The extraction of phenolic compounds was done according to Santacruz et al. (2020). An amount of 15 g of the previously ground cocoa pericarp was mixed with 150 mL of ethanol (95% v/v) and kept under stirring at 130 rpm for 24 h at 25 °C. Afterwards, the sample was centrifuged (SIGMA 2-16P, Germany) for 10 min at 4000 rpm and the supernatant was filtered by vacuum, saving the filtered liquid fraction.

The quantification of phenolic compounds was done according to the Folin-Ciocalteu method proposed by Mahmood et al. (2011). A stock solution was prepared by mixing 10 mL of the previous filtered fraction with 5 mL of ethanol (95% v/v) and distilled water to complete a total volume of 100 mL. From the stock solution, 0.1 mL was mixed with 0.5 mL of the Folin-Ciocalteau reagent, allowing it to stand for 5 min. Afterwards, 1 mL of a sodium carbonate solution (5%) was added and made up to 25 mL with distilled water. The resulting solution was left in dark for 1 h and its absorbance at 760 nm was measured (Spectrophotometer Jenway 6320D, China). The quantification of phenolic compounds was performed using a calibration curve with gallic acid standards, and the results were expressed in mg of gallic acid equivalent (GAE) g^-1 cocoa pericarp.

The antibacterial activity of phenolic compounds against Salmonella spp. was performed according to Santacruz and Castro (2018) and the Clinical and Laboratory Standards Institute (CLSI, 2009). Salmonella spp. was inoculated into Petri dishes using Salmonella-Shigella agar (HiMedia Laboratories, India) as culture medium and incubated at 37 °C for 24 h. Filter paper disks (Fisher Scientific Q2) of 5 mm diameter were immersed in a solution of phenolic compounds at a specific concentration (see experimental design). The disks were dried and placed in the center of the Petri dishes containing Salmonella spp. and incubated at 37 °C for 24 h. Afterwards, the zone of inhibition of the growth by phenolic compounds was measured in triplicate.

Bacterial growth was performed according to Puupponen et al. (2001). 10 mL of fresh growth medium (Salmonella- Shigella agar) were inoculated with 5% of culture (frozen Salmonella spp. was transferred to liquid media and incubated for 24 h). Phenolic compounds, from alcoholic extraction, were added to the culture media to give final concentrations of 0.5, 1 or 5%. The culture was manually shaken, and afterwards, it was incubated at 37 °C for 24 h. Bacterial growth was followed by taking seven samples from the culture during the incubation period of 24 h (0, 0.5, 1, 2, 4, 6, and 24 h). The samples were plated in Petri dishes containing Salmonella-Shigella agar and incubated at 37 °C for 24 h before bacterial count was recorded.

Fresh cheese was immersed in a 1% phenolic compounds solution for 15 min. Afterwards, cheese samples were stored for 8 days, overnight under refrigeration conditions and during the daytime at room temperature (approximately 25 °C).

Microbiological analysis of Salmonella in cheese was performed on samples after 0, 2, 4, 6 and 8 days of storage, according to ISO 6579: 2002 method (ISO, 2002). For the test, 10 g of cheese were weighed and placed in a sterile resealable plastic bag, before adding 90 mL of sterile peptone water. The sample was homogenized manually, and 1 mL of the liquid was transferred to a glass test tube before adding 9 mL of peptone water (10^-1 dilution). The process was repeated to get a dilution of 10^-4, and immediately 1 mL of the solution was inoculated into Petri dishes and incubated at 37 °C for 24 h prior CFU counting.

Cheese samples were previously analyzed to verify the absence of Salmonella. Afterwards, the cheese was immersed for 15 min in a 1% phenolic compounds solution before sensory analyzes. First, a duo-trio test ISO 10399: 2004 (ISO, 2004) with 16 semi-trained panelists was done to determine organoleptic differences between cheese samples treated with phenolic compounds and a control sample (cheese with no immersion). Subsequently, attributes of firmness, color, odor and flavor were evaluated using a five-point hedonic scale, with scores ranging from "I like it a lot" (5) to "I dislike it a lot" (1).

Puncture tests were performed according to Castro et al. (2017), using cheese cubes of edge 2 cm, with slight modifications. A Shimadzu texturometer (Model EZ-LX, Japan) together with a stainless-steel probe of 3 mm diameter and 8 cm length were utilized. The probe speed was 10 mm s^-1 and the maximum penetration force was recorded. Results were the average of three measurements.

A completely randomized design with a unifactorial arrangement was used. The concentration of phenolic compounds obtained by alcoholic extraction at 3 levels (0.5, 1 and 1.5%) was established as the independent variable (Medrano, 2019), whereas the zone of inhibition and the bacterial growth measurement were the dependent variables. The concentration of phenolic compounds with the best Salmonella inhibition was applied to cheese samples, before sensory analysis.

ANOVA and the significance of the difference between means was determined by Tukey test (P<0.05) using InfoStat statistics software (Infostat version 2014, Argentina). All measurements were performed in triplicate.

Results of the duo-trio test showed that 15 of 16 panelists did not find difference between cheese previously immersed in a solution of phenolic compounds and the control sample. Texture analysis displayed that penetration force of control sample was 1.02 N, which was lower than cheese treated with phenolic compounds, showing values between 1.92 and 1.99 N (P<0.05). Instrumental texture of cheese does not have a single dominant characteristic and exhibit at least two primary characteristics (e.g. chewiness and cohesiveness) (Meullenet et al., 1998).

Figure 2 shows the responses of the panelists to the five-point hedonic test about cheese treated with phenolic compounds. Firmness and odor obtained the highest ratings among the evaluated characteristics, since 56.3% of the panelists classified them as "I like it". For the flavor, 37.5% of the panelists gave a rating of "I like it" and an equal percentage for the color of the cheese with the label of "I neither like nor dislike". The lower color rating could be due to the fact that phenolic compounds developed a yellow color in cheese, color that did not change along storage.

Figure 2

Based on present results, it is difficult to conclude if color had influence either in odor or flavor. Moreover, visual and auditory appreciations can modify the flavor of food but are not intrinsic to it (Auvray and Spence, 2008). Koch and Koch (2008) stated that "In fact, it may be that color has nothing to do with the taste of food or drink". Meanwhile, Bayarri et al. (2001) suggested that "the possible influence of color on flavor perception is under discussion and no clear conclusions have been attained yet".