**1. Introduction**

*Listeria monocytogenes* is a human pathogen that may cause listeriosis, a foodborne infection with a low morbidity and a high mortality rate (20–30%) [1]. The presence of *L. monocytogenes* in raw meat does not cause major public health problems since meat is generally consumed after cooking at temperatures above 70 ◦C. However, contaminated raw meat when used as raw material for food products that in their production process fail to eliminate the pathogen may present a safety risk [1]. In addition, the presence of *L. monocytogenes* in raw meat constitutes restrictions on international trade.

Contamination of meat with *Listeria monocytogenes* is a consequence of the production process [2]. In addition, *L. monocytogenes* can survive and grow in vacuum-packed meat cuts stored at temperatures between 0 and 4 ◦C; therefore, different strategies are applied in abattoirs to minimize bacterial contamination [3,4]. Among the different strategies, lactic acid (LA) application is accepted because it does not present risks to consumer health. The maximum concentration of LA allowed is 5% (*m*/*v*) [5]. UV-C light irradiation (UV-C) stands out for its low cost, non-generation of potentially hazardous chemical residues, and low carbon footprint [6]. In addition, UV-C irradiation is an FDA-approved intervention for surface decontamination of foods [7] (FDA, 2019a).

The application of between 2 and 4% LA on meat was reported to reduce *L. monocytogenes* counts on beef surface [8,9]. Reductions between 0.79 to 1.31 log CFU/cm<sup>2</sup> were obtained in fresh beef when LA was applied from 1% to 4% [9]. Different levels of

**Citation:** Brugnini, G.; Rodríguez, S.; Rodríguez, J.; Rufo, C. Effect of UV-C Irradiation and Lactic Acid Application on the Inactivation of *Listeria monocytogenes* and Lactic Acid Bacteria in Vacuum-Packaged Beef. *Foods* **2021**, *10*, 1217. https://doi.org/ 10.3390/foods10061217

Academic Editors: Huerta-Leidenz Nelson and Markus F. Miller

Received: 1 May 2021 Accepted: 19 May 2021 Published: 28 May 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 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 (https:// creativecommons.org/licenses/by/ 4.0/).

*L. monocytogenes* reductions have been observed and were associated with factors such as variabilities among strain sensitivity towards stress and forms of LA application [10,11].

UV-C radiation (200 to 280 nm) has been used for decontamination of food surfaces [12]. The ability of UV-C to inactivate *L. monocytogenes* has also been reported, being strain dependent and showing a direct correlation between UV-C dose and *Listeria monocytogenes* reduction [13–15]. In addition, UV-C radiation can penetrate the packaging material usually used on meat and meat products such as transparent polypropylene and polyethylene bags [16] and cause significant *L. monocytogenes* reduction on food [13].

Antimicrobial interventions may affect fresh meat color, which is considered to be the single most important characteristic influencing consumer's purchase decisions [17,18]. Negative effects on meat color are the major problem associated with the use of lactic acid, especially at high concentration [19]. In contrast, UV-C (118–590 mJ/cm2) on fresh meat does not appear to cause detrimental color changes [15].

Other bacteria present on the beef surface may be affected by LA and UV-C [15,20]. In refrigerated vacuum-packed meat, lactic acid bacteria (LAB) are the ones that develop the most, being responsible for the production of strong lactic acid off-odors when counts reach 10.000.000 UFC/g at the end of shelf life [21–23]. Thus, knowing the effect of LA and UV-C application on LAB may be relevant to improve vacuum packed beef shelf life.

In the past few years, special attention has been given to experimental design and response surface methodology (RSM) to optimize conditions in different systems [24,25]. These modeling tools enable the study of the simultaneous effects of different factors and their interactions on experimental characteristics. This strategy has not been widely used to study the effects of the LA and UV-C combination on vacuum-packed beef. To date, only one report was found describing a similar strategy to study the effects of LA and UV-C on *Salmonella typhimurium* reduction on sliced Brazilian dry-cured loin [26].

In the present work, it was hypothesized that the combined application of LA at low concentrations and UV-C after vacuum packaging might achieve a significant level of reduction in *L. monocytogenes* contamination on beef with a minimal impact on meat color and would contribute to its shelf life by reducing meat LAB counts. To test this hypothesis, a two-factor central composite design and response surface methodology (RSM) were used to optimize the concentration of LA and the UV-C dose applied to vacuum-packed meat that will reduce the amount of *L. monocytogenes* and LAB without significant effects on meat color.

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

### *2.1. Meat Samples*

Eye of round *(Semitendinosus Muscle*) cuts were obtained from a local abattoir. Meat samples were not decontaminated prior to the study. Meat was cut by hand into square pieces of 10 g measuring 5 × 5 cm2. Each piece was individually inoculated with *L. monocytogenes* and treated according to the experimental design.

### *2.2. L. monocytogenes Culture Preparation*

A strain of *L. monocytogenes* (LM 100A1) previously isolated and characterized in our laboratory was used for this study [27]. The culture was prepared by growing LM 100A1 overnight at 35 ◦C, to the stationary phase, in tryptic soy broth (Oxoid Ltd., Hampshire, UK) supplemented with 0.6% yeas<sup>t</sup> extract (Oxoid Ltd., Hampshire, UK). The overnight culture was diluted with butterflied buffer to 6.1 log CFU/mL.

### *2.3. Preparation of Lactic Acid Solution*

The lactic acid solutions were prepared by diluting a concentrated lactic acid solution (85% *m*/*v*) (PURAC®, Corbion, Montevideo, Uruguay) with sterilized distilled water to make 2.5%, 5.0% and 6.0% ( *m*/*v*) lactic acid solution. Fresh solutions were prepared prior to each test.

### *2.4. UV-C Irradiation*

The specifications of the UV-C lamp used were: 30 W T6 tubular 254 nm with UV germicidal lamp (Code ZW30S19W-Z894, Cnlight Co., Ltd., Guangdong, China), diameter 19 mm, length 894 mm and UV intensity at one meter of 107 μW/cm2. Intensity at the application distance was 3.137 mW/cm<sup>2</sup> measured with a ZED Smart Meter s/N 800,009 (EN61326-1-2013) and the reference sensor D-SICONORM-LP-REF-500 W/m2.

Before each trial, the UV-C lamp was preheated for 20 min to stabilize the UV-C emission. UV-C treatments did not increase the surface temperature of the meat to greater than 20 ◦C.

### *2.5. Experimental Design*

A two-factor central composite design with five central points, 2 replicates of factorial points and 2 replicates of axial points were used. The experimental design matrix and all data analysis were performed using Design-Expert® (Version 10, Stat-Ease Inc., Minneapolis, MN, USA). The independent variables were lactic acid concentration (X1) and UV-C dose (X2), and the dependent variables or response variables were *L. monocytogenes* (LM) log reduction (Y1), Lactic Acid Bacteria (LAB) log reduction (Y2), and Chroma value (Y3). The design matrix consisted of 21 experimental runs including 8 factorial points and 8 axial points with five replicates at the center point (Table 1). Observed responses were fitted to first order, second order and quadratic models. Models were selected by the Sequential Sum of Square Method and assessed based on statistically significant coefficients and *R*<sup>2</sup> values using ANOVA technique, with a significance level of α = 0.05. For each response variable (Y), a second-order polynomial model equation was defined:

$$\mathbf{Y} = \beta\_0 + \beta\_1 \mathbf{X}\_1 + \beta\_2 \mathbf{X}\_2 + \beta\_{11} \mathbf{X}\_1^{\top} + \beta\_{22} \mathbf{X}\_2^{\top} + \beta\_{12} \mathbf{X}\_1 \mathbf{X}\_2 \tag{1}$$

where Y is the measured response of the dependent variables, X1 and X2 are the independent variables, β0 is the intercept, β1 and β2 are the linear coefficients, β11 and β22 are the squared coefficients, and β12 is the interaction coefficient.


**Table 1.** Central composite experimental design matrix and observed responses.

a Mean of three values per sample.

Response surface methodology (RSM) included the generation of 3D response surface and contour plots to study the overall relationships and interactions between independent variables and response factors.

### *2.6. Sample Treatments*

According to the experimental design, 21 pieces of 10 g of meat were inoculated with 5.8 log of CFU of the strain LM100A1. Then, 500 μL of the inoculum were disposed on the meat surface and spread with a bent glass rod.

After 10 min, inoculated meat pieces were treated with 1.5 mL of lactic acid solutions from 0 to 6 ( *m*/*v*) %. The LA was disposed on the inoculated side, drop by drop covering the entire surface of the meat samples. Then the samples were vacuum packaged in a multi-laminar (EVA, PVDC, PE) thermo-shrinkable bag with a 76% UV-C transmission rate (Cryovac® BB 2620; 50 μm thick, oxygen permeability of 20 cm<sup>3</sup> m<sup>−</sup>2, 24 h, at 23 ◦C, and 75% RH; and maximum carbon dioxide permeability of 100 cm<sup>3</sup> m<sup>−</sup>2, 24 h, at 23 ◦C, and 75% RH) by use of a vacuum-packaging machine SAMMIC model V-410SGI (Spain).

After packaging, each side of the samples were exposed to 3.137 mW/cm2, at 7 cm from the emitting lamp, for 53, 105 and 127 s, achieving doses of UV-C of 165, 330 and 398 mJ/cm<sup>2</sup> respectively. The UVC-dose range was selected considering the reported UV-C doses applied on beef that did not affect meat color [15], the application conditions, distance from the lamp and duration of the exposures, to be easily implementable in industrial production lines. A second set of samples was prepared for color measurements.

### *2.7. Microbiological Analyses*

After treatments, samples were homogenized in sterile bags with 90 mL of Butterfield buffer, appropriate dilutions were plated by duplicate on PALCAM Listeria Selective Agar (Oxoid Ltd., Hampshire, UK) incubated at 37 ◦C for 48 h for *L. monocytogenes* and on MRS Agar (Oxoid Ltd., Hampshire, UK) incubated anaerobically at 30 ◦C for 72 h for LAB. Colonies were counted and log transformed. Log reductions of *L. monocytogenes* and LAB per gram of meat compared to samples with no treatments were calculated.

### *2.8. Instrumental Color*

At twenty-four hours post treatments, color measurements were performed 30 min after opening the packages. Instrumental lean color (CIE L\*: brightness, a\*: redness and b\*: yellowness) was measured with a Minolta chromameter CR-400 (Konica Minolta Sensing Inc., Tokyo, Japan) using a C illuminant, a 2◦ standard observer angle and 8 mm aperture size, and calibrated with a white tile before use. Three measurements from each sample were taken and the mean value was calculated. Chroma value was calculated as C\*= √ (a\*<sup>2</sup> + b\*2).

### *2.9. Optimization and Model Validation*

The optimized conditions, lactic acid concentration and UV-C dose, were obtained by applying the following constraints on the response factors: (i) to maximize *L. monocytogenes* log reduction; (ii) to maximize LAB log reduction; and (iii) Chroma value > 20, according to MacDougall, et al. 1982 [28] (values above 20 indicate bright red beef).

To validate the proposed model, three experiments were carried out using the optimized conditions as the checkpoint. Experimental responses (log reduction of LM100A1 per gram and log reduction of LAB per gram) of the checkpoint were compared to the predicted results from the fitted models to evaluate the precision of the polynomial equations.

#### *2.10. Evolution of L. monocytogenes, LAB, pH and Instrumental Color of LA/UV-C Treated Meat Vacuum Packed and Stored at 4* ◦*C for 8 Weeks*

Three experiments were carried out using the optimized conditions (LA 2.6% ( *w*/*v*) and UV-C 330 mJ/cm2). Meat samples were treated as explained in 2.6., untreated samples were used as a control group. Samples were stored at 4 ◦C and analyzed for LM, LAB at initial time (week 0) and every two weeks until week 8. Instrumental color was measured at week 0 and week 8. An additional set of samples was prepared for pH determination using a Hanna® model 9025c pH meter with a surface electrode (HI1413B). The T-test

was used to compare the data for control and LA/UV-C samples. Data were expressed as mean ± standard deviation. Significance was determined at the *p* < 0.05 level.
