Applied Microbiology and Biotechnology

, Volume 78, Issue 6, pp 983–990

Investigation on the infection mechanism of the fungus Clonostachys rosea against nematodes using the green fluorescent protein

Authors

  • Lin Zhang
    • Laboratory for Conservation and Utilization of Bio-resourcesYunnan University
    • Key Laboratory for Microbial Resources of the Ministry of EducationYunnan University
  • Jinkui Yang
    • Laboratory for Conservation and Utilization of Bio-resourcesYunnan University
    • Key Laboratory for Microbial Resources of the Ministry of EducationYunnan University
  • Qiuhong Niu
    • Laboratory for Conservation and Utilization of Bio-resourcesYunnan University
    • Key Laboratory for Microbial Resources of the Ministry of EducationYunnan University
  • Xuna Zhao
    • Laboratory for Conservation and Utilization of Bio-resourcesYunnan University
    • Key Laboratory for Microbial Resources of the Ministry of EducationYunnan University
  • Fengping Ye
    • Laboratory for Conservation and Utilization of Bio-resourcesYunnan University
    • Key Laboratory for Microbial Resources of the Ministry of EducationYunnan University
  • Lianming Liang
    • Laboratory for Conservation and Utilization of Bio-resourcesYunnan University
    • Key Laboratory for Microbial Resources of the Ministry of EducationYunnan University
    • Laboratory for Conservation and Utilization of Bio-resourcesYunnan University
    • Key Laboratory for Microbial Resources of the Ministry of EducationYunnan University
Applied Genetics and Molecular Biotechnology

DOI: 10.1007/s00253-008-1392-7

Cite this article as:
Zhang, L., Yang, J., Niu, Q. et al. Appl Microbiol Biotechnol (2008) 78: 983. doi:10.1007/s00253-008-1392-7

Abstract

The fungus Clonostachys rosea (syn. Gliocladium roseum) is a potential biocontrol agent. It can suppress the sporulation of the plant pathogenic fungus Botrytis cinerea and kill pathogenic nematodes, but the process of nematode pathogenesis is poorly understood. To help understand the underlying mechanism, we constructed recombinant strains containing a plasmid with both the enhanced green fluorescent protein gene egfp and the hygromycin resistance gene hph. Expression of the green fluorescent protein (GFP) was monitored using fluorescence microscopy. Our observations reveal that the pathogenesis started from the adherence of conidia to nematode cuticle for germination, followed by the penetration of germ tubes into the nematode body and subsequent death and degradation of the nematodes. These are the first findings on the infection process of the fungal pathogen marked with GFP, and the developed method can become an important tool for studying the molecular mechanisms of nematode infection by C. rosea.

Keywords

Clonostachys roseaGreen fluorescent protein (GFP)Protoplast transformationNematodeInfection

Introduction

Plant-parasitic nematodes cause severe damages to agriculture and forests, amounting to 78 billion US dollars per year over the world (Barker 1998). Synthetic chemical pesticides have been widely used against nematodes. However, such pesticides can cause a series of problems including environment pollution and long-term residue issues. As a result, increasing attentions have been paid to biological control agents. Nematophagous fungi, one of the natural enemies of nematodes, have been proposed as biological agents to control the harmful nematodes (Siddiqui and Mahmood 1996; Nordbring-Hertz et al. 2000; Yang et al. 2007).

The fungus Clonostachys rosea (syn. Gliocladium roseum) is a widely distributed facultative saprophyte in the soil (Schroers et al. 1999). Morandi et al. (2003) found that C. rosea could suppress sporulation of the plant pathogenic fungus Botrytis cinerea and be used for the control of botrytis blight. Dong et al. (2004) reported that certain chemicals isolated from C. rosea showed strong nematicidal activities against several nematodes, Caenorhabditis elegans, Panagrellus redivivus, and Bursaphelenchus xylophilus. In our previous reports, two extracellular serine proteases (Lmz1 and PrC) were isolated from C. rosea and identified as important factors in fungal pathogenicity (Zhao et al. 2005; Li et al. 2006). However, possible mechanisms involved in the infection of nematodes by C. rosea have not yet been fully understood.

The marker protein green fluorescent protein (GFP) has been expressed in a wide variety of organisms including bacteria, fungi, plants, and animals (e.g., Chalfie et al. 1994; Casper and Holt 1996; Spellig et al. 1996). Several recent studies reported using GFP as a tool to study the interaction between pathogenic fungi and their hosts (Cantone and Vandenberg 1999; Inglis et al. 2000; Atkins et al. 2004; Lu et al. 2004; Ying and Feng 2006; Jiang et al. 2007). Among them, Atkins et al. (2004) developed a transformation system for the parasitic fungus Pochonia chlamydosporia by using plasmids carrying the GFP marker gene gfp and the hygromycin resistance gene hph. Ying and Feng (2006) reported a blastospore-based transformation system and integrated phosphinothricin resistance and GFP genes into the genome of Beauveria bassiana. These results suggested that it is possible to develop a GFP transformation system for investigating the pathogenic mechanisms of C. rosea against nematodes.

In this study, we described the transformation and expression of the egfp gene in the nematophagous fungus C. rosea, and we then used the transformed GFP as a marker to investigate the pathogenesis of this fungus against nematodes by monitoring its expression in the infection process.

Materials and methods

Organisms and cultural conditions

The nematophagous fungus C. rosea (strain 611), isolated from a field soil sample in Yunnan Province, was used in this study. The fungus was cultured on potato dextrose agar (PDA) medium, and had been deposited in the China General Microbiological Culture Collection Center (CGMCC 0806).

The saprophytic nematode P. redivivus was grown in an oatmeal medium at 26–28°C. The P. redivivus was washed thoroughly with 50 mM sodium phosphate buffer (pH 7.0) before being used in the assays.

Plasmid construction

Plasmid pEGFP-C1 was purchased from BD Biosciences Clontech (USA) and this plasmid contains the egfp gene. A pair of oligonucleotide primers egfpfor (5-ATCGCCATGGTGAGCAAGGGCGAG-3) and egfprev (5-ATCGCCATGGCTTGTACAGCTCGTCC-3) containing the NcoI restriction site (underlined) was designed based on the nucleotide sequence of egfp in plasmid pEGFP-C1 (GenBank accession number U55763). Polymerase chain reaction (PCR) was performed using the plasmid pEGFP-C1 as the template and the PCR product was cloned into vector pMD18-T (Takara, Japan). The egfp gene fragment was inserted into the NcoI site of the fungal expression vector pAN52 (provided by Dr. PJ Punt). The presence and the direction of the inserted DNA fragment was confirmed by PCR using the forward primer egfpfor based on the sequence of egfp and reverse primer trpCrev (5-ATCGAAGCTTGAGTGGAGATGTGGAGTG-3) according to the trpC sequence. The constructed plasmid was named pANG and this plasmid contained the promoter of gpdA, the egfp coding gene, and the termination sequence of the trpC gene. Moreover, the hygromycin-resistant gene hph was isolated from the plasmid pBS/Sk-hph (provided by Dr. Zhang YJ) using XbaI digestion and inserted into the plasmid pANG to generate recombinant plasmid pANGH3. The pANGH3 was used for subsequent transformation experiment.

Fungal transformation

Fungal transformation was performed based on the methods of Tunlid et al. (1999), Lu et al. (1994), and Xu et al. (2005) with the following modifications. Firstly, the fungus was incubated on PDA plate for 6–7 days, and washed using sterile water containing 2% Tween-80 to obtain conidia. Conidia were incubated in 500 μl of sterile TG liquid medium (per liter: 10 g tryptone, 10 g glucose) in an eppendorf tube on a rotary shaker (180 rpm) at 26°C for 36 h. After that, the mycelia were harvested and washed with 500 μl of the MN solution (0.3 M MgSO4, 0.3 M NaCl). The mycelia were suspended in 500 μl MN buffer containing 5 mg mL−1 snailase (Beijing Jingke Company, China) and 5 mg mL−1 cellulase (Yakult Honsha Co. Ltd., Japan). The suspension was incubated on a shaker (180 rpm) at 28°C overnight, and the protoplasts were collected by centrifugation (6,000 rpm, 10 min) at 4°C. After washed with 500 μl KTC buffer (1.2 M KCl, 10 mM Tris–HCl, pH 7.5, 50 mM CaCl2), the protoplasts were resuspended in 100 μl KTC and used for transformation immediately.

For transformation, 100 μl protoplasts (about 8.0 × 107) were mixed with 50 μl (about 10 μg) of linear vector pANGH3 (linearized by HindIII). After incubation on ice for 30 min, an equal volume (150 μl) of PTC (10 mM Tris–HCl, pH 7.5, 50 mM CaC12, 50% w/v PEG6000) was added and mixed gently. The mixture was incubated at 28°C for 1 h, and was then added to 10 ml of the PDSSA medium (PDA supplying with 10 g L−1 molasses and 0.4 M saccharose) containing 600 μg mL−1 of hygromycin B (Roche Corporation, Germany), and poured into Petri dishes. The plates were incubated at 26°C to select for the growth of transformants. In this experiment, conidia were also used to prepare the protoplasts, and transformation was performed following the same procedures.

DNA extraction

Successful transformants were transferred onto PDA containing 600 μg mL−1 of hygromycin B overlaid with a cellophane sheet (Flexel, Atlanta, GA, USA) and incubated for 10 days at 26°C. The mycelium was separated from the medium by removing the cellophane from the agar. The mycelia were ground in liquid nitrogen. Genomic DNA was then extracted using the method of Raeder and Broda (1985).

Validation of transformant using PCR

Primers prCfor (5-TCCGAATTCATGCGCGTTTCTGCTCTC-3) and prCrev (5-AGCGGATCCCAGAAGAGGGACATTTACAC-3) were designed based on the known sequence of protease PrC (Liang, unpublished data), and used to amplify the gene prC in transformants to exclude the possibility of contamination. In addition, the gene egfp was simultaneously amplified to validate the insertion of egfp in the genome.

Southern analysis of transformants

The integration events in the four transformants were analyzed by southern blot. The hybridization probe here was the partial fragment of the gene egfp. The probe was labeled with digoxin and the subsequent southern hybridization was conducted using Digoxigenind-UTP Kits (Roche, Germany) and nylon membrane (Roche, Germany) according to the instructions.

Mitotic stability of the transformant

The stability of hygromycin B resistance in the transformant was tested according to the method of Tunlid et al. (1999). The transformant was incubated on PDA plate without hygromycin B for at least five continuous transfers. The transformant was monitored using fluorescence microscopy for the expression of GFP.

Observation of C. rosea infection against nematode P. redivivus

The bioassay of recombinant C. rosea strains against nematode P. redivivus was done as follows. The transformant of C. rosea, strain 611g-4, was incubated on PDA at 26°C for 5–6 days. A block of 2 cm2 medium in the center of the plate was removed and 20–30 nematodes were added to the empty block (Gao et al. 1996). The fungus then grew into the space and the nematode specimens were examined after incubating at 26°C for another 2–5 days. The infection process was observed under a Nikon 800 Eclipse microscope (Nikon Corporation, Japan) equipped for epifluorescence with a mercury lamp and an excitation filter of 450–490 nm (blue light) and a barrier filter of 515 nm.

Results

Plasmid construction

During the construction of the recombinant plasmid, it was very important to validate the direction of the inserted egfp DNA. This is because only a single enzyme digestion could be used here. In order to ensure the expression of egfp, two methods were designed to choose the right plasmid. Three plasmids (pANG18, pANG19, and pANG20) were validated by the digestion using restriction enzyme Bsp1407I. The Bsp1407I was localized at the 1422nt position in plasmid pAN52 and 717nt position in gene egfp. The insertion localization of NcoI was at 2305nt position. If the insertion direction was right, the digestion by Bsp1407I would produce a fragment about 1,600 bp (2,305 − 1,422 + 717 = 1,600). In Fig. 1a, lane 3 showed the expected size of plasmid pANG18. In contrast, lanes 1 and 2 showed only a fragment of approximately 910 bp (2,305 − 1,422 + 744 − 717 = 910) when digested by Bsp1407I (Fig. 1a). The correct insertion and orientation of plasmid pANG18 was further confirmed by PCR. A fragment of about 1,500 bp was amplified (Fig. 1b, lane 2) using the egfp forward primer (egfpfor) and trpC reverse primer (trpCrev), and no product was amplified using the egfp reverse primer (egfprev) and trpC reverse primer (trpCrev; Fig. 1b, lane 1). Therefore, the gene egfp was cloned into the plasmid pAN52, and the plasmid pANG18 was successfully constructed as expected. Similarly, the hygromycin-resistant gene hph was isolated from the plasmid pBS/Sk-hph using XbaI and inserted into the plasmid pANG to generate the recombinant plasmid pANGH3.
https://static-content.springer.com/image/art%3A10.1007%2Fs00253-008-1392-7/MediaObjects/253_2008_1392_Fig1_HTML.gif
Fig. 1

Validating the direction of egfp gene within plasmid pANG by enzyme digestion and PCR: a Three clones were digested by enzyme Bsp1407I. Lane M DNA marker (DL2000), lane 1 pANG20 digested by Bsp1407I, lane 2 pANG19 digested by Bsp1407I, lane 3 pANG18 was digested by Bsp1407I. b pANG18 was further confirmed by PCR. Lane M DNA marker (DL2000), lane 1 PCR result of the negative control using primers egfprev and trpCrev, lane 2 PCR result using primers egfpfor and trpCrev

Transformation and validation of transformants

The protoplast was prepared in 1.5-mL eppendorf tubes in order to avoid the contamination. The protoplast yield of C. rosea was 0.3–0.8 × 107 mL−1. The protoplast regeneration frequency was not quantitatively analyzed, and the reasons were analyzed in the “Discussion”.

Transformants became visible after incubation for 7–10 days at 26°C. Interestingly, when varying the volume of medium at 7, 10, 15, and 18 mL per transformation plate (90 mm), the number of transformant was 1, 3, 0, and 0, respectively, and the transformation frequencies were separately 0.1, 0.3, 0, and 0 transformants μg−1, which suggested oxygen was an important factor influencing the transformation frequency of C. rosea. However, no transformant was found using conidia preparations, which suggested that all transformants were obtained from the protoplasts produced by mycelium but not by conidia.

Four transformants were validated using PCR. A 750-bp fragment was amplified from the plasmid pEGFP-C1 and four random transformants (611g-1, 611g-2, 611g-3, and 611g-4) using primers egfpfor and egfprev. The presence of the fragment was consistent with our expectation of positive cloning (Fig. 2a, lanes 1–5). Similarly, a 1.5-kb fragment was amplified from the wild-type strain and the four transformants using primers prCfor and prCrev. The results suggested that the egfp gene is successfully inserted into the genomes of those transformants.
https://static-content.springer.com/image/art%3A10.1007%2Fs00253-008-1392-7/MediaObjects/253_2008_1392_Fig2_HTML.gif
Fig. 2

a Transformants validated by PCR. Lane M DNA marker (DL2000), lane 1 plasmid pEGFP-C1 (positive control) used as template and PCR was performed using primers egfpfor and egfprev, lanes 2–5 the genomic DNA of four transformants (611g-1, 611g-2, 611g-3, and 611g-4) were used as templates, respectively, and PCR was performed using primers egfpfor and egfprev. Lane 6 positive control: wild-type strain 611 (positive control) was used as template and PCR was performed using primers prCfor and prCrev. Lanes 7–10 the genomic DNA of four transformants (611g-1, 611g-2, 611g-3, and 611g-4) were used as templates, respectively, and PCR was performed using primers prCfor and prCrev. b. Southern blot analysis of the transformants. Genomic DNAs from C. rosea strains 611g-1 (lane 1), 611g-2 (lane 2), 611g-3 (lane 3), and 611g-4 (lane 4) were digested with Bsp1407I and hybridized to a DIG-labeled DNA probe consisting of an egfp coding sequence, respectively. Filled triangles indicate the hybridization bands

Southern blot analysis of transformants

The genomic DNA of the four transformants was digested using Bsp1407I, respectively, and hybridized with the internal coding sequence probe of the egfp gene. Hybridization bands 2, 2, 2, and 3 were observed in the four strains, respectively (Fig. 2b). The Bsp1407I was localized at the 717nt position in gene egfp and 3′ terminal of the egfp gene, so each hybridization band was corresponding with a single copy egfp, which suggested that 2, 2, 2, and 3 copies of gene egfp were inserted into the genomes of transformants (611g-1, 611g-2, 611g-3, and 611g-4), respectively.

Mitotic stability and the morphological analysis of transformants

The four transformants (611g-1, 611g-2, 611g-3, and 611g-4) showed a high degree (100%) of mitotic stability. In comparison to the wild-type parental strain, these transformants showed similar morphological properties on non-selective medium and retained their stable GFP expression after incubated continuously on PDA plate for five transfers (Fig. 3). However, transformants grew slower than the wild-type strain. The colony diameter of transformants was about 2.0 cm after incubation on PDA for 7 days at 26°C, and that of the wild type was 9.0 cm. The morphological properties of the transformants did not obviously differ from those of the wild-type strain.
https://static-content.springer.com/image/art%3A10.1007%2Fs00253-008-1392-7/MediaObjects/253_2008_1392_Fig3_HTML.gif
Fig. 3

Comparison of mycelia of transformant strain 611g-4 under light and fluorescence. a Mycelia of the transformant 611g-4 after five transfers, observed under the light microscopy. Scale bar = 50 μm. b Mycelia of the transformant 611g-4 after five transfers, observed under the fluorescence microscopy. Scale bar = 50 μm

GFP expression and infection against nematodes

The transformant strain 611g-4 of C. rosea began to attack the nematodes after co-incubation for 1 day on a PDA plate (Fig. 4). The conidia of the fungus were found secreting a glutinous substance, leading to the conglutination of the nematode. It was difficult for the conglutinated nematodes to escape from the mucous liquid containing abundant conidia. When the nematodes were exhausted from their escape attempts and stopped moving, conidia could germinate and penetrate the body of the nematodes, and grow by digesting the tissue of nematodes. Finally, the nematodes were degraded gradually by the fungus.
https://static-content.springer.com/image/art%3A10.1007%2Fs00253-008-1392-7/MediaObjects/253_2008_1392_Fig4_HTML.gif
Fig. 4

The infection process observed under the fluorescence microscopy. (1) Conidia adhered to the cuticle of nematodes and bourgeoned. Arrow 1, conidia adhered to the cuticle of nematodes; arrow 2, conidia bourgeoned and mycelia invaded into the body of nematodes. (2) Mycelia grew inside the body of nematodes. Arrow, mycelia inside the nematodes. (3) Mycelia grew out from the body of the nematodes. Arrow, mycelia outside the nematodes. (4) The fungus produced abundant conidia. Arrow, conidiophore and conidia. In (1)–(4), all A’s were observed under light microscopy, and B’s were observed under fluorescence microscopy. Scale bar = 20 μm

Discussion

C. rosea is capable of rapid growth and producing excessively large number of conidia in a short time. For several reasons, the protoplast preparation of C. rosea was more difficult than for other filamentous fungi. Firstly, C. rosea grew very fast and large numbers of conidia were produced within 24–36 h, and it was difficult to remove the abundant conidia by filtration repeatedly. The conidia could easily become the false positives if the protoplast regeneration frequency was estimated, resulting in incalculable protoplast regeneration frequency. In addition, the large numbers of conidia competed for nutrition with protoplasts and thus affected the regeneration of protoplasts. Secondly, 200 μg mL−1 of hygromycin B can completely inhibit the mycelial growth of the fungus. However, the concentration of hygromycin B has to be increased to 600 μg mL−1 to kill most conidia. Thirdly, the high oxygen demand for this fungus can also contribute to a low transformation frequency for C. rosea. Since the number of transformants is typically counted using visible colonies on plates, due to the limited oxygen requirements, the transformants may not grow to form colonies even if the DNA fragments were integrated into the fungal genome. In order to improve the transformation frequency of C. rosea, it is essential to control the volume of PDSSA medium at 7–10 mL per plate (90 mm). Moreover, compared with the wild-type strain, the transformants grew slower. The slow growth may be caused by factors such as the abundant expression of the egfp gene or other genes involved in mycelia growth but were disrupted during random insertional mutation of egfp (Xu et al. 2005). From the result of southern blot analysis (Fig. 2b), the size of hybridization bands in four transformants (611g-1, 611g-2, 611g-3, and 611g-4) was different, which suggested that the egfp gene was inserted into the genomes of those transformants randomly.

Nematophagous fungi possess the unique ability of attacking living nematodes and consuming them. To capture nematodes, they produce special adhesive and mechanical traps, or they use conidia as infectious agents (Barron 1977). Recognition and adhesion were the first steps in the infection. However, little is known about the molecular mechanisms of recognition and adhesion. To date, only lectin has been reported to be involved in the recognition process. The interests of studying lectins in nematode-trapping fungi came from an observation that the interaction between Arthrobotrys oligospora and nematodes was mediated by a GalNAc (N-acetyl-d-galactosamine)-specific fungal lectin binding to receptors present on the nematode surface (Nordbring-Hertz and Mattiasson 1979). Subsequently, a carbohydrate-binding protein from the capture organs of the fungi, not present on hyphae, was isolated and partially characterized (Borrebaeck et al. 1984). Similar experiments have indicated that lectins likely play a role in the adhesion to host surfaces by a number of parasitic and symbiotic fungi (Nordbring-Hertz and Chet 1986). Moreover, increasing evidences showed that extracellular hydrolytic enzymes including proteases, collagenase, and chitinase may be involved in nematode cuticle penetration and host cell digestion (e.g., Tunlid et al. 1994; Åhman et al. 2002; Yang et al. 2007). Recently, two extracellular serine proteases (Lmz1 and PrC) were isolated from C. rosea. Both genes were found involved in C. rosea infection against nematodes (Zhao et al. 2005; Li et al. 2006). In the present study, we found that the conidia of C. rosea play an important role in infection against nematodes. The conidia excrete a mucous liquid that can stick to the cuticle of nematodes. Microscope observations suggested that the glutinous substance could prevent the nematodes escape from the adhesive areas (supplied material) formed by the secretions of the conidia.

The transformant of C. rosea 611g-4 was used to observe the infection events against nematodes under fluorescence microscopy. Overall, the process could be divided into four steps as follows. First, when mixed with nematodes in plates for 1–2 days, the conidia adhered to the cuticle of nematodes, and nematodes were immobilized and dead. Then, the conidia germinated and invaded into the body of the nematodes with the help of hydrolytic enzymes, such as protease and chitinase (Lopez-Llorca et al. 2002; Zhao et al. 2005; Li et al. 2006; Gan et al. 2007). Second, the mycelia propagated rapidly inside the nematodes using the decomposed tissue as nutrients on the third day of treatment. Third, after another 1–2 days, a great mass of mycelia was formed inside the coelom of nematodes, and these mycelia began to proliferate to the outside of the nematodes. And fourth, the nematode was degraded gradually during the subsequent 2 days, conidiophores were produced, and abundant conidia were released for the next infection cycle.

Although research on the biological control of nematodes by fungi has been ongoing for decades, few commercial agents are available for field use. The main problem is a lack of knowledge on the ecology of the biocontrol agents in the field, including their germination, proliferation, control efficiency, and sensitivity to population variation (Stirling et al. 1998). Low temperature scanning electron microscopy (Jansson et al. 2000) and fluorescence microscopy observation techniques (Saxena and Lysek 1993; Jensen et al. 1998) have been developed to visualize these fungi and demonstrate their activity in natural habitat. However, knowledge on the molecular mechanism of nematophagous fungal infection against nematodes in nature is still limited due to the lack of suitable techniques. In this study, the egfp gene was expressed successfully in C. rosea by protoplast transformation. The knowledge obtained and the tools developed here should help develop an effective method to investigate the ecological properties of nematophagous fungi and their antagonist processes against nematodes.

Acknowledgments

We sincerely thank Dr. PJ Punt (TNO Nutrition and Food Research Institute, Netherlands) and Dr. Zhang YJ (Biotechnology Research Center of Southwest University, China) for providing plasmids pAN52 and pBS/Sk-hph. This work was funded by the National Basic Research Program of China (approved no. 2007CB411600), by projects from the National Natural Science Foundation of China (approved nos. 30630003, 30570059 and 30660107), and by the Department of Science and Technology of Yunnan Province (approval nos. 2004C0001Z, 2005NG05).

Supplementary material

View video
ESM 1

(WMV 999 kb)

Copyright information

© Springer-Verlag 2008