1. Introduction
Killer yeast strains have the characteristic of secreting toxins of proteinaceous nature that are lethal to sensitive yeast cells and filamentous fungi. The killer phenomenon in yeasts was first discovered in Saccharomyces cerevisiae [1] but soon shown to be present in many other genera of yeast [2–4]. Until now, the killer phenomenon has been discovered in almost one hundred species of more than twenty genera [5]. Over 11 different killer toxins have been described, and they are produced by representatives of such genera as Hanseniaspora, Pichia, Saccharomyces, Torulaspora, Ustilago, Williopsis, etc. (Table 1) [6–10].
Killer yeasts are widespread in nature where they can be found in percentages that several fold exceed those found in laboratory strains, indicating the existence of a competitive advantage. Microbes have evolved to compete over the neighborhoods of their same niche. One of the mechanisms to gain advantage that have been described is shown by some killer yeast through the presence of dsRNA viruses. These mycoviruses encode killer toxins that provide benefits to the producing cells by killing sensitive yeast cells. Killer toxin production has been related in S. cerevisiae with the presence of two dsRNA viruses: L-A, the helper virus, and the M killer virus that encodes a killer toxin that determines its phenotype (K1, K2, K28 or Klus). Although killer toxins could imply an important competitive advantage, it has also been demonstrated that there is a fitness cost for carrying mycoviruses [11].
In addition to the cytoplasmic dsRNA viruses mentioned above, a variety of yeast species (Kluyveromyces lactis, P. acaciae, P. inositovora, D. robertsiae, Trichosporon pullulans, etc.) produce linear dsDNA plasmids, probably not encapsulated, that encoded killer toxins. Furthermore, chromosomally encoded killer toxins (Williopsis saturnus, Pichia anomala, Pichia kluyveri, Pichia membranifaciens, etc.) are also found widespread in nature. Most of these toxins exceed the focus of the presented review but their characteristics and potential applications can be reviewed in the tables showed (Tables 1–3).
The reasons why killer yeasts are immune to their own toxins remain to be elucidated for most killer systems. However, in particular cases it has been solved. For example, K28 immunity occurs via the ubiquitination of re-internalized mature toxin and unprocessed precursor moieties complexes that are degraded by the proteasome [12]. Furthermore, Pichia acaciae, Kluyveromyces lactis and Debaryomyces robertsiae cytoplasmic virus like elements encode toxic anticodon nucleases along with specific proteins that confer toxin immunity [13,14]. The potential use of killer yeasts and their toxins has been intended for various fields of application such as the alcohol fermentation industries (brewery, winery, and distillery), fermented vegetables, biological control of post-harvest diseases, yeast bio-typing, as antimycotics in the medical field and they have been used as model systems to understand eukaryotic polypeptide processing and expression of eukaryotic viruses [10–21].
Currently, around 1500 different species of yeasts are known [22] and one of the largest yeast genera is the genus Pichia within the context of number of species. The number of species included in Pichia genus has changed considerably since Hansen (1904) first described the genus Pichia [22–25]. Pichia species are widely distributed and are present in natural habitats and also as spoilage yeasts in several foods and beverages [26,27] and even, human pathogens [28,29]. Killer phenomenon has been discovered and studied in several species of Pichia that show killer activity against yeasts or fungi of clinical concern (Table 2). Toxins of the genus Pichia are either associated with cytoplasmic genetic elements, such as dsDNA virus-like elements or chromosomally encoded [30–32]. Gene products of both extranuclear and chromosomally encoded toxins, are diverse in their molecular mass. There has been described monomeric toxins of Pichia species from less than 3 kDa (P. ohmeri killer toxin) to trimeric toxins of more than 187 KDa (P. acaciae PaT) [29,33,34], although, similar to what occurs globally for killer toxins, the molecular mass is not generally high (20–50 KDa) in Pichia toxins.
The stability of most killer toxins has been shown to be restricted to acidic pH values and low temperatures [35–38]. In this regard, toxins of the Pichia genus are not different from the rest of killer toxins; to date, no Pichia toxins with stability to high temperature or wide pH ranges have been described.
Known killer toxins display diverse modes of actions, such as membrane-damaging agents, glucanase activity, inhibitors of β-1,3-glucansynthase, cell cycle arrest, inhibition of calcium uptake, or tRNase activity; and toxins of the Pichia genus exhibit most of them [39–42].
In initial works, it was found that the majority of yeasts isolated from spontaneously fermenting olive brines possessed the killer character and that strains of Pichia membranifaciens, the dominant species, were particularly active in the presence of salt, eventually influencing the development of some spontaneous fermentations [43,44]. The reasons for such an increased killer activity was elucidated in the study of the mechanism of action of the first P. membranifaciens toxin analyzed, called PMKT (produced by the strain CYC 1106). Additional studies indicated that P. membranifaciens CYC 1086, presented a different killer behavior, secreting PMKT2. This review is about the killer system of P. membranifaciens; and in particular, it focuses in the findings described during the last 20 years about PMKT and PMKT2, until now, the only toxins described for this species. Significant progress in the determination of the nature of the killer toxins, their toxic mechanism and approaches for the potential use of such toxins has only been done very recently. In this review, are concentrated the studies carried out for the characterization on PMKT and PMKT2, their primary and secondary receptors in sensitive strains, toxin mechanisms of action, and cellular responses of sensitive cells to killer toxins. Finally, the potential applications and future perspectives of killer toxins, with special focus in P. membranifaciens killer toxins, are described and discussed.