MoMih1 is indispensable for asexual development, cell wall integrity, and pathogenicity of Magnaporthe oryzae

Asexual spore serves as essential inoculum of rice blast during the disease cycle, and differentiation of young conidium from conidiophore is intimately regulated by cell cycle. Mih1 encodes a dual-specificity phosphatase that involved in the G2/M transition of the mitotic cell cycle by regulating the Cdk1 activity in eukaryotes. Till now, the roles of Mih1 homologue, however, remain unclear in Magnaporthe oryzae. We here functionally characterized the Mih1 homologue MoMih1 in M. oryzae. MoMih1 is localized to both the cytoplasm and nucleus and can physically interact with the CDK protein MoCdc28 in vivo. Loss of MoMih1 led to delayed nucleus division and a high level of Tyr15 phosphorylation of MoCdc28. The MoMih1 mutants showed retarded mycelial growth with a defective polar growth, less fungal biomass, and shorter distance between diaphragms, compared with the KU80. Asexual reproduction altered in MoMih1 mutants, with both abnormal conidial morphogenesis and decreased conidiation. The MoMih1 mutants severely attenuated the virulence to host plants due to the impaired ability of penetration and biotrophic growth. The incapability of scavenging of host-derived reactive oxygen species, which was possibly ascribed to the severely decreased extracellular enzymes activities, were partially associated with deficiency of pathogenicity. Besides, the MoMih1 mutants displayed also improper localization of retromer protein MoVps26 and polarisome component MoSpa2, and defects of cell wall integrity (CWI), melanin pigmentation, chitin synthesis, and hydrophobicity. In conclusion, our results demonstrate that MoMih1 plays pleiotropic roles during fungal development and plant infection of M. oryzae.


Introduction
Rice blast, caused by the fungal pathogen Magnaporthe oryzae (synonym to Pyricularia oryzae), is one of the most destructive diseases that seriously threatens the rice production worldwide (Wilson and Talbot, 2009). The blast fungus differentiates threecelled pyriform spores as the primary inoculum, and their dispersal is essential for the epidemiology of this disease (Fernandez and Orth, 2018). The blast disease begins at the attachment of those asexual spores to the susceptible rice leaves, and under permitted condition, they germinate and develop a specialized dome-shape appressorium at the tip of germ tube, which helps to rupture the rice leaf cuticle cells (Sabnam and Barman, 2017). After penetration, the invasive hyphae grow biotrophically in the initially infected plant cells and thereafter followed by a necrotrophic phase, which resulted in cell death and appearance of disease lesions on the plants (Ebbole, 2007;Zhang et al., 2016).
The development of both pyriform spores and appressorium is one of the most complicated morphological and physiological events (Zhou et al., 2009;Liu et al., 2010;Oses-Ruiz et al., 2021). The switches from conidiophore growth to young conidium differentiation, and germ tube to melanized appressorium are intimately linked to cell cycle regulation in M. oryzae (Liu et al., 2010;Saunders et al., 2010). During conidiation, two rounds of mitosis and cytokinesis have occurred to develop the mature pyriform conidia, in which each cell contains a single nucleus (Liu et al., 2010;Li et al., 2018). Blocking of Dam1, a subunit of DASH complex involved in cell cycle, delayed the mitotic progression and impaired sporulation under permitted conditions (Shah et al., 2019). Cdc14 is a conserved dual-specificity phosphatases that directly accelerates the cytokinesis during cell division (Meitinger et al., 2012). In M. oryzae, the CDC14 homologue is essential for mitosis and cytokinesis during asexual spore differentiation, and its disruption results in defective mitotic exit, and incorrect septum formation during conidium and appressorium development in M. oryzae (Li et al., 2018). Besides, the Pombe Cdc15 homology (PCH) MoCDC15, which is required for cytokinesis by directly regulating the formation of actomyosin ring, has also been identified for asexual spore development and plant infection of M. oryzae (Goh et al., 2011). During infection process, appressorium mediated plant infection, which is crucial for the full virulence of M. oryzae, is tightly regulated by the cell cycle checkpoint (Saunders et al., 2010). Mutation of two checkpoints such as NIM1 and NimA could prevent the fungal infection of rice plants (Saunders et al., 2010). What's more, the block of cell cycle by prohibiting S-phase checkpoint or mitotic exit also prevents appressorium-mediated plant infection, further highlighting the essential role of cell cycle regulation in determining the full virulence of M. oryzae (Saunders et al., 2010;. Though successful proceeding of cell cycle is critical for the development of asexual spores and infection structures, the precise regulatory mechanism of the cell-cycle check point controlled by the special regulator, have yet to be definitively understood in M. oryzae. Protein phosphorylation and dephosphorylation are two most important post-translational modifications that regulate the cellular processes, such as cell cycle, DNA replication, and gene transcription, and have mainly been regulated by protein kinase and protein phosphatase in eukaryotes (Zhao et al., 2007;Li et al., 2012). Dualspecificity phosphatases are recognized to dephosphorylate tyrosine and serine/threonine residues within the substrate, and thus control a serial of fundamental physiological processes such as cell growth, differentiation, and apoptosis (Tonks, 2006). Mih1 encodes a dualspecificity phosphatase that activates the enzymes activity of cyclindependent protein kinase (Cdk1) by removing the phosphate group on the catalytic subunit Cdc28, then promotes G2/M transition of mitotic cell cycle (Camps et al., 2000;Pal et al., 2008). In fungi like Schizosaccharomyces pombe, Ustilago maydis and Aspergillus nidulans, the Cdc25 encodes a dual-specificity protein phosphatase that is essential for the G2/M phase transition, and the disruption of Cdc25 makes the mutant postpone entry into mitosis and develop abnormally large cell (Lew and Kornbluth, 1996;Sgarlata and Perez-Martin, 2005). However, in budding yeast Saccharomyces cerevisiae, inactivation of Mih1, a homolog of S. pombe Cdc25, could only lead to a moderate delay of mitotic entry, which is inconsistent with the previous findings in other fungi, suggesting a potential exist of redundancy of the dephosphorylation of Cdk1 in different eukaryotes (Pal et al., 2008). In addition, Mih1 has been confirmed to control the phosphorylation state of Vps26, a subunit of retromer subcomplex that directly recognizes the sorting signal in cargo transport, and mutation of Mih1 lead to decreased affinity of retromer to the targets, defective polar growth and pathogenesis of the fungi (Riquelme, 2013;Zheng et al., 2015;Cui et al., 2017;Qu et al., 2022). Taken together, the cell cycle was differentially regulated in the fungi when responses to diverse environmental stimuli, and understanding the roles of dual-specificity phosphatase Cdc25 would provide novel clues to insight of regulatory mechanism of the cell cycle in eukaryotes.
In this study, we report a crucial role for a putative cell division control protein Cdc25 which shows high similarity to the S. cerevisiae Mih1 during fungal development and pathogenesis in M. oryzae. Our results demonstrate that the loss of MoMih1 led to delayed nucleus division and a high level of Tyr15 phosphorylation of MoCdc28. The MoMih1 mutant was severely reduced in polar growth and asexual reproduction. The virulence of MoMih1 mutants were significantly attenuated due to the impaired ability of penetration and biotrophic growth. The MoMih1 mutants displayed also improper localization of retromer protein MoVps26 and polarisome component MoSpa2, and defects of cell wall integrity (CWI), melanin pigmentation, chitin distribution, hydrophobicity, and extracellular enzymes activities. Taken together, our results indicate that MoMih1 is an important regulator required for the fungal development and plant infection of M. oryzae.

Materials and methods
Strains, cultural conditions, and phenotype assays The wild-type strain Guy11 was used the parental strain to delete the MoGin4 and MoCyc2 gene, while the KU80 strain was used as the parental strain to disrupt the MoMih1 gene in M. oryzae (Villalba et al., 2008). Mycelia used for microscopic examination, DNA and RNA isolation, protein extraction were cultured in liquid CM at 28°C for 24 to 48 h. To assay vegetative growth, mycelial plugs of KU80, MoMih1-41, MoMih1-44, and MoMih1c were respectively inoculated on complete medium (CM), minimal medium (MM), oatmeal medium (OM), and rice straw decoction and corn medium (RDC) (Qi et al., 2012), and cultured at 28°C in darkness for 5 d. To measure fungal biomass, four mycelial plugs of tested strains were cultured in 150 mL liquid CM at 28°C for 3 d, then froze dry to measure dry weight.
Conidia suspension of KU80, MoMih1-41, MoMih1-44 and MoMih1c were dropped on hydrophilic coverslips to develop mycelium, then stained with CFW to examine septum formation. For melanin pigmentation, all strains were cultured on CM for 5 d and/or in liquid CM for 7 d at 28°C in darkness. To investigate asexual reproduction, all strains were cultured on RDC medium, and the conidia were collected, and counted as the reference described (Guo et al., 2017). Conidial morphology and size, and the average numbers of septa per conidia were recorded using a light microscope. To evaluate the responses to different stressors, all strains were inoculated on CM plates supplemented with or without 5 mM H 2 O 2 , 200 mg/mL CFW, 200 mg/mL CongoRed, and 5 mM caffeine, respectively, and then cultured at 28°C for 5 d. To detect CWI, equal amount of mycelia of tested strain was treated with 7.5 mg/mL lysing enzyme, and the release of protoplasts was recorded after enzyme treatment for 30, 60, and 90 min.

Phylogenetic analysis
To investigate the phylogenetic relationship among the orthologues involved in the cell cycle in different species, the amino acid sequences of each orthologue in S. cerevisiae were used as a prey to search the NCBI database, all the orthologous proteins in diverse species were downloaded, and aligned using the Muscle (Edgar, 2004). The aligned sequences were imported into software IQ-TREE 2, and thereafter phylogenetically analyzed to evaluate the evolutionary relationship among those orthologues using the maximum likelihood method (Minh et al., 2020).

Targeted gene deletion and complementation
The 1200 bp DNA fragments, which respectively flanked the upstream and downstream of the MoMih1 gene, were amplified from the M. oryzae genomic DNA with primers MoMih1-1F/ MoMih1-2R and MoMih1-3F/MoMih1-4R (Table S1). The hygromycin B cassette (HPH) was amplified from plasmid pCB1003 with primers HPH-F/HPH-R (Table S1). The above DNA fragments were fused to the enzymes digested (XbaI and HindIII) pKO1B vector (Lu et al., 2014) using the gap repair system in yeast FY834. The resultant plasmid pKO1B-MoMih1-HPH was transformed into Agrobacterium tumefaciens strain AGL1, and the MoMih1 deletion mutants was generated using the A. tumefaciensmediated transformation (ATMT) of M. oryzae. The putative MoMih1 mutants were further verified by PCR, RT-PCR, and Southern blot.
To generate complemented strain, a 3030 bp DNA fragment containing a native promoter (1180 bp) and a full length of MoMih1 coding region (1850 bp) was amplified using primers Mih1pF/ Mih1pR (Table S1) and transformed into yeast strain XK125 together with XhoI linearized pYF11. The complemental vector pYF11-MoMih1 was introduced into the MoMih1-44 mutant by PEG-mediated transformation of M. oryzae. Both MoGin4 (MGG_02810) and MoCyc2 (MGG_07065) deletion mutants were generated following the above-described methods.

Localization of MoMih1 in M. oryzae
To observe the localization of MoMih1 in M. oryzae, the full length of MoMih1 coding region (1853 bp) was amplified with primers RP27Mih1pF/Mih1pR (Table S1) and ligated with XhoI linearized pYF11. The resultant plasmid pYF11-MoMih1-eGFP was transformed into Guy11. To visualize the colocalization of MoMih1 with the nucleus signal, the plasmid containing H1-mRFP fusion gene was transformed into strain expressing MoMih1-eGFP. The derivative strains, which expressed MoMih1-eGFP or both MoMih1-eGFP and H1-mRFP, were examined under a confocal laser scanning microscope (CLSM).

Localization of MoVps26 and MoSpa2 in MoMih1 mutant
The full length of MoVps26 coding region (1098 bp) was amplified from genomic DNA of Guy11 with primers Vps26pF/Vps26pR (Table S1), and then ligated to XhoI digested pYF11 following the above-described procedures. The plasmid pYF11-MoVps26-eGFP was respectively introduced in KU80 and MoMih1-44 mutant, and then hyphal tips of each strain expressing MoVps26-eGFP fusion protein were examined under CLSM. To evaluate whether the microtubuledestabilizing drug affected the localization of MoVps26-eGFP, the above transformants were cultured in liquid CM with or without 0.4 mg/mL benomyl, and then hyphal tips of them were examined under CLSM.
To determine the localization of MoSpa2 in MoMih1 mutant, the full length of MoSpa2 coding region (3.0 kb) was cloned with primers Spa2pF/Spa2pR (Table S1) and ligated into pYF11 as the procedures described above. The plasmid pYF11-MoSpa2-eGFP was then introduced into KU80 and MoMih1-44 mutant, respectively, and hyphal tips of the derivative strain expressing MoSpa2-eGFP were examined under CLSM.

Plant infection
To test pathogenicity, conidial suspensions (1 × 10 4 spores/mL) of KU80, MoMih1-41, MoMih1-44 and MoMih1c were dropped on barley leaves (Hordeum vulgare cv Golden Promise; one-week-old), and/or sprayed on rice seedlings (Oryza sativa cv CO-39; two-week-old), respectively. The fungi-inoculated plants were placed in a moist chamber at 28°C in dark for 24 h, and then shifted to a photoperiod of 12 h light for another 96 h. To examine fungal infection, conidial suspension (1 × 10 4 spores/mL) was dropped on lower epidermis of barley leaves, cultured at 28°C for 24 h and/or 48 h, and then invasive hyphae were viewed under light microscopy. The DAB staining was performed following the previous methods (Guo et al., 2017).

Chitin content determination
The chitin contents of KU80, MoMih1-41, MoMih1-44 and MoMih1c were determined following the previous method (Yin et al., 2016). 5 mg freeze-dried fungal powder was mixed with 1 mL 6% KOH solution, water bathed at 80°C for 90 min, and then centrifuged (16 000g, 10 min). The sediment was washed by 1 mL PBS for three times, resuspended in 0.5 mL Mcilvine's buffer (pH 6), and then added 100 mL chitinase (C6137, Sigma-Aldrich, Shanghai, China) in suspension. After reacting in water bath at 37°C for 16 h, 100 mL sodium borate solution (0.27 M, pH 9) was added in the tubes, then boiled at 100°C for 10 min. 1 mL diluted Ehrlich's reagent solution (10-folds) was finally added to the reaction tubes, and then incubated at 37°C for 20 min. The absorbance value of each sample was determined at OD 585 . A standard curve was prepared using the D-GlcNAc (A8625, Sigma-Aldrich, Shanghai, China). This experiment was repeated three times.

Extracellular enzyme activities assay
The strains of KU80, MoMih1-41, MoMih1-44 and MoMih1c were respectively inoculated on CM plates with or without 0.2 mM ABTS (2,2'-Azinobis-(3-ethylbenzthiazoline-6-sulphonate)), placed in chamber at 28°C for 24 h, and then recorded the purple hydrolysis circle on the media. In liquid CM, extracellular enzyme activities, including the laccases and peroxidases, were determined following the previous procedures (Chi et al., 2009). These experiments were repeated three times.

Quantitative real-time PCR
The total RNA of each sample was extracted and reversely transcribed following the previous procedures (Guo et al., 2016). The expression of target genes involving in melanogenesis, conidiation, hydrophobicity or genes encoding peroxidase, laccase, chitin synthase were detected using the BIO-RAD CFX96 touch qRT-PCR system (Bio-Rad, Hercules, California, USA) according to the established procedures (Guo et al., 2016). All the primers used here are listed in Table S1.

Assays for nuclear distribution and autophagy
To determine the roles of MoMih1 on cytokinesis, the vector pMoC H245Y -H3-mRFP was respectively transformed in KU80 and MoMih1-44 mutant via ATMT to generate KU80-H3-mRFP and MoMih1-H3-mRFP strains. To visualize nuclear distribution, both mycelia and conidia of each tested strain were cultured as described above, stained with CFW, and then visualized by CLSM.
To determine the roles of MoMih1 during autophagy, the vector pMoC H245Y -eGFP-ATG8 was transformed into KU80 and MoMih1-44 mutant, respectively. The derivative strains of KU80-eGFP-ATG8 and MoMih1-eGFP-ATG8 were respectively grown in liquid CM at 28°C for 36 h, then transferred to nitrogen starvation (SD-N) medium for another 2 or 4 h. The eGFP-MoAtg8 fusion protein and free GFP were detected using the anti-GFP antibodies, and analyzed by the ImageJ software to calculate the ratio of free GFP and fusion eGFP-MoATG8 (Zhu et al., 2018).

Yeast two-hybrid assay
The full-length of the coding sequences of MoMih1 and MoCdc28 were cloned, and inserted into the bait construct pGADT7 (630442, Clontech, Dalian, China) and prey construct pGBKT7 (630443, Clontech, Dalian, China), respectively. Both the resultant bait construct pGADT7-MoMih1 and prey construct pGBKT7-MoCdc28 were co-transformed into yeast strain AH109 (MATa,112,, and proteins interaction was validated following the described methods (Zheng et al., 2021a).

BiFC assay
The MoMih1-C YFP and MoCdc28-N YFP fusion constructs were generated by ligating the coding sequences of MoMih1 and MoCdc28, which were amplified with primers pHZ68Mih1F/ pHZ68Mih1R and pHZ65Cdc28F/pHZ65Cdc28, into pHZ68 and pHZ65, respectively. The vector pairs of MoMih1-C YFP and MoCdc28-N YFP, MoMih1-C YFP and N YFP, C YFP and MoCdc28-N YFP, and C YFP and N YFP were simultaneously transformed into Guy11. YFP signals were examined under the CLSM. The primers are listed in Table S1.

Assay for Cdc28 phosphorylation
To determine the phosphorylation level of Cdc28, total protein of the tested strains was separated by 12% SDS-PAGE and immunoblotted with an anti-phospho-Cdc2 (Tyr 15 ) antibody (Cell Signaling Technology). The anti-Cdc2 antibody (Santa Cruz Biotechnology) was used as a negative control.

Results
Identification and deletion of MoMih1, MoGin4 and MoCyc2 gene in M. oryzae The proceeding of cell cycle is essential for the precise initiation of conidiophore and appressorium, and is accurately regulated by the cell cycle-related genes in filamentous fungus (Saunders et al., 2010;Yue et al., 2017). In this study, a family of cell cycle-related proteins was identified and phylogenetically grouped into 17 clades ( Figure 1A). Previously, two cell cyclerelated genes, including MoCks1 (Cks1 family) and CHM1 (CLA4 family), were individually deleted and functionally characterized in M. oryzae (Li et al., 2004;Yue et al., 2017). In this study, to investigate the roles of the rest 15 cell cycle-related genes in M. oryzae, a highthroughput gene deletion system (Lu et al., 2014) was used following the homologous gene disruption strategy ( Figure S1A). At present, three of them, including MoMih1 ( The orthologues of cell cycle and morphogenetic proteins in M. oryzae. (A) Phylogenetic analysis and classification of cell cycle and morphogenetic related genes in diverse species. IQ-TREE 2 was used to establish a phylogenetic tree based on the full length of these proteins using the maximum likelihood method. (B) The mycelial growth of the tested strains. The colony diameters of the tested strains were assayed on MM medium at 5 dpi.
The pathogenic assay of the tested strains. Mycelial plugs of the tested strains were inoculated on 7-day-old barley leaves, and the diseased leaves were recorded at 5 dpi.

MoMih1 is required for vegetative growth of M. oryzae
To validate the biological role of MoMih1 in M. oryzae, a complemental strain MoMih1c was subsequently generated by reintroducing the full length of MoMih1 into MoMih1-44 mutant, and verified by RT-PCR ( Figure S1C). The mycelial growth rate of MoMih1 mutants were significantly reduced in comparison to the KU80 and MoMih1c (Figures 3A, B). In liquid CM, mycelial pellets of MoMih1 mutants were much smaller but denser than that of KU80 and MoMih1c ( Figure 3C). The average length of sub-apical cell of the MoMih1 mutants was approximately 50% shorter than that of KU80 and MoMih1c ( Figures 3D, E). Besides, the MoMih1 mutants significantly decreased the fungal biomass compared with that of KU80 and MoMih1c ( Figure 3F). These results indicate that MoMih1 is required for the vegetative growth of M. oryzae.

MoMih1 is required for conidiation and conidial morphology of M. oryzae
The aerial hyphae growth and conidiophore development were compared, and the MoMih1 mutants developed less fluffy aerial hyphae than that of KU80 and MoMih1c ( Figure 4A). Microscopic examination showed that most of the mutants develop abnormal conidiophores, bearing significantly less conidia than that of KU80 and MoMih1c ( Figure 4B). The conidial production of MoMih1 mutants was apparently decreased compared with KU80 and MoMih1c ( Figure 4C). To further understand the potential role of MoMih1 on conidiation, the expression of sporulation-related genes was investigated, and all of them significantly downregulated in the MoMih1 mutant, compared with that in KU80 strain ( Figure 4D). Besides, the MoMih1 mutants showed abnormal conidial morphology ( Figure 4E), with evidently increased conidial length, compared with KU80 and MoMih1c ( Figure 4F). Meanwhile, the MoMih1 mutants developed a greater number of one or two-celled conidia, compared with that by KU80 and MoMih1c ( Figure 4G).

MoMih1 is required for the pathogenicity of M. oryzae
The pathogenicity of KU80, MoMih1 mutants and MoMih1c was conducted on both rice and barley leaves. The MoMih1 mutants significantly attenuated the virulence on host plant leaves, compared with KU80 and MoMih1c ( Figures 5A, B). The infection ability of the MoMih1 mutants was impaired, with most of them exhibiting incapability of rupturing the barley leaf epidermis cells, compared with KU80 and MoMih1c ( Figure 5C). The penetration rate of MoMih1 mutants was approximately 60% less than that of by KU80 and MoMih1c ( Figure 5D). At 48 hpi, the invasive hyphae of the MoMih1 mutants were mostly restricted in the initially invaded cells, in contrast to the free expanding growth

MoMih1 is required for scavenging of host-derived ROS during infection
Successful removal of host-derived reactive oxygen species (ROS) from plant infection site is an essential prerequisite to ensure the infection of M. oryzae (Guo et al., 2010;Guo et al., 2011;Wang et al., 2017). The sensitivity of the MoMih1 mutants to hydrogen peroxide (H 2 O 2 ) was firstly verified, with all of them showing incremental sensitivity to H 2 O 2 , compared to KU80 and MoMih1c ( Figures S5A, B). Since the MoMih1 mutants developed a great amount of tiny necrotic lesions on the rice leaves ( Figure 5A), the ROS burst at the infection site was thus monitored among the tested strains. An average 51.80 ± 3.34% of the plant cells infected by MoMih1 mutant could be stained by DAB, compared with 3.71 ± 1.28% and 2.00 ± 1.79% of the KU80 and MoMih1c, respectively (Figures 5F, G), suggesting the critical role of MoMih1 in detoxifying the host-derived ROS during plant infection. The potential reason for decreased ability to scavenge host-derived ROS was thereafter explored, and the extracellular enzyme activity, including peroxidases and laccases, was almost abolished in MoMih1 mutants, compared with KU80 and MoMih1c ( Figures  S5C, D). The transcriptions of five putative peroxidase and laccase encoding genes were significantly down regulated in the MoMih1 mutants, compared with KU80 ( Figure S5E).

MoMih1 is required for CWI in M. oryzae
Intact cell wall of M. oryzae is a guarantee for the proper infection on rice plants (Feng et al., 2021), the sensitivity of MoMih1 mutants to different cell wall stressors was therefore tested. The MoMih1 mutants are more sensitive to those stressors than that of KU80 and MoMih1c (Figures 6A, B). Besides, the protoplasts released by the MoMih1 mutants were also significantly increased compared with KU80 and MoMih1c at each given time ( Figure 6C), implying a severe defect of CWI in MoMih1 mutants.
Chitin, one of the main components of fungal cell wall, is essential for the CWI of M. oryzae (Yin et al., 2016;Guo et al., 2017). After staining with CFW, the blue fluorescence was mostly The colony diameter of the tested strains grown on four artificial media were measured at 5 dpi, and then statistically analyzed. The error bars represent the SD of three replicates, whereas "**" represents a significant difference (p < 0.01). congregated at the hyphal tip of KU80 and MoMih1c, however, patches of bright fluorescence were unevenly distributed on the hyphal wall but not the tips of the MoMih1 mutants ( Figure 6D). Quantitative detection of chitin contents showed, interestingly, that chitin contents of the MoMih1 mutant were significantly increased, compared with that of KU80 and MoMih1c ( Figure 6E). The expression of seven chitin synthase genes was assayed, with MoChs1 and MoChs2 showing increased expression, but MoChs5 and MoChs7 displaying decreased expression, compared with that of in KU80 ( Figure 6F).

MoMih1 is essential for pigmentation and hydrophobicity in M. oryzae
The MoMih1 mutants were defective in melanin pigmentation compared with KU80 and MoMih1c ( Figure 7A). The melanin biosynthesis related genes, such as MoRSY1, MoALB1, MoBUF1 and MoHNR1, were significantly downregulated in the MoMih1 mutants compared with that in KU80 ( Figure 7B). Moreover, the MoMih1 mutants were more sensitive to tricyclazole, a specific inhibitor of melanin biosynthesis ( Figures 7C, D), implying that MoMih1 was required for melanin biosynthesis in M. oryzae.
The hydrophobicity was determined on the fungal culture of KU80 and MoMih1 mutants, respectively. The water droplets remained on the mycelial culture of KU80 and MoMih1c after 24 hpi, in contrast to the immediate soaking into the mycelial cultures by the MoMih1 mutants ( Figure S6A). Due to the wettable phenotype of the mutants, the transcription of hydrophobinencoding genes including MPG1, MHP1, and two MHP1 homologs (MGG_09134 and MGG_10105) was verified, and all of them showed remarkably down-regulated expression in MoMih1 mutant, compared with that of KU80 ( Figure S6B).

MoMih1 is required for polarized growth in M. oryzae
In view of the retarded growth of the MoMih1 mutants, the polar growth was thus examined. Approximate 40% mutant mycelia displayed abnormally swollen tips, compared with KU80 ( Figures 8A, B). When stained with CFW, the blue fluorescence was mostly converged at the mycelial tips of KU80 and MoMih1c, but not the swollen tips of MoMih1 mutants ( Figure 8C), implying an inactive chitin synthesis in the mycelial tips of MoMih1 mutant. Spa2 is a component of the polarisome and functions in regulating hyphal morphology and extension during fungal polar growth (Virag and Harris, 2006), therefore, the MoSpa2-eGFP fusion protein were examined in KU80 and MoMih1-44, respectively. The MoSpa2 can accurately localize to the normal growing hyphal tip of KU80, but not the swollen tips of the MoMih1 mutant ( Figure 8D), suggesting MoMih1 is required for polarisome formation during the polarized growth of M. oryzae.

MoMih1 is required for subcellular localization of MoVps26 in M. oryzae
In S. cerevisiae, Mih1 directly regulates the phosphorylation of Vps26, a subunit of the retromer subcomplex, and thereby modulates the binding affinities of the retromer to specific cargo (Cui et al., 2017), we thus examined the cellular location of MoVps26 in the MoMih1 mutant. A dispersive distribution of the MoVps26-eGFP was observed in the hyphal cytoplasm of MoMih1 mutant, compared with the mostly punctate structures in strain KU80 ( Figure 9A). The immunoblotting with anti-GFP showed that the proteolysis of MoVps26-eGFP has increased in the MoMih1 mutant ( Figure 9B). The mobility of MoVps26-eGFP depends on microtubules, and could be prevented by the microtubuledestabilizing agent in M. oryzae (Zheng et al., 2015). When treating with benomyl, the MoMih1 mutant showed both obviously increased sensitivity to fungicide and convergent localization of MoVps26-eGFP nearby the hyphal tips, compared with that of KU80 ( Figures 9C, D, E). Besides, the ratio of the hyphae containing the swollen tips has remarkably increased in the MoMih1 mutants ( Figure 9F), compared with control strain.

MoMih1 regulates the MoCdc28 phosphorylation in M. oryzae
In S. cerevisiae, protein tyrosine phosphatase Mih1 control the cell cycle progress through affecting the phosphorylation levels of Cdc28. In M. oryzae, a putative catalytic subunit of Cdk1 protein named as MoCdc28 (MGG_01362) was identified to directly interact with MoMih1 based on yeast two-hybrid (Y2H) and BiFC assay (Figures 10A, B). Moreover, western blot analysis with anti-phospho-Cdk1 antibody also revealed that the Cdk1 phosphorylation signal in the MoMih1 mutant was much stronger than that in the KU80 (Figure 10C), suggesting an essential role of MoMih1 in removing the phosphate group on MoCdc28 during fungal development in M. oryzae.

MoMih1 is required for nuclear segregation in M. oryzae
To determine whether MoMih1 affects the cell cycle of M. oryzae, nuclear segregation was examined in the strains KU80-H3-mRFP and MoMih1-H3-mRFP. Most of the nuclei showed delayed separation in the mycelium of MoMih1 mutant, compared with KU80 ( Figure 11A). However, nuclear segregation seems normally proceed in the MoMih1 mutant during the appressorium formation, which is similar with the wild type strain KU80-H3-mRFP ( Figure 11B). affecting the fungal autophagy, the autophagy was compared between the MoMih1 mutant and KU80. When grown in CM, both the KU80-eGFP-Atg8 and MoMih1-eGFP-Atg8 strains, which expressed eGFP-MoAtg8 fusion protein, showed similar amount of autophagosomes within hyphae ( Figure 11C). After transferring into minimum medium minus nitrogen (MM-N), interestingly, the MoMih1 mutant accumulated similar autophagosomes to that of in wild-type KU80-eGFP-Atg8 ( Figure 11C). Western blot showed that the non-selective autophagy in KU80-eGFP-Atg8 strain was comparable with MoMih1-eGFP-Atg8 strain ( Figure 11D), suggesting that MoMih1 is dispensable for non-selective autophagy in M. oryzae.

Discussion
Successful proceeding of mitosis is necessary for the growth and development of eukaryotes (Hindley et al., 1987). In M. oryzae, cell cycle regulates the appressorium morphology and plant infection, and the completion of mitosis is critical for the full virulence of this fungus . In M. oryzae genome, a series of cell cycle regulators that is homologous to proteins involved in cell cycle regulation in S. cerevisiae, were identified (Oses-Ruiz and Talbot, 2017), and some of them have been verified to participate in the growth, development, sporulation and pathogenicity of this fungus (Saunders et al., 2010;Yue et al., 2017;Li et al., 2018), suggesting the essential roles of the cell cycle regulation during fungal development and plant infection of M. oryzae. In S. pombe, the dual-specificity phosphatase Cdc25 work together with Wee1 kinase to regulate the phosphorylation of Cdk1 to control the proceeding of mitosis, and the dephosphorylation of Cdk1 by Cdc25 promotes the cells to enter mitosis (Dunphy, 1994). In S. cerevisiae, deletion of Mih1, a homolog of Cdc25 in fission yeast, results in abnormally enlarging cells and moderately delayed mitotic process (Pal et al., 2008). Besides, Cdc25 and its homologs play also important roles in fungal growth and pathogenicity in many other filamentous fungi, and for example, the disruption of Cdc25 could incur retarded growth, diminished sporulation, and attenuated ability to infect the hosts by Beauveria bassiana and U. maydis (Sgarlata and Perez-Martin, 2005;Qiu et al., 2015), implying the critical conserved role of this Cdc25 protein during the growth and pathogenesis of those fungi. In M. oryzae, a dual-specificity phosphatase name as MoMih1 has been identified in the fungal genome, however, the underlying mechanism regulated by this cell cycle protein has yet to be understood till now. In this study, we found that MoMih1 is conserved in structures that are unique to Cdc25 and Mih1 phosphatases in S. pombe and S. cerevisiae, respectively. MoMih1 is required for normal mitosis and fungal development in M. oryzae, and deletion of MoMih1 lead to decreased activity of extracellular enzyme and ability to scavenge active oxygen of M. oryzae. These results indicate that MoMih1 plays an important role in determining the fungal morphogenesis and virulence of M. oryzae.
The hyphae of the filamentous fungal pathogens are highly polarized cells, and their polar growth at the hyphal tip is a prerequisite for vegetative growth, development and plant infection of those pathogens (Steinberg, 2007;Guo et al., 2015). In fungal pathogens, polarized growth depends on a specialized protein subcomplex called polarisome (Harris et al., 2005). The accurate localization of Spa2, a central component of polarisome, is required for establishment of proper hyphae polarized growth in filamentous fungal pathogens (Meyer et al., 2008;Li et al., 2014;Cao et al., 2021;Zheng et al., 2021b;Zhu et al., 2022). In this study, approximate 40% of the vegetative hyphae of the MoMih1 showed abnormally swollen tips, and the central localization of MoSpa2-eGFP observed in the polarisome of KU80 was completely abolished in the swollen hyphae tips or asymmetrically localized in the twisted hyphae tips of MoMih1 mutant. In view of the critical role of MoMih1 in protein dephosphorylation, and improper localization of Spa2 homologs to polarisome resulted in defective hyphal tip growth in M. oryzae and other phytopathogens (Li et al., 2014;Cao et al., 2021;Zhu et al., 2022), we speculate that MoMih1 might regulate polarized growth by dephosphorylation of certain components of the polarisome, such as MoSpa2, during fungal development in M. oryzae. Besides, continuous polarized tip growth also requires the constant delivery of cargo to the cell apex (Pantazopoulou et al., 2014). MoVps26, a subunit of retromer subcomplex that participate in intracellular transport of secretory vesicles cargo (Zheng et al., 2015), showed punctate structures at or near the vacuolar of KU80, but dispersed, cytoplasmic signals in the hyphae of M. oryzae. Based above, the abolishment of MoSpa2 and disordered localization of MoVps26 that affected by MoMih1 might be the reason for the abnormal polarized tip growth in M. oryzae.
The massive formation and diffusion of asexual spore directly determines the epidemic degree of the rice blast in the disease cycle (Zhang et al., 2016). Young conidia are produced on the tip of conidiophores differentiated from the fluffy arial hyphae in an axial manner (Liu et al., 2010). The fluffy growth of arial hyphae is proportional to the development of conidiophores, and essential for asexual sporulation of M. oryzae (Jeon et al., 2008;Guo et al., 2017;Matheis et al., 2017). In M. oryzae, deletion of the MoMih1 produced less arial hyphae, and significantly reduced conidiophores and conidia, suggesting a critical role of MoMih1 in determining the asexual development in the blast fungus. The significant downregulation of two transcription factors MoFLBC and MoFLBD, which are essential for aerial hyphae development and spore formation (Matheis et al., 2017), respectively, might be partially responsible for defects of conidiophore development and conidial formation in MoMih1 mutant. The MoACR1, a stagespecific negative regulator for conidiation, is essential for the establishment of sympodial pattern of spore formation, and a mutation of this protein lead to produce head-to-tail conidia in M. oryzae (Lau and Hamer, 1998). Besides, blocking of MoSsadh also completely abolished the conidiation due to the loss of conidiophore development in M. oryzae (Guo et al., 2011). We here found that the transcription level of those two genes MoMih1/MoVps26-eGFP was recorded by amending with or without benomyl, and the rate of abnormal mycelial tips was statistically analyzed. The error bars represent the SD of three replicates, whereas "**" represents a significant difference (p < 0.01).
significantly decreased in the MoMih1 mutant, further suggesting an important role of MoMih1 in conidiophore development. Taken together, our results implied that MoMih1 might function as a key upstream regulator to stimulate the transcription of conidiation related genes during asexual development of M. oryzae. The scavenging of ROS at the infection site of the plants is a prerequisite for the successful fungal penetration during plantmicrobe interaction (Molina and Kahmann, 2007;Chi et al., 2009;Guo et al., 2010;Guo et al., 2011;Zhang et al., 2021). In M. oryzae, the basic leucine zipper (bZIP) family transcriptional factors MoAP1 and Moatf1 are two important regulators in response to oxidative stressors, and deletion either of them resulted in attenuation of degrading the host-derived ROS, and thus reduced the pathogenicity of the mutants (Guo et al., 2010;Guo et al., 2011). In this study, deletion of MoMih1 attenuated the fungal virulence, with both decreased penetration and retarded invasive hyphae extension during plant infection. The MoMih1 mutant was more sensitive to hydrogen peroxide, and almost lost the secretion of extracellular laccases and peroxidases that help to scavenge the host-derived ROS. Since ROS generally acts as the second messenger that activate the defense pathway during plant microbe interactions (Molina and Kahmann, 2007;Guo et al., 2010;Guo et al., 2011), we speculated that MoMih1 may participate in fungal pathogenicity by regulating the extracellular enzyme activity to detoxify host-derived ROS and then facilitated the fungal penetration and invasive hyphae extension in M. oryzae.
The plant infection of M. oryzae is mediated by development of a functional appressorium (Wilson and Talbot, 2009;Zhang et al., 2011;Ryder et al., 2019). The proceeding of cell cycle is critical for the formation of appressorium and secondary infection hyphae (Saunders et al., 2010;. In S. cerevisiae, the checkpoint that inhibits anaphase onset mainly depends on the activity of Cdk1, and the Mih1 has been proved to remove the phosphate group of Cdk1 Y19 , which as a result activated the Cdk1 activity, and promote the transition of mitosis in S. cerevisiae (Pal et al., 2008). In this study, MoMih1 directly interacted with the Cdk1 homologue MoCdc28 in vivo, and deletion of MoMih1 resulted in incremental numbers of nucleus in many hyphal cell compartments of the fungus and significantly increased phosphorylation level of MoCdc28 Y15 in MoMih1 mutant, confirmed that MoMih1 functioned as important cell cycle regulator to proceed the mitosis during development of M. oryzae. This finding agrees with the role of Cdc25 in S. pombe and U. maydis, in which the deletion of Cdc25 strongly blocked the initiation of M phase, further suggesting a conserved role of Cdc25 tyrosine phosphatases in controlling the cell division in diverse fungi (Sgarlata and Perez-Martin, 2005). Interestingly, nuclear division arrest was not identified during appressorium development, we attributed this to the rapid mitosis whose duration time is too short to be captured during conidia germination in MoMih1 mutant. In S. cerevisiae, Mih1 localized to the nucleus and cytoplasm through the mechanism of nucleocytoplasmic shuttle, and in most situation functioned in the nucleus (Yano et al., 2013). In this study, MoMih1 was localized to both the nucleus and cytoplasm at mycelial stage, whereas only located in the cytoplasm at conidial stage, indicating the conserved role of MoMih1 during fungal development, and this spatial expression difference at diverse stages might reflect a critical MoMih1 physically interacts with MoCdc28 and regulates the Cdk1 phosphorylation in M. oryzae (A) Yeast two-hybrid assay. The coding sequences of MoMih1 and MoCdc28 were inserted into the vector pGADT7 and pGBKT7, respectively, and then co-transformed into yeast strain AH109. Yeast cells co-transformed with pGADT7/pGBKT7-MoCdc28, pGADT7-MoMih1/pGBKT7, pGADT7-T/pGBKT7-53, and pGADT7-T/pGBKT7-Lam were used as control, respectively. (B) Detection of interaction between MoMih1 and MoCdc28 by bimolecular fluorescence complementation (BiFC) assay. A pair of MoCdc28-N YFP and C YFP-MoMih1 constructs were co-transformed into wild type strain Guy11. Construct pairs of MoCdc28-N YFP and C YFP, C YFP-MoMih1 and N YFP, and N YFP and C YFP were co-transformed into Guy11 to serve as negative controls. YFP signals were detected under the CLSM. (C) Detection of the Cdk1 phosphorylation in the MoMih1 mutant. Western blot was carried out to detect Cdk1 Y 15 P (for phosphorylation signal) and Cdk1 in protein extracts using the anti-phospho-Cdc2 (Tyr15) and anti-Cdc2 antibodies, respectively. Note that the phosphorylation signal was intensified in MoMih1 deletion mutant.
role of MoMih1 in determining polar growth, cargo transport and full virulence of M. oryzae.
In conclusion, this study revealed the role of MoMih1 in the development of M. oryzae, which laid a foundation for further study on the role of MoMih1 in the cell cycle of M. oryzae.

Data availability statement
The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding author.

Author contributions
MG conceived the experiments. SL and XG performed the experiments. MG and SL wrote the manuscript. JM and SW helped to analyze the data and revise the manuscript. All authors contributed to the article and approved the submitted version.

Funding
This study was granted by the Foundation for the Excellent Talents of Anhui Agricultural University (Grant No: RC2015002 to

Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher's note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpls.2023.1146915/ full#supplementary-material To validate the deletion of MoMih1 in M. oryzae, genomic DNA of KU80 and MoMih1 deletion mutants were digested with SphI and hybridized with MoMih1-LF probe; To validate the copy number of HPH in MoMih1 deletion mutants, genomic DNA of KU80 and MoMih1 deletion mutants were digested with HindIII, and hybridized with the HPH probe.

SUPPLEMENTARY FIGURE 3
Bioinformatic analysis of MoMih1 and Cdc25 homologues. Bioinformatic analysis revealed that MoMih1 contains a conserved HCXXXXXR sequence and DCR motif unique to Cdc25 and Mih1 phosphatases in S. pombe and S. cerevisiae, respectively. The conserved DCR motif (aa 392-394) and the HCXXXXXR consensus sequence (aa 434-441) are marked with the black frame.

SUPPLEMENTARY FIGURE 4
The subcellular localization of MoMih1 in M. oryzae. The conidia (A), mycelia (B), and invasive hyphae (C) of the transformants expressing the MoMih1-eGFP fusion protein were examined under a CLSM.

SUPPLEMENTARY FIGURE 5
The MoMih1 mutants are hypersensitive to hydrogen peroxide in M. oryzae. (A). Mycelia growth under hydrogen peroxide. The KU80, MoMih1 mutants and MoMih1c were inoculated on CM with or without 5 mM H 2 O 2 and cultured at 28°C for 5 days. (B). Statistical analysis of the inhibition rate under hydrogen peroxide. "**" stands for the significant difference among the tested strains (p < 0.01), and error bars represent SD. (C). Laccase activity test. The laccase activity was monitored in CM supplemented with 0.2 mM 2, 2'azino-di-3-ethylbenzathiazoline-6-sulfonate (ABTS). (D). Measurement of the laccase and peroxidase activity. (E). Transcription of peroxidases and laccases encoding genes. The expression of putative peroxidases and laccases encoding genes was examined by qRT-PCR in the KU80 and MoMih1 mutants.

SUPPLEMENTARY FIGURE 6
MoMih1 is required for the hydrophobicity of M. oryzae. (A). Hydrophobicity assay for the MoMih1 mutants. Drops of water were inoculated on the colony surfaces of the tested strains. An easily-wettable phenotype of MoMih1 mutants was observed and photographed at 24 hpi. (B). The transcriptional levels of hydrophobin genes. The expression of four hydrophobin genes, such as MGG_09134, MGG_10105, MoMPG1, and MoMHP1, was significantly reduced in the MoMih1 mutant. "**" mean significant difference in gene expression (p< 0.01), and error bars represent SD.