Strain and plasmid construction.

Strains constructed for this study are listed in Table 1. To obtain Δlfd-2 strains expressing cytoplasmic markers, the his-3 A strain (FGSC 6103) was used as the female in crosses with Δlfd-2 a. Progeny carrying the deletion mutation and the his-3 marker were used for transformation, with either pMF272 (30, 31) or the modified version of pMF272 containing mCherry instead of green fluorescent protein (GFP), as described in Palma-Guerrero et al. (22). Macroconidia were transformed by electroporation as described by Margolin et al. (32).

To obtain double mutants with ΔPrm1, crosses were performed with a ΔPrm1 strain complemented at the his-3 locus with ccg1-Prm1-gfp (A24) (Table 1). ΔPrm1 mutants are sterile both as a female and as a male due to dominant postfertilization defects in the ΔPrm1 mutant (21). The Δfig1 mutant is female sterile (33); therefore, it was used as a male in crosses.

For Δlfd-2 complementation and GFP tagging of LFD-2 for localization studies, the open reading frame (ORF) of NCU02191 was amplified by PCR, as annotated by the N. crassa database (http://www.broadinstitute.org/annotation/genome/neurospora/MultiHome.html), using genomic DNA with primers that incorporated XbaI/PacI sites. Due to the low expression levels of lfd-2 in germlings and hyphae, we constructed an lfd-2 allele under the regulation of the tef1 promoter by replacing Prm1 with lfd-2 in the tef1-Prm1-GFP plasmid (provided by A. Fleissner, Technische Universitat Braunschweig, Germany) using the XbaI/PacI sites.

The plasmid containing sec61-mCherry under the ccg-1 promoter was constructed by introducing the ORF of sec61 (NCU08897) into the modified version of pMF272 containing mCherry instead of GFP, as described in Palma-Guerrero et al. (22), using the XbaI/PacI sites.

A strain expressing the mCherry-tagged endoplasmic reticulum (ER) protein ERV25 (NCU01342) was generated by targeting a constitutively expressed version of ERV25-mCherry-6× His to the CSR1 locus of a mutant strain lacking the erv25 open reading frame. To achieve this, the pCSR1 vector (34) was modified to contain promoter and terminator elements, as well as an expanded multiple cloning site, into which mCherry-6× His and erv25 fragments were sequentially cloned. Primer pair OTS336/OTS337 (AAACTGCAGAGCGCATTTCCGACGTTAAG and AAAGCGGCCGCTTTGATT TCTGTGATGTG), incorporating PstI and NotI restriction sites, respectively, was used to amplify a 1-kb promoter fragment immediately upstream of the predicted open reading frame of the gene encoding glyceraldehyde 3-phosphate dehydrogenase (GPD) from Myceliophthora thermophila genomic DNA (nucleotides 1,538,499 to 1,539,504 of chromosome 7). Insertion of this promoter fragment into the pCSR1 vector resulted in plasmid pCSR1-GPD. This plasmid was further modified to incorporate a new multiple cloning site (MCS) and the trpC terminator from Aspergillus nidulans. Primer pair OTS340/OTS341 (AAAGCGGCCGCCCTGCAGGAATCGATCCTAGGTGTACAGCATGCTTAATTAATTTAATAGCTCCATGTCAACAA and AAACCGCGGGGAGCATTCACTAGGCAACCATGGT) and template plasmid pMFP26 (35) were used to generate a multiple cloning site (MCS)/terminator fragment with NotI and SacII sites at the 5′ and 3′ ends, respectively. The new MCS, appended onto primer OTS340, included NotI, SbfI, ClaI, AvrII, BsrGI, SphI, and PacI restriction sites. Cloning of the MCS/terminator fragment into pCSR1-GPD yielded plasmid pTSL126B. To create a version of this plasmid containing an mCherry-6× His tag, primer pair OTS415/OTS416 (AAAGCATGCGTGAGCAAGGGCGAGGAGGATAAC and CGCTTAATTAACTAGTGGTGGTGGTGGTGGTGGCTGCCCTTGTACAGCTCGTCCATGCC), incorporating SphI and PacI sites, respectively, was used to amplify a Neurospora codon-optimized mCherry-6× His fragment from plasmid pMFP26. Cloning of this fragment resulted in plasmid pTS184J. Subsequently, primer pair OTS387/418 (AAAGCGGCCGCATGGCATCCCTAAAGTCGCTG and AAACCTAGGAATAAGATGCTTAGATCTAAGCAG), incorporating NotI and AvrII restriction sites, respectively, was used to amplify the erv25 open reading frame for insertion into pTS184J, resulting in plasmid pTSL190A. Plasmid pTSL190A was linearized and transformed into the Δerv25 deletion strain FGSC 18824. Transformants were selected on medium containing cyclosporine.

Protein localization.

Protein localization was analyzed in germlings prepared as described above or in hyphal samples growing on VMM agar plates. Strains containing the vacuolar marker red fluorescent protein (RFP)-VAM3 and the Golgi membrane marker RFP-VPS52 (36) were used for colocalization studies. Heterokaryotic cells were obtained by mixing equal amounts of the two strains expressing the different proteins on VMM slant tubes. After conidiation, conidial suspensions were prepared and examined by confocal microscopy using a Leica SD6000 microscope with a 100×, 1.4-numeric-aperture oil immersion objective equipped with a Yokogawa CSU-X1 spinning disk head and 488-nm and 561-nm lasers controlled by Metamorph software (Molecular Devices).

Phylogenetic analyses.

The N. crassa LFD-2 protein sequence was used to search the JGI (http://jgi.doe.gov/) and FungiDB (http://fungidb.org/fungidb/) databases. JGI blast searches were conducted using the default settings (e value of <1e⁻⁵) with the scoring matrix set to BLOSUM62. FungiDB blast parameters matched those used for JGI database searches. Sequences with significant similarity (more than 33% amino acid identity) to LFD-2 (NCU02191) were selected for further analysis, with the exception of NCU07762 and Batrachochytrium dendrobatidis (GP3.005680), which were used as outgroups. Protein sequences were aligned using MAFFT (37), and gaps were trimmed by TrimAl (38) and further improved by Gblock (39) to remove highly divergent regions. Mview (40) was used to view alignments, and percent identities in alignment were calculated. Phylogenetic trees were built by PhyML V 3.0 (41) using LG mode (42), and the number of relative substitution rate categories was set to 16. The trees were visualized and converted with Figtree V 1.40 (http://tree.bio.ed.ac.uk/software/figtree/).

Protein domain prediction.

Pfam, InterPro, and SMART (43,–45) were used to predict conserved protein domains. Predicted domains supported by all databases are shown. TMHMM Server v. 2.0 (46) was used for transmembrane domain prediction. Signal peptides were detected by SignalP 4.1 (47). N-linked glycosylation sites were predicted by NetNGlyc (http://www.cbs.dtu.dk/services/NetNGlyc/), and O-glycosylation sites were predicted by YinOYang, DictyOGly, and NetOGlyc (48,–50).

Lysis is due to defects in plasma membrane integrity.

The lysis phenotype of adhered germlings in the ΔNCU01697 and ΔNCU02191 mutants could be due to defects in cell wall remodeling, which occurs at the point of adherence of germlings and is required for subsequent membrane fusion, or to defects in plasma membrane integrity during membrane merger. In S. cerevisiae, the growth defect of cell wall remodeling mutants is remediated by the addition of an osmotic stabilizer, such as sorbitol, to the growth medium but does not suppress the cell lysis phenotype of membrane fusion mutants, such as the prm1Δ mutant (15). To differentiate whether cell lysis during germling fusion is due to a defect in plasma membrane integrity versus cell wall integrity, we performed fusion assays in VMM (29) supplemented with 1 M sorbitol. As shown in Fig. 1B, osmotic stabilization failed to suppress cell lysis observed in VMM in both ΔPrm1 and Δlfd-1 mutants. In the two new mutants, the addition of sorbitol also failed to suppress the cell lysis phenotype of adhered germlings, suggesting that lysis is due to a defect in membrane integrity in these mutants and not a defect in cell wall integrity.

Characterization of the lfd-2 mutant.

The N. crassa ΔPrm1 and Δlfd-1 mutants show increased cell lysis of adhered germling pairs under conditions of reduced extracellular calcium (22). Therefore, we tested whether the frequency of cell lysis increased under reduced calcium conditions in the two deletion strains that showed lysis during cell-cell fusion. As shown in Fig. 1B, only the strain carrying a deletion of NCU02191 showed a significant increase in lysis of germling pairs under reduced extracellular calcium concentrations. Due to the similarity in germling fusion phenotype between ΔPrm1, Δlfd-1, and ΔNCU02191 strains, we focused on NCU02191, which we named lfd-2 for late fusion defect-2.

To obtain quantitative data for cell fusion events, we constructed duplicate strains where one Δlfd-2 strain carried a cytoplasmic GFP marker and another Δlfd-2 strain carried a cytoplasmic mCherry marker. Five hours postgermination, the fusion phenotype between green and red fluorescent Δlfd-2 germling pairs was analyzed. Whereas 100% of the adhered wild-type germling pairs fused and showed mixed green and red fluorescence, fusion between adhered Δlfd-2 germlings was reduced to 81% (Fig. 2). The reduction in fusion frequency in Δlfd-2 germling pairs was due to a significant increase in cell lysis (14% of the fusion pairs) rather than an increase in the class of adhered but nonfused germlings, as was observed for both ΔPrm1 and Δlfd-1 strains (Fig. 2B and ​andC).C). However, similar to ΔPrm1 and Δlfd-1 mutants (22), increasing the calcium concentration suppressed the lysis phenotype of adhered lfd-2 germlings (Fig. 2A).

Open in a separate window

FIG 2

Characterization of cell lysis frequencies of adhered germlings in Δlfd-2, Δlfd-1; Δlfd-2, and ΔPrm1; Δlfd-2 mutants under normal and altered calcium conditions. (A) Fusion (black bars), lysis (white bars), and blocked fusion (nonfused; gray bars) frequencies of adhered Δlfd-2 germlings at different calcium (Ca²⁺) concentrations. B) Fusion (black bars), lysis (white bars), and blocked fusion (gray bars) for single (Δlfd-1 or Δlfd-2) and double (Δlfd-1; Δlfd-2) mutants under different calcium concentrations. (C) Fusion (black bars), lysis (white bars), and blocked fusion (gray bars) of adhered germlings from single mutants (ΔPrm1 or Δlfd-2) versus the double mutant (ΔPrm1; Δlfd-2) under different calcium (Ca²⁺) concentrations. Bars indicate standard deviations.

The similarity in lysis phenotype between Δlfd-2, ΔPrm1, and Δlfd-1 mutants suggested that these proteins function in a similar process during cell fusion. To assess this hypothesis, we performed an epistasis test by constructing Δlfd-1 Δlfd-2 and ΔPrm1 Δlfd-2 double mutants. Compared with the cell lysis frequency in Δlfd-1 or Δlfd-2 adhered germlings (8% and 14%, respectively), the Δlfd-1 Δlfd-2 germlings showed an additive increase in lysis frequency (23%) (Fig. 2B, white bars). This lysis value increased significantly under reduced calcium concentrations (31%). The deletion of lfd-2 reduced the adhered, nonfused germling phenotype of Δlfd-1 mutants, with a concomitant increase in cell lysis, but did not affect the frequency of the class of fused germlings (Fig. 2B).

N. crassa germlings lacking Prm1 showed a high percentage of adhered but unfused germlings (~44%) (Fig. 2C) (21, 22), of which ~11% showed a lysis phenotype. An additive cell lysis phenotype also was observed in the ΔPrm1 Δlfd-2 germlings (26%) compared to ΔPrm1 or Δlfd-2 germlings alone (11% and 14%, respectively) (Fig. 3B, white bars). The cell lysis frequency of the ΔPrm1 Δlfd-2 germlings also increased significantly (34%) under conditions of reduced calcium. However, the frequency of germlings that underwent productive cell fusion in the ΔPrm1 Δlfd-2 germlings was not significantly different from that of ΔPrm1 germlings alone (Fig. 2C). Similar to the Δlfd-1 Δlfd-2 germlings, the absence of lfd-2 from ΔPrm1 germlings resulted in a substantial reduction in the adhered but nonfused germling class, which was replaced by an increase in the class of germlings that showed cell lysis.

Open in a separate window

FIG 3

Characterization of cell lysis frequencies of adhered germlings in Δfig1, ΔPrm1; Δfig1, Δfig1; Δlfd-1, and ΔPrm1; Δfig1; Δldf-1 mutants under normal and altered calcium conditions. (A) Frequency of fusion (black bars), lysis (white bars), or nonfusion (where adhered germlings are blocked at cell fusion; gray bars) of Δfig1 germlings under different calcium concentrations. (B) Frequency of fusion (black bars), lysis (white bars), or nonfusion (where adhered germlings are blocked at cell fusion; gray bars) of ΔPrm1; Δfig1, Δfig1; Δlfd-1, and ΔPrm1; Δlfd-1; Δfig1 germlings. Bars indicate standard deviations.

By transmission electron microscopy analyses, it was previously shown that a significant portion of ΔPrm1 and Δlfd-1 adhered germlings show stable appressed plasma membranes, which often display cellular invaginations (21, 22). Our data suggest that in the absence of LFD-2, the appressed membranes representative of the adhered, nonfused class of ΔPrm1 and Δlfd-1 germlings are destabilized, resulting in an increase in cell lysis in the ΔPrm1 Δlfd-2 and Δlfd-1 Δlfd-2 germling pairs. This result is different from that observed for ΔPrm1 Δlfd-1 mutants, where the increase in cell lysis occurs as a result of a decrease in cell fusion frequencies but the percentage of adhered, nonfused germlings was unaffected (22) (Fig. 3B).

The N. crassa fig1 mutant displays a germling lysis phenotype.

In S. cerevisiae, Fig1p is a plasma membrane protein that controls pheromone-induced calcium influx and is a component of the low-affinity calcium uptake system (LACS) (53). fig1Δ mating pairs show an ~25% reduction in cell fusion and show calcium-dependent cell lysis (12). A FIG1-like homolog recently was identified in N. crassa, and a strain bearing a deletion of fig1 displays a mating type-specific defect in the production of female reproductive structures (33). Due to its mating cell fusion defect and cell lysis phenotype in S. cerevisiae, we examined the role of fig1 during germling fusion in N. crassa. The fig1 locus (NCU02219) encodes a 267-amino-acid (aa) protein with 4 transmembrane domains. To obtain quantitative data for cell fusion events, we constructed Δfig1 strains carrying GFP or mCherry cytoplasmic markers. Δfig1 germlings showed normal chemotropic behavior and adherence. However, adhered Δfig1 germlings showed significant cell lysis (12.7%) (Fig. 3A), a phenotype very similar to that of Δlfd-2 mutants. As for fig1Δ mutants in S. cerevisiae and for ΔPrm1, Δlfd-1, and Δlfd-2 mutants in N. crassa, increasing extracellular calcium suppressed the cell lysis defect of Δfig1 germlings (Fig. 3A). Similar to the Δlfd-2 mutant, the frequency of adhered but nonfused Δfig1 germlings was not significantly different from that of the wild type. Complementation of Δfig1 with the fig1 open reading frame under the regulation of the tef1 promoter recovered WT levels of cell lysis (4.6% ± 1.2%), confirming that the lysis phenotype detected in the mutant resulted from the absence of FIG1.

To assess epistasis interactions between Prm1, lfd-1, and fig1, we constructed double and triple mutants and quantitatively assessed fusion and lysis in adhered germlings. The ΔPrm1 Δfig1 mutant showed a significant increase in cell lysis (~40%) relative to ΔPrm1 or Δfig1 germlings (~11% and ~12.7%, respectively), indicating a synergistic effect of the deletion of both genes. The adhered germling class, and particularly the fused germling class, were reduced in the ΔPrm1 Δfig1 germlings. In contrast, the Δfig1 Δlfd-1 mutants showed an additive effect on cell lysis (~22%) (Fig. 3B). As with the Δlfd-1 Δlfd-2 germlings, the increase in cell lysis of the Δfig1 Δlfd-1 germlings was due to a decrease in the class of adhered, nonfused germlings, but not in the frequency of germlings that underwent productive cell fusion (Fig. 3B). The triple mutant (ΔPrm1 Δlfd-1 Δfig1) showed a phenotype similar to that of ΔPrm1 Δlfd-1 germlings (Fig. 3B).

To determine whether cell lysis and death observed during germling fusion also occurs during hyphal fusion in an N. crassa colony, we assessed lysis in a mature colony by staining with the vital dye methylene blue. In a wild-type colony, no staining by methylene blue was observed in fungal compartments at hyphal fusion junctions (Fig. 4A; circles indicate fusion junctions). However, in a ΔPrm1 colony, cell lysis at approximately half of the hyphal fusion points was observed (Fig. 4B, red circles). The vital dye staining showed compartmentalization at the fusion point, suggesting that the septal pores surrounding the fusion point were blocked in these cells. For Δlfd-1 and Δlfd-2 colonies (Fig. 4C and ​andD)D) as well as for Δfig1 and ΔNCU01697, no lysis at hyphal fusion points within a colony was detected.

Open in a separate window

FIG 4

Evaluation of lysis and compartmentation during hyphal fusion in a wild-type colony versus ΔPrm1, Δlfd-1, and Δlfd-2 mutant colonies. (A) Lysis at hyphal fusion points in a wild-type colony was assessed by staining with the vital dye methylene blue; black circles show productive fusion events (no cell lysis). (B) Lysis at hyphal fusion points in a ΔPrm1 colony was assessed by staining with the vital dye methylene blue. Red circles show cell lysis that occurred at fusion points; black circles show productive fusion events. (C) Lysis at hyphal fusion points in a Δlfd-1 colony was assessed by staining with the vital dye methylene blue. Black circles show productive fusion events. (D) Lysis at hyphal fusion points in a Δlfd-2 colony was assessed by staining with the vital dye methylene blue. Black circles indicate productive fusion events. Scale bars, 50 μm.

We acknowledge Andre Fleissner (Technische Universitat Braunschweig, Germany) for the tef-1 promoter vector and Barry Bowman (University of California, Santa Cruz) for the RFP-VPS52 and RFP-VAM3 strains.

This work was supported by a National Science Foundation grant (MCB 1121311) and a National Institutes of Health grant (R01GM060468) to N.L.G. We acknowledge the use of materials generated by NIH program project grant P01GM068087, functional analysis of a model filamentous fungus. A.P.G. was the recipient of a fellowship from Fundação Calouste Gulbenkian (104210) and a short-term fellowship from EMBO (329-2012).