MoSec13 combined with MoGcn5b modulates MoAtg8 acetylation and regulates autophagy in Magnaporthe oryzae

ABSTRACT

Macroautophagy/autophagy is an evolutionarily conserved cellular degradation process that is crucial for cellular homeostasis in Magnaporthe oryzae. However, the precise regulatory mechanisms governing autophagy in this organism remain unclear. In this study, we found a multiregional localization of MoSec13 to the vesicle membrane, endoplasmic reticulum, nucleus, and perinucleus. MoSec13 negatively regulated autophagy through specific amino acid residues in its own WD40 structural domain by interacting with MoAtg7 and MoAtg8. We also found that the histone acetyltransferase MoGcn5b mediated the acetylation of MoAtg8 and regulated autophagy activity. Subsequently, we further determined that MoSec13 regulated the acetylation status of MoAtg8 by controlling the interaction between MoGcn5b and MoAtg8 in the nucleus. In addition, MoSec13 maintained lipid homeostasis by controlling TORC2 activity. This multilayered integration establishes MoSec13 as an essential node within the autophagic regulatory network. Our findings fill a critical gap in understanding the role of Sec13 in autophagy of filamentous fungi and provide a molecular foundation for developing new therapeutic strategies against rice blast fungus.

ABBREVIATIONS BFA: brefeldin A; BiFC: bimolecular fluorescence complementation; CM: complete medium; CMAC: 7-amino-4-chloromethylcoumarin; Co-IP: co-immunoprecipitation; COPII: coat complex II; GFP: green fluorescent protein; HPH: hygromycin phosphotransferase; MM-N: nitrogen-starvation conditions; NPC: nuclear pore complex; PAS: phagophore assembly site; PE: phosphatidylethanolamine; UPR: unfolded protein response.

Introduction

Autophagy is an essential cellular process that plays a pivotal role in maintaining cellular homeostasis, regulating metabolic pathways, and mediating responses to stressors [1]. This highly conserved mechanism is primarily orchestrated by Atg (autophagy related) proteins, which coordinate the formation and function of autophagosomes. Among these Atg proteins, Atg8 is particularly crucial for the biogenesis of the autophagic membrane, facilitating the expansion and closure of the phagophore through its lipidation and interaction with various other autophagy-related proteins [2,3]. The newly synthesized Atg8 precursor protein undergoes cleavage by Atg4, resulting in the generation of free Atg8 protein harboring an active Gly residue. Subsequently, regulated by the ubiquitin activating-like enzyme Atg7 and the ubiquitin conjugating-like enzyme Atg3, Atg8 is conjugated to phosphatidylethanolamine (PE) to form Atg8–PE, which anchors to both the inner and outer membranes of the phagophore and remains attached to the inner membrane after autophagosome maturation [4]. Recent studies have highlighted the importance of post-translational modifications of Atg proteins [5,6], especially the acetylation of Atg8 in modulating its subcellular localization between the cytoplasm and the nucleus [7–9]. This dynamic shuttling is critical for the proper activation of autophagy, enabling cells to respond effectively to fluctuations in nutrient availability and cellular stress. Therefore, an in-depth understanding of the regulatory mechanisms of Atg8, including its acetylation, is essential for elucidating the molecular mechanisms of autophagy.

In recent years, researchers have debated the role of the coat complex II (COPII) in providing membrane components for autophagosome biogenesis. Traditionally, COPII vesicles are associated with exporting proteins from the endoplasmic reticulum (ER) to the Golgi apparatus. However, emerging evidence suggests that components of this machinery may also contribute to autophagosome formation [10–12]. The assembly of the COPII complex initiates with the activation of the small GTPase Sar1. In its GTP-bound active state, Sar1 induces spontaneous membrane curvature by inserting its N-terminal amphipathic helix into the membrane [13–16]. Following this, Sar1 recruits the Sec23-Sec24 heterodimer to form the “pre-budding complex”. Finally, the Sar1-Sec23-Sec24 complex interacts with the Sec13-Sec31 heterotetramer, facilitating membrane bending and producing a cage-like outer coat that results in the formation of a complete vesicle [17]. Importantly, the COPII complex is abundant in lysophospholipids, which enhance Sar1 binding to the membrane and promote COPII complex assembly [18]. Thus, a key aspect of COPII complex formation lies in its ability to decrease membrane curvature rigidity, a property predominantly regulated by lipid composition. This interplay between COPII components and membrane dynamics highlights the importance of COPII for the autophagic process and emphasizes the necessity of further exploring the functional role of COPII in cellular homeostasis.

The outbreak of rice blast disease causes extensive reductions in rice yields, posing a threat to global food security [19]. The causative agent of this devastating disease, M. oryzae, employs a crucial infection structure known as the appressorium to invade rice [20]. The development and functionality of the appressorium rely on the regulated process of autophagy, facilitating the degradation of stored materials within conidia to acquire essential nutrients [21]. Although autophagy has been recognized as an important process in the pathogenicity of M. oryzae, the specific role of the COPII complex in this context remains to be fully characterized.

In this study, we unveiled the intricate molecular mechanisms by which MoSec13 regulates autophagy. As a core component of both the COPII complex and the NPC complex, MoSec13 exhibits dual functionality, directly interacting with MoAtg7 and MoAtg8 via amino acid residues within its WD40 domain to modulate autophagy activity. Furthermore, MoSec13 extends its regulatory influence beyond canonical autophagy pathways by governing TORC2 signaling, thereby regulating lipid homeostasis. Parallel investigations revealed that MoGcn5b-mediated acetylation of MoAtg8 serves as a critical post-translational modification governing autophagic flux. Strikingly, MoSec13 significantly modulates the interaction between MoAtg8 and MoGcn5b, thereby controlling the acetylation status of MoAtg8. Collectively, these findings position MoSec13 as a multifunctional nexus integrating vesicle trafficking, nuclear transport, and lipid metabolism to regulate autophagy. Targeting this pathway may offer novel therapeutic strategies to combat M. oryzae-induced rice blast disease.

Results

MoSec13 interacts with MoAtg7 and MoAtg8 to negatively regulate autophagy

The coordination of the COPII system in response to the initiation of autophagosomes remains an enigma. In our previous study, we reported that the inner shell subunit of COPII vesicles, MoSec24B, is involved in regulating the fusion of late autophagosomes with vacuoles [22], which positions the COPII complex as one of the key players in autophagy. The unique ability of the COPII coat to selectively recognize and package specific membrane proteins and autophagy-related factors prompted us to investigate potential autophagy adaptor proteins. We focused our attention on MoSec13, the outer shell component of COPII, due to its multiple WD40 domain. The WD40 structural domain of Sec13 was conserved (Fig. S1A), and MoSec13 exhibited a closer genetic relationship to Sec13 in Neurospora crassa and Botrytis cinerea (Fig. S1B). Notably, MoSec13 consisted of five WD40 structural domains, whereas most Sec13 proteins contained six (Fig. S1C). This suggested that WD40 domains may be lost or modified during evolution in MoSec13.

The subcellular localization of MoSec13 was observed by microscopy, and we found that MoSec13 fused to the green fluorescent protein (GFP) colocalized with vesicle membrane structures stained with FM 4–64 at the conidia stage (Figure 1A). Using MoLhs1-DsRed as a marker to label the ER [23], we found that MoSec13-GFP-tagged vesicle membrane structures also colocalized with this organelle (Figure 1B). Although a small intranuclear pool of Sec13 has been observed previously, it has not been studied. Next, we introduced MoH₂B-mCherry as a nuclear marker and found that MoSec13-GFP localized in part to the nuclear membrane and nucleus (Figure 1C). These results suggest that MoSec13 localizes at intranuclear sites in addition to the ER and vesicle membranes, suggesting that its multi-regional localization may serve to integrate cellular signals.

Figure 1.

MoSec13 interacts with MoAtg7 and MoAtg8 to negatively regulate autophagy. (A) Observation of the subcellular localization of MoSec13-GFP during the conidial stage using confocal microscopy (Fv3000, 60×oil). FM 4–64 was utilized to label vesicular membranes; scale bar: 5 μm. ImageJ software was employed for the analysis of colocalization between MoSec13-GFP and FM 4–64 staining, and GraphPad prism software was utilized for generating linear peak plots. (B) Colocalization analysis of MoSec13-GFP and MoLhs1-DsRed during the conidial stage was conducted using confocal microscopy (zeiss LSM780, 63×oil). MoLhs1-DsRed served as a marker for the ER; scale bar: 5 μm. The colocalization analysis of MoSec13-GFP and MoLhs1-DsRed was performed using ImageJ software, and GraphPad prism software was utilized for generating linear peak plots. (C) Colocalization analysis of MoSec13-GFP and MoH₂B-mCherry during the conidial stage was conducted using confocal microscopy (Fv3000, 60×oil). MoH₂B-mCherry served as a marker for the nucleus; scale bar: 5 μm. The colocalization analysis of MoSec13-GFP and MoH₂B-mCherry was performed using ImageJ software, and GraphPad prism software was utilized for generating linear peak plots. (D) Co-IP assays were conducted to identify interactions between MoSec13 and MoAtg7, as well as MoSec13 and MoAtg8. Protein supernatants derived from co-expressing strains, including GFP and Flag-MoAtg7, GFP and Flag-MoAtg8, MoSec13-GFP and Flag-MoAtg7, and MoSec13-GFP and Flag-MoAtg8, were individually incubated with GFP agarose beads for a duration of 4 h. Subsequently, the eluates were subjected to immunoprecipitation using anti-GFP and anti-flag antibodies for detailed analysis. Note that in panel D the paired analyses were run on different gels. Thus, blots 1 and 3 were run on the same gel, whereas blots 2 and 4 were run together on a single different gel; accordingly, the positions of the molecular mass markers differ between gels. (E) Affinity-isolation assays were conducted to identify interactions between MoSec13 and MoAtg7, as well as MoSec13 and MoAtg8. Purified GST, GST-MoAtg7, and GST-MoAtg8 proteins were mixed with purified His-MoSec13 and incubated with GST beads for 4 h. The eluates were then examined through immunoprecipitation using anti-GST and anti-His antibodies for further analysis. As explained for panel D, two different sets of gels were used for the paired analyses; hence, the molecular mass markers, in particular for 27 kDa, are at different positions between the two sets of gels. (F) Fluorescence localization maps of GFP-MoAtg8 were generated for wild-type Guy11 and the ∆Mosec13 mutant hyphae in both liquid CM and MM-N medium. Vacuole staining was achieved using 7-amino-4-chloromethylcoumarin (CMAC); scale bar: 10 μm. (G) The degradation of the fusion protein GFP-MoAtg8 was observed in both wild-type Guy11 and the ∆Mosec13 mutant mycelium under conditions of liquid CM and MM-N medium. Immunoprecipitation analysis was conducted using anti-GFP and anti-ACTB antibodies. (H) The level of macroautophagy was quantified using the formula GFP:(GFP+GFP-MoAtg8). Data from three replicate experiments were analyzed using a t-test analysis, P*** < 0.001, P** < 0.01. (I) The lipidation levels of MoAtg8 were assessed in both wild-type Guy11 and the ∆Mosec13 mutant under conditions of liquid CM and MM-N medium. (J) The lipidation assessment of MoAtg8 was determined by the ratio of MoAtg8–PE:ACTB. Data from three replicate experiments were analyzed using a t-test analysis, P** < 0.01.

Autophagy is a critical cellular process regulated by a network of Atg proteins. Among these, Atg7 functions as an E1-like enzyme, essential for the ubiquitin-like conjugation of Atg8, which facilitates the expansion and maturation of phagophore membranes [4,24]. Understanding the interaction of Atg7 and Atg8 with regulatory proteins is key to unveiling the molecular mechanisms of autophagosome formation, so we conducted co-immunoprecipitation (co-IP) assays to investigate the interaction between MoSec13 and both MoAtg7 and MoAtg8. Our findings revealed that MoSec13 effectively co-precipitated with both Flag-MoAtg7 and Flag-MoAtg8, corroborated by affinity-isolation assays (Figure 1D and E). Interestingly, we observed that nitrogen-starvation conditions (MM-N) considerably diminished the interactions of MoSec13 with MoAtg7 and MoAtg8 (Fig. S2A-D).

To further elucidate the role of MoSec13 in autophagy, we generated a ∆Mosec13 mutant (Fig. S2E-G). The ∆Mosec13 mutant affected the growth and conidiation of the rice blast fungus (Fig. S3), disrupting various stages of its life cycle and resulting in a compromised ability to infect host plants effectively (Fig. S4). The dynamic changes in the degradation of MoAtg8 can serve as a representation of the autophagic degradation process [25]. Consequently, we introduced the fusion protein GFP-MoAtg8 into both wild-type and mutant strains for assessment. Under standard conditions, numerous GFP-MoAtg8 fluorescent dots aggregated around the vacuole (stained with 7-amino-4-chloromethylcoumarin [CMAC]) periphery in the wild-type strain Guy11, signifying the presence of autophagosomes (Figure 1F). In contrast, the ∆Mosec13 mutant displayed a notable absence of these dots, with GFP-MoAtg8 fluorescence having entered the vacuole. When exposed to nitrogen-deficient conditions, both the wild type and the mutant exhibited a similar pattern, wherein GFP-MoAtg8 fluorescence showed a clear overlap with the vacuoles (Figure 1F).

In a western blot analysis, we observed similar results. In this case, we monitored cleavage of the GFP-MoAtg8 protein. The population of GFP-MoAtg8 located on the autophagosome inner membrane is delivered to the vacuole. MoAtg8 is then rapidly degraded, whereas GFP is relatively stable; the appearance of free GFP therefore reflects autophagic flux. Notably, there was a significant presence of free GFP protein under normal conditions (0 h of MM-N) in the ∆Mosec13 mutant. Upon starvation induction, only slight changes were noted in the bands corresponding to the fusion and free proteins (Figure 1G and H). In contrast, the wild-type Guy11 strain was essentially devoid of free GFP prior to autophagy induction resulting from the shift to nitrogen starvation. Additionally, the MoAtg8–PE content in the ∆Mosec13 mutant was significantly elevated compared to the wild type under both conditions (Figure 1I and J). Furthermore, we observed that the expression of 10 autophagy-related target genes was significantly upregulated in the ∆Mosec13 mutant compared to the wild type (Fig. S5A). These results indicate that MoSec13 serves as a vital negative regulator within the autophagy process, influencing the dynamics of autophagy presumably through its interactions with MoAtg7 and MoAtg8.

The repressive activity of MoSec13 on autophagy depends on specific amino acid residues within the WD40 domain

To identify the specific amino acid residues through which MoSec13 binds to MoAtg7 and MoAtg8, we constructed models of the MoSec13-MoAtg7 and MoSec13-MoAtg8 complexes using AlphaFold2 (Figure 2A and B). Predictive modeling suggested that Met1, Ser19, Asp198, and Arg222 of MoSec13 formed critical hydrogen bonds with MoAtg7 (Figure 2C), and Ser111, Gly134, Asn138, and Gln220 established hydrogen bonds with MoAtg8 (Figure 2D). Notably, these sites were situated within a WD40 domain (Figure 2C and D). Therefore, we mutated these residues to alanine and generated two mutants, MoSec13[4A–1] (corresponding to MoSec13^{M1,S19,D198,R222A}) and MoSec13[4A–2] (corresponding to MoSec13^{S111,G134,N138,Q220A}). Affinity-isolation and co-IP assays revealed that MoSec13[4A–1] was unable to interact with MoAtg7 (Figure 2E and F), whereas MoSec13[4A–2] exhibited weak interactions with MoAtg8 (Figure 2G and H). Furthermore, both ∆Mosec13[4A–1]-Flag and ∆Mosec13[4A–2]-Flag strains failed to rescue the growth and pathogenicity defects of the ∆Mosec13 mutant (Figure 3A-D). Thus, these findings indicated that the interactions between MoSec13 and MoAtg7 or MoAtg8 are indeed critical for normal M. oryzae physiology.

Figure 2.

MoSec13 interacts with MoAtg7 and MoAtg8 through specific amino acid residues within the WD40 domain. (A) The AlphaFold2-model-predicted interaction sites between MoSec13 and MoAtg7, where MoSec13 is represented in green, MoAtg7 in blue, and the interaction sites between the two proteins in red. (B) The AlphaFold2-model-predicted interaction sites between MoSec13 and MoAtg8, with MoSec13 in green, MoAtg8 in blue, and the interaction sites between the two proteins in red. (C) The specific interaction sites predicted by AlphaFold2 for MoSec13 binding to MoAtg7 are located at Met1, Ser19, Asp198, and Arg222 of MoSec13. Asp198 and Arg222 are located at a WD structural domain of MoSec13. (D) The specific interaction sites predicted by AlphaFold2 for MoSec13 binding to MoAtg8 are located at Ser111, Gly134, Asn138, and Gln220 of MoSec13; all of these residues are located at a WD40 structural domain of MoSec13. (E) Affinity-isolation assays were conducted to identify interactions between MoSec13[4A–1] and MoAtg7. MoSec13[4A–1] was derived from the mutation of the four sites Met1, Ser19, Asp198, and Arg222 to alanine. (F) Co-IP assays were conducted to identify interactions between MoSec13[4A–1] and MoAtg7. (G) Affinity-isolation assays were conducted to identify interactions between MoSec13[4A–2] and MoAtg8. MoSec13[4A–2] was derived from the mutation of the four sites Ser111, Gly134, Asn138, and Gln220 to alanine. (H) Co-IP assays were conducted to identify interactions between MoSec13[4A–2] and MoAtg8.

Figure 3.

MoSec13 regulates growth, pathogenicity and autophagic activity through specific amino acid residues within the WD40 domain. (A,B) Growth of the wild-type strain Guy11, the ∆Mosec13 mutant, the complemented strain ∆Mosec13-C and (A) ∆Mosec13-C[4A–1] or (B) ∆Mosec13-C[4A–2] on CM medium for 4 days. (C,D) Pathogenicity evaluation of the wild-type strain Guy11, the ∆Mosec13 mutant, the complemented strain ∆Mosec13-C and (C) ∆Mosec13-C[4A–1] or (D) ∆Mosec13-C[4A–2] on barley leaves caused by mycelial plugs. (E,F) The degradation of the fusion protein GFP-MoAtg8 was observed in both wild-type Guy11, the complemented strain ∆Mosec13-C and (E) ∆Mosec13-C[4A–1] or (F) ∆Mosec13-C[4A–2] mycelium under conditions of liquid CM and MM-N medium. Immunoprecipitation analysis was conducted using anti-GFP and anti-ACTB antibodies. The level of macroautophagy was quantified using the formula GFP:(GFP+GFP-MoAtg8). Data from three replicate experiments were analyzed using a t-test analysis, P**** < 0.0001, P*** < 0.001, P** < 0.01, ns p > 0.05. (G,H) The lipidation levels of MoAtg8 were assessed in both wild-type Guy11, the complemented strain ∆Mosec13-C and (G) ∆Mosec13-C[4A–1] or (H) ∆Mosec13-C[4A–2] under conditions of liquid CM and MM-N medium. The lipidation assessment of MoAtg8 was determined by the ratio of MoAtg8–PE:ACTB. Data from three replicate experiments were analyzed using a t-test analysis, P** < 0.01, P* < 0.05, ns p > 0.05.

To further elucidate whether these specific amino acids within the WD40 domain are responsible for the autophagic function of MoSec13, we monitored the degradation of GFP-MoAtg8 and the lipidation of MoAtg8. We found that compared to Guy11 and the complemented strain ∆Mosec13-C, the amount of the full-length GFP-MoAtg8 chimeric proteins was significantly reduced, accompanied by an increase in free GFP protein in ∆Mosec13-C[4A–1] and ∆Mosec13-C[4A–2] under both normal and nitrogen-deficient conditions (Figure 3E and F). Additionally, ∆Mosec13-C[4A–1] and ∆Mosec13-C[4A–2] exhibited significantly higher levels of MoAtg8–PE than seen in the Guy11 and ∆Mosec13-C strains under both conditions (Figure 3G and H). These results suggest that MoSec13 negatively regulates the autophagy process by interacting with the autophagy core proteins MoAtg7 and MoAtg8 through four different amino acid residues within the WD40 domains.

MoGcn5b-mediated acetylation of MoAtg8 regulates autophagy

Gcn5, a histone acetyltransferase, is instrumental in modulating intracellular acetylation modifications, thereby influencing the dynamics of autophagy [9,26]. In Fusarium graminearum, FgGcn5 negatively regulates autophagy by acetylating Atg8, whereas in M. oryzae, MoGcn5 acetylates Atg7, inhibiting starvation- and light-induced autophagy [9,26]. We observed that MoGcn5b exhibited nuclear localization when co-expressed with H₂B-mCherry, which further suggests that MoGcn5b exerts functions in the nucleus (Figure 4A).

Figure 4.

MoGcn5b-mediated acetylation of MoAtg8 regulates autophagy. (A) Colocalization analysis of GFP-MoGcn5b and MoH₂B-mCherry during the conidial stage was conducted using confocal microscopy (Fv3000, 60×oil). MoH₂B-mCherry served as a marker for the nucleus; scale bar: 5 μm. The colocalization analysis of GFP-MoGcn5b and MoH₂B-mCherry was performed using ImageJ software, and GraphPad prism software was utilized for generating linear peak plots. (B) MoAtg8 was acetylated by MoGcn5b in vitro. The GFP-MoGcn5b protein, purified from M. oryzae, was separately incubated with GST-MoAtg7 and GST-MoAtg8 proteins, both purified from E. coli. Following the incubations, the reaction solutions were subjected to analysis through immunoprecipitation using anti-AcK, anti-GFP, and anti-GST antibodies. (C) A yeast two-hybrid assay was conducted to identify the interaction between MoGcn5b and MoAtg8. Plasmids MoGcn5b-AD and MoAtg8-BD, MoGcn5b-AD and pGBKT7, and MoAtg8-BD and pGADT7 were assessed for growth in SD-Leu-Trp and SD-Leu-Trp-Ade-His media. Plasmids pGBKT7–53 and pGADT7-T were used as positive controls. (D) A co-IP assay was conducted to examine the interaction between MoGcn5b and MoAtg8. Protein supernatants derived from co-expressed strains, including GFP and mCherry-MoAtg8, and GFP-MoGcn5b and mCherry-MoAtg8, were individually incubated with GFP agarose beads for a duration of 4 h. Subsequently, the eluates were subjected to immunoprecipitation using anti-GFP and anti-mCherry antibodies for detailed analysis. (E) The acetylation level of MoAtg8 in wild-type Guy11 and ∆Mogcn5b mutant was examined in CM liquid medium and MM-N liquid medium. The acetylation level was assessed by the formula AcK:GFP-MoAtg8. (F) Data from three replicate experiments were analyzed using a t-test analysis, P*<0.05. (G) Colocalization of GFP-MoAtg8 and MoH₂B-mCherry under nutrient-rich and nitrogen-deficient conditions using confocal microscopy (Fv3000, 60×oil). Scale bar: 5 μm. (H) The degradation of the fusion protein GFP-MoAtg8 was observed in both wild-type Guy11 and the ∆Mogcn5b mutant mycelium under conditions of liquid CM and MM-N medium. Immunoprecipitation analysis was conducted using anti-GFP and anti-ACTB antibodies. (I) The level of autophagy was quantified using the formula GFP:(GFP+GFP-MoAtg8). Data from three replicate experiments were analyzed using a t-test analysis, P***<0.001, P*<0.05, ns p > 0.05. (J) The lipidation levels of MoAtg8 were assessed in both wild-type Guy11 and the ∆Mogcn5b mutant under conditions of liquid CM and MM-N medium. (K) The lipidation assessment of MoAtg8 was determined by the ratio of MoAtg8–PE:ACTB. Data from three replicate experiments were analyzed using a t-test analysis, P**<0.01, P*<0.05.

To investigate whether MoGcn5b is responsible for the acetylation of MoAtg7 and MoAtg8, we exogenously added GFP-MoGcn5b purified from M. oryzae and found that it specifically acetylated GST-MoAtg8 but not GST-MoAtg7, both purified from Escherichia coli (Figure 4B). The interaction of MoGcn5b and MoAtg8 was further confirmed in vivo and in vitro (Figure 4C and D). We performed immunoprecipitation of GFP-MoAtg8 and utilized an anti-acetyl-lysine antibody to assess the acetylation status via western blotting. Our results showed that the acetylation of GFP-MoAtg8 was significantly reduced in the ∆Mogcn5b mutant (Figure 4E and F), suggesting that MoGcn5b plays a critical role in this modification. During starvation conditions, where we observed a marked decrease in the acetylation levels of MoAtg8 (Figure 4E and F), we found a reduction in nuclear-localized MoAtg8 where it predominantly redistributed to the cytoplasm (Figure 4G). Furthermore, we noted that the turnover of GFP-MoAtg8 (Figure 4H and I) and lipidation capacity of MoAtg8 (Figure 4J and K) were diminished in the ∆Mogcn5b strain. This observation suggests that MoAtg8 is an important acetylation target for MoGcn5b during autophagy.

MoSec13 coordinates the acetylation of MoAtg8 by modulating the interaction between MoAtg8 and MoGcn5b

The interaction between MoSec13 and MoAtg8, along with the established association between MoAtg8 and MoGcn5b, suggests an unknown regulatory network between these proteins. Because MoSec13 is a component of the NPC complex and is implicated in the autophagic process, particularly in cargo recognition [27], it raises the question of whether MoSec13 also interacts with MoGcn5b. Our mass spectrometry analysis identified MoGcn5b in the enrichment of MoSec13-GFP using GFP agarose gel beads (Table S1). Further investigations confirmed direct interactions between MoSec13 and MoGcn5b through yeast two-hybrid, affinity-isolation, and co-IP assays (Figure 5A-C). Additionally, bimolecular fluorescence complementation (BiFC) and H₂B-mCherry colocalization studies revealed that MoSec13 and MoGcn5b interacted within the nucleus (Figure 5D). To investigate the distribution of the interaction between MoSec13 and MoAtg8, we performed an additional BiFC assay. Colocalization analysis showed that the fluorescence of MoSec13-MoAtg8 overlapped with MoH₂B-mCherry (Figure 5E), while the small, dotted fluorescence of MoSec13-MoAtg8 colocalized exclusively with mCherry-MoApe1 (an autophagic cargo protein and a marker for the phagophore assembly site [PAS]) and was dispersed within the red fluorescence of mCherry-MoAtg17 (a component of the MoAtg1 kinase complex and another marker for the PAS; Figure 5F). These results indicated that MoSec13 interacted with MoAtg8 not only at the PAS but is also detectable within the nucleus. This interaction was also observed in the isolated cytoplasmic and nuclear fractions via western blot analysis, and weakened with starvation induction (Fig. S5B and C).

Figure 5.

MoSec13 coordinates the acetylation of MoAtg8 by modulating the interaction between MoAtg8 and MoGcn5b. (A) A yeast two-hybrid assay was conducted to identify the interaction between MoSec13 and MoGcn5b. Plasmids MoGcn5b-AD and MoSec13-BD, MoGcn5b-AD and pGBKT7, and MoSec13-BD and pGADT7 were assessed for growth in SD-Leu-Trp and SD-Leu-Trp-Ade-His media. Plasmids pGBKT7–53 and pGADT7-T were used as positive controls. (B) An affinity-isolation assay was conducted to identify the interaction between MoSec13 and MoGcn5b. Purified proteins GST and GST-MoGcn5b were mixed with purified protein His-MoSec13 and incubated with GST beads for 4 h. The eluates were then examined through immunoprecipitation using anti-GST and anti-His antibodies for further analysis. (C) A co-IP assay was conducted to identify the interaction between MoSec13 and MoGcn5b. Protein supernatants derived from co-expressed strains, including GFP and Flag-MoGcn5b, and MoSec13-GFP and Flag-MoGcn5b, were individually incubated with GFP agarose beads for a duration of 4 h. Subsequently, the eluates were subjected to immunoprecipitation using anti-GFP and anti-flag antibodies for detailed analysis. (D) Fluorescence microscopy was performed to observe YFP fluorescence of MoGcn5b-YFPC and YFPN-MoSec13, MoGcn5b-YFPC and YFPN, YFPC and YFPN-MoSec13, and colocalization of MoGcn5b-MoSec13 with H₂B-mCherry during the mycelial period. Colocalization analysis was performed using ImageJ software, and GraphPad prism software was utilized for generating linear peak plots for panels D-F. Scale bar: 10 μm. (E) Confocal microscopy (Fv3000) was performed to observe YFP fluorescence of YFPC-MoAtg8 and MoSec13-YFPN, YFPC-MoAtg8 and YFPN, and YFPC and MoSec13-YFPN, and colocalization of MoAtg8-MoSec13 with H₂B-mCherry during the mycelial period. Scale bar: 5 μm. (F) Confocal microscopy (Fv3000) was performed to observe the colocalization of MoAtg8-MoSec13 with mCherry-MoApe1 and mCherry-MoAtg17 during the mycelial period. Scale bar: 5 μm. (G) The acetylation level of MoAtg8 in wild-type Guy11 and ∆Mosec13 mutant strains was examined in CM liquid medium and MM-N liquid medium. (H) The acetylation level was assessed by the formula AcK:GFP-MoAtg8. Data from three replicate experiments were analyzed using a t-test analysis, P** < 0.01, P* < 0.05. (I) Detection of the strength of MoGcn5b-MoAtg8 interactions in wild-type Guy11 and ∆Mosec13 mutant cells. Wild-type and ∆Mosec13 mutant strains with GFP-MoGcn5b and mCherry-MoAtg8 double tags were incubated in CM liquid medium for 36–48 h. The extracted total protein supernatant was incubated with GFP agarose gel beads for 4 h and the eluate obtained was analyzed by immunoprecipitation with anti-GFP antibody and anti-mCherry antibody. (J) Interaction strength was assessed with the formula mCherry-MoAtg8:GFP-MoGcn5b. Data from three replicate experiments were analyzed using a t-test analysis, P** < 0.01.

To further investigate whether MoSec13 is involved in the regulation of MoAtg8 acetylation, we assessed the acetylation levels of MoAtg8 in the ∆Mosec13 strain. In the ∆Mosec13 mutant, MoAtg8 acetylation was notably higher than in the wild type, and this elevated level persisted even during starvation conditions, opposite to the trend for the Guy11 strain (Figure 5G and H). Consistent with this, we found that the interaction between MoGcn5b and MoAtg8 was enhanced in the ∆Mosec13 mutant (Figure 5I and J). The absence of MoSEC13 facilitates the interaction between MoAtg8 and MoGcn5b, promoting the acetylation of MoAtg8. This result suggests a role for MoSec13 in modulating the acetylation status of MoAtg8 by influencing the interaction between MoAtg8 and MoGcn5b.

MoSec13 regulates reticulophagy and mitophagy

Sec13 is a versatile protein that plays crucial roles across various cellular compartments [28,29]. Protein transport between the ER and the Golgi relies on COPII vesicles [30,31], comprising two main components: the inner shell component, Sec23-Sec24, and the outer shell component, Sec13-Sec31 [30,32]. Both affinity-isolation and co-IP assays confirmed the interaction between MoSec13 and MoSec31 in M. oryzae (Figure 6A and B). The unfolded protein response (UPR) is activated upon ER stress [33]. Consequently, we investigated the splicing status of the transcription factor MoHac1 in both wild-type and mutant strains to assess UPR activation. Under normal conditions, MoHac1 exists in an unspliced state (uMoHAC1). Upon activation of the UPR due to the accumulation of misfolded proteins, the spliced state of MoHac1 (sMoHAC1) is augmented [33]. The results indicated a significant enhancement in the level of sMoHAC1, along with a notable upregulation in the expression of UPR-associated target genes in the ∆Mosec13 mutant compared to the wild type (Figure 6C and D). To further corroborate that the disruption of the COPII-mediated protein transport pathway activates the UPR pathway, we conducted experiments utilizing brefeldin A (BFA), an agent known to inhibit the export of proteins from the ER, thus inducing an ER stress response. Our findings revealed a significant upregulation of UPR-associated target genes at the transcriptional level in the wild-type strain after the application of BFA, mirroring the observations made in the mutant strain (Figure 6E). The aforementioned findings imply that the deletion of MoSEC13 leads to impaired functionality of the COPII complex, triggering the activation of the UPR.

Figure 6.

MoSec13 regulates reticulophagy and mitophagy. (A) Affinity-isolation assay to detect MoSec13 and MoSec31 protein interaction. Purified GST and GST-MoSec31 proteins were mixed with purified His-MoSec13 and incubated with GST beads for 4 h. The eluates were then examined through immunoprecipitation using anti-GST and anti-His antibodies for further analysis. (B) A co-IP assay was used to detect MoSec13 and MoSec31 protein interaction. Protein supernatants derived from co-expressed strains, including GFP and MoSec31-flag, and MoSec13-GFP and MoSec31-flag, were individually incubated with GFP agarose beads for a duration of 4 h. Subsequently, the eluates were subjected to immunoprecipitation using anti-GFP and anti-flag antibodies for a detailed analysis. (C) PCR was employed to identify the splicing state of the UPR transcription factor MoHac1 in both wild-type Guy11 and the ∆Mosec13 mutant strains. ACTB was utilized as an internal reference gene. The unspliced state of MoHac1 is denoted as uMoHAC1, while the spliced state is represented as sMoHAC1. (D) The expression levels of UPR pathway-related target genes in the ∆Mosec13 mutant were assessed using qRT-PCR. Data analysis was conducted using a t-test, P**** < 0.0001, P*** < 0.001, P** < 0.01. (E) The expression levels of target genes associated with the UPR pathway were examined via qRT-PCR in the wild-type strain upon the addition of BFA. Subsequent data analysis was conducted using the t-test, P**** < 0.0001, P*** < 0.001. (F) The degradation of the fusion protein GFP-MoSec62 was assessed in both wild-type Guy11 and the ∆Mosec13 mutant strains under conditions of liquid CM and added DTT. (G) Reticulophagy levels were determined using the formula GFP:(GFP+GFP-MoSec62). Data from three replicate experiments were analyzed using a t-test analysis, P*** < 0.001, P** < 0.01. (H) The degradation of the Porin protein was examined in both wild-type Guy11 and the ∆Mosec13 mutant strains under conditions of liquid CM and MM-N medium. (I) Mitophagy levels were quantified by determining the ratio of Porin to the internal reference ACTB. Data from three replicate experiments were analyzed using a t-test analysis, P** < 0.01, P* < 0.05.

When the UPR pathway is unable to alleviate ER disruption, the autophagy pathway is triggered [34,35]. We noted alterations in the degradation of the ER protein MoSec62, observing a notable acceleration in the degradation of GFP-MoSec62 even under nutrient-rich conditions in the ∆Mosec13 mutant, as compared to the wild type (Figure 6F and G). This observation indicated an elevated level of selective autophagic degradation of the ER (reticulophagy) following the deletion of MoSEC13. Given the potential for interaction between the ER and mitochondria through mitochondria-ER contact points, we decided to explore whether the heightened reticulophagy activity in the absence of MoSEC13 affects selective mitochondrial turnover via autophagy (mitophagy). We consequently investigated the degradation of the mitochondrial marker protein Porin. Our findings revealed that in the wild type, Porin continued to be degraded progressively until it eventually disappeared with prolonged starvation induction (Figure 6H and I). In contrast, the ∆Mosec13 mutant exhibited sustained high protein levels, with discernible bands persisting even after 24 h of induced starvation. These results demonstrate that MoSec13 can also regulate selective autophagy pathways.

MoSec13 is involved in lipid homeostasis regulation

Lysophosphatidylinositol/lysoPI facilitates COPII complex budding by reducing the bending stiffness of the membrane and augmenting the binding of Sar1 and Sec13-Sec31 [18]. To ascertain whether MoSec13 in M. oryzae was linked to lipid metabolic processes, we conducted a quantitative lipidome analysis. A large number (656) of lipids were identified in the experimental samples (Table S2), of which 418 lipids and 34 lipid classes differed between the wild-type Guy11 strain and the ∆Mosec13 mutant (Fig. S6 and Fig. S7A). Quantitative analysis of lipid classes revealed that phosphatidylcholine/PC, phosphatidylethanolamine/PE, phosphatidylinositol/PI, phosphatidylserine/PS, lysophosphatidylcholine/LPC, lysophosphatidylethanolamine/LPE, lysophosphatidylinositol/LPI, and lysophosphatidylserine/LPS were drastically reduced in ∆Mosec13 mutant compared to the wild type (Fig. S7B and C). An annotated map of the Kyoto Encyclopedia of Genes and Genomes/KEGG metabolic pathways showed that MoSec13 participated in regulating sphingolipid metabolism, glycerolipid metabolism, and autophagy (Fig. S7D). Upon further analysis, significant upregulation of lipid expression was observed in sphingolipid metabolism, while there was a notable downregulation of lipid expression in the glycerolipid metabolism and autophagy pathways (Fig. S7E).

It is worth noting that we observed a clear rise in sphingolipid content in the ∆Mosec13 mutant compared to the wild-type strain Guy11 (Fig. S8A). To validate this, we conducted tests using myriocin, an agent that inhibits sphingolipid synthesis, and found that the ∆Mosec13 mutant exhibited resistance to myriocin compared to both the wild-type strain Guy11 and the complemented strain ∆Mosec13-C (Fig. S8B and C). While sphingolipids and sterols typically participate together to maintain cytoplasmic membrane homeostasis, the question arises as to whether the elevated sphingolipid levels in the mutant have any impact on sterols. To investigate this, we inoculated the strains with the sterol synthesis inhibitor amphotericin B, as well as sterol-targeting azoles tebuconazole and terbinafine. The results revealed a significant inhibition of growth in the ∆Mosec13 mutant compared to the wild-type strain Guy11 and complemented strain ∆Mosec13-C (Fig. S8D and E). In summary, the deletion of MoSEC13 leads to abnormal changes in lipid content, which affect lipid homeostasis.

TORC2 activity is compromised in the ∆Mosec13 strain

In yeast, the TORC2 complex regulates sphingolipid biosynthesis; the phosphorylation level of its downstream kinase Ypk1, is commonly measured to indicate TORC2 activity [36]. With the confirmation that MoYpk1 undergoes phosphorylation by TORC2 at the S619 site in M. oryzae [37,38], we examined the phosphorylation levels of MoYpk1 in both wild-type and mutant strains. Under standard culture conditions, the deletion of MoSEC13 led to a notable reduction in the phosphorylation level of MoYpk1 (Figure 7A and B). Following a 3h treatment with myriocin, an elevation in the phosphorylation level of MoYpk1 was observed in the wild-type strain Guy11. However, the phosphorylation level of MoYpk1 remained unaltered and consistently low in the ∆Mosec13 mutant (Figure 7A and B). This observation suggests that MoSec13 may serve as an upstream regulator of TORC2.

Figure 7.

TORC2 activity is compromised in the ∆Mosec13 strain. (A) The phosphorylation level of MoYpk1 was assessed in the wild-type strain Guy11 and the ∆Mosec13 mutant under both CM and myriocin-treatment conditions. (B) Data from three replicate experiments were analyzed using a t-test analysis, P*** < 0.001, P* < 0.05. (C) The colony morphology of the wild-type strain Guy11, the ∆Mosec13 mutant, and the complemented strain ∆Mosec13-C after 8 days of cultivation in CM medium supplemented with 0.5 M NaCl, 0.5 M KCl, and 1 M sorbitol. (D) Growth diameter analysis of the wild-type strain Guy11, the ∆Mosec13 mutant, and the complemented strain ∆Mosec13-C in medium supplemented with NaCl, KCl, and sorbitol. The data underwent a t-test analysis using GraphPad Prism software, P**** < 0.0001. (E-H) The phosphorylation levels of (E) Hog1 (detected with anti-MAPK/p38) and (G) MoYpk1 were assessed in the wild-type strain Guy11 and the ∆Mosec13 mutant under both CM and 0.5 M NaCl-treatment conditions. Data from three replicate experiments were analyzed using a t-test analysis, P**** < 0.0001, P*** < 0.001, P** < 0.01, P* < 0.05, ns p > 0.05.

Yeast cells exhibit the capacity to coordinate and integrate both the Hog1 and TORC2-Ypk1 pathways in response to external hyperosmotic stress [39,40]. We first treated all strains with NaCl, KCl and sorbitol and found that the growth of the ∆Mosec13 mutant was significantly inhibited in all cases (Figure 7C and D). We further assessed the changes in the phosphorylation level of Hog1 under NaCl treatment and found that the phosphorylation levels of Hog1 in the ∆Mosec13 mutant exceeded those of the wild type when induced by hyperosmotic stress (Figure 7E and F). Simultaneously, we investigated the alterations in the phosphorylation level of MoYpk1 and observed that the phosphorylation levels of MoYpk1 in the ∆Mosec13 mutant consistently lagged behind that of the wild type (Figure 7G and H). These findings indicate that MoSec13 may regulate the osmo-sensor(s) upstream of TORC2.

Discussion

The spatial distribution of Sec13 as a component of the COPII and NPC complexes in the ER and nucleus has been recognized for a long time, but its significance in autophagy remains underappreciated. In this study, we identified the unexpected role of MoSec13 in the autophagy mechanism in M. oryzae (Figure 8). Moreover, deletion of MoSEC13 resulted in decreased TORC2 activity and abnormal lipid content.

Figure 8.

A model of MoSec13 regulation in M. oryzae. MoSec13, as a component shared by COPII and NPC complexes, not only negatively regulates autophagy by modulating the interactions of specific amino acid residues in its own WD40 structural domain with MoAtg7 and MoAtg8, but also regulates MoAtg8 acetylation status in the nucleus by binding to MoGcn5b. Moreover, MoSec13 can control TORC2 activity to regulate lipid homeostasis.

COPII and NPC complexes dominate secretory transport and nucleocytoplasmic exchange, and are central nodes for intracellular substance transport [41,42]. Both of them play critical roles in basic cellular functions, stress response and diseases. Autophagy is usually initiated in nutrient-repressed environments and aids cell survival by degrading protein aggregates or damaged organelles [6,43]. In the plant pathogenic fungi, autophagy regulators involved in pathogenicity have been intensively investigated, such as Atg proteins, the ESCRT complex, deacetylated protein MoSnt2, and sterol-binding proteins MoVast1 and MoVast2 [37,38,44–46]. Although COPII and NPC complex-related components have been found to be involved in the autophagy pathway in recent years, the role of Sec13, a shared subunit of these two complexes, in autophagy has rarely been reported [12,47,48]. In this study, we report that MoSec13 participates in the autophagy regulatory mechanism of pathogenicity in M. oryzae. MoSec13 is involved in the growth and pathogenicity of M. oryzae by interacting with the autophagy core proteins MoAtg7 and MoAtg8 through specific amino acid residues that negatively regulate the autophagic process (Figures 2 and 3). Moreover, we found that MoSec13 functions in the nucleus (Figures 1C and 5). Specifically, MoSec13 affects the acetylation status of MoAtg8 by regulating the interaction between the histone acetyltransferase MoGcn5b and the autophagy-related protein MoAtg8 (Figure 5). However, given that MoSec13 is associated with both the COPII and NPC complexes, and its important role in autophagy and pathogenicity of M. oryzae, it remains to be elucidated whether the function of MoSec13 in the autophagy pathway and its role in the nucleus is dependent on these two complexes. The need to clarify this specific regulatory mechanism may require further characterization of other key components of COPII and the NPC.

The acetylation of nuclear LC3/Atg8 plays a pivotal role in determining the cellular localization of this protein, which is crucial for its function in autophagy [8,49]. Atg8/LC3 massively migrates from the nucleus to the cytoplasm under starvation conditions to initiate autophagy [8]. The nuclear-cytoplasmic translocation of MoAtg8 indicates the potential involvement of specific molecular factors in this transport process. The intracellular dynamics and targeting of Sec13 between the nucleus and cytoplasm may coordinate and regulate functions across these compartments [29], suggesting its potential role as a chaperone protein. Our study indicated that starvation induced a redistribution of MoAtg8 into the cytoplasm (Figure 4G), which is accompanied by a decreased interaction between MoSec13 and MoAtg8 (Fig. S2C and D). This reduction occurred simultaneously in both the cytoplasm and the nucleus (Fig. S5B and C), and indicated that MoSec13 might play a potential role in facilitating the translocation of MoAtg8 from the former to the latter. However, it appeared that MoSec13 did not mediate the starvation-induced relocalization of MoAtg8 within the cytoplasm (Figure 4, Fig. S2, and Fig. S5), which may instead be facilitated by other chaperone proteins, such as a fungal analog of TP53INP2/DOR [8]. Considering the significance of Atg8, the redundancy of this process ensures the proper execution of autophagy. However, how MoSec13 acts as a chaperone protein in mediating MoAtg8 function and whether this process involves reversible transport between the cytoplasm and the nucleus need to be elucidated in the future.

The TOR signaling pathway is known in part for its conserved role in controlling cell growth and autophagy through the distinct TORC1 and TORC2 complexes [50–55]. While the role of TORC1 in autophagy regulation is well-established, the functions of TORC2 remain less understood [56]. Recently, discoveries in the rice blast fungus point to two autophagy regulators, MoVast1 and MoVast2, that can regulate TORC2 activity. MoVast2 combined with MoVast1 coregulates lipid homeostasis by modulating TORC1 and TORC2 activity [37,38]. Our data reveal that MoSEC13 disruption attenuates TORC2 activity and triggers sphingolipid intermediates hyperaccumulation (Figure 7 and Fig. S8A). Intriguingly, the ΔMosec13 mutant exhibited resistance to the sphingolipid biosynthesis inhibitor myriocin (Fig. S8B and C), implying compensatory sphingolipid production via TORC2-independent pathways. This adaptation may involve transcriptional activation of sphingolipid synthases or rerouting of precursor metabolites. Critically, sterol-targeting antifungals severely compromised ∆Mosec13 viability (Fig. S8D and E), suggesting that sterol biosynthesis may compensate for sphingolipid overload to preserve membrane integrity. That is, the cell adapts to the absence of TORC2 activity by dynamically reprogramming the lipid synthesis pathway, but this compensation is highly dependent on the synergistic support of sterols. Once both are simultaneously restricted, the system collapses. Therefore, we propose that MoSec13 may act as a membrane tension sensor responsible for the coordination of sphingolipid-sterol homeostasis. While TORC2 inhibition is a downstream consequence of MoSEC13 deletion, its direct mechanistic connection to autophagy regulation remains unresolved. It is necessary to focus on the downstream substrate of TORC2, MoYpk1, for characterization, and because it is difficult to knock out MoYPK1 in M. oryzae, in situ mutagenesis could be introduced in the future to find and validate the key functional sites.

Collectively, we demonstrated that a multiple-localized MoSec13 acted as a negative regulator of autophagy by interacting with MoAtg7 and MoAtg8 through specific amino acid residues via its WD40 domain. We also showed that the histone acetyltransferase MoGcn5b interacted with MoAtg8 and acted as a regulator of MoAtg8 acetylation. Our results showed that MoSec13 regulated the acetylation status of MoAtg8 by controlling the interaction between MoGcn5b and MoAtg8. Furthermore, MoSec13 modulates TORC2 activity to regulate lipid homeostasis. Such layered information integration identifies MoSec13 as an integral part of the autophagic regulatory framework.

Materials and methods

Fungal strains, culture conditions, and phenotypic analysis

In this study, Guy11 served as the wild-type strain of M. oryzae. All strains were inoculated on complete medium (CM;10 g D-glucose,2 g peptone 140, 1 g casamino acid, 1 g yeast extract, 6 g NaNO₃, 1.52 g KH₂PO₄, 0.52 g KCl, 0.52 g MgSO₄·7 H₂O, 1 ml vitamin solution, 1 ml trace elements and 15 g agar in 1 L of distilled water, pH 6.5) [57] and incubated at 25°C under a light cycle of 16 h light and 8 h dark for a duration of 8 days [58]. Questions about the details of the methods used for conidia germination, appressorium formation, and the measurement of pathogenicity can be referred to in the article by Qian et al. [22].

Generation of MoSEC13 gene deletion and complementary system

MoSEC13 gene knockout was primarily achieved by the high-throughput knockout system proposed by Lu et al. [59]. The procedure involved the following steps: The total genome of M. oryzae was utilized as a template to amplify a 1500-bp fragment upstream of MoSEC13 (UF) and a 1500-bp fragment downstream of MoSEC13 (DF). The plasmid pCB1003 vector [60] was employed as a template to amplify the hygromycin phosphotransferase (HPH) gene fragment. These three fragments (UF, DF, and HPH) were then assembled into the linearized pKO3A vector [59] using ligase. The constructed plasmid was transformed into the wild-type strain Guy11 through Agrobacterium tumefaciens-mediated transformation/ATMT, and the obtained transformants were selected on plates containing both hygromycin (Sangon Biotech, A600230–0001) and 5-fluoro-2’-deoxyuridine (Sigma-Aldrich, F0503) antibiotics. For the MoSEC13 gene complementary system, a fragment of the MoSEC13 gene lacking the TAA stop codon was amplified using the wild-type strain Guy11 DNA as a template. This fragment was then integrated into the linearized vector pKD5 [61] using ligase. The complementation vector was introduced into the ΔMosec13 mutant strain using the ATMT method described above. Transformants were selected on plates containing sulfonylurea antibiotic (Yeasen, 41015ES80).

Fluorescence position observation

To determine the fluorescent localization of MoSec13, we obtained strains with a single-fluorescent label of MoSec13-GFP, a double-fluorescent label of MoSec13-GFP and MoLhs1-DsRed, and a double-fluorescent label of MoSec13-GFP and MoH₂B-mCherry. The red and green fluorescence localization of the strains at the conidial stage was observed by microscopy. Conidia of the strain with the MoSec13-GFP single-fluorescent label were stained with FM 4–64 (MCE, HY-103466) for 10 min. The excitation and emission wavelengths are the following: GFP, 488 nm and 510 nm; FM 4–64, 558 nm and 634 nm; DsRed, 557 nm and 579 nm; mCherry, 561 nm and 610 nm.

Quantitative reverse transcription-polymerase chain reaction (qRT-PCR) analysis

RNA was extracted from the mycelium of the wild-type strain Guy11 and the ΔMosec13 mutant by the Trizol method (TaKaRa, 9109). Subsequently, cDNA was reverse transcribed from the extracted RNA using the PrimeScript RT reagent kit (TaKaRa, RR047B). Primers for detecting gene expression were designed through the Integrated DNA Technologies website (Table S3). TUBULIN-, ACTIN and 40S were selected as the internal reference genes, and the transcript levels of the target genes were detected using TB Green Premix Ex Taq (TaKaRa, RR420Q). For details on data processing, please refer to the article by Qu et al. [62].

Growth assays with stress factors

To examine the impact of distinct stressors, we inoculated the wild-type strain Guy11, the ΔMosec13 mutant, and the complemented strain ∆Mosec13-C on CM solid medium supplemented with specific agents: 1 μM myriocin (APExBIO, B6064), 0.5 M NaCl, 0.5 M KCl, 1 M sorbitol (Sangon Biotech, A100691–0500), 800 μg/ml Congo Red (Sangon Biotech, A600324–0050), 0.0075% SDS, 100 μg/ml CFW (Yuanye Bio-Technology Co., Ltd, S26637), 8 mm H₂O₂, 4 μM amphotericin B (APExBIO, B1885), 0.25 μg/ml tebuconazole (MCE, HY-B0852), or 0.25 μg/ml terbinafine (MCE, HY-17395A). These strains were then placed in an incubator at 25°C for 8 days for subsequent data collection and photographic documentation.

Macroautophagy and selective autophagy assays

For macroautophagy assays, the mycelium of the wild-type strain Guy11 and the ΔMosec13 mutant with a GFP-MoAtg8 fluorescent label was cultured in liquid CM medium for 36–48 h [63]. A suitable amount of the mycelium was then transferred to MM-N medium (10 g D-glucose, 1.52 g KH₂PO₄, 0.52 g KCl, 0.52 g MgSO₄·7 H₂O, 1 ml vitamin solution, and 1 ml trace elements in 1 L of distilled water, pH 6.5) for 3 h to induce autophagy, after which it was collected. Changes in the fluorescence localization of MoAtg8 under nutrient-sufficient and nutrient-deprived conditions were observed for Guy11 and ΔMosec13. The vacuole was labeled with the specific molecular probe CMAC (Thermo Fisher Scientific, C2110). To further examine the degradation of GFP-MoAtg8 in the wild-type strain Guy11 and the ΔMosec13 mutant, we used the anti-GFP antibody (Abcam, ab32146) to detect the free GFP band and the GFP-MoAtg8 fusion band. The final results were presented in terms of the ratio of the free GFP band to the combined intensity of the free GFP and full-length GFP-MoAtg8 chimera bands. To detect the lipidated form of MoAtg8, we used the anti-Atg8 antibody (MBL, PM090) to examine protein extracts from cells in nutrient-rich and nutrient-deprived states. All protein samples were analyzed with anti-ACTB/actin (ABclonal Technology, AC004) as a control.

For reticulophagy assays, mycelium from the wild-type strain Guy11 and the ΔMosec13 mutant with a GFP-MoSec62 fluorescent label was cultured in liquid CM medium for 36–48 h [64], and then maintained in CM medium containing DTT for 6 h for induction and collected. The final results were presented as the ratio of the free GFP band to the combined intensity of the free GFP and full-length GFP-MoSec62 chimera bands. For mitophagy assays, the mycelium of the wild-type strain Guy11 and the ΔMosec13 mutant was cultured in liquid CM medium for 36–48 h, then maintained in BM-G medium for 30 h to enrich mitochondria, and finally induced for mitophagy in MM-N medium for 12 and 24 h. The anti-Porin antibody (produced by ABclonal Technology) was used to monitor the level of mitophagy and ACTB was used as a control.

Design of MoSec13[4A–1] and MoSec13[4A–2]

The amino acid sequences of the protein complexes (MoSec13-MoAtg7 and MoSec13-MoAtg8) to be predicted were obtained. Protein complexes were modeled using the ColabFold platform of AlphaFold2 [65]. The PyMOL script “InterfaceResides” was used to screen the amino acid residues bound between the two proteins of the MoSec13-MoAtg7 complex and the MoSec13-MoAtg8 complex. The binding amino acid residues predicted for the MoSec13-MoAtg7 protein complex were located at Met1, Ser19, Asp198, and Arg222 of MoSec13.The binding amino acid residues predicted for the MoSec13-MoAtg8 protein complex were predicted to be located at Ser111, Gly134, Asn138, and Gln220 of MoSec13. We mutated these two sets of amino acids to alanine resulting in MoSec13[4A–1] and MoSec13[4A–2], respectively.

In vitro and in vivo assays to detect protein interactions

To ascertain protein interactions in vitro, we employed both the yeast two-hybrid system and an affinity-isolation system. In the yeast two-hybrid system, positive control plasmids pGBKT7–53 and pGADT7-T, reciprocal protein plasmids MoAtg8-BD and MoGcn5b-AD, and MoGcn5b-AD and MoSec13-BD, and self-activating control plasmids MoAtg8-BD and pGADT7, MoGcn5b-AD and pGBKT7, and MoSec13-BD and pGADT7, were cotransformed into the Y₂HGold strain (Clontech 630,489). The transformed strains were then plated on SD-Leu-Trp and SD-Leu-Trp-Ade-His media and incubated for 4 days to observe the results. In the affinity-isolation system, GST glutathione agarose beads (Sangon Biotech, C600031–0005) were employed. GST-tagged recombinant proteins purified from E. coli (including GST-MoAtg7, GST-MoAtg8, GST-MoGcn5b, and GST-MoSec31) and His-tagged recombinant proteins (including His-MoSec13, His-MoSec13[4A–1], or His-MoSec13[4A–2]) were mixed with the GST beads and incubated for 4 h at 4°C. The supernatant obtained was analyzed by detection with anti-GST antibody (HuaBio, EM80701) and anti-His antibody (HuaBio, R1207–2).

To ascertain protein interactions in vivo, we employed co-IP and the BiFC method. For co-IP, we first obtained strains with double tags of GFP and Flag-MoAtg7, GFP and Flag-MoAtg8, GFP and mCherry-MoAtg8, GFP and Flag-MoGcn5b, GFP and MoSec31-Flag, MoSec13-GFP and Flag-MoAtg7, MoSec13[4A–1]-GFP and Flag-MoAtg7, MoSec13-GFP and Flag-MoAtg8, MoSec13[4A–2]-GFP and Flag-MoAtg8, GFP-MoGcn5b and mCherry-MoAtg8, MoSec13-GFP and Flag-MoGcn5b, and MoSec13-GFP and MoSec31-Flag. Total protein was then extracted from these strains, and the protein supernatant obtained was co-incubated with GFP beads (Smart-Lifesciences, SA070005) for 4 h before the final immunoblotting assay. The supernatant obtained was analyzed by detection with anti-GFP antibody (Abcam, ab32146), anti-FLAG antibody (HuaBio, M1403–2), and anti-mCherry antibody (HuaBio, HA500049). For the BiFC analysis, we cotransformed A. tumefaciens with YFPN and MoGcn5b-YFPC, YFPC and YFPN-MoSec13, MoGcn5b-YFPC and YFPN-MoSec13, YFPN and YFPC-MoAtg8, YFPC and MoSec13-YFPN, and YFPC-MoAtg8 and MoSec13-YFPN tags into the wild-type strain Guy11. The obtained transformants were photographed using confocal microscopy to observe the fluorescence localization. The excitation wavelength of YFP is 514 nm and the emission wavelength is 527 nm.

Wide-target quantitative liposome analysis

To analyze liposomes from the wild-type strain Guy11 and ΔMosec13 mutant, mycelia were harvested after 36–48 h of incubation in CM liquid medium. The extracted sample supernatants were subjected to analysis using LC-MS/MS on an ultra-high-performance liquid chromatograph ExionLC™ AD coupled with a tandem mass spectrometer QTRAP® 6500. For details on data processing, please refer to the article by Zhu et al. [37].

Detection of protein acetylation in vitro

The GFP-MoGcn5b protein purified from M. oryzae was incubated with purified GST-MoAtg7 and MoAtg8 proteins in vitro, respectively, in the presence of 100 µM acetyl coenzyme A (Sigma-Aldrich, A2056). The reaction mixture included 5×HAT buffer (250 mm Tris-HCl, pH 8.0, 0.5 mm EDTA, 5 mm protease inhibitor [Sangon Biotech, A610425–0025], 5 mm dithiothreitol, 5 mm sodium butyrate, 250 mm KCl, 25% glycerol). The reaction was conducted at 37°C for 1 h, with gentle mixing every ten min. Following completion of the reaction, 5×protein loading buffer was added to the reaction solution, boiled, and the reaction results were observed through immunoblotting.

Phosphorylation and acetylation level detection in vivo

Changes of MoYpk1 phosphorylation level in the wild-type strain Guy11 and the ΔMosec13 mutant were assessed using anti-phospho-MoYpk1 (S619) and anti-MoYpk1 antibody (produced by ABclonal Technology). Changes of the MoHog1 phosphorylation level were assessed with anti-phospho-MAPK/p38 antibody (Cell Signaling Technology, 9211). Changes in the acetylation level of Atg8 were detected using anti-acetyl lysine (AcK) antibody (Abcam, ab190479) [26]. Protein loading in all samples was confirmed using the ACTB antibody. The ImageJ software was employed for the analysis of results in relation to gray values.

Statistical analysis

Western blot bands were quantified using ImageJ software. Band intensities were normalized to the corresponding loading control. Data are expressed as mean ± standard deviation/SD of at least three independent biological replicates. Data were statistically analyzed using GraphPad Prism 8. Unpaired two-tailed student’s t-test was used for comparison between two groups. P-value <0.05 were considered statistically significant, P-value >0.05 were not statistically significant.

Funding Statement

This study was supported by the Key Research and Development Program of Zhejiang [2024SSY0104], the National Key Research and Development Program of China [2023YFD1400202], the National Natural Science Foundation of China [32270201], the Lingyan Research and Development Project of Zhejiang Province [2022C02029], and the National Institutes of Health [GM131919].

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/15548627.2025.2499289