Transformation of Epichloë typhina by electroporation of conidia

Background Choke, caused by the endophytic fungus Epichloë typhina, is an important disease affecting orchardgrass (Dactylis glomerata L.) seed production in the Willamette Valley. Little is known concerning the conditions necessary for successful infection of orchardgrass by E. typhina. Detection of E. typhina in plants early in the disease cycle can be difficult due to the sparse distribution of hyphae in the plant. Therefore, a sensitive method to detect fungal infection in plants would provide an invaluable tool for elucidating the conditions for establishment of infection in orchardgrass. Utilization of a marker gene, such as the green fluorescent protein (GFP), transformed into Epichloë will facilitate characterization of the initial stages of infection and establishment of the fungus in plants. Findings We have developed a rapid, efficient, and reproducible transformation method using electroporation of germinating Epichloë conidia isolated from infected plants. Conclusions The GFP labelled E. typhina provides a valuable molecular tool to researchers studying conditions and mechanisms involved in the establishment of choke disease in orchardgrass.


Introduction
Orchardgrass (Dactylis glomerata L.) is an important forage grass species. About 97% of orchardgrass seed used for pastures and hay in North America is produced in the Willamette Valley in western Oregon. Choke disease, caused by the endophytic fungus Epichloë typhina, was first reported in the Willamette Valley during the mid 1990s [1]. Infected plants remain asymptomatic during most of the year. In the spring, E. typhina proliferates within reproductive tillers, encasing the developing seed head in a dense mycelial mat (stroma). However, the stem continues to grow, revealing an elongated, white stromal mass that resembles a small cattail. Conidia of one or two mating types are produced on the surface of each stroma. Fertilization requires the transfer of conidia of one mating type to stroma of the opposite mating type. This is typically accomplished by flies in the genus Botanophila [2]. Following fertilization, perithecia develop within the stroma surface [3,4]. The perithecia produce ascospores, which are dispersed by wind and cause new plant infections. The pathogen has not been shown to be transmitted through the seed in orchardgrass, but it is seed transmitted in other grasses [3][4][5][6].
The recent introduction and rapid spread of choke disease is a serious problem for orchardgrass seed producers in the Willamette Valley [1,7,8]. In England, where the disease has occurred for many years, the number of fields with choke disease increased from zero to a few during the first year of production to 33-81% by the second to fifth year of seed production [9]. In France, choke was reported to affect up to 30% of the tillers in a field by the fourth year of seed production [10,11]. Since the first incidence of choke reported in Oregon in the mid 1990s [1], the fungus has spread tõ 90% of orchardgrass seed production fields and has caused yield losses of up to 65% in individual fields [8].
Very little is known about the conditions necessary for successful infection of orchardgrass by E. typhina. Fertilization of the fungal stroma is facilitated by flies [2,12], but fly density was not correlated with reproductive success of the fungus in western Oregon [2,13]. Fertilization of E. typhina by ascospores has been reported recently and may contribute to the rapid spread of choke in orchardgrass [14]. Attempts to infect orchardgrass foliage or flowers with conidia or ascospores were not successful [4]. Germination of E. typhina ascospores and conidia on the cut ends of seed stalks of orchardgrass and growth of hyphae down the pith has been reported [4][5][6], although it was not established whether these plants ultimately became infected through the stalks. Infection of young tillers by ascospores or conidia produced by ascospores under very favorable experimental conditions was recently reported [15]. It is not known when or how infections occur in the field or how long the latent period is between infection and manifestation of symptoms in the field.
Detection of E. typhina early in the disease cycle can be difficult due to sparse distribution of hyphae in the plant. Therefore a sensitive method to detect the fungus in plants would provide an invaluable tool for elucidating the conditions and establishment of the infection in orchardgrass. Green Fluorescent Protein (GFP) has been utilized extensively as a marker to aid in the development of fungal transformation systems and to examine early stages in plant/fungal interactions [reviewed in [16], [17][18][19][20][21][22]]. Translation fusions between GFP and fungal genes of interest have also been utilized to determine gene expression patterns and to investigate the role of specific genes in the infection process [16,23]. Transformation of E. typhina with the GFP marker gene will facilitate the characterization of the initial stages of infection and progression of the disease in plants [16]. There are several different methods available for fungal transformation, such as chemical induction, Agrobacterium-mediated transformation, particle bombardment and electroporation [ [16,[24][25][26][27], reviews and references within, [28,29]]. Chemical transformation of protoplasts of Epichloë festucae has been reported and was used to determine the expression patterns of the lolitrem biosynthetic genes [30] and to look at the infection process of E. festucae in Lolium perenne [31]. Electroporation of germinating conidia has been used successfully to transform a range of fungal species [32][33][34]. We report here a simple and rapid method using electroporation to transform germinating Epichloë conidia isolated from infected plants. The GFP labelled E. typhina provides a valuable molecular tool for researchers studying the infection process in orchardgrass. The transformation protocol will also be valuable for gene function studies in Epichloë-plant interactions.

Transformation vector
The pCT74 GFP expression vector [16] was used as an intact circular plasmid or as a linearized plasmid for transformation experiments. The plasmid was digested overnight at 37°C with Xho I to linearize the plasmid. The digested DNA was precipitated with 1/10 volume of 3 M NaOAc, pH 5.2, and 2 volumes of EtOH. One μL of blue dextran (10 mg/mL) was added to aid in visualization of the pellet. The precipitated DNA was resuspended in sterile distilled water at a final concentration of 0.5 μg/μL.

Collection and preparation of conidia for transformation
Orchardgrass plants known to be infected with E. typhina were grown in the greenhouse under 16 hr of light at 21°C. To induce flowering, the plants were vernalized in a growth chamber at 8°C with 8 hr of light for 12 weeks, then at 15°C for 2 additional weeks under 18 hr of light, and then returned to the greenhouse. Conidia from newly formed unfertilized stromata ( Figure 1) were collected in about 10 mL of sterile water using a sterile artist paintbrush. The brush was dipped into sterile deionized water, gently brushed along the surface of each stroma to collect conidia in the brush, then shaken in the water to dislodge the spores. Approximately 10 6 conidia/mL were obtained when conidia from 5-10 stroma were collected in one tube. Isolated conidia were gently shaken in sterile deionized water for 4 hr at room temperature at which time >90% of the conidia had germinated (Additional File 1). Suspended germinated conidia were transferred into sterile microfuge tubes (1.5 mL/tube) and centrifuged at~16,000 × g for 45-60 s. Approximately 1.45 mL of the supernatant was discarded. The conidia were resuspended in cold sterile electroporation buffer (1 mM Hepes, 50 mM mannitol, pH 7.5) to a final volume of 1 mL.

Sensitivity of untransformed conidia to hygromycin B
The plasmid used for transformation contains the hygromycin B phosphotransferase gene which confers resistance to hygromycin and enables the selection of transformed colonies. To determine the concentration Figure 1 Epichloë typhina infection of orchardgrass. Conidia are collected from the whitish, felt-like stroma that develops on reproductive tillers as they elongate, just prior to seed head emergence. of hygromycin to be used for selection, conidia (~5000 conidia/plate) were placed on increasing concentrations of hygromycin B, from 0-400 mg/L, on Corn Meal Malt Agar (CMMA -19 g corn meal agar, 2 g yeast extract, 20 g malt extract and 5 g agar per liter). Since conidia were collected from plants grown in the greenhouse, 150 μg/mL of ampicillin or ticarcillin was added to the selection medium to inhibit bacterial growth.
Determination of optimal electrical field strength for conidia transformation Electroporation was performed using the Bio-Rad Gene Pulser II on 100 μL of suspended conidia (~5000 conidia) in a 0.2 cm cuvette. The percent survival of germinating conidia at various combinations of voltage settings between 1 and 2 kV, resistances between 400-800 ohms, and 25 μF capacitance was tested. Percent survival at each resistance level (400, 600, 800 ohms) at the three voltages was determined by calculating the number of colonies from each sample (based on colony counts obtained from plating out serial dilutions from each sample) and comparing that to the number of colonies obtained from non-electroporated germinating conidia.

Transformation of conidia
Approximately 10 5 -10 6 conidia in 100 μL of sterile electroporation buffer were combined with 0.1-2.0 μg of circular or linearized DNA in sterile 500 μL microfuge tubes and incubated on ice for 20 min. The DNA/conidial mixture was then transferred into a cold 0.2 cm electroporation cuvette and subjected to a specific electrical field using Bio-Rad Gene Pulser II. Immediately following the pulse, the conidia were transferred to a sterile 15 mL snap-cap tube containing 900 μL of cold sterile Regeneration Medium (RM -14.5 g mannitol, 0.4 g yeast extract and 1.5 g potato dextrose broth per 100 mL distilled water). The conidia were incubated on ice for 20-30 min and were then placed on a rotating shaker (150-180 rpm) at room temperature for 5 hours prior to plating on selective solid medium (200 μL/ plate). Since conidia were collected from plants grown in the greenhouse, 150 μg/mL of ampicillin or ticarcillin was added to the selection medium to inhibit bacterial growth. Hygromycin B-resistant colonies started to develop in 3-4 weeks. Colonies were examined with a LEICA Fluorescent Dissecting Microscope MZFLIII using a PLANAPO 1.0X Lens to confirm GFP expression in E. typhina transformants. In order to visualize the untransformed control fungus, low intensity white light was used in addition to the UV light.
Southern analysis to determine copy number DNA was isolated using a modified CTAB (hexadecyltrimethylammonium bromide) protocol [35] including treatment with RNase. The DNA was treated with Nucleon PhytoPure™resin, a component found in the Illustra Nucleon Phytopure™Genomic DNA extraction kit (GE Healthcare UK Limited, Buckinghamshire, UK) which covalently binds polysaccharides. Two μL of the resin was added per 100 μL of solution, mixed gently, and then centrifuged to pellet the resin. The DNA remained in the supernatant. Approximately 5 μg of DNA was digested with EcoRI, using the manufacturer's supplied buffer, and separated in a 1.2% agarose gel. DNA was transferred to a Hybond-N+ membrane (Amersham; GE Healthcare) by downward capillary transfer following standard protocols [36]. PCR was used to synthesize a digoxigenin (DIG; Roche, Cat. No. 11573152910) labelled probe for a portion of the GFP gene following the recommendations of the company (https://www.roche-applied-science.com). Primers were designed using Primer3 software [37]. Primers for probe synthesis were GFPF64 5'-GACGTAAACGGCCA-CAAGTTC and GFPR671 5'-GAACTCCAGCAGGAC-CATGTG producing an amplicon of~600 bp (between nucleotide 64 and 671 of the GFP open reading frame). Hybridization and detection were performed following the protocol of Engler-Bloom et al. [38] with modifications described by Krueger and Williams [39].

Sensitivity of untransformed conidia to hygromycin B
When conidia (~5000 conidia/plate) were placed on increasing concentrations of hygromycin B on CMMA, many colonies were present on 0 and 100 μg/mL hygromycin, but only a few colonies grew at 200 μg/mL, and no colonies grew at 300 and 400 μg/mL hygromycin ( Figure 2). Based on this analysis, hygromycin B concentrations of 200 and 300 μg/mL were used for selection of transformants.

Determination of optimal electrical field strength for conidia transformation
Most reports indicate that the best electroporation conditions to use for transformation are those that result in 40-60% survival of the cells. With increasing levels of resistance, the 40-60% survival rate occurred at lower voltages. Approximately 50% of the conidia survived using 1 kV at 800 Ω, 1.2 kV at 600 Ω, and between 1.2 and 1.5 kV at 400 Ω (Additional file 2).

Transformation of conidia
Hygromycin B-resistant colonies started to develop in 3-4 weeks. Transformed colonies grew to a much larger size than untransformed background colonies (Figure 3). Transformation efficiency was greatly affected by use of an intact versus linearized vector and the age of the stroma. The linearized plasmid resulted in a much higher frequency of transformation than the circular plasmid which yielded only two transformants. It was also important that conidia were collected from newly formed stroma, as conidia isolated from more mature stroma were not transformable. The greatest number of transformants was obtained at 400 Ω resistance and 1.25 kV at a capacitance of 25 μF. At resistances of 600 and 800 Ω, a voltage of 1.0 kV gave a higher number of transformants compared to what was obtained at 1.25 kV and 1.5 kV (Table 1).
Over 100 colonies were isolated and over 25 were chosen and stored long term. The percentage of hygromycinresistant colonies expressing GFP was 70-85% and did not follow a specific pattern in relation to voltage or resistance applied (Table 1). GFP expression levels were variable and different colony morphologies were observed (Figure 4). Over 80% of the transformed lines continued to express GFP after successive transfers and storage. Fifteen randomly selected lines were verified as E. typhina by PCR using species specific primers (ACT1-F 5'-CCCCGCTC GGCTC-3' and ACT1-R 5'-GCCCAGGCCGAAAAA TTA-3') to produce a 481 bp amplicon from the actin 1 locus which was sequenced [40].

Southern analysis to determine copy number
Southern analysis was performed to determine copy number for transformed fungal colonies. Approximately 70% of the lines tested contained single inserts and approximately 23% had two or three inserts ( Figure 5). These 12 lines were found to be stable and were still expressing GFP one year after being established.

Conclusions
An efficient method for the transformation of germinating conidia of E. typhina by electroporation was