Development of a greenhouse-based inoculation protocol for the fungus Colletotrichum cereale pathogenic to annual bluegrass (Poa annua)

Turfgrass hosts

Poa annua biotypes 98226, 99112 and 9712 were seeded (0.2 g per pot) in 10 cm pots filled with Fafard Canadian Grow Mix 2 (Agawam, MA). All biotypes were originally collected as seed from putting greens in the US. Biotypes were selected based on their susceptibility to anthracnose disease in turfgrass research plots at the Pennsylvania State University in State College, Pennsylvania (Dave Huff, pers. comm., 2011). Prior to inoculation, each biotype was tested for the presence of endophytic fungi. Colletotrichum cereale can survive asymptomatically in host tissue (Crouch et al., 2009b), thus the presence of endophytic C. cereale strains could confound real-time PCR results. In addition, each biotype was screened for existing mycorrhizal associations. For endophyte testing, 2 g of leaf tissue was randomly selected from 10 cm pots, trimmed to fit inside a 25 mL polypropylene tube (BD Falcon, Bedford, MA), then washed for 2 min in 70% EtOH, 2 min in 10% commercial NaClO, three rinses for 1 min in sterile dH₂O, and 10 s in 95% EtOH. Surface sterilized tissue was plated onto malt extract agar (MEA, Fisher Scientific, Pittsburgh, PA), and incubated at room temperature under continuous light for 3 weeks. For mycorrhizae testing, 0.5 g of P. annua roots were selected at random from established plants. Harvested roots were washed with dH₂O and stained for mycorrhizae following a modified procedure of Phillips & Haymanm (1970). Briefly, 1.5 cm root segments were heated in 10% KOH for 1 h at 90 °C. Cleared root segments were rinsed in tap water for 2 min, heated for 1 h in 20% HCl, followed by a final heating in 0.1% trypan blue (Roche Applied Science, Indianapolis, IN) for 30 min. Stained roots were mounted on microscope slides and visualized using an Olympus BX41 clinical 71 microscope (New York/New Jersey Scientific, Middlebush, NJ).

Fungal isolates

Two isolates of C. cereale were used in this study. Isolate TCGC5-63 (clade A) was collected from P. annua in 2002 from a golf course putting green in Temecula, CA, whereas isolate HF217CS (clade B) was collected from P. annua in 2012 from turfgrass research plots in North Brunswick, NJ. Fungal specimens are deposited in Centraalbureau voor Schimmelcultures (Utrecht, The Netherlands). Prior to inoculation, isolates were removed from −80 °C storage, where they were maintained as desiccated conidia on silica gel in 1 mL cryogenic tubes (Fisher Scientific, Pittsburgh, PA), and placed on potato dextrose agar (PDA, Fisher Scientific, Pittsburgh, PA) under continuous light at room temperature. After fungal growth covered the PDA plates (∼8–10 days), 10 mL of sterile dH₂O was poured onto the plate and fungal mycelia and spores were gently removed by scraping with a sterile glass rod. A sterile inoculation loop was used to streak the fungal suspension onto a fresh PDA plate. Streaked plates were not sealed with parafilm and were placed under direct light at room temperature to encourage sporulation (∼3–4 days).

Inoculation protocol

All P. annua biotypes were inoculated with each C. cereale isolate in separate treatments, with three replications per biotype. The inoculation protocol consisted of a 20 mL conidial suspension (10⁻⁶ conidia/mL) of C. cereale and 10% potato dextrose broth (PDB, Fisher Scientific, Pittsburgh, PA). The conidial suspension was sprayed onto the foliage of 8 wk post-emergent P. annua plants using a hand atomizer. Non-inoculated plants were sprayed with 20 mL 10% PDB as a control. Following inoculation, plants were placed in a custom-made plexiglass mist chamber inside a growth chamber for 24 h. The plexiglass chamber measured 91.5 cm × 63.5 cm × 76.2 cm and was constructed with 7.9 mm plexiglass. A humidifier (Herrmidifier, Phoenix, AZ) atomized water into water vapor, providing a source of moisture in the chamber. The humidifier ran for a duration of 1 h every 2 h. Following incubation, plants were removed from the plexiglass chamber and remained in the growth chamber under conditions of 12 h daylight (500 µE m⁻² s⁻¹), 80% RH, 30 °C day, and 26 °C night until symptom development (∼7 to 10 days). All plants were watered every other day at the base to avoid wetting inoculated foliage during the course of the experiments. The experiment was replicated twice under the same conditions described above and included negative controls. To test the effects of moisture and plant growth medium on the ability of C. cereale to infect P. annua, the experiment was also repeated under the same growth chamber conditions two additional times, with variations in moisture and plant growth substrate. In one experiment, inoculated plants were seeded in Grow Mix 2, but no humidifier was used during the inoculation period. In the other experiment, plants were seeded in sand and inoculated and moisture was supplied as previously described.

Confirmation of infection

Infection was confirmed in three ways: (1) inoculated plants were visually inspected for signs of the fungus (appressoria and acervuli), (2) C. cereale was re-isolated from inoculated tissue after symptom development, and (3) inoculated plants were tested for the presence of C. cereale using real-time PCR, clade-specific assays developed using the Apn2 marker (Beirn, Clarke & Crouch, 2014). This assay confirms to the Minimum Information for Publication of Quantitative Real-Time PCR Experiments guidelines; complete details describing the assay can be found in Beirn, Clarke & Crouch (2014). For pathogen isolations and real-time PCR, plant tissue was surfaced sterilized prior to plating or DNA analysis with 10% commercial NaClO for 2 min, 70% EtOH for 2 min, rinsed in sterile dH₂O, and allowed to dry in a sterile hood. Half of the dried tissue was plated onto PDA, while the remaining tissue was used for real-time PCR analysis.

DNA was extracted from the remaining tissue using the OmniPrep DNA Extraction Kit (G-BioSciences, St. Louis, MO) and a modified manufacturer’s protocol. Briefly, 0.5 g of plant tissue was placed in a 2 mL microcentrifuge tube with eight 2.5 mm glass beads (BioSpec Products, Barletsville, OK) and beat in a BioSpec bead-beater (Barletsville, OK) on the medium setting for six minutes. The manufacturer’s protocol was followed for the remaining steps. Real-time PCR reactions were performed as previously described (Beirn, Clarke & Crouch, 2014) using a Cepheid SmartCycler (Cepheid, Sunnyvale, CA) in 25 µl Cepheid tubes. Briefly, Cepheid’s Smartmix HM lyophilized PCR master mix was used for all reactions with a final probe concentration of 2 µM and primer concentration of 20 µM. Cycling conditions were as follows: initial denaturation at 95 °C for 120 s, followed by 45 cycles of 95 °C for 5 s, 60 °C anneal for 30 s, and 72 °C extension for 1 s. Samples were considered positive if the fluorescent threshold (30) was crossed prior to cycle 40 and negative if samples produced cycle threshold (C_T) values equal to zero.

Mycorrhizae and endophyte testing

No visual signs of endophytic fungi or mycorrhizae were found in any of the P. annua biotypes used in this study.

Inoculation results

Appressoria formation was observed on P. annua biotypes 99112 and 98226 ∼24 h after inoculation with both C. cereale isolates on plants grown in potting mix and sand (Fig. 1A). Appressoria were randomly distributed on leaf surfaces and were dark brown/black and rounded or lobed, measuring 9–10 µm by 6–9.5 µm. Disease symptoms were not observed until 6 to 9 days after appressoria formation on P. annua biotypes 99112 and 98226, and were similar for plants seeded in potting mix or sand. Although leaf lesions were never present, plants inoculated with C. cereale isolate TCGC5-63 or isolate HF217CS appeared chlorotic when compared to uninoculated controls (Fig. 1B). Acervuli with diagnostic black setae were only present on plants inoculated with C. cereale isolate TCGC5-63 (Fig. 1C). Numerous falcate, hyaline conidia were found in conjunction with acervuli. P. annua biotype 9712 never developed any leaf symptoms or signs and was negative for all tests for the presence of C. cereale. Negative controls did not develop any disease symptoms, and did not exhibit any signs of fungal colonization.

Both isolates of C. cereale were re-isolated from symptomatic plant tissues on PDA. Mycelium was white at first, then turning slightly gray with age (Fig. 1D). Numerous acervuli were formed on the agar surface, forming in the older parts of the colony first. Conidia appeared orange colored in mass and measurements conformed to those reported for C. cereale (Crouch, Clarke & Hillman, 2006). Cultures of C. cereale were recovered from all symptomatic plants, regardless of whether they were seeded in potting mix or sand. Control plants generated no isolates of C. cereale upon culturing.

No infection was present in inoculated plants where the humidifier was not used.

Real-time PCR confirmation

Leaf tissue samples were taken from all plants to either confirm infection or, in the case of the controls, to verify the absence of infection. Real-time PCR confirmed infection for both C. cereale isolates HF217CS and TCGC5-63 on P. annua biotypes 99112 and 98226, with average C_T values of 32.90 and 32.04, respectively. Control plants and inoculated biotype 9712 generated C_T values of 0.00. DNA sequences of the ITS region amplified from the isolated fungi matched C. cereale isolates in the GenBank database with 100% similarity (data not shown).