(C) PLOS One [1]. This unaltered content originally appeared in journals.plosone.org.
Licensed under Creative Commons Attribution (CC BY) license.
url:https://journals.plos.org/plosone/s/licenses-and-copyright

------------



The small-secreted cysteine-rich protein CyrA is a virulence factor participating in the attack of Caenorhabditis elegans by Duddingtonia flagrans

['Nicole Wernet', 'Karlsruhe Institute Of Technology', 'Kit', 'South Campus', 'Institute For Applied Biosciences', 'Dept. Of Microbiology', 'Karlsruhe', 'Valentin Wernet', 'Reinhard Fischer']

Date: 2021-11

Nematode-trapping fungi (NTF) are a diverse and intriguing group of fungi that live saprotrophically but can switch to a predatory lifestyle when starving and in the presence of nematodes. NTF like Arthrobotrys oligospora or Duddingtonia flagrans produce adhesive trapping networks to catch and immobilize nematodes. After penetration of the cuticle, hyphae grow and develop inside the worm and secrete large amounts of hydrolytic enzymes for digestion. In many microbial pathogenic interactions small-secreted proteins (SSPs) are used to manipulate the host. The genome of D. flagrans encodes more than 100 of such putative SSPs one of which is the cysteine-rich protein CyrA. We have chosen this gene for further analysis because it is only found in NTF and appeared to be upregulated during the interaction. We show that the cyrA gene was transcriptionally induced in trap cells, and the protein accumulated at the inner rim of the hyphal ring before Caenorhabditis elegans capture. After worm penetration, the protein appeared at the fungal infection bulb, where it is likely to be secreted with the help of the exocyst complex. A cyrA-deletion strain was less virulent, and the time from worm capture to paralysis was extended. Heterologous expression of CyrA in C. elegans reduced its lifespan. CyrA accumulated in C. elegans in coelomocytes where the protein possibly is inactivated. This is the first example that SSPs may be important in predatory microbial interactions.

Pathogenic microorganisms are living at the expense of their host organisms and immediate killing of the host may be disadvantageous. Therefore, many bacterial or fungal pathogens developed an arsenal of small-secreted proteins during the colonization to modulate their host for instance to suppress its defense reactions. This allows a biotrophic phase, at least for some time. Some higher eukaryotic “pathogens” also live at the expense of their hosts but are often predators. In this case, quick killing is followed by digestion. In the case of predatory fungi, one could expect a similar situation, quick killing followed by digestion. However, the genome of such fungi encodes many putative small-secreted proteins, and we show here that one of them indeed appears to be secreted into the host and contributes to virulence. The protein is produced in the trapping devices of the fungus and especially in the penetration peg right after entering the nematode. Heterologous expression in Caenorhabditis elegans reduced the lifespan of the worms. This protein is to our knowledge the first characterized small-secreted protein in the predatory relationship between fungi and nematodes.

Funding: The work was funded by the Karlsruhe Institute of Technology (KIT). N.W. received her salary from KIT. V.W. was funded by the Deutsche Bundesstiftung Umwelt (DBU). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

In this paper, we analyzed a D. flagrans SSP with a cysteine-rich region and named it CyrA (cysteine-rich protein A). The gene is highly upregulated in traps and the protein is secreted into C. elegans at a bulbous structure of the penetration peg. CyrA is important for fungal virulence and causes a reduction of the lifespan when expressed in C. elegans. This is the first example for the characterization of a SSP from a nematode-trapping fungus and paves the way for further studies to unravel the molecular interplay between fungi and nematodes.

Although the necessity of SSPs in predatory interactions is not obvious, many genes of NTF encode such small proteins, and many of the genes are transcriptionally upregulated during the nematode attack and are found in the pathogen-host interaction (PHI base) database e.g. in Arthrobotrys oligospora and Monacrosporium haptotylum [ 15 , 27 , 28 ]. This database contains microbial pathogenicity, virulence and effector genes which were experimentally characterized [ 29 ]. One good example is the gEgh16/gEgh16H gene family in M. haptotylum, for which a role in pathogenesis was described in Blumeria graminis and Magnaporthe grisea [ 30 – 32 ].

Secretion of the SSPs can follow the conventional secretion pathway or an alternative route, as it was shown in the rice-blast fungus Magnaporthe oryzae. Here, the cytoplasmic effector Pwl2 is secreted in a Golgi-independent manner, while the apoplastic effector Bas4 uses the conventional pathway [ 23 ]. Secretion of the cytoplasmic effector depends on the exocyst complex which is therefore essential for plant infection [ 23 , 24 ]. The complex is important for the correct spatiotemporal regulation of exocytosis and guides the docking and tethering of vesicles to the target membrane [ 25 ]. Another example of different secretion pathways for SSPs are Phytophthora sojae and Verticillium dahlia effectors that suppress salicylate-mediated innate immunity in planta [ 26 ].

Most NTF belong to the monophyletic group in the order Orbiliales (Ascomycota), and the co-evolution with their nematode prey dates back 400 million years. Their occurrence in many taxonomic groups is an indicator that they evolved independently serval times during evolution [ 16 ]. They are optimal model systems to study predator-prey interactions between eukaryotes and the underlying interspecies communication [ 17 , 18 ]. It is known that NTF induce trap formation in the presence of nematodes after sensing nematode-specific ascarosides [ 18 , 19 ]. On the other hand, the fungi lure the nematodes to the traps by secreting volatiles that are attractive to the worms [ 17 , 19 ]. After nematodes are trapped, hyphae penetrate the cuticle, secrete many hydrolytic enzymes and digest the worms [ 20 ]. An open question is if proteins with no obvious enzymatic activity are also required for the interaction or if the fungus quickly kills the worm and digests the dead material afterwards. In other pathogenic associations such proteins exist and since they are mostly rather small proteins and are secreted into the host, they were named small-secreted proteins (SSPs). They are also called virulence factors or effectors depending on the organismic interaction type and if there is a longer biotrophic phase, where effector proteins e.g. suppress the defense reactions of the host [ 21 ]. They are also crucial in microbial symbiotic interactions [ 22 ].

Nematodes cause agricultural losses of 80 billion US dollars worldwide [ 4 ]. This is a devastating number especially today, where we must face the needs of a growing population. The use of nematicides is limited because of their negative effect on the environment and health concerns [ 5 , 6 ]. Nematodes are important plant pathogens but can also be a problem in animals. Livestock, like sheep or cows infected with pathogenic nematodes have a reduced productivity and lifespan [ 7 , 8 ]. Nematode eggs are spread onto the pasture with the feces where they go through their life cycle and then are re-ingested during grazing leading to re-infection. This vicious cycle can be broken by feeding pellets containing NTF spores to the animals [ 2 , 9 ]. Duddingtonia flagrans is well suited for such an application because it produces robust and resistant chlamydospores besides adhesive, three-dimensional trapping networks [ 10 – 12 ]. It has already been established successfully as a biocontrol agent in horses, cattle and other animals [ 13 , 14 ]. The chlamydospores survive the passage through the gastrointestinal tract and are then able to germinate in the feces where they reduce the number of nematodes. The genome of this fungus has been sequenced and annotated recently, and molecular and cell biological methods were developed [ 15 ]. This opens new avenues for basic research but may also help to optimize the fungus as biocontrol agent.

Nematode-trapping fungi (NTF) are carnivorous microorganisms that can trap and digest nematodes with sophisticated trapping structures. Most fungi in this versatile group can live saprotrophically and are able to switch to a predatory lifestyle while others are obligate pathogens [ 1 ]. A broad variety of diverse trapping structures is found in different species, such as adhesive networks, constricting- and non-constricting rings and adhesive knobs. NTF play an important role in the regulation of the nematode population in soil in almost all known ecosystems [ 1 ]. Further, as natural antagonists of nematodes they have powerful potential to be used as biocontrol agent against animal and plant pathogenic nematodes [ 2 , 3 ].

Results

CyrA is a small-secreted protein with a cysteine-rich region The analysis of the D. flagrans secretome revealed a large arsenal of SSPs that are potentially involved in the virulence against nematodes [15]. Since there are currently no detailed reports about the function of such proteins in the interaction between fungi and nematodes, we analyzed if candidate proteins were specific to nematophagous fungi, shown to be transcriptionally induced in RNAseq analyses of other nematophagous fungi and predicted to be involved in the attack by EffectorP [33]. One protein meeting these criteria is characterized in this analysis. The CyrA protein is composed of 155 amino acids and harbors a putative secretion signal at the N-terminus (1–21) (Fig 1A). It contains no conserved domains, and orthologous proteins can be found only in the two nematode- trapping fungi A. oligospora and M. haptotylum. To test the functionality of the signal peptide, we fused the protein to Aspergillus nidulans Laccase C lacking its own signal peptide and expressed it in D. flagrans. If secreted, the laccase catalyzes the oxidation of the substrate ABTS in the medium to the more stable state of the cation radical [34]. This is indicated by a blue-green color. After 48 h incubation of the mutants on low nutrient agar (LNA) containing 1 mM ABTS at 28°C the colonies were surrounded by a blue-green color indicating the activity and secretion of the laccase into the medium and therefore functionality of the predicted signal peptide (Fig 1B). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 1. CyrA is a secreted cysteine-rich protein induced during trap formation. (A) Scheme of CyrA. The 155 amino acid long protein contains a 21 amino acid long signal peptide and a cysteine-rich region with eight cysteine residues. (B) The CyrA-LccC expressing D. flagrans strain sNH08 and the WT were grown on LNA with 1 mM ABTS for 48 hours. (C) Quantitative real time PCR analysis of cyrA expression in D. flagrans hyphae grown on LNA and in hyphae and traps co-cultivated for 24 h with C. elegans. The three biological replicates are displayed individually, and the error bar indicates the standard deviation of three technical replicates, P-value = 0.0342. The expression was normalized to actin. (D) Spatial analysis of cyrA expression using a transcriptional reporter assay. The h2b-mCherry reporter was expressed under the constitutive h2b (sNH14) or the cyrA promoter (sNH21), respectively. Pictures of traps and captured nematodes were taken after 24 h of co-incubation. Scale bar = 100 μm. Pictures of uninduced mycelium were taken after 24 h at 28°C. Scale bar = 10 μm. (E) The mean fluorescence of single nuclei of the reporter strains in traps (ind.) and vegetative mycelium (unind.) was measured in different pictures taken with the same settings using ImageJ (arbitrary units). cyrA(p) uninduced n = 39, cyrA(p) induced n = 48, h2b(p) uninduced n = 30, h2b(p) induced n = 30. P-value h2b(p) induced vs. cyrA(p) induced < 0.0001. https://doi.org/10.1371/journal.ppat.1010028.g001

cyrA is up-regulated during the interaction with C. elegans If cyrA plays a role during the infection process, the expression of the gene could be specifically induced. To determine any differential expression of cyrA we performed quantitative real-time PCR (qRT PCR) of RNA extracted from uninduced mycelium and mycelium after co-incubation with C. elegans (Fig 1C). Induction of the expression of the cyrA gene was observed after 24 h. The expression data varied between biological replicates with fold changes of 5.7 (± 0.7, n = 3), 7.9 (± 1.4) and 18.9 (± 2.6). These differences between the replicates reflect differences in trap numbers and/or trapped worms. To obtain further evidence for the upregulation of cyrA, a reporter assay was used to visualize the activity of the cyrA promoter microscopically in hyphae (Fig 1D). A fusion protein comprised of histone H2B and mCherry was expressed in D. flagrans under the control of the cyrA promoter. The H2B protein was used to target mCherry into nuclei to make it easier to distinguish any fluorescent signal from background fluorescence. Traps were induced by co-incubating the mycelium with C. elegans on LNA-microscopy slides, and a control was incubated without worms. Strong fluorescence of the nuclei was observed in traps as well as in trophic hyphae growing inside of captured C. elegans. Fluorescence decreased in mycelium further away from traps and was very weak in areas with no traps and in non-induced mycelium. The mean fluorescence of single nuclei was quantified using ImageJ (Fig 1E). The same construct but under the control of the h2b promoter was used as a control for a constitutively expressed gene. In the corresponding mycelium all nuclei showed similar fluorescent signal intensities. This shows that the expression of cyrA is highly upregulated during trap formation and colonization of the worms.

CyrA localizes in trap specific vesicles To investigate the localization of CyrA during the infection, the protein was fused to GFP. The fusion protein was fully functional, because it rescued the virulence phenotype of a cyrA-deletion strain (see below). Over-expression of the CyrA-mCherry fusion protein with the constitutive A. nidulans oliC as well as under the native promoter revealed the localization of the protein in moderately dynamic speckles in the traps, mainly at the inner rim of the traps (Fig 2A). The fusion protein was mainly visible when nematodes were already captured, just some empty traps showed a strong CyrA-mCherry signal. To investigate if the localization at the inner rim of the trap is a general occurrence in traps and whether these foci correspond to known vesicle populations or organelles, endosomes (RabA), exosomes (BroA), clathrin-coated vesicles (ClaH), peroxisomes (GFP-SKL), endoplasmic reticulum (GFP-KDEL) and nuclei (H2B-GFP) were visualized in the traps (Fig 2B). All vesicle populations and organelles were evenly distributed as expected and the ER showed no unconventional arrangement. Peroxisomes were very abundant in the trap cells. None of these organelles showed the same localization pattern at the inner side of the trap observed in the CyrA-mCherry strain, as shown for exosomes and endosomes (Fig 2C). These results indicate that the CyrA foci are trap-specific structures and suggest a trap-specific secretion mechanism that differs from conventional secretion at the tip. In non-induced mycelium the C-terminal fusion protein was either not visible or localized in vacuoles, suggesting degradation everywhere but in the traps (Fig 2D). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 2. Localization of CyrA and different vesicles and organelles in traps of D. flagrans. (A) The CyrA-mCherry expressing strain sNH25 was co-incubated with C. elegans on LNA slides for 24 h at 28°C to induce traps. The cell wall was stained with calcofluor white (CFW). (B) Visualization of vesicles and organelles through GFP fusion proteins. Clathrin coated vesicles (ClaH-GFP, strain sNH66), exosomes (BroA-GFP, sNH52), endosomes (RabA-GFP, sNH29), peroxisomes (GFP-SKL, sNH16), the ER (GFP-KDEL, sNH22) and nuclei (H2B-mCherry, sNH14). Scale bars = 10 μm. (C) BroA-GFP and RabA-GFP, respectively, were expressed in the CyrA-mCherry expressing strain. Scale bars = 10 μm. (D) cyrA-mCherry was expressed under the constitutive oliC-promoter (sNH25) but the localization of the fusion protein in speckles was only apparent in traps (arrow). In the surrounding vegetative mycelium, CyrA-mCherry localized in vacuolar structures. (E) CyrA was expressed without its signal peptide sequence fused to GFP (strain sNH41). (F) The 21 amino acid long signal peptide of CyrA was fused to GFP (strain sNH42) and the construct was expressed under the constitutive oliC promoter. https://doi.org/10.1371/journal.ppat.1010028.g002 Next, we constructed a strain expressing CyrA without its signal peptide fused to GFP (Fig 2E). The fusion protein without SP localized in the cytoplasm of vegetative mycelium and traps, indicating a loss of trap identity. Because the SP appeared to carry the information for the correct protein localization, we fused only the SP (aa 1–21) to GFP. In this case, the SP was sufficient to reconstitute the initial localization at the inner side of the trap. In vegetative mycelium, diffuse localization in the cytoplasm was observed, but no vesicles were present without traps (Fig 2F). This observation was interesting because one would expect GFP to localize in the ER after the signal peptide is cleaved off after entering the organelle. This indicated that the SP is sufficient to guide proteins to specific vesicles in the traps and that the secretory pathway in traps is distinct from the conventional one in vegetative hyphae.

The exocyst complex is essential for CyrA localization in the infection bulb To further investigate the mechanism of virulence factor secretion in D. flagrans we deleted the exoA gene, coding for an orthologue of the exocyst complex component Exo70. We chose Exo70 for our analysis because it is already known to be involved in the secretion of cytoplasmic effectors in M. oryzae [23]. The exoA gene was deleted in a strain expressing CyrA-mCherry (Fig 5A). The mutant was confirmed via Southern-blot analysis and showed a growth phenotype compared to the wild type (Fig 5B and 5C). For the complementation, the mutant strain was transformed with a plasmid encoding the full-length gene under control of the native promoter. A virulence assay based on the long-term observation of single captured nematodes was performed, determining the time point of paralysis after the capturing event (Fig 5D). Spores of the respective strains were inoculated on LNA microscopy slides together with the C. elegans N2 strain and incubated for 24 hours at 28°C to induce traps. The worms were then washed off and C. elegans BAN126 L4 larvae were added to the mycelium with already formed traps. Trapping networks were observed using a spinning disk confocal microscope and pictures were taken every 2 minutes for 12 hours. The time a worm got stuck in a trap was determined as starting point. The time when all movement ceased, was defined as the paralysis point. While the D. flagrans wildtype took 103 min [SD ± 46] to paralyze the L4-larvae and the complementation strain 108 min [SD ± 46], the ΔexoA-deletion strain paralyzed the worms slower with 166 min [SD ± 75]. This result demonstrates that the exocyst complex plays an important role in D. flagrans virulence. Microscopic analysis of CyrA-mCherry in the ΔexoA-deletion strain showed, that CyrA still localized in dynamic spots at the inner side of the traps but did not accumulate at the infection bulb immediately after penetration (Fig 5E). This is also in accordance with the prolonged time to paralysis of the cyrA-deletion strain, which took 184 min [SD ± 92] and thus shows a similar impairment of virulence as the exoA mutant. Some CyrA-mCherry accumulations at the infection bulb were visible in old, already fully colonized nematodes over 5 hours after the penetration. While in the wildtype control and the complemented strain the fusion protein showed its characteristic localization after 60 min (S3 Movie), this was not the case in the mutant strain even though penetration was clearly visible in the DIC channel. Magnification of the interaction zone showed that CyrA-mCherry accumulated at the infection site after 10 minutes but was not able to enter the nematode after the infection bulb was established, instead the protein seemed to remain at the outer edges of the infection site (Fig 5F and S4 Movie). This observation indicates that the exocyst complex is necessary for the secretion of virulence factors at the infection bulb. It could play an important role in tethering the proteins to the specific membrane. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 5. The exocyst component exoA is necessary for CyrA accumulation at the infection bulb. (A) Scheme of the deletion strategy for exoA. The gene was deleted via homologous recombination by flanking the hygromycin resistance cassette (hph) with 1 kb upstream (LB) and downstream (RB) flanks of exoA. The LB was used as a probe. (B) Southern-blot of genomic DNA of the ΔexoA-deletion mutant (strain sNH60) and WT using the probe indicated in (A). (C) D. flagrans wildtype and the ΔexoA-deletion strain were grown for 7 days at 28°C on PDA. (D) Virulence assay with the D. flagrans wildtype (WT), the ΔexoA-deletion mutant, the complementation strain (Re) and the ΔcyrA-deletion mutant. The strains were co-incubated with a mixed C. elgans N2 population on thin LNA-slides for 24 h at 28°C. After trap formation the N2 worms were washed off the slides and a synchronized population of Ban126 L4-larvae was added to the traps just before the microscopical observation. Pictures of single traps were taken every 5 minutes for 20 hours. The time from the capturing event to full paralysis of the nematodes was taken. The statistical significance was calculated using a student’s t-test (*** = p-value< 0.0001 n [WT] = 35, n [ΔexoA] = 34, n [Re] = 17, n [ΔcyrA] = 27). The error bar indicates the standard deviation. (E) Visualization of CyrA-mCherry in wild type (sNH65), the ΔexoA-mutant strain (sNH60) and the complementation strain (ΔexoA+exoA). The red box indicates the penetration area and the arrow points to the penetration site. (S2 and S3 Movies) (F) Enlargement of the interaction zone of the cyrA-mCherry expressing ΔexoA-mutant strain in (E). Initially, CyrA accumulates at the infection site after 10 minutes (star). The infection bulb is established after 30 minutes but CyrA does not accumulate in the bulb, instead some signals are visible at the outer rim of the entrance site. After 60 min trophic hyphae have colonized the nematode but the fusion protein is not visible. Scale bars = 10 μm. https://doi.org/10.1371/journal.ppat.1010028.g005

Deletion of cyrA affects virulence To gain further insights into the molecular function of CyrA, cyrA was deleted via homologous recombination (Fig 6A and 6B). The cyrA-deletion mutants displayed no vegetative growth phenotype, and trap formation was not affected (Fig 6C). A virulence assay was performed with wild type D. flagrans and the cyrA-deletion strain using the C. elegans strain Ban126 as prey to visualize progression of the infection and cell death (Fig 6D). Captured worms that are still alive displayed clear fluorescence in the nuclei. The GFP signal disappears when the cells die. We observed that indeed the nematode cells surrounding the trophic fungal hyphae died. There was no difference in the overall digestion process in the ΔcyrA-deletion strain as compared to wild type. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 6. The cyrA gene is required for full virulence. (A) Scheme of the deletion strategy. The wild type cyrA-gene locus (upper panel) is replaced by the 1.8 kb hygromycin resistance cassette resulting in the ΔcyrA-mutant locus (lower panel). Southern-blot analysis using a 1 kb fragment from the LB as probe and the restriction enzyme PstI for the digestion of the genomic DNA. (B) Confirmation of the ΔcyrA deletion using Southern-blot analysis. Genomic DNA was digested with PstI, and the LB was used as probe. (C) The D. flagrans WT and the ΔcyrA-deletion strain (sNH11) were grown for seven days at 28°C on PDA. (D) Virulence assay with wild type (WT), the ΔcyrA-mutant strain and the strain re-complemented with the cyrA ORF (sNH27) under its native promoter (Rec.). Multiple traps were observed and the time from the capturing event to full paralysis was taken. Error bars indicate the standard deviation. A student’s t-test was performed for statistical analysis (*** = p-value < 0.0001; n [WT] = 29 n [KO] = 26 n [Rec] = 52). https://doi.org/10.1371/journal.ppat.1010028.g006 After the initial experiments, long-term observations of single captured nematodes were performed as described above. This experiment was carried out with wild type, the deletion mutant, and a re-complemented strain using C. elegans BAN126 L3-larvae. For the complementation, the mutant strain was transformed with a plasmid containing the full-length ORF including the regulatory regions, and death was determined by loss of the nuclear GFP signal. We also used a CyrA-GFP encoding plasmid for re-complementation to prove that the fusion protein is biologically active. C. elegans L3-larvae captured by D. flagrans wild type were paralyzed after 49 min [SD ± 27] and worms captured by the complementation strain after 62 min [SD ± 39]. The mutant strain paralyzed the worms within 95 min [SD ± 44]. The overall time until death of the mutant strain was with 206 min [SD ± 141] not significantly different compared to wild type, taking 146 min [SD ± 80] (Fig 6D). This indicates that CyrA directly or indirectly plays a role in either the penetration or the paralysis of the worms. The same assay was performed with L4-larvae, and the wildtype strain took 103 min [SD ± 46] to digest the larger nematodes whereas the cyrA mutant took 184 min [SD ± 92] (Fig 5D). This is in accordance with the prolonged time to paralysis in the ΔexoA-deletion strain, in which CyrA cannot accumulate at the infection bulb and therefore shows the same phenotype as the ΔcyrA-deletion strain.

[END]

[1] Url: https://journals.plos.org/plospathogens/article?id=10.1371/journal.ppat.1010028

(C) Plos One. "Accelerating the publication of peer-reviewed science."
Licensed under Creative Commons Attribution (CC BY 4.0)
URL: https://creativecommons.org/licenses/by/4.0/


via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.