, Volume 236, Issue 6, pp 1849–1861

Thaxtomin A-deficient endophytic Streptomyces sp. enhances plant disease resistance to pathogenic Streptomyces scabies

  • Lan Lin
  • Hui Ming Ge
  • Tong Yan
  • Yan Hua Qin
  • Ren Xiang Tan
Original Article

DOI: 10.1007/s00425-012-1741-8

Cite this article as:
Lin, L., Ge, H.M., Yan, T. et al. Planta (2012) 236: 1849. doi:10.1007/s00425-012-1741-8


Each plant species in nature harbors endophytes, a community of microbes living within host plants without causing any disease symptom. However, the exploitation of endophyte-based phytoprotectants is hampered by the paucity of mechanistic understandings of endophyte-plant interaction. We here reported two endophytic Streptomyces isolates IFB-A02 and IFB-A03 recovered from a stress-tolerant dicotyledonous plant Artemisia annua L. After the determination of their non-pathogenicity at the genomic level and from the toxin (thaxtomin A, TXT) level, the endophytism of both isolates was supported by their successful colonization in planta. Of the two endophytes, IFB-A03 was further studied for the mechanism of endophyte-conferred phytoprotection owing to its plant growth promotion in model eudicot Arabidopsisthaliana. Using the endophyte-Arabidopsis co-cultivation system into which pathogenic Streptomyces scabies was introduced, we demonstrated that IFB-A03 pre-inoculation could activate the salicylic acid (SA)-mediated plant defense responses upon pathogen challenge. Moreover, IFB-A03 was shown to partially rescue the defense deficiency in eds5 (enhanced disease susceptibility 5) Arabidopsis mutants, putatively acting at the upstream of SA accumulation in the defense signaling pathway associated with the systemic acquired resistance (SAR). These data suggest that endophytic Streptomyces sp. IFB-A03 could be a promising candidate for biocontrol agents against S. scabies—a causative pathogen of common scab diseases prevailing in agronomic systems.


Artemisia annua L. Endophyte Streptomyces sp. Arabidopsis thaliana Plant defense 



Thaxtomin A


Salicylic acid


Enhanced disease susceptibility 5


Systemic acquired resistance


Plants in nature encounter a multitude of antagonists and beneficial organisms in their lifetime. While plant responses to stimuli from external bio-environment (such as pathogenic microbes and herbivores) have been the subject of considerable investigation, at least a category of biotic agents existing in plant internal niches has been nearly neglected for decades; however, it is emerging as one of the hottest topics in the field of plant–microbe interaction, which is the so-called endophyte. As a community of harmless microorganisms colonizing inside plant tissues and present in much less amounts than pathogens, endophytes are ubiquitous in terrestrial plants (Tan and Zou 2001; Arnold et al. 2003; Thomas and Soly 2009). Some of them are believed to confer beneficial effects by promoting plant growth and enhancing plant adaptability to environmental stresses during long-time coevolution, which are exemplified by the grass-inhabiting fungal endophyte Neotyphodium sp. (Tan and Zou 2001), rice-residing endophytic bacterium Pantoea agglomerans (Feng et al. 2006), and some N2-fixative bacteria such as Gluconacetobacter diazotrophicus and Serratia marcescens endophytic to the sugarcane and rice, respectively (Pan and Vessey 2001; Gyaneshwar et al. 2001). Despite increasingly growing evidence for fungal and bacterial endophytes, no endophytic actinobacteria were reported until the first Streptomyces sp. was isolated from a grass Lolium perenne (Guerny and Mantle 1993). Subsequent discoveries of endophytic Streptomyces spp. indicate that they can be novel sources of bioactive metabolites including antibiotics, as underpinned by a series of elaborate work from Strobel’s group (Castillo et al. 2002, 2003; Ezra et al. 2004). In contrast to soil-habitating actinobacteria, some of which develop incompact relationships with plant roots (known as rhizospheric species), a portion of actinobacteria may colonize inside the plant tissues and thus establish intimate associations with the host, which are termed endophytic actinobacteria, as exemplified by Frankia spp. distinctly forming nitrogen-fixative nodules in roots (Siciliano et al. 1998).

On the other hand, phytopathogenic actinobacteria, though constituting only a minority of plant-associated actinobacteria, may result in damage to plant cultivation and crop yields (Lerat et al. 2009). The representative pathogenic species are known to be Streptomyces scabies, S. acidiscabies, and S. turgidiscabies, causing common scab diseases in potato and some vegetables (King et al. 1991; Loria et al. 1995, 1997). Their pathogenicity has been ascribed to the biosynthesis and secretion of thaxtomin A (TXT), a phytotoxin which may result in dramatic cell swelling, reduced seedling growth, and inhibition of cellulose synthesis in plants as exemplified in Arabidopsis thaliana (Scheible et al. 2003). The TXT biosynthesis genes including nos gene are reported to reside on a pathogenicity island (PAI), which is conserved and horizontally transmissible among different species of the Genus Streptomyces, thereby accounting for the emergence of new pathogenic Streptomyces spp. in agronomic cultivation systems (Healy et al. 1999; Bukhalid et al. 1998). In addition to the well-known virulence factor gene nec1 (Bukhalid et al. 1998; Loria et al. 1995), the nos gene, which encodes nitric oxide synthase (NOS) essential to the ultimate nitration during TXT biosynthesis, represents another key PAI determinant in pathogenic streptomycetes (Kers et al. 2004).

Early in 1929, McKinney proposed a resistance phenomenon defined later as “cross-protection” based on the observation that tobacco plants pre-inoculated with mild virus strains might protect those plants from attack by a virulent strain of tobacco mosaic virus (TMV) (MacKenzie and Tremaine 1990). Despite the promising potential, the application of “cross-protection” became a dilemma primarily owing to the suspicion that mild virus strains protective for one plant species may cause serious diseases on the other(s) (Fulton 1986; Jackson and Talor 1996). The plant systemic acquired resistance (SAR) was subsequently discovered as indicated by an enhanced resistance to secondary pathogenic challenge in distal (systemic) tissues of tobacco after pre-inoculation with TMV (Durrant and Dong 2004). The scope of SAR has been broadened because it can be activated by necrosis-causing pathogens and some non-pathogenic rhizobacteria, and the evoked resistance is long-term and effective in systemic tissue against wide-spectrum pathogens including viruses, bacteria, and fungi (Conrath et al. 2002; Durrant and Dong 2004). The establishment of plant SAR requires an elevation in endogenous salicylic acid (SA) level, and its onset is associated with the expression of many pathogenesis-related (PR) genes such as PR-1, and PR-5 (Durrant and Dong 2004).

In view of a paucity in the reports regarding endophytic actinobacteria other than Frankia spp., and the necessity to investigate the interaction between plant and Streptomyces spp. as pathogenic streptomycetes are threatening worldwide crop/vegetable productions (Lerat et al. 2009), this study initiates from the isolation and characterization of endophytic streptomycetes from a highly stress-tolerant herb Artemisia annua L.. The selection of A. annua for isolating actinobacterial endophytes is based on two reasons: (1) it has been ethnobotanically used as a widely accepted folk medicine for its antimalaria, anti-schistosomiasis, antitussis, expectoration, and antiasthma activities in Asia (Tan et al. 1998; Bhakuni et al. 2001); (2) some of the plant stress-tolerant and/or medicinal properties are believed to result from the endophytes habitating inside (Castillo et al. 2002; Kuffner et al. 2010; Strobel and Daisy 2003), which can also explain the well-known notion that the low stress tolerance of axenic plants is ascribed, at least in part, to the absence of endophytic microbes (Hallmann et al. 1997). Historically, endophytes have been considered as weakly virulent or “latent” pathogens to plants (Hallmann et al. 1997). However, recent studies have demonstrated that at least some of them play beneficial roles in plant development and health, such as plant growth promotion and alleviation of disease symptoms caused by plant pathogens (Hallmann et al. 1997; Hasegawa et al. 2006). Since most plant-associated actinobacteria are non-pathogenic (Lerat et al. 2009), the protection of plants from pathogenic Streptomyces sp. could be expected from either endophytic Streptomyces-produced antimicrobials and/or the pre-colonization of the endophyte (Tokala et al. 2002; Cao et al. 2005; Hasegawa et al. 2006).

In this study, two culturable endophytic streptomycetes were isolated from the aerial tissues of healthy Artemisia annua L., which possesses the stress tolerance to chilling, drought, and pest attacks. Our attention was paid to actinobacterial endophytes because (1) studies regarding actinomycetes endophytic to dicotyledonous plants (particularly A. annua) are spare, and the work afforded by Strobel’s group and Zhao et al. (2010) were focused on antibiotics production and microbial taxonomic status, respectively, rather than the mechanism of the endophyte interacting with host plant defense responses; (2) neither IFB-A02 nor IFB-A03 possesses pathogenicity when re-introduced into A. annua (original host) in the greenhouse (Fig. S2); and (3) they can enter and colonize into the alternative host Arabidopsis thaliana.

Given that S. scabies is a primary phytopathogen which is the potential PAI donor species among the genus Streptomyces in nature (Lerat et al. 2009), and also encouraged by Franco’s previous work relating to wheat endophytes (Conn et al. 2008), we subsequently investigated the presumable phytoprotective effects of endophytic Streptomyces using model plant Arabidopsis thaliana, a member of dicotyledonous plants which Artemisia annua L. belongs to. We further elucidated the molecular mechanism underlying endophyte-plant interaction in response to S. scabies attack with an additional intention to determine whether plant defensive hormones SA and/or jasmonic acid (JA) participate in the main signaling pathways associated with enhanced plant resistance. The results herein demonstrate that pre-inoculation of endophytic Streptomyces sp. IFB-A03 may activate the SA-mediated defense responses in Arabidopsis thaliana, putatively acting toward the upstream of SA accumulation in the SA signaling pathway leading eventually to SAR.

Materials and methods

Isolation of endophytic actinobacteria

The plants of Artemisia annua L. were collected from the fields of Nanjing suburb (Sept. 2007) and Tianmu Lake (Mar. 2008), Jiangsu Province, P. R. China. The plant aerial tissues (stems and leaves) were thoroughly cleaned, surface sterilized as described in Coombs and Franco (2003), and immediately ground in 5 ml of sterile physiologic saline (pH 7.0), followed by the serial dilution plating onto the modified Gause No. 1 synthetic media supplemented with kanamycin (25 μg ml−1) and nystatin (25 μg ml−1) (referred to as the isolation media). All plates were incubated at 27–28 °C for up to 4 weeks. Actinomycete colonies were examined using a stereoscope (LED RING Illuminator JSZ6D, China) and picked based on morphological features and colors of pigmentation.

To validate the efficacy of the surface sterilization protocol, the last-run rinsing water was plated and incubated onto the same isolation media as did each plant specimen, and surface-sterilized plant tissue fragments were rolled onto the surface of isolation plates and incubated equally.

Phylogenetic analyses

Genomic DNA was extracted from the 48-h-cultured mycelia of each endophytic isolate as described in Wang et al. (1996). The 16S rDNAs of two isolates were PCR amplified using actinobacterial universal primers (Table S1) and sequenced. Phylogenetic analyses based on 16S rDNA sequences were done by searching Genbank by means of BLAST and then confirmed by performing the multiple sequence alignment, and the tree was constructed by means of the neighbor-joining (NJ) method in Mega 4.0 (Tamura et al. 2007). The stability of tree topology was evaluated by means of BOOTSTRAP based on 1,000 pseudoreplications. The 16S rRNA gene sequences for the two endophytic actinobacterial isolates IFB-A02 and IFB-A03 have been deposited in the GenBank database (accession no. HQ317204-317205).

Microscopic examination of endophytes in planta

The roots and stems of 6- or 7-week-old endophyte-pre-inoculated Arabidopsis plantlets were subjected to sample preparation (See Supplementary Material). Ultra-thin sections of 60–70 nm were examined under an H-7560 transmission electron microscope (TEM, Hitachi, Japan) operating at 80 kV.

Pre-treatment of Arabidopsis aseptic seedlings with endophyte

The endophyte-Arabidopsis co-cultivation system was established at the seed phase or early seedling stage. To optimize its establishment, surface-sterilized seeds were inoculated comparatively by (1) treating with the spore suspension (2 × 108 spores ml−1) for 1.5–2 h followed by uniform sowing on MS agar, and (2) sowing on the sterile MS agar to germinate into 4–6 leafed seedlings that were transferred to the sterile jars with the pre-loaded MS agars containing 200 μl of spore suspension (2 × 108 spores ml−1). The seedlings were cultivated until 5-weeks old as described in the “Cultivation of the sterile Arabidopsis seedlings” of Supplementary Material. The two protocols gave similar efficacy (Fig. S4).

SA application of Arabidopsis

Six-week-old Arabidopsis plants were sprayed with 2 mM SA at two lower leaves 3 days before challenge of pathogen, Streptomycesscabies ACCC 41024 (a phytopathogen preserved at Agricultural Culture Collection of China, Beijing, referred to as S. scabies hereafter unless otherwise stated), and the systemic leaves were harvested at 48 h post challenge. Control-treated plants were applied with sterile distilled H2O.

Pathogen challenge of endophyte-preinoculated Arabidopsis

Following initial pre-inoculation with the endophyte, pathogenic infection was performed with S. scabies using vertical petri dish system as described in Lehr et al. (2008) with some modifications. One half of petri dish was cast by adding the carbon source-free ISP4 agar media (15–20 mL, pH 5.8), while leaving the other half empty. The 5-week-old seedlings pre-inoculated with IFB-A02 and IFB-A03 were transferred onto ISP4 plates. After 6 days, the roots of endophyte-inoculated and -uninoculated Arabidopsis seedlings were challenged with 200 μl of S. scabies spore suspension (1 × 108 spores ml−1), while an equal volume of sterile distilled water was supplied to the control. Leaves were harvested 5 days post challenge and immediately frozen in liquid nitrogen for followup RNA and/or protein extractions.

Semi-quantitative RT-PCR analyses of defensive gene

The extracted RNAs (as detailed in Supplementary Material) from the plant materials tested were converted into first-strand cDNA using PrimeScript™ Reverse Transcriptase kit (Takara Bio., Dalian, China) following the manufacturer’s protocol. One microliter of cDNA was amplified for PR-1, PR-5, and/or PDF1.2 gene using SYBR®Premix ExTaq™ (Takara Bio., Dalian, China) with specific primers outlined in Table S1. Arabidopsis housekeeping gene Actin was used as an internal control. PCR products were subjected to gel electrophoresis and images recorded. Representative results from two independent experiments are shown.

Protein expression analyses

The protein extracts (20 μg) were separated by SDS-PAGE and transferred to a nitrocellulose membrane (Millipore, USA). For immunoblotting, the membranes were incubated with a rabbit polyclonal anti-PR1 antibody followed by incubation with horseradish peroxidase (HRP)-conjugated anti-IgG antibody raised in goat (KeyGEN, Nanjing, China). Specific protein bands were detected using ECL western-blotting kit (Amersham Biosci., UK) and images recorded.

Assays for defensive traits of Arabidopsis

Peroxidase activity was examined as described in “Supplementary Material.” The in-planta release of reactive oxygen species (ROS), primarily superoxide (O2−.), was detected via nitroblue tetrazolium (NBT) method (Tanaka et al. 2006). The leaves of endophytic spore-inoculated and -uninoculated Arabidopsis were collected at 2 h post pathogen challenge and trimmed, followed by vacuum infiltration and incubation (1 h) in the sodium phosphate buffer (0.05 M, pH 7.5) containing 0.05 % NBT and finally immersed in ethanol. The NBT-stained specimens were imaged by a compound microscope (Zeiss, Germany).

Measurement of plant endogenous salicylic acid (SA)

Fresh systemic leaves were collected from 5-week-old IFB-A03-pretreated and -untreated eds5 Arabidopsis mutants after the pathogenic S. scabies attack, and immediately liquid nitrogen frozen with each specimen consisting of 300 mg (fresh weight) of leaf tissues. Sample preparation followed the described protocol with slight modifications (Engelberth et al. 2003). After dissolved in 1 ml of anhydrous methanol, the SA in the residue derived from each sample was determined as reported (Tamaoki et al. 2008) with its authentic material at 0.5 μg ml−1 as a standard by LC–MS on the instrument (see above) using aqueous methanol (10 % → 100 % within 12 min) as the mobile phase. The data were compared in Fig. 7.

Others including microbial culture, cultivation of the sterile Arabidopsis seedlings, assay of key PAI determinants (nos and nec1), bioassay of phytotoxin TXT, sample preparation for transmission electron microscopy (TEM), peroxidase activity assay, 1H-NMR analysis of microbial culture supernatant of endophyte, RNA extraction, and primers for PCR are detailed in “Supplementary Material.”

Results and discussion

Isolation and identification of Streptomyces spp. IFB-A02 and IFB-A03

Actinomycetes were recovered from surface-sterilized Artemisia annua L. aerial parts sampled from two distant sites. Isolates were at least 15 endophytic microbes, most of which were fungi and bacteria. Two actinomycetes, designated as IFB-A02 and IFB-A03, were obtained and subjected to further phylogenetic studies. These two isolates resembled morphologically Streptomyces species (Fig. 1a, b) including the spore-chain features, production of volatile geosmin, and texture of the colony surface. By 16S rDNA-based phylogenetic analysis, IFB-A02 and IFB-A03, were found to match, to the most extent, database entries from the genus Streptomyces. The phylogenetic tree that co-included a set of reference Streptomyces spp., such as S. coelicolor and S. lividans along with pathogenic S. scabies, S. acidiscabies and S. turgidiscabies, was constructed (Fig. 1c). As shown in Fig. 1c, IFB-A02 and IFB-A03, though belonging to the same genus Streptomyces, were not close relatives in hierarchy. However, from both the phenotypic and phylogenetic aspect, neither of IFB-A02 and IFB-A03 described herein was similar to the newly reported Streptomyces artemisiae sp. nov. by Zhao et al. (2010) whose work was dealt primarily with the taxonomic position of this novel species of Streptomyces, the scope and emphasis of which clearly differ from ours.
Fig. 1

The ISP2 (International Streptomyces Project 2) plating cultures of IFB-A02 (a) and IFB-A03 (b) isolated from Artemisia annua L.; c phylogenetic tree of IFB-A02 and IFB-A03 based on the 16S rDNA sequencing by the neighbor-joining method. The sequence data of the known Actinomyces species included in the tree were retrieved from the Genbank sequence database. ATCC is the abbreviation for American Type Culture Collection, Rockville, MD, USA, and accession numbers for the sequences are as follows: Microbispora amethystogenes, GI 1305426; Nocardioides albus, GI 44818; Streptomyces scabies, GI 971124; S. turgidiscabies WI04-05A, GI 220681900; S. acidiscabies A512, GI 290467510; S. lividans, GI 33323645; S. coelicolor A3(2), GI 32141095; S. galilaeus, GI 10038680; S. argenteolus, GI 10039263; S. setonii, GI 971124; S. caviscabies ATCC 51928, AF 112160; S. albolongus, GI 90960241; S. cavourensis subsp. cavourensis, GI 90654239; S. flavofungini, GI 146759931; S. globisporus subsp. globisporus, GI 90959882; Streptomyces sp. VTT E-99-1330 (A83), GI 16611983; Streptomyces sp. MP47-91, GI 161137727; Streptosporangiacae PA147, AF 223347. The bootstrap values from 1,000 pseudoreplications are shown at each of the branch points on the tree. Bar 0.01 nucleotide (nt.) substitution per nt. position

Endophytism of IFB-A02 and IFB-A03

Taken account into their recovery from thoroughly surface-sterilized plant tissues and their non-pathogenicity when re-introduced into the original host A. annua in the greenhouse (Fig. S2), IFB-A02 and IFB-A03 are most likely endophytes. This was further substantiated by the successful colonization of both strains in the alternative host Arabidopsisthaliana (a model eudicot) as examined by TEM. Inside the new host, they were predominantly found in root (Fig. 2a, b) and stem (Fig. 2e, f) tissues, whereas no internal or surface microbial colonization were discerned in the sectioned root and stem tissues of un-inoculated aseptic seedlings (Fig. 2c, d). Moreover, the Arabidopsis seedlings intercellularly colonized by either IFB-A02 or IFB-A03 within the stem tissues appeared healthy and had intact cells with normal cytoplasm (Fig. 2e, f).
Fig. 2

Transmission electron micrographs of ultra-thin transverse sections from endophyte-inoculated Arabidopsis root and stem tissues in comparison to the un-inoculated counterparts. Micrographs from sections of Arabidopsis roots co-cultivated with IFB-A02 (a) and IFB-A03 (b) show endophytic (En) colonization within host cells. In a, the insertc indicates that the un-inoculated counterpart (control) is microbe free in the intercellular spaces while mitochondria (M) being the predominant organelles in the cytoplasm encompassed by the cell wall (c.w.) Micrographs df represent the stem sections of un-inoculated, and IFB-A02- and IFB-A03-preinoculated Arabidopsis, respectively; the former showed the absence of actinobacteria in the cytoplasm and intercellular spaces (d), and the inoculated endophytes were detectable in the intercellular space of stem parenchyma tissue in the latter two cases (e, f). Chloroplasts are indicated by narrow white arrows. Scale bars 0.5 μm in a and c, 0.2 μm in b, 1 μm in df

In parallel to a known pathogen S. scabies, these two strains were screened for the genomic presence of the key PAI determinants, nos and nec1. As shown in Fig. S3, the genomic absence of both nos and nec1 in IFB-A02 and IFB-A03 excluded their phytopathogenicity. Moreover, the pathogenicity of scab-inducing Streptomyces spp. including S.scabies, was essentially dependent on the TXT production (King et al. 1991). The highest in vitro TXT yield was documented to be achieved in the oatmeal broth (OMB, Loria et al. 1995). We thus evaluated the TXT-producing capacity of the two isolated strains with S. scabies co-assayed as a positive control cultured in the OMB. The LC–ESI–MS analyses of the culture extracts under an extracted ion chromatography (EIC) mode showed that neither IFB-A02 nor IFB-A03 generated any detectable metabolite identical in molecular weight to TXT (Fig. 3a, b), which was abundant in the extract of S. scabies (the main peak in Fig. 3c).
Fig. 3

LC–ESI–MS (liquid chromatography hyphened with an electrospray ionization mass spectrometry) assay for phytotoxin production under an EIC (Extracted Ion Chromatography) mode. The ethyl acetate extracts derived from the OMB cultures of IFB-A02 (a) and IFB-A03 (b) contained no metabolite giving the protonated molecular ion (m/z 439) of thaxtomin A (TXT), which is rich in that of S. scabies (c, positive control)

As for the two isolates IFB-A02 and -A03, their genomic absence of nec1 and inability to produce TXT are in accord with the results by Coombs and Franco (2003) concerning endophytes from wheat roots. In contrast to the thin-layer-chromatography (TLC) technique in Coombs and Franco (2003), this study used LC–ESI–MS analysis with more accuracy for detecting toxin TXT at trace level. Notwithstanding the plant sources of endophytic isolation and phylogenetic position of endophytes streptomycetes, the lack of nec1 in the genome and of TXT production at the chemical level may be served as in vitro markers for differentiating endophytic streptomycetes from pathogenic ones, of which S. scabies is the representative. Therefore, successful in-planta colonization of IFB-A02 and -A03 as aforedescribed, along with the results of in vitro assays, has provided corroborant evidence for their endophytism.

The endophytic IFB-A02 and IFB-A03 were tested for their effects on plant using model eudicot Arabidopsis thaliana. Our preliminary work showed that the culture filtrates of both endophytes, when applied to the seeds, could promote the seedling growth and plant disease resistance (Fig. S1). The in vivo investigation concerning how IFB-A02 and IFB-A03 affect plant physiological traits was conducted by co-cultivation of the respective strain with Arabidopsis from the seed phase. To our surprise, co-cultivation of Arabidopsis with IFB-A03 was advantageous over that of IFB-A02 in terms of the germinating vigor, first leaf emergence and biomass of mature plants (Fig. S1). This could be stemmed from the differential affinity to the alternative host of the two phylogenetically different endophytes (Fig. 1c), which underpinned as well their difference in generating ROS and modulating peroxidase activity as stated below.

Improved defensive capacity of Arabidopsis by endophytic pre-inoculations

Following the disclosure of endophytism and differential benefits of IFB-A02 and IFB-A03 to the alternative host, whether endophytic pre-inoculation may affect defense responses of Arabidopsis thaliana to pathogenic S. scabies was investigated. The plant defense responses are manifold including structural enhancements like cell wall reinforcement, callose deposition and epidermal trichome induction, and metabolic/cellular regulations such as production of phytoalexins, oxidative burst, modulation in peroxidase activity and as well as transcriptional/expressional changes in defensive genes associated with defense signaling pathways that are mediated by phytohormone SA and/or JA (van Loon et al. 2006; Lin and Tan 2011).

Peroxidase activity

This was assayed in endophyte-preinoculated and -uninoculated intact Arabidopsis leaves challenged equally by S. scabies. As shown in Fig. S5, pre-inoculation of Arabidopsis with IFB-A02 evoked more increases in peroxidase activity in leaves than that with IFB-A03 as compared to un-inoculated control (2.3-folds vs. 1.8-folds). Peroxidase plays an important role in regulating the ROS level during plant defense responses and conferring resistance to wide-spectrum pathogens (Bindschedler et al. 2006; Wojtaszek 1997). The discerned increment in peroxidase activity suggested that the endophyte pre-inoculation may potentiate the alternative host’s resistance to invasive pathogen by activating on the defense-related enzyme.

ROS detection

Generation of ROS is an early event in plant response to stresses (Rodriguez et al. 2008). Upon pathogen attack, IFB-A03-preinoculated Arabidopsis leaves demonstrated moderate increases in ROS production (Fig. 4c), while IFB-A02-preinoculated counterparts showed the strongest (Fig. 4b) relative to the un-inoculated counterparts (Fig. 4a). This is in line with the higher increment in peroxidase activity of IFB-A02-treated leaves than that of IFB-A03-treated ones as mentioned above.
Fig. 4

Detection of ROS release by nitroblue tetrazolium (NBT) staining method. In comparison to un-inoculated Arabidopsis leaves (a), IFB-A02- (b) and IFB-A03-preinoculated (c) counterparts showed stronger ROS production as indicated by the dark-blue, water-insoluble formazan precipitates (arrows) arising from the oxidation of NBT by the superoxide radicals

While a microburst of ROS contributes to the SAR state by activating defense responses at a low level throughout the plant (Durrant and Dong 2004), an elevated ROS burst to a relatively greater extent (exceeding the threshold) may be harmful owing to the resultant oxidative damage to DNA, lipids, and proteins (Rodriguez et al. 2008), thereby driving the cells to programed death. The distinct difference between the two endophytes in affecting peroxidase activity and ROS release was in accord with phenotypic observations that the IFB-A02-preinoculated seedlings was inferior in growth to those treated with IFB-A03. Accordingly, our further investigation was performed principally with IFB-A03 in view of its acceptable phytoprotective effects for host. Moreover, its spore-bearing feature would add to its application potential as biopesticides.

Pre-inoculation with IFB-A03 induces SAR-related gene transcripts in Arabidopsis

After ensuring the colonizability and acceptable protective effects of IFB-A03 in Arabidopsis, we investigated the main defense mechanism underlying endophyte-conferred phytoprotection by examining whether transcripts of PR genes altered after IFB-A03 inoculation. The RT-PCR analyses showed that IFB-A03 inoculants could induce transcripts of PR-1 and PR-5 in systemic leaves relative to the untreated control (Fig. 5a) though not to the extent of increment induced by direct SA application (Fig. 5b). Moreover, IFB-A03 pre-inoculation could trigger PR-1 protein expression in systemic leaves (Fig. 5c). Likewise, the extent of increment in PR-1 expression induced by IFB-A03 pre-inoculation was not as great as that done by SA treatment (Fig. 5d). In addition, the analysis of in vitro metabolites of endophyte IFB-A03 revealed no detectable SA signals (Fig. S6). Taken together, endophytic IFB-A03 may potentiate the inoculated plant to activate the defense responses in association with SAR, probably through the interplaying of “live” IFB-A03 inoculants with host plant.
Fig. 5

The pathogenesis-related (PR) gene expression in IFB-A03-inoculated (+IFB-A03) and SA-sprayed (+SA) wild-type Arabidopsis thaliana. a The transcriptional induction of PR-1 and PR-5 in systemic leaves (S) of treated plants (in triplicate) relative to untreated (control). bPR-1 and PR-5 transcripts in the leaves of SA-sprayed plants (in duplicate) versus untreated (control). In a, b, Actin was used as an internal control to normalize the amount of cDNA template in RT-PCR analysis. c The expression of PR-1 protein in leaves of treated plants (+) relative to untreated counterparts (−) as analyzed by western blotting using a rabbit polyclonal anti-PR1 antibody. d The expression of PR-1 protein in the leaves of SA-sprayed plants (+) versus untreated (−) by western blotting analysis as aforementioned. In c, d, protein loads were normalized by Coomassie blue (CBB) staining bands corresponding to the ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) large subunit. The experiment was performed twice with similar results

SA involvement in IFB-A03-activated defense responses in Arabidopsis

The IFB-A03-enhanced plant defense responses were further studied using SAR-compromised NahG transgenic and eds5 mutated Arabidopsis plants. As documented, NahGArabidopsis is unable to accumulate SA due to the degradation of SA to catechol. To investigate the role of SA signaling in the endophyte-mediated defense responses, NahG plants pre-inoculated with IFB-A03 were assessed for PR-1 transcripts post S. scabies challenge relative to un-inoculated counterparts. As shown in Fig. S7, no discernable PR-1 transcripts were induced in either endophyte-preinoculated or -uninoculated NahG, indicating that SA is essential for endophytic IFB-A03-conferred plant resistance against pathogen. Arabidopsiseds5 mutants are susceptible to pathogenic infection, showing a reduced SAR response due to the SA signaling impairment with mutation in EDS5 that encodes a transporter protein required for SA accumulation (Durrant and Dong 2004; Nawrath and Métraux 1999). To our observation, IFB-A03-inoculated eds5 seeds were able to germinate normally and grow to the mature plant. Moreover, IFB-A03-inoculated eds5 seedlings displayed more resistance to pathogen S. scabies than their un-inoculated counterparts (Fig. 6a vs. Fig. 6b).
Fig. 6

Activation of Arabidopsis thaliana defense responses by IFB-A03 inoculants. The 6-week-old eds5 mutant seedlings pre-inoculated with IFB-A03 were healthy at 5 days post pathogen attack (a) compared to the same-aged un-inoculated counterparts upon attack by pathogenic S. scabies (b). c Transcriptional alteration of defense-related genes by IFB-A03 pre-inoculation in Arabidopsis. Both WT (in-planta) and eds5 mutant Arabidopsis pre-inoculated with IFB-A03 at seed sowing (WT+ and eds5+, respectively) along with their untreated counterparts (WT and eds5, respectively) were separately grown in sterile MS agar, and 5-week-old seedlings were equally challenged by S. scabies. Systemic leaves of endophyte-treated WT (WT+) and similarly treated eds5 (eds5+) were harvested 5 days post pathogen attack and RT-PCR-analyzed for PR-1, PR-5, and PDF1.2 transcripts in comparison with those of untreated counterparts. Actin shown in the bottom of c is a loading control

To further elucidate the effects of IFB-A03 inoculation on Arabidopsiseds5 mutants in response to pathogen, transcripts of defensive genes PR-1, PR-5, and PDF1.2 were analyzed in the systemic leaves of the IFB-A03-inoculated seedlings of both wild-type (WT) and eds5-mutated Arabidopsis in comparison to their un-inoculated counterparts. As a result, IFB-A03-inoculated eds5 exhibited upregulated transcripts of SAR-related genes PR-1 and PR-5 (Fig. 6c) though to a lesser extent than those observed with the IFB-A03-treated WT. Endophyte-inoculated eds5 also displayed the induced transcript of PDF1.2—a marker gene in JA signaling pathway (Fig. 6c). These observations for IFB-A03-inoculated eds5, which stand as opposed to those documented for eds5 mutants, suggest that IFB-A03 pre-inoculation may partially “rescue” SA-deficient defense pathway in eds5 mutants of Arabidopsis. In addition, the induction of PDF1.2 transcripts implies that JA signaling pathway may be operative to offset the defense responses that are “partially complemented” by the interactions of IFB-A03 inoculants with eds5 Arabidopsis. As suggested by Conn et al. (2008), the streptomycetes may be perceived by plant as “minor” pathogens to trigger an array of defense responses.

Endophytic IFB-A03 acts toward upstream of SA accumulation

Integrated with the enhanced disease resistance observed with IFB-A03-inoculated eds5 mutants, the upregulation in PR-1 and PR-5 transcripts as aforementioned indicates that IFB-A03 probably potentiates the plant to establish the SAR state associated with SA signaling. As documented (Nawrath and Métraux 1999; Durrant and Dong 2004), eds5 mutants have very low levels of SA after pathogen infection. To test whether upstream or downstream of SA endophytic IFB-A03 acts with SA signaling pathway, endogenous SA levels were determined in IFB-A03-treated versus -untreated WT and eds5 mutants. If changes in SA levels occur in the endophyte-treated eds5 relative to untreated counterparts, it holds true for the notion that IFB-A03 may participate in the SA-mediated defense pathway. As shown in Fig. 7, the clear SA elevation was discerned with IFB-A03-treated eds5 relative to untreated counterparts at 0- and 48 h post pathogen attack, suggesting that IFB-A03 may not induce signaling downstream of SA signaling molecules. In addition, IFB-A03 was not an in vitro SA producer based on our preliminary studies of IFB-A03 culture extracts which demonstrated that no detectable signals corresponding to SA was present in its metabolites (Fig. S6). Taken together, these data suggest that IFB-A03 probably potentiate SA signaling pathway at a node upstream of SA accumulation in Arabidopsis thaliana.
Fig. 7

The SA accumulation in response to S. scabies attack. The systemic leaves of IFB-A03-pre-inoculated wild-type (Col-0) and eds5 Arabidopsis thaliana were harvested at 0- and 48 h post attack, with water-treated ones as the control, and assayed for endogenous free SA. Each bar represents the average and SD of three replicates. a Free SA levels in leaves of IFB-A03-treated versus untreated Arabidopsis wild-type (Col-0). “IFB-A03 + Col-0” indicates IFB-A03 pre-inoculated Col-0 plants; b free SA levels in leaves of IFB-A03-treated versus untreated Arabidopsis eds5 mutants. “IFB-A03 + eds5” indicates IFB-A03 pre-inoculated eds5

As reviewed by Schrey and Tarkka (2008), plants enhance their resistance against pathogen ingress through a versatile array of local or systemic defense responses, involving (1) local defenses by structural enhancement of plant cell walls, (2) streptomyte-induced systemic plant resistance by activating SA and JA/ethylene signaling networks, and (3) plant growth promotion correlated with its suppression of pests. Our results indicate that at least two mechanisms, (2) and (3) as mentioned above, may be implicated in endophytic streptomycete-conferred phytoprotection. Mechanism (2) is distinct in light of our data regarding SA accumulation and upregulation of PR transcripts while mechanism (3) acts indirectly to promote plant growth through enhancing the disease resistance against pests, as shown in Figs. S1 and S2. However, the plant-streptomycetes interaction mechanisms may vary with different plant and Streptomyces species. It has been found that the alteration in spruce root architecture is one of mechanism underlying the enhanced disease resistance following streptomycete inoculation (Lehr et al. 2008), referring to mechanism (1) (Schrey and Tarkka 2008).

This study addressed or at least suggested a key event ahead of SA signal accumulation during the SAR pathway in the model plant (Fig. S8). Desired is full revelation of the plant defense signaling component (gene or gene regulator) with which endophytic IFB-A03 interacts to gain more insights into endophyte-plant mutualism.

Streptomycetes are abundant producers of various secondary metabolites such as plant growth regulators, antimicrobials, and siderophores (Conn et al. 2008). From a morphological aspect, they possess such specialized structures as spores which facilitate their survival in nature, as well as in-planta colonization, leading to an advantageous mode of colonization over sessile bacteria. These features of streptomycetes add to their applicative values for biologic control. Previous studies have showed that inoculation of wheat with the endophytic Streptomyces sp. strain EN27 may promote growth and enhance disease resistance both in vitro and in-planta (Coombs 2002).

Previous studies (Conn et al. 2008) demonstrated the involvement of SA in the plant resistance endowed by Streptomyces sp. EN27 in response to pathogenic E. carotovora subsp. Carotovora, operating via an NPR1-independent signaling pathway, based on the data that strain EN27-treated npr1-1 mutants infected with E. carotovora subsp. Carotovora displayed the induction of PR-1, PDF1.2, and the plants showed little disease symptom. Our study indicated that endophytic Streptomyces sp. IFB-A03 may activate the SA-dependent signaling in Arabidopsis thaliana, thereby conferring plant resistance against pathogenic S. scabies. Further investigation is needed to address whether the IFB-A03-activated signaling is NPR1 dependent or not.

In summary, we show here the validation of the ethnobotanical path for recovering the plant defensive microbes such as the newly recognized endophytic Streptomyces sp. IFB-A03 sourced from the stress-adaptive herb Artemisia annua L. The disease resistance endowed by this endophyte is reflected unequivocally by the pre-inoculation test to display, in the alternative host Arabidopsis thaliana, its strengthening effect to retard subsequent infection of pathogenic S. scabies. Concerning its mode of action, IFB-A03 may stimulate plant defenses by acting upstream of SA accumulation in the SA-dependent signaling pathway. While adding renewed information to the feasibility of endophytism-based “cross-protection,” the present work addresses collectively that the endophytic Streptomyces sp. IFB-A03 is a promising candidate for eco-friendly biocontrol agents. As strengtheners of plant defenses, the endophytic streptomycetes merit further consideration as the missing force in the disease resistance against pathogenic Streptomyces sp.


We thank Profs S. Lu (Nanjing Univ.) and D. T. Ren (China Agric. Univ.) for providing seeds of wild-type Col-0 and eds5 mutant, NahG transgenic Arabidopsis thaliana, respectively, and Prof. Z. Hong (Nanjing Univ.) for helpful suggestions on the manuscript. We are also grateful to the two anonymous referees for their comments and advice for the improvement of manuscript. This work was co-financed by grants from NSFC (30821006 & 90813036) and MOST (2009ZX09501-013).

Conflict of interest

The authors declare that they have no conflict of interest.

Supplementary material

425_2012_1741_MOESM1_ESM.pdf (87 kb)
Supplementary material 1 (PDF 87 kb)
425_2012_1741_MOESM2_ESM.jpg (409 kb)
Supplementary material 2 (JPEG 410 kb)
425_2012_1741_MOESM3_ESM.pdf (14 kb)
Supplementary material 3 (PDF 13 kb)
425_2012_1741_MOESM4_ESM.pdf (646 kb)
Supplementary material 4 (PDF 646 kb)
425_2012_1741_MOESM5_ESM.pdf (15 kb)
Supplementary material 5 (PDF 14 kb)
425_2012_1741_MOESM6_ESM.pdf (38 kb)
Supplementary material 6 (PDF 38 kb)
425_2012_1741_MOESM7_ESM.tif (1.2 mb)
Supplementary material 7 (TIFF 1268 kb)
425_2012_1741_MOESM8_ESM.jpg (27 kb)
Supplementary material 8 (JPEG 28 kb)
425_2012_1741_MOESM9_ESM.doc (68 kb)
Supplementary material 9 (DOC 68 kb)

Copyright information

© Springer-Verlag 2012

Authors and Affiliations

  • Lan Lin
    • 1
    • 2
  • Hui Ming Ge
    • 1
  • Tong Yan
    • 1
  • Yan Hua Qin
    • 1
  • Ren Xiang Tan
    • 1
  1. 1.Institute of Functional Biomolecules, State Key Laboratory of Pharmaceutical BiotechnologyNanjing UniversityNanjingPeople’s Republic of China
  2. 2.Department of Bioengineering, Medical SchoolSoutheast UniversityNanjingPeople’s Republic of China

Personalised recommendations