To obtain the homolog of Vps41 sequence from M. oryzae genome database, we used the amino acid sequence of budding yeast to conduct BLASTp search through the now defunct Magnaporthe genome resource¹ and successfully identified Vps41 homologous protein (1357-amino acid) encoded by MGG_03313.7 (MoVps41) and proceeded to perform domain prediction with http://prosite.expasy.org/. Our domain prediction analysis identified two domains motifs namely; the clathrin heavy chain repeat (CHCR) domain and the RING-H2 motif (Supplementary Figure S1). Further sequence alignment analysis showed that MoVps41 shares high sequence homology with the Vps41 proteins from Saccharomyces cerevisiae, Fusarium graminearum, Aspergillus nidulans, Arabidopsis thaliana, Mus musculus, and Homo sapiens (Supplementary Figure S2). The CHCR domain is highly conserved among all the organisms analyzed, with 38 to 72% identity between M. oryzae and mammals, plants and fungi (Supplementary Table S2), whilst the RING-H2 motif appeared absent in the Vps41 sequences from Saccharomyces cerevisiae and A. nidulans (Supplementary Figure S1A and Table S2).

Upon successful generation and confirmation of MoVPS41 deletion mutant (ΔMovps4-18, ΔMovps4-33, and ΔMovps4-34) mutant and complementation (Figure 1), we proceeded further to evaluate growth characteristics of the ΔMovps41 mutant alongside the wild-type and the complemented strain on three different culture medium which includes; CM, CMII, and MM medium. Corresponding results obtained from this assay showed that the deletion of MoVPS41 gene impacted negatively on the vegetative growth of ΔMovps41 mutant and resulted in 20, 22, 64% reduction in the growth of the ΔMovps41 mutant on CM, CMII, and MM medium, respectively (Figures 2A,B), additionally, we showed that the re-introduction of the wild-type MoVPS41 gene into the ΔMovps41 deletion mutant resulted in the generation of ΔMovps41/MoVPS41 complemented strain (C-21) with growth characteristics similar to that of the wild-type strain (Figure 2). From these results we inferred that M. oryzae MoVPS41 is deletable and further positioned that MoVPS41 contribute significantly to efficient vegetative development of the rice blast fungus.

In plant, Vps41 showed high expression activities in vegetative cells and sperm cells of pollen tubes during reproductive development and has been identified as a crucial factor regulating pollen tube–stigma interaction (Hao et al., 2016). To evaluate the possible contribution of MoVps41 to the sexual reproductive life of M. oryzae, we crossed wild-type strain Guy11 (MAT1-2) and mutant strain ΔMovps41 and the ΔMovps41/MoVPS41 complementation strain (C-21) with a standard mating type tester strain KA3 (MAT1-1) by culturing them on oatmeal (OMA) plates under optimum conditions required for perithecia production. Results obtained from this assay showed that Guy11 × KA3 and C-21 × KA3 crosses yielded numerous dark perithecia, whilst, no perithecia was observed under ΔMovps41 × KA3 crosses (Figure 3A). We also examined asexual reproduction characteristics of the MoVPS41 deletion mutant by assessing number and morphology of conidia produced by the ΔMovps41 mutant. Corresponding results obtained from these conidiation assays showed that the quantity of conidia produced by ΔMovps41 mutant was significantly low and constituted less than 0.1% of the quantity of conidia produced by the wild-type and C-21 complementation strain (Figure 3B). Additionally, most of the conidia produced by the ΔMovps41 mutant exhibited septation defects and our records showed 40% of conidia produced by the ΔMovps41 mutant are associated with single septum while conidium without septum accounts for 25.5% and the remaining 34.5% exhibited normal septation (Figures 3C,D). We also performed comparative conidiophore development assay to ascertain if the observed drastic reduction in the quantity of conidia produced by the ΔMovps41 mutant has any relationship with conidiophore-genesis. Results, obtained from this assay showed that the deletion of MoVPS41 equally triggered drastic reduction in the formation of conidiophore the ΔMovps41 mutant (Figure 3E). From these result we posited that MoVPS41 play crucial role in regulating conidiophore-genesis, conidia morphogenesis, sexual and asexual reproduction in M. oryzae.

Because of deficiency in conidiation of the deletion mutants, we conducted pathogenicity assay by inoculating rice and barley seedlings with mycelium plugs obtained from the ΔMovps41 mutant and the Guy11. This investigation showed that unlike the wild-type strain, the ΔMovps41 mutant failed to induce symptoms of blast disease or lesions on susceptible rice seedling (CO39) host plant, while severe symptoms appeared 5-hpi on leaf tissues inoculated with culture block of wild-type and the C-21 complementation strain (Figure 4A). Because barley is also susceptible to M. oryzae blast infection, we decided to examine the virulence of ΔMovps41 on barley leaves with or without wound. From this investigation we noticed that the ΔMovps41 mutant consistently failed to induce blast lesion on both intact and injured barley leaves (Figure 4B). We additionally carried-out root infection assay with rice seedlings, and consistent with results obtained from leaf infection trials, the ΔMovps41 mutant failed to trigger discoloration near the inoculation sites on rice root (Figure 4C). However, the wild-type and complemented C-21 strain readily induced extensive discoloration on rice root with under the same inoculation and incubation conditions, (Figure 4C), from these results we inferred that MoVPS41 mediated vacuolar fusion activities exerts great influence on the pathogenesis of the rice blast fungus by either directly or indirectly facilitating the secretion of virulence factors in M. oryzae.

Records have shown that besides conidia mediated infection, M. oryzae also invades host tissues by deploying appressorium-like structures that develop at hyphal tips (Kong et al., 2013). However, from our pathogenicity assay we noticed that ΔMovps41 failed to initiate hyphae mediated infection on intact leaves of barley and rice seedlings. In view of this observation, we decided to monitor the development of hyphae tip appressorium-like structures in the ΔMovps41 mutant strain and the wild-type strain during growth on glass slide. Results obtained from this examination showed that the ΔMovps41 mutant has no defect with regards to the formation of appressorium-like structures at hyphae (Figures 5A,B). We proceeded further to monitor the penetration abilities of the appressorium-like structures produced by ΔMovps41 mutant by inoculating barley leaves with vegetative hyphae produced by the MoVPS41 deletion mutant. Findings from microscopy examination after 48 hours post inoculation (hpi) revealed that the appressorium-like structures produced by the ΔMovps41 mutant were unable to penetrate barley epidermal cells, whilst appressorium-like structures produced by the wild-type and complemented C-21 strains successfully penetrated barley epidermal cells and formed massive bulbous invasive hyphae inside the invaded epidermal cells (Figure 5C). Additional results obtained from incipient cytorrhysis assays surprisingly showed that the turgor generated by the appressorium-like structures produced by the ΔMovps41 mutant was significantly higher than the level of turgor exhibited by the wild-type and complemented C-21 strains (Figure 5D). From these observations, we subsequently concluded that MoVps41 is required for enforcing the functionality of hyphae tip appressorium-like structures as a potent propagule for initiating hyphae mediated development of rice blast disease.

To further establish the influence of MoVPS41 on the development of conidia base appressorium, we accordingly carried-out appressorium formation assay on appressorium inducing hydrophobic surface with the few conidia obtained from the ΔMovps41 mutant along the wild-type and the C-21 strain. This investigation showed that the ΔMovps41 mutant could form intact appressoria of similar morphology with the appressorium produced by the wild-type and the C-21 strain (Figures 6A,B). However, results obtained from conidia mediated penetration assay conducted by inoculating barley leaves with conidia suspensions prepared with conidia obtained from the ΔMovps41 mutant, the wild-type and C-21 strain, also showed that although the ΔMovps41 mutant could form appressorium on host leaf surface, it was, however, unable to penetrate barley epidermal cells even after extended period of 72-hpi (Figure 6C). These results showed that MoVPS41 is dispensable of appressorium formation in M. oryzae, taken these results together we proposed that MoVPS41 contribute to the development of blast infection by facilitating efficient penetration of host tissues. However, the detailed molecular mechanism underlining this observation is still obscure at this stage and highlights the need for more comprehensive studies on the HOPS complex in the rice blast fungus in the future.

To ascertain the impact of MoVPS41 deletion on vacuolar fusion process in the ΔMovps41 mutant, we observed the morphological pattern of vacuoles in the vegetative hyphae of ΔMovps41 mutant and the wild-type cultured in liquid CM media for 2-days with the aid of transmission electron microscopy (TEM). Results from our microscopy analysis identified numerous and characteristically smaller vacuoles in the ΔMovps41 mutant. However, the vacuoles observed in the wild-type and the complemented strains were typically large (Figure 7A). To further confirm this observation, we stained hyphae produced by the ΔMovps41mutant, Guy11, and complemented strain with CMAC (7-amino-4-chloroethylcoumarin) dye which selectively stains fungal vacuoles. Our vacuole staining bio-assay also stained numerous smaller vacuoles in MoVPS41 mutant while, few but larger size vacuoles were recorded for the wild-type, and the C-21 complemented strain (Figure 7B). These results adequately showed that MoVPS41 is required for homotypic vacuole fusion processes in M. oryzae.

To investigate the sub-cellular localization of MoVps41 protein, a MoVps41-GFP fusion construct generated in our study was transformed into the ΔMovps41 mutant. After screening by PCR, one transformant MOVps41-GFP expressing the MoVps41-GFP construct under control of its native promoter was identified. Similar to the complemented strain C-21, all defects displayed in the ΔMovps41 mutant were rescued in the ΔMovps41/MoVps41-GFP transformant MOVps41-GFP, indicating that the fusion constructs MoVps41-GFP is functional. We used CMAC to stain the hyphal vacuoles of ΔMovps41/MoVps41-GFP transformant MOVps41-GFP and observed the GFP signal under an epifluorescence microscope. MoVps41-GFP signals were observed on the membrane of the vacuoles, and the signals were stronger at the interface between two vacuoles (Figures 8A,B), suggesting that MoVps41 is involved in the homotypic fusion of vacuoles. We also investigated the localization of MoVps41 during appressorium formation on hydrophobic surface and showed that MoVps41 retains its localization to vacuolar membrane during conidia germination and appressoria formation. Furthermore, we examined the localization pattern of MoVps41 during progressive endocytosis and endosomal trafficking processes in conidia, germ tubes and appressoria at time points (0, 4, and 12 h) of appressorium development on a hydrophobic surface by staining with endocytosis oriented dye FM4-64 which stains depict the transport activities occurring on vacuolar membrane during progressive active endocytosis. Consistent with the localization observed for hyphae, conidia, germ tube and appressoria, this also showed FM4-64 epifluorescence signal co-localized with green Florence protein (GFP) signal on vacuolar membranes and exhibited the strongest expression signal at membrane contact sites between two vacuoles (Figure 8C). These results all suggest that MoVps41 is involved in homotypic vacuolar fusion.

Generally, metal ions, including Ca²⁺, Cu²⁺, Fe²⁺, Mn²⁺, and Zn²⁺ play an essential physiological role in the development of fungi (Wang et al., 2003). However, research demonstrations have shown that, the intracellular accumulation of these cations beyond certain physiological limit turn to impact negatively (metal ion toxicity) on fungal growth and development. Sequestering activities of the vacuole play an important role in shielding cells from the harmful effects of toxic metal ions and also crucially mediate the detoxification of these metal ions (Ramsay and Gadd, 1997). To examine the contribution of MoVps41 in vacuolar sequestering and subsequent detoxification of metal ions, we monitored the vegetative growth ΔMovps41 mutant, Guy11 wild-type and the C-21 complemented strain on 1/4 YG growth medium supplemented with defined concentration of divalent metal ions. From this investigation, we observed that ΔMovps41 mutant was highly sensitive to all the metal ions tested and accounted for about 68 to 92% growth inhibition in the ΔMovps41 mutant with 1 mM Cu²⁺ exerting greatest biological growth inhibitory effect (71.2%) on ΔMovps41 mutant compared to 51 and 54% growth inhibition recorded for C-21 complemented strain and Guy11, respectively (Figures 9A,B). These results imply that MoVps41 plays a crucial role in ensuring cellular metal ions homeostasis by facilitating vacuolar mediated sequestering and detoxification of metal ions.

The prevailing osmotic gradient across organelles and membranes has been cited as crucial factors modulating the rate of membrane fission and fusion processes (Brett and Merz, 2008). Accumulating evidence from previous research findings has shown that the signaling activities of HOPS complex proteins are enhanced in response to oxidative stress induced by different types of osmolytes (Brett and Merz, 2008). We therefore evaluated the sensitivity of ΔMovps41 mutant to both osmotic and oxidative stresses by culturing the ΔMovps41 mutant, Guy11 and the C-21 strains on CM plates supplemented with various types of osmotic and oxidative stress inducing osmolytes including; 1 M glycerine, 1 M sorbitol, 1 M NaCl, 1 M KCl, and 10 mM H₂O₂. This investigation revealed that ΔMovps41 mutant display high sensitivity toward all the osmolytes tested, however, we noticed that NaCl, KCl, and CFW exerted the greatest inhibitory effect on the ΔMovps41 mutant (Figures 10A,B). Additionally, we investigated the role of MoVPS41 in enforcing cell wall integrity by monitoring the comparative growth performance of the ΔMovps41 mutant, Guy11 and the C-21 strains on CM plates supplemented with 400 μg/mL Congo red, 0.01% SDS, Compared to the wild-type Guy11 and the C-21 complemented strain, the ΔMovps41 mutant was moderately sensitive to SDS (Figures 10C,D). However, there was no significant difference observed in the growth of ΔMovps41 mutant, Guy11 and the C-21 complemented strain on CM plates supplemented with 400 μg/mL Congo red (Figure 10C). These results demonstrate that MoVps41 plays an essential role in stabilizing membrane integrity in the rice blast fungus under oxidative stress conditions.

The wild-type strain Guy11 and all transformants generated in this study were cultured on complete medium plates (CM: 0.6% yeast extract, 0.6% casein hydrolysate, 1% sucrose, 1.5% agar) at 25°C as described previously (Chen et al., 2008). Additional growth media used in this study, including minimal medium (MM: 6 g NaNO₃, 0.52 g KCl, 0.52 g MgSO₄, 1.52 g KH₂PO₄, 10 g glucose and 15 g agar in 1 L distilled water), 1/4 YG medium (1.3 g yeast extract, 5 g glucose, 15 g agar in 1 liter of distilled water) and oatmeal agar medium (OMA: 30 g oatmeal and 15 g agar in 1 L distilled water). For sexual reproduction, strains were cultured on oatmeal agar medium for 4 weeks (Karthikeyan and Gnanamanickam, 2008). For conidiation, the strains were cultured on rice bran medium (2% rice bran, 1.5% agar, pH 6.0) with a 12 h photophase for 7 days and removed all aerial hyphae and incubated the cultures under light for additional 48 h (Zhong et al., 2015). For DNA extractions and protoplast preparation, strains were grown in the liquid CM in a 150 rpm shaker at 25°C for 2–3 days. Sensitivity assay was conducted by culturing the respective strains on 1/4YG or CM plates supplemented with metal ionic stresses (1 mM Cu²⁺, 3 mM Zn²⁺, 3 mM Mn²⁺, 3 mM Fe²⁺ and 400 mM Ca²⁺), vesicular transport inhibitors (3 mM caffeine and 3 μg/mL monensin), osmotic stresses (1 M sorbitol, 1 M glycerine, 1 M NaCl, and 1 M KCl), oxidative stresses (10 mM H₂O₂), and cell wall stresses [0.01% SDS, 400 μg/mL Congo red and 200 μg/mL Calcofluor white (CFW)].

Split-marker approach (Goswami, 2012; Zheng et al., 2012) was applied to delete the MoVPS41 gene in M. oryzae. A 1.1-kb upstream and a 1.2-kb downstream flanking fragments of MoVPS41 were amplified with primer pairs VPS41-AF/AR and VPS41-BF/BR (Supplementary Table S1), respectively. The resulting PCR products were ligated with hygromycin phosphotransferase (hph) cassette fragments amplified with primers HYG/F+HY/R and YG/R+HYG/R (Supplementary Table S1) by overlapping PCR. For fungal transformation, protoplast preparation and polyethylene glycol (PEG)-mediated transformation of M. oryzae was performed as described by Zhang et al. (2016). Hygromycin-resistant transformants were screened by PCR with primer pairs VPS41-OF+VPS41-OR and VPS41-UA+H853 (Supplementary Table S1). Subsequently, putative ΔMovps41 mutants were analyzed by Southern blotting to confirm the gene replacement event. For complementation assays, a fragment containing the entire MoVPS41 gene, a 2-kb upstream and 1.4-kb downstream sequences was amplified with primers VPS41-CF and VPS41-CR (Supplementary Table S1). The resulting PCR products were co-transformed into ΔMovps41 mutant with pKNT vector. G418-resistant transformants ΔMovps41/MoVPS41 were confirmed by PCR and Southern blotting analysis.

For DNA extraction, genomic DNA was extracted from wild-type strain Guy11, ΔMovps41 deletion mutants and C-21 complemented transformants with the use of CTAB method described by Wang et al. (2017). For Southern blotting analysis, genomic DNA aliquots of 5 μg were completely digested with NcoI enzyme and separated with a 0.7% agarose gel and transferred onto Hybond N+ membrane (Amersham Pharmacia Biotech). The “A-fragments” were amplified with VPS41-AF and VPS41-AR primers (Figure 1 and Supplementary Table S1) and labeled by digoxin (DIG) as probe A. For southern blot analysis, probe labeling, hybridization and detection were performed with DIG High Prime DNA labeling and Detection Starter Kit I (Roche Applied Science), according to the instruction manual.

Conidia harvested from rice bran agar were counted by using hemocytometer and photographed by Olympus BX51 microscope (Olympus Corporation, Japan). Three independent experiments were performed with three replicates each time. For fertility assays, the strains crossed with strain KA3 were cultured on OMA at 25°C for the 7 days with a 12-h photophase and then placed under continuous white light at 16°C for 4 weeks. The border between the mated individuals was examined for the capacity to form perithecia. This experiment was repeated three times. Both leaf and root inoculation was carried-out by inoculating intact and injured barley (Gold Promise) and rice (CO39) leaves with a block media carrying mycelia before incubating them in a moist chamber at 25°C for 5–7 days as described by (Yan et al., 2011).

The development of appressorium-like structures at hyphal tips was assayed in accordance to the procedures described by Kong et al. (2013). To induce appressorium-like structure formation, culture blocks of 4-day-old CM cultures were placed on hydrophobic surfaces, and incubated in moist chamber for 72 h in dark. For penetration assays, the leaves of 10-day-old seedling of barley (Gold Promise) were inoculated with culture blocks of 4-day-old CM cultures for 48 h and subsequent penetration and in vivo development invasive hyphae was examined in line with procedures described by (Liu et al., 2010; Kong et al., 2013).

To generate the MoVPS41-GFP fusion construct, a 2.2-kb DNA fragment containing the entire MoVPS41 gene and its promoter region was amplified with primers VPS41-GF and VPS41-GR (Supplementary Table S1) and inserted into the APaI/HindIII site of pKNTG vector. The resulting fusion construct MoVPS41-GFP was transformed into ΔMovps41 mutant through PEG-mediated transformation. G418-resistant transformants were screened by PCR with primers VPS41-ORF and VPS41-GR (Supplementary Table S1) and examined for GFP signals.

For transmission electron microscopy, observation was carried out as previously described (Wang et al., 2011). For vacuolar staining, hyphae were prepared by growing fragmented mycelia in liquid CM for 3-days at 25°C with shaking at 150 rpm and inoculated with a vacuolar luminal dye CMAC (7-amino-4-chloromethylcoumarin, Sigma–Aldrich) at a final concentration of 10 μM for 20 min as previously described (Ohneda et al., 2002; Menke, 2011). Germinating conidia (0h, 4h and 12 h) on hydrophobic surface were incubated with 32 μM FM4-64 for 30 min to stain the vacuolar membrane. The vacuolar morphology of vegetative hyphae and germinating conidia were observed under the Nikon TiE system (Nikon, Japan).