The Guanine Nucleotide Exchange Factor RIC8 Regulates Conidial Germination through Gα Proteins in Neurospora crassa

Heterotrimeric G protein signaling is essential for normal hyphal growth in the filamentous fungus Neurospora crassa. We have previously demonstrated that the non-receptor guanine nucleotide exchange factor RIC8 acts upstream of the Gα proteins GNA-1 and GNA-3 to regulate hyphal extension. Here we demonstrate that regulation of hyphal extension results at least in part, from an important role in control of asexual spore (conidia) germination. Loss of GNA-3 leads to a drastic reduction in conidial germination, which is exacerbated in the absence of GNA-1. Mutation of RIC8 leads to a reduction in germination similar to that in the Δgna-1, Δgna-3 double mutant, suggesting that RIC8 regulates conidial germination through both GNA-1 and GNA-3. Support for a more significant role for GNA-3 is indicated by the observation that expression of a GTPase-deficient, constitutively active gna-3 allele in the Δric8 mutant leads to a significant increase in conidial germination. Localization of the three Gα proteins during conidial germination was probed through analysis of cells expressing fluorescently tagged proteins. Functional TagRFP fusions of each of the three Gα subunits were constructed through insertion of TagRFP in a conserved loop region of the Gα subunits. The results demonstrated that GNA-1 localizes to the plasma membrane and vacuoles, and also to septa throughout conidial germination. GNA-2 and GNA-3 localize to both the plasma membrane and vacuoles during early germination, but are then found in intracellular vacuoles later during hyphal outgrowth.


Introduction
Since the discovery of the first heterotrimeric G protein in filamentous fungi in the 1990's [1], G proteins have been found to play key roles in diverse fungal processes ranging from asexual and sexual development to pathogenicity of animal and phytopathogenic fungi (reviewed in Li et al, 2007). Most fungi possess three Ga subunits and a single Gb and Gc protein, therefore allowing for the assembly of three different heterotrimers. These three Ga subunits can act independently to regulate separate pathways, leading to differing phenotypes for single Ga mutants. For example, Neurospora crassa GNA-1 is required for normal vegetative growth, aerial hyphae formation and female fertility [2], whereas GNA-3 is required for normal production of asexual spores (conidia) and maturation of sexual spores (ascospores) [3]. In contrast, the N. crassa Dgna-2 mutant displays only a mild phenotype during growth on poor carbon sources [4]. However, loss of GNA-2 exacerbates phenotypes of the Dgna-1 and Dgna-3 mutants, indicating that GNA-2 shares overlapping functions with the other two Ga subunits [5,6]. Indeed, all three G proteins are thought to act together to regulate certain processes, as mutants lacking GNA-1 and GNA-3 or all three Ga subunits are severely impaired in growth on solid medium, inappropriately conidiate in submerged liquid culture and do not produce female reproductive structures [6].
G protein coupled receptors (GPCRs), act as guanine nucleotide exchange factors (GEFs) for Ga subunits, facilitating exchange of GDP for GTP, thereby leading to activation and dissociation from the Gbc dimer (reviewed in Li et al, 2007). However, recently a non-receptor GEF capable of activating Ga proteins, RIC8, has been identified in both animals and some fungi [7,8,9]. In N. crassa, loss of ric8 leads to a severe growth impairment phenotype similar to that in mutants lacking both gna-1 and gna-3 or all three Ga subunit genes [9]. Expression of GTPase-deficient gna-1 or gna-3 alleles rescued many of the defects of the Dric8 mutant during asexual growth on solid medium, and biochemical analyses showed that RIC8 can act as a GEF for both GNA-1 and GNA-3 in vitro, suggesting RIC8 acts upstream of both GNA-1 and GNA-3, particularly during regulation of asexual growth on solid medium [9]. Asexual hyphal growth is important for nutrient scavenging and for the organism to spread throughout the environment. In addition, it is important for encountering a mate of the opposite mating type, which allows the sexual cycle to proceed and produce the environmentally resistant sexual spores (ascospores).
Using a strain expressing a functional RIC8-GFP fusion, we have previously shown that RIC8 is a cytoplasmic protein [9]. Production of Ga-fluorescent protein fusions is problematic, as an N or C-terminal tag can interfere with normal functioning of the Ga protein. However, Dictyostelium discoideum Ga2 and mammalian Ga o have been successfully tagged by insertion of GFP in a fold where it does not interfere with Ga o function [10,11].
In this study we further probe the role of RIC8, GNA-1 and GNA-3 in asexual hyphal growth and development. We analyze conidial morphology and determine conidial germination rates in ric8 and G protein subunit mutants and in Dric8 strains carrying GTPase-deficient alleles of gna-1 or gna-3. We produce strains expressing GNA-1, GNA-2 and GNA-3 proteins as internal TagRFP fusions. We present here the first use of this internal tagging method for localization of Ga proteins in filamentous fungi. Using these strains, we determine the localization pattern of Ga proteins in conidia and during conidial germination and test for colocalization of RIC8 with GNA-1 and GNA-3.

Strains and growth conditions
Neurospora strains used in this study are listed in Table 1. For vegetative growth analysis, strains were grown on Vogel's minimal medium (VM; [12]). To induce the formation of female structures (protoperithecia) required for sexual crossing strains were grown on synthetic crossing medium (SCM; [13]). Cultures were inoculated with conidia and grown as described previously [4,14].

Ga-TagRFP strain construction
To observe the cellular localization of GNA-1, GNA-2 and GNA-3, TagRFP [15] was inserted into a conserved loop region within the Ga protein. This conserved loop region was found to be optimal for insertion of tags with conservation of Ga protein function in Dictyostelium discoideum Ga2 [11] and chinese hamster Gnao [10]; see Fig. 1). Primers were designed to prepare the Ga-TagRFP fusion constructs using yeast recombinational cloning, and are listed in Table 2. TagRFP was amplified from pAL3-Lifeact [15]; provided by Nick Read, University of Edinburgh) and the appropriate Ga N-and C-terminal fragments were amplified from cDNA clones. These fragments were inserted into pRS426 using yeast recombinational cloning [16]. The Ga-TagRFP fusion construct was then subcloned from pRS426 into pMF272 [17] as an EcoRI/XbaI fragment, resulting in replacement of sgfp from pMF272 with the Ga-TagRFP fusion and placing it under the control of the ccg-1 promoter. The fusion constructs were then transformed into his-3 Dgna-1, Dgna-2 or Dgna-3 gene replacement mutants (See Table 1). Transformants were then screened by Southern blotting to ensure correct integration of the construct (data not shown).

Western blot analysis
Western blotting was used to detect Ga-TagRFP fusion proteins in whole-cell extracts prepared from macroconidia. Conidia from 6-7 day flask cultures were collected using sterile water, pelleted and stored at 280uC. After thawing on ice, the conidia were Product #T8215) and then transferred to a mortar. The conidia were vigorously crushed using a mortar and pestle under liquid nitrogen. Roughly equal amounts of ground tissue were then transferred to 2 ml screw cap tubes, and topped off with additional extraction buffer if necessary. Samples were spun at 3,300 RPM (1,0006g) for 10 minutes at 4uC in a microcentrifuge and the supernatants (whole cell extracts) retained. Protein concentration was determined using the Bradford Protein Assay (Bio-Rad, Hercules, CA). Samples containing 50 mg of whole-cell protein were subjected to SDS-PAGE using 10% gels and gels transferred to nitrocellulose membranes as described previously [18]. Western analysis was performed using a polyclonal RFP antibody (#R10367, Invitrogen, Carlsbad, CA) as the primary antibody at a dilution of 1:2000. A horseradish peroxidase conjugate antibody (Bio-Rad) was used as the secondary antibody at a dilution of 1:5000 and chemiluminescent detection was performed as described previously [18].

Microscopy
For analysis of conidial germination, 8610 6 conidia were spread on 100 mm 10 ml VM agarose plates and incubated at 30uC for 0, 4, 6 or 8 h. Cells were then visualized using differential interference contrast (DIC) microscopy using an Olympus IX71 inverted microscope (Olympus America, Center Valley, PA) with a 606 oil immersion objective (NA = 1.42). Images were captured using a QIClickTM digital CCD camera (QImaging, Surrey, British Columbia, Canada) and analyzed using Metamorph software (Molecular Devices Corporation, Sunnyvale, CA). For analysis of conidial anastomosis tubes, conidia were spread on VM agarose, and imaged after 5-16 h at 30uC as detailed above.
For observation of Ga localization, VM agarose plate cultures were prepared as described above. Germinating conidia were analyzed using a Leica TCS SP5 II confocal microscope with a 636 oil objective (NA = 1.40; Leica Microsystems Inc., Buffalo Grove, IL). The Ga-RFP strains were visualized with the Hybrid Detection system (HyD) laser at an excitation of 543 nm, and emission of 565-665 nm.
To confirm vacuolar localization, conidia from the Ga-TagRFP strains were inoculated onto VM agarose plates and incubated as described above. An aliquot containing 30 ml of a 20 mg/ml solution of Oregon Green 488 carboxylic acid diacetate (Carboxy-DFFDA; catalog number O6151; Molecular Probes) was applied to a coverslip. An agarose block containing germinating conidia was inverted onto the coverslip and incubated for 5 minutes at room temperature in the dark. Images were obtained using the Leica TCS SP5 II confocal microscope described above. GFP images were obtained by excitation at 488 nm, with emission collected from 500-535 nm. RFP images were obtained with excitation at 543 nm and emission from 555-700 nm. Images were captured sequentially in order to prevent crosstalk among samples.
The vacuolar and plasma membrane localization of GNA-1-TagRFP was further explored by imaging a heterokaryon that expresses GNA-1-TagRFP and a GFP fusion of the Ca 2+ ATPase, NCA-3 [19]. NCA-3-GFP is known to localize to vacuoles and the plasma membrane [19]. The two strains were co-inoculated onto a VM slant in order to produce a heterokaryon with conidia expressing both fluorescent proteins. Conidia were inoculated onto VM agarose plates as described above and imaged 6 h later using a 543 nm HyD laser and 488 nm laser on the Leica TCS SP5 II confocal microscope. Images were captured sequentially.
Possible co-localization of RIC8 and GNA-1 or GNA-3 was investigated through co-culturing of the RIC8-GFP strain with the GNA-1-TagRFP or GNA-3-TagRFP strains on a VM slant to produce a heterokaryon expressing two fluorescent proteins. Conidia from the heterokaryons were inoculated onto VM agarose plates as described above, followed by incubation at 30uC for 0 or 6 h for GNA-1-TagRFP/RIC8-GFP and 0 or 4 h for the GNA-3-TagRFP/RIC8-GFP fusion strain. Images were obtained by confocal microscopy, as described above.

Statistical analysis
Germination rates of strains relative to wild type were determined using Student's two-sided t test, with values paired by day of analysis [20]. Multiple comparisons within time points were corrected using the false discovery rate (FDR) approach of [21]. Differences in arthroconidiation and germination rates between strains were determined using Student's unpaired twosided t test. Germination rates were first normalized for daily RIC8, Galpha Proteins and Spore Germination PLOS ONE | www.plosone.org differences in wild type germination rates. All statistics were performed in R [22]; code is available on request. Details of the statistical analyses are presented in Tables 3, 4, 5, 6.

Loss of ric8 or both gna-1 and gna-3 leads to overproduction of arthroconidia
We have previously demonstrated that Dgna-1 and Dric8 single mutants, Dgna-1 Dgna-3 double mutants, and the triple Ga mutant have smaller colony sizes than wild type [2,6,9]. The defects of the latter three mutants are similar and much more severe than those of the Dgna-1 strain [6,9]. Smaller colony size could result from slow germination, reduced hyphal extension or both phenomena. In order to determine whether germination defects contribute to the overall reduction in colony size, we investigated early events during germination of conidia in Dric8 mutants and in strains lacking G protein subunit genes.
In N. crassa, arthroconidiation has been proposed to be the default conidiation pathway in mutants with defects in macroconidiation [24]. Arthroconidia are the major cell type used to disseminate many animal fungal pathogens, such as the valley fever fungus Coccidioides immitis and dermatophytes in the genera Mycosporum, Trichophyton and Epidermophyton [25,26,27]. To date, little is known about the regulation of arthroconidiation in these fungi, but it seems clear that RIC8, GNA-1 and GNA-3 impact this process in N. crassa. G protein signaling is required for conidial germination, but is not essential for conidial anastomosis tube formation We next investigated the ability of macroconidia from the various strains to form conidial anastomosis tubes (CATs) and to germinate on solid medium. CATs are small tube-like structures produced by conidia early during colony initiation (rev. in [28]). The results demonstrate that Dric8 and all G protein subunit single and double mutants produce CATs from macroconidia (Fig. 4).
We next analyzed conidial germination (both macroconidia and arthroconidia) in the various strains at 4, 6 and 8 h after plating on Table 2. Primers used in this study.

Name
Anneals to Sequence (59-39)  solid medium. In wild-type, greater than 70% of the conidia have germinated by 4 hr after inoculation, with nearly 100% germination after 8 h (Fig. 5; Table 7). In contrast, germination is significantly reduced in Dric8, Dgna-3, Dgna-1 Dgna-3 and the triple Ga mutants relative to wild type at all time points (Fig. 2, 6). Germination of the Dgna-1 mutant also appears to be slower than wild-type, but this is only statistically significant at 6 h. However, the observation that germination of the Dgna-1 Dgna-3 mutant is significantly lower than the Dgna-3 mutant supports a compensatory role for GNA-1 in conidial germination (Fig. 2, 6). Loss of gna-2 had no effect on conidial germination, and did not exacerbate the germination defect of the Dgna-1 Dgna-3 mutant (Fig. 2). Taken together, the results indicate that RIC8, GNA-3, and to a lesser extent GNA-1, play important roles during conidial germination.
Subsequently, we tested whether constitutive activation of any of the three Ga subunits could rescue the conidial germination defect of Dric8 strains. Introduction of the GTPase-deficient gna-1 or gna-2 alleles did not significantly improve germination, (Fig. 2,  6; Table 4, 5). Strikingly, constitutive activation of gna-3 in the Dric8 background resulted in a significant increase in conidial germination, approaching that of the wild-type strain ( Fig. 6; Table 4, 5). The ability of the GTPase-deficient gna-3 allele to bypass the Dric8 germination defect supports RIC8 acting through GNA-3 to regulate conidial germination in N. crassa. We also probed a possible role for the Gb and Gc subunits during conidial   Table 6. Statistical analysis of germination rates between wild type and Ga-TagRFP expressing strains.    germination (Fig. 3, Table 4). Similar to the results exploring conidial morphology, germination is largely unaffected in Dgnb-1 and Dgng-1 mutants, aside from a slight delay in Dgng-1 4 h after inoculation.

The three Ga proteins localize to the plasma membrane and vacuoles during conidial germination
Having demonstrated that Ga proteins are important for conidial germination, we next investigated the subcellular localization of these proteins during this process. For these experiments, we took advantage of previous studies in Dictyostelium and mammalian cells demonstrating that insertion of a fluorescent tag in a loop region of the Ga subunit does not disrupt the interaction with the Gbc dimer or interfere with Ga function ( Fig. 1; [ 10,11]. We expressed all three Ga-TagRFP fusions under control of the highly expressed ccg-1 promoter from the his-3 locus in the corresponding Ga mutant strain (see Materials and Methods). Expression of GNA-1-TagRFP complemented the conidial germination (Table 6) and female fertility defects and partially complemented the growth rate phenotype of the Dgna-1 mutant (data not shown). Expression of GNA-3-TagRFP in the Dgna-3 mutant restored conidial germination back to wild-type levels at all time points investigated (Table 6).
We confirmed that the three TagRFP constructs are expressed as full-length fusion proteins in conidia using Western analysis with an RFP antiserum (Fig. 7). TagRFP is 27 kD and each Ga is approximately 41 kD. Correspondingly, each of the three Ga fusion proteins migrated close to their predicted molecular mass of 68 kD. Interestingly, although all three genes are expressed under control of the same promoter from a common genomic site, levels of GNA-1 are much higher than those of GNA-2 and GNA-3.
This suggests some post-transcriptional regulation of Ga protein levels, with GNA-1 accumulating to higher levels in N. crassa. In addition, potential degradation products were observed for all three proteins ( Fig. 7; asterisks), suggesting at least some of the variation in protein levels may be due to differences in protein turnover.
We examined the strains expressing TagRFP fusion proteins during conidial germination using confocal fluorescent microscopy (Fig. 8). Conidia were plated on solid medium and examined at four time points (0, 4, 6 and 8 h). At 0 h, all three Ga proteins localized to the plasma membrane and vacuoles (Fig. 8, 9A). The plasma-membrane localization is consistent with results previously observed during cell fractionation studies in our laboratory [2,5,6]; data not shown). The vacuolar localization of GNA-1-TagRFP, GNA-2-TagRFP, and GNA-3-TagRFP was validated by the overlapping signal observed between TagRFP and the carboxy-DFFDA vacuolar dye during confocal microscopy (Fig. 9A). In addition, a strain expressing a GFP-tagged protein, NCA-3, a Ca 2+ ATPase known to localize to vacuoles and the plasma membrane, was combined with the GNA-1-RFP strain to produce a heterokaryon [19]. Many conidia and germinating hyphae from these heterokaryons exhibit green and red fluorescence, indicative of the presence of both NCA-3-GFP and GNA-1-TagRFP. Confocal microscopy of conidia from this heterokaryon confirmed that GNA-1 localizes to the plasma membrane and vacuoles (Fig. 9B). It is unclear whether the vacuolar fluorescence represents a functional localization or a site of protein turnover.
At 4 h of conidial germination, GNA-1 continued to display plasma membrane and vacuolar localization, and could also be observed on the first septum separating the conidium and the developing hypha (Fig. 8). In contrast, GNA-2 and GNA-3 were predominantly found in vacuoles, with less apparent plasma membrane localization (Fig. 8). At 6 h, GNA-1 was present in the  Table 7. Error bars are 6SE for twelve independent experiments (n = minimum of 60 cells). doi:10.1371/journal.pone.0048026.g005 RIC8, Galpha Proteins and Spore Germination PLOS ONE | www.plosone.org plasma membrane, vacuoles and septa. (Fig. 8). GNA-2 exhibited a similar localization as observed at 4 h. Interestingly, in addition to vacuolar localization, GNA-3 was also found in distinct patches on the plasma membrane of the original conidium (Fig. 8). These patches were observed in 50% or more of the cells being sampled. At 6 h, the plasma membrane patches were even more easily observed in GNA-3-TagRFP strains, as the fluorescence in vacuoles and other regions of the plasma membrane began to weaken (Fig. 8). At 8 h, GNA-1-TagRFP fluorescence was still relatively strong in plasma membrane, vacuoles and septa (Fig. 8).
We have previously demonstrated that RIC8 exhibits cytosolic localization in conidia and mature hyphae using fluorescent microscopy with an inverted compound microscope [9]. In this study, we utilized the more discriminating method of confocal fluorescence microscopy to explore possible co-localization of GNA-1-TagRFP and/or GNA-3-TagRFP with RIC8-GFP in   cellular compartments during conidial germination. For these experiments, we cultured strains carrying the RIC8-GFP and GNA-1-TagRFP constructs together to produce heterokaryons expressing both fluorescent proteins. Similar to previous results, RIC8-GFP is cytosolic and excluded from vacant areas that appear to be vacuoles (Fig. 10). In contrast, GNA-1-TagRFP is located in the plasma membrane and vacuoles (Fig. 10) and the merged image did not reveal co-localization with RIC8-GFP in conidia or young germlings (Fig. 10). Similarly, co-localization was also not observed between GNA-3-TagRFP and RIC8-GFP (Fig. 10). In addition, there was no evidence for co-localization between the GNA-3 fluorescent patches and RIC8 (Fig. 10).

Discussion
G proteins regulate nearly every facet of growth and development in fungi. During asexual development, in particular, they have been found to regulate both the timing and level of conidiation (rev. in [29], and in recent years have also been shown to regulate conidial germination [30,31,32,33,34]. Here, we report evidence showing that in N. crassa G protein signaling regulates both conidial development and germination, and this is regulated through the non-receptor GEF RIC8. We observed that loss of any single G protein subunit has no effect on conidial morphology in N. crassa. However, loss of both gna-1 and gna-3 Ga genes leads to a dramatic increase in arthroconidia formation, which is also observed in the Dric8 mutant. Arthroconidia are formed by the fragmentation of vegetative hyphae [23] and their production is proposed to be a default pathway in mutants defective in macroconidiation [24]. Expression of constitutively active (GTPase deficient) GNA-1 or GNA-3 in the Dric8 mutant background does not lead to reduced arthroconidiation, suggesting RIC8 controls arthroconidia formation through Ga-dependent and independent mechanisms. While the exact mechanism by which RIC8 and the Ga proteins negatively regulate arthroconidiation is unclear, it may involve the MAK-1 MAP kinase signaling pathway, as deletion of any of the three kinases of this pathway leads to an overproduction of arthroconidia [35]. Additionally, loss of rgb-1, homologous to the B subunit of type 2A Ser/Thr phosphatases, also leads to production of large amounts of arthroconidia, perhaps through a negative effect on the MAK-1 pathway [24]. Interestingly, in spite of the requirement for G protein signaling genes during conidial development and germination, Dric8 and all G protein subunit single and double mutants were found to produce conidial anastomosis tubes from macroconidia. However, it remains unclear whether the conidia from these mutants that do not germinate are also unable to produce conidial anastomosis tubes.
Conidial germination was found to be significantly reduced in the Dgna-3 mutant, indicating GNA-3 regulates conidial germination in N. crassa, as has been reported for homologs in Penicillium marneffei [34], Aspergillus nidulans [30] and Botrytis cinerea [31]. The germination defect is exacerbated in the Dgna-1 Dgna-3 double mutant, revealing a role for GNA-1 in the absence of GNA-3. GNA-1 homologs have been shown to regulate conidial germination in other fungi [32,33,36]. Germination is reduced in the Dgna-1 mutant relative to wild type, but this is only statistically significant 6 h after inoculation, consistent with GNA-1 playing a more minor role in N. crassa. The observation that GNA-3 is more important during germination in N. crassa than GNA-1 is interesting, given the Dgna-1 mutant displays a greater reduction in hyphal extension rate than the Dgna-3 mutant [2,3]. This suggests that GNA-1 is more important for overall extension of hyphae rather than for early events during germination, while the converse is true for GNA-3. Germination of the Dgna-2 mutant is no different than wild type and germination is similar in Dgna-1 Dgna-2 Dgna-3 triple and Dgna-1 Dgna-3 double mutants, implying that GNA-2 is not required for conidial germination in N. crassa.
Our results demonstrate that the non-receptor GEF RIC8 regulates conidial germination primarily through the GNA-3 Ga subunit, with minor contribution from GNA-1. This observation is consistent with results from GTP binding assays demonstrating Figure 9. All three Ga proteins localize to vacuoles. A. Conidia expressing the corresponding Tag-RFP Ga were inoculated onto 100 mm solid VM at a concentration of 8610 6 conidia per plate. Plates were incubated at 30uC for 4 h (GNA-2 and GNA-3) or 6 h (GNA-1). The vacuolar dye Carboxy-DFFDA was applied at a concentration of 20 mg/ml. Images were captured using the Leica TCS SP5 II inverted confocal microscope. The RFP and GFP panels were merged to create the overlay. B. Conidia expressing both GNA-1-TagRFP and NCA-3-GFP were inoculated onto 100 mm solid VM at a concentration of 8610 6 conidia per plate. Plates were incubated at 30uC for 6 h. Images were captured using the Leica TCS SP5 II inverted confocal microscope. The RFP and GFP panels were merged to create the overlay. All panels are 46 zoom. Bar = 5 mm. doi:10.1371/journal.pone.0048026.g009 that RIC8 exhibits greater GEF activity towards GNA-3 than GNA-1 [9]. Interestingly, in M. oryzae loss of MoRIC8 does not affect conidial germination [7]. Instead, MoRIC8 is required for normal asexual growth, conidiation, appressorium formation and pathogenicity [7].
It is likely that RIC8 and the Ga proteins regulate conidial germination in N. crassa through modulation of cAMP levels [3,9,37,38]. The N. crassa Dric8 mutant has decreased adenylyl cyclase protein and the Dric8 mutation can be suppressed by a mutation in the protein kinase A regulatory subunit [9]. A mutation in the regulatory subunit leads to hyperactivation of the PKA catalytic subunit, thereby bypassing the need for wild-type levels of adenylyl cyclase. The Ga subunit GNA-3 is required to maintain wild-type levels of adenylyl cyclase, further supporting a role for G proteins in cAMP-dependent conidial germination [3]. In B. cinerea and A. nidulans cAMP plays a key role in conidial germination regulated by GNA-3 homologs [31,39].
In this study we pioneered the use of an internal Ga tagging method in filamentous fungi based on a successful approach used in Dictyostelium and mammalian cells, inserting TagRFP into a conserved loop region in the Ga subunits [10,11]. Using this approach we were able to localize GNA-1, GNA-2 and GNA-3 during conidial germination, revealing that all three Ga proteins localize to the plasma membrane in ungerminated conidia and young germlings and that GNA-1 can be detected in septa in older hyphae. All three Ga subunits were also detected in vacuoles, however, it is unclear if this is a functional localization, as proteins are often targeted for degradation by the vacuole [40]. Perhaps the most intriguing finding was the observation of patches of GNA-3-TagRFP on the plasma membrane at later time points during germination of conidia. This change in localization during the transition from conidium to germling may be related to the requirement for GNA-3 during conidial germination; further analysis is necessary to explore this hypothesis.
Using purified proteins in in vitro assays, we have previously demonstrated that RIC8 can bind to and act as a GEF for GNA-1 and GNA-3 [9]. However, confocal microscopy of strains expressing GNA-1-TagRFP or GNA-3-TagRFP and RIC8-GFP did not reveal any evidence for co-localization of either Ga protein with RIC8 in conidia or hyphae, suggesting the interaction may be transient. Studies in Drosophila neuroblasts have demonstrated cytoplasmic localization of Ric-8, while Ga i was found in the apical cortex [41]. Ric-8 was also observed in 'spot-like' structures close to the apical cortex that partially colocalized with Ga i , indicating that their interaction may take place on the cytoplasmic face of the plasma membrane or in the cytoplasm [41].
Interestingly, the GNA-1-TagRFP signal was stronger than that of GNA-2 and GNA-3, and detectable over a longer period, suggesting this protein fusion may be more stable. This conclusion is also supported by the results from western analysis using RFP antibodies. In yeast, the Ste2p GPCR is ubiquitinated and targeted to the vacuole for degradation [42]. Analysis of N. crassa Figure 10. Localization of GNA-1-TagRFP and GNA-3-TagRFP with RIC8-GFP in germinating conidia. Conidia from a fused strain expressing GNA-1-TagRFP or GNA-3-TagRFP and RIC-8-GFP were harvested as described in the Materials and Methods. An aliquot containing 8610 6 conidia was spread on a 100 mm VM solid medium plate. Images were captured using a Leica TCS SP5 II inverted confocal microscope. Conidia were imaged immediately after inoculation on the solid medium plate. Conidia were allowed to germinate for 6 h or 4 h at 30uC before imaging. The first two panels were merged to create the third panel. All panels are 46 zoom. Scale bar = 5 mm. doi:10.1371/journal.pone.0048026.g010 GNA-1, GNA-2 and GNA-3 using UbPred (http://www.ubpred. org/) predicts that GNA-1 and GNA-2 contain potential ubiquitination target sites, which may explain the vacuolar localization of these proteins. Future studies will explore whether this vacuolar localization is functional, and if so investigate the importance of G protein signaling in the vacuole.