Endogenous labeling of cse-8 with gfp was carried out using the Split Marker method (Smith et al., 2011). The oligonucleotides cse-8/Gly-FW and cse-8/Gly-RV (Table 1) were used to directly amplify the ORF of the cse-8 gene using genomic DNA from N. crassa FGSC #988 strain as a template. Likewise, the oligonucleotides FW-lox/3’cse-8 and RV-lox/3’cse-8 ( Table 1 ) were used to amplify the 3’ UTR region of the cse-8 gene. The plasmid pRS416 (Honda and Selker, 2009), was used as a template to amplify the green fluorescent protein gene (gfp) fused to 10 glycines and the hph gene (hygromycin B conferring resistant gene) using the oligonucleotides loxP-R and 10xGly-F ( Table 1 ). Subsequently, fusion PCRs were performed to fuse the PCR products and obtain the constructs cse-8::gfp::hph and hph::cse-8. The resulting constructs were purified after running them by gel electrophoresis (1% agarose), and conidia of N. crassa (FGSC #9718) were transformed with 500 ng of each construct by electroporation in a Biorad Gene Pulser Electroporation using 0.2 mm electroporation cuvettes (600 OHMS, 25 μFD, 1.5 KV).

All fungal or bacterial strains used or generated in this study are listed in Table 2 . N. crassa strains were grown in Vogel’s minimum medium (VMM) supplemented with 1.5% sucrose and 1% agar (Vogel, 1956). For phenotypic characterization, 1x10⁵ conidia were inoculated onto VMM plates supplemented with NaCl (0.8 mM), KCl (0.8 mM), and Congo Red (100 mg/mL) to induce osmotic and cell wall stress. For optimal growth, inoculated flasks and plates were incubated at 30°C (Davis, 2000). For the selection of hygromycin-resistant transformants, conidia were incubated in a recovery solution (1X Vogel’s salts and 2% yeast extract) after electroporation. After 3 hours of incubation, transformed conidia of FGSC #9718 were inoculated onto FGS medium supplemented with hygromycin B (300 μg/mL). The conidia of FGSC #9717 strains were inoculated on FGS plates with or without (for negative controls) histidine (0.25 mg/mL). The FGS medium contains 2% Vogel’s salts, 1% agar, and 10% FGS solution (0.5% fructose, 0.5% glucose, and 20% sorbose). All media components were sterilized by filtration, and the media were autoclaved at 15 lbf/in² for 15 minutes. Synthetic crossing medium (50% SCM 2X solution, 2% sucrose, and 1.5% agar) was used to obtain homokaryotic Δcse-8 strains expressing CHS-1-GFP, CHS-3-GFP, or CHS-5-GFP, and a double mutant strain Δcse-8;Δcse-7.

We used an Olympus SZX12 stereoscopic microscope adapted to a C-HP4 8MP 4K, Full HD C-mount camera and equipped with an Olympus DF PLAPO 1XPF objective to image colony morphology and perithecia. An Olympus Fluoview FV1000 inverted Laser scanning confocal microscope (LSCM) was used to image GFP, dsRED, mCherry, and RFP-tagged proteins, employing 488 nm and 543 nm Lasers. FM4-64, diluted in liquid VMM to a final concentration of 5 µM, was used to stain the NSSG3 strain to identify CSE-8 in subcellular compartments. Nuclei were stained with a stock solution of Hoechst 33258 (100 mg/mL) in distilled water, diluted to a final concentration of 26 µg/mL in filtered PBS buffer (pH=5.2) before use. Stock solutions of 1M 1,4-dithiothreitol (DTT, Sigma-Aldrich) in water and 200 µM tunicamycin (TM, Sigma-Aldrich) in dimethyl sulfoxide (DMSO) were prepared as ER stressors and used at final concentrations of 1.25 µM and 4 µg/mL in VMM, respectively. A stock solution of brefeldin A (BFA, Sigma-Aldrich) in DMSO was prepared at 20 mg/mL and diluted in MMV at a final concentration of 200 µg/mL. This concentration was selected based on prior studies demonstrating its efficacy in disrupting CHS-4-GFP (class III) transport to the SPK in N. crassa (Fajardo-Somera et al., 2015). All samples were incubated at 30°C for 15 to 20 minutes before imaging, following the inverted agar method (Hickey et al., 2004), and were observed with a PLAPLON 60X N.A. 1.42 objective. Images were acquired using FLUOVIEW FV1000 4.0.2.9 software and analyzed with Fiji Image J version 2.1.0/1.53c software. For FRAP (Fluorescence Recovery After Photobleaching) and FLIP (Fluorescence Loss in Photobleaching) analyses of the CSE-8-GFP strain, selective regions of interest (ROI) of the hyphae were selected for photobleaching and overexposed to 52% of the laser intensity for five seconds. Fluorescence intensity profiles at the SPK region during FRAP and FLIP experiments were analyzed in Fiji Image J using the image and batch tools in the Stowers plugins (https://research.stowers.org/imagejplugins/); easyFRAP web (https://easyfrap.vmnet.upatras.gr/) was also used for the analysis of a photobleached ROI in region II. In this case, a non-photobleached ROI in a distal zone of the hypha and an ROI in the background were used as controls. Spinning disk confocal microscopy (SDCM) was performed to observe the dynamics of CSE-8-GFP and CSE-7-mCherry. A Nikon ECLIPSE Ti-E Ti-E/B inverted microscope was used with a Yokogawa CSU-X1 confocal scanner unit, an ANDOR iXon Ultra camera, and an Apo 60X /0.13-0.21 oil immersion objective. Mean shift super-resolution (MSSR, Torres-García et al., 2022) analysis was applied to improve the resolution of the LSCM and SDCM images. We used an amplification parameter (AMP) value of 3 and a 0-order analysis, while the point spread function (PSF) value was obtained using the ‘ImageDecorrelationAnalysis’ plugin. Co-localization profiles were obtained with JACoP (Bolte and Cordelières, 2006).

To further investigate the role of CSE-8 in the CHS secretory pathway and cell wall chitin synthesis, a colorimetric assay was performed to quantify GlcNAc content in the WT, Δcse-8, Δcse-7, and Δcse-8; Δcse-7 strains following the specifications previously described (Morgan and Elson, 1934; Fajardo-Somera et al., 2015).

To analyze the distribution of orthologues of CSE-7 (NCU05720) and CSE-8 (NCU01814) in the fungal kingdom and other organisms, an alignment of CSE-8 and CSE-7 was performed to identify the conserved domains of both proteins. The alignment was then used as input for HMMsearch (https://www.ebi.ac.uk/Tools/hmmer/search/phmmer) to obtain orthologous sequences to the CSE proteins. The resulting sequences were used for phylogenetic analysis, selecting amino acid sequences from representative species of each phylum. The sequences selected were aligned using MAFFT in Jalview v. 2.11.2.5. The resulting alignment was used to construct a phylogenetic tree by maximum likelihood using the Jones-Taylor-Thorton method, with a bootstrap value of 10000 and an amino acid substitution model. The phylogenetic tree obtained was edited in iTOL v6 (Letunic and Bork, 2024; https://itol.embl.de/tree/1589766147326311667424835#). The distribution of transmembrane domains, -sheets, and -helix structures displayed were carried out based on Uniprot (https://www.uniprot.org/) and AlphaFold (https://colab.research.google.com/github/sokrypton/ColabFold/blob/main/AlphaFold2.ipynb). Only structures predicted by AlphaFold with a pLDDT (per-residue measure of local confidence) greater than 50 were considered to outline the secondary structure of the proteins. Alpha-Fold and I-TASSER (https://zhanggroup.org/I-TASSER/) were utilized to predict the protein structures of CSE-8. Molecular docking analyses were performed using the HADOCK SERVER (Yan et al., 2020); http://hdock.phys.hust.edu.cn/) and AlphaFold 3 (Jumper et al., 2021; https://alphafoldserver.com). Predicted protein structures and potential interactions between CHS and CSE proteins were analyzed and visualized in PyMOL Molecular Graphics System, Version 2.5.4, Schrödinger, LLC.

All graphs and statistical analysis presented in this article were performed using GraphPad Prism version 9.

Unlike in yeasts, little is known about the mechanisms governing the vesicular transport of CHSs in filamentous fungi. In order to investigate potential components of the protein machinery involved in this transport, the orthologues of the yeast Chs7 were identified in N. crassa. The function of the Chs7 orthologue CSE-7 (NCU05720) in the transport of CHS-4 (Class IV) has previously been described in N. crassa hyphae (Rico-Ramírez et al., 2018). CSE-7 acts as a cargo receptor for CHS-4, facilitating its transport from the ER to the SPK and septa. We identified a second Chs7 orthologue, NCU01814, and named it CSE-8 (for chitin synthase export chaperone 8) after confirming its role in CHS intracellular traffic.

CSE-8 (Q1K536) was previously annotated as a hypothetical protein and classified as a member of the pfam12271 protein family, which includes proteins with the Chs3 catalytic domain. BLASTp analysis revealed that CSE-8 shares 25.42% identity with Chs7 (J4U2B4) and 31.12% with CSE-7 (Q7SB92; Figure 1A ). Like CSE-7 (358 aa, 39.3 kDa) and Chs7 (316 aa, 34.9 kDa), CSE-8 (299 aa, 33.1 kDa) contains seven transmembrane alpha-helix regions and four cytoplasmic domains with conserved amino acid residues ( Figure 1B ). Despite their structural similarities, CSE-7 is unique in having a long-disordered domain at its C-terminal region ( Figure 1B ).

To assess the conservation of these proteins across the fungal kingdom, we performed an HMM search. These proteins had E-values ranging from 2.5 x 10^-152 to 0.77. Of the identified sequences, 927 were from Ascomycota, 380 from Basidiomycota, 67 from Mucormycota, 45 from Zoopagomycota, 19 from Microsporidia, 26 from Chytridiomycota, 7 from Blastocladiomycota, and 2 from Cryptomycota. Representative species from each phylum were selected to construct the phylogenetic tree, where CSE proteins grouped into two distinct clades, one corresponding to CSE-7 orthologues and the other to CSE-8 orthologues. As expected, CSE-7 clustered with Chs7 from S. cerevisiae, reflecting the conserved identity between these two proteins. Most of the analyzed sequences grouped within the CSE-7 clade, which is characterized by the conservation of a disordered region in the C-terminal part of the protein. Most of the phylogenetically analyzed CSE proteins share common features, including two conserved β-sheets in the N-terminal region and seven predicted transmembrane domains. Most CSE proteins have a molecular weight below 45 kDa, fitting the typical profile of chaperones, which are usually small proteins. Larger CSEs (Rozella allomycis A0A075AX29 and A0A075AVG7, Rhizoctonia solani A0A0K6FLV5, Carpinus fangiana A0A5N6KZK2), are enriched with repeated β-sheet and -helix structures.

Different species had varying numbers of non-redundant sequences containing the characteristic domains of CSE-7 and CSE-8. For example, Basidiobolus meristosporus (Zoopagomycota), Fibularhizoctonia sp. (Basidiomycota), Conidiobulus coronatus (Zoopagomycota), and Absidia rapens (Mucormycota) have 9, 3, 4, and 3 proteins from the pfam12271 family, respectively ( Figure 1B ). Within the Ascomycota, filamentous species such as Aspergillus spp., Penicillium spp., Trichoderma spp., and Fusarium spp. have two genes encoding CSE proteins. Yeast-like and dimorphic species, such as S. cerevisiae and C. albicans, have only one CSE protein distributed in the CSE-7 clade. Only the single CSE of Pneumocystis murina was distributed in the CSE-8 clade. However, the dimorphic fungal pathogen Coccidioides immitis is an exception, possessing two CSE proteins. Another exception was Ustilago maydis, which retains a single copy of the CSE and has three morphological stages during its life cycle, including its yeast-like form (Cabrera-Ponce et al., 2012).

These results suggest that filamentous ascomycetous likely share a conserved mechanism for CHS function, with CSE-7 and CSE-8 serving as auxiliary proteins for the biogenesis of CHS vesicular carriers ( Figure 1 ). Interestingly, one of the sequences identified in the alignment corresponded to a CSE protein from Carpinus fangiana (Viridiplantae, Streptophyta, Betulaceae), commonly known as Fang’s hornbeam. The homology of this sequence was confirmed by E-value (6.2 x 10^-118) and bit score (405). The C. fangiana CSE protein (A0A5N6KZK2) is 709 amino acids long, contains six transmembrane regions, and is predicted to be a multipass membrane protein. Notably, this protein also preserves a homologous region with the α/β hydrolase fold superfamily (IPR029058), a diverse group of hydrolytic enzymes with different phylogenetic origins and catalytic functions. Additionally, it contains a pleckstrin homology domain, which is typically involved in protein targeting and signal transduction pathways. However, no experimental data currently exists about this unique plant CSE.

To determine the subcellular localization of CSE-8 during polarized growth, the N. crassa homokaryon strain expressing CSE-8-GFP was analyzed using LSCM. The subcellular localization pattern of CSE-8-GFP was consistent with previous observations of CHS-tagged fluorescent proteins of N. crassa (Riquelme et al., 2007; Sánchez-León et al., 2011; Fajardo-Somera et al., 2015). There was a prevalence of CSE-8-GFP fluorescence in hyphal regions I and III, more precisely in the SPK and apical regions, as well as in subapical tubular structures ( Figures 2A–C ). FM4-64 staining was used to visualize the localization of CSE-8-GFP within the SPK region, revealing that CSE-8-GFP is concentrated at the SPK core, as shown more clearly through MSSR analysis ( Figures 2D, E ; Supplementary Video 1 ). CSE-8-GFP was also observed at septa in mature hyphae, co-localizing with FM4-64 mainly in the central region of the septum ( Figure 2F ). Additionally, CSE-8 shows partial overlap with FM4-64 in subapical regions of the hyphae, as confirmed by the co-localization plots ( Figures 2G, H ), indicating its involvement in endocytic pathways, particularly in vacuoles (Fischer‐Parton et al., 2000; Hickey et al., 2002). CSE-8-GFP’s appearance in FM4-64-stained organelles is consistent with studies on CHS recycling through the trans-Golgi network in a clathrin-dependent manner, as seen in S. cerevisiae, C. albicans, and A. nidulans (Seaman, 2008; Starr et al., 2012; Sacristan et al., 2013; Hernández-González et al., 2018; Knafler et al., 2019; Jin et al., 2021). Further research on CSE proteins will help clarify whether their recycling is tied to their association with CHS or an unrelated mechanism. These findings support a function of CSE-8 associated with CHS subcellular transport.

We aimed to determine whether CSE-8 localizes to the ER, as its orthologue in S. cerevisiae (Chs7) has been identified as an ER chaperone, and the N. crassa CSE-7 was identified in the nuclear periphery (Trilla et al., 1999; Dharwada et al., 2018; Rico-Ramírez et al., 2018). We used RFP-BiP and dsRED-NCA-1 as markers of the rough ER and nuclear envelope, respectively. NCA-1, a homolog of the SERCA-type Ca²⁺-ATPase found in animal cells, primarily localizes to the nuclear envelope in N. crassa (Bowman et al., 2011). On the other hand, BiP, an HSP70 family protein, functions as an ER chaperone involved in post-transcriptional regulation, protein folding, and the recognition of misfolded proteins destined for the unfolded protein response (UPR) pathway under ER stress (Hendershot et al., 1995). For this work, we renamed the N. crassa GRP-78 protein as BiP (Monnerjahn et al., 2001; Bowman et al., 2011) based on its orthology to the mammalian and plant binding immunoglobulin protein and the yeast Kar2. The co-localization of CSE-8-GFP with dsRED-NCA-1 or RFP-BiP in heterokaryon strains was observed in region III of the hyphae ( Figures 3A, B ). Co-localization analysis confirmed the presence of CSE-8 at ER membranes and, to a lower degree, around the nuclei ( Figure 3C ). Pearson’s coefficients showed a stronger correlation between BiP and CSE-8 than between NCA-1 and CSE-8 ( Figure 3C ), further suggesting that CSE-8 is an ER protein potentially involved in the biogenesis of CHS-carrying microvesicles from the ER, as previously reported for CSE-7 and Chs7.

FRAP and FLIP experiments were carried out to elucidate the biogenesis of CSE-8 in N. crassa hyphae ( Figure 4 ). We first assessed whether photobleaching the apex (within the first 5.63 ± 2.18 μm from the tip) affected the fluorescence of CSE-8-GFP at the SPK, where it accumulates, presumably co-transporting CHS. After photobleaching the SPK region, CSE-8-GFP fluorescence at the SPK recovered about 6.06 ± 0.34 s after photobleaching ( Figure 4A ). Since CSE-8-GFP displayed high fluorescence intensity at distal regions of the hypha ( Figures 2A, B ), we also photobleached a distal ROI (67.33 ± 6.38 μm from the tip) and measured fluorescence intensity at the SPK ( Figure 4B ). Fluorescence loss after photobleaching in the subapical region was observed, including a notable decrease in CSE-8-GFP fluorescence at the SPK. The half-time fluorescence recovery value at the SPK was greater for FLIP when applying the photobleaching at an ROI in region III of the hyphae (t_1/2 = 38.95 s), than for FRAP applied directly at the SPK (t_1/2 = 11.36 s) ( Figures 4C, D ). To support our observations, the photobleaching was also applied to region II of the hyphae (10.99 ± 0.78 μm from the apex), which is characterized by a high abundance of nuclei (Figure 4E). FRAP analysis displayed a t_1/2 value of 52.64 s for the photobleached ROI ( Figure 4F ). Fluorescence at the SPK did not completely disappear following photobleaching at region II, suggesting that most CSE-8-GFP vesicles originate from more distal regions ( Figure 4G ), rich in rough ER (Martínez-Andrade et al., 2024). Additionally, a significant portion of the fluorescence may derive from the network of endomembranous cisternae (NEC), where CSE-7 has been previously localized (Rico-Ramírez et al., 2018). These findings suggest that CHS synthesis occurs in the subapical regions and that efficient transport to the apex is essential for polarized growth.

CSE-8 exhibits a subcellular distribution similar to the one previously reported for CSE-7 by Rico-Ramírez et al. (2018). To compare the dynamics of these two proteins, a heterokaryon strain co-expressing CSE-8-GFP and CSE-7-mCherry was analyzed using SDCM. We used SDCM rather than LSCM in order to observe the dynamics of the two proteins in near real-time and to avoid possible spurious co-localization due to scanning time. When the two fluorescence channels were superimposed, overlapping spots of CSE-8 and CSE-7 were observed at subapical regions of the hyphae ( Figure 5A ; Supplementary Video 2 ), around non-fluorescent round organelles. The MSSR plugin in Fiji (Torres-García et al., 2022) was used to enhance the resolution of the confocal images, confirming a partial co-localization between CSE-8 and CSE-7 ( Figures 5C, D ). In addition, microscopy revealed that some GFP and mCherry vesicle clusters moved independently in both retrograde and anterograde directions ( Figure 5B ). In the kymographs corresponding to the subapical region of the hyphae, it can be seen in more detail that both CSE-8-GFP and CSE-7-mCherry have independent trajectories mainly in the anterograde direction, although retrograde displacements are also observed ( Figure 5E ). These results reveal that CSE-8 and CSE-7, are transported in different vesicle sub-populations. To determine whether the organelles observed near CSE-7 and CSE-8 were nuclei ( Figure 5B ), the nucleic acid dye Hoechst 22358 was used ( Figure 5F ). MSSR revealed clusters of GFP and clusters of mCherry around the nuclei, with some clusters partially overlapping ( Figure 5G ). Remarkably, non-fluorescent organelles surrounded by CSE-7-mCherry and CSE-8-GFP are still observed in the hypha, suggesting that there could be compartments, other than nuclei, related to both proteins.

Genetic crosses were performed between a strain expressing CHS-3-GFP (NSSG6) ( Figure 6A ) and a Δcse-8 knockout strain to investigate potential interactions between CSE-8 and CHS in N. crassa. Homokaryotic strains recovered from the resulting ascospores were analyzed by LSCM to confirm the presence of fluorescence, followed by validation through PCR using primers flanking the CHS-3-GFP and cse-8 gene constructs ( Supplementary Figure 1 ). In homokaryotic Δcse-8 strains expressing CHS-3-GFP, fluorescence appeared in subapical clusters, with an absence of signal at the SPK and septa ( Figures 6B, C ). A previous study including FRAP analysis of CHS-3-GFP and CHS-6-GFP revealed that photobleaching near the hyphal tip caused only transient disruption in CHS localization at the SPK (Riquelme et al., 2007). Furthermore, the localization of CSE-8-GFP within the lumen of globular vacuoles is consistent with earlier findings for CHS-3-GFP (Riquelme et al., 2007). These results support the hypothesis of intracellular transport of CSE-8 in association with CHS-3. Additionally, Δcse-7 and Δcse-8 strains expressing CHS-1-GFP or CHS-5-GFP were obtained ( Supplementary Figure 2 ). Since no disruption in CHS transport to polarized growth sites was observed for CHS-1 (class III) or CHS-5 (class V), which are unique to filamentous fungi, it is suggested that these CHS may employ alternative transport routes, bypassing ER-to-Golgi COPII vesicles, as proposed in previous studies (Riquelme et al., 2007; Sánchez-León et al., 2011; Rico-Ramírez et al., 2018). Thus, while CSE proteins may still function as chaperones assisting with CHS folding at the ER, they do not appear to be essential for CHS transport to apical regions in filamentous fungi, suggesting the evolution of alternative transport mechanisms.

To explore the possible interaction mechanism between CHS-3 and CSE-8, AlphaFold 3 and the HDOCK server were used to predict potential interaction sites. The model with the lowest average distance values between the atoms was selected and compared to the docking results obtained from AlphaFold 3, which yielded a pTM score of 0.68. A CHS-3 dimer obtained from SwissProt was used for the interaction models with CSE-8 to better predict their real interaction ( Supplementary Figure 3A ). This model of the N. crassa CHS-3 dimer is supported by crystallographic data on CHS structures from S. cerevisiae and C. albicans (Chen et al., 2022; Ren et al., 2022; Chen et al., 2023) and is consistent with the structure of Chs1 (class I) from the oomycete Phytophthora sojae and Chs2 (class I) from Candida albicans (Chen et al., 2022; Ren et al., 2022). In order to obtain more accurate results, the models displayed in Supplementary Figure 4 were used as controls.

According to the interaction model, the 5th, 6th, and 7th transmembrane domains of CSE-8 interact with transmembrane regions 1, 2, 6, and 7 of one CHS-3 unit ( Supplementary Figures 3B, C ). The most stable in silico interaction was predicted between the beta-sheet LPLC domain of CSE-8 and the charged amino acids glutamate, aspartate, and arginine located at the C-terminal end of CHS-3 ( Supplementary Figures 3C, D ). Additionally, potential interactions between a pair of hydrophobic amino acids in the second unit of the CHS-3 dimer were suggested ( Supplementary Figure 3D ). This interaction model aligns with experimental results on the Shr3 chaperone, whose transmembrane domains protect charged amino acids on the N-terminal end of the amino acid permease Gap1 (Kota et al., 2007). Interestingly, the in silico interaction between the β-folded structures of CSE-8 and the transmembrane domains of CHS-3 mirrors behaviors observed in other chaperones. These proteins use their β-sheet structures to facilitate protein folding, prevent accumulation, and contribute to protein oligomerization (Sun and MacRae, 2005; Karamanos et al., 2020).

To further investigate the role of CSE-8 in CHS-3 transport, we examined their subcellular transport under conditions of ER stress induced by DTT and TM ( Figure 7 ). The minimum inhibitory concentrations for observing the effects of these stressors on protein transport were determined to be 1.25 mM for DTT and 4.25 μm/mL for TM. As a positive control for ER stress, the N. crassa heterokaryon strain expressing CSE-8-GFP and RFP-BiP was grown ( Supplementary Figures 5A, B ). TM produced a similar effect to DTT on RFP-BiP distribution ( Supplementary Figures 5E, F ). In subapical regions near the tip, globular clusters of both proteins were observed, indicating that CSE-8-GFP is retained in the ER under stress conditions ( Figures 7A, B ). RFP-BiP, an established ER stress reporter, accumulated in subapical globular bodies in hyphae treated with DTT or TM, consistent with Kar2-sfGFP localization in stressed yeast cells (Lajoie et al., 2012). DTT also produced similar effects in hyphae expressing CSE-7-GFP or CHS-4-GFP, supporting the notion that CSE proteins play a comparable role in vesicular trafficking from the ER ( Supplementary Figure 6 ). Likewise, CHS-3-GFP accumulated in subapical regions under DTT or TM stress, with complete SPK disruption observed at the hyphal tip. CHS-3-GFP appeared as an apical vesicle crescent and in smaller clusters than those presented by CSE-8-GFP ( Figure 7B ).

Despite DTT exposure, CSE-8-GFP and CHS-3-GFP still reached the hyphal tip, forming a disorganized structure like an apical crescent at the apex. FM4-64 staining revealed complete disruption of the SPK, including its outer layer, which surrounds the core visualized by CSE-8-GFP in non-treated cells ( Supplementary Figures 5C, D ). TM did not severely disrupt the SPK, likely due to its primary effect on the folding of glycosylated proteins. It is noteworthy that, under ER stress, CSE-8-GFP predominantly appeared in globular structures resembling globular vacuoles along the hyphae. These globular vacuoles were more pronounced in the CSE-8-GFP strain than in the CHS-3-GFP strain. In contrast, CHS-3-GFP was observed in smaller clusters near the tip. These findings suggest two key conclusions: first, CSE-8’s presence in the NEC could indicate its localization within the ER of this network. Second, the significant accumulation of CSE-8-GFP in globular vacuoles along the hyphae may result from the activation of the unfolded protein response degradation or other autophagic pathways. Further studies are required to determine the exact pathways CSE-8 and CHS-3 follow under stress ER conditions. However, these experiments confirm that CSE-8 is an ER protein based on its subcellular distribution in response to stress.

Severe loss of polarity, hyperbranching, and excessive septa formation in DTT- and TM-treated hyphae underscore the importance of ER integrity for polarized growth ( Figures 7C, E ). CSE-8 and CHS-3 transport appear essential for maintaining this growth, likely activated in response to ER stress to sustain essential cellular processes like chitin synthesis. The constant supply of S. cerevisiae CHS7 under ER stress was identified among genes up-regulated under UPR cell conditions induced by TM and DTT (Travers et al., 2000; Weichet et al., 2020). In the same study, genes corresponding to exomer subunits Chs5 and Chs6 were also up-regulated in response to ER stress. These findings suggest a possible activation of cse genes and other selected genes in response to ER stress to support key developmental processes, including chitin synthesis. Further studies on the response of cse-8 and cse-7 genes under ER stress conditions may help explain the role of CSE proteins in CHS-3 and CHS-4 transport from the ER.

The next step in elucidating the vesicular trafficking pathway of CHS-3, using CSE-8 as a chaperone, was to investigate whether CHS-3 could exit the ER and enter COPII vesicles. Brefeldin A (BFA) inhibits COPII vesicle formation and, consequently, the ER-to-Golgi vesicular transport pathway, leading to the disruption of Golgi cisternae and accumulation of coatomer subunits in ER transition membranes, forming BFA bodies (Orci et al., 1993).

A strain expressing the YPT-1-RFP construct was used as a control and grown in VMM medium poisoned with BFA to ensure Golgi disruption. In this strain, YPT-1-RFP localized to the SPK and accumulated in globular bodies within the subapical region ( Figure 8 ), a hallmark of Golgi membrane disruption. Similarly, CSE-8 accumulated in BFA-induced bodies along the hypha, with prominent accumulations into BFA bodies in the 1^st and 2^nd regions ( Figures 8A, C, D ). CHS-3-GFP clusters were also observed near the tip ( Figure 8B ). BFA treatment led to an increase in fluorescence intensity across all regions, confirming the disruption of CSE-8-GFP, YPT-1-RFP, and CHS-3-GFP transport ( Figures 8E-G ). These results indicate that CSE-8 is transported to the Golgi in COPII vesicles, following the canonical transport route to the hyphal tip. This is consistent with previous observations for Chs3, where the cargo receptor Erv14 is required for its exit from the ER in COPII vesicles (Sacristan et al., 2013).

To investigate the function of CSE proteins, we characterized corresponding mutants to identify any distinct phenotypic differences. No significant differences in colonial growth were observed in the Δcse-8 and Δcse-7 single mutants ( Figures 9A, B ). However, the Δcse-8; Δcse-7 double mutant exhibited a slower growth rate than the single mutants and the WT strain ( Figures 9A, B ). Growth of the Δcse-8 and Δcse-7 mutants was significantly impaired under osmotic stress conditions induced by high concentrations of NaCl and KCl ( Figure 9B ). Furthermore, both the double mutant and Δcse-7 were more sensitive to Congo red (CR)-induced stress than the WT. CR is a molecule known for its antifungal activity, as it can inhibit chitin synthesis and form complexes with chitin and glucans present in the cell wall (Bartnicki-Garcia et al., 1994; Ram and Klis, 2006). Taken together, these results indicate that the absence of cse-7 and cse-8 renders N. crassa hyphae more sensitive to osmotic stress-induced changes in turgor pressure, as well as to antifungal agents targeting cell wall and chitin synthesis, directly linking both genes to the mechanisms of chitin synthesis in cell wall. The average number of branches at a selected distance from the tip of the main hyphae was lower in the mutants tested ( Figure 9C ). However, statistical analyses did not reveal any significant difference in this morphological characteristic. Levels of the chitin monomer GlcNAc were consistent across all three mutants, each showing approximately 50% less GlcNAc than the WT strain ( Figure 9D ). Notably, knockout mutants Δchs-5, Δchs-6, Δchs-7, and RIP mutant chs-1 — mutants in chs specific to filamentous fungi — display a severely affected radial growth phenotype (Yarden and Yanofsky, 1991; Fajardo-Somera et al., 2015). These findings suggest that, at least in N. crassa, CSE proteins are unlikely to be directly involved in the transport of CHSs unique to filamentous fungi. However, the possibility that CSE interacts with these specific CHS cannot be entirely excluded, as the significant reduction in GlcNAc levels observed in Δchs-1, Δchs-6, and Δchs-7 knockout strains (Fajardo-Somera et al., 2015) is similar to the results seen here for the single and double knockout mutants for CSE proteins. Further experiments are needed to investigate the potential relationship between CSE and other CHSs.

The absence of cse-8 also appeared to affect sexual reproduction in N. crassa, resulting in perithecia with two beaks ( Figure 9E ). We termed this phenotype “shaka” due to its resemblance to the hand gesture used by surfers in Hawaii. Around 20% of the perithecia produced in homozygous and heterozygous sexual crosses exhibited the shaka phenotype ( Figure 9F ), although this did not affect the production of viable ascospores. Despite there being no previous research on this shaka phenotype, chs-3 is essential for the production of mature perithecia (Fajardo-Somera et al., 2015), suggesting that the absence of cse-8 could affect sexual development in N. crassa. The developmental timing of mature perithecia does not differ from that of WT fruiting bodies (Lehr et al., 2014). Although cse-8 expression varies during perithecium development (Lehr et al., 2014), it does not appear to be essential, as functional beaks and viable ascospores are still formed in Δcse-8 mutants. Therefore, cse-8 provides an interesting model for studying ostiole biogenesis in ascomycete fungi.