Mutants in genes encoding conserved eukaryotic signal transducing proteins have been very helpful in efforts to understand the environmental mediated control of development and the sensory pathways needed to detect the host and establish invasive growth. Several such mutants have been constructed for the maize pathogen Cochliobolus heterostrophus, agent of Southern corn leaf blight [1-4]. In Ascomycetes for which sufficient sequence information is available, there are three Gα encoding genes, one Gβ and one Gγ gene. Deletion of the MAP kinase gene CHK1 [3] or the Gβ gene CGB1 [1] has a vast effect on growth and development and is drastically reducing virulence under all conditions tested. Mutants in CGA1, a heterotrimeric G protein Ga subunit, produce conidia that germinate as abnormal, straight-growing germ tubes forming few appressoria [5]. Nevertheless, these mutants can cause normal lesions on plants, unlike other filamentous fungal plant pathogens in which functional homologues of CGA1 are required for full virulence [2]. This demonstrates that appressorium formation is not essential for virulence. Indeed, inoculation with mycelium results in growth on the leaf surface followed by penetration into the leaf without noticeable appressorium formation: sometimes aggregates of mycelia localize to stomatal apertures, but it seems that direct penetration of the epidermis is also possible. Detailed examination indicated that under some host physiology conditions, CGA1 disruption and deletion mutants are considerably less virulent [6]. In addition disruption of the CGA1 gene causes aerial growth formation and spores aggregation that indicates a possible role for CGA1 in regulation of hydrophobin secretion [5]. Determination of Cochliobolus heterostrophus hydrophobins expression in cga1 mutants provided the molecular evidence for the role of CGA1 in suppression of hydrophobins expression [5].

Although, Gene disruption studies are an efficient way to identify the role of signaling components such as the G-protein subunits and the MAPK cascade, a constant activation of desired genes became, in recent years, a powerful genetic tool to accomplish the information resulting from the disruption experiment and to identify new functions. Site specific mutagenesis (such as Q204-L, G42-R and R178-C) designed to constitutively activate Gα signaling was reported in C. parasitica [7], M. grisea [8], N. crassa [9] and in A. nidulans [10]. Conversion of one of these amino acids abolishes GTPase activity, which in turn would constitutively activate G protein signaling. Here we constructed a constitutively activated Gα allele (cga1^Q204L) to investigate the role of CGA1 in developmental processes. In particular, we examined its influence on hydrophobin assosiated traits.

cga1^Q204L construction and verification. In order to construct a constitutively activated CGA1 mutant we used site-directed mutagenesis (Q204L mutation) to change glutamine to leucine at position 204. A plasmid carrying the CGA1 gene followed by the Bar resistance cassette (pCGA1-Bar, Figure 1(a)) was used as a template for Site-directed mutagenesis (Figure 1(a), black frame) and afterward for cloning in E. coli cells. The mutagenesis and the cloning success were verified by PCR amplification of the CGA1 mutated gene and by sequencing. Vectors containing the Q204L mutation were prepared from the plasmid using two unique restriction sites Apa1 or Bstx1 (6 kb linear fragment) or HindIII that excluded the pBluescript vector (3 kb linear fragment) (Figure 1(b)).

The plasmid was then transformed into two strains of C. heterostrophus, WT and cga1. The mutated colonies were grown on selective media (CM-Bar) for several transfers. DNA extracted from both strains, cga1 and the WT, was obtained by PCR amplification. Mutant in the background of cga1 was verified by amplifying the transformed sequence using the forward Cga1 B4xho primer, and the Cga1-r reverse primer (Figure 1(a)). Since the cga1 mutation was created by deleting 18 bp upstream to the CGA1 promoter after Xho restriction site and insertion of a hygromicin cassette instead, an unsuccessful transformation will result in the presence of a sole copy of the disrupted CGA1 gene. So a PCR reaction with these primers will result in a 2854 bp product (772 bp CGA1 minus 18 bp of the coding region plus 2100 bp Hyg resisting cassette). Successful transformants are expected to be carrying an additional copy of the complete CGA1 gene, with the point mutation. So the same PCR will produce additional band, 809 bp long, which will allow us to distinguish between successful and unsuccessful transformants (Figure 2).

Since the new cga1^Q204L mutant has two copies of the cga1 gene (the activated one and the original disrupted one) a PCR with the primers Cga1-f and Cga1-r will result in a mixture product of both genes. In other words a mixture of adenine (A) and thymine (T) is expected. Several isolated mutants (WT and cga1 in the background) showed resistance to bialaphos antibiotics, and proved by PCR to have the bar expression cassette together with the CGA1 gene (as shown for the QL18a mutant strain, Figure 2). Nevertheless, only one mutation, QL18a (CGA1 in the background, created by plasmid digested with Hind III) showed mixture of adenine (A) and thymine (T) in a Cga1-f—Cga1-r PCR reaction product (Figure 3).

The WT strain transformation with the same linear plasmid led to at least one bar resistance mutant named QL10. This mutant proved by PCR to carry the bar expression cassette and the additional CGA1^Q204L gene but the resultant products that were sent for sequencing didn’t contain the mixture of adenine (a) and thymine (t) as expected (Figure 3).

cga1^Q204L phenotype characterization. The cga1^Q204L mutation strain (QL18a) was characterized by appearance of white aerial hyphae (Figure 4), a WT proximately sporulation (Figure 5) but hyphal straight growth, with no apparent appressorium formation (Figure 5). Interestingly these phenotypes resemble the cga1 phenotypes. A “water drop assay” conducted to test the colony hyphae absorbency phenotype that may indicate hydrophobins secretion pattern (Figure 4, lower panel). Here also no obvious difference was found between cga1 and the cga1^Q204L mutants and both strains showed hydrophobic colony surface.

Former examination showed that cga1 strains have a significant sensitivity to Sorbitol (1M) osmotic stress in comparison to the WT strains [4]. The cga1^Q204L may show resistance similar to the WT or even more as described in N. crassa [9]. So this trait may provide us with a better insight of the role of cga1 in mediating osmotic

stress response. Although this expectation, the constantly activated cga1 mutant showed the same sensitivity to Sorbitol (1M) osmotic stress as CGA1 disruption strains (data not shown).

In most published works employing activated Gα alleles, it has not been shown, biochemically, that the transgene encodes a protein lacking GTPase activity, or that it activates downstream effectors such as adenylate cyclase (an example is [7]). Such work can support the resulting phenotype conclusion. Nevertheless, phenotypes characterization, based on site specific mutagenesis (such as Q204-L, G42-R and R178-C), designed to constitutively activate Gα signaling, has been widely used to investigate the role of genes encoding heterotrimeric G-protein α subunits (Gα) in filamentous fungi. Constitutively activated mutants in this gene were used to study the Gα’s role in Schizophyllum commune [11], Hypocrea jecorina [12], Ustilago maydis [13], Penicillium chrysogenum [14] and Cryphonectria parasitica [7].

Comparative analysis of the phenotypic traits exhibited by fungal strains containing null or activated Gα alleles has been used by a number of laboratories to identify putative signaling-related functions [7-10]. Free Gβg may cause a signal in both the Gα null and activated mutant strains and the phenotypic traits of both strains are affected mainly by the manipulated Gα gene. Indeed in some instances phenotypes of the Gα null and activated mutant strains were different. In N. crassa [9] the Gα activated mutant (gna-1^R178C and gna-1^Q204L) has longer, abundant aerial hyphae, less conidia per aerial hyphae, greater colony dry weight mass, lower carotenoid secretion, higher cAMP levels and increased sensitivity to heat and hydrogen peroxide-induced oxidative stress than wild-type strains while the null mutant Dgna-1 presents the opposite phenotypes. Furthermore, the permanent activation of Dgna-1 abolished the osmotic sensitivity, the lower extension rate and the female sterility of the null Dgna-1 mutant. In A. nidulans [10], constitutive signaling of α-subunit of G protein (fadA^G42R) resulted in proliferation and block of sporulation in contrast to the null mutant (fadA) that was characterized by reduced growth with normal sporulation. Other studies demonstrated that some similar phenotypes result from constant activation or silencing of the Gα subunit. In M. grisea activated and null mutants of the Gα subunit magB exhibited similar phenotypes in terms of reduced conidiation, sexual reproduction and virulence [8]. These findings were supported by a recent work in C. parasitica [7]. Pigmentation, conidiation, and virulence negative regulation were completely compromised in both Gα (cpg-1) null mutant and activated CPG-1 strains (QL and RC). Nevertheless dissimilar responses were identified in the two C. parasitica mutant strains when subjected to a variety of stresses.

In this work we constructed an activated cga1 allele, Q204L. Although the success of the mutagenesis was proved by PCR and sequencing, this mutant presents similar phenotypes to the null cga1 mutant. These phenotypes include colony growth rate on CM or on CM containing 1.5 M sorbitol (hyper osmotic stress), sporulation, hyphae straight growth, aerial hyphae growth and hydrophobicity. The results were presented here indicating a possible role for CGA1 as a stabilizer of these traits.

I am grateful to Dr. Benjamin A. Horwitz (TechnionIsrael Institute of Technology, Israel) for his guidance and much helpful advice. I thank Oren Schaedel for his technical assistance and Dr. Sophie Lev for many helpful suggestions.