**1. Introduction**

Alcoholic fermentation is the process of monosaccharide's conversion to ethanol and CO2. Therefore, the anaerobic metabolism of *Saccharomyces cerevisiae* is the main cause of wine fermentation. However, in spontaneous wine production other yeasts participate such as *Hanseniaspora, Candida, Pichia*, and *Metschnikowia* genera, among others [1]. Nevertheless, these yeasts have lower fermentation capacity, and are not able to grow in high ethanol concentration conditions, given that the *Saccharomyces* genus is the predominant during the final stages of fermentation [1].

Among the unfavorable growth wine conditions, several yeasts are capable to proliferate and generate undesired characteristics in the final product. *Brettanomyces bruxellensis* has been described as the main spoilage yeas<sup>t</sup> during the maturity stage of wine in barrels [2,3]. This yeas<sup>t</sup> has the capacity of transforming hydroxycinnamic acids into vinyl and ethyl derivates, which produce off-flavors in wine [4,5]. Additionally, this aromatic defect can be produced in early stages of fermentation by other yeasts such as *Pichia guilliermondii* [2]. Among these, there are strains which can transform *p*-coumaric acid in 4-vinylphenol in similar proportions as described for *B. bruxellensis*, being that *P. guilliermondii* is a potential problem for winemaking [6]. Because of this, in this industry the use of sulfites is a widespread strategy to control growth of undesired microorganisms. Nevertheless, several

strains are resistant and the use of sulfites in high quantities is potentially unsafe for human health [7]. As a result, alternative strategies such as antimicrobial peptides (AMPs) have been proposed to biocontrol spoilage microorganisms [8–12]. AMPs are low molecular mass peptides with amphipathic characteristic which can affect the growth of several microorganisms by permeabilization of plasmatic membranes and/or by increasing the reactive oxygen species [13–15]. Previously, Peña et al. [16,17] have described antimicrobial peptides production in *Candida intermedia*, which reduce the viability of different *B. bruxellensis* strains in a laboratory medium without affecting the growth of fermentative yeas<sup>t</sup> *S. cerevisiae*. However, it has not ye<sup>t</sup> been described how these peptides affect *B. bruxellensis* and if they are able to inhibit the growth of *P. guilliemondii*. Thus, the aim of this work was to explore the cellular damage produced by *C. intermedia* LAMAP1790 peptides above *B. bruxellensis* and determine if they can control the growth of yeas<sup>t</sup> *B. bruxellensis* and *P. guilliermondii* using laboratory culture mediums and synthetic must. This knowledge will allow to determine the antimicrobial peptides produced for *C. intermedia* as a possible biocontroller in the wine industry.

#### **2. Materials and Methods**

#### *2.1. Strains and Culture Media*

The strains of *B. bruxellensis* LAMAP2480, *C. intermedia* LAMAP1790, *Pichia guilliermondii* LAMAP3202, LAMAP3203, *S. cerevisiae* BY4741, and EC1118 were obtained from the culture collection at the Laboratory of Biotechnology and Applied Microbiology, University of Santiago de Chile. *C. intermedia* LAMAP1790 was isolated in Chile from must in the early stages of fermentative process [18] and *B. bruxellensis* LAMAP2480 was isolated from Chilean wine [19]. Both strains of *P. guilliermondii* were isolated from Argentinian vineyards. The strain LAMAP3202 and LAMAP3203 was characterized by Sangorrín et al. (2013), labeled as P7 and P8 strains respectively [20]. All strains used in this work were grown on GYEB media (yeast extract 5 g/<sup>L</sup> and glucose 20 g/L, adjusted to pH 5.0 with 100 mmol/L phosphate-citrate buffer) [21].

#### *2.2. Obtained Supernatant with Antifungal Activity of C. intermedia and Characterization of the Protein Nature of This Activity*

To obtain the supernatant with antifungal activity from *C. intermedia* LAMAP1790, the yeas<sup>t</sup> was inoculated in 100 mL GYEB medium during 48 h at 28 ◦C with orbital agitation at 120 rpm. Then, the culture was centrifuged during 10 min at 5900× *g* to obtain saturated culture supernatant. Afterward, a cut-off of total proteins present in the supernatant was done by means of ultrafiltration in devises *Amicon*® *Ultra-15* with 10 kDa cutoff (Merck-Millipore®, Darmstadt, Germany). In this work, the antifungal supernatant is defined as the fraction obtained from ultrafiltration which only contains proteins of molecular mass under 10 kDa. This antifungal supernatant was sterilized using disposable filters with 0.22 μm pore size (Membrane Solutions LLC®, Windham, NH, USA) and stored at −20 ◦C to be used later. To determine whether the antifungal activity is related with the presence of peptides with molecular mass under 10 kDa, the antifungal supernatant was treated with 2 mg/mL protease of *Streptomyces griseus* (Sigma-Aldrich®, St. Louis, MO, USA) during 4 h at 37 ◦C.

#### *2.3. Determination of the Cellular Damage Produced on B. bruxellensis by Exposure to Antifungal Supernatant of C. intermedia*

The obtained antifungal supernatant of *C. intermedia* LAMAP1790 was assessed to determine if it produces: (a) membrane permeability or (b) rise of the reactive oxygen species (ROS) on exposed *B. bruxellensis* cells during different periods, similar to described by [22] and [23]. Then, 3 × 10<sup>5</sup> *B. bruxellensis* LAMAP2480 cells were exposed individually to 1 mL of sterile antifungal supernatant and incubated during 12 h and 24 h at 28 ◦C. As positive control, a similar number of cells with 600 μg/mL zymolyase 100*T* (Amsbio®, Abingdon, OX, UK) was inoculated at 37 ◦C during 2 h and then exposed to 30% H2O2during 30 min. As negative control, the same concentration of cells was

used and at 28 ◦C in buffer HEPES saline 1× pH 7.0 (70 mM NaCl, 0.75 mM Na2HPO4, 25 mM HEPES) were incubated. After treatments, the cells were washed 3 times with buffer HEPES saline 1× pH 7.0. To facilitate the observation, yeas<sup>t</sup> was stained with calcofluor white (Sigma®) in 1:1 proportion with KOH to 10% p/v. The membrane permeability was assessed by means of staining with 2 μM propidium iodide (Sigma®) and the accumulation of ROS was determined by means of staining with 10 μM 6-carboxy-2,7-dichlorodihydrofluorescein diacetate (C400; Thermo-Scientific®, Waltham, MA, USA). The fluorescent cells were observed using the epifluorescence microscope Moticam Pro BA410 (Motic®, Xiamen, China), with 40× fluorescence microscope objective lent.

#### *2.4. Screening the Antifungal Activity of C. intermedia LAMAP1790 on B. bruxellensis, P. guilliermondii, and S. cerevisiae*

The qualitative determination of the antifungal activity of *C. intermedia* LAMAP1790 on *B. bruxellensis* LAMAP2480, *P. guilliermondii* LAMAP3203, LAMAP3203 and *S. cerevisiae* EC1118 strains was carried out following the methodology used by [16]. For this, 1 × 10<sup>5</sup> cells from each strain were inoculated in 25 mL warm agar MBA (5 g/<sup>L</sup> yeas<sup>t</sup> extract, 5 g/<sup>L</sup> peptone, 20 g/<sup>L</sup> glucose and 15 g/<sup>L</sup> agar, adjusted to pH 5.0 with 100 mmol/L phosphate-citrate buffer and supplemented with 0.03 g/<sup>L</sup> of methylene blue). Each inoculated media was plated into petri dishes. A surface inoculation was carried out using 10 μL of 1 × 10<sup>8</sup> cells/mL suspension from *C. intermedia* LAMAP1790. As control, the same surface inoculation of *S. cerevisiae* BY4741 was carried out. The plates were incubated for 7 days at 28 ◦C and every assay was evaluated six times. The qualitative determination was done by observation and measuring the inhibition halo present in the plates.

#### *2.5. Antifungal Activity of Low Mass Peptide Fraction Obtained from C. intermedia Antifungal Supernatant against B. bruxellensis, S. cerevisiae and P. guilliermondii in Synthetic Must*

The obtention of 100X concentrated low mass peptide fraction (under 10 kDa) was performed by lyophilization (IlShineBioBase® freeze dryer, Dongducheon-si, Gyeonggi-do, Korea) of 3 L to sterile antifungal supernatant derived from cultures of *C. intermedia* LAMAP1790 in GYEB medium. The total protein quantification in the fraction was done according to [24]. The evaluation of the antifungal activity was done using simultaneous inoculation of *S. cerevisiae* EC1118 and *B. bruxellensis* LAMAP2480 or *P. guilliermondii* LAMAP3202 in synthetic grape must, (100 g/<sup>L</sup> glucose, 100 g/<sup>L</sup> fructose, 5 g/<sup>L</sup> maleic acid, 0.5 g/<sup>L</sup> citric acid, 3 g/<sup>L</sup> tartaric acid, 0.75 g/<sup>L</sup> potassium phosphate, 0.5 g/<sup>L</sup> potassium sulfate, 0.155 g/<sup>L</sup> calcium chloride, 0.25 g/<sup>L</sup> magnesium sulfate, 0.2 g/<sup>L</sup> sodium chloride, 4 mg/<sup>L</sup> manganese sulfate, 1.5 mg/<sup>L</sup> calcium pantenoate, 2 mg/<sup>L</sup> nicotinic acid, 0.25 mg/<sup>L</sup> thiamine hydrochloride and 0.003 mg/<sup>L</sup> biotin; pH 3.5) [25], Previously, each strain was adapted to the media using a procedure described by [26]. To the antifungal assays, 5 mL synthetic must was inoculated with 1 × 10<sup>2</sup> cells of *S. cerevisiae* EC1118, *B. bruxellensis* LAMAP2480 or *P. guilliermondii* LAMAP3202 strains individually (determined by direct yeas<sup>t</sup> count in Neubauer chamber), and supplemented with 1 μg of low mass peptide fraction. As a control, the same procedure was carried out, but the medium was supplemented with 1 μg of total proteins obtained from the concentrate sterile culture supernatant of *S. cerevisiae* BY4741 (*IlShineBioBase*® *freeze dryer*, Dongducheon-si, Gyeonggi-do, Korea). Each assay was incubated for 21 days, and every 3 days a cellular count of the cultures was carried out on YPD agar plates (5 g/<sup>L</sup> yeas<sup>t</sup> extract, 5 g/<sup>L</sup> peptone, 20 g/<sup>L</sup> glucose and 20 g/<sup>L</sup> agar) incubated for 7 days at 28 ◦C. The count of spoilage yeas<sup>t</sup> in the mixed culture was performed in YPD agar plates supplemented with 0.01% *v*/*v* of cycloheximide, according to [27].
