The Phosphatase Cascade Nem1/Spo7-Pah1 Regulates Fungal Development, Lipid Homeostasis, and Virulence in Botryosphaeria dothidea

Botryosphaeria canker and fruit rot caused by the fungus Botryosphaeria dothidea is one of the most destructive diseases of apple worldwide. Our data indicated that the phosphatase cascade Nem1/Spo7-Pah1 plays important roles in the regulation of fungal growth, development, lipid homeostasis, environmental stress responses, and virulence in B. dothidea. ABSTRACT Protein phosphatase complex Nem1/Spo7 plays crucial roles in the regulation of various biological processes in eukaryotes. However, its biological functions in phytopathogenic fungi are not well understood. In this study, genome-wide transcriptional profiling analysis revealed that Nem1 was significantly upregulated during the infection process of Botryosphaeria dothidea, and we identified and characterized the phosphatase complex Nem1/Spo7 and its substrate Pah1 (a phosphatidic acid phosphatase) in B. dothidea. Nem1/Spo7 physically interacted with and dephosphorylated Pah1 to promote triacylglycerol (TAG) and subsequent lipid droplet (LD) synthesis. Moreover, the Nem1/Spo7-dependently dephosphorylated Pah1 functioned as a transcriptional repressor of the key nuclear membrane biosynthesis genes to regulate nuclear membrane morphology. In addition, phenotypic analyses showed that the phosphatase cascade Nem1/Spo7-Pah1 was involved in regulating mycelial growth, asexual development, stress responses, and virulence of B. dothidea. IMPORTANCE Botryosphaeria canker and fruit rot caused by the fungus Botryosphaeria dothidea is one of the most destructive diseases of apple worldwide. Our data indicated that the phosphatase cascade Nem1/Spo7-Pah1 plays important roles in the regulation of fungal growth, development, lipid homeostasis, environmental stress responses, and virulence in B. dothidea. The findings will contribute to the in-depth and comprehensive understanding of Nem1/Spo7-Pah1 in fungi and the development of target-based fungicides for disease management.

However, to date, only a few systematic studies regarding the biological functions of Nem1/Spo7 complex have been reported for filamentous fungi. Functional analysis of the phosphatome in Aspergillus fumigatus preliminarily showed that NemA is involved in regulating responses to stresses mediated by a cell wall-damaging agent and iron starvation (32). In phytopathogenic fungi, Nem1/Spo7 complex regulates LD biogenesis, disruption of the complex caused pleiotropic defects on hyphal growth, mycotoxin production, oxidation stress response, and pathogenicity in Fusarium graminearum (33,34). In Magnaporthe oryzae, only Nem1 and not Spo7 has been identified, and Nem1 is important for mycelial growth and conidiation but has no effect on host infection (35).
In this study, genome-wide transcriptional profiles analysis showed that the Nem1 homologue in B. dothidea was significantly upregulated and predicted to play an important role during the infection process. Therefore, a deep understanding of the fundamental biology of Nem1/Spo7 complex in B. dothidea will promote the development of new fungicides for apple ring rot management. We identified and dissected the biological functions of the HAD family phosphatase complex Nem1/Spo7 and its substrate Pah1 in B. dothidea. Our studies demonstrated that the PA phosphatase Pah1 interacted with and was dephosphorylated by Nem1/Spo7 and functioned as a transcriptional repressor of key nuclear membrane biosynthesis genes to regulate nuclear membrane morphology in B. dothidea. Moreover, dephosphorylated Pah1 promoted TAG and subsequent LD biosynthesis. In addition, DNem1, DSpo7, and DPah1 mutants displayed obvious defects in vegetative growth, asexual development, stress responses, and virulence in planta. Collectively, the results show that phosphatase cascade Nem1/Spo7-Pah1 was important for hyphal growth, cellular development, lipid metabolism, and pathogenicity in B. dothidea.

RESULTS
Identification and sequence analysis of Nem1 and Spo7 in B. dothidea. To identify candidate fungicide target proteins important for pathogenicity of B. dothidea, we performed genome-wide transcriptome analysis of the wild-type (WT) strain LW03 cultured in yeast extract-peptone-dextrose (YEPD) for 24 h with or without transference into liquid bark medium (to simulate infection conditions) for 12 h. Gene expression profile analysis revealed that Nem1 of B. dothidea encoded by KAF4313976.1 (NCBI:protein accession number KAF4313976.1) was upregulated about 74-fold in hyphae with transference into liquid bark medium compared to the hyphae in YEPD alone (Fig. 1), suggesting that the phosphatase Nem1 in B. dothidea may play crucial roles during infection process. Next, we retrieved the Spo7 orthologue of B. dothidea by BLAST search of the B. dothidea genome database (https://www.ncbi.nlm.nih.gov/biosample/SAMN13735646) with the Saccharomyces cerevisiae Spo7 protein sequence as a query. Sequence analysis revealed that NEM1 and SPO7 of B. dothidea are predicted to encode proteins with 511 and 443 amino acids (aa), sharing 48% and 30% sequence similarity with yeast Nem1 and Spo7, respectively. Phylogenetic analysis by IQTREE (http://www.iqtree.org/) of Nem1/Spo7 proteins in B. dothidea and other organisms indicated that the phosphatase catalytic subunit Nem1 and its regulatory subunit Spo7 are evolutionarily conserved from yeast to human (see Fig. S1A and B in the supplemental material). Structural analysis showed that Nem1 contains a transmembrane (TM) domain, a C-terminal region (designated CTR), and a typical catalytic HAD domain with the DLDET catalytic motif, which is highly conserved in S. cerevisiae ( Fig. 2A). Spo7 possesses three conserved regions (CR1 to CR3), two TM domains, and a conserved hydrophobic Leu-Leu-Ile (LLI) sequence comprising residues 50 to 52 in CR1 (Fig. 2B). In addition, yeast complementation assays showed that B. dothidea NEM1 could not restore the growth defect of yeast NEM1 deletion mutant on yeast extract peptone raffinose galactose (YPRG) medium containing caffeine (Fig. S2), indicating that the function of Nem1 at least partially differs between yeast and B. dothidea.
Nem1 interacts with Spo7 to form phosphatase complex Nem1/Spo7. To determine the interaction relations between Nem1 and Spo7 of B. dothidea, we conducted a coimmunoprecipitation (co-IP) analysis. Briefly, the total proteins were isolated from the wild-type strain (LW03) transformed with Nem1-3ÂFlag and/or green fluorescent protein (GFP)-Spo7, and proteins copurifying with the anti-GFP agarose were analyzed by Western blotting with the monoclonal anti-Flag antibody. As shown in Fig. 2C, Nem1 interacted with Spo7 in the co-IP assay. In addition, split-ubiquitin membrane-based yeast two-hybrid assays were performed to further confirm the association between Nem1 and Spo7. As shown in Fig. 2D, Nem1 could directly interact with Spo7. Moreover, the absence of CTR in Nem1 (Nem1DCTR) completely abolished the interaction between Nem1 and Spo7 ( Fig. 2D and E). To determine the effects of the hydrophobicity of the LLI sequence in Spo7 on the interaction pattern among Nem1 and Spo7, we individually or simultaneously mutated the three amino acids to alanine (A) and arginine (R); the resulting proteins were called Spo7 L50A , Spo7 L51A , Spo7 I52A , Spo7 3A , Spo7 L50R , Spo7 L51R , Spo7 I52R , and Spo7 3R , respectively. Alanine conserves the hydrophobic property, while arginine conserves the hydrophilic property. Co-IP analysis showed that the alanine-substituted forms of Spo7 (i.e., Spo7 L50A , Spo7 L51A , Spo7 I52A , and Spo7 3A ) exhibited a level of interaction with Nem1 similar to that of wild-type Spo7, while the interaction levels of the arginine-substituted forms Spo7 L50R , Spo7 L51R , Spo7 I52R , and Spo7 3R were significantly reduced (72, 68, 55, and 93%, respectively) (Fig. 2F). The split-ubiquitin membrane-based yeast two-hybrid assays showed consistent results (Fig. 2D). These results indicate that Nem1 interacts with Spo7 to form a phosphatase complex in B. dothidea. The CTR of Nem1 and the hydrophobicity of the LLI sequence in Spo7 are crucial for interaction between Nem1 and Spo7.
Identification of the phosphatidic acid phosphatase Pah1. To identify the substrate of Nem1/Spo7 complex, we sought candidate proteins which interact with Nem1/Spo7 by conducting affinity capture mass spectrometry (MS) as described in Materials and Methods. In this assay, we found that protein encoded by KAF4301686.1 (NCBI:protein accession number KAF4301686.1) was copurified with both Nem1-GFP and GFP-Spo7 (Table S1). Bioinformatics and phylogenic analysis showed that the KAF4301686.1-encoded protein is homologous to S. cerevisiae phosphatide phosphatase Pah1 (Fig. S1C), and it is therefore designated Pah1 of B. dothidea. PAH1 is predicted to encode a protein with 795 aa, sharing 54% sequence similarity with yeast Pah1. Structural analysis showed that Pah1 consists of a conserved N-LIP domain and HAD-like domain with the DIDGT catalytic motif, which are required for PA phosphatase activity. In addition, Pah1 contained a conserved amphipathic helix (AH) domain and an acidic tail (AL, aspartate/glutamate-rich sequences) in the N terminus and C terminus, respectively (Fig. 3A).
Nem1/Spo7 complex interacts with and dephosphorylates Pah1. To verify the interactions between Pah1 and Nem1/Spo7 complex in B. dothidea, co-IP assays were conducted, and the results showed that both Nem1 and Spo7 interacted with Pah1; Nem1 was not necessary for the interaction among Pah1 and Spo7 (Fig. 3B), while loss of Spo7 resulted in the complete abolishment of interaction between Pah1 and Nem1 ( Fig. 3C). Furthermore, AL of Pah1 was essential for its interaction with Spo7 (Fig. 3D). The interaction patterns between Nem1/Spo7 complex and Pah1 were further examined by the split-ubiquitin membrane-based yeast two-hybrid assays. Consistent with the results of the co-IP assays, Nem1 only physically interacted with Spo7 and not with Pah1, while Spo7 directly interacted with both Nem1 and Pah1, and lack of AL in Pah1 abolished its interaction with Spo7 (Fig. 3E). Collectively, these results implied that Nem1/Spo7 complex physically interacted with Pah1. Next, to confirm whether Pah1 is the catalytic substrate of Nem1/Spo7 complex in B. dothidea, we first generated deletion mutants of NEM1 (DNem1) and SPO7 (DSpo7) using a homology recombination strategy ( Fig. S3A and B) and then determined the phosphorylation patterns of Pah1 in wild-type strain LW03 and DNem1 and DSpo7 mutants by Phos-tag assays as described in Materials and Methods. Briefly, the Pah1-Flag fusion construct was transformed into LW03 and the DNem1 and DSpo7 mutants and the mobility shifts of Pah1-Flag in the resulting strains (LW03::Pah1-Flag, DNem1:: Pah1-Flag, and DSpo7::Pah1-Flag) were tested by immunoblotting assay using an anti-Flag antibody. As shown in Fig. 3F, the intensity of fast-migrating bands, which are correlated with dephosphorylated isoforms of Pah1, were significantly decreased after deletion of NEM1 or SPO7, indicating that Pah1 was dephosphorylated by Nem1/Spo7 complex in B. dothidea. Co-IP assays revealed that loss of Nem1 had no effect on the interaction between Spo7 and Pah1. (C) Co-IP assays showed that Spo7 was essential for interaction between Nem1 and Pah1. (D) Pah1 physically interacted with Spo7 by its AL domain in co-IP assays. (E) Split-ubiquitin membrane-based yeast two-hybrid analysis of the interaction between Nem1, Spo7, and wild type or mutated forms of Pah1. (F) Nem1/Spo7 complex was essential for dephosphorylation of Pah1. The Pah1-Flag protein from the wild-type strain (LW03) or the DNem1 or DSpo7 mutant was subjected to Phos-tag sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE; top) and normal SDS-PAGE (bottom), followed by immunoblotting with anti-Flag antibody.

Microbiology Spectrum
The Nem1/Spo7-Pah1 cascade regulates vegetative growth and asexual development. To test the detailed functions of the Nem1/Spo7-Pah1 cascade in B. dothidea, we generated single-deletion or complementation mutants of the cascade (DNem1, DSpo7, DPah1, DNem1-C, DSpo7-C, and DPah1-C) as described in Materials and Methods, and the mutants were confirmed by PCR and Western blotting assays (Fig. S3).
To determine the role of Nem1/Spo7-Pah1 in mycelial growth, LW03, DNem1, DSpo7, and DPah1 mutant, and complemented strains were cultured on potato dextrose agar (PDA) and minimal medium (MM). As shown in Fig. 4, after incubation at 25°C for 3 days, the DNem1 mutant exhibited a colony morphology similar to that of the DSpo7 mutant, which grow slower on PDA and MM than LW03 and their complemented strains (with growth reductions of 17% and 27% on PDA and MM, respectively). Compared with the DNem1 and DSpo7 mutants, the DPah1 mutant displayed a significantly more serious defect in hyphal growth rate, which was decreased to 14% of that of LW03.
Asexual reproduction plays important roles in the life cycle of B. dothidea and epidemic apple ring rot. To investigate the roles of the Nem1/Spo7-Pah1 cascade in asexual development, conidial production by each strain was evaluated on carrot agar conidium induction medium. After 7 days of incubation with black light exposure at 25°C, no fruiting bodies (FBs) were generated by the DPah1 mutant, and the DNem1 mutant produced numbers of FBs similar to those produced by the DSpo7 mutant, which were obviously fewer than that of LW03 and complemented strains ( Fig. 5A and B and Fig. S4). Moreover, the conidia in FBs generated by the DNem1 and DSpo7 mutants exhibited abnormal morphologic characteristics (Fig. 5C). The average length of conidia produced by the DNem1 and DSpo7 mutants were much shorter than those of LW03 and complemented strains (Fig. 5D). In addition, the abnormal conidia of the DNem1 and DSpo7 mutants exhibited a lower germination rate and shorter germinal tubes than those of the LW03 and complemented strains in the presence of 2% (wt/vol) sucrose ( Fig. 5E to G). Collectively, the results showed that the Nem1/Spo7-Pah1 cascade of B. dothidea is important for hyphal growth and asexual development.
The Nem1/Spo7-Pah1 cascade plays crucial roles in LD accumulation. To elucidate the functions of the Nem1/Spo7-Pah1 cascade in lipid droplet (LD) biogenesis, we observed the accumulation patterns of LDs using the LD staining dyes Nile red and BODIPY (boron-dipyrromethene). As shown in Fig. 6A and Fig. S5A, histochemical analysis showed that no visible LDs were observable in the hyphae of the DPah1 mutant, and the DNem1 and DSpo7 mutants contained much fewer LDs than LW03 and complemented strains. In yeast, dephosphorylated Pah1 upon Nem1/Spo7 complex catalyzes PA to produce DAG, which is then acylated to produce TAGs, the most prominent neutral lipids in LDs (36). Therefore, TAG quantification was performed and results showed that the TAG contents in the mycelia of the three deletion mutants (DNem1, DSpo7, and DPah1) were reduced to very low values, of which the DPah1 mutant decreased the most (Fig. 6B). In addition, we tested the LD accumulation patterns in conidia of each strain. As shown in Fig. 6C and Fig. S5B, the LD number was seriously decreased in conidia of the DNem1 and DSpo7 mutants compared with those in LW03 and complemented strains. Overall, these results indicate that the Nem1/Spo7-Pah1 cascade plays crucial roles in TAG and LD synthesis. Dephosphorylated Pah1 upon Nem1/Spo7 functions as a transcriptional repressor to regulate nuclear envelope morphology. During the process of protoplast preparation for mutant complementation, we observed that the size of protoplasts produced by the DNem1, DSpo7, and DPah1 mutants were obviously bigger than those of LW03 and complemented strains (Fig. S6), suggesting the phosphatase cascade Nem1/ Spo7-Pah1 may be involved in regulating protoplast membrane growth. Previous studies revealed that yeast Pah1 is a key regulator of nuclear membrane growth during the cell cycle. To investigate the functions of the B. dothidea Nem1/Spo7-Pah1 cascade in regulating nuclear membrane morphology, the nuclear membrane reporter B. dothidea Sec63 fused with GFP was transferred into LW03 and the DNem1, DSpo7, and DPah1 mutants. As shown in Fig. 7A, the DPah1 mutant exhibited seriously enlarged and irregularly shaped nuclei, and the DNem1 and DSpo7 mutants showed a consistent but milder defect on nuclear envelope morphology compared with the DPah1 mutant. The major lipid components of the nuclear membrane in yeast are phosphatidylinositol (PI) and phosphatidylcholine (PC). Ino1 is the rate-limiting enzyme for PI synthesis, and Opi3 is the enzyme catalyzing the final steps in the production of PC (18). To elucidate the mechanism of Nem1/Spo7-Pah1 in regulating nuclear membrane expansion in B. dothidea, quantitative real-time PCR (qRT-PCR) experiments were performed. As shown in Fig. 7B, the mRNA , and OPI3 showed a significant upregulation in the DNem1, DSpo7, and DPah1 mutants with respect to LW03. Moreover, the increase in the mRNA levels was higher in the DPah1 mutant than in the DNem1 or DSpo7 mutant, consistent with the phenomenon that the DPah1 mutant displayed more severe nuclear membrane expansion than the DNem1 or DSpo7 mutant. In addition, deletion of INO1 (whose mRNA levels increased most) obviously repressed nuclear membrane expansion in most cells of the DNem1, DSpo7, and DPah1 mutants (Fig. 7A), implying that dephosphorylation of Pah1 upon Nem1/Spo7 complex affects nuclear envelope morphology by controlling the expression of key nuclear membrane synthesis genes.
To further explore whether dephosphorylated Pah1 could bind to the promoters and function as a transcriptional regulator of nuclear membrane synthesis genes, we performed chromatin immunoprecipitation (ChIP) of the Pah1-GFP fusion, followed by qRT-PCR analysis. The results showed that Pah1 in fact associated with promoters but not the open reading frames (ORFs) of INO1, INO2, and OPI3, and the enriched amounts of Pah1 in these promoters were approximately 4.5-to 7.5-fold over the background levels detected at the promoter of a reference gene, GAPDH (encoding glyceraldehyde-3-phosphate dehydrogenase). What is more, the association of Pah1-GFP with promoters of the three genes was disrupted in the DNem1 and DSpo7 strains (Fig. 7C). These results suggest that dephosphorylated Pah1 by Nem1/Spo7 complex functions as a transcriptional repressor of nuclear membrane biosynthesis genes to maintain normal nuclear morphology in B. dothidea.
The Nem1/Spo7-Pah1 cascade is important for full virulence of B. dothidea. B. dothidea infects both branches and fruits of apples. To determine the role of the To test whether the attenuated virulence of the DNem1 and DSpo7 mutants was caused by their growth defect, the colony morphology of each strain was tested on apple fruit and bark media, which were similar to the nutritional conditions during the infection process. As shown in Fig. S7, the amounts of aerial mycelia of the DNem1 and DSpo7 mutants were similar to those of LW03 and complemented strains, and the growth rates of deletion mutants were decreased slightly on both fruit and bark media. These results suggest that the main reason for the reduced pathogenicity of the DNem1 and DSpo7 mutants is not growth defect. Previous studies showed that plants produced phytoalexin such as benzoxazolin-2-one (BOA) during the defense response to various biotic stresses (37). In this study, we found that both the DNem1 and DSpo7 mutants displayed increased sensitivity toward O-aminophenol (BOA detoxification intermediate) compared with LW03 and complemented strains (Fig. S8). These results suggest that the phosphatase cascade Nem1/Spo7-Pah1 was important for pathogenicity of B. dothidea, and Nem1/Spo7 complex regulates virulence at least partially by overcoming phytoalexin produced by plants during the infection process.

DISCUSSION
Nem1, a HAD family phosphatase number, serves as the catalytic subunit, which binds to its regulatory subunit Spo7 to form phosphatase complex Nem1/Spo7, has been reported to play critical roles in various biological processes in eukaryotes (18,27,38,39). However, the function of Nem1/Spo7 (CTDNEP1-NEP1R1 in mammalian cells) complex in filamentous fungi remains obscure. In this study, we found that the Nem1 homologue in B. dothidea was significantly upregulated during the infection process, implying that Nem1 is important for pathogenicity of B. dothidea, which prompted us to further investigate the biological functions of Nem1/Spo7 complex in B. dothidea.
We identified the Nem1/Spo7 homologue in B. dothidea, and structural analysis showed that Nem1 contains a transmembrane domain and a characteristic HAD-like domain. Although the HAD-like domain is divergent in Nem1 homologues of different organisms, the phosphatase activity conferred by catalytic motif DXDX(T/V) is evolutionarily conserved Role of the Phosphatase Nem1/Spo7-Pah1 in B. dothidea Microbiology Spectrum (19,40). Consistently, B. dothidea Nem1 possesses catalytic motif DLDET in the HAD-like domain, which is exactly the same as in yeast (30). Spo7 has two transmembrane domains and three other conserved regions. The split-ubiquitin membrane-based yeast two-hybrid and co-IP assays showed that Nem1 physically interacted with Spo7, and the CTR of Nem1 and the hydrophobicity of the LLI sequence in Spo7 are indispensable for their interaction.
To identify the catalytic substrate of Nem1/Spo7 complex, we first sought a candidate protein that interacts with Nem1 and Spo7. By performing affinity capture MS, co-IP, and yeast two-hybrid assays, we identified a phosphatidic acid phosphatase, Pah1, which physically interacted with Nem1/Spo7 complex by Spo7, and the acidic tail sequence of Pah1 was essential for the interaction with Spo7. Then the Phos-tag assay was performed, and the results showed that Pah1 was indeed dephosphorylated by Nem1/Spo7. Previous studies reported that dephosphorylated Pah1 catalyzes the conversion of PA to DAG, which is further converted to TAG for storage in LDs. In this study, we tested the TAG and LD accumulation patterns in single-deletion mutants of Nem1/Spo7-Pah1 cascade, and the results showed that both TAG contents and LD numbers were seriously decreased in the DNem1 and DSpo7 mutants. Under the same conditions, the DPah1 mutant exhibited a more severe defect, which is consistent with their orthologues in yeast (41) and filamentous fungi (33). These results suggested that the role of Nem1/Spo7-Pah1 in regulating TAG and subsequent LD synthesis was evolutionarily conserved. In addition, both the DNem1 and DSpo7 mutants of B. dothidea showed declined tolerance to the Demethylation inhibitor (DMI) fungicide tebuconazole (Fig. S9), which is widely used for the control of apple ring rot. Given that tebuconazole interferes with the synthesis of ergosterol, which is converted to sterol ester (SE) (14), another important component in LDs, we speculate that the response defects of the DNem1 and DSpo7 mutants toward tebuconazole may be associated with their lipid abnormalities, and the underlying mechanism needs to be further studied.
Notably, we report for the first time that single depletion of Nem1/Spo7-Pah1 cascade resulted in abnormal increases in the size of protoplasts. In addition, all the three single-deletion mutants of Nem1/Spo7-Pah1 displayed abnormal nuclear membrane expansion, and the DPah1 mutant exhibited a more severe defect than the DNem1 and DSpo7 phosphatase mutants. qRT-PCR and ChIP-qPCR assays revealed that dephosphorylated Pah1 upon Nem1/Spo7 could bind to the promoters of genes encoding key enzymes for nuclear membrane synthesis, which exhibited a significant upregulation in the DNem1, DSpo7, and DPah1 mutants. Consistently, the increase in the mRNA levels was higher in the DPah1 mutant than in the DNem1 and DSpo7 mutants. Moreover, knocking out of the most upregulated gene (INO1) repressed nuclear membrane expansion and restored a nearly normal spherical nucleus in most cells of all three deletion mutants. These results indicate that the dephosphorylated Pah1 functions as a transcriptional repressor of nuclear membrane synthesis genes to regulate nuclear envelope morphology. Interestingly, knockout of AnNem1 (Nem1 homologue in Aspergillus nidulans) led to small nuclei (42); in contrast, deletion of Nem1 in B. dothidea yielded expansive nuclei with a bigger size, suggesting the diverse functions and mechanisms of Pah1 proteins in regulating nuclear size of eukaryotes.
It is worth noting that Nem1 could not complement the stress response defects of yeast Nem1 deletion mutants. In addition, recent advances in the budding yeast and higher eukaryotes have been increasingly connecting LD dynamics to the regulation of autophagy, and TAG, as the key component in LDs, is demonstrated to be important for autophagy induction after nutrient starvation (43)(44)(45). Deletion of yeast Nem1, Spo7, or Pah1 seriously reduces TAG contents and LD numbers and finally affects the initiation of autophagy (26). However, in this study, our results showed that although the TAG and LD numbers were significantly decreased, autophagy occurred normally in the DNem1, DSpo7, and DPah1 mutants (data not shown). These results indicate that the roles of the Nem1/Spo7-Pah1 cascade in regulating autophagy vary in different organisms despite their conserved functional domains.
In filamentous fungi, such as A. fumigatus, M. oryzae, and F. graminearum, Nem1 is involved in regulating mycelial growth and conidiation but not in conidial morphology or germination (32)(33)(34)(35). Nevertheless, in this study, disruption of either Nem1 or Spo7 affected hyphal growth and conidiation as well as conidial morphology and germination. Moreover, F. graminearum Nem1/Spo7 (FgNem1/Spo7) complex plays important roles in regulation of pigment metabolism but not in response to cell wall-damaging stress mediated by Congo red or calcofluor white (CFW) in F. graminearum, while B. dothidea Nem1/Spo7 single-deletion mutants displayed normal pigment and hypersensitivity toward CR and CFW, which is similar to the case with in A. fumigatus. Moreover, A. fumigatus NemA (AfNemA) is involved in regulating responses to stress mediated by iron starvation, while the DNem1 and DSpo7 mutants exhibited no defects in sensitivity toward deficiency of iron in B. dothidea. In addition, FgNem1/Spo7 complex is critical for full virulence of F. graminearum mainly by affecting mycotoxin deoxynivalenol (DON) production and response to plant immunity mediated by reactive oxygen. In contrast, knockout of M. oryzae Nem1 (MoNem1) does not affect the pathogenicity of M. oryzae despite its lower growth rate. In this study, our results showed that disruption of the B. dothidea Nem1/ Spo7-Pah1 cascade significantly reduced the virulence of B. dothidea on both apple fruit and branch, and the defects in phytoalexin tolerance may contribute to the reduced virulence of the DNem1 and DSpo7 mutants. We could not determine the sensitivity of the B. dothidea DPah1 mutant toward various environmental pressures due to its extremely declined hyphal growth rate and complete loss of asexual reproduction. Therefore, it is difficult to analyze the reasons causing the abolished virulence of the DPah1 mutant, but it is certain that growth defects are one of the important reasons for the decline of the pathogenicity. These results suggest that Nem1/Spo7 complex plays diverse roles in regulating development, secondary metabolism, stresses responses, and pathogenicity in different fungal species. Taken together, the results indicate that the phosphatase cascade Nem1/Spo7-Pah1 presents great functional differentiation among diverse species despite the existence of several conserved domains; it is speculated that there is a pattern of evolution in which different species adapt to survive, reproduce, and spread.
In conclusion, our results indicate that protein phosphatase complex Nem1/Spo7 interacted with and dephosphorylated the PA phosphatase Pah1. Dephosphorylated Pah1 was active and promoted the conversion of PA to TAGs and subsequent LD synthesis. Moreover, Nem1/Spo7-dependently dephosphorylated Pah1 could bind to the promoters of key nuclear membrane biosynthesis genes and functioned as a transcriptional repressor to regulate nuclear membrane morphology. In addition, phenotypic analyses showed that the Nem1/Spo7-Pah1 cascade was involved in regulating hyphal growth, asexual development, stress responses, and pathogenicity of B. dothidea (Fig. 9). The results of this study will provide a theoretical basis for using the phosphatase cascade Nem1/Spo7-Pah1 as a target for phosphatase inhibitor or fungicide development and RNA interference (RNAi) for the management of apple ring rot, and they also provide insights on comprehensive functional analysis of phosphatase complex Nem1/Spo7 in fungi.

MATERIALS AND METHODS
Transcription profiling assays. Total RNA was isolated using TRIzol reagent (TaKaRa Inc., Dalian, China) according to the manufacturer's instructions. The sequencing libraries were generated using the NEBNext Ultra RNA library prep kit for Illumina (New England Biolabs [NEB], Ipswich, MA, USA) as described previously (46). The clustering of the index-coded samples was performed on a cBot cluster generation system using TruSeq cluster kit v3-cBot-HS (Illumina, Bologna, Italy) according to the manufacturer's instructions. The library preparations were sequenced on an Illumina Nova platform, and 150bp paired-end reads were generated. Differential expression analysis was performed using the DESeq2 R package (1.16.1).
Affinity capture MS analysis. Nem1-GFP and GFP-Spo7 were transferred into the LW03, and the resulting transformants were used for protein extraction as described previously (47). The supernatant protein was transferred into a sterilized Eppendorf tube. Anti-GFP agarose (40 mL; Chromo Tek, Munich, Germany) was added to capture Nem1-GFP or GFP-Spo7 interacting proteins following the manufacturer's instructions. After incubation overnight, the agarose was washed three times with TBS (20 mM Tris-HCl, 500 mM NaCl [pH 7.5]). Proteins binding to the beads were boiled with 90 mL of TBS supplemented with 15 mL of 10% SDS. After centrifugation, proteins in the supernatant were digested with trypsin as described previously (47). Tryptic peptides were analyzed by mass spectrometry as previously described (48).

Microbiology Spectrum
Fungal strains and culture conditions. B. dothidea strain LW03 (LXS030101) was single-spore isolated from infected red Fuji fruit showing typical symptoms of apple ring rot in Shandong Province and was used as the wild-type (WT) strain for transformation experiments throughout this study. Strains were grown on potato dextrose agar (PDA; 200 g of potato, 20 g of glucose, 10 g of agar and 1 L of water), minimal medium (MM; 0.5 g of KCl, 2 g of NaNO 3 , 1 g of KH 2 PO 4 , 0.5 g of MgSO 4 Á7H 2 O, 0.01 g of FeSO 4 Á7H 2 O, 30 g of sucrose, 200 mL of trace element, 15 g of agar, and 1 L of water [pH 6.9]), bark medium (100 g of bark from Fuji apple tree and 15 g of agar in 1 L of water), and fruit medium (200 g of Fuji apples and 15 g of agar in 1 L of water) for mycelial growth tests. For fruiting body and conidium induction, each fungal strain was grown on carrot agar medium (200 g of carrot and 15 g of agar in 1 L of water) with near-UV light (wavelength, 365 nm; HKiv, Xiamen, China). Mycelia used for protein extraction or fluorescence observation were grown in yeast extract-peptone-dextrose (YEPD) medium (10 g of peptone, 3 g of yeast extract, 20 g of D-glucose, made up to 1 L [pH 6.7]). For determination of fungal growth under phytoalexin stress conditions, mycelial plugs (5 mm in diameter) taken from the periphery of a 3-day-old colony of each strain were inoculated onto PDA amended or not with phytoalexin stress mediator O-aminophenol at concentrations indicated in the figure legends. After 3 days of incubation at 25°C, colony diameter in each plate was measured. Each experiment was repeated three times.
Generation of deletion and complemented mutants. Gene deletion vectors were constructed using a strategy based on double-joint PCR (49). Primers used to amplify the flanking sequences of each gene are listed in the Table S1 in the supplemental material. Briefly, the knockout constructs were transformed into wild-type strain LW03 protoplasts by employing polyethylene glycol (PEG)-mediated protoplast transformation (50). Putative gene deletion mutants were identified by PCR with relevant primers (Table S2).
To construct the Nem1-GFP fusion cassette, the NEM1 fragment containing the native promoter and open reading frame (ORF; without stop codon) was amplified with primers P19 and P20 (Table S1) and the resulting PCR products were cotransformed with XhoI-digested pYF11 into yeast strain XK1-25 (51). The recombined NEM1-GFP fusion vector was constructed with the alkali-cation yeast transformation kit (MP Biomedicals, Solon, OH, USA). Subsequently, the Nem1-GFP fusion vector was recovered from the yeast transformants by using the yeast plasmid extract kit (Solarbio, Beijing, China) and then transferred into E. coli strain DH5a. The GFP-SPO7, PAH1-GFP, PAH1-FLAG, SEC63-GFP, and NEM1-FLAG fusion vectors were generated similarly. Finally, the NEM1-GFP, GFP-SPO7, and PAH1-GFP plasmids were introduced into the deletion mutants for generation of complemented strains. SEC63-GFP plasmid was transformed into LW03 and various deletion mutants to obtain fluorescence-labeled strains.
Histochemical analysis of LDs and quantification of TAG production. Nile red or boron-dipyrromethene (BODIPY) was used for lipid droplet (LD) staining. For Nile red staining, hyphae or conidia of each strain were incubated in Nile red staining solution (20 mg mL 21 ) of polyvinylpyrrolidone and 2.5 mg mL 21 of Nile red oxazone (Sigma-Aldrich, St. Louis, MO, USA) in 50 mM Tris-maleate buffer (pH 7.5) for a few seconds, and then LDs were examined under a microscope with an episcopic fluorescence attachment. For BODIPY staining, hyphae or conidia were stained for 10 min in the dark with 1 mM BODIPY493/503 (Thermo Fisher Scientific, MA, USA) dissolved in 0.1Â phosphate-buffered saline (PBS) buffer. After washing twice with 1 Â PBS buffer, LDs were examined with a confocal microscope (TCS SP5; Leica, Germany). Each experiment was repeated three times.
To determine TAG content, fresh hyphae were digested with cellulase (Kaiyang, Shanghai, China), lysozyme (Kaiyang) and Driselase (Sigma-Aldrich) for 4 h in 0.7 M NaCl to prepare for protoplasts. Then triglyceride quantification kit E1013 (Applygen, Beijing, China) was used to quantify the TAG concentration The phosphatidic acid (PA) phosphatase Pah1 is phosphorylated by multiple protein kinases, and then the phosphorylated Pah1 is dephosphorylated by protein phosphatase complex Nem1/Spo7. Subsequently, the dephosphorylated Pah1 catalyzes PA conversion to diacylglycerol (DAG) and subsequent TAG and LD synthesis. Moreover, dephosphorylated Pah1 could bind to the promoter of key nuclear membrane biosynthesis genes and functioned as a transcriptional repressor to regulate nuclear membrane morphology. In addition, the Nem1/Spo7-Pah1 cascade was involved in regulating hyphal growth, asexual development, and pathogenicity of B. dothidea.
Role of the Phosphatase Nem1/Spo7-Pah1 in B. dothidea Microbiology Spectrum after resuspension of the 50 mg of protoplasts in 1 mL of lysis buffer according to the manufacturer's instructions. The protein concentrations in the resulting lysis supernatant were determined by the bicinchoninic acid method with protein quantification kit P1511 (Applygen) as a reference. The experiment was repeated three times. Western blotting. For total protein extractions, mycelia were ground in liquid nitrogen. Approximately 200 mg of finely ground mycelia powder was resuspended in 1 mL of extraction buffer (50 mM Tris-HCl [pH 7.5], 100 mM NaCl, 5 mM EDTA, 1% Triton X-100, 2 mM phenylmethylsulfonyl fluoride [PMSF]) and 10 mL of protease inhibitor cocktail (Sangon, Shanghai, China). After homogenization with a vortex shaker, the lysate was centrifuged. The supernatant was mixed with protein loading buffer and boiled for 5 min. The resulting proteins were separated by 12.5% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to a polyvinylidene fluoride (PVDF) membrane. To test the expression pattern of Nem1, Spo7, and Pah1 in complemented strains, monoclonal anti-GFP antibody 32146 (Abcam, Cambridge, MA, USA) was used at a 1:100,000 dilution for immunoblot analyses. The samples were also detected with monoclonal anti-glyceraldehyde-3-phosphate dehydrogenase (anti-GAPDH) antibody EM1101 (Hangzhou HuaAn Biotechnology Co., Ltd.) as a reference. Incubation with a secondary antibody and chemiluminescent detection were performed as described previously (52). Each experiment was repeated three times.
Co-IP assay. The GFP and Flag fusion constructs were transformed in pairs into LW03. Transformants expressing the fusion constructs were confirmed by Western blotting. In addition, the transformants bearing a single fusion construct were used as references. For coimmunoprecipitation (co-IP) assays, total proteins were isolated and incubated with the anti-GFP agarose (Chromo Tek, Martinsried, Germany). Proteins eluted from agarose were analyzed by Western blotting detection with monoclonal anti-Flag antibody A9044 (Sigma-Aldrich, St. Louis, MO, USA) and monoclonal anti-GFP antibodies. Each protein sample was also detected with anti-GAPDH antibody as a reference. The experiment was repeated three times.
Phos-tag assay. After 10 5-mm mycelial plugs of each strain were cultivated in YEPD medium at 25°C for 42 h, hyphae of each strain were harvested for protein extraction. The resulting protein samples were resolved on 8% SDS-polyacrylamide gels prepared with 25 mM Phos-tag binding reagent acrylamide (APEÂBIO, F4002) and 100 mM MnCl 2 (the separating gel included 1. . Gels were electrophoresed at 20 mA/gel for 4 h. Prior to protein transfer, gels were first equilibrated three times in transfer buffer containing 5 mM EDTA for 5 min and further equilibrated in transfer buffer without EDTA for 5 min two times. Protein transform from the Mn 21 -Phos-tag acrylamide gel to the PVDF membrane was performed for 5 h at 100 V at 0°C, and finally, the membrane was analyzed by Western blotting with the anti-Flag antibody. Each protein sample was also separated by 12.5% SDS-PAGE and detected with anti-Flag antibody as a reference. The experiment was repeated three times. Split-ubiquitin membrane-based yeast two-hybrid assay. In order to construct plasmids for the split-ubiquitin membrane-based yeast two-hybrid assay, coding sequence of each tested gene was amplified from cDNA of LW03 with the primer pairs indicated in Table S1. The wild-type or mutated form of the NEM1, SPO7, or PAH1 cDNA fragment was fused to the C-terminal or N-terminal half of ubiquitin (Cub) and the artificial transcription factor LexA-VP16 (Dualsystems Biotech, Zurich City, Switzerland). In addition, NEM1 or PAH1 was fused to the mutated C-terminal or N-terminal half of ubiquitin (Nub) (Dualsystems Biotech, Switzerland). The pairs of yeast two-hybrid plasmids were cotransformed into S. cerevisiae strain NMY51 following the LiAc/single-stranded DNA/PEG transformation protocol (53). Transformants were grown at 30°C for 3 days on synthetic medium lacking leucine (Leu) and tryptophan (Trp) and then transferred to medium stripped of Trp, Leu, histidine (His), and adenine (Ade) to assess binding activity. The experiments were repeated three times.
qRT-PCR analysis. For total RNA extraction, the harvested mycelia of each strain were ground in liquid nitrogen, total RNA was then isolated from mycelia of each sample with TaKaRa RNAiso reagent (TaKaRa Biotechnology Co., Dalian, China), and 10 mg of each RNA sample was used for reverse transcription with a RevertAid H Minus first-strand cDNA synthesis kit (Fermentas Life Sciences, Burlington, ON, Canada). The expression levels of INO1, INO2, and OPI3 were determined by quantitative real-time PCR (qRT-PCR) using the primers listed in Table S1. For each sample, the ACTIN gene was measured as an internal control using qPCR (0.4 mL of primer-F, 0.4 mL of primer-R, 10 mL of SYBR [TaKaRa Biotechnology Co., Dalian, China], 1 mL of cDNA and 8.2 mL of H 2 O) and an RT-PCR program (initial denaturation at 95°C for 3 min, followed by 40 cycles of denaturation at 95°C for 20 s, annealing at 53°C for 25 s, extension at 72°C for 30 s, and final extension at 72°C for 10 min). The experiment was repeated three times independently.
ChIP assays. Chromatin immunoprecipitation (ChIP) was performed as described previously (54). Briefly, mycelia were cross-linked with 1% formaldehyde for 10 min and then stopped with 125 mM glycine. Subsequently, samples were grounded in liquid nitrogen and suspended in lysis buffer with protease inhibitor. DNA was sheared into 200-to 500-bp fragments in a Bioruptor Plus (UCD-300; Liege, Belgium). After centrifugation, the supernatant was diluted with 10Â ChIP dilution buffer. The DNA fragments' supernatants were incubated with the GFP antibody 290 (Abcam, Cambridge, MA, USA) as the IP sample or incubated with anti-IgG1 antibody MA1-10406 (Invitrogen-Thermo Fisher Scientific, MA, USA) as the mock sample. In addition, protein A agarose beads (sc-2001; Santa Cruz Biotechnology, Dallas, TX, USA) were added into IP and mock samples before incubation with antibodies to remove the background. The beads were washed after a second addition for immunoprecipitation, and the immunoprecipitated complexes were then eluted from beads. Then 5 M NaCl and 2 mL of RNase were added into ChIP and mock samples to reverse crossing-linking and digested with proteinase K. Next, an equal volume of phenol-chloroform-isoamyl alcohol was added into each sample to precipitate DNA. Finally, the IP or mock DNA samples were quantified by qPCR assays with primers listed in Table S2. Relative enrichment levels were determined using the fold enrichment method. The experiment was repeated three times.
Determination of asexual reproduction and pathogenicity assays. To detect asexual development, aerial hyphae of 4-day-old cultures of LW03, DNem1, DSpo7, and complemented strains and 7day-old cultures of DPah1 grown on carrot agar plates were pressed down with 500 mL of 0.25% Tween 60 and incubated under black light (wavelength, 365 nm; HKiv Co., Ltd., Xiamen, China); fruiting body formation and conidium production were assayed after incubation at 25°C for 2 weeks.
The pathogenicity of B. dothidea was analyzed on young Fuji apple fruits and 1-year-old branches. To test the virulence of each strain on wounded apple fruits, holes were punched into the apple skin (5 mm in diameter, 3 mm in depth), and mycelial plugs (5 mm in diameter) of each strain were taken from the edge of a 3-day-old colony grown on PDA and then transferred into the holes of the apple fruits. For pathogenicity tests on branches, the phloem was exposed after removal of 5-mm-diameter samples of epidermis of branches, and then the mycelial plugs of each strain were transferred onto the exposed phloem. In addition, water agar plugs without mycelia were used as negative controls. The inoculated fruits and branches were placed under high-humidity conditions (95%) at 25°C for 7 days and 10 days, respectively. These experiments were repeated three times independently, and 20 samples were used for each strain each time.
Data availability. The RNA sequencing data set is available in supplemental material.