Identification of DHOase in M. oryzae

By using the protein-protein basic local alignment search tool (BLAST) in the NCBI database (https://www.ncbi.nlm.nih.gov/), we found that the protein in M. oryzae encoded by MGG_12634 and the DHOase from the de novo pyrimidine nucleotide biosynthesis pathway in S. cerevisiae named Ura4 showed 43.67% in amino acid sequence homology. In addition, the homologous protein of the yeast Ura4 in M. oryzae was unique. According to the name of the enzyme in yeast, we named the corresponding homologous protein MoPyr4 in M. oryzae. Then, we analyzed the homology of DHOase in different fungi and constructed a phylogenetic tree using DNAMAN v.8 and MEGA 7.0.26 software. Homology analysis suggested that MoPyr4 in M. oryzae was genetically close to DHOase in most other fungi besides the model fungus S. cerevisiae, indicating the evolutionary conservation of DHOase (Figure S1A, B).

To study the biological function of MoPyr4 in M. oryzae, the MoPYR4 gene was knocked out using a high-throughput gene knockout strategy [30]. We constructed the pKO3A knockout vector with the hygromycin resistance gene HPH and transferred it into Guy11, the wild-type of M. oryzae, by Agrobacterium tumefaciens-mediated transformation (ATMT). When the target MoPYR4 gene was replaced with HPH in the Guy11 genome according to the homologous replacement principle, we confirmed the deletion of MoPYR4 by polymerase chain reaction (PCR) and quantitative real-time polymerase chain reaction (qRT-PCR) and finally obtained the ΔMopyr4 mutants (Figure S2A, B). Moreover, to confirm that the phenotypic and physiological differences shown by the ΔMopyr4 mutants were indeed caused by deletion of the MoPYR4 gene, we inserted the MoPYR4 gene into the genome of the ΔMopyr4 mutant via ATMT and obtained the complemented strain ΔMopyr4::MoPYR4.

MoPyr4 is required for the growth and conidiation of M. oryzae

To determine whether the absence of MoPyr4 affects fungal development, we first measured the colony diameter and found that the growth rate of the ΔMopyr4 strain on complete medium (CM) was significantly slower than that of the Guy11 and ΔMopyr4::MoPYR4 strains (Fig. 1A and D). Moreover, the ΔMopyr4 strain could not grow on basic minimal medium (MM) (Fig. 1A). In addition to the colony growth rate, the aerial mycelium of ΔMopyr4 was significantly sparser and fluffier (Fig. 1B). Next, we examined the conidia production of each strain. As shown in Fig. 1E, the conidiation of ΔMopyr4 was dramatically reduced, displaying an order of magnitude difference from that of Guy11 and ΔMopyr4::MoPYR4. The conidia produced by ΔMopyr4 were not greater than 3% of those produced by the wild-type and complemented strains [conidiation of Guy11 = (133.0 ± 13.9) × 104 conidia/mL, conidiation of ΔMopyr4 = (3.5 ± 0.9) × 104 conidia/mL, conidiation of ΔMopyr4::MoPYR4 = (153.3 ± 15.3) × 104 conidia/mL]. Consistent with this result, the conidiophores of ΔMopyr4 were sparser than those of Guy11 and ΔMopyr4::MoPYR4, and fewer conidia could be observed on the conidiophores of ΔMopyr4 (Fig. 1C). The above results indicate that MoPyr4 is essential for the growth and conidiation of M. oryzae.

Fig. 1

Hyphal growth and conidiation of the ΔMopyr4 mutant. (A) Hyphal growth of the Guy11, mutant, and complementary strains on CM and MM plates. (B) Growth of aerial hyphae of the Guy11, mutant, and complementary strains on CM plates. (C) Conidia and conidiophore growth of the Guy11, mutant, and complementary strains. Scale bar, 100 μm. (D) Statistical analysis of colony diameters of the Guy11, mutant, and complementary strains on CM plates. The data were calculated from three replicates. Asterisks indicate statistically significant differences (t-test, **** P < 0.0001). (E) Statistical analysis of the conidiation in the Guy11, mutant, and complementary strains. The data were calculated from three replicates. Asterisks indicate statistically significant differences (t-test, **** P < 0.0001)

MoPyr4 is required for the pathogenicity of M. oryzae

To further investigate the function of MoPyr4 in the pathogenicity of M. oryzae, we inoculated the three strains (Guy11, ΔMopyr4, and ΔMopyr4::MoPYR4) on two different host plants, barley and rice. First, the same-sized mycelial plugs were cut and inoculated on detached barley leaves. The disease spots on the barley leaves showed that Guy11 and ΔMopyr4::MoPYR4 both caused severe large brown lesions, and the areas around the lesions on the leaves turned yellow. However, ΔMopyr4 caused only small, weak lesions on the barley leaves, and leaf yellowing was not obvious (Fig. 2A and D). Next, the conidial suspensions were inoculated on isolated barley leaves for 4 days (d). The results showed that the disease spots on the barley leaves caused by Guy11 and ΔMopyr4::MoPYR4 were large and dark, with the leaves turning yellow, and the lesions of adjacent conidial droplets were almost connected to one piece. In contrary, the lesions caused by ΔMopyr4 could not expand from one conidial droplet to adjacent droplets, and the percentage of the lesion area was significantly smaller (Fig. 2B and E). To simulate the pathogenic process in the field more accurately, we sprayed a conidial suspension of the three strains on 2-week-old potted rice seedlings (susceptible Oryza sativa cv. CO-39). The typical rhombic fusion lesions were caused by the wild-type and complemented strains on the rice seedlings at 7 days post-incubation (dpi), and the leaves were yellow. In contrast, the lesions of the ΔMopyr4 mutant were significantly smaller and independent, and the leaves still looked green (Fig. 2C and F). Taken together, the above results indicate that MoPyr4 plays an important role in the virulence of M. oryzae on host plants.

Fig. 2

Pathogenicity of the ΔMopyr4 mutant. (A) Disease spots on detached barley leaves inoculated with mycelial plugs of the Guy11, mutant, and complementary strains. (B) Disease symptoms on isolated barley leaves inoculated with conidial suspensions (20 µL, 5 × 104 conidia/mL) of the Guy11, mutant, and complementary strains. (C) Disease symptoms of rice seedlings inoculated with conidial suspensions spray (2 mL, 5 × 104 conidia/mL) of the Guy11, mutant, and complementary strains. (D, E, F) Statistical analyses of the percentage of the lesion area per leaf caused by mycelial plugs (D), conidial suspensions (E), and conidial suspensions spray (F) of the Guy11, mutant, and complementary strains. The data in D and E were calculated from three replicates. The data in F were calculated from 15 replicates. Asterisks indicate statistically significant differences (t-test, *** P < 0.001, ** P < 0.01, **** P < 0.0001)

The addition of exogenous UMP restored the growth and conidiation of the ΔMopyr4 mutant

The enzymes of the de novo pyrimidine nucleotide biosynthesis pathway perform their catalytic functions in turn and first synthesize UMP [4, 5]; therefore, we wondered whether exogenous UMP addition could restore the phenotype of the ΔMopyr4 mutants. First, 5 mM UMP was added to the CM, and the colony diameter of each strain was measured at 7 dpi. The results showed that the growth rate of the ΔMopyr4 mutant on CM plates significantly increased to the level of the wild-type and complemented strains after the addition of exogenous UMP (Fig. 3A and B). Moreover, the growth of the ΔMopyr4 mutants was completely restored on the MM plates supplemented with 5 mM exogenous UMP (Fig. 3A and B). In addition, the aerial mycelia of ΔMopyr4 were no longer sparse or fluffy on the CM plates supplemented with UMP, and their growth was restored to levels consistent with those of Guy11 and ΔMopyr4::MoPYR4 (Fig. 3C). Next, we found that the conidiation of ΔMopyr4 recovered to the same level as that of Guy11 or ΔMopyr4::MoPYR4 after the addition of exogenous UMP (Fig. 3E). Consistent with the results of conidiation recovery, the growth of the conidiophores of ΔMopyr4 supplemented with exogenous UMP were consistent with the wild-type and complemented strains (Fig. 3D). These results show that the exogenous UMP restores the growth and conidiation of ΔMopyr4, and confirm that MoPyr4 plays an important role in the de novo pyrimidine nucleotide biosynthesis pathway.

Fig. 3

Complementation of the growth of the ΔMopyr4 mutant after supplementation with exogenous UMP. (A) Hyphal growth of the Guy11, ΔMopyr4 mutant, complementary strain, and the ΔMopyr4 mutant supplemented with 5 mM exogenous UMP on CM and MM plates. (B) Statistical analysis of colony diameters of the Guy11, ΔMopyr4 mutant, complementary strain, and the ΔMopyr4 mutant supplemented with 5 mM exogenous UMP on CM and MM plates. The data were calculated from three replicates. Asterisks indicate statistically significant differences (t-test, **** P < 0.0001). (C) Aerial hyphal growth in the Guy11, ΔMopyr4 mutant, complementary strain, and the ΔMopyr4 mutant supplemented with 5 mM exogenous UMP on CM plates. (D) Conidia and conidiophore growth of the Guy11, ΔMopyr4 mutant, complementary strain, and the ΔMopyr4 mutants supplemented with 5 mM exogenous UMP on CM plates. Scale bar, 100 μm. (E) Statistical analysis of conidiation in the Guy11, ΔMopyr4 mutant, complementary strain, and the ΔMopyr4 mutant supplemented with 5 mM exogenous UMP. The data were calculated from three replicates. Asterisks indicate statistically significant differences (t-test, *** P < 0.001, **** P < 0.0001)

The addition of exogenous UMP restored the pathogenicity and infection process of the ΔMopyr4 mutant

To further explore the reasons for the reduced pathogenic ability of ΔMopyr4, we observed the development of invasive hyphae in barley leaves, and analyzed the recovery of pathogenicity and the appressorium-mediated infection process of ΔMopyr4 with exogenous UMP.

We cut mycelia plugs of the same size from 7-day-old colonies of Guy11, ΔMopyr4::MoPYR4, and two types of the ΔMopyr4 mutants grown on CM plates and CM plates supplemented with exogenous UMP, and inoculated them on isolated barley leaves for 4 days. The ΔMopyr4 from the CM plate produced only weak disease spots, while the ΔMopyr4 from the CM + UMP plate led to severe brown lesions and leaf yellowing on the barley leaves, which was consistent with the findings for Guy11 and ΔMopyr4::MoPYR4, suggesting that the pathogenicity of ΔMopyr4 was restored by the exogenously added UMP (Fig. 4A). Next, in addition to the normal CM plates, we collected the conidia of the ΔMopyr4 strain on CM plates supplemented with exogenous UMP and diluted them using ddH2O supplemented with UMP. The conidial suspensions were then inoculated on isolated barley leaves for 4 days. As predicted, the pathogenic ability of the mutant conidia recovered to the level of that of the wild-type and complemented strains after exogenous UMP supplementation (Fig. 4B).

Fig. 4

Restoration of pathogenicity and penetration in the ΔMopyr4 mutant after supplementation with exogenous UMP. (A) Disease spots on cut barley leaves inoculated with mycelial plugs of the Guy11, ΔMopyr4 mutant, complementary strain, and the ΔMopyr4 mutant supplemented with 5 mM exogenous UMP. (B) Disease symptoms on isolated barley leaves inoculated with conidial suspensions (20 µL, 5 × 104 conidia/mL) of the Guy11, ΔMopyr4 mutant, complementary strain, and the ΔMopyr4 mutant supplemented with 5 mM exogenous UMP. (C) Penetration and growth of infectious hyphae in detached barley leaves inoculated with conidial suspensions (20 µL, 5 × 104 conidia/mL) of the Guy11, ΔMopyr4 mutant, complementary strain, and the ΔMopyr4 mutant supplemented with 5 mM exogenous UMP. ‘lt’ indicates long-term addition of UMP both to the CM plates and conidial suspensions. ‘st’ indicates short-term addition of UMP only to the conidial suspensions after washing from the CM plates without UMP. Scale bar, 20 μm. (D) Penetration assays on cut barley leaves were quantified and statistically analyzed. Scale bar, 20 μm

Moreover, we observed the formation of penetration pegs and the expansion of invasive hyphae inside detached barley leaves at 36 and 48 h after conidial inoculation under a microscope. To distinguish the degree of influence of the exogenous UMP on the infection ability of the mutants, the exogenous UMP addition was divided into long-term supplement (lt) and short-term supplement (st) groups (lt, the ΔMopyr4 mutants were cultured on CM plates supplemented with exogenous UMP before the conidia were washed off, and the exogenous UMP was added to the conidial suspension before inoculation onto the leaves; st, the ΔMopyr4 mutants were cultured on CM plates without exogenous UMP, and the exogenous UMP was added to only the conidial suspension before inoculation). As shown in Fig. 4C and D, the process of invasive hyphal formation and expansion was divided into four types (type 1, no invasive hyphae formed; type 2, only a single invasive hypha was formed; type 3, the invasive hyphae formed two or three branches; and type 4, the invasive hyphae produced multiple branches and expanded to adjacent cells). More than 90% of the appressoria of the wild-type and complemented strains formed invasive hyphae at 36 h post-inoculation (hpi), and the invasive hyphae branched and expanded rapidly. In contrast, the infection process of the appressoria of the ΔMopyr4 strain was extremely slow. At 48 hpi, approximately 70% of the invasive hyphae of Guy11 and ΔMopyr4::MoPYR4 were type 4, less than 20% were type 3, and very few appressoria were type 1. However, in the ΔMopyr4 mutant, approximately 80% of the invasive hyphae were type 1 or type 2, and only a few invasive hyphae branched into type 3 or type 4, showing that the infection process was still slow and lagging. These results suggest that the defects in the formation and expansion of invasive hyphae mediated by the appressorium were the main reason for the reduced pathogenicity of the mutant. Moreover, the addition of exogenous UMP, especially the long-term supplement, had a recovery effect on the infection process of the ΔMopyr4 mutant (Fig. 4C and D).

In conclusion, the ΔMopyr4 mutant has defects in development of invasive hyphae during infection, and it can be partly restored by addition of exogenous UMP.

The appressorium formation and appressorium turgor pressure were defective in the ΔMopyr4 mutant but were restored with exogenous UMP

The formation and internal turgor accumulation of the appressorium are the keys for M. oryzae to invade host plants. Therefore, the following experiments further explored the specific reasons for the decreased infection ability of the ΔMopyr4 mutant from these two aspects, and verified the recovery effects of exogenous UMP on the recovery of these defects.

To induce appressorium development, we collected the conidia of Guy11, ΔMopyr4::MoPYR4, and ΔMopyr4 grown on CM plates and ΔMopyr4 grown on CM + UMP plates and then inoculated them on artificial hydrophobic surfaces. The appressorium formation was observed at 4, 8, 12, and 24 hpi. The results showed that the appressorium formation rate of ΔMopyr4 was significantly lower than that of the wild-type and complemented strains at all time points, and at 24 hpi, almost all conidia of wild-type and complemented strains formed mature appressoria, while the appressorium formation rate of the mutant strain was less than 90% (Fig. 5A and B). Moreover, the lengths of the germ tubes of the Mopyr4 mutant were much longer when most of the appressoria were formed, and the conidia that failed to form appressorium showed abnormal growth of the germ tube and inability to expand the tip (Fig. 5A). Therefore, there was a defect in the appressorium formation of ΔMopyr4. In addition, the long-term exogenous addition of UMP effectively rescued the impairments in the appressorium formation of ΔMopyr4 (Fig. 5A and B).

Fig. 5

Restoration of appressorium formation and turgor pressure in the ΔMopyr4 mutant after supplementation with exogenous UMP. (A) Appressorium formation of the Guy11, mutant, mutant supplemented with 5 mM UMP, and complementary strains observed at different time points. Scale bar, 20 μm. (B) Statistical analysis of the appressorium formation rates of the Guy11, mutant, mutant supplemented with 5 mM UMP, and complementary strains at different time points. The data were calculated from three replicates. Asterisks indicate statistically significant differences (t-test, *** P < 0.001, **** P < 0.0001). (C) Immunoblotting of the phosphorylation level of Pmk1 in Guy11 and the ΔMopyr4 mutant. The phosphorylation level of Pmk1 was examined with an anti-P-Pmk1 antibody. The unphosphorylation level of Pmk1 was examined with an anti-Pmk1 antibody. The protein GAPDH was used as a loading control. The level of phosphorylated Pmk1 was calculated using the Formula P-Pmk1/Pmk1. (D) Appressorium collapse of the Guy11, mutant, mutant supplemented with 5 mM UMP, and complementary strains observed in 1 M, 2 M, or 3 M of glycerol solutions. Scale bar, 20 μm. (E) Statistical analysis of the appressorium collapse rates of the Guy11, mutant, mutant supplemented with 5 mM UMP, and complementary strains in 1 M, 2 M, or 3 M of glycerol solutions. The data were calculated from three replicates. Asterisks indicate statistically significant differences (t-test, ** P < 0.01, *** P < 0.001)

According to the above results, the ΔMopyr4 mutant exhibited defects in appressorium formation, penetration peg formation, and invasive hyphal expansion (Figs. 4C and 5A and B, and 4D). Considering the regulation of these three infection-related processes by the Pmk1-MAPK signaling pathway [19], the phosphorylation levels of Pmk1 were detected in Guy11 and ΔMopyr4. Figure 5C shows that the level of phosphorylated Pmk1 in the ΔMopyr4 strain was abnormally elevated, as determined by western blotting (WB), suggesting abnormal activation of the Pmk1-MAPK signaling pathway in the mutant strain.

Sufficient internal turgor pressure generated from glycerol accumulation in mature appressoria is a necessary condition for generating mechanical force to push the penetration peg through the host surface [17]. According to the excessive length of the germ tube of ΔMopyr4 during appressorium formation observed in Fig. 5A, we speculated that the transport efficiency of materials from conidia to the appressorium might be reduced, which could therefore affect turgor pressure accumulation inside the appressoria. Therefore, we performed an appressorium collapse experiment to measure the turgor pressure in the appressorium. As expected, the collapse rate of the appressoria of ΔMopyr4 was significantly greater than that of Guy11 and ΔMopyr4::MoPYR4 under different glycerol concentrations. The difference between the mutant and the wild-type strains was most significant under the 1 M glycerol treatment, as the collapse rate of the appressoria of ΔMopyr4 was approximately 14% greater than that of Guy11 (Fig. 5D and E). This result indicates that the internal turgor accumulation was impaired in the appressorium of ΔMopyr4. In addition, the long-term exogenous addition of UMP to ΔMopyr4 reduced the collapse rate of the appressorium to the same level as that in Guy11 and ΔMopyr4::MoPYR4 (Fig. 5D and E), suggesting a recovery of the defect in turgor pressure accumulation in the appressoria of ΔMopyr4.

Taken together, the ΔMopyr4 mutant has defects in appressorium formation and internal turgor pressure accumulation in the appressoria, and these can be restored by long-term addition of exogenous UMP.

The degradation and transport of glycogen and lipid droplets were defective in the ΔMopyr4 mutant but were restored with exogenous UMP

The accumulation of turgor pressure inside appressoria requires the autophagic degradation of glycogen and lipids and their successful transport from conidia to appressoria through germ tubes [31, 32]. Since the appressoria of the ΔMopyr4 mutant failed to accumulate sufficient internal turgor pressure (Fig. 5D and E), we continued to test whether there were impairments in the degradation and transport of glycogen and lipid droplets in the mutant. We induced the develoment of Guy11, ΔMopyr4, and ΔMopyr4::MoPYR4 appressoria on transparent hydrophobic slides and then used KI/I2 solution and the fluorescent dye boron dipyrromethene (BODIPY) to stain glycogen and lipid droplets, respectively, before observation under a microscope at 0, 8, 16, and 24 hpi. Figure 6A and B show that glycogen biosynthesis in the conidia of the mutant did not differ from that in the conidia of the wild-type. However, the glycogen in the conidia of ΔMopyr4 was significantly more than that of Guy11, and the glycogen in the appressoria of ΔMopyr4 was significantly less than that of Guy11 at 8 hpi. When glycogen in most of the conidia and appressoria of Guy11 were both degraded at 16 and 24 hpi, the glycogen degradation rates in the conidia and appressoria of ΔMopyr4 were both significantly lower (Fig. 6A and B, and 6C). These results revealed obvious defects in the degradation and transport of glycogen by ΔMopyr4 during the appressorium development. As shown in Fig. 6D and E, and 6F, compared with those of Guy11, the degradation and transport of lipid droplets in ΔMopyr4 also tended to be impaired. Moreover, long-term supplementation with exogenous UMP effectively reversed the above defects in ΔMopyr4 (Fig. 6).

Fig. 6

Restoration of glycogen and lipid droplet degradation and transport in ΔMopyr4 after the long-term addition of exogenous UMP. (A) Localization of glycogen in the conidia and appressoria of the Guy11, mutant, mutant supplemented with 5 mM UMP in the long term, and complementary strains at different time points during appressorium development. KI/I2 solution was used to stain glycogen at 0, 8, 16, and 24 hpi before observation. Scale bar, 10 μm. (B, C) Statistical analysis of the percentage of conidia (B)/appressoria (C) containing glycogen in the Guy11, ΔMopyr4, ΔMopyr4(+ UMP lt), and ΔMopyr4::MoPYR4 at different time points. The data were calculated from three replicates per treatment. At least 100 conidia/appressoria were counted per replicate. ‘ns’ indicates no statistically significant difference. Asterisks indicate statistically significant differences (t-test, ns P > 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001). (D) Localization of lipid droplets in conidia and appressoria of the Guy11, mutant, mutant supplemented with 5 mM UMP in the long term, and complementary strains at different time points during appressorium development. BODIPY was used to stain lipid droplets at 0, 8, 16, and 24 hpi before observation under a fluorescence microscope. Scale bar, 10 μm. (E, F) Statistical analysis of the percentage of conidia (E)/appressoria (F) containing lipid droplets in Guy11, ΔMopyr4, ΔMopyr4(+ UMP lt), and ΔMopyr4::MoPYR4 at different time points. The data were calculated from three replicates per treatment. At least 100 conidia/appressoria were counted per replicate. ‘ns’ indicates no statistically significant difference. Asterisks indicate statistically significant differences (t-test, ns P > 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001)

Taken together, there are delayed degradation and transport of glycogen and lipid droplets in the ΔMopyr4 mutant, and these could be rescued by long-term addition of exogenous UMP.

MoPyr4 responds to oxidative stress and colocalizes with peroxisomes

Reactive oxygen species (ROS) burst is an important strategy in plant immunity that not only directly inhibits the growth of pathogens but also serves as an important signal to trigger other immune responses. Therefore, pathogens need to scavenge ROS to ensure the growth and expansion of invasive hyphae in plants [33, 34]. Figure 4C and D show the lower infection rate of ΔMopyr4 in host plants and the defects in the expansion of invasive hyphae to adjacent cells (type 4). Therefore, we considered the possibility of defects in the ROS clearance ability of ΔMopyr4 during infection. As a result, we explored the sensitivity of the mutants to oxidative stress. We added paraquat (1 mM), hydrogen peroxide (H2O2, 5 mM), and rose bengal (RB, 50 µM) to CM plates and measured the inhibitory effects of these oxidative stress factors on the wild-type, mutant, and complemented strains. As shown in Fig. 7A and B, the relative inhibition rates of ΔMopyr4 on the H2O2- and RB-supplemented plates were significantly greater than those of Guy11 and ΔMopyr4::MoPYR4, indicating that the mutant was more sensitive to H2O2 and RB. Interestingly, ΔMopyr4 was significantly less sensitive to paraquat, and the agent had almost no inhibitory effect on the mutant, which was reversed for the other two agents. Taken together, these results indicate that the deletion of the MoPYR4 gene results in significant changes in the susceptibility of M. oryzae to various oxidative stresses.

Fig. 7

Hyphal growth of the ΔMopyr4 mutant on CM supplemented with oxidative stress factors and co-localization of MoPyr4 with peroxisomes. (A) Hyphal growth of the Guy11, mutant, and complementary strains on CM plates supplemented with 1 mM paraquat, 5 mM H2O2 and 50 µM RB. (B) Statistical analysis of relative growth inhibition rates of the Guy11, mutant, and complementary strains on CM plates supplemented with oxidative stress factors. The data were calculated from three replicates. Asterisks indicate statistically significant differences (t-test, *** P < 0.001, **** P < 0.0001). (C) Partly co-localization of MoPyr4-GFP with MoPts1-dsRed. Confocal fluorescence microscopy images (Zeiss LSM880, 63 × oil) of co-expressed dsRed-tagged protein and GFP-labeled protein were acquired. The overlapping fluorescence signals of GFP and dsRed in the merged image are framed with white borders and magnified in the magnified image, with white arrows denoting co-localization. Scale bar, 5 μm. (D) Line-scan graph showing the fluorescence intensities of green and red fluorescence signals, with black arrows denoting co-localized areas

Several reports have shown that one of the major metabolic functions of peroxisomes is ROS scavenging, which is very important for fungal infection [33,34,35]. To investigate the relationship between MoPyr4 and peroxisomes, we observed the conidial distribution of MoPyr4-GFP and MoPts1-dsRed-labeled peroxisomes in M. oryzae via confocal fluorescence microscopy. The fluorescence microscope showed that MoPyr4 and MoPts1 were both scattered throughout the cytoplasm of the conidia (Fig. 7C). Although there was no complete co-localization, a certain number of obvious co-localization sites between MoPyr4 and MoPts1 were found, showing a partial co-localization pattern and implying a close relationship between them. We amplified and pointed out the regions of overlapping fluorescence signals and made a line-scan graph representing the fluorescence intensity, as shown in Fig. 7C and D. Overall, MoPyr4 partly colocalizes with peroxisomes in M. oryzae.

MoPyr4 regulates the phosphorylation of Osm1 in response to hyperosmotic stress

In addition to oxidative stress, the accumulation of degradation products caused by plant cell death during pathogen invasion exposes fungi to an environment with increased osmotic pressure, so it is also critical for pathogenic fungi to recognize and respond to external hypertonic environments [36]. Therefore, we investigated the sensitivity of the mutant to hypertonic stress. We calculated the relative inhibition rates of the hypertonic stress factors sodium ion (NaCl, 0.6 M), potassium ion (KCl, 0.6 M), and sorbitol (1 M) on Guy11, ΔMopyr4, and ΔMopyr4::MoPYR4, and found that the ΔMopyr4 mutant was significantly more sensitive to high concentrations of all three hypertonic stress factors (Fig. 8A and B). These results indicate that MoPyr4 plays an important role in the adaptability of M. oryzae to hypertonic stress.

Fig. 8

Hyphal growth of the ΔMopyr4 mutant strain on CM supplemented with hyperosmotic stress factors and the phosphorylation level of Osm1 in the ΔMopyr4 mutant. (A) Hyphal growth of the Guy11, mutant, and complementary strains on CM plates supplemented with 0.6 M NaCl, 0.6 M KCl, and 1 M sorbitol. (B) Statistical analysis of the relative growth inhibition rates of the Guy11, mutant, and complementary strains on CM plates with hyperosmotic stress factors. The data were calculated from three replicates. Asterisks indicate statistically significant differences (t-test, ** P < 0.01, *** P < 0.001). (C) Immunoblotting of the phosphorylation level of Osm1 in the Guy11 and mutant strains. The vegetative hyphae of the Guy11 and mutant strains were grown in liquid CM supplemented with 0.6 M NaCl for 0, 30, 60, or 90 min for hyperosmotic stress induction. The phosphorylation level of Osm1 was examined with an anti-P-Osm1 antibody. The protein GAPDH was used as a loading control. (D) Changes in the relative phosphorylation level of Osm1 in Guy11 and the mutant over time. The relative level of phosphorylated Osm1 was calculated using the following formula: P-Osm1/GAPDH. The data were calculated from three replicates. ‘ns’ indicates no statistically significant difference. Asterisks indicate statistically significant differences (t-test, ns P > 0.05, * P < 0.1, *** P < 0.001)

The Osm1-MAPK signaling pathway in M. oryzae is homologous to the Hog1-MAPK signaling pathway in yeast, mediating the fungal response to hypertonic stress [37, 38]. We hypothesized that MoPyr4 is involved in the response of M. oryzae to hypertonic stress by regulating this pathway. Therefore, we detected differences in the phosphorylation levels of Osm1 in the mutant and wild-type strains under hypertonic conditions. The WB results showed that the uninduced vegetative hyphae of Guy11 maintained the phosphorylation of Osm1 at a relatively low level, and once transferred to media supplemented with 0.6 M NaCl to induce a hypertonic stress reaction, the phosphorylation level of Osm1 increased significantly within 30 min, and then decreased gradually within 1 h (Fig. 8C and D). There was no significant difference in the Osm1 phosphorylation level between ΔMopyr4 and Guy11 when the strains were not induced, but once induced by 0.6 M NaCl, the Osm1 phosphorylation level of ΔMopyr4 increased rapidly with increasing amplitude greater than that of Guy11 in the first 30 min and decreased slowly from 30 to 60 min. As a result, the phosphorylation levels of Osm1 in the mutant were significantly greater than those in the wild-type at both 30 and 60 min, suggesting that the Osm1-MAPK signaling pathway in the MoPYR4 deletion mutant was abnormally active under hypertonic stress (Fig. 8C and D). These results show that MoPyr4 regulates the adaptability of M. oryzae to hypertonic stress and the relative phosphorylation level of Osm1.

MoPyr4 interacts and partly co-localizes with MoAtg5

MoPyr4, encoded by the gene MGG_12634, was identified among the proteins enriched by immunoprecipitation of MoAtg5-GFP identified by mass spectrometry (MS) (Table S1), thus, the following experiments were performed to verify the association between MoPyr4 and MoAtg5, the core protein of the autophagy pathway. First, we performed a yeast two-hybrid assay and preliminarily tested the interaction between MoPyr4 and MoAtg5 (Fig. 9A). Next, although a co-immunoprecipitation (Co-IP) assay failed to verify the interaction in vivo, a GST pull-down experiment provided additional evidence for this interaction (Fig. 9B). To observe the close relationship between MoPyr4 and MoAtg5 more directly, we co-expressed the fusion expression cassettes MoPyr4-GFP and MoAtg5-mCherry in M. oryzae and observed the fluorescence signals in the conidia. As shown in the confocal fluorescence microscopy images, MoPyr4-GFP and MoAtg5-mCherry were dispersed in the conidial cytoplasm, and some punctate and clustered locations of MoAtg5 were highly correlated with those of MoPyr4. The merged image and the line-scan graph of fluorescence intensities displayed the partial co-localized sites of the two proteins (Fig. 9C and D). In summary, these results prove the interaction and partial co-localization between MoPyr4 and MoAtg5.

Fig. 9

The interaction and co-localization of MoPyr4 with MoAtg5. (A) Yeast two-hybrid assay for examining the interaction between MoPyr4 and MoAtg5. The pair of plasmids pGADT7-T and pGBKT7-53 were used as the positive control. (B) GST pull-down assay between MoPyr4 and MoAtg5. GST-MoPyr4 and Flag-MoAtg5 were expressed in vitro. The proteins were purified, incubated with GST beads, and eluted. GST-MoPyr4 was detected with an anti-GST antibody, and Flag-MoAtg5 was detected using an anti-Flag antibody via WB. (C) Partial co-localization of MoPyr4-GFP with MoAtg5-mCherry. Confocal fluorescence microscopy images (Zeiss LSM880, 63 × oil) of co-expressing MoAtg5-mCherry and GFP-labeled MoPyr4 were captured in conidia. The overlapping fluorescence signals of GFP and mCherry in the merged image are framed with white borders and magnified, with white arrows denoting co-localization. Scale bar, 5 μm. (D) Line-scan graph showing the fluorescence intensities of green and red fluorescence signals, with black arrows denoting co-localized areas

MoPyr4 positively regulates autophagic degradation

To further explore how MoPyr4 regulates autophagy in M. oryzae, we examined the level of autophagic degradation in ΔMopyr4 and Guy11 under starvation conditions. During autophagic degradation, autophagosomes fuse with the vacuole, and the GFP-Atg8 labeled on the inner membranes of autophagosomes enters the vacuole, in which the Atg8 part of the fusion protein can be degraded by hydrolases, while the cleaved GFP is relatively resistant to hydrolysis and thus cannot be degraded. As a result, changes in the levels of free GFP and GFP-Atg8 are widely used to reflect intracellular autophagic affluxes, and GFP-Atg8 has become a well-used marker for autophagy activity detection [39, 40]. At present, the conserved function of GFP-Atg8 in M. oryzae has been confirmed [39, 40]. Therefore, the vegetative hyphae of the mutant and wild-type strains expressing GFP-MoAtg8 were precultured in liquid CM for 48 h and then transferred to liquid synthetic defined medium without amino acids and ammonium sulfate (SD-N) for starvation induction for 0, 3, or 6 h. First, we collected the vegetative hyphae at different time points to extract total protein for WB. After measuring the efficiency of free GFP cleavage from GFP-MoAtg8 in Guy11, we found t