The Exocyst Regulates Hydrolytic Enzyme Secretion at Hyphal Tips and Septa in the Banana Fusarium Wilt Fungus Fusarium odoratissimum

ABSTRACT

Hyphal polarized growth in filamentous fungi requires tip-directed secretion, while additional evidence suggests that fungal exocytosis for the hydrolytic enzyme secretion can occur at other sites in hyphae, including the septum. In this study, we analyzed the role of the exocyst complex involved in the secretion in the banana wilt fungal pathogen Fusarium odoratissimum. All eight exocyst components in F. odoratissimum not only localized to the tips ahead of the Spitzenkörper in growing hyphae but also localized to the outer edges of septa in mature hyphae. To further analyze the exocyst in F. odoratissimum, we attempted single gene deletion for all the genes encoding the eight exocyst components and only succeeded in constructing the gene deletion mutants for exo70 and sec5; we suspect that the other 6 exocyst components are encoded by essential genes. Deletion of exo70 or sec5 led to defects in vegetative growth, conidiation, and pathogenicity in F. odoratissimum. Notably, the deletion of exo70 resulted in decreased activities for endoglucosidase, filter paper enzymes, and amylase, while the loss of sec5 only led to a slight reduction in amylase activity. Septum-localized α-amylase (AmyB) was identified as the marker for septum-directed secretion, and we found that Exo70 is essential for the localization of AmyB to septa. Meanwhile the loss of Sec5 did not affect AmyB localization to septa but led to a higher accumulation of AmyB in cytoplasm. This suggested that while Exo70 and Sec5 both take part in the septum-directed secretion, the two play different roles in this process.

IMPORTANCE The exocyst complex is a multisubunit tethering complex (MTC) for secretory vesicles at the plasma membrane and contains eight subunits, Sec3, Sec5, Sec6, Sec8, Sec10, Sec15, Exo70, and Exo84. While the exocyst complex is well defined in eukaryotes from yeast (Saccharomyces cerevisiae) to humans, the exocyst components in filamentous fungi show different localization patterns in the apical tips of hyphae, which suggests that filamentous fungi have evolved divergent strategies to regulate endomembrane trafficking. In this study, we demonstrated that the exocyst components in Fusarium odoratissimum are localized not only to the tips of growing hyphae but also to the outer edge of the septa in mature hyphae, suggesting that the exocyst complex plays a role in the regulation of septum-directed protein secretion in F. odoratissimum. We further found that Exo70 and Sec5 are required for the septum-directed secretion of α-amylase in F. odoratissimum but with different influences.

INTRODUCTION

Exocytosis constitutes the late stage of fungal secretion, where distinct cargo proteins are directed to different cell surface sites and subsequently translocated into the extracellular space (1). Several conserved protein families regulate each step of exocytosis to maintain the fidelity of the tethering process, i.e., multisubunit tethering complexes (MTCs) that serve as the first, long-range, reversible connection between a vesicle and its target membrane (2). The exocyst complex is the MTC for secretory vesicles at the plasma membrane and contains eight subunits, Sec3, Sec5, Sec6, Sec8, Sec10, Sec15, Exo70, and Exo84, which are well defined in eukaryotes from yeast (Saccharomyces cerevisiae) to humans (3, 4). All exocyst subunits are encoded by a single gene in fungi and animals, whereas plants often have multiple genes for individual exocyst components (4, 5).

Although the basic secretory machinery is conserved among eukaryotes, organisms have evolved divergent strategies to regulate endomembrane trafficking, which may explain why we find different subcellular localizations of the exocyst complex in various fungi. In Saccharomyces cerevisiae, exocyst components localize at the site of bud emergence and the tip of small daughter cells and subsequently relocate to the mother-daughter connection during cell cycle progression (6, 7). In the filamentous fungi Neurospora crassa, Ashbya gossypii, Magnaporthe oryzae, and Aspergillus oryzae, the exocyst subunits are all located at the hyphal tips, but various patterns of localization have been found in these fungi (8–11). For instance, the movement of exocyst components in A. gossypii showed a correlation with hyphal growth rate, which was not observed in other filamentous fungi (9). Cellular expansion primarily occurs at the hyphal tip in filamentous fungi; therefore, the localization of exocyst proteins at hyphal tip is consistent with the requirement of exocytosis. However, recent studies have shown that exocytosis can occur not only at the hyphal tip but also subapically, especially for some hydrolytic enzymes that are secreted from septa (11, 12). In A. oryzae, an exocyst subunit, Sec3, is considered to be involved in α-amylase secretion at the septum (11).

Recent studies suggest that the exocyst complex plays a specialized role in the regulation of fungal growth and pathogenicity with diverse functions in yeast and filamentous fungi (8–11). Deletion of the exocyst subunit sec5 homolog in N. crassa did not result in a lethal phenotype as in S. cerevisiae, although a severe growth defect was observed. Disruption of the exo70 gene in the rice blast fungus M. oryzae led to defects in the secretion of cytoplasmic effectors and, ultimately, pathogenicity (13). In Botrytis cinerea, the deletion of exo70 significantly reduced fungal growth, production of conidia and sclerotia, and pathogenicity (14).

Fusarium oxysporum f. sp. cubense (Foc), the fungal pathogen causing Fusarium wilt of banana (Musa spp.), is one of the most devastating plant diseases in the world. Based on host cultivar specificity, Foc can be classified into four races (15). Foc tropical race 4 (TR4) is a serious emerging threat to banana cultivation due to its strong virulence, affecting nearly all banana cultivars, including Cavendish, which accounts for 28% of local consumption as well as 15% of export products (16). Several studies have found that Foc has a polyphyletic origin; therefore, Maryani et al. recently revised the taxonomy of Foc and designated Foc TR4 as Fusarium odoratissimum (16). In this study, we systematically analyzed and explored the biofunctions of the exocyst and its role in the secretion process in F. odoratissimum.

RESULTS

Δexo70 and Δsec5 exhibit defects in vegetative growth, conidiation, and virulence of F. odoratissimum.

We identified eight exocyst components in Saccharomyces cerevisiae and subsequently analyzed the homologs of these exocyst components in F. odoratissimum (see Table S1 in the supplemental material). We also targeted all eight genes for the knockout experiment but only obtained exo70 and sec5 gene deletion mutants, which we designated Δexo70 and Δsec5, respectively. We failed to get single gene deletion mutants for the other 6 exocyst component genes from at least 5 independent transformations. Deletion of these genes may be lethal because of the gene replacement efficiency of the homologous recombination approach in F. odoratissimum, and homologs of these 6 genes also cannot be deleted in M. oryzae (10); additional details are provided in the supplemental material (Fig. S1). Biological phenotypes, including vegetative growth, conidiation and virulence, were analyzed among the wild-type, Δexo70, Δsec5, and complemented strains to study the functions of Exo70 and Sec5 in F. odoratissimum. When these strains were incubated on complete medium (CM) and minimal medium (MM) solid agar plates for 3 days, the colonies of Δexo70 and Δsec5 strains were exhibited smaller colony diameters than those of the wild-type and complemented strains (Fig. 1A). However, we further monitored the vegetative growth of the wild-type, Δexo70, Δsec5, and complemented strains in liquid CM and MM after incubation for 3 days. We showed that there is no distinguishable difference in mycelial weight and hyphal morphology among each strain (Fig. 1B and C), suggesting that the deletion of exo70 or sec5 affects the growth rate of F. odoratissimum on solid media but not in the liquid media. We further evaluated how Δexo70 and Δsec5 strains respond on CM agar plates supplemented with the cell wall stress agents Congo red (CR) and calcofluor white (CFW), the cell-membrane stress agent SDS, and the osmotic stress agent NaCl. We learned that Δexo70 and Δsec5 show higher tolerance to CR, CFW, SDS, and NaCl treatments than the wild-type and complemented strains (Fig. S2), suggesting that the deletion of exo70 or sec5 results in higher resistance against environmental stress factors in F. odoratissimum.

FIG 1.

Vegetative growth of the wild-type (WT) and the exo70, and sec5 deletion mutant (Δexo70 and Δsec5) and complemented (Δexo70-C and Δsec5-C) strains. (A) Colonies of the above-listed strains were grown on complete medium (CM) or minimal medium (MM) agar plates for 3 days. The hyphae on the CM plate were photographed by microscopy. (B) Statistical analysis of colony diameters of the indicated strains on CM or MM agar plates. (C) The dry weight of mycelia of the above-listed strains incubated in complete medium (CM) or minimal medium (MM) for 3 days. (D) Conidiation from the indicated stains on SNA plates was quantified. Three independent biological experiments were performed for each strain with three technical replicates. Error bars represent the standard deviation (SD) of three replicates, and values on the bars followed by the same letter are not significantly different at P = 0.01.

Conidia play a key role in the fungal disease life cycle, and rapid propagation of conidia as an inoculum source can lead to a subsequent higher rate of infection and disease development (17). To assess conidiation in Δexo70 and Δsec5, the strains were cultured on Spezieller Nährstoffarmer agar (SNA) medium for 7 days before we quantified conidia. The assay showed that conidia production in Δexo70 and Δsec5 was dramatically reduced compared with that of the wild-type and complemented strains (Fig. 1D). However, the loss of exo70 or sec5 did not lead to morphological defects or reduced germination rate (Fig. S3).

We evaluated the virulence of Δexo70 and Δsec5 on banana plant roots and leaves. Two months postinoculation, we observed noticeable vascular discoloration in the corms of the banana plantlets inoculated with conidia of the wild-type and complemented strains. Meanwhile, the banana plantlets inoculated with the two mutants showed less discoloration in the corm, with Δexo70 causing fewer necrotic symptoms than Δsec5 (Fig. 2A). Disease index analysis confirmed the observations (Fig. 2A). Each strain was inoculated on slightly wounded banana leaves to further assess the invasive growth capacity. We found that the leaves inoculated with the wild-type and complemented strains showed discernible necrosis and lesions, while Δexo70 caused visibly less cell necrosis on banana leaves (Fig. 2B). The necrotic lesions caused by Δsec5 were larger than those caused by Δexo70 but smaller than those caused by the wild-type strain (Fig. 2B). These results confirmed that the deletion of exo70 has a greater adverse impact on F. odoratissimum virulence than the deletion of sec5.

FIG 2.

Pathogenicity of the wild-type (WT) and the exo70, and sec5 deletion mutant (Δexo70 and Δsec5) and complemented (Δexo70-C and Δsec5-C) strains. (A) Disease symptoms on corms of banana tissue plantlets were recorded 2 months after inoculation with the indicated strains and negative-control water. The percentage of plants in each disease index was scored for corm symptoms. 0 represents no browning in the corm, 1 represents 1% to 20% browning area in the corm, 2 represents 20% to 40% browning area in the corm, 3 represents 40% to 60% browning area in the corm, and 4 represents 60% to 100% browning area in the corm (36). (B) Necrosis spots on banana leaves were recorded 5 days after inoculation with the indicated strains and negative-control water. The bottom numbers indicate the relative necrosis size caused by the indicated strains compared with the WT strain.

The exocyst complex localizes to the hyphal tip and septum in F. odoratissimum.

In order to further analyze the mechanisms of F. odoratissimum exocyst, the subcellular location of all exocyst components were observed by tagging each with green fluorescent protein (GFP) at the carboxyl terminus using the split marker technique (8). However, the Sec15-GFP did not express well in the transformants, so we fused GFP to the amino terminus of Sec15 and finally succeed in seeing the fluorescent signals of GFP-Sec15. These modified fusion proteins were expressed from the endogenous loci, and all were functional. All generated strains displayed wild-type morphology, including vegetative growth, conidiation, and pathogenicity (Fig. S4). The localization of each exocyst subunit in growing F. odoratissimum hyphae was monitored, and all subunits localized to a crescent structure at the growing hyphal tip (Fig. 3). In filamentous fungi, polarized secretion is essential for fungal apical growth, and this process involves the polarized traffic of secretory vesicles to the vesicle supply center (VSC), a vesicle-dense region in the hyphal tip (18). The Spitzenkörper acts as a VSC directing secretory vesicles for cell wall biogenesis to the hyphal tip, which can be stained with FM4-64 (19). Thus, we further observed the Spitzenkörper by dying it with FM4-64 and found the cellular localization of the Spitzenkörper as a bright spot at the center of the apical dome (Fig. 3). When these two signals were merged, exocyst proteins localized more broadly at the hyphal tip and did not completely overlap the Spitzenkörper. A line scan of the fluorescence signals further confirmed the above-described observation (Fig. 3). In addition, we observed that all exocyst components can also localize to mature hyphal septa (Fig. 4, Fig. S5). All exocyst subunits localized as a ring structure at the septum when scanned with confocal microscopy (Fig. 4), suggesting that the exocyst complex localizes only to the outer rim of the septum.

FIG 3.

The subcellular localization of exocyst components in F. odoratissimum. Images of F. odoratissimum hyphae were generated with laser scanning confocal microscopy. The dye FM4-64 was used to label the Spitzenkörper. Epifluorescence micrographs were overlaid to observe relative localization, and a line scan graph was generated at the position shown by the arrow to show FM4-64 (red) and GFP fluorescence (green). Bar = 5 μm.

FIG 4.

The subcellular localization of exocyst components at the septum in F. odoratissimum. Images of F. odoratissimum hyphae were generated with laser scanning confocal microscopy. Bar = 5 μm.

In the filamentous fungi M. oryzae and N. crassa, cellular localization for 7 of the 8 exocyst components was systemically observed, while homologs of the last exocyst component, Sec10, could not be labeled with GFP for an unknown reason in these two fungi (8, 10). In our study, we succeeded in labelling all eight exocyst components with GFP and observed their cellular localization in F. odoratissimum (Fig. 3 and 4). To further assess the composition of the exocyst in F. odoratissimum, Sec10-GFP was affinity-purified from hyphal protein extracts and analyzed by liquid chromatography-tandem mass spectrometry (LC-MS/MS). These affinity purification (AP)-MS raw data (Table S2) were filtered against a control data set obtained from GFP-expressing cells by precipitating GFP-tagged proteins with functions unrelated to the exocyst. The remaining seven exocyst subunits were identified by immunoprecipitation with Sec10, consistent with the predicted octameric nature of the complex.

The localization of the exocyst at the hyphal tip is dependent on microtubules and the actin cytoskeleton, while septum localization is only dependent on microtubules.

We studied the behavior of Sec3-GFP and Exo70-GFP after depolymerization of the actin and microtubule (MT) cytoskeletons with the treatment of Latrunculin A (LatA, an actin cytoskeleton inhibitor) and methyl benzimidazole carbamate (MBC, a microtubule inhibitor), respectively. We found that the subcellular localization of Exo70-GFP and Sec3-GFP to septa and hyphal tips disappeared when hyphae were treated with MBC for 1 h. In addition, only hyphal tip localization of Exo70-GFP or Sec3-GFP was affected by LatA treatment (Fig. 5). These observations suggest that the hyphal tip localization of the exocyst is dependent on the microtubule and actin cytoskeleton, while the localization of the exocyst at the septum is only dependent on the microtubule.

FIG 5.

The subcellular localization of Exo70-GFP and Sec3-GFP in F. odoratissimum after 1 h of treatment with dimethyl sulfoxide (DMSO) (as a control), F-actin inhibitor (LatA), and microtubule inhibitor (MBC). Bar = 10 μm.

Exo70 plays a more direct role than Sec5 in the secretion of extracellular hydrolases.

To further test whether Exo70 and Sec5 are involved in the regulation of exocytosis, secreted proteins from the wild-type, Δexo70 and Δsec5 strains were analyzed by SDS-PAGE. While the overall levels of secreted protein in the three strains were similar, we did notice some differences in the protein band patterns of the three samples (Fig. 6A). Secreted extracellular hydrolases, e.g., cellulase, xylanase, pectinase, and amylase, are important for the infection process in plant-pathogenic fungi and are particularly needed to overcome physical obstacles such as the plant cell wall. To test whether the exocyst participates in the secretion of extracellular hydrolases, we used Czapek’s medium supplemented with bran to induce the production of cellulase and amylase. Based on previous studies (20, 21), filter paper can be used as a substrate to detect the total cellulase activity, including endoglucanase, exoglucanase, and β-glucosidase, so filter paper activity (FPA) can represent the activity of the total cellulase. Subsequently, we evaluated the activity of cellulase by analyzing endoglucanase (EG) and FPA. The deletion of exo70 resulted in decreased activities for EG, FPA, and amylase and compared to the wild-type and complemented strains (Fig. 6B to D). However, the loss of sec5 did not affect the activities of EG and filter paper enzymes, but led to a slightly reduced amylase activity. These results suggest that Exo70 plays a more direct role in the secretion of extracellular hydrolases.

FIG 6.

Exo70 and Sec5 are required for protein secretion in F. odoratissimum. (A) Silver-stained gel of secreted proteins in the wild-type (WT) strain and exo70 and sec5 deletion mutant (Δexo70 and Δsec5) strains cultured in liquid MM for 72 h. (B to D) The activities of the filter paper enzymes (B), amylase (C) and EG enzymes (D) secreted from the indicated strains after 7 days of cultivation in Czapek’s medium supplemented with bran were detected using the DNS method. One unit of enzymatic activity is defined as 1 μg/min reducing glucose released from the substrate at pH 4.6 at 50°C. Error bars represent the standard deviation (SD) of three replicates, and values on the bars followed by the same letter are not significantly different at P = 0.01.

Exo70 and Sec5 influence α-amylase cellular localization in F. odoratissimum.

Previous studies found that fungal exocytosis occurs not only at hyphal tips but also at septa, especially for exocytosis of select hydrolytic enzymes (11, 22). In A. oryzae, the important enzyme α-amylase (AmyB) can be secreted from septa, and AmyB colocalizes with the exocyst component Sec3 (11). In F. odoratissimum, deletion of exo70 or sec5 resulted in reduced amylase activity, especially in Δexo70. Thus, we questioned whether the Exo70 or Sec5 is important for the regulation of septum-directed secretion of α-amylase in F. odoratissimum. When we followed the AmyB-GFP localization pattern, we observed GFP signals on the cell membrane and septa. Furthermore, AmyB-GFP only accumulated at the edge of septa, which is in accordance with the localization of exocyst to septa in F. odoratissimum. Then, we observed AmyB-GFP localization in the Δexo70 and Δsec5 strains. In Δexo70, AmyB-GFP localization was not observed in the cell membrane and septum but, rather, accumulated in the cytoplasm (Fig. 7). The septal localization of AmyB-GFP in the Δsec5 strain was similar to that in the wild-type strain, but a higher GFP signal in the cytoplasm was observed. These observations suggested that Exo70 plays a more important role in regulation of the secretion of α-amylase at the septum, which in accordance with the previous results of amylase activities in Δexo70 and Δsec5. In addition, we inoculated the wild-type strain, Δexo70, Δsec5, and the complemented strains on the yeast extract peptone agar (YPA; 3 g/liter yeast extract, 5 g/liter peptone, and 15 g/liter agar) medium plates containing starch (0.4%) or glucose (0.4%) as the carbon source, and after 3 days of incubation we found that the Δexo70 and Δsec5 colonies on YPA with starch were significantly smaller than the Δexo70 and Δsec5 colonies on the YPA with glucose, while the wild-type and the complemented strains did not show this difference (Fig. S6), suggesting that Δexo70 and Δsec5 have reduced ability to utilize the starch, which is consistent with our previous finding that Δexo70 and Δsec5 have defects in secretion of amylase.

FIG 7.

Exo70 is required for the secretion of α-amylase (AmyB). The subcellular localization of AmyB-GFP in the WT and exo70 and sec5 deletion mutant (Δexo70 and Δsec5) strains. Bar = 10 μm.

The SNARE protein Snc1 works in conjunction with the exocyst complex to regulate the secretion of α-amylase at the septum.

In S. cerevisiae, the exocyst works downstream of the Rab GTPases Sec4, Cdc42, and Rho3 and tethers secretory vesicles to the plasma membrane prior to the actual fusion step mediated by SNARE proteins, such as Sso1 and Snc1 (23). We asked whether these conventional exocytic regulators also participate in F. odoratissimum septum-dependent exocytosis. To test this hypothesis, we identified SNARE protein homologs Sso1 (FOIG_01471) and Snc1 (FOIG_02364), along with the Rho GTPase homolog Rho3 (FOIG_11072), in F. odoratissimum for functional study. Loss of Sso1 severely inhibited vegetative growth in F. odoratissimum, while the deletion of Snc1 and Rho3 had a minimal impact on F. odoratissimum growth (Fig. 8A and B). We expressed the AmyB-GFP fusion protein in these mutants, and we found that AmyB-GFP proteins localize to the membrane and septa in the Δrho3 and Δsso1 strains, which was similar to the localization of AmyB-GFP in the wild-type strain. However, in the Δsnc1 strain, AmyB-GFP also accumulated in the cytoplasm but did not localize to the membrane or septa. These results suggested that Snc1 participates in the regulation of AmyB localization in F. odoratissimum (Fig. 8C). We analyzed α-amylase activity in these strains but observed no difference when the wild-type and complemented strains were compared, indicating that Snc1 is involved in septal secretion of AmyB but is not essential for this function.

FIG 8.

Snc1 is involved in the secretion of α-amylase (AmyB). (A) Colonies of the wild-type (WT), Δrho3, Δsnc1, and Δsso1 and complemented (Δrho3-C, Δsnc1-C, and Δsso1-C) strains on CM agar after 3 days. Bar = 30 μm (B) Statistical analysis of colony diameter of the indicated strains on CM medium after 3 days. Error bars represent the standard deviation (SD) of three replicates, and values on the bars followed by the same letter are not significantly different at P = 0.01. (C) The subcellular localization of AmyB-GFP in WT, Δrho3, Δsnc1, and Δsso1 strains. Bar = 10 μm.

DISCUSSION

In this study, we characterized the functions of the exocyst complex in the banana pathogen F. odoratissimum. We generated deletion mutations in two genes, exo70 and sec5, suggesting that other exocyst-encoding genes are essential for F. odoratissimum. Published reports show that the roles of the eight exocyst subunits we targeted in this study vary in different filamentous fungi and yeasts, and many components are essential for viability (8–11, 24–27). In S. cerevisiae, among the eight exocyst-encoding genes, only sec3 deletion produced a viable mutant strain (28). Deletion of sec3 is also not lethal in Candida albicans (26), but all eight exocyst subunits were deemed essential for growth in A. gossypii (9). However, in the filamentous fungal pathogen M. oryzae, deletion of exo70 or sec5 was not lethal when eight exocyst components were targeted (10). Most exocyst subunits were also essential for viability in N. crassa, but sec5 mutants are viable (8). In B. cinerea and Verticillium dahliae, homologs of exo70 are nonessential genes (14, 29). Notably, in the plant fungal pathogens M. oryzae, B. cinerea, V. dahlia, and F. odoratissimum, deletion of exo70 or sec5 homologs resulted in impaired growth, development, and virulence (10, 14, 29), suggesting that these two genes are relatively conserved among these fungi.

In filamentous fungi, it is generally recognized that exocytosis, which delivers cell membrane and wall components in addition to secretory enzymes, takes place at the hyphal tips. Systematic observation of an exocyst localization pattern in filamentous fungi confirmed the hyphal tip location of exocyst in M. oryzae, A. gossypii, N. crassa, and F. odoratissimum, but the detailed localization patterns show some differences (8–10). In this study, we observed that all eight exocyst components display consistent localization patterns in the hyphal tips as a crescent but distinct from the Spitzenkörper (Fig. 3), which is similar to the localization of the exocyst in M. oryzae (10). In A. gossypii, the exocyst subunits Sec3, Sec5, and Exo70 localized as a cortical cap at the hyphal tip in slowly growing cells but to the Spitzenkörper in rapidly growing hyphae (9). In N. crassa, the exocyst complex occupies two key locations, with Sec5, Sec6, Sec8, and Sec15 localizing as a crescent at the hyphal tip, while Exo70 and Exo84 closely associate with the peripheral part of the Spitzenkörper (8).

In addition to the hyphal tip location of M. oryzae exocyst, Gupta et al. (10) found that the M. oryzae exocyst localizes specifically in a ring form at the appressorium pore and is necessary for appressorium repolarization and host cell invasion during host infection. F. odoratissimum is a soilborne vascular wilt fungus, and during the infection of banana roots, hyphae penetrate into the roots from the intercellular space of the epidermis with no appressorium or hyphopodium-like structure (26). In this study, we also observed that all eight exocyst components localize to the hyphal septum, and this localization has not been reported in A. gossypii, N. crassa, or M. oryzae (8–10). In A. oryzae, Hayakawa et al. (11) found that the accumulation of Sec3 in septa contributes to septum-directed secretion. We observed that all F. odoratissimum exocyst subunits localized at the edge of the septa as a ring, which is consistent with the suggested exocyst complex role in tethering secretory vesicles to the plasma membrane.

In filamentous fungi, additional evidence suggests that fungal exocytosis can occur at additional sites in hyphae, including the septum. In A. oryzae and Aspergillus niger, several hydrolytic enzymes have been found to accumulate at the septa, including RNase T1 (RntA), glucoamylase (Gla), and α-amylase (AmyB) (11, 12, 22). Certain plasma membrane transporters also localized at the septa, including the purine/xanthine permease AoUapC and the general amino acid transporter AoGap1 in A. oryzae and the plasma membrane H⁺-translocating ATPase (PMA-1) in N. crassa (11, 30). Moreover, these septum-localized proteins were shown to be secreted from septa (11, 30). In N. crassa and M. oryzae, chitosan, the product of chitin deacetylation, is localized to hyphal septa, and the corresponding M. oryzae chitin deacetylase Cda1 also localized to septa (31, 32). Deacetylation of chitin by Cda1 can lead to the loss of structural integrity and therefore increase cell wall permeability, allowing hydrolytic enzymes to better access substrates (32). Thus, efficient secretion from septa may be dependent upon higher cell wall permeability.

One question that has not been clarified in our study or in the literature is whether the protein machinery mediating exocytosis also plays a role in septum-directed secretion. In A. oryzae, Hayakawa et al. (11) found that the accumulation of AmyB-eGFP at the septum was largely abolished in the t-SNARE sso1 conditional mutant under repressive conditions, suggesting that Sso1 is required for α-amylase secretion. In our study, the deletion of sso1 and rho3 did not affect AmyB-GFP localization, while deletion of snc1 led to abnormal AmyB-GFP localization but did not impact α-amylase activity, suggesting that Snc1 is involved, but not essential, in F. odoratissimum α-amylase secretion. Collectively, the data suggest that apical exocytosis and septum-directed exocytosis only partially share molecular machinery, including some SNARE proteins, and the exocyst complex, but further investigation is needed to better understand the subapical septal secretion mechanism.

MATERIALS AND METHODS

Fungal strains, medium, and culture conditions.

Fusarium odoratissimum strain 58, which was isolated, identified, and sequenced in our previous studies (33, 34), was used as the wild-type (WT) strain in this study. The genome information of the WT strain was previously deposited into BIG Sub under accession number PRJCA001282. All strains used in this study were stored as mycelial suspensions in 20% glycerol solution at −80°C. Complete medium (CM) (20× nitrate salts, 50 ml liter⁻¹; 1,000× trace elements, 1 ml liter⁻¹; 1,000× vitamin solution 1 ml liter⁻¹; glucose, 10 g liter⁻¹; peptone, 2 g liter⁻¹; yeast extract, 1 g liter⁻¹; Casamino Acids, 1 g liter⁻¹; and agar, 15 g liter⁻¹) adjusted to pH 7 and minimal medium (MM) (NaNO₃, 6 g liter⁻¹; KCl, 0.52 g liter⁻¹; MgSO₄·7H₂O, 0.312 g liter⁻¹; vitamin B1, 0.01 g liter⁻¹; 1,000× trace elements, 1 ml liter⁻¹; glucose, 10 g liter⁻¹; and agar, 15 g liter⁻¹) adjusted to pH 6.5 were used to test vegetative growth. Spezieller Nährstoffarmer agar (SNA; 1 g KH₂PO₄, 1 g KNO₃, 0.5 g MgSO₄·7H₂O, 0.5 g KCl, 0.2 g glucose, 0.2 g sucrose, 20 g agar, and 1 liter water) medium was used for conidiation assays (35).

Construction of gene deletion mutants and complementation strains.

The double-joint PCR approach was used to generate the gene-replacement constructs for the deletion mutants in F. odoratissimum (36). The primers used to amplify upstream and downstream fragments of all exocyst complex subunits (including sec3, -5, -6, -8, -10, and -15 and exo70 and -84), rho3, snc1, and sso1 are listed in Table 1. For gene complement, the entire target genes (without stop codon), including their native promoter regions, were amplified by PCR and transformed with XhoI-digested pYF11 vector using the yeast gap repair approach (37). The resulting transformants carrying the neomycin-resistant marker were introduced into the gene mutant protoplasts, and the resulting transformants were selected in Geneticin (150 μg/ml)-containing medium. The methods for fungal transformation and mutation screening were described previously (38).

TABLE 1.

Primers used in this study

Namea	Sequence (5′–3′)b
sec3-up	F:AGGACCAGGGTCTTCCTTGA
	R:TTGACCTCCACTAGCTCCAGCCAAGCCCAGGAGCGTTAGGATGAGCA
sec3-down	F:GAATAGAGTAGATGCCGACCGCGGGTTTACCACTGCCAGGAAGCTGA
	R:ATGCCCTTTCGCACTTCCTT
sec3-id	F:CCCTTTCAGATGACATCGCTT
	R:TTGCCTGTCTGGCATTCCT
sec5-up	F:TCGAATGGCCAGCATTATGA
	R:TTGACCTCCACTAGCTCCAGCCAAGCCGTGGCGTCGCATACTCGATA
sec5-down	F:GAATAGAGTAGATGCCGACCGCGGGTTGCTTCGCGATGTCATTCAA
	R:TCTTCACAGGCCGATGCTTAT
sec5-id	F:AGCGATGCGTGACAGCATT
	R:AACTTGACGGAGAAGGCATT
sec6-up	F:AAGAAGACCCCACGAGCA
	R:TTGACCTCCACTAGCTCCAGCCAAGCCTCTTGTCTAGATCGTCGGGAT
sec6-down	F:GAATAGAGTAGATGCCGACCGCGGGTTGGACGAGTCATGATTTGCGTT
	R:TCGGAACCAACCAGCAGAA
sec6-id	F:GATGGAACGACCTGCAGTAA
	R:TTCGTGTCGCCGATATAATG
sec8-up	F:TGGTGAGGGACCTCCTAAAA
	R:TTGACCTCCACTAGCTCCAGCCAAGCCAGTATGAGGTTGGGAGATGGC
sec8-down	F:GAATAGAGTAGATGCCGACCGCGGGTTAAGACGATACGTGATGCACGA
	R:AGGTTCATTCTAGACCGACGA
sec8-id	F:CATCCAGGATCGAGATTCCAT
	R:TGTCTGCCACATGTGCTCACT
sec10-up	F:ATGAGGACGTGGATTTGGTT
	R:TTGACCTCCACTAGCTCCAGCCAAGCCTGTTTCTGATGAGCGTAGAGG
sec10-down	F:GAATAGAGTAGATGCCGACCGCGGGTTAGAGGCGGATCATGGGAAA
	R:TTCACAACGATCTCAGGTACG
sec10-id	F:TCCAGCGTACTCTTGAGCTCG
	R:ACCACTCAAAACGCTCTGAA
sec15-up	F:TTCATCGTGTACAGCGGCTA
	R:TTGACCTCCACTAGCTCCAGCCAAGCCACGCATAATCGTCCACAGCA
sec15-down	F:GAATAGAGTAGATGCCGACCGCGGGTTACGATAATGAGACCATTGGGA
	R:TGCCCAACGAGGAATTTGA
sec15-id	F:TGAGCTCTCAATACCTGGGA
	R:TCGGTTTATGCCAAATCGTG
exo70-up	F:TTTGGACGATGTTCTTCAGGG
	R:TTGACCTCCACTAGCTCCAGCCAAGCCTGAATAGCATCCGTTCAGGT
exo70-down	F:GAATAGAGTAGATGCCGACCGCGGGTTCATAGGGCAAAGGGGGTAATT
	R:ATATCTCCGGTCCCTCCCA
exo70-id	F:CACGCGGAGCACTCATAACT
	R:AAGATCGGGGAAACCGGTTG
exo84-up	F:TGTAGCTTCGCTGGTCTGTGA
	R:TTGACCTCCACTAGCTCCAGCCAAGCCTTGGGCAAGTGTTGCAAGTA
exo84-down	F:GAATAGAGTAGATGCCGACCGCGGGTTATGTTGGTTGAGGTGGATGGA
	R:TGAAATGACGTTGATGCGGT
exo84-id	F:TGAGGCTGGCTTGTTCGGTA
	R:GCACTTCTACACTGAACCGC
rho3-up	F:GGGACGAAGATTAGAGATGG
	R:CATTCATTGTTGACCTCCACTAGCTCCATCTGACAAATGTGATTCAAA
rho3-down	F:GCAAAGGAATAGAGTAGATGCCGACCGACATGATACGATACGAGCGT
	R:GCAAAGGAATAGAGTAGATGCCGACCGACATGATACGATACGAGCGT
rho3-id	F:GAACCCACCGTCTTTGAGAA
	R:AGGTATCGGAGGGCGTTTAT
snc1-up	F:GCGTCTGTGACCGTCATTGT
	R:CATTCATTGTTGACCTCCACTAGCTCCATTTGGCTGTGGTTTGGAGAT
snc1-down	F:GCAAAGGAATAGAGTAGATGCCGACCGTTGTGGCTACTCGTTAGTAG
	R:CTCCGTTGCATCATGATAGA
snc1-id	F:GATGACGCCTCCTTTTGGTA
	R:GACTCTGTTGGCTCCTCGAC
sso1-up	F:TATTGGACCCTGCATCGCCG
	R:CATTCATTGTTGACCTCCACTAGCTCCACTCGACGCTACCTATCCCTC
sso1-down	F:GCAAAGGAATAGAGTAGATGCCGACCGGTAGGAGTCGAACGATGAGA
	R:CTCCCGAAGGCGAGTCGAAC
sso1-id	F:AGAACAACCCTCACGTCACC
	R:GCCACACCAATCTCCTCGTT
HYG	F:TGGAGCTAGTGGAGGTCAACAATGAATG
	R:GTATTGACCGATTCCTTGCGGTCCGAA
HY	F:GATGTAGGAGGGCGTGGATATGTCCT
	R:CGGTCGGCATCTACTCTATTCCTTTGC
sec3-GFP	F:TACCACTGCCAGGAAGCTGA
	R:GCCTCCGCCTCCGCCTCCGCCGCCTCCGCCACGGAATGCGGCCTTGACGTC
sec3-3′ UTRF	F:GAATAGAGTAGATGCCGACCGCGGGTTGTCTGGTGCTTAGAGAGAATG
	R:ATTGTTGTGCGGTGTCTGCTT
sec5-GFP	F:ACGAATGCCAGCCAGCTT
	R:GCCTCCGCCTCCGCCTCCGCCGCCTCCGCCGGTTGAGATACCGGACAGAC
sec5-3′ UTR	F:GAATAGAGTAGATGCCGACCGCGGGTTAAGGCTCTGGATCCGCATATT
	R:CCAACAACACTTCAAAGGCA
sec6-GFP	F:AACGTCGAGGGCTCCAACT
	R:GCCTCCGCCTCCGCCTCCGCCGCCTCCGCCCTTGACCCTGCTCATAATAGT
sec6-3′ UTR	F:GAATAGAGTAGATGCCGACCGCGGGTTCCGAGGACGAGTCATGATTTGCR:TGCTGATGAGGATTTTGCCA
sec8-GFP	F:TCAGAAGAGATGGACTGGGAA
	R:GCCTCCGCCTCCGCCTCCGCCGCCTCCGCCTGTCTGCCACATGTGCTCACT
sec8-3′ UTR	F:GAATAGAGTAGATGCCGACCGCGGGTTATGGACCAAGACGATACGTG
	R:ATGCTGCAAAATGGATGACG
sec10-GFP	F:AGGATGCTTATATCGAGCGTC
	R:GCCTCCGCCTCCGCCTCCGCCGCCTCCGCCAAGACCACTCAAAACGCTCTG
sec10-3′ UTR	F:GAATAGAGTAGATGCCGACCGCGGGTTGCGTTGGTGCAAGAGGCGGA
	R:TGGAACTTGCTCGTGCATCA
sec15-GFP	F:ACATCCGGAACTTCCTGAAT
	R:GCCTCCGCCTCCGCCTCCGCCGCCTCCGCCTGTTCGGTTTATGCCAAATCGT
sec15-3′ UTR	F:GAATAGAGTAGATGCCGACCGCGGGTTTTTAGTACTGCAACGATAATGA
	R:TCGCTCCCGATCAAGACTA
exo70-GFP	F:GAACGTTGCGTGAACTCAAT
	R:GCCTCCGCCTCCGCCTCCGCCGCCTCCGCCGGAAGCCAAACTGGAAAACA
exo70-3′ UTR	F:GAATAGAGTAGATGCCGACCGCGGGTTGCATAGGGCAAAGGGGGTAAT
	R:ATGACGATGACCTCTATGGCT
exo84-GFP	F:GGACGCAATAAGCTTGCAGAT
	R:GCCTCCGCCTCCGCCTCCGCCGCCTCCGCCTGACAACCCAAGTCCGACAG
exo84-3′ UTR	F:GAATAGAGTAGATGCCGACCGCGGGTTACAGGAGATTTAAGATGTTG
	R:TCGAGGGTGGGGATCGTAGTA
amyB-GFP	F:ACTCCCTGCGACATCGACTA
	R:GCCTCCGCCTCCGCCTCCGCCGCCTCCGCCGCTAGAGGCAACCAAGACCT
amyB-3′ UTR	F:GCAAAGGAATAGAGTAGATGCCGACCGATGGCATAAAAATGGGGTTG
	R:GTTCGGCCCAATATCAAGAT
10*Gly-GFP	F:GGCGGAGGCGGCGGAGGCGGAGGCGGAGGCATGGTGAGCAAGGGCGAG
	R:TTGACCTCCACTAGCTCCAGCCAAGCCTTACTTGTACAGCTCGTC

^a“Up” and “down” indicate PCR primers to amplify upstream and downstream fragments for the construction of targeted gene deletion mutant, respectively; “id” indicates PCR primers for identification of targeted gene deletion transformed strains; “GFP” and “UTR” indicate PCR primers used to amplify the GFP-tagged fragment of the target gene; HYG/F, HY/R, and HY/F, HYG/R amplify upstream and downstream of the HPH fragment.

^bF, forward; R, reverse.

Vegetative growth and conidiation assays.

For growth assays on medium plates, fresh mycelial plugs (5 mm diameter) of each strain taken from the periphery of a 3-day-old colony were inoculated on CM or MM agar plates. After a 3-day incubation at 25°C, the colonies of each strain were measured and photographed. For the mycelial weight experiment, fresh mycelial plugs (5 mm diameter) of each strain taken from the periphery of a 3-day-old colony were inoculated on CM or MM liquid plates. After a 3-day incubation at 25°C, the mycelial dry weight was determined. For conidiation assays, five fresh mycelial plugs of each strain were inoculated on SNA plates and incubated at 28°C for 7 days. The number of conidia was determined for each strain using a hemocytometer. These experiments were repeated three times independently.

Virulence assays.

Banana plantlets (Cavendish banana, AAA cultivar) at the five-leaf stage were cultivated in a glasshouse (temperature, 25°C; light, 12 h; humidity, 75%). For banana root infection, F. odoratissimum strains were cultured in potato dextrose broth (PDB) medium for 3 days to induce spore formation. Before inoculation, the banana roots were wounded to facilitate infection; the wounded roots were immersed in a conidia suspension (10⁶ conidia ml⁻¹) for 8 h, planted in vermiculite, and maintained in a growth chamber. A total of 10 banana plantlets were used for each treatment. After inoculation for 2 months, the internal score was used to measure the discolored corm area of individual plants (39). For banana leaf infection assays, fresh mycelial plugs (5 mm diameter) of each strain were injected into wounded banana leaves after surface sterilization using 75% ethyl alcohol, and infection symptoms were observed after inoculation for 5 days. These experiments were repeated three times independently.

Secretory protein extraction and detection.

For secretory protein assays, three fresh mycelial plugs (5 mm diameter) were inoculated into a 250-ml flask containing 100 ml of MM broth. After culturing for 72 h, the filtrated supernatant was centrifuged at 5,000 rpm at 4°C for 10 min to remove mycelium and conidia, and then protease inhibitor was added to the supernatant. Subsequently, the extracellular proteins were precipitated for 8 h on ice with ice-cold 70% sodium sulfate (wt/vol). The protein samples were collected by centrifugation at 16,000 rpm at 4°C for 20 min and solubilized in lysis buffer (10 mM Tris-HCl, 1 mM EDTA, pH 7.5). Then, the protein solution was purified with protein centrifugal filters (Millipore; Amicon Ultra). For the relative abundance assay, the mycelium was separated from the supernatant and dried to determine the mass. The amount of supernatant used for protein extraction was adjusted based on the dry mass obtained from the cultures. Extracted secretory proteins were detected using the fast silver stain kit (Sangon, China).

Activity detection for extracellular hydrolases.

For detection of extracellular hydrolase activities, three fresh mycelial plugs (5 mm diameter) were inoculated into a 250-ml flask containing 100 ml of Czapek’s (CS) medium with 2% bran for approximately 7 days at 25°C. Mycelia were completely removed by filtration, and the culture filtrates were used for the measurement of extracellular enzyme activities. The activities of amylase and cellulase were determined by using the 3,5-dinitrosalicylic acid (DNS) method with simple modification as previously described (20, 21). The dry weights of the harvested mycelia were measured to normalize the enzyme activities.

Generation of GFP fusions.

We searched the F. odoratissimum genome for orthologs of all known exocyst components previously identified in S. cerevisiae (Table S1) and designed primers to tag the endogenous genes with GFP using a split marker gene replacement procedure (8). Eight subunits of the exocyst complex were labeled with GFP at the C terminus, except Sec15, which was labeled at the N terminus. The AmyB proteins were labeled by amplifying sequences from the AmyB-GFP fusion vector with the primers listed in Table 1, cloned into the pKNT GFP vector using a one-step cloning kit (Vazyme Biotech, China), and verified by sequence analysis (40).

Affinity purification and mass spectrometry (AP-MS).

To identify putative exocyst-interacting proteins in F. odoratissimum, the wild-type strain was transformed with Sec10 fluorescently labeled with GFP, and the resulting transformant was used for protein extraction. In addition, the strain transformed only with GFP was used as a control. Protein extraction was performed as a previous described protocol (41). Approximately 50 μl of anti-GFP agarose (Abmart, Shanghai, China) was added to capture Sec10-interacting proteins following the manufacturer’s instructions. After incubation at 4°C overnight, the agarose was washed three times with 500 μl of TBS (20 mM Tris-HCl and 500 mM NaCl, pH 7.5). Proteins binding to the beads were immediately eluted with 60 μl of elution buffer (0.2 M glycine, pH 2.5). The eluant was instantly neutralized with 3 μl of neutralization buffer (1.5 M Tris, pH 9.0). The protein eluates were sequenced by The Beijing Genomics Institute (Beijing, China).

Light and epifluorescence microscopy.

Fresh mycelia were collected for live cell imaging by shaking mycelia for 24 h in CM medium. A Nikon A1R laser scanning confocal microscope system (Nikon, Japan) or an Olympus BX51 microscope (Olympus, Japan) was used for live-cell imaging. GFP excitation was performed with 488 nm light (emission [Em.] 525/40 nm). Hyphal tips were visualized by staining with FM4-64 at a final concentration of 10 μg/ml with 405 nm light (Em. 452/45 nm) and photographed.

Data availability.

The data that support the findings of this study are available from the corresponding authors upon reasonable request.