María Teresa Fernández-Espinar

Study of the authenticity of commercial wine yeast strains by molecular techniques

mtDNA restriction analysis has been carried out with 45 different commercial Saccharomyces wine yeast strains. The analysis with Hinf I provided unique profiles for 17 of the 45 strains and can therefore be considered as individual strains. Nevertheless, among the remaining 28 strains, only eight mtDNA restriction patterns appeared. These strains were subjected to electrophoretic karyotyping and PCR amplification of delta sequences. We concluded that the maximum discriminatory power was obtained when the results of the three techniques were combined, giving 13 different composite patterns for the 28 strains under study. The results showed evidence of mistakes during production or fraudulent practices by yeast producers, since only 30 individual strains have been identified among the 45 Saccharomyces wine yeast strains commercialised by different companies. Additionally, commercial starters of Saccharomyces uvarum and Saccharomyces bayanus have been re-identified as Saccharomyces cerevisiae.

In vivo virulence of commercial Saccharomyces cerevisiae strains with pathogenicity-associated phenotypical traits.

Two commercial Saccharomyces cerevisiae strains, a bakers strain and the bio-therapeutic agent Ultralevure, have been proposed as a possible exogenous source of human colonization (de Llanos et al., 2004, 2006a). Moreover, these strains express phenotypical traits associated to pathogenicity (de Llanos et al., 2006b). Taking into account that both commercial preparations represent an important source of living S. cerevisiae cells we have performed an in vivo study to evaluate whether there is a potential safety risk to humans. Their virulence was compared with that of other commercial strains with less virulent traits, and with clinical isolates, using two murine models (BALB/c and DBA/2N mice). Burden determination in the brain and kidneys showed that the ability to disseminate, colonize and persist was manifested not only by clinical isolates but also by commercial strains. Among these, the bakers strain and Ultralevure were able to cause the death of BALB/c mice at rates similar to those shown by two of the clinical isolates. These results highlight the pathogenic potential of these strains and show that four-week-old BALB/c mice are an appropriate murine model to study the virulence of yeasts with low or moderate pathogenicity. Furthermore, we have shown the positive effect of an immunosuppressive therapy with cyclophosphamide in the virulence of the bakers strains and Ultralevure but not in the rest of the commercial strains under study. The data suggest that although S. cerevisiae has always been considered a GRAS microorganism, commercial preparations should include only those strains shown to be safe in order to minimize complications in risk groups.

Differences in activation of MAP kinases and variability in the polyglutamine tract of Slt2 in clinical and non-clinical isolates of Saccharomyces cerevisiae.

The concept of Saccharomyces cerevisiae as an emerging opportunistic pathogen is relatively new and it is due to an increasing number of human infections during the past 20 years. There are still few studies addressing the mechanisms of infection of this yeast species. Moreover, little is known about how S. cerevisiae cells sense and respond to the harsh conditions imposed by the host, and whether this response is different between clinical isolates and non‐pathogenic strains. In this regard, mitogen‐activated protein kinase (MAPK) pathways constitute one of the major mechanisms for controlling transcriptional responses and, in some cases, virulence in fungi. Here we show differences among clinical and non‐clinical isolates of S. cerevisiae in the level of activation of the MAPKs Kss1, which controls pseudohyphal and invasive growth, and Slt2, which is required for maintaining the integrity of the cell wall under stress conditions and in the absence of stimulating conditions. Moreover, we report for the first time the existence of length variability in SLT2 alleles of strains with a clinical origin. This is due to the expansion in the number of glutamine‐encoding triplets in the microsatellite region coding for the polyglutamine (poly‐Q) tract of this gene, which range from 12 to more than 38 repetitions. We suggest that this variability may influence biological features of the Slt2 protein, allowing it to adapt swiftly in order to survive in unusual environments. Copyright

Pathogenic potential of Saccharomyces strains isolated from dietary supplements.

Saccharomyces cerevisiae plays a beneficial role in health because of its intrinsic nutritional value and bio-functional properties, which is why it is also used as a dietary supplement. However, the perception that S. cerevisiae is harmless has changed due to an increasing number of infections caused by this yeast. Given this scenario, we have tested whether viable strains contained in dietary supplements displayed virulence-associated phenotypic traits that could contribute to virulence in humans. We have also performed an in vivo study of the pathogenic potential of these strains using a murine model of systemic infection by intravenous inoculation. A total of 5 strains were isolated from 22 commercial products and tested. Results highlight one strain (D14) in terms of burden levels in brains and kidneys and ability to cause death, whereas the other two strains (D2 and D4) were considered of low virulence. Our results suggest a strong relationship between some of the virulence-associated phenotypic traits (ability to grow at 39°C and pseudohyphal growth) and the in vivo virulence in a mouse model of intravenous inoculation for isolates under study. The isolate displaying greatest virulence (D14) was evaluated in an experimental murine model of gastrointestinal infection with immunosuppression and disruption of mucosal integrity, which are common risk factors for developing infection in humans, and results were compared with an avirulent strain (D23). We showed that D14 was able to spread to mesenteric nodes and distant organs under these conditions. Given the widespread consumption of dietary supplements, we recommend only safe strains be used.

Evaluation of the Ability of Polyphenol Extracts of Cocoa and Red Grape to Promote the Antioxidant Response in Yeast Using a Rapid Multiwell Assay

Saccharomyces cerevisiae has been used as a model organism to study the capacity of cocoa and red grape extracts to trigger an antioxidant response. A methodology adapted to microtiter plates has been developed to monitor yeast growth after culture preincubation with food ingredients and exposure to oxidative stress by hydrogen peroxide and menadione. This methodology proved effective in measuring the ability of cocoa and red grape extracts to promote an antioxidant response in yeast, and also the prospect of conducting dose-response studies. Additionally, the method has proven useful to perform studies with mutant strains lacking genes that may be related to the mechanism of action underlying the antioxidant properties. Thus, in a single assay, it is possible to elucidate the sensitivity of strains to oxidative stress, the ability of an ingredient to promote an antioxidant response, and the possible implication of certain genes. Results of assays using strain hst3Δ showed that the antioxidant protection provided by exposure to cocoa and red grape extracts was not present in the strain lacking gene HST3 when H2 O2 and menadione were used as oxidizing agents. This effect was previously reported for cocoa extract only, with H2 O2 as stressor. Moreover, the results showed that the mutant strain hst3Δ is more resistant to menadione and H2 O2 in the absence of preincubation with cocoa and red grape extract, hinting at the possible implication of sirtuin Hst3 in the antioxidant cellular response.