Functional expression and cellular localization of cercosporin-resistance proteins fused with the GFP in Cercospora nicotianae.

The Cercospora nicotianae pdx1 and crg1 genes were previously identified as genes required for resistance to the singlet oxygen ((1)O(2))-generating toxin cercosporin. The pdx1 gene has subsequently been shown to be required for pyridoxine biosynthesis, but both the precise biochemical function of the PDX1 protein and the function of the CRG1 protein remain undefined, as both sequences lack defined enzymatic domains or cofactor-binding sites. The gfp gene encoding green fluorescent protein was translationally fused with pdx1 and crg1. Transformation of these constructs into strains mutant in these respective genes resulted in green-fluorescent transformants complemented for the mutant phenotype. Microscopic studies revealed that in transformants transformed with gfp alone, fluorescence was distributed evenly throughout the cytoplasm and excluded from the vacuoles. Expression of PDX1::GFP either under the constitutive Aspergillus nidulans gpdA promoter or its own native promoter was visualized as distinct fluorescent circular structures in the cytoplasm, suggesting that PDX1::GFP was probably localized in the intracellular vesicles. Expression of CRG1 fused with GFP at either its N- or C-terminus resulted in low green fluorescence, compared with that of GFP alone or PDX1::GFP. The green fluorescence of either of the CRG1::GFP fusion proteins was barely observable in transformants and was generally seen as a few scattered regions of fluorescence in the hyphae. Southern blot analysis indicated multiple copies of the constructs were integrated into the fungal genome. Northern analysis revealed that pdx1:: gfp and crg1:: gfp were each expressed as an intact transcriptional unit. Cell fractionation followed by immunoblotting against a GFP antibody showed that GFP alone and PDX1::GFP were detected exclusively in the cytoplasmic fraction. The two CRG1::GFP proteins were barely detected in the cytoplasmic fraction and not at all from the membrane fraction, a result inconsistent with microscopic observation and computer sequence analysis, which suggests that CRG1 is a membrane protein.