Killer Factor as a Mechanism of Interference Competition in Yeasts Associated with Cacti

Abstract

Many yeasts in the genus Pichia are killers (yeasts that secrete proteins toxic to other yeasts). The genus is subdivided into complexes comprised of species with similar morphologies, physiological abilities, and percentage of (guanine + cytosine) comprising their nuclear DNA. Here, we test the hypothesis that ecological factors influence the types of toxins produced by members of two Pichia complexes (the Pichia kluyveri and Pichia opuntiae complexes). Members of each complex are separated by geography or host range, but all live in decaying cactus tissue. Similarities among killer toxins were investigated by testing strains from each complex in the laboratory for their ability to kill a standard set of 70 yeasts. Principal components analysis demonstrated that the killer phenotype was constant within a species, but not within the complexes. The probability that a toxin would kill a particular yeast depended on some ecological characteristic of the yeasts tested as sensitives (i.e., the region, host plant, or habitat from which the yeasts were collected). We further tested the hypothesis that ecological factors influence killer phenotype by analyzing within—community patterns of killing. Killer strains from two different cacti were tested for their ability to kill a subset of the yeasts actually collected from the same pockets of rotting cactus tissue from which the killer strains were collected or from rot pockets in neighboring cacti. When the yeasts tested for sensitivity were classified a priori by their probability of interacting with each killer strain, it was shown that the sensitivity to killer toxin was correlated with the probability of interaction. Further, the yeast community composition of a rot pocket depended on the presence or absence of a killer strain in that pocket. We demonstrated that killer toxins could change community composition by growing strains of two sensitive yeasts in the presence of a killer or a non—killer strain of P. kluyveri. The killer strain reached a higher cell density than the non—killer strain in the presence of sensitive yeasts. The density of sensitive yeasts grown with a killer strain was lower than that of a strain grown with a non—killer strain of P. kluyveri.

Candida krusei and Kloeckera apis inhibit the causal agent of pineapple fusariosis, Fusarium guttiforme

“…Attachment of yeasts to pathogen hyphae has been described as a factor related to competition for nutrients, hyphae lysis, and disruption of filamentous fungi (Chaurasia et al 2005). Ganter & Starmer (1992) referred to this type of interaction as interference competition.…”

Antagonistic interactions between garden yeasts and microfungal garden pathogens of leaf-cutting ants

“…mycocins) that inhibit the growth of yeast sensitive strains (Carreiro et al 2002). Toxin-secreting yeasts are thought to maintain the stability of some microbial communities (Starmer et al 1987), mediated through interference competition against other yeasts and also plant pathogenic fungi (Ganter and Starmer 1992;Walker et al 1995). Carreiro et al (2002) suggested that interference competition as a possible mechanism controlling yeast population and other antagonistic microbes in attine gardens.…”

Combined use of killer biotype and mtDNA-RFLP patterns in a Patagonian wine Saccharomyces cerevisiae diversity study

“…Moreover, the natural distribution of yeasts producing killer toxins and sensitivity to those toxins were demonstrated to be related to phylogeny as well as the ecological circumstances of the strains (Ganter and Starmer 1992). In winemaking, different killer yeast species and strains have been reported and killer interactions between indigenous yeasts and between indigenous and selected yeasts (starters) could be established during alcoholic fermentation.…”