**Antagonistic Yeasts: A Promising Alternative to Chemical Fungicides for Controlling Postharvest Decay of Fruit**

**Xiaokang Zhang 1,2,**† **, Boqiang Li 1,**† **, Zhanquan Zhang <sup>1</sup> , Yong Chen <sup>1</sup> and Shiping Tian 1,2,\***


Received: 12 July 2020; Accepted: 28 August 2020; Published: 31 August 2020

**Abstract:** Fruit plays an important role in human diet. Whereas, fungal pathogens cause huge losses of fruit during storage and transportation, abuse of chemical fungicides leads to serious environmental pollution and endangers human health. Antagonistic yeasts (also known as biocontrol yeasts) are promising substitutes for chemical fungicides in the control of postharvest decay owing to their widespread distribution, antagonistic ability, environmentally friendly nature, and safety for humans. Over the past few decades, the biocontrol mechanisms of antagonistic yeasts have been extensively studied, such as nutrition and space competition, mycoparasitism, and induction of host resistance. Moreover, combination of antagonistic yeasts with other agents or treatments were developed to improve the biocontrol efficacy. Several antagonistic yeasts are used commercially. In this review, the application of antagonistic yeasts for postharvest decay control is summarized, including the antagonistic yeast species and sources, antagonistic mechanisms, commercial applications, and efficacy improvement. Issues requiring further study are also discussed.

**Keywords:** yeast; biological control; postharvest decay; fruit

#### **1. Introduction**

As an important part of the human diet, fruit provides the body with beneficial vitamins, minerals, organic acids, and antioxidants. Fruits have been shown to have many health-related effects, such as anti-cancer effects, skin protecting effects, and postponing of senescence [1–4]. As orchards are usually far away from urban areas, and the fruit maturity occurs in a relatively short period, leading to a disparity between supply and demand in the market, which necessitates a certain period of storage and transportation to adjust for this disparity. However, postharvest spoilage, which involves rot, nutrient loss, and water content loss, occurs most often during storage and transportation, which leads to considerable economic losses. It has been reported that about 25% of total fruit production is wasted after harvest in developed countries, and the postharvest losses in developing countries account for >50% of total fruit production because of lack of efficient transportation and refrigeration facilities [5].

Fungi are the main cause of postharvest spoilage. Fruit rot can be induced by wound generated during harvesting, packaging, storage, and transportation, as well as the favorable growth conditions for pathogens (e.g., high water and nutrient content, low pH, and decreased resistance after harvest) [6]. During the process of infection, many fungi produce mycotoxins, which may enter the food chain via fresh and processed fruit products and then endanger human health. For example, *Penicillium expansum*,

which causes blue mold in many fruits, leads to not only fruit decay but also the contamination of patulin, a teratogenic, carcinogenic, and immunotoxic mycotoxin [7]. Chemical fungicides have long been used to control postharvest decay. However, overdependence on traditional chemical fungicides has resulted in a variety of problems, such as fungicide residues, environmental pollution, and increased pathogen resistance to fungicides. Therefore, identifying safe and effective approaches to control postharvest fungal disease is urgent.

Since Gutter and Littauer first reported the use of *Bacillus subtilis* to combat citrus fruit pathogens in 1953, the biocontrol capability of microorganisms against postharvest decay has attracted widespread attention [8,9]. Among the various microbial antagonists, yeast and yeast-like fungi occupy an important position as they are environmentally friendly, exhibit good biocontrol efficacy against pathogens, possess adequate stress tolerance, and can potentially be genetically improved; additionally, there is a well-developed system for culturing, fermentation, storage, and handling of these antagonistic yeasts [10]. Moreover, yeasts have been used in food and beverage production for thousands of years and currently play an important role in the food industry. Thus, the utilization of yeasts is generally considered safe, and easily acceptable by market. With the great properties and application superiority, antagonistic yeasts are considered as a promising alternate to synthetic chemical fungicides [5,9]. Over the past few decades, great progresses have been made in biological control based on antagonistic yeasts, including strain isolation and screening, mode of action, improvement of biocontrol efficacy, and formulation. Particularly, several antagonistic yeasts with excellent biocontrol performance have been developed and registered as commercial products. Nonetheless, the widespread use of yeast antagonists to manage postharvest diseases still faces many challenges. A deeper understanding of the mode of action of antagonistic yeasts in postharvest biocontrol system is still needed; the inconsistency of performance of antagonistic yeasts under commercial conditions need be overcome; the market penetration of products is difficult.

Here, a comprehensive overview of the applications of antagonistic yeasts in postharvest decay control is presented, including the features of antagonistic yeasts, antagonistic mechanisms, efficacy improvement, and commercial applications. The latest research results are highlighted, and issues requiring further study are also discussed.

#### **2. Features of Antagonistic Yeasts**

Yeasts are a group of eukaryotic fungi, most of which are unicellular and reproduce by budding [11]. There are also a variety of phylogenetically different groups of yeast-like fungi, such as *Aureobasidium pullulans*. Antagonistic yeasts (also known as biocontrol yeasts) refers to yeast or yeast-like fungi that can inhibit or interfere with the growth, development, reproduction, or activity of phytopathogens. Wilson and Wisniewski summarized the criteria for the selection of ideal biocontrol agents in 1989 [12]. With the extensive research on antagonistic yeasts, the screening criteria for antagonistic yeasts have gradually improved [13]. An ideal antagonistic yeast should be genetically stable, have simple nutrient requirements, be effective in adverse environmental conditions and at low concentrations, and be effective against multiple fungal pathogens on various fruits [6,9]. Moreover, an antagonistic yeast should have favorable commercial potential: It should be able to grow on an inexpensive growth medium, be easy to store and dispense, and be compatible with other physical and chemical treatments (e.g., controlled atmosphere, low/high temperature, chemical fungicides/pesticides, and phytohormones [5]. As for biosafety, a desirable antagonistic yeast would be environmentally friendly, have no pathogenicity regarding the host fruits, produce no metabolites that are harmful to humans, and be unable to cause infection in humans [5,9].

The isolation and screening process is the first step in the development of a biocontrol agent. Most antagonistic yeasts were isolated directly from fruit surfaces [14,15], but they have a wider distribution in nature, such as on leaves and roots and in seawater and soil (even Antarctic soil) [16–20]. So far, a large number of antagonistic yeasts have been isolated and screened. Some of them have been widely studied, such as *Candida* spp., *Cryptococcus* spp., *Metschnikowia* spp., *Pichia* spp., *Rhodotorula* spp., and yeast-like fungus *A. pullulans*, and several species, such as *Candida oleophila*, *Candida sake*, *Metschnikowia fructicola*, *A. pullulans*, *Saccharomyces cerevisiae*, and *Cryptococcus albidus*, have been developed as commercial products [21–28]. They have been demonstrated to antagonize common postharvest pathogens, including *Botrytis cinerea*, *Penicillium* spp., *Rhizopus stolonifer*, *Colletotrichum* spp., *Monilinia fructicola, Alternaria alternata*, and *Aspergillus niger*. Representative antagonistic yeasts that were isolated from various sources and are used for the management of postharvest diseases are shown in Figure 1.

**Figure 1.** Representative antagonistic yeasts from various sources used for the management of postharvest decay. Species that have been already in commercial use are highlighted in red.

#### **3. Mechanisms of Action**

Elucidating the mechanisms of action is the foundation for the development and application of antagonistic yeasts [29]. Compared with the impressive results achieved regarding the identification of antagonistic yeasts, the study of their mechanisms of action is relatively slow due to the complexity of the postharvest biocontrol system. In this system, the antagonistic yeasts, pathogenic fungi, and fruit hosts interact with each other under the influence of the environment, and the influence of the epiphytic microbiome should also be taken into consideration (Figure 2) [12,29–31].

The antagonistic yeasts are likely to function via multiple mechanisms, including competition for nutrients and space, mycoparasitism, induction of host resistance, production of volatile organic compounds (VOCs), and toxins [9,10,31]. With the increase in the number of annotated yeast genomes and the development of "omics" technologies and transformation technologies, the modes of action of antagonistic yeasts will be further deciphered in the near future [32–34].

**Figure 2.** Schematic diagram of the possible interactions among components of the biocontrol system, including the pathogens, antagonistic yeasts, host, epiphytic microbiome, and environment. Antagonistic yeasts can inhibit pathogens through competition for nutrient and space, mycoparasitism, VOCs, and killer toxins. Conversely, pathogens also compete with antagonistic yeasts for nutrient and space to affect their colonization and growth. In addition, antagonistic yeasts can induce the resistance of hosts to inhibit infection, while reactive oxygen species (ROS) produced by hosts may be an oxidative stress to yeasts. During the interaction between fruit hosts and pathogens, hosts can resist the pathogen attack through oxidative burst, innate immune system, and antifungal metabolites, while pathogens can suppress host resistance through pathogen-associated molecular patterns (PAMPs), effectors, phytotoxins, pH modification, and suppression or stimulation of the oxidative burst. The epiphytic microbiome on hosts is also associated with the host resistance. Moreover, environmental conditions have a wide influence on the pathogenicity of pathogens, the efficacy of antagonistic yeasts, and the resistance of hosts.

#### *3.1. Competition for Nutrients and Space*

Both postharvest pathogens and antagonistic yeasts require nutrients (e.g., carbohydrates and nitrogen) and space to colonize and develop. Therefore, the competition for nutrients and space has been considered the primary mode by which antagonistic yeasts suppress postharvest fungal pathogens [5,29]. Once the antagonistic yeasts come into contact with the surface of the injured fruit, they will occupy the wounds and rapidly deplete the nutrients, which limits the germination of fungal spores [35,36]. After that, other mechanisms of action (besides the competition for nutrition and space) cooperatively come into operation to control the postharvest pathogens [8].

Carbon, nitrogen, and iron ions are the main nutrients needed for the growth of microbes. Compared with carbohydrates, nitrogen is considered to be a key factor limiting the growth of postharvest fruit pathogens, because most fruits are rich in sugar but limited in nitrogen sources such as amino acids. The application of exogenous amino acids reduced the antagonistic effect of the yeast *A. pullulans* against *Penicillium expansum* on apple fruit, indicating the importance of nitrogen competition to biocontrol efficacy [37]. Moreover, iron plays a crucial role in the growth and virulence of pathogens. Iron is a component of cytochromes, other heme proteins, and non-heme proteins; it is also a cofactor of various enzymes in fungal cells [9,38]. The yeast *Metschnikowia pulcherrima* can

produce iron chelators to compete for the iron required by pathogens, thus strongly inhibiting the growth of the pathogens [39]. Parafati et al. also proposed that the consumption of iron ions plays an important role in the biocontrol effect of *M. pulcherrima* [40]. Some antagonistic yeasts can also produce siderophores to compete for iron in a low-iron microenvironment, thus inhibiting the germination and growth of pathogens. For example, rhodotorulic acid is a dihydroxamate siderophore produced by *Rhodotorula glutinis* that improves the biocontrol against *P. expansum* [41]. Siderophores produced by *A. pullulans* plays an important role in yeast growth and pathogen inhibition under iron deficiency environment [13,42].

Biofilms are dense microbial communities attached on fruit surfaces, and encapsulated by polymeric extracellular matrix (ECM) [43]. Formation of biofilm is considered as another strategy utilized by antagonistic yeasts to compete for space and nutrient [10,44]. Scherm et al. found that the biofilm formation of *S. cerevisiae* M25 was directly related to its biocontrol effect, with only the *S. cerevisiae* cells collected during the biofilm formation phase effectively controlling *P. expansum* on apples [45]. Biofilm formation has also been hypothesized to be a key mechanism of action of *Metschnikowia citriensis* against *Penicillium digitatum* and *Penicillium italicum* on citrus fruit [46]. Notably, it was reported that *Pichia fermentans* formed biofilms and inhibited postharvest decay in apple fruits but caused rapid decay in peach fruits in the absence of a plant pathogen [47], indicating the potential risk of dimorphic antagonistic yeast becoming pathogens.

#### *3.2. Mycoparasitism*

Mycoparasitism refers to the phenomenon of antagonistic yeasts feeding on fungal pathogens via attaching to the fungal pathogen hyphae and then secreting cell wall-degrading enzymes to destroy or lyse the fungal structures. Especially in the case of nutritional deficiencies, antagonistic yeasts tend to absorb nutrients from pathogenic cells, leading to the death of these "prey" cells. During mycoparasitism, a variety of enzymes are involved in the degradation of the fungal pathogen cell wall, especially β-1,3-glucanase (GLU), chitinase (CHT), and proteases [29], and these secreted enzymes are thought to play an important role in biocontrol [48]. Wisniewski et al. first reported the mycoparasitism of *Pichia guilliermondii* [49]. They observed that the yeast strongly adhered to the *B. cinerea* mycelium and caused hyphal collapse, which was presumably due to a lectin-like interaction. It has also been reported that both *Pichia membranefaciens* and *C. albidus* can attach to and degrade the hyphae of *P. expansum, M. fructicola*, and *R. stolonifer* [50]. Banani et al. found that the chitinase gene, *MfChi*, of the yeast *M. fructicola* was significantly induced by cell wall of the postharvest pathogen *M. fructicola*, and MfChi-overexpressing *Pichia pastoris* inhibited the brown rot of peach fruits [51]. In *C. oleophila*, GLU was demonstrated to be associated with inhibiting conidial germination and hyphal growth of *P. expansum* [52].

#### *3.3. Induction of Host Resistance*

Induction of host resistance, as one of the major mechanism of antagonistic yeasts for postharvest decay control in fruits, has also been extensively studied [53,54]. Antagonistic yeasts have been reported to act as biological elicitors in the interactions with fruit hosts [29,48]. Treatment with antagonistic yeasts can increase the expression of defense-related genes and enhance the activities of defense-related enzymes. Strongly induced activities of defense-related enzymes, such as CHT, GLU, phenylalanine ammonia-lyase (PAL), and peroxidase (POD), have been reported to be responsible for the biocontrol efficacy of *Cryptococcus laurentii* on postharvest decay caused by *A. alternata*, *M. fructicola*, and *P. expansum* [55–57]. Chan et al. found that the antagonistic yeast *P. membranaefaciens* could induce the activities of three pathogenesis-related (PR) proteins, which may contribute to the resistance improvement of peach fruit to *P. expansum* [33]. Similarly, induced expression of defense-related genes and the activities of defense-related enzymes by *W. anomalus* were considered as one of the possible mechanisms in inhibiting blue mold decay caused by *P. expansum* in pears [58].

Moreover, application of antagonistic yeasts can enhance activity of antioxidant enzymes, which may alleviate oxidative damage cause by reactive oxygen species (ROS) produced by hosts in response to pathogen infection. *P. membranaefaciens* has been reported to affect the activities of antioxidant enzymes, including POD, catalase (CAT), glutathione peroxidase (GPX), superoxide dismutase (SOD), and polyphenol oxidase (PPO), in peaches and sweet cherry fruits after inoculation with *P. expansum* [33,59]. Additionally, four antagonistic yeasts (*P. membranaefaciens*, *C. laurentii*, *Candida guilliermondii*, and *R. glutinis*) have been reported to increase the activities of POD and CAT, upregulate the expression of the corresponding genes, and reduce the levels of protein carbonylation in peach fruits caused by *M. fructicola* [60].

Antagonistic yeasts can also induce changes in secondary metabolites and cell structure related to disease resistance. Droby et al. found that the application of *C. oleophila* increased the levels of the phytoalexins umbelliferone, scoparone, and scopoletin in grape fruit peels [61]. El-Ghaouth et al. found that the antagonistic yeast *Candida saitoana* could induce host cell deformation, generate mastoid structures, and consequently inhibit *B. cinerea* infection [62].

Multiple mechanisms may be simultaneously involved in the resistance induction by antagonistic yeasts. For example, several antagonistic yeasts, such as *C. laurentii* [63], *P. membranaefaciens* [33], *P. guilliermondii* [64], *R. glutinis* [60], and *R. paludigenum* [25], induced changes in activities of both defense-related enzymes and antioxidant enzymes in fruit. Induction of disease resistance by antagonistic yeasts is also affected by pathogens and environmental conditions. As shown in Figure 2, there are complex interactions between the hosts, pathogens, antagonistic yeasts, and environment, which remains to be elucidated.

#### *3.4. Production of VOCs and Killer Toxins*

Compared to filamentous fungi, yeasts have a lower secretory capacity and produce only few secondary metabolites. Nevertheless, VOCs and killer toxins are metabolites that have been reported to exhibit antifungal activity.

VOCs are volatile compounds with low molecular weight (<300 Da), low polarity, and high vapor pressure. Some antagonistic yeasts can produce VOCs, and the mixture of VOCs has been proposed to play an important role in the control of postharvest pathogens under airtight conditions [48,65]. It was reported that *Candida intermedia* 410 inhibited the growth of *B. cinerea* on strawberries by releasing VOCs without direct contact; the absorption of VOCs by activated carbon abolished the biocontrol activity of *C. intermedia* 410 [16]. Two strains of *A. pullulans* (L1 and L8) have been reported to produce VOCs to inhibit the growth and infection of postharvest pathogens, including *B. cinerea*, *Colletotrichum acutatum*, and *Penicillium* spp. [66]. Moreover, VOCs have been reported to suppress the mycelial growth, sporulation, and ochratoxin A biosynthesis of *Aspergillus carbonarius* and *Aspergillus ochraceus* [67,68]. VOCs are considered to be potential biological fumigants because of their volatility, which allows them to control postharvest decay without direct contact with the edible commodities. Contarino et al. found that the main VOCs emitted by common antagonistic yeasts include ethyl alcohol, phenylethyl alcohol, 3-methyl-1-butanol, ethyl acetate, and isoamyl acetate [69]. However, VOCs produced by *Muscodor albus* have been reported to cause DNA damage and cytotoxicity in bacterial cells, indicating that some VOCs may be toxic [70]. Therefore, the safety of VOCs should be thoroughly evaluated in future studies.

Several toxins have been reported to be able to control postharvest pathogens, and proteinaceous killer toxins are the most prominent antifungal toxins produced by yeasts [10]. Killer toxins provide a competitive advantage to yeasts, and they can kill fungi (including other yeasts) by a variety of mechanisms, including hydrolyzation of the cell wall, destruction of the cell structure, and inhibition of DNA synthesis [71]. Yeast strains with a particular killer phenotype are immune to their own killer toxins and those in the same class while being lethal to other yeast strains [71]. Owing to this characteristic, killer toxins have long been used in the wine industry to control spoilage yeasts. As natural antifungal proteins, killer toxins are environmentally friendly, nontoxic to mammals, have a good acid tolerance, and have a low probability of inducing resistance. Therefore, killer toxins have been proposed as potential biocontrol agents. Killer toxins produced by *Wickerhamomyces anomalus* BS91 are encoded by the genes *WaEXG1* and *WaEXG2*, showed exoglucanase activity, and associated with biocontrol capabilities against *B. cinerea*, *P. digitatum*, *P. italicum*, *Monilinia fructigena*, and *M. fructicola* [40,72–75]. *P. membranaefaciens* was found to produce killer toxins PMKT and PMKT2 that target (1→6)-β-D-glucans and mannoproteins in pathogen cell walls and thereby inhibit the growth of postharvest pathogens [76]. Moreover, killer toxins produced by *Debaryomyces hansenii* have been reported to suppress human pathogenic *Candida* yeasts, but only within a certain temperature and pH range, indicating the influence of environmental factors on the antifungal activity of killer toxins [77]. Furthermore, the effects of yeast killer toxins on beneficial microorganisms need to be further evaluated, especially regarding microorganisms in the phyllosphere, on edible commodities, and in the human gut.

#### **4. Constraints on the Application of Antagonistic Yeasts, Improvement of Their Biocontrol E**ffi**cacy, and Commercial Application**

#### *4.1. Constraints on the Application of Antagonistic Yeasts*

Over the past few decades, numerous yeasts with antifungal properties have been identified, but only a few have been developed as commercial antifungal products. This has mainly been due to the fact that besides having excellent biocontrol efficacy, for commercial application, an antagonistic yeast needs to meet additional requirements. Many commercial factors restrict the development and commercialization of antagonistic yeasts, including the immature commercialization technology, high development costs, small postharvest market, and low market acceptance [8,78]. Furthermore, as the utilization of antagonistic yeasts to control postharvest decay is an emerging industry, the research on antagonistic yeasts remains insufficient. In particular, although many studies have reported on the biocontrol mechanisms of antagonistic yeasts, the specific mechanisms require further clarification.

Biosafety is one of the main reasons for using antagonistic yeasts instead of chemical fungicides. Most of the identified antagonistic yeasts have been directly isolated from the surface of fruits, and humans are already exposed to these yeasts when they eat fresh fruits and vegetables in their daily lives, so there is often less concern about the biosafety of antagonistic yeasts. However, some yeasts may be the origin of human infection under rare circumstances [79–81]. Therefore, the biosafety of antagonistic yeasts, including their safety related to skin irritation and ingestion, needs to be fully evaluated. Registration is also an obstacle to the commercialization of many antagonistic yeasts. Biocontrol agents must be approved by relevant regulatory agencies before commercial application. Compared with synthetic chemical fungicides, the registration of an antagonistic yeast is less costly and time-consuming, but it is still a factor to be considered in the development process. The registration of an antagonistic yeast requires an accredited safety assessment report and biocontrol efficacy data. Furthermore, the difficulty of registration varies in different regions. For example, the registration of biocontrol agents in the United States takes an average of 2 years, while in Europe, it takes about 7 years [6]. In China, with government incentives, the registration of biocontrol agents takes about 2–3 years.

Compared with chemical fungicides, antagonistic yeasts still need to be improved in many respects, which also limits their commercialization and market acceptance. Antagonistic yeasts are more expensive than chemical fungicides and are inconvenient to use. Moreover, an ideal biocontrol agent for controlling postharvest decay of fruits and vegetables must be highly effective (>95%) [31]. However, according to the reported researches so far, the biocontrol efficacy using antagonistic yeast alone cannot reach the level demonstrated by chemical fungicides. In addition, the biocontrol efficacy of antagonistic yeasts regarding postharvest decay depends on the high activity and reproductive capacity of the yeasts. In addition, there are issues associated with the use of many antagonistic yeasts, such as their unstable antifungal activity, short shelf life, and strict required storage conditions.

#### *4.2. Improvement of the Biocontrol E*ffi*cacy*

As mentioned above, the use of antagonistic yeast alone to prevent postharvest decay is generally inferior to the use of chemical fungicides. Therefore, while identifying new high-efficacy yeast strains, researchers are also constantly searching for effective ways to strengthen the biocontrol efficacy of existing antagonistic yeasts. The combined use of biological control and physical or chemical methods is an effective way to improve the biocontrol efficacy. For example, hot water treatment (HWT) by immersing fruit in a circulating water bath at 42 ◦C for 40 min improved the biocontrol efficacy of the antagonistic yeasts *C. guilliermondii* and *P. membranaefaciens* without affecting their growth [82].

Salicylic acid (SA) is an important hormone in plants that is related to the induction of the plant response against pathogens [54]. Qin et al. found that SA treatment increased the antagonism of *R. glutinis* against *P. expansum* and *A. alternata* in sweet cherry fruits [83]. SA at low concentrations increased the activities of defense-related enzymes but had little effect on the growth of the yeast and the two pathogens. This indicated that the biocontrol efficacy enhanced by SA may be related to the triggering of host resistance. The ability of SA to enhance the biocontrol efficacy of biocontrol microbes has been demonstrated in many yeast species [84–86]. Methyl jasmonate (MeJA) is another phytohormone that can induce host defense responses [74]. MeJA has also been reported to improve the biocontrol effects of antagonistic yeasts [87,88].

Moreover, it has been reported that exogenous application of brassinosteroids or nitric oxide can induce plant host resistance [89,90], but their synergistic effects when used with antagonistic yeasts remain to be studied. Many natural plant extracts can inhibit the growth and development of pathogenic fungi, such as methyl thujate [91,92], hinokitiol [93], and cinnamic acid [94]. Li et al. reported that cinnamic acid improved the biocontrol efficacy of *C. laurentii* [95], which indicates the potential of combined application of natural plant extracts with antagonistic yeasts for controlling postharvest pathogens. Several other microbial metabolites, such as epsilon-polylysine, natamycin, and rapamycin, have been reported to control postharvest pathogens [96–98], and the combined application of microbial metabolites with antagonistic yeasts is worth exploring.

The use of certain chemical reagents or other antifungal methods can also enhance the biocontrol efficacy of antagonistic yeasts. For example, CaCl<sup>2</sup> has been reported to enhance the efficacy of antagonistic yeasts [99–101]. Additionally, chitosan has antifungal properties and can induce host defense responses, and multiple studies have shown that chitosan can enhance the biocontrol efficacy of antagonistic yeasts such as *C. saitoana* [102], *C. laurentii* [103], and *P. membranaefaciens*[104]. Furthermore, inorganic salts (e.g., ammonium molybdate, sodium bicarbonate, and trisodium phosphate) [105–107], minerals (e.g., silicon and boron) [108,109], and sugar protectants (e.g., maltose and lactose) [110] have been reported to enhance the biocontrol efficacy of antagonistic yeasts. It has also been reported that the use of a combination of an antagonistic yeast and a low-dose chemical fungicide can achieve a similar biocontrol efficacy to the use of the fungicide alone at a commercial dosage, which is considered to be an effective method to reduce fungicide use [14].

The mixed application of various antagonistic yeasts is also considered to be an effective way to broaden the antifungal spectrum of biocontrol reagents and to enhance the biocontrol efficacy. Calvo et al. found that the combined application of *R. glutinis* and *C. laurentii* improved their ability to control gray mold on apples [111]. However, it should be noted that compatibility between mixed antagonistic yeasts is necessary to ensure that their normal growth and function are maintained. Moreover, Zhao et al. reported that the heterologous expression of flagellin in *S. cerevisiae* significantly induced resistance in the host plant and improved the biocontrol efficacy of the yeast against *B. cinerea*, which suggests that the heterologous expression of elicitors in yeasts may be an effective strategy to improve the biocontrol efficacy [112].

#### *4.3. Commercial Application*

The commercialization of an antagonistic yeast is a long and costly process requiring extensive testing of toxicology and biocontrol efficacy under commercial conditions. Encouragingly, over the past few decades, a few antagonistic yeasts have been developed and commercialized (Table 1). Aspire (based on *C. oleophila*) and YieldPlus (based on *C. albidus*) are the first-generation commercial antagonistic yeasts [27]. They were available on the market for several years, but they have now been withdrawn due to reasons such as difficulties in market development, low profitability, and inconsistent and low efficacy under commercial conditions [29]. After that, Nexy (another product based on *C. oleophila*) was developed for controlling decay on pome, citrus, and banana, and it was approved for registration throughout the European Union in 2013. Shemer (based on *M. fructicola*) was originally registered in Israel and was successfully used for managing pre- and postharvest diseases on various fruits and vegetables [113]. It was subsequently acquired by Bayer CropScience (Germany) and then sublicensed to Koppert Biological Systems (the Netherlands) to expand its sales [114]. Moreover, Bio-ferm, an Austrian company, developed two products based on *A. pullulans* strains DSM 14940 and DSM 14941, Blossom Protect (Boni-Protect) and Botector. With the mode of action of competition for nutrients and space, Blossom Protect is used to control postharvest decay caused by several fungal pathogens in pome fruit, while Botector is mainly used against gray mold in grape, strawberry, and tomato.


**Table 1.** Antagonistic yeast-based commercial products developed for the management of postharvest pathogens (adapted from [9] and [115] with modification).

#### **5. Conclusions and Perspectives**

The environmental pollution and health hazards caused by chemical fungicides have attracted increasing attention from regulatory agencies and consumers, and there is now global interest in reducing or eliminating the use of chemical fungicides. As a potential substitute for chemical fungicides, antagonistic yeasts have been extensively studied over the past few decades, and considerable progress has been made regarding the identification and development of antagonistic yeasts. However, so far, the use of antagonistic yeast alone is still insufficient to completely replace chemical fungicides. There remain many aspects of antagonistic yeasts that could be improved, even for the few commercially available antagonistic yeasts.

Although the application of antagonistic yeasts is limited by many obstacles, there is still great potential for their improvement and development. Due to the regulatory restrictions on chemical fungicides and the declining consumer acceptance of them, it is foreseeable that the use of chemical fungicides will be gradually decreased or even discontinued. The reduction in available products on the market and the increasing demand for safe and effective antifungal products provide opportunities for the development of antagonistic yeast products. The biocontrol efficacy of antagonistic yeasts could be further improved in the future through a variety of strategies, such as combining an antagonistic yeast with a chemical or physical treatment, using multiple antagonistic yeasts, and genetically altering antagonistic yeasts. Moreover, the advancement of molecular biotechnologies and the emergence of "omics" technologies are providing powerful tools for the development and application of antagonistic yeasts.

To promote the commercial application of antagonistic yeasts, efforts can be made in the following aspects: (a) the full verification of biosafety; (b) the in-depth exploration of the involved mechanisms of action; (c) the enhancement and maintenance of biocontrol efficacy under commercial conditions; (d) the development of broad-spectrum antifungal products; (e) the extension of shelf-life; (f) the control of cost and the development of the market; and (g) the understanding of the complex interactions between the components of the biocontrol system, including the antagonistic yeast, pathogen, host, natural microbiome, and environment. Furthermore, gene editing has been considered to be a potentially effective strategy to improve the performance of antagonistic yeasts, though genetically modified microorganisms (GMOs) are restricted due to government policies and low consumer acceptance at present.

**Author Contributions:** S.T. designed the manuscript content; X.Z. and B.L. wrote the manuscript under the coordination of S.T.; Z.Z. and Y.C. provided some data. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work is supported by National Natural Science Foundation of China (31930086; 31530057; 31722043) and The National Key Research and Development Program of China (2016YFD0400902).

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Review Saccharomyces boulardii***: What Makes It Tick as Successful Probiotic?**

#### **Pedro Pais 1,2,**† **, Vanda Almeida 1,2,**† **, Melike Yılmaz 1,2 and Miguel C. Teixeira 1,2,\***


Received: 9 May 2020; Accepted: 2 June 2020; Published: 4 June 2020

**Abstract:** *Saccharomyces boulardii* is a probiotic yeast often used for the treatment of GI tract disorders such as diarrhea symptoms. It is genetically close to the model yeast *Saccharomyces cerevisiae* and its classification as a distinct species or a *S. cerevisiae* variant has long been discussed. Here, we review the main genetic divergencies between *S. boulardii* and *S. cerevisiae* as a strategy to uncover the ability to adapt to the host physiological conditions by the probiotic. *S. boulardii* does possess discernible phenotypic traits and physiological properties that underlie its success as probiotic, such as optimal growth temperature, resistance to the gastric environment and viability at low pH. Its probiotic activity has been elucidated as a conjunction of multiple pathways, ranging from improvement of gut barrier function, pathogen competitive exclusion, production of antimicrobial peptides, immune modulation, and trophic effects. This review summarizes the participation of *S. boulardii* in these mechanisms and the multifactorial nature by which this yeast modulates the host microbiome and intestinal function.

**Keywords:** *Saccharomyces boulardii*; *Saccharomyces cerevisiae*; probiotics; gastrointestinal tract

#### **1. Introduction**

Probiotics are defined as live organisms which, when administered in adequate amounts, confer a health benefit to the host, independently of where the action takes place and of the type of administration. They are normally recommended to help strengthen host systems, for example the gastrointestinal (GI) tract, and assist in the recovery of certain diseases. According to this definition, probiotics in food must contain at least 10<sup>6</sup> CFU/g of viable and active microorganisms, while freeze-dried supplements have shown good results with 10<sup>7</sup> to 10<sup>11</sup> viable microorganisms per day [1–5]. It is also preferable that these are of human origin and that they cannot transfer any antibiotic resistance, pathogenicity or toxicity factors [4].

The most commonly used probiotics comprise lactic acid producing bacteria (*Lactobacillus* spp., *Bacillus* spp., *Bifidobacterium* spp., *Streptococcus* spp., and *Enterococcus* spp.) that are found in the human gastrointestinal tract, usually ingested in fermented foods [4]. These probiotics can be used by themselves or combined with each other, although it should be noted that not all combinations are stable and different strains of the same probiotic bacteria can have different capabilities or enzymatic activities, even if they belong to the same species [4,6]. Probiotic properties widely differ between species, strains or even between strain variants, which means these properties can be strain/variant-specific [4].

The ability of a given organism to display probiotic activity is also dependent on its ability to compete for a host niche. Probiotics must compete with pathogens that adhere specifically to host cells,

such as those of the GI tract, including *Helicobacter pylori* or *Clostridium di*ffi*cile*, but also *Borellia* spp., *Treponema* spp. or *Spirilium* spp. [2]. This means that the competition between probiotic microorganisms and pathogens is dependent on habitat-related idiosyncrasies [2]. Host factors can also influence the effectiveness of a probiotic. Genetic factors, baseline immune functions or microbiome diversity vary among individuals, which together with environmental factors (e.g., diet or stress) account for unique backgrounds where the same probiotic will have distinct outcomes [4].

Several bacteria have been identified as probiotics and their modes of action scrutinized to some extent, but yeasts may also exhibit probiotic properties. The baker's yeast *Saccharomyces cerevisiae* does not seem to present significant advantageous attributes for human health [1]. On the other hand, the closely related *Saccharomyces boulardii* is effective in complementing the treatment of acute gastrointestinal diseases such as diarrhea or chronic diseases such as inflammatory bowel disease (IBD) [7,8]. To date, this is the only yeast used as a probiotic [4] and its probiotic properties are supported by scientific evidence from the *S. boulardii* CNCM I-745 (or *S. boulardii* Hansen CBS 5926) strain produced by Laboratoires Biocodex, highlighted by more than 80 randomized clinical trials [1]. Nevertheless, the efficacy of this strain cannot be extrapolated to other strains, like *S. boulardii* CNCM 1079 [1].

In this review, current knowledge on *S. boulardii* traits that support its probiotic nature and the correlation with distinctive features when compared with the non-probiotic *S. cerevisiae* will be explored. Focus will be given on reviewing the biology, genetics, ability to colonize the human gut and compete with gastrointestinal pathogens as features that may underlie the probiotic activity of *S. boulardii*. Unanswered questions, mostly related to the genetic basis underlying the probiotic phenotype, are discussed.

#### **2.** *S. boulardii* **and** *S. cerevisiae***: Similar but Di**ff**erent**

The budding yeast known as *S. boulardii* is usually referred to as a distinct species within the *Saccharomyces* genus, despite being genetically close and sharing a similar karyotype to the model yeast *S. cerevisiae* [9–11]. Molecular typing studies resorting to pulsed-field gel electrophoresis (PFGE), randomly amplified polymorphic DNA-polymerase chain reaction (RAPD-PCR), and restriction fragment length polymorphisms (RFLP) of non-transcribed spacer (NTS) or internal transcribed spacer (ITS) reveal that *S. boulardii* strains from distinct origins all belong to a clearly delimited cluster within the *S. cerevisiae* species, arguing that they should be considered different strains of the same species [10,12]. Likewise, a DNA/RNA hybridization spotted microarrays study also concluded that *S. boulardii* is a strain of *S. cerevisiae* that has lost all intact Ty1/2 elements rather than a different species [13], while another study identified Ty1/3/4 as absent elements, but not Ty2/5 [11]. Phylogenetic analysis also shows that *S. boulardii* clusters are closely related to *S. cerevisiae* wine strains [11]. In spite of such similarities, microsatellite polymorphisms may provide a way to differentiate both species and identify *S. boulardii* properly [14,15].

Despite the striking relatedness in molecular phylogeny and typing, *S. boulardii* does possess identifiable distinct traits and is physiologically and metabolically distinct from *S. cerevisiae* (Table 1). Namely, *S. boulardii* is incapable of producing ascospores, switching to haploid form, or using galactose as carbon source [11,16–19]. It is more resistant to temperature and acidic stresses, but less resistant to bile salts [12,18].


**Table 1.** Metabolic, physiological and genetic features of *S. cerevisiae* and *S. boulardii*. The data shown was collected from several studies [11–13,16–21].

#### **3.** *S. boulardii* **Genomic Variations Provide Hints for Its Physiological Properties**

*S. boulardii* and *S. cerevisiae* genomes were found to differ in internal regions of lower copy number in three chromosomes: chromosome I (*PRM9*, *MST28*, *YAR047C*, *YAR050W*, *CUP1*, *YAR060W* and *YAR061W*); chromosome VII (*YGL052W* and *MST27*) and chromosome XII (*ASP3* and *YLR156W*). *PRM9*, *MST27* and *MST28* genes encode nonessential membrane proteins specific to the *Saccharomyces sensu stricto* species [18]. *YAR050W* encodes a lectin-like protein that participates in flocculation; Asp3 is a nitrogen catabolite-regulated cell wall L-asparaginase II. *CUP1* had a two times lower number of copies than the average for *S. cerevisiae* species, possibly causing the increased sensitivity to copper in *S. boulardii* when compared to other *S. cerevisiae* strains [18].

Within genes with higher copy number, two functions are well represented: protein synthesis (*RPL31A*, *RPL41A*, *RPS24B*, *RPL2B* and *RSA3*) and stress response (*HSP26*, *SSA3*, *SED1*, *HSP42*, *HSP78* and *PBS2*). It is possible that these genes aid in increased growth rate and pseudo-hyphal switching and in higher resistance to high pH [18]. Duplicated and triplicated genes mostly encode stress response proteins, elongation factors, ribosomal proteins, kinases, transporters and fluoride export, which might aid in *S. boulardii* adaptation to stress conditions [11]. Altered gene copy number and mutations when compared to *S. cerevisiae* in the *SDH1* and *WHI2* genes was associated with increased acetic acid production by *S. boulardii*, correlated with antimicrobial activity [22].

*S. boulardii* was shown to display enhanced ability for pseudo-hyphal switching during nitrogen starvation compared to other *S. cerevisiae* strains [18]. Several genes related to pseudo-hyphal growth have considerably different number of copies: *CDC42*, *DFG16*, *RGS2*, *CYR1* and *CDC25* have higher copy number; *STE11*, *SKM1* and *RAS1* have lower copy numbers [18]. Some of these genes are involved in cyclic adenosine monophosphate (cAMP) pathways, suggesting its hyperactivation can lead to increased pseudo-hyphal growth. As a possible consequence, *S. boulardii* ability to create pseudo-hyphae was observed to be faster and more extensive than *S. cerevisiae* [18].

Variations in the number of repetitive sequences within flocculation genes was also identified in *S. boulardii*, namely in *FLO1*. The encoded flocculin was found to harbor additional copies of residue repeats when compared with most *S. cerevisiae* strains [11]. The Flo8 protein was also found to differ between *S. boulardii* and some *S. cerevisiae* strains where a point mutation results in a truncated protein (including in the reference strain S288c), resulting in defective flocculation and adhesion [23]. Other flocculation genes (*FLO10* and *FLO11*) detected in *S. boulardii* were not found to harbor differences in the copy number and period length of the repeats [11]. The higher maximum number of repeats (e.g., *FLO1*) in *S. boulardii* may affect its adhesion and flocculation ability, as well as sensitivity to stress [11].

Several studies have shown that *S. boulardii* is unable to use galactose as a carbon source, despite harboring all galactose uptake and fermentation genes [16,17,19]. Some studies have proposed that it is able to assimilate, but not ferment, galactose, possibly due to energy requirements [24,25]. More recently, a mutation in the gene *PGM2* was also associated with the inefficient use of galactose [17]. *S. boulardii* is also unable to use palatinose, possibly related with the absence of 3 isomaltase encoding genes (*IMA2*, *IMA3* and *IMA4*), which is involved in palatinose uptake and metabolism [11,19].

#### **4. Adaptation to Host Environment**

Probiotics must be able to endure in adverse conditions. *S. boulardii* optimal growth temperature corresponds to the human host temperature (37 ◦C), while *S. cerevisiae* grows optimally at 30 ◦C. *S. boulardii* is also more resistant to very high temperatures keeping 65% viability after one hour at 52 ◦C, while *S. cerevisiae* loses viability down to 45% [12].

The main obstacles in the stomach are the very acidic pH (2 to 3) and the presence of proteases such as pepsin that kill most microorganisms, including probiotics that enter the organism [26]. Diseases like hypochlorhydria decrease the bactericide properties of the stomach and make the patient more susceptible to infections by *H. pylori* and *Salmonella* spp. and to migrations of pathogenic microorganisms to the small intestine where they establish themselves [26]. In the case of the small intestine, main stressors include the high concentrations of bile salts, pancreatic enzymes, hydrolytic enzymes, pancreatin, organic acids, the integrity of the epithelial and brush border, the immune defense and the native microbiome and its secondary metabolism products (H2S, bacteriocins, organic acids) [27]. Bile salts are detergents produced in the liver from cholesterol and secreted to the intestine to improve nutrient absorption. As detergent like molecules, bile salts can be toxic to GI tract microorganisms by disrupting their cellular membrane lipid bilayer structures [12]. However, some probiotics are able to resist degradation by hydrolytic enzymes and bile salts [6]. For example, *S. boulardii* and *Bacillus coagulans* remain viable after exposure to simulated gastric juice containing pepsin and hydrochloric acid. These probiotics were also seen to be stable to the impact of bile salts [6]. *Bacillus clausii* was partially resistant to these conditions [6]. On the other hand, most *Lactobacillus* and *Bifidobacterium* spp. have reduced viability under exposure to gastrointestinal agents such as pepsin, hydrochloric acid and bile [6].

In vitro testing of probiotic formulations consisting of *S. boulardii* and bacterial probiotics (*Lactobacillus* spp. and *Bifidobacterium* spp.) highlighted the ability of *S. boulardii* to survive GI tract conditions. *S. boulardii* was able to survive after incubation in a gastric-like environment and in an intestinal environment (bile salts, pancreatin, pH 7.0) for 3 h, whereas the viability of the bacterial probiotics was severely impaired [6]. *S. boulardii* is also more resistant to a gastric environment than *S. cerevisiae*, while the viability of both species in an intestinal environment (sodium chloride, pepsin, pancreatin, pH 8.0) is not affected after 1 h [12]. Accordingly, 1 h was enough to show that *S. boulardii* is more resistant to low pH than *S. cerevisiae*, particularly at pH 2.0 [18]. Interestingly, although *S. boulardii* can survive the GI environments, its viability is significantly increased for 2 h if encapsulated by a double layer with sodium alginate and gelatin [28]. Tolerance displayed by *S. boulardii* to bile salts has also been tested. Surprisingly, *S. cerevisiae* is more tolerant to bile salts than *S. boulardii*. However, after 1 h, both species show a tolerance threshold bellow what would be considered as resistance to bile salts [12].

Dynamic modelling of the stomach and small intestine conditions also showed *S. boulardii* to be resilient to gastric and lower intestinal conditions, while modelling of the colon environment revealed the yeast is not able to colonize the colon, but had an individual-dependent effect in the microbiotic profile [29]. Other studies also point to the inability of *S. boulardii* to colonize the gut, suggesting that this yeast does not strongly adhere to intestinal epithelial cells and is quickly removed from the gastrointestinal system in healthy individuals [18]. However, it has been shown to colonize the intestine of gnotobiotic mice after a single administration [21]. This may mean that, although *S. boulardii* can colonize the gut, competition with intestinal microbiome is limiting [18]. Indeed, both *S. boulardii* and other *Saccharomyces* strains were shown to be unable to remain attached to human and mouse epithelial cells, in vitro and in vivo, respectively [18]. However, they do adhere to Caco-2 cells through an extracellular factor, probably secreted mucus [18]. Colonization of the gut was observed to be dependent, both in mice and human, on repeated administration over several days [20,21,30]. Moreover, administration of ampicillin increased *S. boulardii* cell concentration [20], reinforcing the notion that competition with intestinal microbiome plays a relevant role in the establishment of this yeast.

#### **5. Mechanisms of Action**

The gut microbiome is responsible for a multitude of roles, including protection against pathogen colonization, epithelial barrier maintenance or modulation of immune activity [31]. The mechanisms by which gut microbiome homeostasis is maintained are not yet fully understood. Probiotics are believed to display a variety of mechanisms: antitoxin effects, physiological protection, modulation of the normal microbiome, metabolic regulation and signaling pathway modification, nutritional and trophic effects, immune system regulation, pathogen competition, interactions with the brain-gut axis, cellular adhesion, cellular antagonism and mucin production [1,4,31]. *S. boulardii* has been described as participating in a number of these effects as part of its probiotic activity (Figure 1). The genetic basis and mechanistic details that underlie these observations are not fully understood and their clarification could be key to better exploit this yeast and how to potentiate general probiotic activity.

**Figure 1.** Overview of the main modes of action that support S. boulardii probiotic activity in the intestinal epithelium. Studies have described the outcome of S. boulardii administration in pathogen exclusion, antimicrobial properties, immune modulation, and trophic effects. The genetic basis and mechanistic details that underlie these observations are not fully understood and their clarification could be key to better exploit this yeast and how to potentiate general probiotic activity. Pathogen exclusion is mainly achieved by pathogen binding to the yeast cells, rather than competition for epithelial binding sites with the pathogens. The yeast cell wall components responsible for the binding, the correspondent pathogenic receptors and the binding dynamics have not been fully investigated. Antimicrobial action is achieved, at least partially, by the secretion of still unknown proteins with antimicrobial effects. The genes that code for these proteins have not been identified and could provide further clues on the mode of action of S. boulardii. Immune modulation and the effect of S. boulardii on inflammatory pathways has been uncovered to some extent. The mechanistic insights and dynamics of S. boulardii interaction with immune cells still need to be ascertained to better understand the yeast action in the immunological function. Multiple trophic effects have been described to be stimulated by S. boulardii on intestinal epithelial cells. Some pathways have been elucidated, although the multitude of trophic effects suggests concerted action and crosstalk between yeast and host cell sensory pathways.

#### *5.1. Modulation of The Normal Microbiome*

Modulation of the normal microbiome may be favored directly by transiting probiotics which produce antimicrobial substances, or indirectly contribute to immune modulation or gut barrier function [31]. The use of probiotics has typically been applied to reestablish the normal gut microbiome upon dysbiosis. Gut dysbiosis refers to changes in the microbiome's quantitative and qualitative composition. These changes may contribute to a disease state frequently associated to inflammation and can be a result of antibiotic-associated diarrhea, acute infectious diarrhea or IBD [31,32]. Probiotics treatment helps to stabilize the gut microbial community and lead to an improved disease outcome [32]. While some probiotics may become a part of the microbiome, others simply pass through the GI tract and modulate or influence the existing microbiome before exiting the body [31].

Several factors can have deleterious effects on the gut microbiome and hinder its protective role to the host epithelial lining, such as antibiotic use or surgery [1]. This may result in host susceptibility to colonization by pathogens until the normal microbiome is reestablished, which can take several weeks [33]. *S. boulardii* helps to restore the normal microflora in this type of patient and the use of probiotics as modulators of the normal microbiome through colonization during the susceptibility period may work as a surrogate until the normal microbiome is reestablished [34].

#### *5.2. Antimicrobial Activity*

Antagonism against pathogens can be achieved by colonization and exclusion of pathogens, modulation of metabolic and signaling pathways, production of inhibitory compounds or immune modulation [31]. Competition is one of the main mechanisms associated with probiotic activity against gut pathogens: consumption of nutrients by probiotics results in nutrient limitation for pathogenic organisms [31,35]. On the other hand, the ability of probiotics to grow and colonize the gut can lead to a decrease of the gut pH due to the production of metabolites, leading to stressful conditions for pathogens [35]. A possible role for *S. boulardii* in managing pathogenic activity was associated with a protective effect of *S. boulardii* against pathogenic bacteria in yeast-treated mice, although the mode of action is not associated with a reduction of the pathogenic population [21], as well as with another study which observed a protective effect against *Candida albicans* in a murine model [36].

The production of compounds with antimicrobial activity is yet another major mode of action of probiotics. Several components of the probiotic metabolome, such as organic acids, bacteriocins, hydrogen peroxide, diacetyl, or amines limit the growth of pathogenic bacteria [31]. In particular, bacteriocins play a crucial role in the antimicrobial action of probiotic bacteria, especially *Lactobacillus* spp. [35,37–40]. As for the production of antimicrobial substances by other probiotics, *S. boulardii* possibly secretes proteins that reduce*Citrobacter rodentium*adhesion to host epithelial cells by modulating virulence factors [41]. It also displays antimicrobial activity by secreting 54-kDa, 63-kDa and 120-kDa proteins that cleave microbial toxins or reduce cAMP levels. *S. boulardii* can block toxin receptors or function as a decoy receptor for toxins. The 54-kDa serine protease produced by *S. boulardii* cleaves toxins A and B from *C. di*ffi*cile* and the enterocytic receptor to which the toxins bind, which causes inflammation, fluid secretion, mucosal permeability and injury in the intestines [42,43]. Other mechanisms that *S. boulardii* uses against *C. di*ffi*cile* infection are growth inhibition and decreased toxin production due to secreted factors and stimulation of host mucosal disaccharidase activity [44,45]. Another study refers to the ability of *S. boulardii* to inhibit *Escherichia coli* surface endotoxins by dephosphorylation. A 63-kDa alkaline phosphatase targets the lipopolysaccharide (LPS) and contributes to decreased tumor necrosis factor α (TNF-α) cytokine levels [46]. *S. boulardii* also produces a 120-kDa protein that decreases the chloride secretions stimulated by cholera toxin by reducing cAMP levels [47]. *S. boulardii* is also able to adhere to cholera toxin via its cell wall, thus blocking its toxic effects [48]. Despite these observations, the sequencing of *S. boulardii* genomes did not provide a clear identification of the genes encoding these 54-kDa, 63-kDa and 120-kDa proteins [25].

*S. boulardii* also confers protection against the lethal toxin produced by *Bacillus anthracis*. The bacterium causes ulcerative lesions from the jejunum to cecum and uses its toxin to disrupt intestinal epithelium integrity, causing mucosal erosion, ulceration and bleeding [49]. The protective effect of *S. boulardii* is associated with maintenance of barrier function and reduction of harmful physiological responses elicited by the toxin, such as formation of stress fibers [50]. The protective effect is achieved by release of proteases and cleavage of the lethal toxin [50].

Some *S. boulardii* strains are able to produce high concentrations of acetic acid, which was found to exert an inhibitory effect in *E. coli* [22]. In turn, the decrease in pH due to acetic acid production is essential for the antimicrobial activity of short-chain organic acids. The combined effect of high acetic acid concentration and lower pH may be an additional mechanism that makes *S. boulardii* an effective probiotic. Moreover, acetic acid is produced under aerobic conditions by *S. boulardii*. Due to the radial oxygen gradient between the epithelial surface (high oxygen levels) and the center of the gut lumen (low oxygen levels), microorganisms colonizing the epithelial surface have greater availability of oxygen [51]. Since acetic acid is produced under aerobic conditions by *S. boulardii*, its production should be higher near the epithelial surface. During antibiotic treatment and pathogen infection, oxygen concentration also increases in the gastrointestinal tract [52,53], which could support the antimicrobial action of *S. boulardii*.

#### *5.3. Adhesion*

In order for the host not to mechanically eliminate the gut microbiome, it is crucial that its components adhere to host surfaces [1,4]. Some probiotics express surface adhesins that mediate the attachment to the mucous layer by recognizing host molecules such as transmembrane proteins (integrins or cadherins) and extracellular matrix components (collagen, fibronectin, laminin or elastin) [1,4]. Probiotics can also influence the production of mucin and the barrier function of the intestine, thus hindering adhesion and consequent invasion of pathogenic microorganisms [54].

Mucin is produced by epithelial cells to avert adhesion by pathogenic bacteria, whereas successful probiotics should be able to adhere to the intestinal mucous, as is the case of *S. boulardii* [13,55]. The adhesion of *S. boulardii* to the mucus membrane contributes to reducing the availability of binding sites for pathogens [13]. Five *S. cerevisiae* cell wall proteins (encoded by *CIS3*, *CWP2*, *FKS3*, *PIR3* and *SCW4*) were found to mediate adhesion of the yeast cells to the pathogenic bacteria *E. coli*, *Salmonella enterica* serovar *typhimurium* (*S. typhimurium*) and *Salmonella enterica* serovar *typhi* (*S. typhi*) [55]. Other studies have shown that these bacteria are also bound to *S. boulardii* [55–58]. Additionally, *S. boulardii* also inhibits *C. di*ffi*cile* adhesion to epithelial cells and displays inhibitory activity on *Entamoeba histolytica* adhesion to erythrocytes [59,60]. This interaction limits the ability of pathogens to bind directly to the intestinal receptors and proceed with host invasion. In fact, *S. boulardii* hinders epithelium invasion by *S. typhimurium* due to steric hindrance caused by its larger size as compared to bacteria [61]. As *S. boulardii* does not significantly bind to epithelial cells of healthy individuals and is quickly flushed out, pathogens bound to *S. boulardii* are possibly flushed together with the yeast cells [13,31,55]. However, *S. boulardii* does have several flocculation genes required for protection against environmental stress and biofilm formation [11]. The characterization of this gene family in the context of host adhesion and colonization could provide further insight on the probiotic features of *S. boulardii*.

The ability of *S. boulardii* to bind bacterial pathogens has been associated with the presence of mannose residues in the yeast cell wall [56]. This is a similar mechanism to the previously characterized adhesion of bacterial pathogens to the epithelial surface via mannose residues [62], which is the basis for the addition of exogenous sugars as a strategy to inhibit pathogen adhesion [63]. Cell wall mannan oligosaccharides are a common feature in yeast, but the affinity between *E. coli* and *S. boulardii* is higher than with *S. cerevisiae* [56]. Further investigations revealed that bile salts decrease adhesion of bacteria to yeast cells [55], which can have relevant implications for yeasts as successful probiotics. Accordingly, bile salts also decrease the adhesion of probiotic bacteria to intestine epithelia due to diminished surface hydrophobicity and higher surface potential [64], bolstering how important it is for probiotic microorganisms to evolve adaptation strategies within the host.

#### *5.4. Immune Modulation*

Metabolites produced by the gut microbiome can perform immunomodulatory and anti-inflammatory functions that stimulate immune cells. This ability arises from the interaction between the probiotics and the epithelial cells, dendritic cell monocytes, macrophages and/or lymphocytes [1,31]. Probiotics also promote enhanced phagocytic activity, cell proliferation and production of secretory immunoglobulins IgA and IgM [65].

*S. boulardii* can modulate immunological function by acting as a stimulant or a pro-inflammatory inhibitor. It is capable of modulating the inflammatory process upon *S. typhimurium* infection by decreasing the levels of the pro-inflammatory molecules such as cytokine interleukin 8 (IL-8), mitogen activated protein (MAP) kinases and the (nuclear factor kappa B) NF-κB signaling pathway [58]. An inhibitory effect of *S. boulardii* over MAP kinases and IL-8 levels upon *C. di*ffi*cile* infection was also observed [66]. Likewise, *S. boulardii* contributes to increasing the levels of anti-inflammatory cytokines (IL-4 and IL-10) and decreasing pro-inflammatory cytokines (IL-1β) upon infection with *E. coli* and *C. albicans* [67]. On the other hand, *S. boulardii* was associated with increased IgA and IgG levels in serum in response to *C. di*ffi*cile* toxins A and B [68,69]. *S. boulardii* was also found to attach to the surface

of dendritic cells [70] and modulate the expression of toll-like receptors (TLRs) and cytokines [70–72]. Moreover, *S. boulardii* also causes the imprisonment of T helper cells in mesenteric lymphatic nodes, reducing inflammation [73].

Another study found that in the early phase of *S. typhimurium* infection, *S. boulardii* induces pro-inflammatory cytokine production (interferon-γ—IFN-γ) and represses the production of anti-inflammatory cytokines (IL-10) in the small intestine, but increases the levels of both cytokines in the cecum [57]. This suggests that *S. boulardii* can differentially modulate immune activity through the GI tract [57]. Overall, probiotics may be able to persistently modulate both the innate and adaptive immune responses either locally or systemically [1,31]. The data from several studies indicates that *S. boulardii* plays a pivotal role in immune modulation against the most common GI tract pathogens.

#### *5.5. Trophic E*ff*ects*

*S. boulardii* is a modulator of enzyme activity required to maintain a healthy gastrointestinal tract. It exerts trophic effects such as stimulation of brush border membrane digestive enzymes and nutrient transporter activity [74]. Several studies have shown a wide array of trophic effects stimulated by *S. boulardii*: brush border sucrase, lactase, and maltase activities [44,75–78]; iso-maltase activity [78]; glucoamylase and *N*-aminopeptidase activity [76]; leucine-aminopeptidase activity [79]; α,α-trehalase activities in the endoluminal fluid and intestinal mucosa; brush border α-glucosidase [80]; spermine, spermidine and putrescine levels in rat jejunal mucosa [75,77]; adenosine triphosphatase, γ-glutamyl transpeptidase, lipase, and trypsin activities and TNF-α, IL-10, transforming growth factor beta (TGF-β), and secretory IgA [5]; diamine oxidase activities, brush border sodium/glucose cotransporter 1 expression and sodium-dependent glucose uptake [74,77]; GRB2-SHC-CrkII-Ras-GAP-Raf-ERK1,2 transduction pathway in rats and decreased p38 MAPK and NF-κB [81–83].

Probiotics can modulate short chain fatty acids (SCFA: acetate, propionate, or butyrate) and/or branched-chain fatty acid (BCFA: iso-butyrate, 2-methylbutyrate, or isovalerate) synthesis. SCFAs have a complex role in the physiological and biochemical functions in different tissues (intestine, liver, adipose, muscle and brain). *S. boulardii* assists in reestablishing SCFA levels, which are depressed during disease [84,85]. Acetate and butyrate are major SCFAs in intestinal epithelial cells, playing a role in barrier function, anti-inflammatory and immune modulation pathways [86,87]. A study reported that a short-term treatment (6 days) with *S. boulardii* diminishes the incidence of diarrhea in patients receiving enteral nutrition by increasing SCFA levels, particularly butyrate [84]. SCFAs can also present antimicrobial activity, and a study probing several *S. boulardii* and *S. cerevisiae* strains for inhibitory effects in *E. coli* described the production of acetic acid exclusively by *S. boulardii* as an antimicrobial mechanism [22]. Moreover, acetate also stimulates T regulatory cells, induces mucus secretion gene expression, inhibits proinflammatory cytokine CXCL8 and serves as a substrate for the production of butyrate by the microbiome [22].

The activity of many digestive enzymes (sucrase-iso-maltase, maltase-glucoamylase, lactase-phlorizin hydrolase, alanine aminopeptidase and alkaline phosphatase) and nutrient transporters (sodium-glucose transport proteins) may be induced by polyamines secreted by *S. boulardii* [74]. *S. boulardii* secretes polyamines that promote RNA binding and stabilization and, hence, growth and protein (lactase, maltase, sucrase, among others) synthesis [74]. These molecules are also able to shield lipids from oxidation and boost SCFA activity. Polyamines may also affect kinase activities and external signal transduction pathways, therefore modulating the GRB2-SHC-CrkII-Ras-GAP-Raf-ERK1,2 and the PI3K pathways [74]. They can also aid the generation of specific transcripts by interacting with DNA [74]. All of these polyamine functions lead to a general polyamine-triggered metabolic activation in order to regenerate brush border damage and maturation of enterocytes [74,75,80]. Not only does *S. boulardii* induce the enzymatic activities of lactase-phlorizin hydrolase, α-glucosidases, alkaline phosphatases and aminopeptidases, but it also increases glucose intestinal absorption, one of the products of lactose degradation [74]. Production of lactase by the host, partially stimulated by *S. boulardii*, mediates lactose degradation thus alleviating lactose intolerance.

#### **6.** *S. boulardii* **Safety and Clinical E**ffi**cacy**

Although many probiotics are documented as safe, common safety issues regarding the use of probiotics include: transfer of antibiotic resistance genes, translocation of live organisms from the intestine to other sites of the body, persistence in the intestine and development of adverse reactions [1]. Most of these concerns have been dismissed when evaluating *S. boulardii* safety. *S. boulardii* is not known to acquire resistance genes, unlike bacterial probiotics such as *Lactobacillus* spp. [88,89]. Animal studies show that there is reduced translocation in the treatment with *S. boulardii* when compared with *S. cerevisiae* [90–92]. *S. boulardii* does not persist in the intestine after three to five days after discontinuation of the ingestion, according to pharmacokinetic studies [20]. The data available from 90 clinical trials assessing the efficacy and safety of *S. boulardii* has been thoroughly assessed elsewhere [1]. Randomized and controlled trials clearly show the absence of any serious adverse reactions, while only some presented moderate adverse reactions, such as constipation in patients with *C. di*ffi*cile* infection [93]. Although fungemia is viewed as a potential problem, there were no fungemia cases reported in clinical trials [1]. *S. boulardii*-associated fungemia was observed in patients with serious co-morbidity factors and central venous catheters, which responded well to fluconazole or amphotericin B therapy [91,94–96]. Importantly, *S. cerevisiae*-associated fungemia has a worse prognosis than that caused by *S. boulardii* [97].

Clinical trials have investigated the efficacy of *S. boulardii* in the improvement of several GI conditions' outcome. This yeast was seen to improve the outcome of several diarrhea diseases, including pediatric diarrhea, antibiotic-associated diarrhea, acute diarrhea, traveler's diarrhea caused by bacterial, viral or parasites, and enteral nutrition-related diarrhea [1,15,98]. *S. boulardii* also improves the outcome in patients suffering from *H. pylori* or *C. di*ffi*cile* infections by helping bacteria eradication, preventing relapses, reducing adverse reactions, and reducing treatment-associated diarrhea [1,15,98]. IBD is a prevalent GI tract disorder associated with inflammatory diarrheal diseases such as ulcerative colitis, pouchitis and Crohn's disease [1]. Clinical trial data points to a possible role of *S. boulardii* in reducing treatment relapses [1,15,98], which are frequent in these conditions, although further studies are required to reach compelling conclusions. Irritable Bowel Syndrome (IBS) symptoms also improve with *S. boulardii* administration. It is a condition frequently characterized by abdominal bloating, abdominal pain, and disturbed intestinal transit. These symptoms were shown to be alleviated in 50% of patients upon *S. boulardii* use [99].

#### **7. Conclusions**

*S. boulardii* is a probiotic yeast with proven efficacy in the treatment of GI conditions, especially when used as an adjuvant to antibiotic treatment. Present data indicate that the benefits of *S. boulardii* appear to be transient and independent of host gut colonization, differentiating its mode of action from other widely used bacterial probiotics. The absence of colonization appears to correlate with pathogen binding as a mechanism to halt pathogen colonization, rather than competitive exclusion due to yeast adhesion. Genomics studies have contributed to pinpoint distinct genome features that mainly confer on *S. boulardii* the ability to resist host stresses, conferring higher viability through GI passage than observed for other common probiotics. *S. boulardii* also elicits a complex immunomodulatory effect with roles in fine-tuning immunological pathways during pathogen infection or chronic diseases. This yeast also contributes to the homeostasis of the normal microbiome and plays a relevant role in modulating secretory functions by intestinal epithelial cells, thus benefitting nutritional requirements of the host. Overall, *S. boulardii* displays a multifactorial role as a probiotic, with proven efficacy and safety in alleviating the symptomology of a number of GI conditions. However, there is a significant knowledge gap between *S. boulardii* phenotypic effects and the underlying genetic basis, especially when compared to *S. cerevisiae*. What are the 54-kDa, 63-kDA and 120-kDA proteins secreted by *S. boulardii* that cleave microbial toxins or reduce cAMP levels? What are the proteins responsible for higher adhesion of *S. boulardii* to pathogenic bacteria, when compared to *S. cerevisiae*, especially considering the differences in flocculin encoding genes? What are the mechanisms that allow *S. boulardii*

to overcome the negative impact of bile salts during host adaptation? What are the proteins or cellular components that mediate immune recognition of *S. boulardii* and modulation of the immune response? These questions remain unanswered. Further research on the genetic basis of *S. boulardii* probiotic activity will certainly increase our understanding of this fascinating yeast, while providing important clues for the selection and optimization of even more powerful probiotic fungi.

**Author Contributions:** P.P. and M.C.T. delineated manuscript organization. P.P., with contributions from V.A. and M.Y., wrote the manuscript, under the coordination of M.C.T. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by "Fundação para a Ciência e a Tecnologia" (FCT) [Contract PTDC/ BII-BIO/28216/2017], as well as by Programa Operacional Regional de Lisboa 2020 [LISBOA-01-0145 -FEDER-022231—The BioData.pt Research Infrastructure]. Funding received by iBB from FCT (UIDB/04565/2020), and from Programa Operacional Regional de Lisboa 2020 (LISBOA-01-0145-FEDER-007317).

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **References**


Combination With Standard Antibiotics for Clostridium difficile Disease. *JAMA J. Am. Med. Assoc.* **1994**, *271*, 1913–1918. [CrossRef]


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Review* **Application of Probiotic Yeasts on** *Candida* **Species Associated Infection**

**Lohith Kunyeit 1,2,3, Anu-Appaiah K A 1,2 and Reeta P. Rao 3,\***


Received: 8 August 2020; Accepted: 19 September 2020; Published: 25 September 2020

**Abstract:** Superficial and life-threatening invasive *Candida* infections are a major clinical challenge in hospitalized and immuno-compromised patients. Emerging drug-resistance among *Candida* species is exacerbated by the limited availability of antifungals and their associated side-effects. In the current review, we discuss the application of probiotic yeasts as a potential alternative/ combination therapy against *Candida* infections. Preclinical studies have identified several probiotic yeasts that effectively inhibit virulence of *Candida* species, including *Candida albicans*, *Candida tropicalis*, *Candida glabrata*, *Candida parapsilosis*, *Candida krusei* and *Candida auris.* However, *Saccharomyces cerevisiae* var. *boulardii* is the only probiotic yeast commercially available. In addition, clinical studies have further confirmed the in vitro and in vivo activity of the probiotic yeasts against *Candida* species. Probiotics use a variety of protective mechanisms, including posing a physical barrier, the ability to aggregate pathogens and render them avirulent. Secreted metabolites such as short-chain fatty acids effectively inhibit the adhesion and morphological transition of *Candida* species. Overall, the probiotic yeasts could be a promising effective alternative or combination therapy for *Candida* infections. Additional studies would bolster the application of probiotic yeasts.

**Keywords:** *Candida albicans*; non-albicans *Candida* species; *Candida auris*; *Saccharomyces boulardii*; *Saccharomyces cerevisiae*; aromatic alcohols

### **1. Introduction**

The fermented foods are a rich source of beneficial microorganisms, and they have a long history of exhibiting health benefits, particularly *S. cerevisiae* and lactic acid bacteria (LAB). Their safety is evidenced by consumption of fermented foods and beverages over centuries. Today, it is well accepted that the rich microbial profile of fermented food provides more than just nutrition. For example, functional activity of microorganisms in food helps enhance the bio-availability of micronutrients, improving the sensory quality and shelf life of the food, degrading anti-nutritive factors (such as trypsin inhibitors and phytate degradation), enriching antioxidant and antimicrobial compounds, and fortifying health-promoting bioactive compounds [1,2]. These attractive microbial activities in the fermented foods have been a draw in the field of probiotics.

Characteristics of bacterial strains such as *Lactobacillus* and *Bifidobacterium* species have been extensively studied and commercially available as probiotic supplements. Yeasts, which are also common in fermented foods, remain largely unexplored for probiotic potential. We and other researchers have observed that yeasts that originate from fermented sources such as apple cider, wine, fermented coconut palm, and fermented dairy products survive the harsh condition of the gastrointestinal (GI) tract and retain the ability to attach to intestinal epithelium [3–5]. More recently, live bacteria have

been used in fecal transplants to prevent and/or treat several GI complications [6]. The probiotic bacteria, such as lactic acid bacteria (LAB) and *Bifidobacterium* species, have effectively treated several GI complications, including candidiasis [7,8]. However, other than *Saccharomyces boulardii,* potential probiotic yeasts such as *S. cerevisiae* and several other non-*Saccharomyces* yeasts are largely unexplored use as biotherapeutics, specifically for *Candida* infections. In reviewing the current literature here, we focus on the biotherapeutic potential and mechanism(s) of action of beneficial yeasts against *Candida* infections.

The vast majority of fungal infections are caused by *Candida albicans*, a polymorphic commensal yeast as well as some non-albicans *Candida* species. Disease range from superficial infections, such as cutaneous and mucosal, to life-threatening bloodstream infections (BSI), or invasive deep tissue infections. Superficial infections usually affect the nails, skin, and mucosal membrane of the host and are recalcitrant to treatment. For example, vulvovaginal candidiasis (VVC) has infected 75% of women population at least once in their lifetime. Furthermore, a small population (5–8%) suffers from at least four recurrent VVC per year [9].

Bloodstream infection (BSI) and other invasive*Candida* infections cause high morbidity and mortality especially among immune-compromised patients [10]. *Candida* species are the fourth-leading cause of nosocomial infections in the world, and *Candida* BSI attributes to 35% mortality rate in all the *Candida* associated infections [11]. Furthermore, the National Nosocomial Infection Surveillance System (NNIS), USA, has revealed total 27,200 nosocomial infections between January 1980 through April 1990, among these *C. albicans* and non-albicans *Candida* species were involved total 19,621 (72%) of the overall infections [12].

Though *C. albicans* is a major commensal yeast flora of the GI tract, non-albicans *Candida* species such as *Candida glabrata*, *Candida tropicalis*, *Candida parapsilosis,* and *Candida krusei* have been frequently identified in a healthy individual's gut. On the other hand, among 15–20 pathogenic non-albicans *Candida* species, *Candida glabrata*, *Candida tropicalis*, *Candida parapsilosis,* and *Candida krusei* are predominant constituting 35–65% of the overall infections [13]. As an opportunistic pathogen, certain groups of immune-compromised individuals have a higher susceptibility towards *Candida* infection. Invasive *Candida* infections are also closely associated with advanced medical techniques such as medical implants and stents [14]. For instance, the patients who are on antibiotic therapy and chemotherapy, central venous catheters, total parenteral nutrition, extensive surgery, burns, renal failure and hemodialysis, or mechanical ventilation are at a major risk for superficial and invasive *Candida* infections [14].

#### **2. Morphological Transition and Metabolic Flexibility Promote Virulence of** *Candida* **In Vivo**

As a polymorphic yeast, *C. albicans* and few non-albicans *Candida* strains, such as *C. tropicalis* and *C. glabrata,* exhibit multiple morphological structures such as yeast form, germ tubes, pseudo-hyphae, and/or hyphae that play a key role in the infection. For example, filamentous morphology is well-known for epithelial invasion and is primarily involved in biofilm formation [15]. Yeast form cells are planktonic and are important for dissemination. Once they attach, they initiate germ tubes, pseudo-hyphae, and/or hyphae that enhance adhesion to surfaces. Attachment to abiotic surfaces initiates biofilm formation. Biofilms on implanted medical devises may lead to invasive fungal infections—a major risk factor for *Candida* infection-associated mortality [16]. Attachment to live cells (such as epithelium) causes damage, evokes an immune response and ultimately gains access to deeper tissues. Therefore, the polymorphism of *Candida* is an important consideration in its infectious outcomes.

The host's innate immunity is a major factor in fungal clearance, normally through a process called phagocytosis where immune cells ingest and biochemically eliminate the pathogens [17]. However, the switch from yeast to filamentous form is a common escape mechanism of *Candida* species [18]. Therefore, *C. albicans* filament has less susceptibility for phagocytosis by innate immune cells than the yeast form [19]. In addition, metabolic flexibility of *C. albicans* facilitates colonization by adapting

to varying nutritional availability [20]. For instance, in case of *Candida* meningoencephalitis (*Candida* infection in brain tissue), glucose and vitamins are the major nutrient sources for the pathogen, while in liver, it utilizes glycogen as a nutrient source [9]. A study revealed that adaptation to alternative carbon sources such as lactate and other nutrient sources increased environment stress response and virulence [21]. All of these attributes make *C. albicans* and non-albicans *Candida* species a unique pathogen among the microbial community.

#### **3. Drug Resistance Is a Major Hurdle to Antifungal Therapy**

Antifungal drugs used to treat *Candida* associated infections, work either by killing or inhibiting the growth of *Candida* species. A sparse number of antifungal classes such as polyenes, azoles, and echinocandins are used depending on conditions of invasive *Candida* infections [22]. Multiple *Candida* strains have already developed resistance to these drugs making this a public health threat [23]. For example, surveillance data from health-care facilities revealed widespread fluconazole resistance among clinical isolates of both *C. albicans* and non-albicans *Candida* strains [24–26]. Azoles such as fluconazole is a first-line antifungal drug that is used extensively for therapy and prophylaxis against *Candida* infections. Several resistant mechanisms have been connected to drug-resistant *Candida* species including overexpression of drug efflux pumps, alteration in drug targets, and changes in membrane sterol composition [22]. The structural heterogeneity of *Candida* biofilm has a major significance in clinical context due to higher resistance against most antifungal agents. Furthermore, these drugs can be toxic for the patients with several side effects that include GI disturbances, hepatotoxicity, and neurotoxicity due to their target resemblance to its host cell, antifungal metabolism in liver and cross drug interaction in the host [27,28].

More recently, multi-drug resistant *Candida auris* has emerged as a "super bug" posing significant clinical challenges and a major threat to public health. *Candida auris,* is often involved in the nosocomial bloodstream infection world-wide [29]. *C. auris* has been shown to last in the hospital settings and spread from person-to-person by direct contact or contaminated surfaces [23]. In addition, *C. auris* is closely related with *Candida haemulonii* and is often misidentified as such. This requires a specialized laboratory method for identification [23], further delaying implementation of infection control. Therefore, now more than ever, there is an urgent need for effective alternatives to conventional modes to treat *Candida* infections.

Some attempts have been made to using specific diets that avoid high sugar-containing food such as bakery products, milk, and dairy product. The claim is that it reduces *Candida* colonization of the GI tract [30]. Intestinal overgrowth of *C. albicans* contributes to Crohn's disease that affects 1.6 million Americans [31,32]. *C. albicans* overgrowth is caused by an imbalance in the intestinal microbiota and host immune status. To restore the balance and modulate host immunity, foods rich in antioxidants and other nutritional supplements have been suggested [30,33]. More recently, studies on the human microbiome have opened new insight into the role of the resident gut microbiota in physical health and mental wellbeing. Applications of beneficial microbes as fecal transplants [34] or fermented milk products [35] for the treatment of irritable bowel syndrome (IBS) and irritable bowel disease (IBD) has gained traction. Here we discuss the potential of probiotic yeasts against *Candida* virulence and pathogenesis.

#### **4. Use of Probiotics as Biotherapeutics**

As stated by Hippocrates, "let food be thy medicine and medicine be thy food". Today, the idea of food and/or diet is not just extended towards mere survival or hunger satisfaction. The health-conscious population deeply cares about additional aspects including health improvement and prevention of diseases. In this context, functional food plays a significant role where, the concept of food has not only intended to provide humans with necessary nutrients, but also to prevent diseases and increase physical and mental well-being. Probiotic, considered as a functional food, is mostly consumed in the form of traditional fermented food products such as milk products, fermented vegetables, and meats [36].

Probiotics are defined as "live microorganisms which, when consumed in adequate amounts, confer health benefits on the host" [37]. The archived scientific documents have explained the diverse positive effects of probiotics on a wide range of diseases and disorders including lactose indigestion, diarrhea, immune modulation, inflammatory bowel syndrome, constipation, infection, allergy, serum cholesterol, blood pressure, and reduction of urinary tract infections [38]. In addition, the Human Microbiome Project by National Institute of Health (NIH), USA, changed the views on beneficial microbial research; it exposed the influence of gut microbiome and human health during various infections and disease conditions including, mental health.

#### **5. Interaction of Probiotics Yeast and** *Candida* **Species**

Several reports suggest that probiotic bacteria are effective against GI complications such as diarrhea, leaky gut syndrome, as well as *Helicobacter pylori* and *Clostridium di*ffi*cile* infections [39,40]. However, *Saccharomyces cerevisiae* var. *boulardii* is the only yeast currently available for human use as probiotics. Its efficacy against *Candida* has been explored previously. Specifically, pathogen-free mice that were infected with *C. albicans* and subsequently treated with *S. boulardii* prevented the translocation of *Candida* to internal organs [41–43]. These groups further confirmed that *S. boulardii* effectively reduced *C. albicans* translocation colonization and inflammation in in vivo models.

Clinical reports around the use of probiotic yeasts are limited. One study, reports that oral administration of *S. boulardii* to infants reduced the fungal colonization and invasive fungal infections [44]. Another study conducted in preteen children focused on the effects of probiotics against *Candida* infection. They used a probiotic cocktail of yeast and bacteria in combination with prebiotics and demonstrated a reduction in colonization of *C. albicans* [45].

#### **6. Probiotic Yeasts Exhibit Multiple Inhibitory Mechanisms against** *Candida* **Species**

Pre-clinical and/or clinical studies indicate that *S. boulardii* and other potential probiotic yeasts ameliorate complications associated with *Candida* infection by mechanisms outlined in Table 1. However, there was a lack of specific mechanistic insights on how these probiotic yeasts interact with *Candida* species especially in the context of a live host. Pathogens in GI tract induce necrosis and apoptosis of intestinal epithelia by reducing the production of mucin or its degradation. Pathogens also downregulate IgA and other proteins of the tight junction thereby increasing intestinal permeability [46,47]. *S. boulardii* has been shown to increase IgA production in *Clostridium di*ffi*cile* colitis and antibiotic-associated diarrhea in mice model [48]. *S. boulardii* also decreases epithelial necrosis, apoptosis, and increases the production of antioxidant enzymes such as superoxide dismutase, catalase, and glutathione peroxidase in mouse models in necrotizing enterocolitis in mice [49]. In addition, *S. boulardii* activates the intestinal epithelial restoration in GI tract [50]. Together these cellular responses may contribute to its beneficial properties and prevent *Candida* infection.

**Table 1.** List of probiotic yeasts and its mechanisms against virulence and pathogenesis of *Candida* species.



\* Potential probiotic yeast, not commercialized.

#### *6.1. Immunogenic Response and Anti-Virulence Ability of Probiotic Yeasts*

Since resistance to antifungal drugs has emerged as a significant problem, researchers have explored alternative means of treating recalcitrant fungal infections. Modulation of host immunity is one avenue that is being considered as an alternative [56,57]. For example, *S. boulardii* has been shown to reduce pro-inflammatory cytokines such as IL-1β and TNF, and increase anti-inflammatory cytokines IL-4 and IL-10 during *Candida* infection [42,58]. Other alternative therapies target virulence strategies such as adhesion and filamentation of *C. albicans* [59]. These maybe used to treat abiotic surfaces to deter microbes from binding. Probiotics also have the ability to inhibit virulence factors of the pathogen. We and others have demonstrated that cells, as well as the cell-free secretome of probiotic yeasts such as *S. boulardii, S. cerevisiae,* and a non*-Saccharomyces* yeast *Issatchenkia occidentalis* inhibit adhesion, filamentation, and biofilm development of *C. albicans* [52] and other non-albicans *Candida* species such as *C. tropicalis, C. krusei, C. glabrata,* and *Candida parapsilopsis.* Biofilms are complex multispecies structures that include *C. albicans* among other microbes [60,61]. Probiotics yeasts have been shown to be effective against fungal biofilms composed of *C. albicans* and non-albicans *Candida* species [53]; however, no studies have been focused on their efficacy on cross-kingdom biofilms. These studies implicated the involvement of yeast metabolite(s) in inhibiting adhesion and/ or morphological transition in vitro [53]. These studies also indicate that probiotic yeast affect a broad spectrum and not limited to *C. albicans;* rather, it can inhibit virulence across the *Candida* genus.

Cultured intestinal epithelial models such as Caco-2, Intestin 407 and HT-29 have been extensively used to study microbial interactions or host-microbe interactions. These cell lines recapitulate various features of the intestinal epithelial surface including the formation of villi, production of mucus, and antibodies such as IgA [62]. We and others have demonstrated that probiotic yeasts effectively reduce adhesion of *C. albicans* and non-albicans *Candida* species to these cultured epithelial cell lines [52,53]. In addition, yeast *S. boulardii* has been shown to pose a barrier and preserve the integrity of the epithelium by the reduction of pro-inflammatory cytokines in the intestine [40,52].

Even though live probiotic cells are known to play a significant role in preventing virulence of *C. albicans*, the role of exact cellular components involved are less investigated. For example, administration of cell wall components of *S. cerevisiae* reduced the *Candida* associated inflammation and colonization in animal models [55]. Interestingly, one of the four *S. cerevisiae* strains used in this study (strain Sc-4) increased the mortality and inflammation in the host, suggesting strain-specific effects of the probiotic yeasts against *Candida* species [55]. Such strain specificity as also been reported in the interaction of *Lactobacillus* strains with *C. albicans* [63]. Furthermore, heat-killed *S. cerevisiae* reduced the vaginal colonization of *C. albicans* when applied against vaginal candidiasis in a murine model [54]. These effects could either mediated by yeast cell wall components such as β-glucan or simply that the biomass of heat-inactivated probiotic cells form a physical barrier that occludes host factors that facilitate *C. albicans* attachment (Figure 1A) [54].

**Figure 1.** Probiotic yeast either form a physical barrier on epithelial surfaces (**A**) or secretes bioactive metabolite (**B**) to inhibit the adhesion and morphological transition of *Candida* species on epithelial cells. Further, suitable probiotic yeasts cell number is required for the effective inhibition of *Candida* virulence in the host GI tract (**C**).

#### *6.2. Role of Small Bioactive Metabolites in Probiotic Action*

Beneficial microbes or probiotics in the intestine are thought to control pathogen overgrowth by competing for limited nutrients. There is a growing body of literature that supports the notion that inhibitory function is primarily mediated by secreted small molecules with suitable probiotic cell number (Figure 1B,C) [53,64]. Microorganisms produce metabolites that have been shown to alter the course of an infection by synergistic or antagonistic interactions with infectious agents. Such metabolites include hydrogen peroxide, bacteriocins, and organic acids that effectively inhibit the virulence and growth of various *Candida* species [64,65] (Table 2). On the other hand, few interesting microbial metabolites, such as tyrosol and indole-3 acetic acid, trigger the filamentation in *C. albicans* [66,67]. Small molecules derived from bacteria have been evaluated for activity against *Candida* virulence and pathogenesis. For example, lectins of lactobacilli and bifidobacterial strains isolated from humans have been shown to inhibit the growth of drug-resistant *C. albicans* [68]. The Gram-positive pathogenic bacteria, *Enterococcus faecalis,* produces a peptide called EntV which has been shown to reduce *C. albicans* virulence [69]. Furthermore, organic acids such as acetic acid and lactic acid have been shown to enhance antifungal treatment of *C. albicans* and *C. glabrata* [70]. Many *Lactobacillus*, *Bifidobacterium,* and yeasts strains produce these organic acids. *S. boulardii* produces several bioactive compounds such as *Saccharomyces* anti-inflammatory factor (SAIF), anti-toxin factors, short-chain fatty acids, bioactive proteins of 54 kDa, and 120 kDa which play a major role in preventing bacterial infections [38,71]. However, there has been very limited knowledge on probiotic yeast metabolites on *Candida* species. Recently a group

showed that yeast *S. boulardii* metabolite capric acid (Decanoic acid)—a saturated fatty acid, inhibits the filamentation of *C. albicans* interaction [52].

In natural habitats, potential interaction of microbial communities has been a key element for the ecological dynamics. Bacteria and eukaryotic microorganisms exhibit both symbiotic and/or antagonistic interaction in the natural environment. In fact, *C. albicans* co-exists with other non-albicans *Candida* species or bacteria in the biofilm as well as the human GI tract. These inter-species interactions between C. albicans and other microbes typically affect filamentation of *C. albicans*. For instance, certain secretory molecules of *Salmonella typhimurium* and *Streptococcus mutants* inhibit cell growth and filamentation of *C. albicans* in the co-culture conditions [72,73]. Another well studied bacterium is *Pseudomonas aeruginosa*, where bacterial toxin phenazine inhibits the filamentation of *C. albicans* [74,75].

The morphological transition of yeast has been controlled by cell density and/or quorum sensing molecules. Apart from bacteria, the quorum sensing mechanism is also well studied in yeast such as *C. albicans* and *S. cerevisiae*. Farnesol and tyrosol are known cell density molecules in *C. albicans* which controls the morphological transition. Similarly, yeasts such as *S. cerevisiae* and other many non-*Saccharomyces* yeast produce alcoholic signaling molecules called phenylethanol and tryptophol. An abundant usage and availability of well-curated genetic database indicate that *S. cerevisiae* has gained more attention on quorum sensing mechanisms than the non-*Saccharomyces* yeast strains. There are few studies claiming that factors such as low nitrogen content and cell density play a significant role in the production of phenylethanol and tryptophol in *S. cerevisiae* and regulates its morphological transition mechanism [76]. Furthermore, these signal molecules are controlled by the expression of *ARO8*, *ARO9,* and *ARO10*, where *ARO8* and *ARO9* encode the aromatic aminotransferases and *ARO10* encodes the aromatic decarboxylase reaction [77,78].


**Table 2.** Microbial metabolites and its functions against *Candida* species.

Several research groups have predicted and/or observed an antagonistic nature of aromatic alcohols, phenylethanol, and tryptophol against fungi. Winters et al., (2019) reported that high concentrations of *S. cerevisiae*inhibited non-*Saccharomyces* strains in mixed cultures and under fermentation conditions [78]. Although there were direct evidence of inhibition due to these secondary metabolites, commercially procured phenylethanol and tryptophol have been shown to inhibit filamentation of *C. albicans* [77]. This result is bolstered by the observation that administration of tryptophol enhances survival of *Galleria mellonella* larval that are infected with *Candida* [80]. Furthermore, a cocktail of phenylethanol, isoamyl alcohol, E-nerolidol, and farnesol provides protection against *Candida* infection in a murine model of infection [81]. Together these studies establish a paradigm for inhibition of fungal virulence that is mediated by aromatic alcohols.

#### **7. Gaps in our Understanding of Biotherapeutic Application of Probiotics for** *Candida* **Infection**

Probiotic yeasts yield several positive outcomes in in vitro, ex vivo, and in vivo readouts during colonization of *Candida* species. Information about their effect during systemic infection is an area that needs further investigation. Numerous animal and handful of clinical experiments have revealed that probiotics and metabolites such as short-chain fatty acids, tryptophol and phenylethanol play an abundant role in human health and diseases. However, the origin of these metabolites is ill-defined and their effects on clinical manifestations of *Candida* infection need further investigation. These studies would provide substantive information to improve biotherapeutic properties of beneficial microbes against *Candida* infections.

Emergence of drug resistance and complications associated with side effects have sparked interest in alternative therapies. Applications of food-derived yeasts have been shown to have positive outcomes against *C. albicans* and non-albicans *Candida* species virulence and infection in pre-clinical and clinical settings. Food-derived beneficial yeasts are also generally safe and pose an effective alternative to traditional antifungals. They may also be used in combination therapy with conventional antifungal drugs since the synergistic effect of probiotics and antifungal agents would prevent emergence of drug resistance.

**Funding:** This research was funded by NIH-NCCIH, grant number NIH-NCCIH 1R15AT009926-01, DST-INSPIRE program, Department of Science and Technology, Government of India, award number DST/INSPIRE Fellowship/2013/553 and Fulbright-Nehru doctoral fellowship, United States–India Education Foundation (USIEF), India, award number 2310/DR/2018-2019.

**Acknowledgments:** We thank the Director, CSIR-Central Food Technological Research Institute (CFTRI) for encouragement and research support. This work is partially supported by NIH-NCCIH 1R15AT009926-01 grant to RPR. LK is grateful to the INSPIRE program, Department of Science and Technology, Government of India and Fulbright-Nehru doctoral fellowship, United States–India Education Foundation (USIEF), India for the financial support for his doctoral research.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Review*
