The so locus is required for vegetative cell fusion and postfertilization events in Neurospora crassa

Abstract

The process of cell fusion is a basic developmental feature found in most eukaryotic organisms. In filamentous fungi, cell fusion events play an important role during both vegetative growth and sexual reproduction. We employ the model organism Neurospora crassa to dissect the mechanisms of cell fusion and cell-cell communication involved in fusion processes. In this study, we characterized a mutant with a mutation in the gene so, which exhibits defects in cell fusion. The so mutant has a pleiotropic phenotype, including shortened aerial hyphae, an altered conidiation pattern, and female sterility. Using light microscopy and heterokaryon tests, the so mutant was shown to possess defects in germling and hyphal fusion. Although so produces conidial anastomosis tubes, so germlings did not home toward wild-type germlings nor were wild-type germlings attracted to so germlings. We employed a trichogyne attraction and fusion assay to determine whether the female sterility of the so mutant is caused by impaired communication or fusion failure between mating partners. so showed no defects in attraction or fusion between mating partners, indicating that so is specific for vegetative hyphal fusion and/or associated communication events. The so gene encodes a protein of unknown function, but which contains a WW domain; WW domains are predicted to be involved in protein-protein interactions. Database searches showed that so was conserved in the genomes of filamentous ascomycete fungi but was absent in ascomycete yeast and basidiomycete species.

FIG. 1.

Macroscopic phenotype of the so mutant. The so (soft) mutant exhibits short aerial hyphae and an altered conidiation pattern. (A) Agar flask cultures. Left, wild type (FGSC 2489); right, so mutant (FGSC 542). (B) Agar plate cultures. Top, wild-type strain (FGSC 2489); bottom, so mutant (FGSC 542).

FIG. 2.

A comparison of colony morphology between the wild type and a so mutant. Hyphae imaged by confocal microscopy after staining with FM4-64 (A to D) or Calcofluor M2R (E, F). (A and B) Morphology of hyphae at the colony periphery. (C to F) Morphology of hyphae within sub-peripheral colony regions. Asterisks in panel C indicate fusion events. Daggers in panel D indicate contact of so hyphae but absence of fusion events. Strains are wild type (FGSC 2489) in A, C, and E and so mutant (FGSC 542) in B, D, and F. Bars = 20 μm (A to D) and 10 μm (E, F).

FIG. 3.

The so mutant shows defects in germling fusion. (A) Fusion (indicated by arrow) between germinating conidia of wild-type FGSC 2489 in stationary liquid culture. (B) so (FGSC 542) conidia in stationary liquid culture germinate at a comparable time to that of a wild-type strain but do not show germling fusion events. If germlings appeared to have physical contact (arrow), no cell fusion events were observed. The inset shows the area of physical contact at a higher magnification. (C) Confrontation between so (FGSC 542) and a wild-type (Wt) strain (R12-60) on solid medium. The wild-type strain is labeled by H1-GFP (H1) (white nuclei). While homing and fusion between wild-type conidia were observed, (black arrow), a reaction between so conidia or so and wild-type conidia did not occur. CATs, however, did form in so strains (white arrow) but were formed at a reduced frequency compared to that in a wild-type strain. Bars = 10 μm.

FIG. 4.

so conidial anastomosis tubes do not home and fuse. (A) The CATs of two wild-type (R12-60) macroconidia home toward each other (0 min). The lower germling was manipulated by optical tweezers and moved slightly to the right (5 min). The two CATs subsequently homed back toward each other and fused (59 min). Note that the CATs increased in width after manipulation. (B) CATs from so (FGSC 542) and wild-type (R12-60) macroconidia, which were orientated toward each other via manipulation by optical tweezers, were followed in a time course. The growth of the wild-type and so conidia toward each other was not due to homing, and when the CATs touched they did not fuse. Note that the germlings showed some drifting and rotation but fail to show chemotropic interactions. The wild-type germlings were distinguished from the so germlings by labeling with nucleus-targeted H1-GFP (data not shown). Bar = 5 μm.

FIG. 5.

Trichogynes of so fuse with microconidia of the opposite mating type. Microconidia containing H1-GFP-tagged nuclei (R12-60 and R11-03) were placed on top of wild-type (FGSC 2489) (A) and so (FGSC 542) (B) protoperithecia of the opposite mating type, respectively. The interaction between trichogynes and microconidia was evaluated by microscopy 22 h (A1, B1) and 44 h (A2, B2) after inoculation. (A) Trichogynes from the WT strain (FGSC 2489) also show attraction to wild-type microconidia of the opposite mating type (R12-60) (black arrow). The GFP-labeled nuclei disappeared after contact between a wild-type trichogyne (FGSC 2489) and a conidium of the opposite mating type (R12-60) (white arrow). (B) Trichogynes from so (FGSC 542) also show attraction to wild-type microconidia of the opposite mating type (R11-03) (black arrow). Loss of nuclear GFP fluorescence was also observed (B1 versus B2; white arrow). Bar = 10 μm.

FIG. 6.

Cloning strategy for so. Contiguous cosmids in a region of ∼130 kbp around the so locus on the right arm of LGI. Flanking markers aro-8 and arg-13 are indicated. H39A5 contains five predicted open reading frames (based on our analysis and http://www-genome.wi.mit.edu/annotation/fungi/neurospora). Boxes indicate ORFs (not drawn to scale) on contig 3.145 designated NCU02793.1 to -02797.1. Each ORF was cloned and transformed individually into a so mutant (R9-01) to assess complementation.