Multiple Colletotrichum species commonly exhibit focal effector accumulation in a biotrophic interface at the primary invasion sites in their host plants

ABSTRACT

Fungal plant pathogens deploy a suite of secreted proteins, called effectors, to facilitate successful infection. Several fungal pathogens have been reported to secrete and accumulate their effector proteins in the host–pathogen interfacial spaces. Previously, we reported that the strain 104-T of the cucurbit anthracnose pathogen Colletotrichum orbiculare secretes and accumulates mCherry-tagged effectors along with the formation of ring-shaped fluorescence signals beneath the appressoria. However, it was unclear whether these effector accumulation patterns occur in other C. orbiculare isolates and other species belonging to the Colletotrichum genus. Here, we investigated the effector localization during host infection of C. orbiculare MAFF306589, C. trifolii MAFF305078, which infects alfalfa, and C. higginsianum MAFF305635, which infects Brassicaceae plants. We generated effector–reporter lines of each species, which constitutively expressed mCherry-tagged CoDN3 effector (CoDN3:mCherry). Immunoblotting analysis of the liquid culture fluids of the generated lines detected CoDN3:mCherry, which confirmed secretion of CoDN3:mCherry by fungal cells. Via inoculation assays in the corresponding host plants, we detected ring-shaped CoDN3:mCherry fluorescence around the appressorial invasion sites in all tested reporter lines. These results suggest that pathogens in the Colletotrichum genus have evolutionarily conserved the trait of effector secretion in the infection stage irrespective of differences in their hosts.

Although most higher plants have elaborate and robust defense mechanisms against attempted infection by numerous phytopathogenic fungi, via adaptation to plants, fungi can suppress the immune responses and manipulate the environmental status of the host plants by deploying a suite of secreted proteins called effectors.^1–3 Although many fungal effectors are known to interact with their target plant proteins located in the cytoplasm, it is unknown how these fungal effectors enter the host plant cells.²

Several fungal pathogens have been reported to secrete and accumulate their effector proteins into specific host–pathogen interfacial spaces.⁴ Magnaporthe oryzae is the causal pathogen of rice blast disease. The pathogen invades aboveground tissues of Gramineous plants using melanized appressoria, then develops primary invasive hyphae and establishes temporal biotrophic interaction with the host plant.⁵ M. oryzae has been reported to accumulate effector proteins in the host–pathogen interfacial space to form dot-like structures, named the biotrophic interfacial complex (BIC).⁶ The BIC is initially observed at the tip of the primary invasive hyphae. When an invasive hypha starts to differentiate into a bulbous shape, the BIC remains as a dot signal beside the first bulbous invasive hypha.^6,7

Many Colletotrichum species form melanized appressoria before developing primary invasive hyphae in host cells, and further expansion of the invasive hyphae eventually causes anthracnose disease in various plants.⁸ We previously investigated the localization of effectors secreted by C. orbiculare (wild-type (WT) strain 104-T), the causal agent of cucurbitaceous anthracnose disease, during host infection, by analyzing C. orbiculare transgenic lines expressing effectors tagged with mCherry. We found a ring-like fluorescence signal derived from the mCherry-tagged effectors beneath the appressoria.⁹ Further analyses revealed that C. orbiculare accumulates effectors in a host–pathogen biotrophic interfacial space around the neck of the primary invasive hyphae beneath the appressoria in its host plants, cucumber and melon.^9,10 The observed effector localization in C. orbiculare is distinct from the BIC detected in M. oryzae, which suggests potential diversity in host–pathogen interfacial structures into which fungal effectors are secreted. We also revealed that the disruption of SEC4 encoding an exocytosis-related component in C. orbiculare reduced virulence and impaired effector delivery to the ring-like signal interface, implying biological significance of the effector delivery to the interface for the host infection.⁹

Although we previously reported effector protein accumulation of C. orbiculare strain 104-T (MAFF240422), it remains unclear whether other C. orbiculare isolates and other species belonging to the Colletotrichum genus exhibit similar accumulation patterns of effectors. To examine this aspect, we generated reporter strains of several Colletotrichum species that express the effector CoDN3 tagged with mCherry (hereafter, CoDN3:mCherry) under control of the constitutive promoter (TEF promoter) of Aureobasidium pullulans.^9,11

CoDN3 is an effector of C. orbiculare 104-T that suppresses the cell death caused by another C. orbiculare effector NIS1 in Nicotiana benthamiana.¹² CoDN3 also suppresses cell death in N. benthamiana induced by another C. orbiculare effector NLP1.^13,14 CoDN3 interacts with plant cytoplasmic calmodulin protein, and the calmodulin binding domain of CoDN3 is necessary for the suppression of cell death in N. benthamiana.¹⁴ These findings suggest that CoDN3 is likely to be translocated into the host cells and to act there.

We newly generated effector reporter lines of C. orbiculare MAFF306589, C. trifolii MAFF305078, which infects alfalfa, and C. higginsianum MAFF305635, which infects Brassicaceae plants such as Brassica rapa var. perviridis (hereafter, Komatsuna (Japanese name)) and Arabidopsis thaliana. To generate the CoDN3:mCherry line of C. trifolii, we introduced the plasmid pBATTEFP_DN3:mCherry, which carries both the TEF promoter:CoDN3:mCherry:SCD1 terminator and a bialaphos resistance gene in C. trifolii.⁹ For C. orbiculare and C. higginsianum, we constructed the plasmid pCB1004TEFP_DN3:mCherry to use hygromycin for the transformant selection. Briefly, to generate this plasmid, the NotI–KpnI fragment of the TEF promoter:CoDN3:mCherry:SCD1 terminator was cut out from pBATTEFP_DN3:mCherry and inserted into the NotI–KpnI site of pCB1004, which carries a hygromycin resistance gene.¹⁵

We first investigated whether CoDN3:mCherry proteins are secreted from the generated transgenic lines of C. orbiculare, C. trifolii, and C. higginsianum. For this purpose, we also generated transgenic lines of each Colletotrichum species that express cytosolic RFP (red fluorescent protein) under the control of the TEF promoter. The WT, CoDN3:mCherry lines, and RFP lines of each species were cultured in liquid medium (0.3% yeast extract, 0.3% casamino acids, and 0.5% sucrose), and the vegetative hyphae and culture medium were separated by filtering through double gauze. Then, the culture medium was filtered through a Millex-GP syringe filter (0.22 μm, Millipore), the proteins were concentrated with acetone precipitation, and the protein pellet was suspended in extraction buffer (50 mM Tris-HCl pH 7.4, 150 mM NaCl, 5% glycerol, 0.5% Triton X-100) containing 1% protease inhibitor cocktail (P8215, Sigma). Vegetative hyphae were flash-frozen in liquid nitrogen, homogenized using zirconia balls (3 mm, Nikkato), and suspended in extraction buffer. The sample solutions of vegetative hyphae were centrifuged at 21,000 g at 4°C to remove debris. The presence of CoDN3:mCherry or RFP in the samples was analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and an immunoblotting assay.

CoDN3:mCherry proteins were detected in the culture filtrates derived from the CoDN3:mCherry lines of all Colletotrichum species. By contrast, CoDN3:mCherry protein was not detected in vegetative hyphae of each line (Figure 1a). Cytosolic RFP, which does not have a secretion signal, was detected only in the vegetative hyphae (Figure 1a). These results indicate that CoDN3:mCherry protein is secreted by the CoDN3:mCherry lines of C. orbiculare, C. trifolii, and C. higginsianum.

Figure 1.

mCherry-tagged CoDN3 effectors secreted by multiple Colletotrichum pathogens accumulated around the hyphal neck region of primary invasive hyphae together with the formation of a ring-shaped fluorescence signal

(a) Detection of CoDN3:mCherry in culture fluid of each Colletotrichum reporter line. C. orbiculare, C. trifolii, and C. higginsianum lines expressing cytosolic RFP or CoDN3:mCherry were incubated in liquid medium for 2 days. The parental wild-type (WT) isolates of each species were also incubated as a control. Proteins were extracted from the culture filtrates (CF) and vegetative hyphae (VH) separately. Equal amount of proteins (about four µg) from CF and VH were loaded in each lane for SDS-PAGE followed by immunoblotting to detect cytosolic RFP or CoDN3:mCherry protein. RFP and mCherry were detected using Anti-RFP Monoclonal Antibody (Invitrogen; clone RF5R) and antibody to mouse IgG conjugated with peroxidase (SeraCare Life Sciences Inc.). Chemiluminescence was detected using Amersham ECL Prime Western Blotting Detection Reagent (GE Healthcare). The upper panels of each sample show the immunoblotting data and the lower panels show the corresponding regions stained with PageBlue protein staining solution (Thermo Scientific). (b) Accumulation patterns of CoDN3:mCherry protein secreted by multiple Colletotrichum species. Each reporter line expressing CoDN3:mCherry was inoculated onto the cotyledons of the corresponding host plants, and the inoculated samples were observed when the pathogens developed primary biotrophic hyphae. Images were taken at 4 days postinoculation (dpi) for C. orbiculare (104-T and MAFF306589) and C. trifolii (MAFF305078), and at 38 h postinoculation (hpi) for C. higginsianum (MAFF305635). The ring-shaped signals are indicated by arrows. DIC, differential interference contrast; PH, primary invasive hypha; MA, melanized appressorium. Bars = 10 μm. (c) The ratio of ring signal detection for appressoria that formed primary invasive hyphae. Each strain was inoculated onto the cotyledons of the corresponding host plants, and the inoculated samples were observed when the pathogens developed primary invasive hyphae (at 4 dpi for C. orbiculare and C. trifolii and at 38 hpi for C. higginsianum). At least 60 appressoria that formed primary invasive hyphae were observed under the laser scanning confocal microscope for each experiment. The mean and SD were calculated from the results of three independent experiments. WT indicates the wild-type isolates without the CoDN3:mCherry gene.

Next, we inoculated a conidial suspension (10⁶ conidia/mL) of CoDN3:mCherry-expressing lines of C. orbiculare, C. trifolii, and C. higginsianum onto the abaxial sides of cotyledons of cucumber (Cucumis sativus cv Suyo), alfalfa (Medicago sativa), and Komatsuna (B. rapa var. perviridis cv Kaoru), respectively. We then peeled off the lower epidermal layer of the cotyledons with tweezers and observed them under a laser scanning confocal microscope (Olympus FluoView FV1200 with Olympus 60 × U Plan S-Apo 1.35 numerical aperture objective).

We detected ring-shaped fluorescence signals of the CoDN3:mCherry around the appressorial invasion sites of the CoDN3:mCherry line of C. orbiculare MAFF306589 as shown for the C. orbiculare strain 104-T. By contrast, the nontransgenic WT MAFF306589 did not exhibit ring-shaped fluorescence signals (Figure 1b,c). These results suggest that the CoDN3:mCherry accumulation pattern with the ring-shaped signal is not specific to 104-T but is likely to be common in C. orbiculare.

We also found the ring-shaped CoDN3:mCherry signal in the cases of inoculation of the C. trifolii reporter lines in alfalfa and the C. higginsianum reporter lines in Komatsuna (Figure 1b,c). This finding suggests that the CoDN3:mCherry accumulation pattern with the ring-shaped fluorescence signal is a common phenomenon in the Colletotrichum genus. Although the detection of such ring signals was not reported for C. higginsianum–A. thaliana interactions in the previous study,¹⁶ the use of a constitutive TEF promoter and a different host plant Komatsuna, instead of A. thaliana, in this study may have enabled us to detect the ring-shaped fluorescence in C. higginsianum.

In this study, we generated effector reporter lines expressing CoDN3:mCherry of multiple species belonging to the Colletotrichum genus. The immunoblotting analysis of CoDN3:mCherry expressed in each species confirmed that CoDN3:mCherry was secreted by fungal cells. We next found that all tested reporter lines focally accumulated the CoDN3:mCherry effector around the neck region of the primary hyphae, which was detected as ring-shaped fluorescence signals, during their biotrophic interaction with their host plants.

In the Colletotrichum genus, the orbiculare clade to which C. orbiculare and C. trifolii, but not C. higginsianum, belong is thought to be diverged from all other clades at the bottom of the entire phylogenetic tree of this genus.¹⁷ Thus, it is possible that ring-shaped effector accumulation via the secretion center in the hyphal neck region is a conserved mechanism among the Colletotrichum genus and derived from their common ancestor.

Whereas M. oryzae and many Colletotrichum fungi, including C. orbiculare, utilize melanized appressoria to penetrate the cuticle and cell wall in the aboveground parts of their hosts, some soilborne plant pathogens and arbuscular mycorrhizal fungi use another adherent structure, known as hyphopodia, to invade their host plant root tissues.^5,8,18–22 Hyphopodia are structures formed on epiphytic hyphae from which the invasive hypha develops.²³ Interestingly, a recent study reported that effector candidates, labeled with GFP, of a root-infecting pathogen Verticillium dahliae accumulate at the neck of invasive hyphae beneath hyphopodia, which exhibit ring-shaped fluorescence signals that are reminiscent of the CoDN3:mCherry signal.²⁴ Actin and an exocyst-related component Sec8 have been shown to localize at the neck of invasive hyphae beneath the hyphopodia of V. dahliae.²⁴ This observation is also consistent with the findings that actin and an exocyst-related component SEC4 localized at the primary hyphal neck under the appressoria of C. orbiculare.⁹ Although the appressoria of Colletotrichum species and the hyphopodia of V. dahliae are morphologically different, these findings suggest that the mechanisms underlying effector secretion in the primary invasion sites of both pathogens might have substantially overlapping aspects.

The two genera Colletotrichum and Verticillium belong to the same order Glomerellales.²⁵ On the other hand, M. oryzae belongs to Magnaporthales (not Glomerellales) and accumulates its cytoplasmic effector proteins within BICs, as described above.⁶ M. oryzae does not appear to accumulate effectors as ring-shaped signals in the primary hyphal neck region, as seen in Colletotrichum species and V. dahliae. We currently assume that M. oryzae has a different strategy for effector secretion toward the host–pathogen interface compared with Colletotrichum fungi and V. dahliae. However, Colletotrichum species and M. oryzae commonly use melanized appressoria to infect the aboveground parts of hosts, whereas V. dahliae uses hyphopodia to infect the roots of the host. Thus, it is plausible that in planta effector secretion machineries have evolved in plant pathogenic fungi independent of the innovation of fungal cell structures that are used to penetrate their host plant cells.

Interestingly, M. oryzae and C. graminicola, the causal pathogen of anthracnose diseases of maize, have been reported to form hyphopodia when inoculated to host plant roots, although they form appressoria pigmented with melanin in aboveground plant tissue.^18–20 It is important to investigate further the effector accumulation patterns in the host–pathogen interface, where hyphopodia of these pathogens develop invasive hyphae. Sordariomycetes includes many plant parasitic genera, such as Fusarium and Cryphonectria, in addition to Colletotrichum, Magnaporthe, and Verticillium. It is also necessary to examine the extent to which the ring-shaped effector accumulation pattern is conserved among plant pathogens within Sordariomycetes. Finally, we emphasize the importance of investigating the link between the interfacial accumulation of effectors and the effector-mediated infection strategy, for example, whether the interfacial accumulation of effectors is involved in effector translocation into host plant cells.

Funding Statement

This work was supported by the Asahi Glass Foundation and Grants-in-Aid for Scientific Research [21H04725 18H02204 18H04780].

Disclosure Statement

The author has declared that no competing interests exist.