Micheline Wésolowski-Louvel

The RAG4 gene encodes for the sole transmembrane glucose sensor of Kluyveromyces lactis. A rag4 mutation leads to a fermentation-deficient phenotype (Rag− phenotype) and to a severe defect in the expression of the major glucose transporter gene RAG1. A recessive extragenic suppressor of the rag4 mutation has been identified. It encodes a protein (KlRgt1) 31% identical to the Saccharomyces cerevisiae Rgt1 regulator of the HXT genes (ScRgt1). The Klrgt1 null mutant displays abnormally high levels of RAG1 expression in the absence of glucose but still presents an induction of RAG1 expression in the presence of glucose. KlRgt1 is therefore only a repressor of RAG1. As described for ScRgt1, the KlRgt1 repressor function is controlled by phosphorylation in response to high glucose concentration and this phosphorylation is dependent on the sensor Rag4 and the casein kinase Rag8. However, contrary to that observed with ScRgt1, KlRgt1 is always bound to the RAG1 promoter. This article reveals that the key components of the glucose-signaling pathway are conserved between S. cerevisiae and K. lactis, but points out major differences in Rgt1 regulation and function that might reflect different carbon metabolism of these yeasts.

In Kluyveromyces lactis, the casein kinase I (Rag8p) regulates the transcription of glycolytic genes and the expression of the low-affinity glucose transporter gene RAG1. This control involves the transcription factor Sck1p, a homologue of Sgc1p of Saccharomyces cerevisiae. SGC1 is known to interact genetically with ScGCR1 and ScGCR2, which code for regulators of glycolytic gene expression. Therefore, we studied the role of KlGCR1 and KlGCR2 genes in K. lactis. The Klgcr1 null mutant could not grow on glucose when respiration was blocked by antimycin A (Rag⁻ phenotype). In contrast, the Klgcr2 null mutant could grow under the same conditions, although at a reduced rate. In both mutants, the transcription of glycolytic genes was affected, while that of ribosomal protein genes was not modified. Furthermore, the transcription of the glucose permease genes was also found to be affected in the two mutants, although dissimilarly. While RAG1 transcription decreased at high glucose concentrations, the expression of the high-affinity glucose permease gene HGT1 was unexpectedly impaired under gluconeogenic conditions, in the absence of glucose. Gel mobility shift assays performed with purified maltose-binding protein–KlGcr1p showed that KlGcr1p could interact directly with the promoters of the glycolytic genes, but not with the promoters of the glucose permease genes. Thus, the control exerted by KlGcr1p and KlGcr2p upon glucose transporter genes is probably indirect.

The RAG3 gene of Kluyveromyces lactis, a homolog of PDC2 of Saccharomyces cerevisiae, is known to be a regulator of the pyruvate decarboxylase gene KlPDC1. We have identified new target genes for Rag3p. The RAG3 gene product was found to be required for the transcription of two genes of the biosynthetic pathway of thiamine (a cofactor of pyruvate decarboxylase). Conversely, the RAG3 gene product partially repressed the expression of the pyruvate dehydrogenase gene KlPDA1. Therefore, RAG3 may act as a general regulator in the balance of the two alternative pathways of pyruvate metabolism in yeast.