Nucleosome Assembly Protein 1, Nap1, Is Required for the Growth, Development, and Pathogenicity of Magnaporthe oryzae

Magnaporthe oryzae is the causal agent of rice blast, leading to significant reductions in rice and wheat productivity. Nap1 is a conserved protein in eukaryotes involved in diverse physiological processes, such as nucleosome assembly, histone shuttling between the nucleus and cytoplasm, transcriptional regulation, and the cell cycle. Here, we identified Nap1 and characterized its roles in fungal development and virulence in M. oryzae. MoNap1 is involved in aerial hyphal and conidiophore differentiation, sporulation, appressorium formation, plant penetration, and virulence. ΔMonap1 generated a small, elongated, and malformed appressorium with an abnormally organized septin ring on hydrophobic surfaces. ΔMonap1 was more sensitive to cell wall integrity stresses but more resistant to microtubule stresses. MoNap1 interacted with histones H2A and H2B and the B-type cyclin (Cyc1). Moreover, a nuclear export signal (NES) domain is necessary for Nap1’s roles in the regulation of the growth and pathogenicity of M. oryzae. In summary, NAP1 is essential for the growth, appressorium formation, and pathogenicity of M. oryzae.


Introduction
Magnaporthe oryzae (syn. Pyricularia oryzae)-induced rice blast is a highly destructive rice disease capable of causing massive yield reductions in rice or wheat crops worldwide [1]. At the beginning of infection, a three-celled conidium adheres to the rice surface, germinates under suitable conditions (appropriate temperature and humidity), and forms a germ tube, which subsequently differentiates into a melanized, dome-shaped appressorium stimulated by external hydrophobic signals and hardness [2,3]. During appressorium maturation, the glycogen granules and lipid droplets in conidia are gradually degraded and translocated to appressoria, resulting in the accumulation of a high concentration of glycerol, which generates a turgor pressure of up to 8 MPa [4]. The immense turgor provokes great mechanical pressure in an appressorium and promotes its further development into a penetration peg that quickly develops into invasive hyphae in plant cells. After approximately 5-7 days, the typical necrotic spots of rice blast appear on plant surfaces. At this time, a large number of secondary conidia are released from conidial stalks, and start the next round of the infection cycle [5]. Nap1, nucleosome assembly protein 1, was first identified in HeLa cell extracts due to its function in stimulating nucleosome assembly [6]. Nap1 is conserved in eukaryotes Using the gene replacement strategy, we knocked out NAP1 in the wild-type M. oryzae strain 70-15 ( Figure S1A). We verified that the insertion copy number of HPH into ΔMonap1 mutant genome was single by qPCR (Table S1). Based on ΔMonap1, we characterized the role of NAP1 in the growth and pathogenicity of M. oryzae. To determine whether the mutant phenotype is caused by the deletion of MoNAP1, we built a complementation strain of ΔMonap1 (Monap1c) by transforming the full-length genomic copy of MoNAP1 into ΔMonap1. We also complemented ΔMonap1 with the NAP1 gene of S. cerevisiae (Ynap1c) (Figure S1B-F). In the following section, we describe the biological function of MoNap1 in detail.

NAP1 Is Involved in Sporulation and Appressorium Formation in M. oryzae
The ΔMonap1 mutant grew more slowly than the wild type 70-15 in CM, exhibiting reductions in mycelial growth of approximately 20%. The colony diameter of ΔMonap1 was 4.24 ± 0.04 cm, while that of the wild type was 5.18 ± 0.02 cm at 9 dpi (Figure 2A,B). In addition, the colonies of the mutant appeared thinner and whiter than those of the wild type ( Figure 2A). ΔMonap1 is more susceptible to senescence, which can be observed by the collapse and autolysis of the aerial hyphae at the center of the colony (Figure 2A), indicating that MoNap1 may play a role in the aging process. However, the growth of ΔMonap1 in MM was comparable to that of the wild type 70-15 ( Figure 2C,D). ΔMonap1 produced significantly fewer spores, approximately 1/8 of the wild type ( Figure 2E). In ΔMonap1, the mRNA levels of two transcription factor genes FLBC and CON7, which are involved in the regulation of sporulation, were significantly downregulated ( Figure S2). In the wild type, many conidiophores bearing spores in a typical sympodial mode were observed, while ΔMonap1 produced fewer conidiophores and fewer spores in each conidiophore ( Figure 2F). The appressorial morphology of ΔMonap1 was distorted. The appressoria of the wild type were usually circular or dome-shaped, while approximately 61.5 ± Using the gene replacement strategy, we knocked out NAP1 in the wild-type M. oryzae strain 70-15 ( Figure S1A). We verified that the insertion copy number of HPH into ∆Monap1 mutant genome was single by qPCR (Table S1). Based on ∆Monap1, we characterized the role of NAP1 in the growth and pathogenicity of M. oryzae. To determine whether the mutant phenotype is caused by the deletion of MoNAP1, we built a complementation strain of ∆Monap1 (Monap1c) by transforming the full-length genomic copy of MoNAP1 into ∆Monap1. We also complemented ∆Monap1 with the NAP1 gene of S. cerevisiae (Ynap1c) (Figure S1B-F). In the following section, we describe the biological function of MoNap1 in detail.

NAP1 Is Involved in Sporulation and Appressorium Formation in M. oryzae
The ∆Monap1 mutant grew more slowly than the wild type 70-15 in CM, exhibiting reductions in mycelial growth of approximately 20%. The colony diameter of ∆Monap1 was 4.24 ± 0.04 cm, while that of the wild type was 5.18 ± 0.02 cm at 9 dpi (days post inoculation) (Figure 2A,B). In addition, the colonies of the mutant appeared thinner and whiter than those of the wild type ( Figure 2A). ∆Monap1 is more susceptible to senescence, which can be observed by the collapse and autolysis of the aerial hyphae at the center of the colony (Figure 2A), indicating that MoNap1 may play a role in the aging process. However, the growth of ∆Monap1 in MM was comparable to that of the wild type 70-15 ( Figure 2C,D). ∆Monap1 produced significantly fewer spores, approximately 1/8 of the wild type ( Figure 2E). In ∆Monap1, the mRNA levels of two transcription factor genes FLBC and CON7, which are involved in the regulation of sporulation, were significantly downregulated ( Figure S2). In the wild type, many conidiophores bearing spores in a typical sympodial mode were observed, while ∆Monap1 produced fewer conidiophores and fewer spores in each conidiophore ( Figure 2F). The appressorial morphology of ∆Monap1 was distorted. The appressoria of the wild type were usually circular or dome-shaped, while approximately 61.5 ± 11.4% of the appressoria in ∆Monap1 were abnormal ( Figure 3A,B) and smaller than those in the wild type (the diameter of appressoria in the wild type was 11.86 ± 0.34 µm, whereas that in the ∆Monap1 mutant was 8.98 ± 0.07 µm) ( Figure 3C). The germ tubes of the ∆Monap1 mutants were significantly longer than those of the wild type when germinated on hydrophobic plastic coverslips ( Figure 3A,D). The germ tube of the wild type was 16.04 ± 1.55 µm in length, whereas that of ∆Monap1 was 57.89 ± 7.07 µm ( Figure 3D). Spore germination was not affected by the MoNAP1 deletion at 4 hpi (hours post inoculation); however, the appressorial formation rate in ∆Monap1 was lower than that in the wild type at 24 hpi (89.8 ± 3.2% in ∆Monap1 compared to 99.0 ± 0.7% in the wild type) ( Figure 3E). In addition, turgor pressure of appressoria at 24 hpi was not significantly different between ∆Monap1 mutants and wild-type ( Figure S3).  11.4% of the appressoria in ΔMonap1 were abnormal ( Figure 3A,B) and smaller than those in the wild type (the diameter of appressoria in the wild type was 11.86 ± 0.34 μm, whereas that in the ΔMonap1 mutant was 8.98 ± 0.07 μm) ( Figure 3C). The germ tubes of the ΔMonap1 mutants were significantly longer than those of the wild type when germinated on hydrophobic plastic coverslips ( Figure 3A,D). The germ tube of the wild type was 16.04 ± 1.55 μm in length, whereas that of ΔMonap1 was 57.89 ± 7.07 μm ( Figure 3D). Spore germination was not affected by the MoNAP1 deletion at 4 hpi; however, the appressorial formation rate in ΔMonap1 was lower than that in the wild type at 24 hpi (89.8 ± 3.2% in ΔMonap1 compared to 99.0 ± 0.7% in the wild type) ( Figure 3E). In addition, turgor pressure of appressoria at 24 hpi was not significantly different between ΔMonap1 mutants and wild-type ( Figure S3).     The complementation strain Monap1c recovered the mutant's defects in mycelial growth, sporulation, conidiophore formation, and conidium differentiation, indicating that the defects in growth and conidiation of ∆Monap1 were caused by the deletion of NAP1. We also complemented ∆Monap1 with the NAP1 gene of S. cerevisiae and found that NAP1 in S. cerevisiae can restore the sporulation and virulence of the ∆Monap1 mutant but not the growth of ∆Monap1 ( Figure 4A-D).
The complementation strain MoNAP1c recovered the mutant's defects in mycelial growth, sporulation, conidiophore formation, and conidium differentiation, indicating that the defects in growth and conidiation of ΔMonap1 were caused by the deletion of NAP1. We also complemented ΔMonap1 with the NAP1 gene of S. cerevisiae and found that NAP1 in S. cerevisiae can restore the sporulation and virulence of the ΔMonap1 mutant but not the growth of ΔMonap1 ( Figure 4A-D).  The data were analyzed by GraphPad Prism 8.0 and significant differences compared with the wild type were estimated by multiple t tests: ** p < 0.01. (C) Conidiation in the wild type, ∆Monap1, Ynap1c, and Monap1c strains. At least 150 spores were counted per replicate. n = 3 independent biological replicates. Error bars represent the standard deviations. The data were analyzed by GraphPad Prism 8.0 and significant differences compared with the wild type were estimated by multiple t tests: ** p < 0.01. (D) Disease symptoms on leaf explants of barley inoculated with mycelial plugs from wild type, ∆Monap1, Ynap1c, and Monap1c. The pictures were photographed at 4 dpi.

NAP1 Is Required for Virulence in M. oryzae
The ∆Monap1 mutant displayed reduced virulence on barley leaves and rice seedlings. We evaluated the fungal pathogenicity using three inoculation methods. At 4 dpi, the mycelial plugs of the wild-type 70-15 and the complementation strain Monap1c caused yellow and brown blast lesions and rotten plant tissues, while the mycelial plugs of ∆Monap1 did not ( Figure 5A). When inoculated on detached barley leaves with 20 µL spore suspensions (5 × 10 4 conidia mL −1 ) for 4 days, the disease lesions caused by ∆Monap1 were apparently weaker than those caused by the wild type and the complementation strain ( Figure 5B). When sprayed on 2-week-old rice seedlings (CO-39) with conidial suspensions of 5 × 10 4 conidia mL −1 , the ∆Monap1 mutant caused small and restricted lesions in contrast to the typical spindle-like, gray centered, and merged blast lesions caused by the wild type and complementation strain at 7 dpi ( Figure 5C). The lesion areas caused by ∆Monap1 were significantly lower than those caused by the wild type and Monap1c (those caused by ∆Monap1 were 19 ± 8.1%, while those caused by the wild type and Monap1c were 52 ± 4.6% and 53 ± 7.0%, respectively) ( Figure 5D). Furthermore, we measured the penetration rate of the wild type, ∆Monap1, and Monap1c at three time points (24 hpi, 36 hpi, and 48 hpi) ( Figure 5E). At 24 hpi, 28.4 ± 1.7% and 32.9 ± 5.3% of appressoria penetrated into barley cuticle cells and formed invasive hyphae (IH) structures with the wild type and complementation strain Monap1c. However, only 3.1 ± 1.8% of ∆Monap1 appressoria formed IH structures. At 36 hpi and 48 hpi, the appressorial penetration rate of ∆Monap1 was still lower than that of the wild type and the complementation strain Monap1c ( Figure 5F). Moreover, the complemented ∆Monap1 with the NAP1 gene of S. cerevisiae can restore the virulence of the ∆Monap1 mutant ( Figure 4D).

∆Monap1 Is Sensitive to Cell Wall Integrity Stresses and Resistant to Microtubule Stresses
The integrity of the cell wall, which is the first barrier contacting the external environment, plays a crucial role in the homeostasis of M. oryzae [28]. To assess the role of MoNAP1 in cell wall integrity stress, the growth of the wild-type 70-15, ∆Monap1 and Monap1c strains was measured in CM with three cell wall integrity stresses, including 75 µg mL −1 calcofluor white (CFW), 600 µg mL −1 Congo red (CR), and 0.004% sodium dodecyl sulfate (SDS) ( Figure 6A). The relative growth rate of mycelial growth of ∆Monap1 was significantly decreased compared to that of the wild type on CFW and CR ( Figure 6B). In contrast, in the presence of 0.004% SDS, ∆Monap1 showed an increased relative growth rate compared to that of the wild type. In addition, ∆Monap1 was more resistant to bleomycin (BLM) than the wild type ( Figure 6A,B).
In S. cerevisiae, the ∆nap1 mutant showed increased resistance to benomyl, a microtubule destabilizing drug that causes mitotic arrest at high concentrations, suggesting that Nap1 plays a role in regulating microtubule dynamics [19]. Similar to yeast, the ∆Monap1 mutant grew faster than the wild type in plates containing 15 µg mL −1 benomyl in M. oryzae ( Figure 6A,B). In addition to benomyl, ∆Monap1 was more resistant to high temperature. ∆Monap1 exhibited a lower growth defect than the wild type at 32 • C ( Figure 6A,B). When wild-type conidia are treated with HU, a DNA replication inhibitor, appressorium formation is blocked [29]. We inoculated wild-type 70-15, ∆Monap1 and Monap1c in CM containing 2 mM HU and found that ∆Monap1 had a higher relative growth rate than the wild type and Monap1c ( Figure 6A,B).

3, 7662 9 of 21
GraphPad Prism 8.0 and significant differences compared with the wild type were estimated by multiple t tests: ** p < 0.01.

∆Monap1 Is Sensitive to Cell Wall Integrity Stresses and Resistant to Microtubule Stresses
The integrity of the cell wall, which is the first barrier contacting the external environment, plays a crucial role in the homeostasis of M. oryzae [28]. To assess the role of MoNAP1 in cell wall integrity stress, the growth of the wild-type 70-15, ∆Monap1 and Monap1c strains was measured in CM with three cell wall integrity stresses, including 75 μg mL −1 calcofluor white (CFW), 600 μg mL −1 Congo red (CR), and 0.004% sodium dodecyl sulfate (SDS) ( Figure 6A). The relative growth rate of mycelial growth of ∆Monap1 was significantly decreased compared to that of the wild type on CFW and CR ( Figure 6B). In contrast, in the presence of 0.004% SDS, ∆Monap1 showed an increased relative growth rate compared to that of the wild type. In addition, ∆Monap1 was more resistant to bleomycin (BLM) than the wild type ( Figure 6A,B).

Nap1 Is Necessary for Proper Septin Ring Organization in M. oryzae
A critical requirement for appressorium morphology and function in M. oryzae is the recruitment and organization of septin-dependent cytoskeletal components [30]. In C. albicans, Nap1 plays a role in septin ring organization, and the deletion of NAP1 affects the stability of the septin ring [18]. The abnormal appressorial morphology of ∆Monap1 prompted us to observe the localization of the septin ring in the appressorium. We visualized the septin ring by expressing Sep3-GFP and Sep5-GFP fusion proteins in the wild type and mutants. In the wild type, Sep3-GFP and Sep5-GFP showed a ring structure around the appressorium pore, while in the ∆Monap1 mutant, Sep3-GFP and Sep5-GFP were evenly distributed and appeared similar to a round pie at the bottom of the appressorium at 24 hpi (Figure 7). stability of the septin ring [18]. The abnormal appressorial morphology of ∆Monap1 prompted us to observe the localization of the septin ring in the appressorium. We visualized the septin ring by expressing Sep3-GFP and Sep5-GFP fusion proteins in the wild type and mutants. In the wild type, Sep3-GFP and Sep5-GFP showed a ring structure around the appressorium pore, while in the ∆Monap1 mutant, Sep3-GFP and Sep5-GFP were evenly distributed and appeared similar to a round pie at the bottom of the appressorium at 24 hpi (Figure 7).

Nap1 Binds Histones H 2 A and H 2 B and G2/Mitotic-Specific B-Type Cyclin Cyc1 in M. oryzae
As a nucleosome assembly protein, Nap1 can tightly bind histones H 2 A and H 2 B in yeast [31], Drosophila [21], and tobacco [32]. To characterize its biochemical properties in M. oryzae, we detected whether MoNap1 bound core histones. Using an in vitro GST pull-down assay, we identified the interaction between MoNap1 and histones H 2 A and H 2 B. The recombinant proteins GST-H 2 A and GST-H 2 B pulled down FLAG-MoNap1, but GST did not ( Figure 8A). Furthermore, we detected in vivo protein interactions between MoNap1 and H 2 A using a bimolecular fluorescence complementation (BiFC) assay. MoNap1 was fused to the C-terminal fragment of YFP (YFP CTF ), and H 2 A was fused to the N-terminal fragment of YFP (YFP NTF ). The two vectors expressing MoNap1-YFP CTF and YFP NTF -H 2 A were cotransformed into the wild type; as negative controls, MoNap1-YFP CTF and YFP NTF as well as YFP CTF and YFP NTF -H 2 A were cotransformed into the wild type. Only transformants coexpressing MoNap1-YFP CTF and YFP NTF -H 2 A showed YFP signals in the cytoplasm, but the transformants coexpressing two pairs of control vectors did not ( Figure 8B).
In budding yeast, Nap1 acts with Clb2 to perform mitotic functions and suppress polar bud growth [19]. In M. oryzae, a B-type cyclin gene, CYC1, is the homolog of CLB2 [33]. The different sensitivities of ∆MoNap1 to benomyl and temperature prompted us to investigate the relationship between Nap1 and the Cyc1 protein. In the pull-down experiments, MoNap1 was pulled down by Cyc1 ( Figure 8C). Thus, MoNap1 physically interacted with Cyc1 in M. oryzae. N-terminal fragment of YFP (YFP NTF ). The two vectors expressing MoNap1-YFP CTF an YFP NTF -H2A were cotransformed into the wild type; as negative controls, MoNap1-YFP C and YFP NTF as well as YFP CTF and YFP NTF -H2A were cotransformed into the wild type. Onl transformants coexpressing MoNap1-YFP CTF and YFP NTF -H2A showed YFP signals in th cytoplasm, but the transformants coexpressing two pairs of control vectors did not (Figur 8B).

The NES of MoNap1 Is Dispensable for Localization and Binding the Histone Core Proteins H 2 A and H 2 B but Required for Growth and Pathogenicity
In both yeast and Drosophila, Nap1 has been reported to be a nucleocytoplasmic shuttling protein. However, in M. oryzae, MoNap1 mainly localized in the cytoplasm of conidial, appressorial, and hyphal cells ( Figure 9A). Additionally, the treatment with leptomycin B (LMB, a specific nuclear export inhibitor) [34] did not result in the obviously increased accumulation of Nap1-GFP in the nucleus ( Figure 9B).

The NES of MoNap1 Is Dispensable for Localization and Binding the Histone Core Proteins H2A and H2B but Required for Growth and Pathogenicity
In both yeast and Drosophila, Nap1 has been reported to be a nucleocytoplasmic shuttling protein. However, in M. oryzae, MoNap1 localized in the cytoplasm of conidial, appressorial, and hyphal cells ( Figure 9A). Additionally, the treatment with leptomycin B (LMB, a specific nuclear export inhibitor) [34] did not result in the accumulation of Nap1-GFP in the nucleus ( Figure 9B).  In yeast, Nap1 shuttles via a leucine-rich NES [20], and this short peptide sequence is conserved in many species, including nematode Nap1 (nNap1), Drosophila Nap1 (dNap1), and human Nap1 (hNap1). MoNap1 contains a 16-amino acid sequence in the N-terminal region that is similar to the NES of the yeast Nap1 protein ( Figure 10A). To investigate whether these sequences function as active NES signals, we constructed mutated Nap1 ∆NES in which Nap1 lacked the corresponding NES region. However, the deletion of the NES sequence in MoNap1-GFP (MoNap1 ∆NES -GFP) did not make Nap1 preferentially localize in the nucleus ( Figure 10B).
We examined the possible relationship between histone binding and the NES sequence of Nap1. Therefore, we performed GST pull-down assays with FLAG-MoNap1 ∆NES and core histones. We found that MoNap1 ∆NES bound core histones similar to MoNap1 ( Figure 10C). Therefore, the NES region is not involved in H 2 A/H 2 B binding.
Nevertheless, we found that Monap1 ∆NES exhibited a phenotype similar to that of ∆Monap1, including reduced growth and virulence ( Figure 10D-F).
is conserved in many species, including nematode Nap1 (nNap1), Drosophila Nap1 (dNap1), and human Nap1 (hNap1). MoNap1 contains a 16-amino acid sequence in the N-terminal region that is similar to the NES of the yeast Nap1 protein ( Figure 10A). To investigate whether these sequences function as active NES signals, we constructed mutated Nap1 ∆NES in which Nap1 lacked the corresponding NES region. However, the deletion of the NES sequence in MoNap1-GFP (MoNap1 ∆NES -GFP) did not make Nap1 localize in the nucleus ( Figure 10B).

Discussion
Nap1 is a well-conserved protein in eukaryotes, such as mammals, Drosophila, and yeast; however, its roles in filamentous fungi are not well understood. M. oryzae, an important plant pathogenic fungus, is often used as a model organism in investigating pathogenic fungal-host molecular interactions. In this study, we characterized the biological roles of Nap1 in M. oryzae by knocking out NAP1 and found that Nap1 plays diversified roles in aerial hyphal development, conidiophore and spore differentiation, appressorium formation, and virulence. NAP1 in S. cerevisiae can eliminate the defects in the sporulation and virulence of the ∆Monap1 mutant, suggesting that Nap1s in M. oryzae and S. cerevisiae are homologous in both biological function and protein sequence.
In M. oryzae, ∆Monap1 displayed reduced spore production, abnormal appressoria, and decreased infection growth and virulence. The reduced conidiation in ∆Monap1 is due to its fewer conidiophores and fewer spore numbers on each conidiophore. In M. oryzae, several transcription factor genes (COS1, CONX2, MSTU1, GTA1, GCC1, FLBC, HOX2, and CON7) [35][36][37] and autophagy genes, such as ATG5 [38] and TEA4 [39], are involved in the regulation of sporulation. The mRNA levels of two genes, FLBC and CON7, were significantly downregulated, suggesting a possible pathway regulating conidiation via FLBC and CON7. In addition, melanin promotes sporulation in the wild type 70-15 [40]. The lower melanin contents in the aerial hyphae of ∆Monap1 are also responsible for its reduced spore production. M. oryzae penetrates the rice cell wall by a melanized, domeshaped appressorium with a large turgor. In ∆Monap1, the appressorium morphology is abnormal and smaller than that in the wild type. Normal appressorium function depends on the microtubule arrangement and F-actin polymerization organized by septin [41]. The septin ring determines the polarity of the appressorium, which is required for plant penetration [42]. Septin GTPases can regulate the organization and functions of microtubule [43]. Moreover, in mammals, the stability of septin disks is dependent on intact microtubules [44]. In C. albicans, Nap1 is associated with the assembly of septin rings, and NAP1 deletion leads to changes in septin dynamics [18]. In tobacco, Nap1 colocalizes with the mitotic spindle and the phragmoplast [12] and interacts with tubulins [32]. Benomyl, a microtubule inhibitor, destabilized microtubules in cells [45]. In M. oryzae, the deletion of MoNAP1 led to increased resistance to benomyl and an increased proportion of abnormal appressoria, suggesting its function in microtubule arrangement. Abnormal cytoskeleton and appressorium morphology cause an abnormal septin ring [30,46]. ∆Monap1 showed an abnormal distribution of Sep3 and Sep5 at the bottom of the appressorium at 24 hpi. This finding is consistent with the aberrant appressorium morphology of ∆Monap1. Thus, the reduced virulence of ∆Monap1 may be the result of an abnormal appressorium morphology, which is probably related to the altered microtubule dynamics.
A dramatic change in microtubule dynamics occurs when cells enter mitosis. Spindle orientation is a microtubule-dependent process during mitosis that determines the division plane. Mitosis is induced specifically by kinase complexes that contain B-type cyclins and a cyclin-dependent kinase [19]. In yeast, Nap1 interacts with the B-type cyclin Clb2, which is involved in the assembly of the mitotic spindle. Nap1 is required for the ability of Clb2 to suppress polarized bud growth. In M. oryzae, Nap1 interacts with the B-type cyclin Cyc1. The deletion of NAP1 alters the effect of benomyl on microtubule stability and alters microtubule stability during mitosis. In budding yeast, when the wild-type cells were cultured in media containing 11 µg mL −1 benomyl, the cells were unable to assemble a functional mitotic spindle. However, ∆nap1 cells do not exhibit a benomyl-induced mitotic delay [19]. This finding indicates that the deletion of NAP1 either increases microtubule stability or reduces the requirements for microtubule function during mitosis. In M. oryzae, when conidia are treated with HU, a DNA replication inhibitor, the cells are arrested at the S phase, and appressorium formation is blocked [29,33]. We found that ∆Monap1 also had an increased resistance to HU. Temperature is one of the key factors affecting division of cells, and high temperature inhibits the cell cycle in Chlamydomonas reinhardtii [47]. In A. nidulans, the NimA gene encodes a protein kinase necessary for mitosis [48]. An equivalent gene mutant MonimA E37G in M. oryzae grew normally at 26 • C but showed a reversible growth defect at 32 • C [49]. However, ∆Monap1 exhibited a higher relative growth rate at 32 • C. Therefore, it is possible that MoNap1 is involved in the regulation of mitosis in M. oryzae.
Chitin [(1-4)-β-linked N-acetylglucosamine] is a major structural cell wall component in ascomycetes, and most chitin is fully acetylated and associated with (1-3)-β-glucan and (1-6)-β-glucan [50,51]. In this study, ∆Monap1 was sensitive to two chitin-binding anionic dyes, CFW and CR, which inhibited the assembly of enzymes that connect chitin to (1-3)-β-glucan and (1-6)-β-glucan [52], indicating that the loss of MoNAP1 affects the CWI pathway. The CWI pathway controls the cellular remodeling process in response to environmental challenges in fungi [28,52]. Bleomycin (BLM) is a DNA-damaging agent that acts by not only inducing DNA strand breaks [53] but also destroying cell wall components through oxidation and preventing the formation of fungal septum and cytokinesis [54][55][56]. The nrp1-1 nrp2-1 mutant plants were significantly more sensitive to the bleomycin treatment and increased levels of DNA damage than the wild-type plants in Arabidopsis [57]. Conversely, our results showed that ∆Monap1 increased resistance to BLM. The difference in the response to BLM may be due to the differences in nucleus and cell wall components or the differences in action sites of BLM between Arabidopsis and M. oryzae.
The localization of Nap1 in the cell has been an interesting issue, and Nap1 exhibits nucleocytoplasmic shuttling in many species. The distinct subcellular localization of Nap1 in the cell reflects functional diversity and complexity. Nap1 is a nucleocytoplasmic shuttling protein in yeast that is primarily localized in the cytoplasm at the steady state but localized in the nucleus when a leucine-rich nuclear export sequence (NES) is mutated [20]. Nap1 binds histone H 2 A/H 2 B dimers in the cytoplasm and participates in the import of histones into the nucleus [58]. The analysis of Nap1 phosphorylation revealed that Nap1 is phosphorylated at 11 sites in vivo in S. cerevisiae [59]. In vitro, Nap1 is a substrate for the phosphorylation of CK2 at three serines. Normal S-phase progression requires the reversible phosphorylation of Nap1, and the phosphorylation of Nap1 by CK2 appears to facilitate its import into the nucleus [59]. However, we did not observe increased nuclear localization of MoNap1 after deleting a putative NES, which is consistent with the observation in C. albicans, where the deletion of the NES also did not result in NAP1 accumulation in the nucleus [18]. Interestingly, in the BiFC assay, we found that the fluorescent signal was also mainly concentrated in the cytoplasm, which may be related to the fact that NAP1 binds histones in the cytoplasm to form a complex and participates in histone import into the nucleus [60].
In summary, we characterized the roles of Nap1, nucleosome assembly protein 1, in M. oryzae by molecular genetic assays and found that MoNAP1 is required for fungal development and virulence probably through its involvement in microtubule dynamics.

Strains and Culture Conditions
Magnaporthe oryzae wild-type strain 70-15 (a strain artificially crossed in the laboratory) and the ∆Monap1 mutant, the complementary strain of ∆Monap1 (Monap1c), were cultured in complete medium (CM) or minimal medium (MM) at 25 • C under a 16 h light and 8 h dark cycle [61]. For the stress tests, the strains were cultured in CM supplemented with 0.004% sodium dodecyl sulfate (SDS), 75 µg mL −1 calcofluor white (CFW), 600 µg mL −1 Congo red (CR), 35 µg mL −1 bleomycin (BLM), 15 µg mL −1 benomyl, or 2 mM hydroxyurea (HU) at 25 • C in darkness for 9 d. For the temperature-sensitive stresses, the strains were cultured in CM at 32 • C under a 16 h light and 8 h dark cycle for 9 days.

Knockout of NAP1 in M. oryzae
An upstream fragment and a downstream fragment of NAP1 from M. oryzae genomic DNA were amplified with the primers Up-F/Up-R and Down-F/Down-R, respectively. The primers used in this study are listed in Supplementary Table S1. A hygromycin B resistance gene, HPH, was cloned using the primers HPH-F and HPH-R. Three DNA fragments were fused into a knockout cassette in HindIIIand XbaI-digested pKO3B [62] using a fusion enzyme (Vazyme, Nanjing, China). Then, the knockout cassette was transformed into the wild-type strain via Agrobacterium tumefaciens-mediated transformation (ATMT), and the null mutants were confirmed using a previously reported method [36,62]. First, the transformants were screened in selective CM containing 0.5 µM 5-fluoro-2 -deoxyuridine (F2dU) and 200 µg mL −1 hygromycin B. Second, the genomic DNA of the transformants was isolated using an improved CTAB method. Then, the null mutants were identified using double PCR in which MoNAP1 was amplified using the primer set S-F/S-R with β-TUBULIN (primers: Tbl-gF/Tbl-gR) as a positive control. The successful recombination of HPH into the deleted MoNAP1 site was confirmed by another PCR using the primer set L-F/HPH-CKR. Third, the insertion copy number of HPH in the mutant genome was determined by quantitative real-time PCR (qPCR) using β-TUBULIN as a control (primer sets qHPH-F/qHPH-R and qtub-F/qtub-R).
For the complementary strain Monap1c, a 3.68 kb native full NAP1 gene containing a promoter, coding sequence (CDS), and a terminator was cloned from M. oryzae wildtype genomic DNA by the primer set NAP1c-F/NAP1c-R and inserted into the EcoRI and XbaI-linearized vector pKD3, which contains a bialaphos resistance gene (BAR) [63]. The complemented DNA fragment was transformed into ∆Monap1 via ATMT, and the transformants were selected in selection medium containing 750 µg mL −1 glufosinate ammonium. The expression of MoNAP1 in the complementation strains was confirmed at the mRNA level by reverse transcription polymerase chain reaction (RT-PCR). To build a mutated MoNAP1 in which the NES sequence was deleted (MoNAP1 ∆NES ), two segments (NAP1 1−430 bp and NAP1 478−1549 bp ) with primers (NAP1 ∆NES -F1/NAP1 ∆NES -R1 and NAP1 ∆NES -F2/NAP1 ∆NES -R2) were amplified in the genomic DNA of M. oryzae, fused to the BamHI-XbaI site of pKD3, and then transformed into ∆Monap1 via ATMT. To complement the NAP1 of S. cerevisiae into M. oryzae, a 2.35 kb NAP1 gene was cloned from S. cerevisiae genomic DNA using the primer set yNAP1-F/yNAP1-R and inserted into the BamHI and XbaI-linearized vector pKD5, which contains a sulfonylurea resistance gene (SUR) [63]. The resulting construct was transformed into ∆Monap1 by ATMT.

Phenotypic Characterization
The mutant phenotype was assayed according to previous reports [64,65]. To compare mycelial growth, a 5-mm agar block of 8-day-old wild type, ∆Monap1, and complementation strain Monap1c were inoculated in 7-cm plates containing 17.5 mL CM and cultured at 25 • C with a 16 h light and 8 h dark phase. The colony diameter was measured, and the colonies were photographed 9 days post-inoculation (dpi). Regarding sporulation, conidia were collected from 9-cm plates containing 30 mL CM at 12 dpi and counted with a counting chamber. Regarding conidiophore development, media containing vegetative hyphae were sliced into slender pieces and recultured under continuous light at 25 • C for 24 h [65]. To measure conidial germination and appressorium formation, 25 µL of spore suspension (1 × 10 5 conidia mL −1 ) were dropped on plastic coverslips and incubated at 22 • C under dark conditions. Conidial germination and appressorium formation were observed at 4 hpi and 24 hpi, respectively.
For the mycelial virulence assays, 5-mm mycelial pellets were inoculated on excised leaves of 7-day-old barley (Hordeum vulgare) for 4 days. The virulence assays of rice seedlings (Oryza sativa cultivar CO39) were performed by spraying spore suspensions (5 × 10 4 conidia mL −1 ) in 0.2% (w/v) gelatin on 14-day-old rice seedlings. The rice seedlings were cultured in a wet container under dark conditions at 22 • C for 48 h and then recultured under a 16 h light and 8 h dark cycle at 25 • C for 3 days. To assess the disease lesion severity (disease score), the area of a 5-cm-long leaf with the most serious disease lesions in each seedling was counted [65]. Regarding host penetration, 20 µL of spore suspensions (5 × 10 4 conidia mL −1 ) of the wild-type, ∆Monap1, and Monap1c strains were dropped onto 7-day-old barley leaves and cultured at 25 • C. At 24 hpi, 36 hpi, and 48 hpi, the leaves were collected, decolored by methanol, fixed in alcoholic lactophenol, and observed under a microscope [64].

Observation of Fluorescence Fusion Proteins in M. oryzae
The fluorescence fusion protein Nap1-GFP was constructed by cloning a 1.5-kb NAP1 via PCR with a primer set (NAP1-GFP-F/NAP1-GFP-R) from the wild-type genomic DNA and fused with GFP in the BamHI and SmaI sites of pKD3-GFP [46]. To express Nap1 ∆NES -GFP, two segments (NAP1 1−430 bp and NAP1 478−1549 bp ) of the NAP1 coding sequence were amplified from the genome with primers (NAP1 ∆NES -GFP-F1/NAP1 ∆NES -GFP-R1 and NAP1 ∆NES -GFP-F2/NAP1 ∆NES -GFP-R2) and fused together with GFP in the pKD3 vector digested with XbaI. The fluorescent fusion protein expression vector was transformed into ∆Monap1 via ATMT. The fluorescence signals were observed under a laser scanning confocal microscope (FV3000). To observe the effects of LMB on the localization of Nap1, 20 nM LMB were added to the conidial suspension, and the fluorescence signals were observed at 3 hpi, 8 hpi, and 24 hpi. To localize Sep3 and Sep5 in the wild type and ∆Monap1 mutant, the vectors pKD5-Sep3-GFP and pKD5-Sep5-GFP into which SEP3 or SEP5 was inserted into the pKD5-GFP vector were transformed into the wild type and ∆Monap1, respectively [46].

Pull-Down Assays and Bimolecular Fluorescence Complementation (BiFC) Assay
For the pull-down assays, the cDNA fragments of MoNAP1 and MoNAP1 ∆NES were inserted into pET21 containing a 3 × FLAG tag, and the cDNA fragments of H 2 A, H 2 B, and CYC1 were inserted into pGEX-4T containing a GST tag. The expression vectors were transformed into Escherichia coli BL21 separately. A BiFC analysis was used to observe the protein-protein interactions in vivo in M. oryzae. The plasmids used in the BiFC assay were constructed as previously described [66]. In brief, a TUBULIN promoter and a C-terminal fragment spanning approximately amino acids 155 to 238 of YFP (YFP CTF ) were inserted into the vector pKD9-HPH to form the plasmid pKD9-YFP CTF -HPH. The plasmid pKD9-YFP CTF -HPH was digested with BamHI and SmaI and then ligated with the MoNAP1 fragment amplified with a primer pair (NAP1-YFP CTF -F/NAP1-YFP CTF -R), resulting in the final pKD9-MoNAP1-YFP CTF -HPH plasmid. Another N-terminal fragment spanning approximately amino acids 1 to 154 of YFP (YFP NTF ) was inserted into the vector pKD5-SUR and a TUBULIN promoter to form the plasmid pKD5-YFP NTF -SUR. The plasmid pKD5-YFP NTF -SUR was digested with SalI and then ligated with the H 2 A fragment amplified with a primer pair (H 2 A-YFP NTF -F/H 2 A-YFP NTF -R), resulting in the final pKD5-YFP NTF -H 2 A-SUR plasmid. Next, the pKD9-MoNAP1-YFP CTF -HPH fusion vector was transformed into the 70-15 strain in combination with pKD5-YFP NTF -H 2 A-SUR. The transformants were screened in CM containing hygromycin B and sulfonylurea, and at least three independent transformants were examined under an LSM710nlo laser scanning confocal microscope (Zeiss, Jena, Germany). Funding: This work was funded by the Key R&D projects of Zhejiang Province (grant no. 2021C02010) and the National Natural Science Foundation of China (grant nos. 31871908). This work was also supported by a grant Organism Interaction from Zhejiang Xianghu Laboratory to FCL.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: All data supporting the findings of the current study are available within figures and supporting information. All strains generated during this study are available from the corresponding author upon reasonable request.