET12567/pJTU1278-5601 or
E. coli ET12567/pJTU1278-7671) were cultured in LB medium containing antibiotics (100 μg mL
−1 Ampicillin, 50 μg mL
−1 Kanamycin, 30 μg mL
−1 Chloramphenicol) until OD
600nm reached 0.4–0.6. The cells were collected by centrifuging at 8000×
g and washed with fresh LB medium, then collected and suspended in 500 μL of sterile water. For the preparation of
Streptomyces competent cells, the spores of
Streptomyces sp. AMCC400023 were collected from the OBB plates and placed in 5 mL TES buffer (10 mM Tris HCl, 1 mM EDTA, 0.1 mM SDS). After heat shock and ice bath, 5 mL YT medium was added and cultured at 37 °C for 2.5 h. Then, the spores were collected by centrifugation and resuspended with sterile water, mixed with the cultures of
E. coli ET12567/pJTU1278-5601 or
E. coli ET12567/pJTU1278-7671, and spread on the MS plates. After culturing at 30 °C for 2–3 d, 1 mL of antibiotic solution (0.25 mg mL
−1 thiostrepton and 0.5 mg mL
−1 nalidixic acid) was added and then incubated for 16 h. The colonies were transferred to MS medium supplemented with 0.25 mg mL
−1 tryptophan and 0.5 mg mL
−1 nalidixic acid, then cultured at 30 °C until confirmed by PCR amplification and sequencing, using primers listed in
Table S3.
2.6. Plant Virulence Assays
To assess the virulence phenotype of Streptomyces sp. AMCC400023 and mutants, an in vitro radish seedling assay was performed. Radish seeds were surface sterilized with 75% ethanol for 8–10 min and rinsed 3–4 times with sterile distilled water. The seeds were allowed to germinate for about 24–48 h in the dark, in a petri dish containing a moistened filter paper. Homogenously germinated seeds with a similar shoot length were placed into a glass tube of 1% agar water, with 3 seeds per tube. An appropriate amount of spore suspension with an OD value of 1.0 was inoculated in 100 mL of OBB for 5 d, then 200 µL of the culture was inoculated with newly sprouted radish seedlings, and the same amount of OBB medium was used as blank control. The tubes were incubated at 24 °C for 6 d, with intermittent light (16 h light and 8 h dark). Each experiment was performed in triplicate. To perform statistical analysis, we employed the Kruskal-Wallis test and Mann-Whitney test. Additionally, we applied the Benjamini-Hochberg procedure to correct the false discovery rate (FDR).
The potato pot trial was conducted from August to December 2020 at Shandong Agricultural University. Fifteen healthy potato tubers of variety Favorita were selected, surface sterilized with 1% NaClO for 5 min, and rinsed 3 times with sterile distilled water. After germination, potato seedlings with a similar shoot length were planted in a 35 × 30 (cm) plastic pot with 10 kg of soil.
Streptomyces sp. AMCC400023 and the mutant derivatives were cultured on Gause’s No. 1 medium at 28 °C for 10 d. Spores were collected with sterile distilled water and inoculated in OBB media at 28 °C for 7 d. The cultures were adjusted to the same OD value by fresh OBB, then inoculated during the potato expansion period and the final concentration of the
Streptomyces strains was 10
6 CFU/cm
3. Each treatment was repeated in fifteen pots. Samples treated with OBB medium and sterile distilled water worked as a blank control. After 90 d of cultivation, the disease index of potatoes was calculated.
2.7. Thaxtomin Quantification
With similar spore inoculation,
Streptomyces sp. AMCC400023 and its mutants were inoculated into OBB medium that was prepared in the same batch, and cultured at 28 °C and 180 rpm for 6–7 d. The cultures were filtered, the supernatant was extracted 3 times with an equal volume of ethyl acetate for 12 h, and then an appropriate amount of anhydrous sodium sulfate was added to remove the water. After rotary evaporation, the residue was dissolved in 2 mL of acetonitrile and filtered with a 0.22 μm membrane to get the thaxtomin solution. Liquid chromatography with tandem mass spectrometry (4000 QTRAP; AB SCIEX, Framingham, MA, USA) was used to quantify the thaxtomin content in solution with 0.3 mL min
−1 flow rate of an isocratic mobile phase of 40:60 acetonitrile/water, both of them containing 2 mM ammonium acetate and 0.1% acetic acid. Thaxtomin was detected according to the method described previously [
27]. The mycelia of the culture strains were dried and weighed. After dividing the total thaxtomin production by the dry weight of strains, the average thaxtomin production of culture strains was obtained. All experiments were repeated three times with different biological samples of
Streptomyces strains.
2.8. Transcriptomic and RT-qPCR Analyses
Streptomyces sp. AMCC400023 was cultured in TDMc medium that contained 7 g L−1 of cellobiose. Based on the cell growth and thaxtomin accumulation curves, we collected samples from 36 h, 72 h, and 120 h, with three biological replicates in each group. Total RNA was isolated by using TRIzol RNA extraction kit (Thermo Fisher Scientific, New York, NY, USA) according to the manufacturer’s protocol. Assessment of the RNA integrity was performed on the Agilent 4200 Tape Station (Agilent, Santa Clara, CA, USA); meanwhile, RNA concentration and purity were measured on the Thermo NanoDrop One (Thermo Fisher Scientific). Epicentre Ribo-Zero rRNA Removal Kit was used to remove the ribosomal RNA, and the cDNA libraries were constructed according to the protocol of Illumina’s NEBNextő Ultra II Directional RNA Library Prep Kit and sequenced using Illumina HiSeq 2500 (Illumina, San Diego, CA, USA).
The quality control process used Fastp v0.22.0 software, and clean reads were then mapped to the reference genome using Bowtie2 v2.4.4. FPKM values were calculated for each gene using cufflinks and transformed to TPM values. The DESeq R package was used to identify DEGs with a threshold of P < 0.05 and foldChange > 2 or foldChange < 0.5. Heatmap analyses were performed using the complexHeatmap v2.18.0 R package.
The expression levels of
GEO5601 and
GEO7671 were further verified by RT-qPCR, with primers listed in
Table S3. For PCR amplification, a total volume of 10 μL containing 5 μL of 2× PerfectStartTM Green qPCR SuperMix, 1 μL cDNA, 0.2 μL gene-specific primer (
Table S3), and 3.6 μL ddH
2O, was used. The program was 30 s at 94 °C, followed by 45 cycles with 5 s at 94 °C, and 30 s at 60 °C. To normalize the gene expression, 16S rDNA was used as an endogenous control.
4. Discussion
Pathogenic
Streptomyces strains generally have two or three sets of
cebR-cebEFG-bglC gene clusters for cello-oligosaccharides transport. As the signal receptor, the solute-binding protein CebE captures the external signal and functions as the doorway to trigger the regulation network of pathogen virulence [
25]. In the well-studied strain
S. scabiei 87–22, three putative CebEs were identified and their functions for the transporting of cellobiose/cellotriose were verified. Apart from some non-pathogenic
Streptomyces species, the second CebE (
SCAB2421) seems only present in the pathogenic
S. scabiei [
26]. In this pathogenic strain, two CebE homologs of SCAB57751 and SCAB2421 both serve as the signal receptors, with SCAB57751 being the most important one. Based on the transcriptional analysis and mutant virulence assay, the PAI-localized
SCAB77271 did not show significant physiological function.
Unlike strain S. scabiei 87–22, Streptomyces sp. AMCC400023 employs two sets of transporters for uptake of cello-oligosaccharides. Being most closely related to the functional important proteins SCAB57751 and CebEreti, we suspected that CebE5601 is more likely to fulfill the major biological functions of cellobiose/cellotriose recognition and binding in strain Streptomyces sp. AMCC400023. Experimentally, several lines of proof confirmed our speculation for the central status of CebE5601 in binding cello-oligosaccharides, not only for nutrient consumption but also as a signal in triggering thaxtomin synthesis and pathogenic virulence.
Firstly, CebE
5601 showed much higher binding affinities toward cellobiose/cellotriose in the MST assay, whereas CebE
7671 had more than 1000-fold higher
KD value. Maybe due to the assay bias of different methods or protein difference, when comparing the binding affinity of SCAB57751 that was measured by the tryptophan intrinsic fluorescence, a difference of 2–3 magnitudes existed between SCAB57751 and CebE
5601 [
25]. Also, in
S. scabiei 87–22, SCAB57751 showed an apparent binding preference towards cellotriose but not cellobiose (2.1 nM and 14.2 nM, respectively). It was supposed that the higher affinity to cellotriose could be a key adaptation for this bacterium to perceive cellotriose as the major signal for triggering thaxtomin synthesis instead of a nutrient uptake response [
25]. By contrast, and worth noticing, no apparent difference in binding values between cellobiose and cellotriose was observed for the CebE
5601 protein of
Streptomyces sp. AMCC400023. Species differences would be an applicable explanation for such affinity and need to be further addressed.
Secondly, as proven by transcriptomic and RT-qPCR analyses, the
cebE clusters of
GEO5601 are significantly more highly expressed throughout the cultivation process in TDMc liquid medium. Also, compared to the later exponential period (72 h),
GEO5601 was highly upregulated during the midst of the exponential phase when the strain grew vigorously (36 h). In combination with the constant and low expressions of the
GEO7671 gene cluster, we propose that the
GEO5601 transport system is functional in cellobiose uptake and substantially contributes to the nutrient consumption of the cells. After the later exponential period, the restored expression of
GEO5601 in the 120 h samples further confirmed its functions in thaxtomin synthesis. This result is partly consistent with the findings in strains
S. scabiei 87–22 and
S. scabiei EF-35. When grown in the minimal starch medium supplement with 5 g L
−1 of cellobiose, the expressions of
SCAB2421, especially
SCAB57751, were both upregulated [
31,
32].
Thirdly, in-frame deletions of two cebE genes resulted in different responses in plant infection. Through radish seedling assay, we found that the growth of the seedlings infected by two CebE mutants was better than that of the wild-type strain, and both showed alleviated pathogenic symptoms, which preliminarily indicated that the knocking-out of the cebE did have an impact on pathogenicity. Remarkably, the seedling growth of ∆GEO5601 was better than that of ∆GEO7671, implying that CebE5601 might be more important for its pathogenicity. As radish seedling assay can only partly reflect the pathogenicity of Streptomyces and the whole infection process is more complicated, we further performed the potato pot assay. Consistent with these results, the thaxtomin production of ∆GEO7671 did not change significantly compared with that of the wild-type strain, but the thaxtomin production of ∆GEO5601 showed a significant decrease, with only 11% of the wild-type yield retained.
In addition to the central status of CebE
5601 in cello-oligosaccharides transport and strain virulence, the specific location (PAI) of
GEO7671 makes it an attractive target to explore its physiological functions in pathogenic
Streptomyces strains. With weak binding affinities to cellobiose/cellotriose, the seriously impaired spore formation of Δ
GEO7671 implies that it might be involved in the spore development process. Meanwhile, the absence of CebE
7671 modestly impaired cell growth and consequently alleviated the pathogenicity of
Streptomyces sp. AMCC400023. In combination with the 4.1-fold upregulation of
GEO7671 in the late growth stage (120 h), we propose that this PAI-located CebE transporter may also be involved in strain virulence, maybe not for signal reception, but is functional in the sporulation and thus makes it conducive to occupying a specific ecological niche. In the identified
GEO7671 gene cluster (
Figure 2a), a regulator of the TexR/AcrR family is adjacent to the
GEO7671 gene. Referring to the finding that a TexR/AcrR regulator in
S. coelicolor mediates the development of aerial hyphae and spores, we suspect that this gene regulation would be an explanation for the impaired sporulation of Δ
GEO7671 and needs to be further elucidated [
33,
34]. Integrating the physiological roles of CebE
5601 and CebE
7671, we propose that the pathogenic
Streptomyces sp. AMCC400023 relies on two sets of CebEFG-BglC transporters for growth and pathogenicity, with distinct contributions to strain colonization, growth dynamics, and virulence (
Figure 7). Specifically, under cellobiose/cellotriose signals, CebE
5601 plays a pivotal role in sugar uptake, nutrient consumption, thaxtomin synthesis, and infection symptom manifestation. Meanwhile, CebE
7671 assumes a novel biophysical function of spore formation and potentially contributes to the colonization of expanding plant tissue.