The tail domain of the Aspergillus fumigatus class V myosin MyoE orchestrates septal localization and hyphal growth

The notoriety of filamentous fungi as pathogens can largely be attributed to their ability to form hyphae. Hyphae are highly polarized complexes that grow by apical extension and are separated into compartments by septa, which help to differentiate apical polarized tip extension (apical compartments) and lateral branching (subapical compartments) (Riquelme, 2013; Harris, 2011; Takeshita, 2016). Despite these two processes being integral to survival, surprisingly little is known about the molecular mechanisms behind them. Polarized hyphal growth involves the delivery of secretory vesicles to sites of cell growth by motor proteins, such as kinesin, myosin and dynein (Takeshita, 2016; Takeshita et al., 2014). Myosins are a family of ATP-dependent proteins that are divided into classes on the basis of their structure (Hammer and Sellers, 2012). Class V myosins are actin-based cargo transporters that carry necessary cell wall-building components to the hyphal tip and contain four domains: the head that generates the force; the neck, which functions as a regulatory domain; the coiled-coil that aids in dimerization of the protein; and the tail, which functions as a cargo-binding region (Hammer and Sellers, 2012).

In the model yeast, Saccharomyces cerevisiae, one of its two type V myosins, Myo2p, is essential and deletion of its tail domain is lethal (Catlett and Weisman, 1998). The Myo2p tail is sufficient for localization to the bud site, and overexpression of the tail leads to a dominant-negative phenotype due to interference with the function of the endogenous Myo2p (Reck-Peterson et al., 1999). Furthermore, point mutations in Myo2p lead to vacuolar inheritance defects and secretory vesicle distribution, demonstrating its essential role in cargo trafficking (Schott et al., 1999; Catlett and Weisman, 1998). Additional work in the fission yeast Schizosaccharomyces pombe showed that distinct regions of the tail of one of its two type V myosins, Myo52, are required for cell polarity, cell division and Myo52 movement throughout the cytoplasm (Mulvihill et al., 2006).

The S. cerevisiae tail domain is phosphorylated in vivo and cargo binding to a vertebrate class V myosin is regulated by phosphorylation (Legesse-Miller et al., 2006; Karcher et al., 2001), suggesting that post-translational modification plays an important role in the function of class V myosins. We have previously demonstrated that the class V myosin MyoE in the human pathogenic filamentous fungus Aspergillus fumigatus is required for hyphal polarization and proper septation, as well as virulence in a murine model of invasive aspergillosis (Renshaw et al., 2016), making it critical to understand how these processes work. This background prompted us to examine the importance of the tail domain of MyoE in A. fumigatus and to also understand how post-translational modification may affect its function.

In this study, we generated iterative truncations of the tail domain of MyoE and determined that the tail domain is required for radial growth, conidiation, and MyoE's location at the septa and hyphal tips, the two most active areas of the growing fungus. Additionally, we describe for the first time the phosphorylation and acetylation status of all domains of a class V myosin. A. fumigatus MyoE is phosphorylated in vivo at eight residues and acetylated at one residue under basal conditions, and is phosphorylated at one additional site in the absence of the catalytic subunit of the serine/threonine phosphatase calcineurin. Similar to the locations of MyoE, calcineurin A (CnaA, UniProt ID: Q4WUR1) also localizes at the hyphal tips and septa (Juvvadi et al., 2008). Generation of nonphosphorylatable, nonacetylatable or phosphomimetic mutants of these sites did not result in phenotypic differences to the wild-type strain, with the exception of one phosphorylation site in the ΔcnaA background strain. Taken together, these findings demonstrate the importance of the tail domain and phosphorylation/dephosphorylation in the localization and function of MyoE as a regulator of hyphal growth and septation.

To define the importance of specific areas of the tail domain for MyoE function, we generated a series of C-terminal truncations of MyoE (named MyoE^Δ1084, MyoE^Δ1160, MyoE^Δ1249, MyoE^Δ1389) and determined requirements for (1) radial growth, (2) conidiation, (3) septation and (4) localization to the hyphal tip and septa (Fig. 1A). To confirm proteins were stably expressed, we subjected protein extracts to western blot analysis by using an anti-GFP antibody (Fig. 1B and S1A). The MyoE^Δ1084 and MyoE^Δ1160 proteins were stable and comparable to the wild-type strain; however, while the MyoE^Δ1249 protein showed a faint band, the MyoE^Δ1389 protein showed no band (Fig. 1B). These data indicate that the MyoE^Δ1249 protein was not as stable as the MyoE^Δ1084 and MyoE^Δ1160 truncated proteins and that the MyoE^Δ1389 protein was degraded due to improper folding of the truncated protein. To compensate for protein loss, we then generated myoE^Δ1249 and myoE^Δ1389 strains with the truncated proteins under the control of the strong constitutively active promoter pOtef (Langfelder et al., 2001). The myoE^pOtef-Δ1249 strain showed more protein than the original truncation strain while the myoE^pOtef-Δ1389 strain showed no protein band, indicating that the MyoE^Δ1249 protein, although overexpressed, was still being degraded. Thus, we continued our analyses using the myoE^pOtef-Δ1249 strain and did not further analyze the strain with the MyoE^Δ1389 truncation. All tail truncation mutants displayed reduced radial growth in comparison to the wild-type strain (P<0.001) but none was as severely growth restricted as the ΔmyoE strain (P<0.001) (Fig. 1C,D).

The ΔmyoE strain appeared white on agar and exhibited significantly reduced conidiation compared to the wild-type strain (Renshaw et al., 2016), and the tail truncation mutant strains were less pigmented than the wild-type strain. Quantification of conidial production per mm² to account for the decreased radial growth revealed that all three tail truncation mutants exhibited reduced conidial production in comparison to the wild-type strain when grown on agar (P<0.001) (Fig. 1E). Conidial production in the tail truncation mutant strains was partially remediated by sorbitol (P<0.01), suggesting a cell wall defect.

MyoE is required for proper septation frequency (Renshaw et al., 2016); thus, we assessed septation in the tail mutant strains. We used Aniline Blue, which stains β-(1,3)-glucan in the cell walls, and found that septa had normal morphology; however, some strains had irregular interseptal distances. The myoE^Δ1160 strain exhibited greater apical compartment length (P<0.05) and the myoE^pOtef-Δ1249 strain had smaller apical compartment length (P<0.001), while the myoE^Δ1084 strain had apical compartments similar to those in the wild-type strain (P>0.05) (Fig. 1F). All three tail truncation strains had smaller sub-apical compartment lengths in comparison to that in wild-type strain (P<0.001).

MyoE localizes to the hyphal apex as motile, dot-like structures near the hyphal tip and throughout the cytoplasm, as well as two stable discs on either side of each septa (Renshaw et al., 2016). We visualized MyoE in the tail truncation mutants by tagging each with GFP. Deletion of any part of the tail domain resulted in loss of MyoE at the septa (Fig. 2A). MyoE localized to the hyphal apex, similar to the wild-type strain; however, hyphal apex localization was a diffuse pattern in the cytoplasm near the hyphal tip. Importantly, although the myoE^pOtef-Δ1249 strain generated slightly less MyoE protein than the wild-type strain, bright MyoE staining was visible at the hyphal tip, suggesting a threshold of MyoE protein level that is necessary for its localization to the hyphal tip, and this threshold is met in the myoE^pOtef-Δ1249 strain. Additionally, no motile dot-like structures were visible. Phenotypes of the tail truncations are summarized in Table 1.

Given the mislocalization of MyoE in the tail truncation mutants, we sought to gain a better understanding of MyoE localization during germination and septum formation through live cell imaging. Time-lapse microscopy revealed that, in the wild-type strain, MyoE localized to the tips of germ tubes as soon as they emerge (Fig. 2B; indicated by white circles). Next, MyoE was located as a ring structure across the hyphae at a possible pre-septation site, as no septum was visible initially. Ten minutes later, a septum was established at the location where the ring structure had been initially visible; subsequently, the ring disappeared and MyoE localization was seen as a double disc on either side of the septum (Fig. 2B; indicated by white arrows) that remained throughout the duration of the imaging. This actin ring-like pattern of MyoE located at pre-septal sites is visible at multiple time points.

We also followed MyoE localization in the myoE^Δ1084 strain by time-lapse microscopy (Fig. 2B; lower panel; indicated by white circle). In this strain, with the entire tail domain removed, MyoE localizes to the hyphal apex as soon as a germ tube emerges and remains stable at the tip. However, no rings were visible and no septal localization was seen at any time point, indicating the requirement of the tail domain for septal localization.

Given that the S. cerevisiae tail domain is essential for protein function and because the truncation of A. fumigatus MyoE tail mislocalized MyoE from the hyphal septum without impacting its localization at the hyphal tip, which is indicative of its partially active nature, we investigated the sequences of the tail domains of A. fumigatus against other fungi. We compared the amino acid sequences of the vacuole-binding domain (amino acids 1295–1305) and the vesicle-binding domain (amino acids 1439–1486) of A. fumigatus with other class V myosins (Fig. 2C). The vacuole-binding domain, which contains the Vac17p-mediated binding site in S. cerevisiae (Catlett et al., 2000), is highly conserved amongst pezizomycotina (Aspergillus nidulans and Neurospora crassa), but is less conserved when compared with other ascomycetes (S. cerevisiae, Candida albicans, and S. pombe), the basidiomycete Cryptococcus neoformans, and humans. The vesicle-binding domain, which contains the secretory vesicle-binding region (Catlett et al., 2000), is identical in A. nidulans, noticeably absent in N. crassa, and contains conserved regions shared among ascomycetes, basidiomycetes and humans. The high similarity of residues in these domains indicates that both the vacuole- and secretory vesicle-binding characteristics of class V myosins are likely conserved among different organisms; however, the A. fumigatus and the S. cerevisiae tail domains are only 41% similar at the amino acid level, suggesting that the differences in function are in the less-conserved regions. To determine if the S. cerevisiae tail domain was sufficient for MyoE function, we attempted heterologous expression of the tail domain of budding yeast instead of the MyoE tail domain; however, codon optimization of the S. cerevisiae tail yielded several possible stop codons and we were unable to produce a full-length protein to test this hypothesis.

Coordinated transport of vesicles and other organelles in cells requires both the microtubule and actin networks. Because MyoE localized as dot-like structures at the hyphal tip, we sought to determine the localization of full-length MyoE and the various tail-truncated versions of MyoE by using reagents that disrupt the actin or microtubule networks. Treatment with cytochalasin A (8 µg/ml) that inhibits actin polymerization or with the microtubule disruptor benomyl (100 µg/ml) after growth for 18 h resulted in mislocalization of MyoE from the hyphal tip in the wild-type strain and all three tail truncation strains (Fig. 3A,B). No specific MyoE localization was seen at the hyphal tip in the presence of cytochalasin in any strain. In the presence of benomyl, the wild-type strain exhibited some concentrated MyoE localization near the hyphal tip; however, this was dissimilar to that seen in the untreated wild-type strain. Importantly, septal localization of MyoE remained stable in the wild-type strain in the presence of either agent but mislocalized in all the truncated strains. To determine if actin was required for transport of MyoE to the hyphal septum, we inoculated conidia in the presence of cytochalasin A (0.25 µg/ml) and allowed them to grow overnight. Growth was extremely inhibited after 18 h; however, MyoE still remained at the septa. Similar localization patterns at the hyphal tip and septa were seen following treatment with the actin-depolymerizing drug latrunculin A in A. nidulans (Zhang et al., 2011). The fact that polarized growth still occurred after 18 h of cytochalasin A treatment suggests that actin is not completely depolymerized and, thus, the possibility that some actin filaments remain near the septa cannot be eliminated.

Post-translational regulation of type V myosins has been reported in yeast and vertebrates. The tail of the S. cerevisiae type V myosin Myo2p was shown to be phosphorylated in vivo and phosphorylation of the vertebrate type V myosin tail domain releases the motor from the organelle in vitro, two observations that suggest that post-translation modification plays an important role in the function of MyoE (Karcher et al., 2001; Legesse-Miller et al., 2006). To test this in A. fumigatus, we determined the phosphorylation and acetylation profile of MyoE after 24 h of growth. MyoE was phosphorylated at eight residues: one in the head and neck domains, four in the coiled-coil domain, and two in the tail domain; acetylation of MyoE occurred at one residue in the head domain (Table 2, Fig. 4A). A-score localization analysis was unable to confirm the exact phosphorylation site for one of the residues due to their proximity to each other and specific trypsin digestion patterns and, thus, we investigated both sites in further analysis (S1036 or S1039).

To assess these residues as possible post-translation regulators, we generated nonphosphorylatable or nonacetylatable mutant strains of each of the sites (Table S1). Phenotypic analyses of these strains determined that mimicking a nonphosphorylated state (S/T to A) or nonacetylated state (K to R) at any one site did not result in a difference in radial growth (data not shown), hyphal branching, hyphal polarity, or septal or tip localization of MyoE when grown under basal conditions (Fig. 4B, myoE^S809A is representative of all strains analyzed). Based on these results, we hypothesized that post-translational modifications are specifically important for stress response regulation. To investigate this we grew the strains on agar supplemented with cell-wall-inhibiting antifungals caspofungin and nikkomycin Z; however, the phosphorylation mutants showed no significant growth difference compared to that of the wild-type strain under these stress conditions (data not shown). Based on our previous data that showed MyoE and calcineurin localization as a dot at the hyphal tip, and two discs on either side of each septa (Renshaw et al., 2016; Juvvadi et al., 2008), we were also interested to examine the impact of calcineurin and Ca²⁺-signaling functions on the localization of MyoE. To decipher this link, we used the calcineurin inhibitor FK506 and the Ca²⁺ chelator EGTA but neither inhibitor affected septal localization of MyoE.

Although FK506 did not affect the localization of MyoE, given the extensive work demonstrating the importance of the serine/threonine phosphatase calcineurin on A. fumigatus growth and septal localization, we were intrigued by the localization of MyoE at the septa and tip that mimicked that of calcineurin. This suggested that calcineurin, although not required for septal localization of MyoE, has a specific and previously undefined role in the dephosphorylation of MyoE. Thus, we also determined the phosphorylation status of MyoE in the absence of cnaA, the gene encoding the catalytic subunit of calcineurin, by using a comparative phosphoproteomic approach. This analysis revealed that MyoE is phosphorylated at one additional residue in its tail domain in the ΔcnaA strain (Fig. 4C,D), although the analysis was unable to determine the specific phosphorylated residue (either T1509 or S1516) due to the proximity of the two residues to each other. We, therefore, generated phosphomimetic strains of both of these individual residues (T1509E and S1516E) to investigate phenotypic effects. The myoE^T1509E strain exhibited radial growth, branching morphology and MyoE localization that were similar to those in the wild-type strain (data not shown, Fig. S2). However, while the myoE^S1516E strain exhibited radial growth that was similar to that in the wild-type strain, MyoE localized at the hyphal tip but not at the septum (Fig. 4E). To ensure that the mislocalization was not simply due to any mutation, we generated a nonphosphorylatable mutant of this site (S1516A). The myoE^S1516A strain exhibited growth and localization indistinguishable from those in the wild-type strain (Fig. 4E), indicating the importance of phosphorylation/dephosphorylation of S1516 for the localization of MyoE at the septa. Western blot analysis for MyoE^S1516A and MyoE^S1516E proteins can be seen in Fig S1B. To better understand if this effect was related to calcineurin, we next determined the localization of MyoE in the absence of cnaA; however MyoE localized as normal to the hyphal tip and septa in this strain (data not shown). Given the analyses in this study of the importance of the MyoE tail domain for maintaining proper septation frequency, we sought to determine if S1516 and, thus, localization at the septa was necessary for septation. As previously demonstrated, we measured the apical and sub-apical compartments in the myoE^S1516E strain. The myoE^S1516E strain has larger apical compartments than the wild-type strain or any of the tail truncations strains examined (Fig. 4F), adding further impetus to the role of phosphorylation of MyoE in regulating the timing of septation. Sub-apical compartments in this mutant were, however, similar in size to those seen in the wild-type strain, as noted in other tail-truncated mutants.

Myosins as molecular motors interact with the key cytoskeletal components (actin and microtubules) and facilitate organelle and vesicular or nonvesicular trafficking via cargo-binding in addition to participating in a wide variety of cellular functions (Hammer and Sellers, 2012). Our previous study, establishing the role of MyoE − the class V myosin in A. fumigatus that is responsible for hyphal growth, polarity and septation, prompted us to examine the mechanistic basis for its regulation (Renshaw et al., 2016). Considering the known functional relevance for the tail domain of class V myosins in S. cerevisiae and phosphorylation as a mechanism of its regulation in vertebrates, in this study we generated a comprehensive library of tail domain truncations and phosphorylation mutants to define the importance of the MyoE tail domain for MyoE function and spatial regulation in A. fumigatus.

The reduction in MyoE^Δ1249- and MyoE^Δ1389-truncated proteins on a western blot suggested that these proteins are degraded in vivo. Proper folding is important for the function of class V myosins (Hammer and Sellers, 2012), and these truncated mutants may alter the folding of the protein, making it a target for degradation. Deletion of any part of the tail domain of MyoE in A. fumigatus resulted in decreased radial growth, yet not to the extent of the ΔmyoE strain, indicating a prominent role for the tail domain in the regulation of apical growth. Importantly, truncations in the tail domain also impacted the spatial distribution of MyoE in the hyphal compartment. While the septal localization of MyoE was completely abolished in all the tail truncations, the myoE^Δ1084, myoE^Δ1160 and myoE^pOtef-Δ1249 strains showed more polarization and diffuse pattern toward the hyphal tips.

It is well established that the type V myosin tail is involved in receptor binding to enable cargo delivery, and this is facilitated through actin-mediated transport. A recent study has indicated that head-to-tail regulation is critical for motor activity of myosin V in S. cerevisiae (Donovan and Bretscher, 2015). Furthermore, vertebrate class V myosin can bind to microtubules and its interaction with kinesin has been shown to increase the processing of each motor (Ali et al., 2008; Ross et al., 2008). Additionally, A. nidulans MyoE has been shown to move bi-directionally, indicating that MyoE moves retrograde on a minus-end microtubule motor, such as dynein (Taheri-Talesh et al., 2012). Based on these studies and as indicated in our model (Fig. 5), we speculate that the absence of tail domain results in loss of MyoE motor regulation, recycling and/or interaction with retrograde motors leading to its mislocalization from the hyphal septum and hyperpolarization at the hyphal tip. The hyperpolarization of the MyoE truncations toward the hyphal tip further suggests that the head domain is bound to actin at the tip but remains inactive, resulting in a phenotype that mimicks the myoE deletion. Mislocalization of MyoE from the hyphal tip following treatment with the actin depolymerizing agent cytochalasin A confirmed the association of MyoE with actin filaments.

As for the impact of MyoE truncations on the hyphal morphology, smaller apical (myoE^pOtef-Δ1249 mutant) and sub-apical septal compartments (myoE^Δ1084, myoE^Δ1160, myoE^pOtef-Δ1249 mutants) were also observed when compared to the wild-type strain and, as previously shown, the ΔmyoE strain (Renshaw et al., 2016), suggesting that the MyoE tail domain is required for the maintenance of proper septation frequency. However, because the hyphal compartments in the tail truncated strains were larger than the ΔmyoE strain's compartments, it is possible that other factors also contribute to septation. Further lending support to the importance of MyoE regarding septation is the loss of septal localization following truncation of the tail domain. We speculate that MyoE is involved in the organization of early septation processes to mark the sites of future septation. Given that septation is tightly linked to branching, it is reasonable to presume that, with smaller compartments and, thus, more septa, we see increased branching in MyoE truncated strains that have abnormal septal frequency. Intriguingly, we also found that MyoE localization at the septa is independent of both actin and microtubules, as localization remained unaltered following exposure to cytochalasin A or benomyl. However, as pre-grown hyphal cultures were treated with the actin or microtubule inhibitors, we cannot rule out the possibility that the cytoskeletal components are required per se for translocation of MyoE from the hyphal tip to the septum.

Although the phosphorylation of type V myosin, Myo2, in S. cerevisiae did not have any impact on cell growth (Legesse-Miller et al., 2006), the phosphorylation of type V myosin tail in vertebrates has been implicated in its function to modulate cargo binding (Karcher et al., 2001). Hence, we investigated the relevance of post-translational modifications for the function of A. fumigatus MyoE, which revealed that MyoE is phosphorylated and acetylated in vivo. We then generated nonphosphorylatable or nonacetylatable strains to determine the importance of these sites but found no significant phenotypic differences in residue-substituted strains. This suggests that no single phosphorylation site is important for MyoE function but does not rule out the possibility that multiple phosphorylation sites work in concert to regulate MyoE. The notable aspect of MyoE localization at the hyphal septum akin to the catalytic subunit of the serine/threonine protein phosphatase calcineurin, prompted us to examine if dephosphorylation by calcineurin regulates MyoE function and septal localization. We, therefore, also determined the phosphorylation status of MyoE in the absence of the catalytic subunit of calcineurin. Mass spectrometry revealed one additional phosphorylated site in the ΔcnaA strain (either T1509 or S1516). We then generated phosphomimetic strains of each of these sites. These strains exhibited radial growth similar to that seen the wild-type strain; however, the myo^ES1516E strain lost localization at the septa and has larger apical compartments than the wild-type strain (Fig. 4F), demonstrating the importance of phosphorylation/dephoshorylation of this specific residue on MyoE septal localization and septation frequency. Given the findings in our current study that show that MyoE localization at the septa is independent of both actin and microtubule networks, and is required for septation frequency, we postulate that MyoE interacts with yet-unknown protein/s at septa that may be necessary to maintain apical septal frequency and the size of the apical compartment. Our future work aims to understand this unknown process. MyoE localized properly at the septa and hyphal tip in the ΔcnaA strain, indicating that calcineurin does not regulate dephosphorylation at S1516.

In conclusion, we have demonstrated the importance of the A. fumigatus MyoE tail domain in the integral processes of hyphal growth and septation. Furthermore, we have shown that phosphorylation plays an important role in localization of MyoE, and that its localization at the septum is microtubule- and actin-independent. Future studies directed towards understanding how MyoE remains stably at the septa and determining MyoE protein–protein interactions might reveal the precise role for MyoE in the regulation septation.

Strains used in this study are listed in Table S1. The A. fumigatus akuB^KU80 or the akuB^KU80 pyrG⁻ uracil/uridine auxotrophic strains were used for strain mutations and the A. fumigatus akuB^KU80::myoE-gfp was used as the wild-type strain (Renshaw et al., 2016). Cultures were grown on glucose minimal medium (GMM) as previously described (Shimizu and Keller, 2001) at 37°C, except where otherwise specified. Escherichia coli DH5α-competent cells were used for cloning.

For MyoE truncation and phosphorylation mutant strain generation, the truncated or mutated cDNAs were amplified and cloned into pUCGH-myoEterm (Renshaw et al., 2016). For the heterologous expression of the S. cerevisiae class V myosin tail, 1 kb upstream of the A. fumigatus myoE tail and the codon-optimized C-terminal 1458 bp of myo4 was cloned into pUC57 (GenScript). The plasmid was linearized with KpnI and BamHI and cloned into pUCGH-myoEterm. The resulting plasmids were linearized with KpnI and XbaI, transformed into the akuB^KU80 strain, and transformants were selected by growth in the presence of hygromycin B. Mutant strains were confirmed by sequencing and western blotting. Protein extracts for western blotting were prepared as previously described (Juvvadi et al., 2011). Approximately 300 µg of protein were resolved by electrophoresis through a 4–20% SDS-polyacrylamide gel (Bio-Rad). Proteins were electroblotted onto a polyvinylidene difluoride membrane (Bio-Rad), stained with Ponceau S for 5 min, and probed with polyclonal rabbit anti-GFP (GenScript, cat. no. A01704) at 1:200 and anti-β-tubulin (GenScript, cat. no. A01203) at 1:2000 as primary antibodies, and rabbit anti-IgG as secondary antibody (1:2000). SuperSignal West Pico chemiluminescent substrate (Thermo Scientific) was used for detection.

Conidia (10⁴) were inoculated on GMM agar, incubated at 37°C, and radial growth was measured every 24 h for 5 days. For conidial quantification, conidia (10⁴) were inoculated onto GMM agar, incubated for 5 days at 37°C, then harvested in 10 ml 0.05% Tween-80 and quantified using a hemacytometer as previously described (Lamoth et al., 2012). All assays were performed in triplicate, and the mean growth rates were compared statistically by Student's t-test using GraphPad Prism (San Diego, CA).

For fluorescence microscopy, 10⁴ conidia were inoculated into 5 ml GMM on sterile coverslips (22×40 mm; no. 1.5) and incubated for 18 h at 37°C. An Axioscop 2 plus microscope (Zeiss) equipped with AxioVision 4.6 imaging software was used to observe strains.

Conidia (10⁴) were cultured on coverslips immersed in 5 ml of GMM broth and incubated at 37°C for 18 h (Juvvadi et al., 2011). For Aniline Blue staining, coverslips were rinsed with GMM, stained with 500 µl of Aniline Blue, and incubated at 25°C for 5 min. Coverslips were rinsed with GMM and observed by fluorescence microscopy.

To measure the interseptal distances, conidia were inoculated onto coverslips and stained with Aniline Blue, as previously described (Juvvadi et al., 2011). The apical compartment was measured as the distance from the hyphal apex to the apical septum. The sub-apical compartment was measured as the distance from the apical septum to the next visible septum. Twenty apical and sub-apical compartments were measured per strain. Statistical analysis was performed using Student's t-test comparing the apical or sub-apical compartments of the wild-type strains to those of the mutant strains.

For live cell imaging experiments, 10⁴ conidia were inoculated into a 35 mm glass-bottom Petri dish (MaTek) in a total volume of 3 ml GMM broth and incubated at 37°C for 8 h. The dish was then shifted to a Zeiss Axio Observer Zi motorized live cell station equipped with a Pecon XL s1 with temperature controlled at 37°C. Differential interference contrast (DIC) and GFP fluorescent images were captured at 10 min intervals from +8 h after inoculation over a period of 18 h by using the 63×/1.4 oil Plan Apochromat DIC objection and GFP filtercube HE38 with a Coolsnap ES2 high-resolution CCD camera. Images were collected by using MetaMorph 7.6.5 software.

To determine the effect of cell wall and other stress conditions on the mutant strains, 10⁴ conidia were inoculated onto GMM agar supplemented with caspofungin (1 or 4 µg/ml), nikkomycin Z (2 µg/ml), EGTA (4 mM), or FK506 (100 ng/ml). Growth was visualized after 3 and 5 days of incubation at 37°C. To assess the effect of anti-actin and anti-microtubule agents on the mutant strains, 10⁴ conidia were inoculated into 5 ml GMM on sterile coverslips (22×40 mm; No. 1.5) and incubated for 17 h at 37°C. Then the medium was removed and either cytochalasin A (8 µg/ml) or benomyl (100 µg/ml) in 5 ml GMM was added and incubated for 1 h. Strains were observed for MyoE localization with an Axioskop 2 plus microscope (Zeiss).

The A. fumigatus strains expressing the myoE-egfp fusion construct under the control of the myoE native promoter (wild-type or ΔcnaA) were grown in GMM liquid for 24 h at 37°C. MyoE proteins were purified by using GFP-Trap^® affinity purification and subjected to liquid chromatography tandem mass spectrometry (LC/MS-MS) analysis as previously described (Vargas-Muniz et al., 2016).

We acknowledge Yasheng Gao (Duke University, Durham, NC) for his technical assistance in performing time-lapse microscopy.