Bistability and hysteresis of the 'Secteur' differentiation are controlled by a two-gene locus in Nectria haematococca

Seven mutants specifically affected in the generation of σ were selected through two different screening strategies. The s1 and s2 mutations completely abolish the generation of σ and of its morphological expression, the Secteur. The remaining five mutations promote its constitutive generation, which determines an intense pigmentation but not growth alteration. The seven mutations map at the same locus, Ses (for 'Secteur-specific'). The s2 mutant was obtained by an insertional mutagenesis strategy, which permitted the cloning of the Ses locus. Sequence and transcription analysis reveals that Ses is composed of two closely linked genes, SesA, mutated in the s1 and s2 mutant strains, and SesB, mutated in the s* mutant strains. SesB shares sequence similarity with animal and fungal putative proteins, with potential esterase/lipase/thioesterase activity, whereas SesA is similar to proteins of unknown function present only in the filamentous fungi Fusarium graminearum and Podospora anserina.The cloning of Ses provides evidence that a system encoded by two linked genes directs a bistable and hysteretic switch in a eukaryote. Atypical regulatory relations between the two proteins may account for the hysteresis of Secteur differentiation.Although it has long been known that cellular memory, or epigenetic states, can be created by emergent properties of regulatory or metabolic networks (see Delbrück in the discussion of [1], and [2,3]), the full implications of this type of inheritance have only recently been understood. To date, pertinent studies focused mainly on phenomena related to chromatin structure and DNA methylation, RNAi and other post-transcriptional gene silencing processes, and prions. However, virtually any physiological process can adopt a bistable (or multistable) behavior, as defined by the ability to adopt two (or more) stable states rather than a range of intermediate states, provided that a positive feedback loop exists (or its related counterpart, the mu