Introduction

Polar extension of hyphal cells, along with branching events, are characteristic hallmarks of the growth of filamentous fungi. Some of these processes in Neurospora crassa are regulated by COT-1 (colonial temperature sensitive-1) (Perkins et al. 1982; Yarden et al. 1992). This protein is a Nuclear DBF2-related (NDR) kinase, a member of the Ser/Thr kinase family that is involved in cell division and morphology spanning unicellular eukaryotes to mammals (Hergovich et al. 2006). Prominent changes in the temperature sensitive cot-1 (ts) mutant include excessive branching as well as increased thickness of the cell wall and septa, which are evident when the strain is grown at restrictive temperatures of ≥32 °C (Yarden et al. 1992; Gorovits et al. 2000). Dysfunction of COT-1 homologues in other fungal species also leads to significant and pleiotropic morphological consequences (Durrenberger and Kronstad 1999; McNemar and Fonzi 2002; Buhr et al. 1996; Scheffer et al. 2005; Johns et al. 2006).

The fungal cell wall is essential for maintaining cell shape and integrity and its maintenance is a dynamic process that requires both deposition of new wall material and removal or rearrangement of existing linkages during growth and differentiation and in response to changing environmental conditions (Bowman and Free 2006; Osherov and Yarden 2010; Riquelme et al. 2011; Teparic and Mrsa 2013). Cell wall remodeling involves the function of biosynthetic enzymes (e.g., chitin and glucan synthases), which act in concert with enzymes exhibiting opposing activities (e.g., chitinases and glucanases). These enzymatic activities must be balanced to provide the cell wall elasticity and strength which allows new growth, branching, and morphological differentiation (Bowman and Free 2006). Based on their significance for fungal proliferation, cell wall remodeling processes are prime potential targets for antifungal pharmaceuticals (Free 2013; Valiante et al. 2015).

The main constituent (30–80 % of cell wall mass) of the fungal cell wall is β-1-3-glucan (Free 2013). The glucan synthase complex (GSC) is comprised of a catalytic subunit (FKS1) and a regulatory subunit (ROH1). In contrast to yeasts, most filamentous fungi have only one FKS1 homolog (Tentler et al. 1997; Thompson et al. 1999; Beauvais and Latge 2001). In N. crassa, FKS-1 (NCU06871) was shown to be required for proper hyphal growth and development (Tentler et al. 1997). Impairing glucan synthase function has been shown to dramatically affect fungal cell morphology and growth and cell wall integrity (Free 2013).

Chitin content in most filamentous fungi has been reported to reach up to 15 % of the cell wall (Free 2013) and most of these fungi have seven or eight chitin synthase (chs) genes (Kong et al. 2012), even though up to ten clades of chitin synthase-encoding genes have been designated across different fungal linages (Li et al. 2016). All of the seven chs genes in N. crassa (Table 1) were found to be nonessential, suggesting redundancy in chs gene function (Beth Din and Yarden 1994; Beth Din et al. 1996; Dunlap et al. 2007; Riquelme and Bartnicki-Garcia 2008; Sanchez-Leon et al. 2011; Fajardo-Somera et al. 2015). Chitinases are also involved in a variety of functions associated with fungal growth (Kuranda and Robbins 1991; Dunkler et al. 2005; Baker et al. 2009). In N. crassa, 13 genes encoding for chitinases have been identified (Tzelepis et al. 2012).

Table 1 Neurospora crassa strains used in this study
Full size table

The cell wall integrity (CWI) pathway is responsible for cell wall remodeling and reinforcement in response to cell wall stresses and is highly conserved among fungal species (Damveld et al. 2005; Fuchs and Mylonakis 2009). A key component of the CWI pathway in Saccharomyces cerevisiae, is the transcription factor Rlm1p (a MADS- box MEF2 type protein), which has been shown to be responsible for expression of numerous genes associated with the cell wall (Jung and Levin 1999). In various other fungi, Rlm1 has been shown to be involved in the regulation of expression of some cell wall-related biosynthetic enzymes (Damveld et al. 2005; Kovacs et al. 2013; Levin 2011; Delgado-Silva et al. 2014), yet the involvement of Rlm1p homologs in the CWI pathway in filamentous fungi is by far less clear than that established in S. cerevisiae.

Ssd1 is a S. cerevisiae mRNA-binding protein involved in translational regulation of cell wall remodeling proteins (Uesono et al. 1997; Hogan et al. 2008; Jansen et al. 2009; Kurischko et al. 2011; Wanless et al. 2014). This protein has been shown to associate with over 150 mRNA species, including those that encode proteins that function in cell wall organization and remodeling, cell separation and the ‘‘SUN’’ family (e.g., Sun4, Sim1, Uth1) of cell wall degrading proteins. Ssd1 can suppress translation of these cell proteins and this activity has been demonstrated to play an important role in cell integrity maintenance under stress conditions. Ssd1 function is negatively regulated by the yeast COT-1 homologue Cbk1 (Hogan et al. 2008; Jansen et al. 2009).

Similar to that found in yeasts, the N. crassa Ssd1 homologue, GUL-1, has also been shown to interact with the corresponding NDR kinase, COT-1. Suppressor analysis of cot-1 (ts) showed that gul-1 partially suppresses the pleiotropic morphology of the mutant (Terenzi and Reissig 1967). Morphological characteristics of gul-1 mutants as well as the gul-1 deletion strain include slow growth rate, reduced germination and hyperbranching (Terenzi and Reissig 1967; Seiler et al. 2006). Orthologues of Ssd1 have also been found in other filamentous fungi. In the plant pathogens Colletotrichum lagenarium and Magnaporthe grisea they were determined to be required for pathogenicity (Tanaka et al. 2007).

Despite the aberrations in cell wall morphology that accompany impaired cot-1 function, it has not yet been established to what extent cot-1 is involved in the regulation of expression of genes encoding cell wall remodeling proteins. In this study, we report on the involvement of cell wall remodeling elements in conferring the cot-1 (ts) phenotype. We propose that changes in the transcript abundance of genes encoding chitin and glucan synthases, as well as chitinases, contribute to the phenotypic changes observed. Furthermore, we show that these changes are partially mediated by gul-1 but expression of the genes analyzed is not altered in a N. crassa rlm-1 mutant.

Materials and methods

Strains, media and growth conditions

Standard procedures and media used in the handling of N. crassa were used as described by Davis (2000) or are available through the Fungal Genetics Stock Center (http://www.fgsc.net/Neurospora/NeurosporaProtocolGuide.htm). N. crassa strains used in this study are listed in Table 1. Strains were grown in either liquid or solid (supplemented with 1.5 % agar) Vogel’s minimal medium with 1.5 % (w/v) sucrose (Vs). When required, the medium was supplemented with 10 μg ml−1 hygromycin B (Duchefa Biochemie, Haarlem, The Netherlands). In general, all mutants produced throughout this study were identified on the basis of morphological characteristics and their genetic nature was confirmed by PCR using specific primers (Table 2). For growth rate measurements, 6 μL of a conidial suspension (2 × 106 conidia/mL) were inoculated on Petri dishes containing Vs and growth was measured after 24 h.

Table 2 PCR primers used in this study
Full size table

qRT-PCR

Total RNA was isolated from mycelia grown for 16 h at 25 °C/28 °C/34 °C or grown for 12 h at 25 °C followed by 4 h at 34 °C, at 150 rpm, in liquid Vs. The RNA samples were purified with the RNeasy Plant Mini Kit (Qiagen, Hilden, Germany) and DNA-free kit (Ambion). 1 μg of purified RNA was used for cDNA synthesis utilizing superscript II RNase H reverse transcriptase (Invitrogen, Carlsbad, California). Relative quantification of transcript abundance was performed on an ABI StepOnePlus Real-Time PCR sequence detection system and software (Applied Biosystems). For chs gene transcript detection, we used primers previously designed to identify unique 80–150 bp amplicons of the highly similar members of this gene family (Koch et al. 2014). PCR mixtures comprised a 10 µM concentration of each primer (Table 3), 5 μL of FAST SYBR green PCR master mix (Applied Biosystems), 2 μL of a 20 ng/μL cDNA solution and nuclease-free water to a final volume of 10 μL. The annealing temperature was 64 °C. Total cDNA abundance in the samples was normalized using the beta tubulin (tub-2) gene (Table 3). In all experiments, samples were amplified in triplicate, and the average cycle threshold was then calculated and used to determine the relative expression of each gene. Relative expression levels were calculated using the 2−ΔCt method.

Table 3 PCR primers used for transcriptional analysis
Full size table

Microscopy

Light and fluorescent microscopy was performed with an EVOS FL auto imaging system (Life Technologies). Strains were grown on slides coated with solid (supplemented with 1.5 % agarose) Vs medium at 25 °C/28 °C for 16 h or grown for 12 h at 25 °C followed by 4 h at 34 °C. To determine general chitin deposition in cell wall and septa the strains were stained with calcofluor white (10 mg/mL). To determine the length between branches we used the ImageJ 1.37 V software package (Rasband, WS, US National Institutes of Health, Bethesda, MD, http://rsb.info.nih.gov/ij/, 1997–2006).

For transmittance electron microscopy (TEM), strains were grown on solid Vogel’s minimal medium. Samples were prepared on the basis of a previously described protocol (Gorovits et al. 2000) with slight modifications. Hyphal samples (0.5 × 0.5 cm squares) were cut from the agar and fixed in 1.5 % KMnO4 (w/v in H2O) for 20 min at room temperature. The samples were rinsed three times in double distilled water (DDW), and three times for 10 min each, in cacodylate buffer and then washed once in DDW. The samples were then washed in DDW and were subjected to en bloc staining with 1 % uranyl acetate in DDW (overnight, at room temperature) to enhance contrast. Samples were washed four times for 10 min in 50 % ethanol, followed by dehydration in increasing concentrations of ethanol consisting of 70, 80, 90, 95 %, (10 min each) followed by 100 % anhydrous ethanol three times (20 min), and twice in propylene oxide (10 min). Following dehydration, the tissues were infiltrated with increasing concentrations of agar 100 resins in propylene oxide, consisting of 25, 50, 75, and 100 % resin (16 h for each step). The tissues were then embedded in fresh resin and polymerized in an oven at 60 °C for 48 h. Embedded sample blocks were sectioned with a diamond knife on an LKB 3 microtome and ultrathin Sects. (80 nm) were collected onto 200 Mesh, thin bar copper grids. The gridded sections were sequentially stained with uranyl acetate and lead citrate for 10 min each and viewed with a Tecnai 12 TEM 100 kV (Phillips, Eindhoven, The Netherlands) microscope equipped with a megaview II CCD camera and Analysis® version 3.0 software (SoftImaging System GmbH, Münster, Germany).

Cell wall carbohydrate analyses

Cell wall samples were prepared as described by Maddi and Free (2010). Briefly, cell walls were isolated from cells grown for 12 h on Vs medium at 25 °C followed by 4 h at 34 °C, at 150 rpm. Carbohydrate analyses were performed at the University of Georgia Complex Carbohydrate Research Center (Athens, GA). For glycosyl linkage analysis, the samples were permethylated, depolymerized, reduced, and acetylated, and the partially methylated alditol acetates were analyzed by gas chromatography-mass spectrometry (GC–MS), essentially as described by Heiss et al. (2009), with minor modifications to accommodate the detection of amino sugars. Approximately 1 mg of cell wall sample was used for linkage analysis. The sample was suspended in 200 μl of dimethyl sulfoxide and left to stir for 3 days. Permethylation was affected by two rounds of treatment with sodium hydroxide (15 min) and methyl iodide (45 min). Following sample workup, the permethylated material was hydrolyzed using 2 M trifluoroacetic acid (2 h in sealed tube at 121 °C), reduced with NaBD4, and acetylated using acetic anhydride/pyridine. The resulting partially methylated alditol acetates were analyzed on an Agilent 7890A GC interfaced to a 5975C MSD (mass selective detector, electron impact ionization mode); separation was performed on a 30 m Supelco SP-2331 bonded phase fused silica capillary column.

Results

Transcript levels of chs family members and fks-1 are increased in a cot-1 (ts) background

To analyze the possible association between cot-1 and genes involved in cell wall biosynthesis, we first examined the transcript levels of chitin and glucan synthases (chs-1–7 and fks-1, respectively) in a cot-1 (ts) background. The transcript abundance of the chs genes and fks-1 was assayed in RNA isolated from hyphae of wild type and cot-1 (ts) strains grown at 25 °C, as well as following a shift to the cot-1 (ts) restrictive temperature (34 °C). On the basis of the differential expression pattern of chs-1–7 in the wild type control, we determined that the N. crassa chs genes can be divided into two major groups, based on their transcript abundance: Highly expressed (chs-1, 3, 4, 5, 7) and those expressed at lower levels (chs-2, 6). When cot-1 (ts) was grown at 25 °C, no distinct differences were observed between the expression patterns of the chs and fks-1 genes, when compared to the wild type (Fig. 1). However, at restrictive temperature, a statistically significant increase in the expression of the highly expressed group of chs genes (chs-1, 2, 3, 4), as well of fks-1, was observed in a cot-1 (ts) background. The most pronounced change was observed in expression of chs-1, where transcript abundance was almost threefold of that measured in the control. Overall, the increases observed in the expression of the cell wall biosynthetic enzyme-encoding genes indicate that they are involved in some of the morphological cell wall defects characteristic of cot-1 (ts).

Fig. 1
figure 1

Expression profiles of chs genes (a, b) and fks-1 ( c) in wild type, cot-1 (ts) Δgul-1 and Δgul-1;cot-1 (ts) mutants grown at permissive temperature (a) or following a shift to cot-1 (ts) restrictive temperature (b), and fks-1 (c) in the different mutants as determined by qRT-PCR. RNA was purified from conidia germinated in liquid Vs medium for 16 h at 25 °C or for 12 h at 25 °C followed by 4 h at 34 °C. The relative expression level of individual genes was analyzed by the 2−ΔCt method with the β-tubulin gene as the internal control for normalization. Bars indicate ±1.96SE (confidence level of 95 %), determined with data obtained from three independent replicates

Full size image

Deletion of chs-5 or chs-7 result in a reduction in cot-1 (ts) hyperbranching

The fact that deletion/inactivation of some (but not all) single chs-encoding genes in N. crassa results in various changes in hyphal growth rates has been previously described (Beth Din and Yarden, 1994; Beth Din et al. 1996; Sanchez-Leon et al. 2011; Fajardo-Somera et al. 2015). To verify the differences in growth rates of the different mutants under the conditions performed in this study, we first carried out comparative growth tests in which all chs deletion strains were examined in parallel. Based on our results, we determined that Δchs-2 and Δchs-4 grew in a manner similar to the wild type. In contrast, Δchs-1, Δchs-3 and Δchs-7 exhibited slower growth (80 % of wild type), while Δchs-5 and Δchs-6 grew at only 35 % of the rate measured in the wild type. Excluding the case of the slow growing Δchs-5, results obtained were in line with those recently described (Fajardo-Somera et al. 2015).

To determine the possible genetic interactions between the genes encoding CHSs and COT-1, we produced seven chs;cot-1 (ts) double mutants. To follow the possible effects of chs gene deletion in a cot-1 (ts) background, we cultured the double mutants at 28 °C, a semi-restrictive temperature, which allows measurable growth of cot-1 (ts) and detecting even subtle morphological changes. Radial growth of most of the double mutants was comparable to that of cot-1 (ts). The exceptions were Δchs-1;cot-1 (ts), Δchs-5;cot-1 (ts) and Δchs-6;cot-1 (ts), which grew at slower rates, probably due to the additive effects of the mutations in cot-1 (ts) and the slower growing Δchs-1, Δchs-5 and Δchs-6 strains. Nonetheless, even though growth rates were different, no visible differences in colony morphology or in hyphal structures was observed between the double mutants and cot-1 (ts).

When cultures were shifted from 28 to 34 °C, all double mutants exhibited a morphology similar to cot-1, excluding Δchs-5;cot-1 (ts) and Δchs-7;cot-1 (ts). These exhibited a less severe colonial phenotype, which was characterized by a reduction in hyphal density and reduced branching frequencies of the colony (Fig. 2). Thus, while the distance between branches in the Δchs-5 and Δchs-7 strains was comparable to the wild type (~160 µM), and was only 10 µM in cot-1 (ts), the distance in the double mutants ranged 30–40 µM. Swelling of some of the hyphal cells was also observed. In all strains, calcofluor white staining patterns were similar, suggesting no drastic changes in chitin deposition had occurred.

Fig. 2
figure 2

Deletion of chs-5 or chs-7 results in partial suppression of the cot-1 (ts) phenotype. Cultures were grown on slides coated with Vs medium for 12 h at 28 °C followed by 4 h at 34 °C. Bar 200 µm

Full size image

Based on these results, we concluded that at least two chitin synthases (chs-5 and chs-7) have a genetic interaction with cot-1 (ts), as determined on the basis of the colony morphology observed.

Two chitinases (chit-1, gh18-5) are expressed in opposing manners in cot-1 (ts)

chit-1 and gh18-5 transcripts were analyzed due to the suggested involvement of the corresponding proteins in the CWI pathway (Seidl 2008; Kovacs et al. 2013; Yamazaki et al. 2008). When grown at 25 °C, RNA levels of both chit-1 and gh18-5 in the cot-1 (ts) mutant and in the wild type strain were comparable (Fig. 3). However, when the wild type was subjected to a 25–34 °C temperature shift, chit-1 expression decreased significantly (by ~60 %). A more marked decrease was observed in cot-1 (ts), where chit-1 expression was less than 20 % of that measured prior to the shift. The expression pattern of gh18-5 in both the wild type and cot-1 (ts) was different than that of chit-1. Transcript levels were not altered in the wild type following the temperature shift. However, in the cot-1 (ts) mutant, following the temperature shift, gh18-5 transcript abundance was nearly tenfold higher.

Fig. 3
figure 3

Expression profiles of chit-1 (a) and gh18-5 (b) in wild type, Δgul-1, cot-1 (ts) and Δgul-1;cot-1 (ts) strains as determined by qRT-PCR. RNA was purified from conidia germinated in liquid Vs medium for 16 h at 25 °C or for 12 h at 25 °C followed by 4 h at 34 °C. The relative expression level of individual gene was analyzed by the 2−ΔCt method with the β-tubulin gene as the internal control for normalization. Bars indicate ±1.96SE (confidence level of 95 %), determined with data from three independent replicates

Full size image

These results show that the two chitinase-encoding genes exhibit opposing expression patterns in a cot-1 (ts) background, and that cot-1 inactivation emphasizes the different roles these genes may play in cell wall remodeling.

Deletion of chit-1 in a cot-1 (ts) background results in a significant reduction in growth rate

As chitinases are components of the cell wall maintenance and remodeling machinery, we first examined the consequences of chit-1 and gh18-5 deletion. Interestingly, we found that the linear growth rates of Δchit-1 and Δgh18-5 did not differ from that of the wild type strain. To determine whether chit-1 and gh18-5 genetically interact with cot-1 (ts), double mutants of the chitinases and cot-1 (ts) were produced. Δchit-1;cot-1 (ts) exhibited a significant reduction in growth rate when compared to the parental strains at the different growth conditions [80 and 70 % of Δchit-1 and cot-1 (ts), respectively]. However, the morphology of the mutant was similar to cot-1 (ts). In contrast to Δchit-1;cot-1 (ts), Δgh18-5;cot-1 (ts) exhibited a similar growth rate to the parental strains at the different growth conditions [e.g., similar to Δgh18-5 at 25 °C and to cot-1 (ts) at 34 °C].

These results demonstrate that under the conditions tested, chit-1, but not gh18-5, genetically interact with cot-1.

Inactivation of cot-1 does not confer changes in rlm-1 expression

The A. nidulans Rlm1A protein has been shown to be involved in stress-related CWI pathway regulation (Kovacs et al. 2013). The most likely homologue (51 % amino acid sequence identity) of this gene in N. crassa (NCU02558) has not yet been functionally analyzed. To investigate the possible involvement of rlm-1 in changes in cell wall remodeling in cot-1 (ts), we analyzed this mutant and studied the expression of the relevant genes in the Δrlm-1 strain.

The N. crassa Δrlm-1 strain grew at a rate that was only 5 % slower than the wild type, suggesting that the requirement for a functional rlm-1 is marginal for proper growth and development of N. crassa under standard conditions.

As cot-1 has been shown to exhibit characteristics of stress-related conditions (Gorovits and Yarden 2003), we examined rlm-1 expression in cot-1 (ts). The expression level of rlm-1 was only slightly increased following a 25–34 °C shift in cot-1 (ts), when compared to the wild type.

We further examined the possible relevance of rlm-1 for cell wall remodeling and measured the transcript abundance of chs, fks-1 and chitinase genes in the Δrlm-1 mutant. In all cases, the expression levels of chs-1–7 and fks-1 in the mutant were similar to those measured in the wild type strain. In addition, no changes in the expression of chit-1 and gh18-5 were observed in the mutant. We concluded that the cellular stresses imposed by impairment of COT-1 do not activate the CWI pathway via rlm-1.

A mutation in gul-1 reduces the severity of the cot-1 (ts) phenotype and is accompanied by altered cell wall remodeling gene expression

Previous studies have shown that the growth defects of cot-1 (ts) are partially suppressed in a gul-1 background, suggesting a functional link between GUL-1 and COT-1 (Terenzi and Reissig 1967; Seiler et al. 2006). To determine the possible role of gul-1 as a mediator between cot-1 (ts) and genes involved in cell wall remodeling, we examined the transcript levels of chs-1–7 and fks-1 in a Δgul-1;cot-1 (ts) mutant. At 25 °C, the expression levels of the cell wall biosynthetic genes were similar to those measured in the parental strains and the wild type (Fig. 1). Following a 25–34 °C temperature shift, chs-1, 2, 3, 4, as well as fks-1 expression, significantly increased in the cot-1 (ts) mutant, as described above (Fig. 1). However, their expression levels in the Δgul-1;cot-1 (ts) and the parental strains were similar to the wild type (Fig. 1). These results show that the increases in cell wall biosynthetic gene expression patterns in a cot-1 (ts) background are dependent on a functional gul-1.

We also examined the transcript levels of chit-1 and gh18-5 in the Δgul-1 and Δgul-1;cot-1 (ts) mutants. When Δgul-1;cot-1 (ts) was grown at 25 °C, no distinct differences were observed between the expression patterns of the chit-1 gene when compared to cot-1 (ts) and the wild type (Fig. 3). However, the expression of chit-1 in a Δgul-1 mutant was only about 30 % of that measured in the wild type. When a 25–34 °C temperature shift was imposed, chit-1 expression decreased significantly in the wild type, yet this gene’s expression was not altered in Δgul-1. Deletion of gul-1 in a cot-1 background did not alter the already reduced expression of chit-1 previously measured in the cot-1 (ts) parent (at 34 °C). Thus, the decrease in the expression of chit-1 in a cot-1 (ts) background is independent of the presence of gul-1.

The expression level of gh18-5 in Δgul-1;cot-1 (ts) was similar to the parental strains and the wild type at 25 °C. However, in contrast to that observed in cot-1 (ts), following a shift to restrictive temperature, the change in transcript abundance of gh18-5 in the Δgul-1;cot-1 (ts) strain was negligible, when compared to the several fold increase observed in the cot-1 (ts) parent (Fig. 3). Taken together, our results suggest that GUL-1 is most likely a functional link between COT-1 and the N. crassa cell wall remodeling proteins CHS-1, 2, 3, 4, as well as FKS-1 and GH18-5.

A mutation in gul-1 suppresses the effect of impaired COT-1 function on the cell wall

To further investigate the contributions of gul-1 in diminishing the effect of the cot-1 (ts) mutation on the cell wall, hyphae were examined using transmission electron microscopy. Hyphal cell walls of the cot-1 (ts) strain were thicker and the septa showed a multilayered structure when compared with the wild type [as was previously described by (Gorovits et al. 2000)]. In the Δgul-1 mutant, the cell wall appeared to be thinner than the wild type, whereas the Δgul-1;cot-1 (ts) double mutant exhibited septa and cell wall morphology which highly resemble that of the wild type (Fig. 4). These observations were supported, at least in part, by carbohydrate cell wall composition analysis (Table 4). The major change observed was that of chitin content, which was increased by almost 40 % in the cot-1 (ts) strain when compared to the wild type. Chitin content was reduced by 10 % in the Δgul-1 strain, but close to wild type levels in the Δgul-1;cot-1 (ts) double mutant. Interestingly, the changes in glucan (the major cell wall carbohydrate constituent) were much lower, while at the same time the deletion of gul-1 resulted in a 20 % reduction in mannan. The Δgul-1 mutant was also found to be more sensitive to a chitin synthase inhibitor (growth rate of ~40 % of the wild type in the presence of 10−5 M Nikkomycin Z). Based on these results, we suggest that GUL-1 contributes to changes in cell wall morphology and composition in a manner that confers the suppressive effect on cot-1.

Fig. 4
figure 4

Transmission electron micrographs of hyphal morphology of wild type, cot-1, Δgul-1 and Δgul-1;cot-1 strains. Cultures were grown on Vs medium for 12 h at 25 °C followed by 4 h at 34 °C. Bar 1000 nm

Full size image
Table 4 Carbohydrate content in the cot-1 (ts), Δgul-1 and Δgul-1;cot-1 (ts) mutants (percent, relative to wild type)
Full size table

Discussion

In this study, we examined the relationship between a key component controlling the polar development of hyphae in the fungus N. crassa and the downstream elements that contribute to determination of cell wall structure.

The Ndr kinase COT-1 is important for cell differentiation and polar morphogenesis in various fungi. One of the outcomes of impairing cot-1 function is the presence of thick hyphal cell walls and septa (Gorovits et al. 1999, 2000; Yarden et al. 1992). This implies the occurrence of changes in the function of the cell wall biosynthetic machinery, predominantly chitin and glucan synthases. Glucan is the major constituent of the fungal cell wall (Maddi et al. 2009; Free 2013). However, even though chitin comprises only a small amount of the fungal cell wall, subtle changes in its concentration can have a large impact on fungal morphology (Hall 2015).

To analyze the possible association between cot-1 (ts) and genes involved in cell wall biosynthesis, we first studied the expression level of the chitin synthases gene family (chs-1–7) in N. crassa. On the basis of the differential expression pattern of these genes, we determined that chs genes can be divided into two major groups: highly expressed: chs-1, 3, 4, 5, 7 (class III, I, IV, V, VII, respectively) and those expressed at lower levels: chs-2, 6 (class II and class VI, respectively). Even though this pattern was repeatedly observed, it is still unclear if this division has functional relevance with regard to cell wall synthesis. In M. oryzae, CHS1-CHS7 designation was based on their orthologs in N. crassa. Nevertheless, the expression patterns of these genes are not identical in both fungi (Kong et al. 2012). Hence, it is likely that similar class designation does not necessarily imply identical expression/function.

Examining the transcription level of chitin synthases and glucan synthase genes showed that there is an increase of 20–250 % in the expression of these genes in a cot-1 (ts) background. These results show that changes of those elements may contribute to the phenotype of cot-1 with regard to the abnormally thick cell wall and septa. To what extent the variation in changes of expression of the different chs genes is due to their functional redundancies, has yet to be determined. This is further emphasized by the analysis of the chs;cot-1 (ts) double mutants (see below).

In S. cerevisiae, deletion of CBK1 alters the pattern of transcription of genes involved in cell wall morphology and those that are involved in digestion of the chitinous septum between mother and daughter cells. Our results were similar to those described by Bidlingmaier et al. (2001), who showed that a chitin synthase encoded by CHS1 was significantly elevated in cbk1D cells. Thus, our results along with evidence obtained in S. cerevisiae and C. albicans (McNemar and Fonzi 2002) suggest that it is possible that the polarized growth defects observed in cot-1 (ts) are due, at least in part, by misregulation of the expression of genes important for cell wall remodeling-related processes.

Characterization of seven chitin synthases deletion mutants in a cot-1 (ts) background revealed that deletion of single chs genes did not result in improved radial growth of cot-1 (ts), probably due to the redundancy of gene family members. chs gene redundancy has been clearly demonstrated in A. nidulans, where chsA (class II) and chsD (class IV) have highly similar functions in conidia formation (Motoyama et al. 1996). Two additional chitin synthases, CsmA (class V) and CsmB (class VII), carry out overlapping functions that are required for hyphal tip growth (Takeshita et al. 2006). Several chs gene double mutants in N. crassa also do not exhibit severe phenotypes, further supporting the partially redundant nature of some of the chs family members (Herold and Yarden, unpublished).

Microscopic analysis of Δchs-5;cot-1 (ts) and Δchs-7;cot-1 (ts) hyphae showed that these mutants exhibited a reduction in branching when compared to cot-1 (ts), an indication for the presence of genetic and functional interactions between them. The N. crassa genes encoding for CHS5 and CHS7 belong to class V and VII chitin synthases, respectively. Both of these classes have a conserved myosin motor-like domain (MMD) at their amino termini (Fujiwara et al. 1997; Riquelme and Bartnicki-Garcia 2008). In A. nidulans, CsmA (class V) has been localized near actin structures at the hyphal tips and at forming septa, and it was shown that the MMD was, in fact, capable of binding to actin (Takeshita et al. 2006). In A. fumigatus, deletion of the CSMA and CSMB genes (both harboring an MMD) induced significant disorganization of the cell wall structure, even though no modifications of chitin content in their conidial cell walls were reported (Jimenez-Ortigosa et al. 2012). Based on changes in actin patch deposition patterns in the cot-1 (ts) and cot-1 (T589A) strains, Ziv et al. (2009) suggested a functional link between COT1 and some of the components regulating the actin cytoskeleton and factors involved in establishing and/or maintaining cell wall integrity. The fact that the two chs genes whose deletion conferred a phenotypic change in cot-1 (ts) encode a protein that harbors an MMD, supports this suggestion.

To address the question whether the cot-1 (ts) phenotype may also be an outcome, at least in part, of impaired regulation of cell wall degrading enzymes, we analyzed the changes in expression of two chitinase encoding genes—gh18-5 and chit-1. The N. crassa GH18-5 belongs to a highly conserved chitinase A group, whose orthologs are present in many ascomycetes (Karlsson and Stenlid 2008). The N. crassa chit-1 gene belongs to the B1 group and its expression was found to be induced by colloidal chitin (Tzelepis et al. 2012). Several reports have suggested that subgroup B chitinases could be responsible for cell wall remodeling in fungi (Adams 2004; Jaques et al. 2003; Hurtado-Guerrero and van Aalten 2007). chit-1 influences growth rate of N. crassa, presumably through a cell wall plasticizing role (Tzelepis et al. 2012). The A. nidulans ortholog ChiA is a GPI-anchored chitinase that is localized at hyphal branching sites and at hyphal tips in conidial germ tubes (Yamazaki et al. 2008). In this study, we found that the expression of both chit-1 and gh18-5 was altered in a cot-1 (ts) background. These changes could well affect hyphal extension and branching in a manner that contributes to the cot-1 (ts) phenotype. These results are comparable to the reported in S. cerevisiae, where the cell separation defects in the Δcbk1 strain were preceded by a drastic reduction in the expression of CTS1 (a chit-1 homologue) (Racki et al. 2000; Bidlingmaier et al. 2001). In Pneumocystis carinii, CBK1 expression stimulated transcription of the CTS1 chitinase which is necessary for cell wall separation in that fungus (Kottom and Limper 2004).

Both Δchit-1 and Δgh18-5 strains were morphologically similar to the wild type. These results are partially in line with a previous study (Tzelepis et al. 2012), where Δgh18-5 was shown to be similar to the wild type. However, under the growth conditions described in that report, Δchit-1 was shown to exhibit several phenotypic defects. The severe phenotype observed in the Δchit-1;cot-1 (ts) strain established the presence of a genetic interaction between the two genes. On the basis of the expression data we further concluded that a functional cot-1 (ts) is required for proper chit-1 expression and that impaired expression of chit-1 confers developmental defects. No genetic interaction was observed between gh18-5 and cot-1 (ts) and the additive effects observed indicate that these elements probably operate in different pathways.

Mechanistic analysis of NDR protein function and identifying proteins that interact with them has been studied in several systems (Hanks and Hunter 1995; Hergovich et al. 2006; Pearce et al. 2010). However, information regarding downstream elements affected by the NDR complex is limited. In this study, we have shown, for the first time, the involvement of cell wall remodeling elements in the cot-1 (ts) phenotype. We propose that this may well be due to changes in the transcription of some genes encoding chitin and glucan synthases and chitinases we have observed.

The fact that cot-1 (ts) undergoes phenotypic suppression when a wide variety of stresses are imposed indicates that when cot-1 function is impaired, cellular stress responses are activated. One explanation for the environmental suppression that occurs under such circumstances is that the balance between the external stresses and the preactivated stress response in the cot-1 mutant enables the development and hyphal extension of the fungus (Gorovits and Yarden 2003). Some of these stresses are known to activate the CWI pathway.

Is rlm-1 involved in mediating CWI element expression in N. crassa? In S. cerevisiae and C. albicans, RLM1 has been shown to play key roles in CWI (Dodou and Treisman 1997; Watanabe et al. 1997; Delgado-Silva et al. 2014). In filamentous fungi, information concerning the involvement of rlm-1 in cell wall remodeling is less clear and in some cases even contradicting (Damveld et al. 2005; Fujioka et al. 2007, Kovacs et al. 2013). We have found that inactivating rlm-1 results in only negligible effects on growth and development of N. crassa. Analyzing the expression of rlm-1 in the cot-1 (ts) mutant demonstrated that rlm-1 is most likely not involved in conferring the cot-1 phenotype. Furthermore, deletion of rlm-1 did not result in changes in chs, fks-1 as well as gh18-5 and chit-1 expression. This is similar to that found in A. nidulans, where most cell wall-related biosynthetic genes were regulated independently of RlmA, with the exception of 1,3-α glucan synthase (Fujioka et al. 2007). However, Kovacs et al. (2013) reported that expression of some of the A. nidulans cell wall hydrolase-encoding genes is, in fact, regulated by RlmA. Overall, results obtained here demonstrate that in spite of the structural conservation of RLM-1, its function in N. crassa is different from that observed in several other fungi.

We also analyzed the significance of gul-1 in conferring the pleiotropic phenotype of cot-1 (ts). We found that the expression of several cell wall biosynthetic enzyme-encoding genes (chs-1, chs-2, chs-3, chs-4 and fks-1) was found to be regulated in a gul-1 dependent manner. When expression levels of chs-1–7 and fks-1 were examined in a gul-1;cot-1 (ts) background, the increase in their expression, as was determined to occur in cot-1 (ts), was significantly curbed. These results further support the hypothesis that miss-expression of the cell wall remodeling machinery contributes to the severity of the cot-1 (ts) phenotype and suggest that gul-1 is a functional link between cot-1 and the cell wall biosynthetic elements in N. crassa.

In yeast, loss of Cbk1 regulation of Ssd1 causes lethal defects in cell wall organization, likely by making the cell wall too rigid to expand with the growing bud (Jansen et al. 2009). To determine whether gul-1 is also a mediator between cot-1 and cell wall degrading elements in a filamentous fungus, we quantified the mRNA levels of gh18-5 and chit-1 in the relevant backgrounds. The increase in the expression of gh18-5 was found to be dependent on a functional gul-1. The decrease in the expression of chit-1 (in a cot-1 background) was found to be independent of the presence of gul-1. Thus, gul-1 is probably not a functional link between cot-1 and chit-1 and gul-1 and cot-1 may affect chit-1 in different, independent, pathways.

Taken together, our results indicate that the presence of gul-1 is required for the increase in the mRNA levels of cell wall remodeling genes in a cot-1 (ts) background. This increase can partially contribute to the morphological defects of cot-1 (ts), and in particular, to the abnormal thick cell wall and septa. Similar results suggest that one of the functions of Cbk1 in hyphal morphogenesis of C. albicans is down-regulation of the transcriptional repressor Nrg1 (a DNA-binding protein involved in filamentous growth) via the gul-1 homologue Ssd1 (Lee et al. 2015). If GUL-1 functions as a translational repressor in N. crassa (as it does in yeast) and is hyperactive when COT-1 is nonfunctional, one expected cellular response could be an increase in some of the cell wall remodeling gene transcription rates. This increase could stoichiometrically overcome the natural transcript-GUL-1 ratio and result in increased accumulation of cell wall carbohydrates.

Some of the results obtained concerning changes in transcript abundance are supported by details of the phenotypic suppression of the cot-1 (ts) phenotype by deletion of gul-1. The Δgul-1;cot-1 (ts) double mutant produce septa and cell wall morphology which highly resemble that of the wild type, in marked contrast to the cot-1 (ts) strain. The morphological observations are further supported by carbohydrate cell wall composition analysis showing the curbing of excess cot-1 (ts) chitin content in a Δgul-1 background (as well as by the hyper sensitivity of Δgul-1 to Nikkomycin Z). Evidence for changes in sensitivity to chitin synthesis inhibitors in gul-1 homologue mutants has also been reported in C. albicans (Song et al. 2008) and C. lagenarium (Tanaka et al. 2007).

Surprisingly, the changes in glucan were much lower. Changes in total glucan content may not reflect deviation in the proportions of β-1,3, mixed β-1,3/1,4-glucan and α-1,3-glucan linkages (which were not obtained in the current analysis). These may differentially contribute to cell wall architecture on the basis of arrangement (rather than quantity of the monomeric constituent).

We have demonstrated the presence of a mechanistic link between COT-1, GUL-1 and cell wall remodeling in N. crassa. Similar interactions are likely to be present in other filamentous fungi. Additional GUL-1-interacting partners have yet to be identified and in what manner the COT-1-GUL-1 pathway may also be involved in cell wall remodeling gene mRNA localization and temporal translation has yet to be determined.