Introduction

Many fungal pathogens of plants, insects and mammals can switch between parasitic, saprophytic and symbiotic lifestyles in response to changing environmental conditions. Understanding the mechanisms by which they switch between these disparate lifestyles has important implications for agriculture and medicine1,2,3,4. Metarhizium robertsii is an excellent example of a fungus with multifactorial lifestyles5. It is well adapted to life in the soil as a saprophyte, and some isolates such as ARSEF2575 (Mr2575) can colonize plant roots and promote plant growth6,7,8,9. Mr2575 is also an entomopathogen with an ability to kill a wide spectrum of insects, and accordingly has been developed as a biocontrol agent against agricultural pests and vectors of human diseases10.

The transition from saprophyte to pathogen is initiated when conidia adhere to the cuticle of a susceptible insect host and produce germ tubes, which differentiate into infection structures called appressoria (sticky holdfasts that attach to the cuticle). The appressoria produce infection pegs which penetrate the cuticle via a combination of mechanical pressure and cuticle-degrading enzymes. The fungus proliferates in the host hemocoel as a yeast-like phase (blastospores), and the insect is killed by a combination of fungal growth and toxins. Hyphae subsequently re-emerge from the cadaver to produce conidia that have the potential to enter into saprophytic, symbiotic or pathogenic lifestyles11.

Many genes involved in the lifestyle switch to pathogen have been experimentally characterized; their gene products are generally expressed at low levels outside the host. They include an adhesin (MAD1) and hydrophobins that are responsible for adherence to the cuticle12,13, two chitin synthases for appressorial formation14, and a large number of cuticle-degrading enzymes for penetration of the cuticle15,16. Gene products associated with colonizing the hemocoel include the cold shock protein CRP1, laccase Mlac1, sterol carrier Mr-NPC2a, the collagen-like protein MCL1, enzymes for anaerobic respiration, and toxic secondary metabolites such as destruxins17,18,19,20,21,22. Major signaling pathway MAPK cascades and cAMP-PKA have been found to regulate both saprophytic growth and pathogenesis23,24. Nevertheless, no mechanisms have been characterized that control M. robertsii’s choice of lifestyle options as a saprophyte or a pathogen.

Here, from analysis of a random T-DNA insertion library22, we identify a membrane anchor protein (Mr-OPY2) that controls the saprophyte-to-pathogen transition of M. robertsii. Mr-OPY2 protein levels are low during saprophytic growth, and when elevated they initiate appressorial formation. Precise regulation of the Mr-OPY2 protein level is achieved via alternative transcription start sites. We further find that Mr-OPY2 controls appressorial formation by regulating a previously unidentified transcription factor, AFTF1 (appressorial formation transcription factor 1).

Results

Identification of the membrane anchor protein Mr-OPY2

From a T-DNA insertion mutant library, we identified a mutant (M2880) that cannot infect insects. Gene mapping showed that the T-DNA in the mutant was inserted in the open reading frame (ORF) of a gene (MAA_03000) that encodes a protein with significant similarity (1e−08) to the membrane anchor protein OPY2p from Saccharomyces cerevisiae (AJV94457), and we designated this gene as Mr-OPY2 (M. robertsii OPY2). Mr-OPY2 is a single copy gene with an 1,305 bp ORF that is interrupted by one intron, and encodes a protein containing 434 amino-acid residues. The deduced Mr-OPY2 has an OPY2 domain (PFAM09463) from amino acids 30 to 68, and a transmembrane domain from amino acids 93–115 as predicted by TMHMM (version 2.0)25, and domains showing significant similarity to CR-A and D in S. cerevisiae Opy2p26. Blast analysis found that homologs of Mr-OPY2 were widely distributed in both pathogenic and saprophytic Ascomycota and Basidiomycota. Phylogenetic analysis showed that the clade containing OPY2 proteins from Ascomycota yeasts was basal to the clade that contained OPY2 proteins of filamentous Ascomycota and Basidiomycota. The OPY2 genes of the filamentous fungi conform to their species tree, consistent with their being derived from a common ancestral sequence (Supplementary Fig. 1a, Supplementary Table 1). We compared the topology of the obtained tree with those of alternative trees using nine tests including the SH-test and the AU-test provided by the program CONSEL27. The nine tests consistently showed that the obtained tree (Supplementary Fig. 1a) to be the best supported (Supplementary Fig. 1b).

Mr-OPY2 and pathogenicity

We disrupted Mr-OPY2 in the wild-type strain (WT) to produce ΔMr-OPY2. The genes and strains used in this study are listed in Table 1. ΔMr-OPY2 was complemented with its genomic clone containing the upstream (1,747 bp) and downstream (245 bp) regions to produce strain C-ΔMr-OPY2. The confirmation of gene disruption and mutant complementation is presented in Supplementary Fig. 2.

Table 1 Transcripts and strains in this study
Full size table

The pathogenicity of M. robertsii was assayed on Galleria mellonella larvae. Inoculations were conducted either by topically applying conidia onto the insect cuticle or by direct injection of conidia into the hemocoel (thus bypassing the cuticle). As with the T-DNA insertion mutant M2880, ΔMr-OPY2 was unable to infect insects via topical application (Fig. 1a), but it was as pathogenic as the WT following injection (Supplementary Fig. 3). C-ΔMr-OPY2 showed WT levels of virulence (Supplementary Fig. 3) following topical application or direct injection (Fig. 1a and Supplementary Fig. 3). Insects killed by either the WT or injected ΔMr-OPY2 were mummified with dark green conidia produced on the cadavers (Fig. 1b).

Fig. 1
figure 1

Pathogenicity of WT, the mutants ΔMr-OPY2 and ΔMr-STE50, and their respective complementation strains. a LT50 (time taken to kill 50% of insects) values when the insects were inoculated by topical application. The bioassays were repeated three times with 40 insects per repeat. Data are expressed as the mean ± SE. Values with different letters are significantly different (n = 3, P < 0.05, Tukey’s test in one-way ANOVA). b Upper panel: mycelial growth and conidiation on the surface of cadavers (scale bar represents 10 mm). Lower panel: mycelia in cross sections of cadavers (scale bar represents 5 mm). Each image is representative of ~ 120 insect cadavers (three replicates with 40 insects per replicate). c Formation of appressoria (stained with Calcofluor Brightener White 2B) against a hydrophobic plastic surface. AP: appressorium; CO: conidium; SP: septum. Left: bright field microscopy. Note: the hyphal tips of ΔMr-OPY2 and ΔMr-STE50 do not swell to form appressoria. Right: fluorescence microscopy. Note: the septum between the appressorium and its appressorial mother cell. Images are representative of at least three independent experiments for each condition. Scale bar represents 10 μm

Full size image

We investigated whether the mutants were able to produce appressoria against insect cuticle and another normally inductive milieu-the hydrophobic surface of plastic petri dishes in the presence of low levels of nitrogenous nutrients28. ΔMr-OPY2 germ tubes meandered across the plastic and hindwings of Locusta migratoria manilensis, but their hyphal tips did not swell to initiate appressorial formation. Fluorescent staining with Calcofluor white Brightener 2B showed that the fluorescent intensity of the ΔMr-OPY2 hyphal tip was not distinguishable from other parts of the hyphae (Fig. 1c), whereas both WT and C-ΔMr-OPY2 produce normal appressoria that fluoresce strongly (Fig. 1c). Therefore, the cell wall structure or composition is altered in the growing hyphal tips of ΔMr-OPY2. In addition, appressorial differentiation of M. robertsii typically occurs after nuclear division with a septum being formed between the appressorium and appressorial mother cell28. This septation was observed in WT and C-ΔMr-OPY2, but not in ΔMr-OPY2 (Fig. 1c). WT and C-ΔMr-OPY2 germlings typically had three septa when the hyphal tips initiated appressorial formation, at which time ΔMr-OPY2 only had two. 18 h after inoculation, when ~ 80% of WT hyphal tips had produced appressoria against plastic surfaces, the hyphae of WT (27.4 ± 1.82 μm) were significantly (t-test, P = 0.007) shorter than ΔMr-OPY2 (42.7 ± 0.29 μm), as ΔMr-OPY2 showed continued polar growth rather than differentiation into appressoria.

Mr-OPY2 has two mRNA variants during infection

Northern blots and 5′ and 3′ RACE (Rapid Amplification of cDNA Ends) were used to analyze RNA during: (1) saprophytic growth (in SDY broth (Sabouraud dextrose broth plus 1% yeast extract)); (2) cuticle penetration (conidia germinating and differentiating appressoria on locust wings) and 3) hemocoel colonization (hyphae cultured in Bombyx mori (silkworm) hemolymph). Northern analysis revealed that a single Mr-OPY2 transcript was produced during saprophytic growth, and an additional less intense smaller band was produced in the two infection stages (Fig. 2a). RACE was performed using the RLM RT-PCR kit (Roche, USA) that only clones intact mRNA with a 7-methyl guanosine cap structure. Only one PCR band was obtained from the three RNA samples with 3′ RACE. Consistent with Northern analysis, 5′ RACE obtained one long band using RNA from M. robertsii mycelia grown in SDY broth, and an additional shorter band from M. robertsii when penetrating cuticle or growing in hemolymph (Fig. 2b). All PCR products were cloned, sequenced and assembled. The long 1,836 bp and shorter 1,452 bp transcripts are designated as Mr-OPY2-L (Genbank number: KY548479), and Mr-OPY2-S (Genbank number: KY548480), respectively. The two mRNA variants contain an identical ORF (designated as major ORF:mORF) that encodes Mr-OPY2 (Fig. 2c), and differ only in that the 5′ UTR of Mr-OPY2-L (designated as 5′ UTRL) is 384 bp longer than that of Mr-OPY2-S (designated as 5′ UTRS). The first nucleotide of Mr-OPY2-L (T) is different from that of Mr-OPY2-S (C), and no intron was found in the region corresponding to the UTRs, indicating that the two mRNA variants result from alternative transcription start sites.

Fig. 2
figure 2

Transcription and translation analyses of the Mr-OPY2 gene. a Representative Northern blot image of RNA isolated from undifferentiated mycelia grown in SDY or hemolymph (HE), and from germlings differentiating appressoria on locust wings (AP). The uncropped Northern blot is shown in Supplementary Fig. 11a. b 5′RACE with the RLM RT-PCR kit. The uncropped image of the agarose gel is shown in Supplementary Fig. 11b. c A diagrammatic representation of differences between the two mRNA variants of Mr-OPY2. Note: the two transcripts share the same major ORF but they have different 5′UTR. d Representative western blot image and analysis to quantify Mr-OPY2 levels in ΔMr-OPY2 and WT grown in SDY or hemolymph (HE) (Upper panel). The OPY2 levels of hemolymph-grown WT mycelium (1.4) is calculated relative to growth in SDY which is set at 1. ΔMr-OPY2 is included as a negative control. Lower panel: a portion of the loading control gel (for the Western blot analysis in upper panel) stained with Coomassie Brilliant Blue. M: Protein ladder (Thermo scientific, USA). The uncropped images of the western blot and the SDS-PAGE gel are shown in Supplementary Fig. 11c, d. e Immunohistochemical staining of the Mr-OPY2 protein in germlings differentiating appressoria against a plastic surface, and in non-differentiating germlings from 1/2 SDY shake cultures. Scale bar represents 10 μm. FITC: Fluorescein isothiocyanate; Merge: FITC and bright field pictures are merged; AP: appressorium. Note: no fluorescent signal in ΔMr-OPY and a weak signal in WT grown in 1/2 SDY. Images are representative of at least three independent experiments for each condition

Full size image

Western blot analysis showed that there was 1.4-fold more Mr-OPY2 protein in hemolymph cultures than in SDY (Fig. 2d). We could not obtain sufficient biomass from differentiating appressoria to extract membrane proteins for western analysis, so we used indirect immunofluoresence (IIF) assays to compare Mr-OPY2 protein levels in appressoria with those in non-differentiating germlings in SDY. The assay results showed that the Mr-OPY2 protein was more abundant in appressoria (Fig. 2e).

To identify differences in copy number between Mr-OPY2-L and Mr-OPY2-S we used two pairs of primers for quantitative RT-PCR (qRT-PCR) analysis of the 5′ UTRL and mORF (Supplementary Fig. 4). No significant differences were found in the levels of 5′ UTRL and mORF during saprophytic growth and the two infection stages. Therefore, expression of the Mr-OPY2-S transcript did not significantly contribute to mORF copy number. Consequently, changes in Mr-OPY2 protein levels during lifestyle transitions are most probably attributable to translational regulation of mORF in Mr-OPY2-L and Mr-OPY2-S.

uORFs mediate translational suppression of mORF in Mr-OPY2-L

The 5′ UTRS of Mr-OPY2-S lacks uORFs (upstream ORFs), whereas two uORFs were identified in the 5′ UTRL of Mr-OPY2-L (Fig. 3a). As uORFs suppress translation efficiency in many genes29,30,31, we investigated the effect of the uORFs on translation of the main ORF in Mr-OPY2-L. To do this, we cloned the genomic DNA fragment (gMr-OPY2) that contains the two exons, the single intron and the termination region (200 bp), and mutated the AUGs to GUGs in the 5′ UTRL to generate gMr-OPY2 ΔAUGs (Fig. 3a, Supplementary Fig. 5a). The knock-in method was used to replace the corresponding region in M. robertsii’s genome with gMr-OPY2 ΔAUGs, producing the strain T-Mr-OPY2-L ΔAUGs. As a control, the corresponding region was also replaced with the unmutated gMr-OPY2 to produce the strain T-Mr-OPY2 AUG (Supplementary Fig. 5a, b). qRT-PCR of SDY mycelial cultures revealed that expression of the Mr-OPY2 ORF was similar (P > 0.05, Tukey’s test in one-way analysis of variance (ANOVA)) in WT, T-Mr-OPY2 AUG and T-Mr-OPY2 ΔAUGs (Supplementary Fig. 5c), whereas western analysis showed that Mr-OPY2 protein levels in T-Mr-OPY2 ΔAUGs were increased ~4-fold compared with the WT and T-Mr-OPY2 AUG (Fig. 3a). No obvious difference in the level of Mr-OPY2 protein was observed between WT and T-Mr-OPY2-L AUG (Fig. 3a). These results show that upstream AUG triplets in the 5′ UTR of the long mRNA variant of Mr-OPY2 reduce translatability and protein levels.

Fig. 3
figure 3

The uORFs suppress translation efficiency of the major ORF in Mr-OPY2-L. a Top panel: graphic representation of WT transcripts Mr-OPY2-L, Mr-OPY2-S and the mutated Mr-OPY2-L (Mr-OPY2-L ΔAUGs) that has GUGs in place of AUGs (highlighted in red). The blue and green labelled sequences in the 5′UTR represent the positions of the two different uORFs (these color conventions are also used in the top panel in b. Middle panel: Mr-OPY2 protein levels in SDY grown cultures of: (1) ΔMr-OPY2; (2) WT; (3) T-Mr-OPY2-L AUG (the native Mr-OPY2 gene was replaced by a native genomic Mr-OPY2 clone); and (4) T-Mr-OPY2 ΔAUGs (the native Mr-OPY2 gene was replaced by a genomic Mr-OPY2 clone with mutated uORFs). Bottom panel: a portion of the loading control gel (for the Western blot analysis in the middle panel) stained with Coomassie Brilliant Blue. M: Protein ladder. b The short Mr-OPY2-S transcript is more efficiently translated than the longer Mr-OPY2-L. Top panel: graphic representation of the constructs containing the transcript Mr-OPY2-L, Mr-OPY2-S, and Mr-OPY2-L ΔAUGs driven by the Pgpd-NUTR. The 'No 5′UTR' indicates that the 5′UTR of the M. acridum gpd gene is excluded. These constructs were transformed into the Mr-OPY2 deletion mutant ΔMr-OPY2. Middle panel: representative Western blot image and analysis to quantify Mr-OPY2 levels in SDY grown cultures of: (1) ΔMr-OPY2; (2) WT; (3) ΔMr-OPY2::Mr-OPY2-L; (4) ΔMr-OPY2::Mr-OPY2-S, and (5) ΔMr-OPY2::Mr-OPY2-L ΔAUGs. Bottom panel: a portion of the loading control gel (for the western blot analysis in the middle panel) stained with Coomassie Brilliant Blue. M: Protein ladder. For western blot analysis (a, b middle panels), the Mr-OPY2 levels are calculated relative to the WT which is set at 1 (ΔMr-OPY2 is a negative control). The uncropped images of the western blots and the SDS-PAGE gels are shown in Supplementary Fig. 12. Images are representative of at least three independent experiments

Full size image

Secondary structures of 5′ UTRs may affect translation of the downstream ORF. We thus predicted the secondary structures of 5′ UTRL and 5′ UTRS using the Sfold Web Server (http://sfold.wadsworth.org/cgi-bin/index.pl). The 5′ UTRL formed a complex secondary structure with many stem-loops, and had a Gibbs free energy ΔG0 37 = −25.6 Kcal mol−1. In contrast, 5′ UTRS formed a simple stem-loop structure with ΔG0 37 = −7.2 Kcal mol−1. The secondary structure of the 5′ UTR of Mr-OPY2-L ΔAUGs (ΔG0 37 = −26.9 Kcal/mol) was similar to that of Mr-OPY2-L (Supplementary Fig. 5d). Secondary structures with free energy higher than ΔG = −30 Kcal/mol are not considered to impair translation32. Therefore, it is unlikely that the secondary structures of 5′ UTRL and 5′ UTRS substantially effect translation of the downstream ORF.

To characterize the effects of 5′ UTRL and 5′ UTRS further, we constructed expression plasmids where the transcripts Mr-OPY2-S, Mr-OPY2-L and Mr-OPY2-L ΔAUGs were placed downstream of the constitutive gpd gene promoter (Pgpd) from M. acridum 33. The 5′ UTR of the gpd gene was not included in the promoter to avoid it impacting translation (Fig. 3b). These plasmids were then transferred into ΔMr-OPY2 (i.e., the native mORF was deleted) to construct strains ΔMr-OPY2ːːMr-OPY2-S, ΔMr-OPY2ːːMr-OPY2-L, and ΔMr-OPY2ːː Mr-OPY2-L ΔAUGs. Strains with similar levels of the mORF transcript were selected for further analysis (Supplementary Fig. 6). In SDY cultures, the level of Mr-OPY2 protein in ΔMr-OPY2ːːMr-OPY2-S was ~4-fold higher than ΔMr-OPY2ːːMr-OPY2-L, indicating that Mr-OPY2-S was more efficiently translated than Mr-OPY2-L. Conversely, ΔMr-OPY2ːː Mr-OPY2-L ΔAUGs and ΔMr-OPY2ːːMr-OPY2-S had similar Mr-OPY2 protein levels (Fig. 3b). Taken together these results suggest that: (1) the 5′ UTRL secondary structure does not affect the translation efficiency of mORF, and (2) uORFs are major factors for suppressing the translation of mORF.

Differences in Mr-OPY2 levels lead to lifestyle transitions

As described above, alternative transcription start sites produce differences in Mr-OPY2 protein levels that characterize saprophytic and pathogenic lifestyles. In order to investigate the impact of Mr-OPY2 protein abundance on lifestyle transitions, we took advantage of a series of truncated Pgpd promoters of the M. acridum gpd gene which express foreign genes at different strengths33. We selected the 404 bp (P404) and 683 bp (P683) long promoters, which include the 5′UTR of the M. acridum gpd gene, to drive the major ORF of Mr-OPY2 in the Mr-OPY2 deletion mutant (ΔMr-OPY2) (Fig. 4a), i.e., the expression of Mr-OPY2 was exclusively controlled by the constitutive gpd promoters. As expected, qRT-PCR analysis showed that expression of the Mr-OPY2 transcript in P683-Mr-OPY2 (driven by the promoter P683) is significantly higher (t-test, P = 0.006) than expression driven by P404 (strain designated P404-Mr-OPY2), and the quantity of Mr-OPY2 protein in P683-Mr-OPY2 was ~ 8-fold higher than in either P404-Mr-OPY2 or WT (Fig. 4a). Three isolates were randomly selected from P404-Mr-OPY2 and P683-Mr-OPY2 colonies. As no intra-strain differences were found between the three isolates of each strain in any assay we provide representative data for a single isolate/strain.

Fig. 4
figure 4

Precise regulation of Mr-OPY2 levels is important for saprophytic growth and infection. a Constructing strains with manipulated Mr-OPY2 levels. Left panel: diagram illustrating constructs where Mr-OPY2’s ORF is driven by the two Pgpd promoters, P683 and P404, with high and low activity, respectively. These constructs were transformed into ΔMr-OPY2 to produce strains where the Mr-OPY2 level was exclusively controlled by the promoters. The 'With 5′UTR' indicates the 5′UTR of the M. acridum gpd gene is included. Middle panel: western blot analysis of Mr-OPY2 levels in the transformants: (1) ΔMr-OPY2; (2) WT; (3) P404-Mr-OPY2; and (4) P683-Mr-OPY2. Mr-OPY2 levels are calculated relative to the WT which is set at 1. ΔMr-OPY2 is a negative control. Right panel: a portion of the loading control gel (for the Western blot analysis in the middle panel) stained with Coomassie Brilliant Blue. M: Protein ladder. The uncropped images of the western blot and the SDS-PAGE gel are shown in Supplementary Fig. 13. b Colony phenotype (left upper panel), conidiophores (left lower panel) and conidial yields (right panel) of strains with different Mr-OPY2 levels. Note: strain P683-Mr-OPY2 with elevated Mr-OPY2 level relative to WT has a fluffy colony with impaired conidiophores and reduced conidial yield. Colony pictures were taken 18 days after inoculation. Scale bar in the left upper panel represents 10 mm, left lower panel 10 μm. In the right panel, values with different letters are significantly different (n = 9, P < 0.05, Tukey’s test in one-way ANOVA). c Pathogenicity of strains with different Mr-OPY2 levels. Left panel: appressorial development on a hydrophobic surface. At each time point, values with different letters are significantly different (n = 3, P < 0.05, Tukey’s test in one-way ANOVA). Right panel: survival curves of G. mellonella larvae infected by the WT, P404-Mr-OPY2 and P683-Mr-OPY2. Control: insects treated with 0.01% Triton X-100 solution. Bioassays were repeated three times with 40 insects per repeat. Conidial yield and appressorium formation assays were repeated three times with three replicates per repeat. Data are expressed as the mean ± SE. Images are representative of at least three independent experiments for each condition

Full size image

During saprophytic growth on PDA (potato dextrose agar) plates, the growth rate and conidial yield of P404-Mr-OPY2 was similar to WT (Fig. 4b). However, the more highly expressing P683-Mr-OPY2 produced a fluffy phenotype, abnormal conidiophores and significantly (P < 0.01, Tukey’s test in one-way ANOVA) lower conidial yield than WT (Fig. 4b). Compared with WT, appressorial formation by P404-Mr-OPY2 and P683-Mr-OPY2 was delayed (Fig. 4c), and the pathogenicity of both P683-Mr-OPY2 and P404-Mr-OPY2 was significantly (P < 0.05, Tukey’s test in one-way ANOVA) reduced (Fig. 4c).

Mr-OPY2 and tolerance to abiotic stresses

As S. cerevisiae Opy2p is involved in tolerance to high osmotic stress (Yamamoto et al., 2016), we investigated whether M. robertsii OPY2 is also involved in tolerance to abiotic stresses. ΔMr-OPY2 germinates and grows in SDY broth or on PDA plates at the same rate as WT and C-ΔMr-OPY2 (Fig. 5a, Supplementary Fig. 7), and this was not altered by adding 0.01% H2O2 to induce oxidative stress (Supplementary Fig. 7). The cell wall-disturbing agent Congo red (1 mg ml−1) produced no significant (P > 0.05, Tukey’s test in One-way ANOVA) differences in growth rate between WT, ΔMr-OPY2 and C-ΔMr-OPY2 on PDA (Supplementary Fig. 7), but the germination rate of ΔMr-OPY2 in SDY was significantly reduced (P < 0.05, Tukey’s test in one-way ANOVA) (Supplementary Fig. 7). Under hyperosmotic stress (SDY plus 0.75 M KCl), WT, ΔMr-OPY2 and C-ΔMr-OPY2 showed similar (P > 0.05, Tukey’s test in One-way ANOVA) germination rates, but the ΔMr-OPY2 hyphae were severely deformed forming a pearl-like shape (Fig. 5a). Similarly, unlike the C-ΔMr-OPY2 and WT, mycelial growth of ΔMr-OPY2 on PDA plates supplemented with 0.75 M KCl was also greatly (P < 0.01, Tukey’s test in One-way ANOVA) inhibited (Fig. 5b). However, the level of Mr-OPY2 protein in WT was the same in SDY and in SDY supplemented with 0.75 M KCl (Fig. 5c). No significant differences were observed between WT, P683-Mr-OPY2 and P404-Mr-OPY2 in their tolerance to oxidative stress, high osmotic stress and Congo red (Supplementary Fig. 8).

Fig. 5
figure 5

Colony phenotypes and osmotic stress tolerance of ΔMr-OPY2, ΔMr-STE50 and WT. a Upper panel: colonies of the strains on PDA (Scale bar represents 10 mm). Middle panel: colonies on PDA plus 0.75 M KCl (scale bar represents 1 mm). Lower panel: bright field images of individual germlings in YE (0.01% yeast extract) plus 0.75 M KCl (Scale bar represents 10 μm). Colony pictures were taken 18 days after inoculation. Germling pictures were taken 16 h after inoculation. b Growth curves of the strains on PDA plates plus 0.75 M KCl. Growth assays were repeated three times with three replicates per repeat. Data are expressed as the mean ± SE. Note: growth of ΔMr-OPY2 and ΔMr-STE50 was greatly reduced. c Upper panel: Western blot analysis of Mr-OPY2 protein levels in ΔMr-OPY2 grown in SDY (1), WT grown in SDY (2) and in SDY plus 0.75 M KCl (3). The band intensity of WT in SDY is set at 1, and WT in SDY with KCl is relative to it; ΔMr-OPY2 is a negative control. Lower panel: a portion of the loading control gel (for the western blot analysis in the upper panel) stained with Coomassie Brilliant Blue. M: Protein ladder. The uncropped images of the Western blot and the SDS-PAGE gel are shown in Supplementary Fig. 13. d Upper panel: yeast two-hybrid confirmation of the physical interaction of Mr-SET50 with Mr-OPY2 and three MAPKKK proteins (SSK2, STE11 and BCK1). Colonies were grown in SD-His-Ade-Leu-Trp + X-α-gal + AbA (Takara, Dalian China). NC: negative control (yeast cells containing the plasmid pGADT7-T and pGBKT7-Lam). PC: positive control (yeast cells containing the plasmid pGADT7-T and pGBKT7-53). Lower panel: Mr-STE50 lacks autoactivation activity. Y2HGold cells with pGBKT7-Mr-STE50 cannot grow in SD-His-Trp-Ade + X-α-gal (Takara, Dalian China). NC and PC are the same as those in the upper panel. Images are representative of at least three independent experiments for each condition

Full size image

Mr-OPY2 controls phosphorylation levels of MAPKs

The S. cerevisiae OPY2 protein regulates the phosphorylation level of MAPKs by complexing with the adaptor protein STE50p that in turn interacts with three MAPKKK proteins26. We used yeast two-hybrid assays to confirm that the M. robertsii Mr-OPY2 physically interconnects with MAA_02467, the M. robertsii ortholog of STE50p (designated Mr-STE50), and thereafter the three MAPKKK proteins Mero-Ste11, Mero-Ssk2 and Mero-Bck1 (Fig. 5d). We disrupted Mr-STE50 to produce ΔMr- STE50, which was complemented with its genomic clone to produce C-ΔMr- STE50. Confirmation of gene disruption and mutant complementation is shown in Supplementary Fig. 2. ΔMr-STE50 produced greyish green conidia compared with the darker green conidia of WT colonies (Fig. 5a). ΔMr- STE50 had reduced tolerance to high osmotic stress (Fig. 5b) and the cell wall-disturbing agent Congo Red (Supplementary Fig. 7), but its tolerance to oxidative stress was unaltered (Supplementary Fig. 7). Like ΔMr-OPY2, ΔMr-STE50 did not produce appressoria, and was consequently unable to kill insects via topical application (Fig. 1a, c), but it showed WT virulence when injected into insects (Supplementary Fig. 3). Mycelia of WT and ΔMr-OPY2 grew out of cadavers commencing 3 days after death. In contrast, no mycelium emerged on the ΔMr-Ste50 infected insects, although cadavers contained mycelia (Fig. 1b). There were no significant (P > 0.05, Tukey’s test in one-way ANOVA) differences between the WT and C-ΔMr-STE50 in any assays.

Previously, we found that the Fus3-MAPK and Hog1-MAPK signaling pathways regulate M. robertsii’s infectivity and tolerance to high osmotic stress24. To investigate if Mr-OPY2 regulates these two MAPK signaling pathways we compared the phosphorylation level of Fus3-MAPK and Hog1-MAPK in WT, ΔMr-OPY2 and ΔMr-STE50 using their respective antibodies. During saprophytic growth in SDY, phosphorylation levels of Hog1-MAPK and Fus3-MAPK were similar in ΔMr-OPY2 and ΔMr-STE50, and slightly lower than in the WT (Fig. 6). Phosphorylation of Fus3-MAPK and Hog1-MAPK was reduced in ΔMr-OPY2 and ΔMr-STE50 growing in hemolymph or under high osmotic stress (Fig. 6). Thus, in hemolymph, phosphorylation of Fus3-MAPK was reduced 1.7-fold and 11-fold relative to WT in ΔMr-STE50 and ΔMr-OPY2, respectively. During appressorial formation, phosphorylation of Hog1-MAPK was reduced in the two mutants; whereas phosphorylation of Fus3-MAPK was unchanged compared wit the WT (Fig. 6).

Fig. 6
figure 6

Regulation of MAPK signaling pathways by Mr-OPY2 and Mr-STE50 during saprophytic growth, pathogenesis and under high osmotic stress. β-Tubulin was used as a loading control. Numbers indicate band intensity for Phos-Fus3p (Phos-Hog1p) relative to total Fus3p (Hog1p). WT values were set to 1. Images are representative of at least three independent experiments for each condition. The uncropped images of the western blots are shown in Supplementary Fig. 14

Full size image

Mr-OPY2 negatively regulates AFTF1 during cuticle penetration

RNA-seq was used to compare WT and ΔMr-OPY2 transcriptomes of undifferentiating hyphae in SDY and hemolymph cultures, and germlings differentiating appressoria on locust wings. WT and ΔMr-OPY2 grown in SDY, hemolymph and locust cuticle differed by 82, 5 and 39 DEGs (differentially expressed genes), respectively. Only one DEG encoding a hypothetical protein was shared in the three growth conditions (Fig. 7a), indicating that Mr-OPY2 regulates distinctive subsets of genes during saprophytic growth, cuticle penetration and colonization of the hemocoel.

Fig. 7
figure 7

Identification and characterization of the appressorial formation transcription factor AFTF1 and its negative regulation by Mr-OPY2. a Venn diagram of RNA-seq analysis showing the distribution of shared DEGs in transcriptomes of WT and ΔMr-OPY2 during growth in SDY (SDY), the insect cuticle (Cuticle) and hemolymph (Hemolymph). Two biological repeats were established for each treatment. b qRT-PCR analysis of Aftf1 expression by the WT grown in hemolymph and on locust cuticle relative to expression during saprophytic growth in SDY (which is set to 1 in the figure). c qRT-PCR analysis of Aftf1 expression by the WT, ΔMr-OPY2, ΔMr-STE50, ΔMero-Hog1, ΔMero-Fus3 and ΔMero-Slt2 during appressorial formation. The expression level in the WT is set to 1. The qRT-PCR analyses were repeated three times with three replicates per repeat. Data are expressed as the mean ± SE. Values with different letters are significantly different (n = 3, P < 0.05, Tukey’s test in one-way ANOVA). d The % germlings differentiating appressoria on a hydrophobic plastic surface in the WT, ΔAftf1 (the Aftf1 disruption mutant), C-ΔAftf1 (the complemented strain of ΔAftf1) and Aftf1 OE (a strain overexpressing Aftf1). Appressorium formation assays were repeated three times with three replicates per repeat. Data are expressed as the mean ± SE. At each time point, values with different letters are significantly different (n = 3, P < 0.05, Tukey’s test in one-way ANOVA). e Survival curves of G. mellonella larvae infected by WT, ΔAftf1, C-ΔAftf1 and Aftf1 OE. Data are expressed as the mean ± SE. Control: insects treated with 0.01% Triton X-100 solution. The bioassays were repeated three times with 40 insects per repeat. f A schematic model of Mr-OPY2-mediated regulation of appressorial formation and osmotic stress tolerance. Mr-OPY2 regulates the phosphorylation level of Fus3 and Hog1-MAPK under high osmotic stress. During appressorial formation, Fus3-MAPK regulates the expression level of Aftf1, but regulation of Aftf1 by Mr-OPY2 is mediated by unidentified components. Slt2-MAPK also regulates Aftf1, but it remains to be determined whether Mr-OPY2 is involved in such regulation

Full size image

During differentiation of appressoria and cuticle penetration, a single transcription factor (MAA_08552) with a Zn2Cys6 fungal type DNA binding domain, designated as AFTF1, was differentially expressed between WT and ΔMr-OPY2. qRT-PCR analysis showed Aftf1 was upregulated ~ 31 fold (P < 0.01, Tukey’s test in one-way ANOVA) on insect cuticle relative to hemolymph and SDY (Fig. 7b). On locust wings, expression of Aftf1 in ΔMr-OPY2 was 7.2-fold higher than WT (P < 0.01, Tukey’s test in one-way ANOVA), but it was reduced > 11-fold (P < 0.01, Tukey’s test in one-way ANOVA) in ΔMr-STE50 and the two MAPK mutants (ΔMeor-Fus3 and ΔMero-Slt2) (Fig. 7c). Both ΔMeor-Fus3 and ΔMero-Slt2 are impaired in appressorial formation24.

In order to investigate the functions of Aftf1, we deleted its ORF to produce ΔAftf1, which was complemented with its genomic clone to produce C-ΔAftf1. Confirmation of gene disruption and mutant complementation is shown in Supplementary Fig. 2. We also constructed strains (designated as Aftf1 OE) overexpressing Aftf1 driven by the constitutive promoter (Ptef) of the translation elongation factor gene from Aureobasidium pullulans 34. Germlings of two randomly selected Aftf1 OE transformants significantly (P < 0.01, Tukey’s test in one-way ANOVA) upregulated (~ 300-fold) transcription of Aftf1 compared with WT when differentiating appressoria on locust cuticle (Supplementary Fig. 9a). No intra-strain differences were found between the two isolates in any assay, and so representative data for a single isolate is provided. Conidiation and saprophytic growth of Aftf1 OE and ΔAftf1 on PDA resembled the WT (Supplementary Fig. 9b, c). However, appressorial formation by ΔAftf1, and to a lesser extent Aftf1 OE, was significantly delayed (P < 0.05, Tukey’s test in one-way ANOVA) (Fig. 7d), significantly (P < 0.05, Tukey’s test in one-way ANOVA) reducing pathogenicity compared to the WT (Fig. 7e). In all the above assays, there were no significant (P > 0.05, Tukey’s test in one-way ANOVA) differences between the WT and C-ΔAftf1 (Fig. 7 and Supplementary Fig. 9). Therefore, we conclude that the precise level of AFTF1 protein controlled by Mr-OPY2 is important for appressorial formation and pathogenicity.

Appressorial production is associated with synthesis of key effector proteins influencing virulence24. RNA-seq analysis showed that four cuticle-degrading proteases (MAA_00460, maa_07827, MAA_07828 and MAA_09637) were down-regulated in ΔMr-OPY2 during penetration of the cuticle. In addition, a cell wall surface protein Mas1 (MAA_08589) was upregulated; its ortholog is an important factor for appressorial formation in Magnaporthe oryzae 35. qRT-PCR analysis showed that the protease MAA_09637 was also down-regulated in ΔAftf1, suggesting that AFTF1 could regulate a subset of the genes regulated by Mr-OPY2 (Supplementary Fig. 10). The expression level of these five genes did not differ significantly (P > 0.05, Tukey’s test in one-way ANOVA) between WT and Atf1 OE.

Discussion

Fungi can live as saprophytes in diverse habitats, on virtually any carbon source, or in commensal, mutualistic or parasitic relationships with plants, animals, insects or other fungi. The mechanisms that regulate transitions between saprophytic, symbiotic and pathogenic growth phases are poorly understood36, with the exception of a transcriptional regulator, Wor1/Ryp1, that governs lifestyle transitions in human pathogenic fungi, C. albicans and Histoplasma capsulatum 37,38, and several plant pathogens including Fusarium verticillioides (Sge1)36. In this study, we report that the anchor membrane protein Mr-OPY2 is dispensable for saprophytic growth of M. robertsii, but is essential for this fungus to initiate development of infection structures (appressoria), which are a prerequisite for infection. Therefore, Mr-OPY2 regulates the transition of M. robertsii from saprophyte to pathogen; to our knowledge, this is the first demonstration that OPY2 regulates the initiation of infection in a pathogenic fungus. Homologs of Mr-OPY2 are conserved in plant and mammal pathogens, suggesting a potential role in regulating pathogenesis in other fungi, especially the many plant pathogenic fungi (e.g. M. oryzae) that also differentiate appressoria.

The mechanism by which Mr-OPY2 controls M. robertsii’s saprophyte-to-pathogen transition involves precise regulation of its protein level, with higher levels during cuticular penetration than saprophytic growth. The Mr-OPY2 disruption mutant, and a manipulated strain (P404-Mr-OPY2) with a constantly low Mr-OPY2 protein level, were both impaired in appressorial formation and thus in pathogenicity. This confirms that upregulation of the Mr-OPY2 protein level is necessary for infection processes. Conversely, a low level of Mr-OPY2 protein is important for normal saprophytic growth because forcefully elevating Mr-OPY2 (the strain P683-Mr-OPY2), with a constitutive promoter produced an abnormal colony phenotype with reduced conidial yield. P683-Mr-OPY2 was also impaired in appressorial formation, indicating that constitutively expressing high levels of Mr-OPY2 protein destabilizes appressorial formation. Mr-OPY2 is also upregulated in the insect hemolymph, but its role following penetration of the cuticle remains to be elucidated because ΔMr-OPY2 resembled WT in its ability to colonize the hemocoel. Western blot analyses of Mr-OPY2 showed two bands in all strains (including WT) that express Mr-OPY2, which were absent in the deletion mutant of Mr-OPY2. One band is close to the theoretical molecular weight of Mr-OPY2 (45.5 kDa), whereas the other is ~50 kDa. The heavier band could result from post-translational modifications; the difference in the intensity of the two bands should result from the difference between the amount of protein modified and unmodified. C. albicans and S. cerevisiae OPY2 proteins were similarly bigger than their theoretical molecular weights due to glycosylation39,40.

The precise regulation of Mr-OPY2 protein levels is achieved via alternative transcription start sites. During pathogenesis, two Mr-OPY2 transcripts were produced; the abundant long transcript was inefficiently translated because its 5′UTR has two small uORFs that were proven by two independent experiments to reduce translation efficiency of the downstream major ORF. As described in other studies29,30,31, small uORFs can modulate ribosomal access to the AUG start codon of the major ORF, decreasing its translation efficiency. Although the short transcript of Mr-OPY2 has a lower copy number than the long one, it is more efficiently translated. Thus, although the long transcript was abundant during both saprophytic growth and pathogenesis, increased levels of the Mr-OPY2 protein during pathogenesis derived from the short transcript. We are not aware of any other reports on regulation of OPY2 protein levels via alternative transcription start sites. However, alternating transcription start sites is a common mechanism for regulating gene expression in fungi41.

Only a single novel transcription factor (AFTF1) is regulated (negatively) by Mr-OPY2 during cuticle penetration. Homologs of AFTF1 are also found in other fungi such as Fusarium oxysporum (EWY87630) and M. oryze (XP_003715433). While upregulation of AFTF1 is critical for appressorial formation, the process was delayed by overexpressing AFTF1. Therefore, a precisely modulated optimal level of AFTF1 is needed to choreograph appressorial formation with Fus3- and Slt2-MAPK upregulating Aftf1 during cuticle penetration and Mr-OPY2 ensuring that the level of AFTF1 is not too high for optimum effect. The precise regulation of Mr-OPY2 is presumably an adaptation to facilitate the precise regulation of AFTF1. The phosphorylation level of Fus3-MAPK is not controlled by Mr-OPY2, suggesting that Mr-OPY2 does not regulate AFTF1 via Fus3-MAPK. Yeast two-hybrid assays showed that the MAPKKK (Bck1) of the Slt2-MAPK cascade interacts with Mr-STE50, which in turn interacts with Mr-OPY2, suggesting that Mr-OPY2 could regulate AFTF1 via Slt2-MAPK. Mr-OPY2 regulates phosphorylation of Hog1-MAPK during appressorial formation, but Hog1-MAPK does not regulate expression of Aftf1 indicating that Mr-OPY2 has additional functions besides regulating AFTF1 (Fig. 7f).

In addition to regulating pathogenesis, OPY2 is also necessary for M. robertsii to tolerate high osmotic stress, but as this stress did not alter Mr-OPY2 protein levels, low levels may be sufficient for stress resistance. Mr-OPY2 regulates M. robertsii’s tolerance to high osmotic stress via the Hog1-MAPK and Fus3-MAPK pathways (Fig. 7f). S. cerevisiae Opy2 protein also mediates tolerance to high osmotic stress via Hog1-MAPK, but the Opy2 deletion mutant can grow under the osmotic stress because the redundant SSK branch also regulates Hog1-MAPK to tolerate osmotic stress26,40,42. The Opy2 protein of C. albicans is not involved in tolerance to high osmotic stress40, or C. albicans’s pathogenicity against a natural host, but by an unknown mechanism does facilitate its ability to kill a non-natural host (the insect G. mellonella) following injection40. In contrast, we found that OPY2 is dispensable for hemocoel infection by the specialized insect pathogen M. robertsii. Therefore, the functions of OPY2 proteins have clearly diversified in fungi, along with their lifestyles. Our phylogenetic analysis confirmed that the OPY2 proteins from unicellular Ascomycota yeasts (including S. cerevisiae and C. albicans) are phylogenetically distant from their homologs in filamentous Ascomycota and Basidiomycota fungi.

In conclusion, we report that Mr-OPY2 controls the saprophyte-to-pathogen transition of the insect pathogen M. robertsii by negatively regulating the transcription factor AFTF1. This work describes a new circuit regulating fungal infection, and advances our insight into the mechanisms underpinning fungal lifestyle transitions.

Methods

Gene cloning and disruption

The flanking sequences of T-DNA were cloned by YADE (Y-shaped adaptor dependent extension) from M. robertsii mutants generated by T-DNA insertion as previously described22. M. robertsii ARSEF2575 and M. acridum ARSEF324 were obtained from the Agricultural Research Service Collection of Entomopathogenic Fungi. The primers used in this study are listed in Supplementary Table 2.

Genes were disrupted based on homologous recombination using our previously developed high-throughput gene disruption methodology43. Disruption plasmids were constructed using Ppk2-OSCAR-GFP and pA-Bar, and confirmation of gene disruption was performed as described43. To complement a gene disruption mutant, the genomic clone of the gene containing the promoter region, ORF and termination region was amplified by PCR and inserted into the pPK2-Sur-GFP43. High fidelity Taq DNA polymerase (KOD Plus Neo, Osaka Japan) was used in PCR reactions to clone DNA fragments, which were confirmed by DNA sequencing. Agrobacterium tumefaciens AGL1 was used for fungal transformation as previously described44.

Histoimmunochemical staining and western blot analysis

In order to prepare an antiserum to Mr-OPY2, a predicted antigenic region (PGPNASPDQIKKHRD), residues 59–73 of the N-terminal domain, was synthesized and conjugated with keyhole limpet hemocyanin. The antibodies were raised in New Zealand White rabbits (Sigma, USA). Total cell membrane protein from fungal cells was prepared using the Mem-PER Eukaryotic Membrane Protein Extraction Reagent Kit (Thermo Scientific, USA), followed by purification with the Pierce sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) Sample Prep Kit (Thermo Scientific). Two identical SDS-PAGE gels were set up, one of which was used for western blot analysis, and the other one was stained with Coomassie Brilliant Blue in order to visualize the sample loading. The antiserum (200 mg ml−1) was used at a 1ː500 dilution to detect Mr-OPY2 protein with western blot analysis.

To analyze phosphorylation levels of Fus3 and Hog1-MAPK in M. robertsii, cell proteins were prepared with TCA (trichloroacetic acid) buffer as described45, and blotted to polyvinyldene difluoride membranes (Bio-Rad, USA). Rabbit anti-phospho-p38 MAPK (Thr180/Tyr182) (Catalogue number: 9211), anti-phospho-p44 MAPK (Catalogue number: 9101) and anti-p38 MAPK (Catalogue number: 9212) (Cell Signaling Technology, MA, USA) were used at a 1ː1000 dilution to detect phosphorylation of Hog1-MAPK, Fus3-MAPK and the protein level of Hog1-MAPK, respectively. To prepare an antiserum to the Fus3-MAPK protein, a predicted antigenic region (CHDPEDEPTAPPTP) was synthesized, and the antiserum was raised in New Zealand White rabbit (HUABIO, Hangzhou, China). The antiserum (0.4 mg ml−1 IgG) was used at a 1ː1000 dilution to detect the Fus3-MAPK protein. β-tubulin was used as a loading control, which was detected using its mouse antibody (Catalogue number: M1305-2, HUABIO) at a 1ː2000 dilution. Bound primary antibodies were revealed using horseradish peroxidase-conjugated goat anti-rabbit IgG (Catalogue number: HA1006, HUABIO) or goat anti-mouse IgG (Catalogue number: HA1006, HUABIO) with a 1ː4000 dilution, and a chemiluminescence detection system (Clarity Western ECL, Bio-Rad).

Appressoria on the hydrophobic surface of a diagnostic slide (Live Focus, Jiangsu, China) were prepared for IIF as described46. Mr-OPY2 antiserum (200 mg ml−1) was diluted 200-fold for primary labeling, and FITC-conjugated-goat anti-rabbit IgG (Catalogue number: HA1004, HUABIO) was used for secondary labeling.

All Histoimmunochemical staining and Western blot analyses were repeated at least three times, and representative results are shown.

Assays of pathogenicity and conidiation and stress tolerance

Bioassays were conducted using last instar G. mellonella larvae (Ruiqingbait Co., Shanghai, China) by topical application or direct injections as described (Zhao et al., 2014). Tolerance to high osmotic stress, oxidative stress and cell wall-disturbing stress was assayed as previously described24. Determination of conidial yields and observation of conidiophores were conducted as previously described24. All assays were repeated three times.

Site mutagenesis

PCR mutagenesis was used to mutate the three AUGs into GUGs and thereby delete the uORFs in the 5′UTR of Mr-OPY2-L. The Mr-OPY2 genomic clone (gMr-OPY2) containing its exons, introns and terminator region was cloned by PCR using high fidelity Taq DNA polymerase into the pUCm-T (Takara, Dalian, China) to produce the plasmid PUC-gMr-OPY2. Two complementary primers (Supplementary Table 2) containing mutated sites flanked by 20 bases on each side were designed and used to amplify PUC-gMr-OPY2. The PCR product was then digested with DpnI (NEB, England) to cut the parental plasmid DNA, whereas the nascent PCR DNA was left intact. The PCR products were then incorporated into the E. coli strain that has an intact dam methylase system. The resulting plasmids thus contained no uORFs, and this was confirmed by DNA sequencing. The AUG at –167 bp (167 bp upstream of the AUG of the major ORF) in gMr-OPY2 was initially mutated to produce gMr-OPY2 ΔAUG167, and the two linked AUGs at –363 bp and –366 bp in gMr-OPY2 ΔAUG167 were then mutated to form gMr-OPY2 ΔAUGs.

Constructing strains with the Mr-OPY2 gene driven by Pgpd

To analyze the translation efficiency of the transcripts Mr-OPY2-L, Mr-OPY2-S and Mr-OPY2-OPY2-L ΔAUGs, each was placed downstream of the promoter Pgpd-NUTR (the gpd gene’s 5′UTR excluded) of the gpd gene from M. acridum ARSEF32433. To do this, Pgpd-NUTR was amplified using the genomic DNA of M. acridum ARSEF324 as the template, which was then inserted into the NdeI/BamHI sites in the plasmid pBARGPE1 (obtained from Fungal Genetic Stock Center) to replace the Aspergillus nidulans gpd promoter, resulting in the plasmid pPgpd-NUTR. The DNA fragment containing Pgpd-NUTR and the TtrpC terminator was excised from pPgpd-NUTR with SpeI/SmaI and inserted into the SpeI/EcoRV sites of the plasmid pPK2-Sur-GFP43 to produce the plasmid pPK2-Pgpd-NUTR.The transcripts of Mr-OPY2-L, Mr-OPY2-S and Mr-OPY2-OPY2-L ΔAUGs were cloned by PCR, digested with EcoRI/SmaI and inserted into EcoRI/EcoRV sites downstream of the promoter Pgpd-NUTR in pPK2-Pgpd-NUTR to produce plasmids pPK2-Mr-OPY2-L, pPK2-Mr-OPY2-S and pPK2-Mr-OPY2-OPY2-LΔAUGs, respectively. The three plasmids were then transferred into Agrobacterium tumefaciens and transformed into the Mr-OPY2 deletion mutant ΔMr-OPY2 to produce the strains ΔMr-OPY2ːːMr-OPY2-S, ΔMr-OPY2ːːMr-OPY2-L and ΔMr-OPY2ːː Mr-OPY2-L ΔAUGs.

The strains P404-Mr-OPY2 and P683-Mr-OPY2 were constructed by placing the major ORF downstream of either the 404 bp promoter (P404) or the 683 bp promoter (P683) of the M. acridum gpd gene (including the 5′UTR of the gpd gene)33. The P404 and P683 promoters were cloned by PCR using the genomic DNA of M. acridum ARSEF324 as the template, and digested with enzymes NdeI/BamHI to replace the gpd promoter of A. nidulans in the plasmid pBARGPE1 and produce plasmids pP404 and pP683. The DNA fragment containing the promoter (P404 or P683) and trpC terminator region was excised with SpeI/SmaI and inserted into the SpeI/EcoRV sites in the pPK2-Sur-GFP plasmid to produce pPK2-P404 and pPK-P683, respectively. The mORF of Mr-OPY2 was cloned by PCR and excised with EcoRI/KpnI, which was then inserted into the corresponding sites downstream of the promoters P404 and P683 in pPK2-P404 and pPK-P683 to form the plasmids pPK2-P404-Mr-OPY2 and pPK2-P683-Mr-OPY2, respectively. pPK2-P404-Mr-OPY2 and pPK2-P683-Mr-OPY2 were then transferred into A. tumefaciens and transformed into the Mr-OPY2 deletion mutant ΔMr-OPY2 to produce the strains P404-Mr-OPY2 and P683-Mr-OPY2.

Transcriptomic analysis using RNA-Seq

RNA-seq was used to profile transcriptomes during saprophytic growth in SDY, appressorial formation (30 h post inoculation of conidia onto the hindwings of L. migratoria) and growth in B. mori hemolymph. Total RNA was extracted with TRIzol reagent (Life Technologies, USA). For saprophytic growth, conidia (1 × 106 conidia/ml) were grown at 26 °C for 36 h in SDY. For appressorial formation, hindwings of L. migrattoria manilensis were surface sterilized in 1% bleach as previously described21. A conidial suspension (200 μl of 1 × 105 conidia/ml) was applied to each hindwing which was then placed on a 1% water agar plate, and incubated at 26 °C for 30 h. For growth in the hemolymph, mycelium from a 36 h SDY culture was collected by filtration and washed with sterile water three times. One gram of mycelium was inoculated into 12 ml of B. mori hemolymph and incubated for 12 h at 26 °C with 100 rpm shaking. The mycelium was then collected by filtration for RNA extraction. Locust wings and silkworm hemolymph were prepared as described24. Two biological repeats were established for each treatment.

Construction of libraries and sequencing with the Illumina HiSeq 2500 platform were performed by Novogene (Beijing, China). After paired-end sequencing, clean reads were obtained using the NGS QC Toolkit and mapped to the draft genome sequence of M. robertsii ARSEF 2316 using the programs TopHat 2.0.647. Reads that aligned uniquely to the reference sequence were used for gene expression quantification using the RPKM method48. Differential expression analysis was performed with DESeq software (version 1.18.0)49, and an adjusted P-value 0.05 (Benjamini–Hochberg method).

Construction of the strain overexpressing Aftf1

The strain overexpressing Aftf1 was constructed by placing its ORF downstream of the constitutive promoter Ptef of the translation elongation factor gene from A. pullulans 34. To do this, the promoter Ptef was cloned by PCR using the plasmid pPK2-Sur-GFP43 as the template, and digested with enzymes NdeI/BamHI to replace the gpd promoter of A. nidulans in the plasmid pBARGPE1 and produce the plasmid pPtef. The DNA fragment containing the promoter and trpC terminator region was excised with SpeI/SmaI and inserted into the SpeI/EcoRV sites in the pPK2-Sur-GFP plasmid to produce pPK2-Ptef. The ORF of Aftf1 was cloned by PCR and excised with EcoRI/EcoRV, which was then inserted into the corresponding sites of pPK2-Ptef to produce pPK2-Ptef-Aftf1. pPK2-Ptef-Aftf1 was then transferred into A. tumefaciens and transformed into the WT M. robertsii. Overexpression of Aftf1 was confirmed by qRT-PCR as described below.

Gene cloning and gene expression analysis

RLM RT-PCR kit was used to clone intact mRNA with a 7-methyl guanosine cap structure (Roche).

For quantitative RT-PCR, complementary DNAs (cDNAs) were synthesized with total RNAs (800 ng per reaction) with ReverTra AceqPCR RT Master Mix (Toyobo, Japan). Each cDNA was diluted 20-fold, and used as a template for qRT-PCR analysis using Thunderbird SYBR qPCR Mix without ROX (Toyobo). The genes act and tef were used as internal standards50. The relative expression level of each gene was determined using the 2−ΔΔCt method51.

Northern blot analyses were conducted using standard methods52. The ORF of the Mr-OPY2 was used as the probe, which was labelled with DIG-DNA Labelling Kit (Roche). Northern blot and qRT-PCR analyses were repeated three times.

Yeast two-hybrid and autoactivation assays

Yeast competent cells were prepared using the Yeast-maker Yeast Transformation System 2 Kit (Clontech, USA), and yeast two-hybrid assays were conducted according to the manufacturer’s instructions (Clontech). The ORFs of the tested genes were cloned by PCR and inserted into the plasmid pGADT7 (Mr-OPY2, Mero-BCK1, Mero-STE11 and Mero-SSK2) or pGBKT7 (Mr-STE50). The ORFs in the plasmids were confirmed by sequencing. The plasmid pGBKT7-Mr-STE50 was transformed into Y2HGold cells, and other plasmids (from pGADT7) were transformed into Y187 cells. After mating, the resulting strains were grown on the medium (SD-His-Ade-Leu-Trp) with X-α-gal and AbA (Aureobasidin A) (Takara). The autoactivation of Mr-STE50 in pGBKT7 was tested using Y2HGold cells. Yeast two-hybrid and autoactivation assays were repeated three times.

Statistical analysis

Tukey’s honestly significant difference test in OriginPro 8.5 program was used in this study (OriginLab, USA)

Data availability

The nucleotide sequences of the long and short Mr-OPY2 transcripts were deposited in the Genbank nucleotide database with accession codes KY548479 and KY548480. The RNA-seq data were deposited in the Genbank database with accession codes SRP105041. All other relevant data supporting the findings of the study are available in this article and its Supplementary Information files, or from the corresponding author upon request.