Dominant active Rac and dominant negative Rac revert the dominant active Ras phenotype in Colletotrichum trifolii by distinct signalling pathways

Authors


E-mail mdickman@unlnotes.unl.edu; Tel. (+1) 402 472 2849; Fax (+1) 402 472 2853.

Summary

The small G-protein superfamily is an evolutionarily conserved group of GTPases that regulate diverse signalling pathways including pathways for growth and development in eukaryotes. Previously, we showed that dominant active mutation in the unique Ras gene (DARas) of the fungal phytopathogen Colletotrichum trifolii displays a nutrient-dependent phenotype affecting polarity, growth and differentiation. Signalling via the MAP kinase pathway is significantly impaired in this mutant as well. Here we describe the cloning and functional characterization of Rac (Ct-Rac1), a member of the Rho family of G proteins. Ct-Rac1 expression is downregulated by DARas under limiting nutrition. Co-expression of DARas with dominant active Rac (DARac) stimulates MAPK activation and restores the wild-type phenotype. Inhibition of MAPK activation suppresses phenotypic restoration suggesting Rac-mediated MAPK activation is responsible for reversion to the wild-type phenotype. We also examined the role of reactive oxygen species (ROS) in these genetic backgrounds. The DARas mutant strain generates high levels of ROS as determined by DCFH-DA fluorescence. Co-expression with DNRac decreases ROS generation to wild-type levels and restores normal fungal growth and development. Pretreatment of DARas with antioxidants or a cytosolic phospholipase A2 inhibitor also restores the wild-type phenotype. These findings suggest that Ras-mediated ROS generation is dependent on a Rac–cPLA2-linked signalling pathway. Taken together, this study provides evidence that Rac functions to restore the hyphal morphology of DARas by regulating MAPK activation and intracellular ROS generation.

Introduction

By cycling between active GTP- and inactive GDP-bound states, small GTP-binding proteins act as molecular rheostats regulating diverse eukaryotic cellular processes ranging from cell proliferation to cellular differentiation (Barbacid, 1987; Boguski and McCormick, 1993; Symons and Takai, 2001). The Ras family has been the most studied and has served as the prototype for the small G-protein superfamily. Numerous studies have shown that signalling through Ras is essential for proper cell growth (Prendergast and Gibbs, 1993; Cox et al., 1994; Pronk and Bos, 1994; Graham et al., 2000; Khosravi-Far et al., 2001). Ras proteins mediate signalling by activation of several downstream effectors, perhaps the most notable being the classic Raf-MEK-MAPK cascade (Gauthier-Rouviere et al., 1990; Leevers and Marshall, 1992; Roberts, 1992; McCormick, 1994; Marais et al., 1995; Marshall, 1995). Recently, several Ras-dependent, Raf-independent effectors were found to be required for elicitation of a number of Ras-modulated responses (Van Aelst and D’Souza-Schorey, 1997; Bos, 1998; Campbell et al., 2000; Gupta et al., 2000; Scita et al., 2000). The Rho-GTPases (Rac, Rho and Cdc42) are important downstream targets in Ras-mediated signalling (Van Aelst and D’Souza-Schorey, 1997; Zohn et al., 1998) and cross-talk between Ras and Rac has been well documented in mammals (Scita et al., 2000).

Rac proteins belong to the Rho-like GTPase family and act to control actin cytoskeleton rearrangement and superoxide generation (Abo et al., 1991; Ridley et al., 1992; Coso et al., 1995; Minden et al., 1995; Qiu et al., 1995). Genetic and biochemical studies have indicated that the activation of Rac-dependent pathways contribute to the mitogenic and oncogenic potential of Ras (Qiu et al., 1995; Joneson et al., 1996). Expression of high levels of human oncogenic Ras (HRasV12) results in apoptosis in both primary and immortalized cells and the induction of apoptosis by HRasV12 is blocked by dominant active Rac and potentiated by dominant negative Rac (Joneson and Bar-Sagi, 1999). More importantly, it has been demonstrated that NIH 3T3 cells stably transformed with HRasV12 leads to a significant increase of intracellular reactive oxygen species (ROS) and this generation of ROS by oncogenic Ras occurs via a Rac-dependent pathway (Sundaresan, 1996; Irani et al., 1997; Cho et al., 2002). Rac was recently identified to play an important role in regulating cellular morphogenesis in the fungal kingdom. The Rac homologue YlRAC1 from the dimorphic fungus Yarrowia lipolytica, is involved in the yeast to hyphal transition, and disruption of YlRAC1 affects cell morphology and impairs hyphal growth but does not affect actin polarization or cell invasiveness (Hurtado et al., 2000); CflB, Rac homologue from the dimorphic fungus Penicillium marneffei, plays a key role in regulating the morphogenesis of both vegetative hyphae and conidiophores (Boyce et al., 2003). However, no Rac homologue has yet been described in fungal phytopathogens.

Previously, we have demonstrated that in Colletotrichum trifolii, a fungal phytopathogen of alfalfa, the unique Ras gene plays a pivotal role in regulating hyphal growth and development. Mutants expressing a dominant active form of Ct-Ras (DARas) exhibited severe defects in polarized growth, aberrant hyphal growth and significantly reduced conidiation in a nutrient-dependent manner (Truesdell et al., 1999). Moreover, DARas induced cellular transformation of mouse fibroblasts and these cells became oncogenic in nu/nu mice (Truesdell et al., 1999), suggesting Ct-Ras is functionally conserved in both mammals and fungi. Whereas much is known about Ras regulation in mammals, there is a lack of the detailed understanding of the effector pathways that couple  Ras to its up-  and downstream mediators in fungi such as C. trifolii.

In this study we sought to determine whether Rac is involved in the regulation of hyphal morphogenesis in the DARas mutant. Following the cloning of the C. trifolii Rac homologue (Ct-Rac1), yeast two-hybrid and in vitro binding assays indicated physical interaction between Rac and Ras, and ‘oncogenic’ DARas inhibited Rac expression in a nutrient-dependent manner. Moreover, we found that co-expression of dominant active Rac (DARac) in the DARas background resulted in phenotypic restoration of the wild type and involved MAPK activation. Additionally, DARas induced elevated levels of ROS which appear to contribute to the aberrant hyphal morphology, as removal of ROS by antioxidants restored normal hyphal growth. Co-expression of dominant negative Rac (DNRac) decreased ROS generation presumably through a Rac-cPLA2-dependent pathway and thereby restored wild-type hyphal morphology. The present data reveal a novel mechanism by which DARas is able to alter hyphal morphology by interfering with the Rac-dependent regulation of MAPK activation and ROS generation.

Results

Isolation of a fungal rac gene

A PCR-based library screening strategy was used to clone a Rac GTPase homologue from a wild-type strain of C. trifolii race 1. The sequenced PCR fragment was shown to have high similarity with other known Rac genes and was used to obtain full length cDNA and genomic clone as described in Experimental procedures. The open reading frame of Ct-Rac1 is predicted to encode a 200 amino acid protein with a molecular weight of 22 kDa. blast searches showed that the deduced Ct-Rac1 amino acid sequence is 87%, 78%, 77%, 76% and 76% identical to P. marneffei CflB (AF515698.1), Y. lipolytica Rac1p (AF176831.1), Ustilago maydis Rac1 (AF495535.1), Drosophila melanogaster Rac1 (NM057602.3) and Mus musculus Rac1 (A60374) respectively. Importantly, the putative Ct-Rac1 amino acid sequence contains all the consensus motifs characteristic of Rac proteins, including the GTPase region, GTP/GDP-binding region and effector binding site. Southern analysis revealed that Ct-Rac1 is a single copy gene in C. trifolii genome (data not shown).

We examined the expression levels of Ct-Rac1 during various stages of fungal growth and development. Total RNA from conidia, germinating conidia, mature appressoria and hyphae were prepared as previously described (Chen and Dickman, 2002) and Northern blots showed that Ct-Rac1 was expressed in all of these stages, but relatively higher expression was observed during vegetative growth (Fig. 1A).

Figure 1.

Characterization of Ct-Rac1 gene.
A. Developmental Northern-blot analysis of Ct-Rac1. Total RNA (20 µg lane−1) was isolated from condidia, germinating conidia, appressoria and vegetatively growing hyphae. Northern blots were probed with Ct-Rac1 cDNA and loading variations of different samples are normalized using the signal intensities of the 25S rRNA from Colletotrichum gloeosporioides as an internal control.
B. Inhibition of Ct-Rac1 expression by DARas under nutrient-limiting conditions. Mycelia were harvested from wild-type C. trifolii and DARas after growth in liquid YPSS media (left panel) or minimal media (right panel) for 6 days, and Northern blots were probed with Ct-Rac1 cDNA. Equivalent RNA loading was assessed by probing the RNA samples with 25S rDNA of C. gloeosporioides.

As mentioned, activated Ct-Ras mutation (DARas) was shown to induce aberrant hyphal morphology under nutrient-limiting conditions (Truesdell et al., 1999). Because Rac acts as a downstream effector of Ras signalling (Van Aelst and D’Souza-Schorey, 1997), we further investigated the effect of DARas on Ct-Rac1 expression. As shown in the right panel of Fig. 1B, sustained Ras signalling resulted in the inhibition of Ct-Rac1 transcription under starvation conditions; Ct-Rac1 mRNA transcripts were barely detectable in DARas-transformed cells. This result suggests that the inhibition of Ct-Rac1 expression may represent an important step in activated Ras-induced abnormal hyphal morphogenesis. Cells expressing DARas are impaired in polarized growth and exhibit an increased hyphal branching phenotype. Considering the role for Rac in actin cytoskeleton rearrangement (Bussey, 1996; Arellano et al., 1999; Hurtado et al., 2000; Boyce et al., 2003), the aberrant hyphal morphology of DARas mutant cells is consistent with the low Rac expression, as observed in these cells.

Ct-Ras interacts with Ct–Rac1

Interaction between full length Ct-Ras and Ct-Rac1 proteins was confirmed using yeast two hybrid and in vitro pull down assays. Ct-Rac1 was fused to the B42 Activation Domain (AD), Ct-Ras and DARas were separately fused with LexA DNA Binding Domain (BD). The Ct-Rac1-AD fusion protein was co-transformed with either Ct-Ras-BD or DARas-BD fusion proteins and the β-galactosidase activity was determined as described in Experimental procedures. Transformants containing empty vector, Ct-Rac1-AD, Ct-Ras-BD, or DARas-BD had very low β-galactosidase activity. By comparison, the transformants containing both Ct-Rac1-AD and Ct-Ras-BD or DARas-BD showed relatively high β-galactosidase activities as compared by the controls (Fig. 2A), suggesting a physical interaction between Ct-Rac1 and Ct-Ras as well as DARas.

Figure 2.

Ct-Rac1 physically interacts with Ct-Ras or DARas in a yeast two-hybrid system and in vitro.
A. Yeast two-hybrid assay. Single transformants and co-transformants were analysed in a liquid β-galactosidase assay and compared with each other (Clontech). One unit of activity is defined as the amount of enzyme required to hydrolyse 1 µmol of ONPG to ο-nitrophenol and d-galactose per min per cell. Each bar represents the mean of six independent transformants. EGY48 corresponds to the untransformed host strain. Transformants with more than one plasmid are separated by slashes.
B. GST pull-down assay. GST-Ct-Ras and GST-DARas (1.0 nmol each) were used to pull down L-[35S]methionine-labelled in vitro translated Ct-Rac1 (20 µl). Lane 1 contained about 20–40% of the input labelled Ct-Rac1. Lanes 2 and 3 are controls using glutathione-Sepharose beads alone and GST, respectively, for pull-downs. Lanes 4 and 5 are pull downs of Ct-Rac1 using Ct-Ras and DARas respectively.

To further evaluate whether Ct-Ras interacts with Ct-Rac1, in vitro binding assays were performed. Ct-Ras and DARas were expressed in Escherichia coli as glutathione S-transferase (GST) fusion proteins and purified using glutathione-Sepharose beads. The identity of these fusion proteins was confirmed using a GST antibody (data not shown). Purified GST-Ct-Ras and GST-DARas fusion proteins were separately incubated with an in vitro translated, [35S]-methionine-labelled Ct-Rac1 protein. Analysis of the bound fraction by SDS–PAGE and autoradiography showed that Ct-Rac1 bound to Ct-Ras and DARas (Fig. 2B, lane 4 and 5).

Morphological aberrations of hyphae by expression of dominant-active Rac (DARac)

To address the possible role of Ct-Rac1 in hyphal growth and development of C. trifolii, we initially examined the effect of the expression of a dominant active form of Rac (DARac) in C. trifolii cells by fungal transformation. Transformants expressing DARac exhibit aberrations in hyphal morphology (Fig. 3B). After 4 days the hyphae became thicker than wild type and hyphal swellings along the branches were observed in this mutant. Interestingly, unlike the DARas mutant, which also exhibits an aberrant hyphal growth in a nutrient-dependent manner, the aberrant hyphal morphology of DARac mutant was nutrient-independent. Additionally, the number of germinating conidia produced by DARac mutant was significantly reduced to approximately 37% of wild-type levels but the level of appressorium formation was essentially the same between mutant and wild type. Thus, these data suggest that similar to DARas, the DARac mutant also alters hyphal morphology and may be positioned in the same signal transduction cascade.

Figure 3.

Ectopic expression of dominant active Rac (DARac) in C. trifolii. Hyphal morphology on YPSS medium. As described in Experimental procedures, an expressing plasmid encoding DARac was transformed into wild-type C. trifolii and the hyphal morphology of transformants was assessed by microscopic observation.
A. Wild-type (untransformed) hyphae after 3 days.
B. Hyphae of DARac mutant after 3 days, note hyphal swellings along the branches (arrows). Bar = 20 µm. Picture shown is representative of 37 individual transformants.

Expression of dominant negative Rac (DNRac) abolishes hyphal growth

Next, we generated a dominant negative form of Rac (DNRac) by exchanging threonine 22 to asparagine. This construct was fused with the conditional Neurospora crassa qa-2 promoter, and introduced into wild-type C. trifolii strain. Northern analysis indicated that the expression of DNRac was markedly induced by quinic acid, whereas expression was significantly reduced and indistinguishable from that of wild type in the presence of glucose (data not shown). When grown in minimal media containing 15 mM glucose, in which qa-2(p)-driven expression of DNRac is repressed, the transformants were indistinguishable from the wild-type (Fig. 4, panel 2); When grown in minimal media containing 15 mM quinic acid to induce expression of DNRac, hyphal growth of these transformants was abolished (Fig. 4, panel 1). The hyphal growth of the wild-type strain in minimal media containing 15 mM quinic acid (Fig. 4, panel 4) was similar as in rich media (Fig. 4, panel 3) but exhibited slower growth. Together with the aforementioned results, we conclude that control of Rac activity appears to be a critical factor in regulating proper hyphal growth and development in C. trifolii.

Figure 4.

Inhibition of hyphal growth by dominant negative Rac (DNRac). Transformant expressing DNRac was grown for 7 days at room temperature in minimal media supplemented with 15 mM quinic acid (MM + QA, panel 1); minimal media supplemented with 15 mM glucose (MM + Glu, panel 2). Controls are the wild-type strain grown in complete media (YPSS, panel 3) and minimal media supplemented with 15 mM quinic acid (MM + QA, panel 4). Picture shown is the representative of 16 individual transformants.

Co-expression of DARac in a DARas background restores wild-type hyphal morphology

To obtain further insight into the signalling mechanisms by which DARas mediates aberrant hyphal morphology under nutrient-limiting conditions, we examined the effect of cotransforming DARac in the DARas mutant. As shown in Fig. 5, DARac significantly suppresses the DARas-induced aberrant hyphal morphogenesis. Thus, the double mutant (DARasRac) exhibits wild-type hyphal growth characteristics under nutrient-limiting conditions (Fig. 5C). Moreover, we found that DARasRac mutant increased sporulation rates (>20%) compared with the extremely low sporulation rate of DARas (<2%), but less than wild type (>90%) (unpubl. data).

Figure 5.

Co-expression of DARac in DARas mutant leads to phenotypic restoration. Cells expressing DARas were co-transformed with an expression vector encoding DARac. The double mutant (DARasRac) was grown in minimal media for 6 days and the morphology of hyphae was observed with light microscopy. Note that DARas-expressing cells caused curled and distorted hyphal morphology (B), whereas DARasRac double mutant restored the wild-type-like hyphal growth (C). The hyphal morphology of the wild-type strain was shown in A. At least 25 double mutants were assessed in this experiment. Bars, 50 µm.

Recent studies have shown that proper function of MAPK- and Rac-dependent pathways is required for efficient mitogenesis or transformation by activated Ras (Qiu et al., 1995; Joneson et al., 1996; Rodriguez-Viciana et al., 1997; Yoo et al., 2001). Thus we reasoned that MAP kinase activity may be involved in the restoration of normal hyphal morphology in the double mutant. Indeed, DARasRac mutant cells pretreated with PD98059, a specific MAP kinase kinase (MEK) inhibitor (Zhang et al., 1997), reverted to aberrant hyphal morphology (Fig. 6A, panel c). As controls, MEK inhibitor treatment had no effect on hyphal morphology of DARas (Fig. 6A, panel d) but severely suppressed hyphal elongation of both DARac mutant (Fig. 6A, panel e) and wild type (Fig. 6A, panel f). The levels of phosphorylated MAPK (active MAPK) after inhibitor treatment were undetectable by immunoblotting (data not shown).

Figure 6.

Active MAP kinase phosphorylation is required for phenotypic restoration in DARasRac double mutant.
A. PD98059 inhibits normal hyphal growth and promotes aberrant hyphal morphology in double mutant. Panel a, DARas mutant were cultured in minimal media for 6 days; Panel b, DARasRac double mutant was cultured in minimal media without PD98059 (10 µM) for 6 days; Panel c, DARasRac double mutant was cultured in minimal media amended with PD98059 (10 µM) for 6 days; Panel d, DARas mutant cells were cultured in minimal media amended with PD98059 (10 µM) for 6 days; Panel e, DARac mutant was cultured in minimal media amended with PD98059 (10 µM) for 6 days; Panel f, wild-type C. trifolii strain was cultured in minimal media amended with PD98059 (10 µM) for 6 days. The hyphae were stained with calcofluor white (1 mg ml−1) to visualize cell walls. All results represent the average of at least three separate experiments using independent transformants. Bars, 20 µm.
B. Rac activates the MAP kinase pathway in DARasRac double mutant. Six days after growth in complete media, hyphae of different strains were harvested by filtration and washed three times with sterilized distilled water and then transferred to minimal media for another 6 days. Cultures from complete (upper panel) or minimal media (lower panel) were harvested and cell lysates were analysed by means of Western blots using an antibody that specifically recognizes the dually phosphorylated (active) forms of ERK1/2 or an anti-ERK2 antibody for equivalent sample loading. Lane 1, DARas mutant; Lane 2, DARac; Lane 3, DARasRac double mutant; Lane 4, DNRac; Lane 5, DARasDNRac double mutant; Lane 6, wild-type C. trifolii race 1. Note that lane 4 was not shown in minimal media because of the abolishment of hyphal growth in DNRac mutant. The data are representative of two independent experiments.

Because the MEK inhibitor impairs wild-type-like hyphal morphology in DARasRac mutant, activation of MAP kinase may therefore account, at least in part, for the maintenance of wild-type-like hyphal growth in this double mutant. Previously, we applied a nested-PCR to isolate a MAPK homologue from C. trifolii genome and the DNA sequence of a 339 bp PCR product showed a high degree of identity with mammalian ERKs including the highly conserved MAPK phosphorylation site (TEY) (unpubl. data). Therefore, we examined MAPK activity of different strains grown in different nutritional conditions using an antibody that selectively recognizes phosphorylated, active forms of ERK-1 and -2. As shown in Fig. 6B, when grown in rich media, sustained MAPK activities were observed in each strain including wild type, DARas, DARac and DARasRac double mutant. However, after 7 days growth in minimal media, MAPK activities varied in each strain. The wild-type C. trifolii strain still maintained high MAPK activity (lane 6); Expression of DARas alone significantly inhibited MAPK phosphorylation (lane 1). Consistent with the wild-type-like phenotype, DARasRac double mutant had equivalent levels of MAPK activity (lane 3) compared with wild-type strain indicating that activation of MAPK correlates with normal hyphal morphology and overexpression of DARac overcomes the inhibitory effect of DARas and thereby induces MAPK activation. In accordance with this observation, we found that expression of DARac alone induces MAPK activation (lane 2) and co-expression of DNRac with DARas (DARasDNRac) significantly reduced MAPK phosphorylation (lane 5), further supporting the notion that activation of Ct-Rac1 correlates with MAPK activation. Concentrations of total MAPK were equivalent in all fungal strains, as detected by anti-ERK2 antibody.

Taken together, these results suggest that activation of MAPK in DARasRac mutant is dependent on signals generated through DARac, which is a downstream mediator of Ras. Although Ct-Rac1 expression was inhibited by DARas, overexpression of DARac bypassed the suppressive effect and induced MAPK activation, and restored wild-type phenotype. Consistent with our results, recent studies in mammalian cells have shown that signals from Rac converge on the Raf-dependent MAPK pathway and cross-talk between Raf- and Rac-dependent pathways is required for mitogenesis (Frost et al., 1997; Tang et al., 1999; Sun et al., 2000; Ingram et al., 2001). Because the MEK inhibitor reverted the wild-type phenotype of DARasRac mutant, and inhibited MAPK activation, presumably effectors upstream of ERK, for example, a fungal Raf or MEK homologue is the target of Rac leading to activation of MAPK.

DARas induces ROS production in C. trifolii

Based on the available evidence, inhibition of MAPK is involved in the aberrant hyphal phenotype in DARas mutant. We therefore reasoned that activation of other effector(s) may also be involved in the hyphal distortion observed in DARas mutant. We considered the possible involvement of reactive oxygen species (ROS), which have been recognized as important regulators of stress responses in many cell types and are known to be downstream effectors of Ras (Irani et al., 1997). Coupled with the functional conservation between Ct-Ras and human Ras, we examined whether or not cells expressing DARas induce production of ROS. Protoplasts generated from DARas cells were loaded with the fluorophore DCFH-DA and the resultant ROS generation was monitored as a function of H2O2-sensitive DCF fluorescence by visualization of the DCF-positive cells by fluorescence microscopy. We found that ROS generation differed between wild-type and DARas cells, as expression of DARas resulted in a significant increase in the intracellular fluorescence (Fig. 7B).

Figure 7.

Enhanced ROS production by DARas accounts for aberrant hyphal morphology under nutrient-limiting conditions. Aliquots of protoplasts were stained with 2′,7′-dichlorodihydrofluorescein diacetate (DCHFDA) to visualize the amount of H2O2 within the cells, as described in Experimental procedures. Figures shown are representative of three independent experiments. A, wild type;, DARas mutant; C, DARac mutant; D, DARasRac double mutant; E, DARasDNRac double mutant; F, DARas mutant treated with 1 µM mepacrine; G, DARas mutant treated with 1 mM NAC; H, DARas mutant treated with 25 µM DPI.

Enhanced production of ROS results in aberrant hyphal morphology in DARas under nutrient-limiting conditions

To investigate whether enhancement of the ROS generation induced by DARas is associated with aberrant hyphal morphology, we assessed the effect of H2O2 on hyphal morphogenesis in the wild-type strain. Spores pretreated with different concentrations of H2O2 were inoculated onto sterile glass microscope slides containing a thin layer of YPSS agar. We found that similar to hyphal morphology of DARas mutant under nutrient-limiting conditions, wild-type C. trifolii strain treated with up to 1 mM H2O2 also resulted in hyperbranched hyphae (Fig. 8, panel 3). Higher levels of H2O2 were found to completely inhibit hyphal growth (data not shown).

Figure 8.

Effect of exogenous H2O2, antioxidants, and DNRac expression on hyphal morphogenesis. Panel 1, wild-type strains were grown for 6 days in minimal media; Panel 2, DARas mutants were grown for 6 days in minimal media; Panel 3, spores of wild-type strains were pretreated with 1 mM H2O2 for 20 min and then inoculated into microscopic slides overlaid with a thin layer of complete media for another 6 days at room temperature; Panel 4, DARas mutants were grown for 6 days on minimal media supplemented with 1 mM N-acetyl-L-cysteine (NAC); Panel 5, DARas mutants were grown for 6 days on minimal media supplemented with 25 µM Diphenyleneiodonium (DPI); Panel 6, DARasDNRac double mutant was grown for 6 days on minimal media. Cell morphologies were observed with light microscopy. Pictures shown are representative of three independent experiments. Bars, 50 µm.

We also treated DARas mutant with the membrane-permeant antioxidant N-acetyl-L-cysteine (NAC) and the NADPH-oxidase inhibitor diphenylene iodonium (DPI). Both NAC (1 mM) and DPI (25 µM) reverted the aberrant hyphal morphology to a wild-type-like hyphal morphology (Fig. 8, panels 4 and 5). In accordance with these observations, treatment of DARas with NAC (Fig. 7G) or DPI (Fig. 7H) completely prevented DCF fluorescence. These findings strongly support the idea that abnormal intracellular ROS generation induced by DARas is required for aberrant hyphal morphology in C. trifolii. Thus, hyphal morphogenesis in DARas is regulated in a ROS-dependent manner.

ROS generation in DARas is through a Rac-linked pathway

To determine in more detail how ROS is generated in the DARas mutant, DNRac was constitutively co-expressed in DARas and hyphal morphology of this double mutant (DARasDNRac) was examined by microscopy. DARasDNRac mutant, when grown in minimal media, exhibited a complete reversion of hyphal phenotype (Fig. 8, panel 6). In line with this observation, this double mutant showed a significantly weaker DCF-fluorescence than that from cells transformed with DARas alone (Fig. 7, panel E). In contrast, DARac mutant showed increased fluorescence signals (Fig. 7C). Treatment of these fungal strains with antioxidant drugs completely suppressed ROS generation (data not shown). Thus, these results indicate that enhancement of intracellular ROS in DARas is generated via Rac expression in an analogous manner to that found in mammalian cells.

A putative cPLA2 homologue in C. trifolii as a downstream effector of Ras-mediated ROS generation

Previous studies in fibroblasts have shown that cytosolic phospholipase A2 (cPLA2) serves as a major downstream mediator of Rac to stimulate the generation of intracellular ROS (Peppelenbosch et al., 1995; Kim et al., 1997; Woo et al., 2000). We speculated that a cPLA2 homologue in C. trifolii might contribute to ROS generation by functioning as a downstream target of Ct-Rac1. To test this possibility, we examined whether cPLA2 inhibition affects the hyphal morphology of DARas. As expected, treatment of DARas with mepacrine, a potent cPLA2 inhibitor (Kim et al., 1997), caused a significant morphological reversion of abnormal hyphal growth in a dose-dependent manner, with normal polarized hyphal elongation and branching (Fig. 9C). Mepacrine had little effect on the morphology of wild-type strain (Fig. 9A). Thus, a conserved Rac-cPLA2-linked pathway might be operative and play a role for signalling via DARas. Taken together, these findings support the notion that a Ras-Rac-cPLA2-dependent pathway mediates intracellular ROS production and accounts for the morphological changes induced by DARas.

Figure 9.

Treatment of DARas mutant with a specific cPLA2 inhibitor (mepacrine) restores the proper hyphal morphology. Spores (10−7–10−8 ml−1) of DARas mutant were inoculated into minimal media supplemented with 1 µM mepacrine at room temperature for 6 days. The morphology of the cells was visualized by light microscopy.
A. Wild-type strain grown in minimal media supplemented with 1 µM mepacrine.
B. DARas mutant grown in minimal media.
C. DARas mutant grown in minimal media supplemented with 1 µM mepacrine. Pictures are representative of three independent experiments. Bars, 20 µm.

Rac-mediated MAPK pathway operates in parallel with Rac-mediated ROS pathway in C. trifolii

To determine if cross-talk between Rac-mediated MAPK activation and Rac-mediated ROS generation occurred, the generation of ROS was examined in DARasRac double mutant. Co-expression of DARac did not induce significantly higher ROS production as determined by DCF-fluorescence in DARasRac mutant (Fig. 7D). We found that ROS levels were slightly decreased when Ct-Rac1 was dominant active in DARas mutant. Presumably, an important role of DARac in the DARas background, is to induce activation of MAP kinase and dampen ROS generation, possibly by activating an unidentified effector-mediated signalling cascade. Moreover, treatment of DARas with antioxidants had no effect on MAPK activation (unpubl. data). Thus, these results suggest that two Rac-mediated signal transduction pathways (MAPK-dependent and ROS-dependent) may function in parallel in C. trifolii for proper hyphal growth and development.

Discussion

Hyphal morphogenesis is a complex developmental programme that accounts for the foraging success of filamentous fungi. Coordinated control of hyphal growth and development in filamentous fungi is achieved by activating intracellular signalling cascades but many of the components regulating hyphal morphogenesis are still not well characterized. Previously, studies in the filamentous fungus Colletotrichum trifolii have shown that Ras plays a pivotal role in regulating hyphal growth and development (Truesdell et al., 1999). Constitutively active DARas resulted in abnormal hyphal morphology, but only under conditions of nutrient limitation. This was reflected by increased hyphal branching, loss of polarized growth, and reduced conidiation, suggesting that mutant Ras has pleiotropic effects on the morphology and growth behaviour of C. trifolii. However, the mechanism by which DARas induces aberrant hyphal morphology has remained largely unknown. Several studies (Bar-Sagi and Hall, 2000) have suggested that besides the Raf-MEK-MAPK pathway, the small Rho GTPases including Rac and Rho contribute to the full Ras biological response. Based on this premise, we show that in C. trifolii, Rac is important in mediating the DARas phenotype and does so via at least two distinct mechanisms. The major findings of this study are: (i) inhibition of MAPK activation by DARas contributes to aberrant hyphal morphology in C. trifolii; (ii) co-expression of DARac induces MAPK activation and results in wild-type phenotypic restoration; (iii) stimulation of ROS generation via a Ras/Rac/cPLA2-dependent pathway is required for the maintenance of aberrant hyphal morphology in DARas; (iv) the Rac-mediated MAPK cascade operates in parallel with Rac-mediated ROS generation. To our knowledge, the findings of this study are the first evidence indicating that Rac serves as a Ras effector to control hyphal growth and development by regulating a MAPK-dependent and a ROS-dependent pathway in filamentous fungi.

Numerous reports (Van Aelst and D’Souza-schorey, 1997; Zohn et al., 1998) indicate that Rac is a key downstream target in Ras-mediated signalling. Functional studies (Rodriguez-Viciana et al., 1997) established that phosphatidylinositol 3-kinase (PI3-K) is often positioned downstream of Ras and upstream of Rac. However, in C. trifolii, our results obtained from yeast two-hybrid and in vitro-binding assays suggest a physical interaction between Ras and Rac. Our data cannot exclude the presence of PI3-K in Ras signalling and it is possible that Ras, Rac, and PI3-K, form a ternary complex in C. trifolii. Importantly, endogenous Rac expression was inhibited by DARas under nutrient-limiting conditions. In eukaryotes, an important role of Rac proteins is to regulate cellular morphogenesis via actin cytoskeleton reorganization and polarized growth (Johnson, 1999). For example, Rac proteins in fibroblasts were found to regulate formation of lamellipodia, a structure with actin-rich cell extensions. Lamellipodia formation was induced by dominant active forms of Rac but inhibited by dominant negative forms of Rac (Nobes and Hall, 1995). More recently, a Rac homologue (cflB) from the dimorphic fungus Penicillium marneffei was shown to be required for cellular polarization during hyphal development. Deletion of cflB resulted in loss of polarized growth, inappropriate septation, distorted actin cytoskeleton and lower germination rates. A dominant negative form of cflB showed a similar phenotype, albeit less severe (Boyce et al., 2003). Similar to the morphological effects on Rac deletion in P. marneffei, mutationally active DARas in C. trifolii also induced abnormal hyphal proliferation, defects in polarized growth and significantly reduced differentiation under conditions of nutrient deprivation. Thus, these data strongly suggest that inhibition of Ct-Rac1 expression in DARas is likely to account for the aberrant cellular morphology and that Rac may play a key role in the regulation of proper hyphal growth and development in C. trifolii. Consistent with this idea, DNRac abolished hyphal growth and development, whereas DARac resulted in thicker and swelling hyhae followed with significantly reduced germination. These findings further indicate that similar to P. marneffei, inappropriate function of Rac in C. trifolii severely affects hyphal morphogenesis.

Accumulating evidence (McCormick, 1994; Marshall, 1995) in mammalian systems has demonstrated that the classic Ras/Raf/MEK/ERK pathway operates in parallel with Ras/Rac signalling. Rac also induces Raf or MEK to activate ERK and thereby mediates cellular transformation (Frost et al., 1996; 1997; Joneson et al., 1996; Tang et al., 1999; Bar-Sagi and Hall, 2000; Stephens and Hawkins, 2000; Clerk et al., 2001; Ingram et al., 2001; Allen et al., 2002). However, the majority of these studies were conducted in animal cells, and a clear example of the physiological relevance of the Rac-mediated ERK activation in filamentous fungus has not been established.

As described in Fig. 6, DARas inhibits MAPK under nutrient-limiting conditions. It is therefore possible that inhibition of MAPK activation may be causal to the aberrant hyphal growth observed in DARas. If so, an active MAPK pathway may be required for normal hyphal growth and development. This hypothesis was confirmed by our studies. Indeed, we found in double mutant DARasRac that constitutive expression of DARac resulted in a significant increase of ERK-like MAPK phosphorylation up to the same levels detected in wild type with concomitant restoration of wild-type-like hyphal growth. Moreover, treatment of this double mutant with PD98059, a pharmacological inhibitor of MAP kinase kinase (MEK), resulted in abnormal hyphal morphology. Thus, on the basis of these data, we propose that co-expression of DARac ameliorates the Ras-mediated suppressive effect and converges on the Raf/Mek pathway to activate MAPK which is coupled to phenotypic restoration. It is worthwhile to note that active MAPK signalling is required for proper hyphal growth and development in C. trifolii and that MAPK activation in this context depends on signals from Ct-Rac1 converging on the Ras/Raf/Mek/MAPK cascade. These data are consistent with recent reports that NIH3T3 fibroblast cells transformed with the human HRasV12 (analogous to DARas), significantly decrease p44/42 MAPK phosphorylation and MAPK activity (Irani et al., 1997) and that Rac-PAK signalling stimulates ERK activation by regulating formation of MEK1-ERK complexes (Eblen et al., 2002).

In addition to regulation of actin polymerization and activation of MAPK, Rac is also a known regulator of ROS generation by regulating the catalytic activity of the NADPH oxidase complex (Bokoch, 1994; Bishop and Hall, 2000). A number of previous studies have demonstrated that overexpression of oncogenic Ras produced elevated intracellular ROS in NIH3T3 cells (Irani et al., 1997), human keratinocyte HaCaT cells (Yang et al., 1999), and human lung Wi-38VA-13 cells ( Liu et al., 2001). Because our previous studies have shown that DARas functions as an oncogene in NIH3T3 cells (Truesdell et al., 1999), it is reasonable to speculate that expression of DARas in C. trifolii induces ROS generation which may directly involve Rac. As expected, we observed that in C. trifolii, DARas cells significantly enhanced ROS production and that this enhancement of ROS generation was inhibited by co-transformation of DNRac (DARasDNRac) as well as by treatment with two antioxidants. Moreover, our results  showed  that  expression  of  DARac  stimulated ROS production. Thus, the enhancement of ROS generation by DARas is mediated by a Ras/Rac-dependent pathway.

The correlation of aberrant hyphal morphology with activated Ras and with elevated ROS production led us to hypothesize that DARas induced aberrant hyphal morphology through enhancement of intracellular ROS generation under nutrient-limiting conditions. In agreement with this is the observation that treatment of DARas with antioxidants (DPI and NAC) restored normal hyphal morphology and co-expression of DNRac resulted in phenotypic restoration by decreasing ROS production. Interestingly, the inhibitory effect of DPI, a flavoprotein inhibitor, suggests that a flavoprotein-binding protein similar to the phagocytic NADPH oxidase may play role in the signalling pathway resulting in phenotypic restoration of DARas mutant. Moreover, we found that treatment of DARas with mepacrine, a potent inhibitor of cytosolic phospholipase A2 (cPLA2), significantly reduced ROS generation and thereby restored the normal hyphal growth. cPLA2 has been known to function as a major downstream mediator of Rac signalling. When activated, Rac activates cPLA2 followed by the generation of ROS (Woo et al., 2000). Taken together, our present work has demonstrated that in addition to MAPK inactivation, increased ROS generation mediated by the Ras/Rac/cPLA2 pathway accounts, at least in part, for the aberrant hyphal morphology of DARas mutant.

There is a growing body of evidence that ROS-induced MAPK phosphorylation and activation in various cells such as vascular smooth muscle cells and NIH3T3 cells are involved in the regulation of cell growth and differentiation (Stevenson et al., 1994; Baas and Berk, 1995). However, our data are contrary to these observations. In C. trifolii, despite enhanced intracellular ROS generation, cells expressing DARas exhibit inactivation of MAP kinase under conditions of nutrient deprivation. We therefore propose that enhancement of ROS generation in DARas activates intracellular signalling pathway(s) that are MAPK-independent. In support of this, we observed that in DARasDNRac double mutant, co-expression of DNRac did not induce MAP kinase activation but did rescue the normal hyphal morphology. Treatment of DARas with antioxidants did not interfere with its suppressive effect on MAP kinase activation (unpubl. data). Thus, this work is consistent with a model described in Fig. 10 where two Rac-linked pathways, ROS-dependent and MAPK-dependent, may function in parallel to regulate hyphal growth and development in C. trifolii. A possible explanation for this type of regulation may centre on the compartmentalization of the two distinct signals.

Figure 10.

Model showing the dual role of Rac in regulating hyphal growth and development in C. trifolii. The model outlines the role of Rac-mediated MAPK activation and ROS generation in Ras signalling and in hyphal morphogenesis. MAP kinase activation is inhibited by DARas whereas constitutive co-expression with DARac converges on Raf or Mek to activate MAP kinase and thus restores wild-type-like hyphal morphology. Reactive oxygen species (ROS) which are enhanced in the DARas mutant, accounts for the aberrant hyphal morphology under conditions of nutrient limitation, whereas co-expression of DNRac significantly reduces intracellular ROS generation and leads to phenotypic restoration, possibly through a Ras-Rac-cPLA2-dependent signalling pathway.

Our studies demonstrate that Rac-mediated ROS production plays a regulatory role in the cellular signal transduction pathway involved in hyphal growth and development in C. trifolii. However, the signalling components targeted by ROS are poorly understood. Recently, enhancement of intracellular ROS generation was shown in other eukaryotic systems to be functionally associated with the regulation of gene expression and the activation of transcription factors including NF-κB, Sp1, Ref-1, AP-1, Nrf2, and p53 (Ishii et al., 2000; Thannickal and Fanburg, 2000; Ozaki et al., 2002). Thus, it is likely that increased ROS generation in DARas may induce aberrant hyphal morphology through the activation of redox-sensitive transcription factors. Identification of these ROS-regulated transcription factors in future work will provide new insights into Ras signalling mechanism and regulation of hyphal growth and development in C. trifolii.

Experimental procedures

Reagents

Hydrogen peroxide (H2O2), mepacrine, N-acetyl-L-cysteine (NAC), diphenyleneiodonium (DPI) and Calcofluor white were purchased from Sigma. 2′,7′-dichlorofluorescein diacetate (DCF-DA) was from Molecular Probes (Eugene, OR). Monclonal mouse anti-phosphorylated ERK1/2 antibody and polyclonal rabbit anti-ERK2 antibody were both purchased from Santa Cruz Biotechnology (Santa Cruz, CA). PD98059 was obtained from Calbiochem (San Diego, CA). All other chemicals were from standard sources and were molecular biology grade or higher.

Strains, media and growth conditions

The C. trifolii strains used in this study are listed in Table 1. Colletotrichum trifolii cultures were routinely grown at 25°C on yeast extract-phosphate-soluble starch (YPSS) agar media (Yang and Dickman, 1997). For the isolation of protoplasts, DNA, RNA, and protein, mycelia were grown in stationary liquid YPSS for 3–7 days. Conidia, germinating conidia, appressoria and mycelia were collected as described previously (Yang and Dickman, 1999). Different culture conditions were used as indicated in the text: rich media (YPSS), minimal media (water agar only, MM), quinic acid (QA)-containing minimal media (QA-MM) (Warwar et al., 2000). Escherichia coli strain DH5α (Invitrogen, Carlsbad, CA) was used for the plasmid DNA transformation.

Table 1.  Fungal strains used in this study.
StrainPlasmidDescriptionReference or source
Colletotrichum trifolii Wild-type C. trifolii race 1Dickman (1988)
DARasCt-RasVal2Dominant active Ct-Ras mutantTruesdell et al. (1999)
DARacCt-RacG17VDominant active Ct-Rac mutantThis study
DNRacCt-RacT22NDominant negative Ct-Rac mutantThis study
DARasRacCt-RasVal2 + Ct-RacG17VDouble mutant overexpressing dominant active Ras (DARas) and dominant active Rac (DARac)This study
DARasDNRacCt-RasVal2 + Ct-RacT22NDouble mutant overexpressing dominant active Ras (DARas) and dominant negative Rac (DNRac)This study

PCR cloning of Ct-Rac1

Two degenerate primers, RAC-F: 5′-GGIGAYGGNGCNGT NGGNAA-3′, RAC-R: 5′-AYTCYTCYTGICCNGCNGTRTC-3′ (Y = C/T, R = G/A, N = G/A/C/T, I = inosine), were designed according to two conserved amino acid motif, GDGAVGK and DTAGQEEY, found in all known Rac proteins. First-strand cDNA was generated from polyadenylated mRNA prepared from a mycelial culture and used as template. Polymerase chain reaction was conducted with the initial five cycles at a low annealing temperature (1 min, 94°C; 1 min, 45°C; 1 min, 72°C), whereas the last 30 cycles were at a higher annealing temperature (1 min, 94°C; 1 min, 52°C; 1 min, 72°C). Polymerase chain reaction products were cloned in TOPO TA vector (Invitrogen, Carlsbad, CA) and sequenced. A PCR clone having homology with known Rac genes was used to screen a C. trifolii genomic library and a C. trifolii cDNA library prepared with RNA from vegetative mycelia. The GenBank accession No. for Ct-Rac1 is AY325889.

Molecular constructs and fungal transformation

Site-directed mutagenesis was used to generate mutant versions of C. trifolii Rac (DARac and DNRac) by Taq PCR-mediated DNA amplification. DARac was generated by substitution of the glycine 17 (G17) of C. trifolii Ct-Rac to valine. Similarly, DNRac was generated by substitution of threonine 22 (T22) to asparagine. All mutagenized DNA fragments were amplified with pfu polymerase (Stratagene, La Jolla, CA) and sequenced. Expression of DARac was driven by the constitutive glyceraldehyde 3-phosphate dehydrogenase (gpd) promoter from pNOM102 (Roberts et al., 1989). Expression of DNRac was driven by the conditional Neurospora crassa quinic acid (qa-2) promoter from pSO-1 (obtained from Dr Oded Yarden). For selection, the hygromycin B phosphotransferase gene expression cassette from a derivative of pHA1.3 (Powell and Kistler, 1990) or the bleomycin resistance gene expression cassette from pAMPH520 (Austin et al., 1990) was subcloned in the two constructs described as above. PEG-mediated fungal transformation was performed as previously described (Yang and Dickman, 1999).

Yeast two-hybrid system

Ct-Rac1 cDNA was cloned into the prey expression vector, pLexA, and Ct-Ras and DARas cDNAs were cloned into the bait expression vector, pB42AD. Using yeast strain EGY48, which harbours a lacZ reporter plasmid, p8op-lacZ, transformation was carried out with 0.1 µg of each expression plasmid in a solution containing 40% polyethylene glycol, 0.1 M lithium acetate (pH 7.5), 10 mM Tris-HCl (pH 7.5), 1 mM EDTA (pH 8.0), and 50 µg of denatured salmon sperm DNA and cultured on SD/Gal/Raf/–His/–Ura/–Trp. A minimum of three independent transformants were streaked and tested for the transactivation of the reporter gene lacZ using a ο-nitrophenyl-β-D-galactopyranoside (ONPG) assay, as described by the manufacturer (Clontech, Palo Alto, CA).

In vitro translation binding assay

In vitro translation of radiolabelled Ct-Rac proteins was performed using a TNT T7 coupled wheat germ extract system (Promega, Madison, WI) in the presence of [35S]-methionine (20 µCi/reaction). Escherichia coli BL21 cells were transformed with plasmids encoding the GST-Ct-Ras and GST-DARas fusion proteins or GST alone. The GST-Ras fusion proteins were purified on glutathione agarose beads (Molecular Probes, Eugene, OR) using standard techniques and then incubated in binding buffer (50 mM Tris-Cl pH 7.4, 10 mM EDTA, 150 mM NaCl) with radiolabelled in vitro translated Ct-Rac1 proteins for 2 h at 4°C. After three washes with binding buffer, bound proteins were eluted by boiling in protein sample buffer, resolved by SDS–PAGE and visualized by Western blotting.

Detection of intracellular H2O2

Intracellular H2O2 was monitored by 2′,7′-dichlorodihydrofluorescein diacetate (DCHF-DA) fluorescence (Ohba et al., 1994). DCFH-DA is lipophilic and thus cell permeable, where it forms polar, non-fluorescent 2′,7′-dichlorofluorescin (DCFH). When ROS are generated, particularly H2O2, DCFH is oxidized to 2′,7′-dichlorofluorescein (DCF), whose green fluorescence is easily observed by fluorescence microscopy. The amount of DCF formed is proportional to the cellular oxidant production. To monitor intracellular H2O2 in C. trifolii, protoplasts from fungal vegetative hyphae were produced as described previously (Yang and Dickman, 1999), stabilized in STC solution, and then incubated for 20 min at room temperature with the H2O2-sensitive fluorophore DCFH-DA (50 µM). The protoplasts were harvested, suspended in STC solution, and applied to a microscopic slide for observation under a Zeiss Axioskop microscope using an immunofluorescent DCF-specific filter. To assess inhibitor effects, protoplasts were pretreated with the respective inhibitor for 30 min before the same procedures.

Western blotting

The isolation of mycelial proteins and Western blotting were performed as described previously (Chen and Dickman, 2002). All solutions used for the preparation of protein extracts were supplemented with fungal protease inhibitor cocktail (Sigma, St Louis, MO). Protein concentration was determined by the method of Bradford (Bradford, 1976), and 10 µg of protein/lane was electrophoresed on a 12% SDS–PAGE after boiling for 3 min in sample buffer. Proteins were transferred to nitrocellulose membranes (Amersham Biosciences, Piscataway, NJ). After electroblotting, the membranes were blocked at 37°C with PBS-Tween containing 5% milk and 1%BSA for 1 h and then incubated for 1 h at room temperature with mouse anti-phosphorylated ERK1/2 primary antibody (1 : 1000). The secondary antibody (1 : 10 000) was anti-mouse Ig conjugated horseradish peroxidase (Pierce, Rockford, IL). Proteins were visualized by enhanced chemiluminescence (Supersignal west pico chemiluminescent system, Pierce). Rabbit anti-ERK2 antibody (1 : 10 000) was used to confirm equivalent loading.

Microscopy

Sterile glass microscope slides were overlaid with a thin (approximately 2-mm) layer of molten agar medium and placed in sterile glass Petri dishes. Spores of the fungal strains diluted with a concentration of 104−105 ml−1, were inoculated onto the slides, and the Petri dishes were incubated in a plastic box lined with water-soaked paper towels to ensure a humid environment. After 2 days of incubation, a coverslip was placed over the resulting hyphae for another 5–6 days of growth. The hyphae were viewed on a Zeiss microscope with differential interference contrast (DIC) optics. Images were captured visa a charge-coupled device camera and processed using AxioVison 3.1 for Windows (AxioCam HR, Thornwood, NY).

Acknowledgements

We thank Steve Harris for a critical reading of this manuscript and helpful comments. This work was supported by the United States-Israel Binational Agriculture Research and Development Fund BARD no. 2814–96.

Ancillary