A Novel Way for Whey: Cheese Whey Fermentation Produces an E ﬀ ective and Environmentally-Safe Alternative to Chlorine

Featured Application: Industrial waste whey, as a low-cost, e ﬃ cient, and environmentally safe disinfectant, with potential applications for minimally processed foodstu ﬀ . Abstract: Cheese whey has been described as an environmental hazard due to its high organic content. Although it has been suggested that whey can be used as food disinfectant, it continues to pose an environmental problem because it still contains a high organic load. Here, we aimed to develop a low-cost, scalable fermentation protocol to produce a disinfectant from dairy waste that has very little organic content and high levels of lactic acid. Fermentation was achieved with industrial whey from ewe, goat, and cow’s milk, using a speciﬁc mesophilic-lactic acid bacteria starter mix over 120 h, which yielded the highest lactic acid production and the lowest lactose content. Antibacterial activity was observed against Listeria monocytogenes , Salmonella enterica, and Escherichia coli O157:H7, plus a total of thirteen other food pathogenic and spoilage strains, and antibacterial activities were determined to be highest after 120 h. We further validated this whey’s application as a disinfectant in shredded lettuce and compared its e ﬃ cacy to that of chlorine, evaluating microbial quality, texture, color, and sensory perception, pH, and O 2 and CO 2 determinations. Results showed that not only was microbial quality better when using our whey solution ( p < 0.05), but also the quality indicators for whey were statistically similar to those treated with chlorine. Hence, our work validates the use of an industrial waste whey as a low-cost, e ﬃ cient, and environmentally safe disinfectant, with potential applications for minimally processed foodstu ﬀ s as an alternative to chlorine.


Introduction
Whey is an industrial by-product resulting from cheese manufacture. Approximately 89% of the milk used for cheese manufacture is processed into whey [1]. Due to its high organic matter content, it holds a high chemical oxygen demand (COD), hence posing a considerable pollution problem [2][3][4][5], meaning that currently there are very restrictive legislations regarding whey disposal [6,7]. Hence, a novel trend for finding emerging alternative uses has risen, with one of them being the reuse of whey as a disinfecting agent [8][9][10]. The increasingly recognized antibacterial potential of whey is mostly due to the presence of lactic acid and antibacterial bioactive peptides [11][12][13], both resulting was performed on Man Rogosa and Sharp medium (MRS broth) (Biokar Diagnostic TM , Beauvais, France), containing 2% (w/v) agar (Difco, Quilaban, Portugal). Incubation was done at 37 • C ± 1 • C for 48 h ± 2 h. The samples of whey were obtained after manufacture of cheese from a mixture of ewe, goat, and cow´s milk, containing 0.04% (w/v) NaCl, pH 6.65 (Lab 850, Schott AG, Mainz, Germany) and 30 g·L −1 lactose, quantified by HPLC (High-performance liquid chromatography) ion exchange chromatography, as in a previous work [41], was used for the fermentation assays. Whey was divided into 500 mL aliquots in Erlenmeyer flasks, and assays were conducted at 37 • C for 120 h. Over time, at regular intervals, 5 mL samples were taken for pH measurement (Lab 850, Schott AG, Mainz, Germany) and HPLC determinations, as previously described by Santos et al. [41].

Biochemical Analysis
The quantification of sugars and acids throughout the fermentation assays were done by HPLC (Waters Corporation, Milford, MA, USA), equipped with a 515 HPLC Pump (Waters Corporation, Milford, MA, USA) and incorporated with a Refractive Index Detector (RID) 486 (Waters Corporation, Milford, MA, USA). Prior to injection, samples were centrifuged at 10.000 rpm/10 min and the supernatants were filtered through a Millipore membrane with a pore size of 0.2 µm. Samples were injected in a Schodex SC-1011 column (Waters Corporation, Milford, MA, USA) and separations were achieved at 50 • C, using 5 mM sulphuric acid as mobile phase (isocratic elution), at a flow rate of 0.6 mL.min −1 . Calibration curves and peak integration were performed as already described by Santos et al. [41].

Minimum Inhibitory Concentration (MIC) Throughout the 120 h Fermentation
In a first approach, antibacterial activities were tested throughout the length of the five-day fermentation, with L. monocytogenes and Escherichia coli O157.H7. Minimum Inhibitory Concentration (MIC) determinations were determined using suspensions of L. monocytogenes and Escherichia coli O157:H7 from overnight cultures, which were prepared in Müller-Hinton Broth medium ((Biokar Diagnostics™ Beauvais, France) adjusted to a final optical density (OD) of 0.12 (2 × 10 8 cfu·mL −1 ) [42], read at 546 nm in a spectrophotometer (Boeco S-20, Hamburg, Germany). Further decimal dilutions were made in the same medium to obtain final suspensions of 2 × 10 5 cfu·mL −1 . MIC values were assessed in sterile 96-well plates (Greiner Bio-one, Germany), using the micro dilution method of Bouhdid et al. [43]. Plates were incubated for 24 h ± 2 h, at 37 • C ± 1 • C, and the OD was read at 546 nm (Synergy HT, Biotek, Winooski, VT, USA) in the beginning of the inoculation and at the end of the assay.

Well-Diffusion Assay
After selecting the day that presented the highest antibacterial activity, fermented whey (100%) was tested on a second screen for its antibacterial activity against several food outburst-related bacterial species, using the well-diffusion assay described by Rizzello et al. [11], modified by incorporation of inoculum into Plate Count Agar (bioMerieux, Marcy l'Etoile, France), 10 mL per plate, overlaid with 10 mL of water-agar (2% w/v). Wells of 15 mm were cut into agar plates and 200 µL of 120 h fermented whey were placed in each well. Plates remained at 4 • C for 4 h to allow diffusion.
To test dose-dependence, we also compared non-fermented whey, fermented whey, and milk against the strain of L. monocytogenes, using diluted solutions of 100, 50, and 25% (v/v). After incubation at 37 • C ± 1 • C, 24 h ± 2 h, plates were measured with a caliper rule for inhibition zone diameters, with averages of three independent trials calculated with standard deviations. As a positive control we used an anionic detergent surfactant and susceptibility was recorded by average diameters > 4 mm [44].

Impact of Fermented Whey as a MP Lettuce Sanitizer
Loose leaf lettuce (Lactuca sativa var. crispa) samples were purchased at a local biological market, always from the same organic grower, on the same day of the experiments. Samples were treated in parallel with whey solution and a 110 ppm chlorine solution by dissolving Amokina ® (Angelini, Portugal) in sterile water, according to the manufacturer instructions. Outer leaves of the lettuces with signs of damage were discarded; the inner ones were cut using a cylindrical metal cutter of 6 cm diameter in order to take representative portions of all parts of the plant tissue. Circles of lettuce leaves were than washed in distilled water. Subsequently, three different solutions were used to sanitize and wash the lettuce shreds: (1) Distilled water acting as the reference; (2) chlorinated solution (110 ppm of Amokina ® ), as previously described; and (3) a 75% (v/v) fermented whey solution in distilled water [41]. The three different treatments were carried out in plastic bags filled with about 200 g of lettuce shreds, from the three different zones of the lettuces (outer, inner, and middle leaves) immersed in 1 L of the sanitizing or washing solutions. Bags were sealed and soaked for 5 min at 4 • C using an incubator with orbital shaking (Panasonic MIR 154, Gunma, Japan). After that, lettuce shreds were rinsed with sterile distilled water to remove sanitizers, and finally, the excess surface water was removed by a handheld salad spinner (IKEA Tokig, Lisbon, Portugal) for about 30 s. Processed lettuce were pooled and packed at 100 g of shredded lettuce per bag in heat-sealing bags (300 × 230 mm) of 30 µm oriented polypropylene (Amcor Flexibles Neocel, Portugal), graciously granted by Campotec SA. A bag prepared with 100 g of lettuce just soaked in distilled water, whose excess was also removed by the salad spinner, was marked as day 0 and served as the reference. To evaluate O 2 and CO 2 changes, three independent bags were separated. Samples were stored at 4 • C for subsequent evaluation of O 2 and CO 2 changes, pH, texture, color, sensory quality, and microbial growth, on processing day (day 0) and after 1, 3, 5, 7, 9, and 10 days of storage. The experiments were performed in three independent trials.

pH Measurement
The samples were prepared by mixing 10 g of each sample with 20 mL of deionized water in sterile stomacher bags (Seward Limited, London, UK) and then homogenized in a Stomacher (Model 400 Circulator, Seward Limited, London, UK) for 2 min at regular speed. Afterwards, the pH of samples was measured with a Metrohm 827 pH Lab with a 6.0228.010 Primatrode, electrode with integrated temperature sensor (NTC) (Metrohm, Herisau, Switzerland).

Texture Analysis
The texture features of lettuce samples were evaluated in a puncture test with a texturometer TA-XTplus (Stable MicroSystems, Godalming, UK) using a 5 kg load cell. Each sample was fixed between a perforated plate and an acrylic ring with a 9.7 mm hole, as in a sandwich, and kept in place by a plastic clip to avoid any slipping. A 2 mm diameter inox probe penetrated (5 mm distance) the sample through the hole in the rin, at 1 mm/s crosshead speed. From the force-distance texturogram, two parameters were evaluated: firmness, as the maximum rupture force (N) in the yy axis; and brittleness, defined as the rupture deformation (mm), corresponding to the distance of the peak on the xx axis.

Color Measurement
Sample color was measured on the CIELAB L*a*b* chromatic space with a Minolta CR-300 colorimeter (Minolta, Osaka, Japan) with standard illuminant D65 and a visual angle of 2 • . Tri-stimulus color coordinates (CIELAB system) were used to measure the degree of lightness (L*) which ranges from 0 (black) to 100 (white), redness (a) ranging from −60 (green) to +60 (red) and yellowness (b) ranging from blue (−60) to yellow (+60). The instrument was initially calibrated using a white ceramic tile (L*97.21; a* 0.13; b* 2.00). The measurement was performed by placing a piece of lettuce directly under the sensor and at least 30 measurements were done by treatment and day of evaluation, covering three different leaves coloration picked by naked eye selection (dark green, median green and light green).

Sensory Evaluation
Evaluation of sensory quality of the samples, after washing with the sanitizing agents, along the ten days of storage, was performed by a panel of ten untrained members. Panelists were required to evaluate changes in visual quality (appearance/freshness), texture, flavor, off-odors, color (browning) and overall acceptability of samples. Samples were scored by an hedonic scale from 1 to 9 with descriptors anchored at both ends (dislike, not characteristic of the product and like very much, very characteristic of the product) to describe attributes considered. The limit of acceptance from the consumer's point of view is set to be 5, and values below this point indicate unacceptable samples.

Microbial Analysis
Enumeration of Aerobic Microorganisms at 30 • C (AM), Psychrotrophic Microorganisms (PM) and Lactic Acid Bacteria (LAB) were performed in duplicate plates as previously described by Santos et al. [41]. Counts were performed for lettuce shreds before and after sanitizing treatments, to monitor microbial development during storage at 1, 3, 5, 7, 9 and 10 days in three independent trials.

Statistical Analysis
Statistical analysis was made by applying variance analysis, the one factor (ANOVA), and post-hoc multiple comparisons (Tukey test).

Longer Fermentation Periods Are Important for the Highest Lactic Acid Production and the Lowest Sugar Content
Although lactic acid production is of importance, the major aim of our research was to obtain a disinfecting agent with immediate application to spin-off as an easy, low-cost, technology. While most studies use short fermentation periods such as 24 h [33,39], in the present work we allowed the fermentation to last up to 120 h. The metabolite profile obtained throughout fermentation period is graphed in Figure 1, which shows the amounts of several metabolites: lactose, galactose, acetic and lactic acids, as well as the pH throughout the 120 h. Results show that although the pH dropped from 6.6 to 3.9 during the first 24 h of fermentation, lactic acid production continued to increase through the 120 h, supporting the need to use longer fermentation times, as opposed to the usual procedure in most studies. Lactic acid was the main acid detected, but acetic acid (0.89 g·L −1 ) was produced after 120 h. Starting from 30 g·L −1 lactose in the unfermented whey, the starter led to a mass conversion rate of lactose into lactic acid of 0.56, while residual lactose was 2.61 g·L −1 . The yield of converting lactose into lactic acid was therefore higher in our work when compared to other works, such as Plessas et al. [45] who reported a 0.47 conversion rate. Also, the fact that lactose was reduced to concentrations lower than 3 g·L −1 makes it suitable to be used in food products as a disinfectant without the problems of lactose intolerance [46].
Appl. Sci. 2019, 9, x FOR PEER REVIEW 6 of 16 lactose into lactic acid was therefore higher in our work when compared to other works, such as Plessas et al. [45] who reported a 0.47 conversion rate. Also, the fact that lactose was reduced to concentrations lower than 3 g·L −1 makes it suitable to be used in food products as a disinfectant without the problems of lactose intolerance [46]. Hence, overall results show that our fermentation protocol has the potential to reduce the environmental impact of whey by reducing its organic content.

Antibacterial Activity Is Also Dependent on the Length of Fermentation
Although several works have already shown that fermented whey can be used as an antibacterial agent in minimally processed salads, being able to reduce overall microorganism growth [8,40], most studies also use fermentation periods around 24 to 72 h [33,39]. Since lactic acid production was higher in longer fermentation periods, we set out to search the best timing along fermentation for higher antibacterial activity. Hence, whey samples were collected daily throughout the 120 h, and firstly tested against reference Gram-positive Listeria monocytogenes 4b and Gramnegative Escherichia coli O157:H7, and MIC were determined for each day (Table 1). Results show that our fermented whey reduced the growth of both bacteria, from day 1 in Escherichia coli O157:H7 and from day 4 in L. monocytogenes 4b, concomitantly to the increase in lactic acid produced ( Figure 1), respectively, of approximately 5 and 15 g·L −1 . However, more importantly, activities were significantly higher on the final day of fermentation, where MICs were more significantly reduced. Thus, our results indicate that fermentation periods longer than 72 h should be preferred in this type of study to obtain the best antibacterial activities.  Hence, overall results show that our fermentation protocol has the potential to reduce the environmental impact of whey by reducing its organic content.

Antibacterial Activity Is Also Dependent on the Length of Fermentation
Although several works have already shown that fermented whey can be used as an antibacterial agent in minimally processed salads, being able to reduce overall microorganism growth [8,40], most studies also use fermentation periods around 24 to 72 h [33,39]. Since lactic acid production was higher in longer fermentation periods, we set out to search the best timing along fermentation for higher antibacterial activity. Hence, whey samples were collected daily throughout the 120 h, and firstly tested against reference Gram-positive Listeria monocytogenes 4b and Gram-negative Escherichia coli O157:H7, and MIC were determined for each day (Table 1). Results show that our fermented whey reduced the growth of both bacteria, from day 1 in Escherichia coli O157:H7 and from day 4 in L. monocytogenes 4b, concomitantly to the increase in lactic acid produced (Figure 1), respectively, of approximately 5 and 15 g·L −1 . However, more importantly, activities were significantly higher on the final day of fermentation, where MICs were more significantly reduced. Thus, our results indicate that fermentation periods longer than 72 h should be preferred in this type of study to obtain the best antibacterial activities.

Five Days-Fermented Whey Presents Dose-Dependent Antibacterial Activities and Is Effective Against Several Bacterial Species
Based on the results in Figure 1 and Table 1, whey fermented for 120 h was selected to further analyze antibacterial activities towards thirteen pathogenic and spoilage strains, and results are presented in Figure 2. Results show that growth inhibition was induced in all strains, either being Gram-negative or Gram-positive species, corroborating a broad range of antibacterial activity. Subsequently, 120 h whey samples were evaluated against L. monocytogenes and assessed at three different dilutions (Table 2) and compared to milk and non-fermented whey. Results show there was a dose-dependent effect in 120 h fermented whey, whereas neither milk nor unfermented whey exhibited an antibacterial effect (p < 0.05) (

Five Days-Fermented Whey Presents Dose-Dependent Antibacterial Activities and Is Effective Against Several Bacterial Species
Based on the results in Figure 1 and Table 1, whey fermented for 120 h was selected to further analyze antibacterial activities towards thirteen pathogenic and spoilage strains, and results are presented in Figure 2. Results show that growth inhibition was induced in all strains, either being Gram-negative or Gram-positive species, corroborating a broad range of antibacterial activity. Subsequently, 120 h whey samples were evaluated against L. monocytogenes and assessed at three different dilutions (Table 2) and compared to milk and non-fermented whey. Results show there was a dose-dependent effect in 120 h fermented whey, whereas neither milk nor unfermented whey exhibited an antibacterial effect (p < 0.05) ( Table 2).

Figure 2.
Inhibition halos of different bacteria exposed to 120 h fermented whey. Chlorine was used as a positive control. Results are the average of three replicates ± standard deviation. Note: NI = No inhibition observed. Table 2. Halos of inhibition of L. monocytogenes growth exposed to 120 h fermented whey, milk, or non-fermented whey diluted in water.

Industrial Whey Is an Effective Disinfecting Agent When Applied to Lettuce, with Better or Similar Results When Compared Chlorine
Because we aimed to test our 120 h fermented whey in a more realistic manner, we set out to determine its efficacy as a lettuce disinfectant, testing quality indices such as texture and color, including its evaluation by a sensory panel, and also assessing its microbiological quality.  Because we aimed to test our 120 h fermented whey in a more realistic manner, we set out to determine its efficacy as a lettuce disinfectant, testing quality indices such as texture and color, including its evaluation by a sensory panel, and also assessing its microbiological quality. Table 3 summarizes the monitoring of the O 2 contents and production of CO 2 in the lettuce bags over the 10 days of cold storage. While oxygen was practically kept constant around 19% for the reference samples treated with water, for shredded lettuce sanitized with fermented whey, a reduction from 19 to 18% on the first day was observed, and then it remained practically constant. With chlorine, the reduction continued until the third day to values around 16%. For the CO 2 production, more pronounced differences are shown, becoming apparent that the bags treated with chlorine showed a higher reduction in O 2 levels and a larger production of CO 2 compared to the reference and whey treated bags. Consumption of O 2 and production of CO 2 could come from the respiration of the lettuce leaves and AM, as well as further production of CO 2 from anaerobic fermenting microorganisms. Although results suggest that water-and whey-treated samples have lower biological activity, the higher changes detected in the samples treated with chlorine could be due to the response of the lettuce cell leaves to the chemical stress of the chlorinated compounds.

pH
From Figure 3, we can observe that there are no major differences between the pH of the different treatments, with a tendency for a slight increase along the 10-day period. The reference, i.e., lettuce washed with distilled water, is the sample where the increase of pH is significantly (p < 0.05 i.e., p = 8.8 × 10 −6 ) higher, from 6.16 to 7.00, but still less than a unit of variation for a storage period of ten days. Martin-Diana et al. [8] reported a similar increase in pH in a study with lettuce and carrots and stated that this was a good index for the potential quality maintenance of the products.
According to Beuchat [47], vegetable products retain adequate quality in a pH range of 5-6.5. Our results show that samples treated with whey exceeded this value (6.57) only on day 9, and the samples treated with chlorine exceeded it on day 7 (6.75).  Table 3 summarizes the monitoring of the O2 contents and production of CO2 in the lettuce bags over the 10 days of cold storage. While oxygen was practically kept constant around 19% for the reference samples treated with water, for shredded lettuce sanitized with fermented whey, a reduction from 19 to 18% on the first day was observed, and then it remained practically constant. With chlorine, the reduction continued until the third day to values around 16%. For the CO2 production, more pronounced differences are shown, becoming apparent that the bags treated with chlorine showed a higher reduction in O2 levels and a larger production of CO2 compared to the reference and whey treated bags. Consumption of O2 and production of CO2 could come from the respiration of the lettuce leaves and AM, as well as further production of CO2 from anaerobic fermenting microorganisms. Although results suggest that water-and whey-treated samples have lower biological activity, the higher changes detected in the samples treated with chlorine could be due to the response of the lettuce cell leaves to the chemical stress of the chlorinated compounds.

pH
From Figure 3, we can observe that there are no major differences between the pH of the different treatments, with a tendency for a slight increase along the 10-day period. The reference, i.e., lettuce washed with distilled water, is the sample where the increase of pH is significantly (p < 0.05 i.e., p = 8.8 × 10 −6 ) higher, from 6.16 to 7.00, but still less than a unit of variation for a storage period of ten days. Martin-Diana et al. [8] reported a similar increase in pH in a study with lettuce and carrots and stated that this was a good index for the potential quality maintenance of the products.
According to Beuchat [47], vegetable products retain adequate quality in a pH range of 5-6.5. Our results show that samples treated with whey exceeded this value (6.57) only on day 9, and the samples treated with chlorine exceeded it on day 7 (6.75).

Texture
Texture is a major quality parameter, and together with color they are crucial for consumer acceptance. Texture measurements evaluated two parameters: resistance to puncture as a force in N, and deformation at rupture in length (mm), generally described as firmness and brittleness (the lower the deformation, the more brittle the material is), respectively. Results are shown in Figure 4a,b. As seen in Figure 4a, the firmness of the lettuce shreds was not substantially affected by any of the treatment and variation within 10 days of storage was quite small. As observed with the pH variation, the reference treatment with water is where the highest difference is found-74 mN, which is significantly (p = 2.2 × 10 −6 ) less firm, and 0.177 mm which is less brittle (p = 0.0027) after 10 days in cold storage, i.e., a reduction of 12% in firmness and 7% in brittleness, whereas with whey there was a significant increase in firmness of 69 mN (p = 8.2 × 10 −6 ), corresponding to a 10% increment and no significant variation on brittleness (p > 0.05), which is due to the calcium ions present in the whey solution that might reinforce the walls of the lettuce cells.

Texture
Texture is a major quality parameter, and together with color they are crucial for consumer acceptance. Texture measurements evaluated two parameters: resistance to puncture as a force in N, and deformation at rupture in length (mm), generally described as firmness and brittleness (the lower the deformation, the more brittle the material is), respectively. Results are shown in Figure 4a,b. As seen in Figure 4a, the firmness of the lettuce shreds was not substantially affected by any of the treatment and variation within 10 days of storage was quite small. As observed with the pH variation, the reference treatment with water is where the highest difference is found-74 mN, which is significantly (p = 2.2 × 10 −6 ) less firm, and 0.177 mm which is less brittle (p = 0.0027) after 10 days in cold storage, i.e., a reduction of 12% in firmness and 7% in brittleness, whereas with whey there was a significant increase in firmness of 69 mN (p = 8.2 × 10 −6 ), corresponding to a 10% increment and no significant variation on brittleness (p > 0.05), which is due to the calcium ions present in the whey solution that might reinforce the walls of the lettuce cells.

Color Measurement
In Table 4, the lightness L* color parameter for shreds of mid-tone, i.e., median green, can be seen. These results are too scattered to be able to differentiate between sanitizing treatments. The same was observed for the other color coordinates a* and b*. Although 30 shots were taken for each measurement and standard deviations are not high, the chances of taking shreds from different color groups are high and conclusive results cannot be drawn from these experiments. Nevertheless, it can be said that color was not dramatically affected by all treatments.

Color Measurement
In Table 4, the lightness L* color parameter for shreds of mid-tone, i.e., median green, can be seen. These results are too scattered to be able to differentiate between sanitizing treatments. The same was observed for the other color coordinates a* and b*. Although 30 shots were taken for each measurement and standard deviations are not high, the chances of taking shreds from different color groups are high and conclusive results cannot be drawn from these experiments. Nevertheless, it can be said that color was not dramatically affected by all treatments. Table 4. Lightness (L*) color parameter of the lettuce after sanitizer treatments, over 10 days in cold storage. Results represent the average of at least three replicate experiments (n = 3) ± SD.

Sensory Evaluation
From the sensory evaluation, one can see the resulting spider diagrams for chlorine and whey treatments in Figures 5a and 5b, respectively. Therefore, from the spider diagrams, one can see that all the attributes were highly scored, with all above the midpoint 5, over the 10 days of cold storage.
To compare the two treatments, overall acceptability was chosen to represent the sensory results, and in Figure 6 it can be seen that there are two major plateaus. The first showed top scores until day 5, and the second with scores around 6 from 7 to the 10th day of cold storage, probably related to the scores obtained for off-odors and flavor. From these results, no differences can be seen for the two treatments, with whey or with chlorine.

Sensory Evaluation
From the sensory evaluation, one can see the resulting spider diagrams for chlorine and whey treatments in Figure 5a and Figure 5b, respectively. Therefore, from the spider diagrams, one can see that all the attributes were highly scored, with all above the midpoint 5, over the 10 days of cold storage.
To compare the two treatments, overall acceptability was chosen to represent the sensory results, and in Figure 6 it can be seen that there are two major plateaus. The first showed top scores until day 5, and the second with scores around 6 from 7 to the 10th day of cold storage, probably related to the scores obtained for off-odors and flavor. From these results, no differences can be seen for the two treatments, with whey or with chlorine.    [48]. However, in general, the initial contamination of vegetables reflects the microbiota environment in which they were grown [49,50]. Before treatment (day 0), the initial AM was 6.10 log cfu·g −1 . Other works have found similar results in whole vegetables [28,51,52]. As can be observed in Figure 7a, AM counts sharply decreased after all treatments (p < 0.05). Throughout the storage, samples treated with whey always presented a significantly lower AM count (p < 0.001) when compared to water and chlorine, independent of time (p < 0.001). This is a strong indication of the effectiveness of fermented whey as an alternative disinfectant.
(a) Aerobic Microorganisms at 30 • C Enumeration of AM (aerobic microorganisms at 30 • C) is an indicator of quality and gives an estimate of total viable populations, both endogenous and contaminant microbiota, since microorganisms are inevitably introduced during manipulations [48]. However, in general, the initial contamination of vegetables reflects the microbiota environment in which they were grown [49,50]. Before treatment (day 0), the initial AM was 6.10 log cfu·g −1 . Other works have found similar results in whole vegetables [28,51,52]. As can be observed in Figure 7a, AM counts sharply decreased after all treatments (p < 0.05). Throughout the storage, samples treated with whey always presented a significantly lower AM count (p < 0.001) when compared to water and chlorine, independent of time (p < 0.001). This is a strong indication of the effectiveness of fermented whey as an alternative disinfectant.  [48]. However, in general, the initial contamination of vegetables reflects the microbiota environment in which they were grown [49,50]. Before treatment (day 0), the initial AM was 6.10 log cfu·g −1 . Other works have found similar results in whole vegetables [28,51,52]. As can be observed in Figure 7a, AM counts sharply decreased after all treatments (p < 0.05). Throughout the storage, samples treated with whey always presented a significantly lower AM count (p < 0.001) when compared to water and chlorine, independent of time (p < 0.001). This is a strong indication of the effectiveness of fermented whey as an alternative disinfectant. (a)

Psychrotrophic Microorganisms
Similarly to AM, psychrotrophic microorganisms (PM) counts were significantly reduced (p < 0.05) in all treatments at day 1 ( Figure 7b). Also, the highest count reduction (p < 0.001) to 2.20 log cfu·g −1 occurred in samples treated with whey. With chlorine, the count reduction found was 1.13 log cfu·g −1 . In the case of the PM, both for the lettuce treated with whey as well as with chlorine, after 5 days in cold storage, there were no significant differences between both treatments in respect to these counts.

Lactic Acid Bacteria
Because whey contains naturally high counts of LAB (in this case 10 7 cfu·g −1 ), which could induce a higher bacterial load into the product, we investigated the evolution of LAB counts in all treatments throughout cold storage time, and results are presented in Figure 7c. As can be seen, although on day 1 there was a significant difference between LAB counts in all treatments, from day 3 onward, the presence of live LAB from fermented whey did not confer a significant difference (p > 0.05) when compared to chlorine and water treatments. This type of bacteria is normally present in MP vegetables in loads from 2 to 6 log cfu·g −1 (Figure 7c; at day 0, our reference sample was around 3.5 log cfu·g −1 ), so LAB counts ended up being similar in all treatments tested, meaning that the introduction of LAB from whey would not pose any noticeable difference to the final product. Furthermore, it has been shown that the presence of these LAB could have a positive antimicrobial effect in lettuce and apple taste [35]. Additionally, the presence of lactic acid, antibacterial peptides, and low pH are contributive factors for the inhibitory effects of whey [53][54][55].

Psychrotrophic Microorganisms
Similarly to AM, psychrotrophic microorganisms (PM) counts were significantly reduced (p < 0.05) in all treatments at day 1 ( Figure 7b). Also, the highest count reduction (p < 0.001) to 2.20 log cfu·g −1 occurred in samples treated with whey. With chlorine, the count reduction found was 1.13 log cfu·g −1 . In the case of the PM, both for the lettuce treated with whey as well as with chlorine, after 5 days in cold storage, there were no significant differences between both treatments in respect to these counts.

Lactic Acid Bacteria
Because whey contains naturally high counts of LAB (in this case 10 7 cfu·g −1 ), which could induce a higher bacterial load into the product, we investigated the evolution of LAB counts in all treatments throughout cold storage time, and results are presented in Figure 7c. As can be seen, although on day 1 there was a significant difference between LAB counts in all treatments, from day 3 onward, the presence of live LAB from fermented whey did not confer a significant difference (p > 0.05) when compared to chlorine and water treatments. This type of bacteria is normally present in MP vegetables in loads from 2 to 6 log cfu·g −1 (Figure 7c; at day 0, our reference sample was around 3.5 log cfu·g −1 ), so LAB counts ended up being similar in all treatments tested, meaning that the introduction of LAB from whey would not pose any noticeable difference to the final product. Furthermore, it has been shown that the presence of these LAB could have a positive antimicrobial effect in lettuce and apple taste [35]. Additionally, the presence of lactic acid, antibacterial peptides, and low pH are contributive factors for the inhibitory effects of whey [53][54][55].

Conclusions
In conclusion, our overall results showed that the established fermentation for industrial whey has the potential to be a low-cost, scalable process to reduce the environmental impact of whey by reducing its organic content, and holds strong potential as an effective disinfecting agent when applied to lettuce, with better or similar results when compared to a 110 ppm chlorine solution. Also importantly, it did not alter the quality parameters of the shredded loose-leaf lettuce, which is highly prone to decay. Color was not substantially affected, and panelists were not able to discriminate from chlorine treatments. Furthermore, lettuce shreds treated with fermented whey showed a slight reinforcement within the cold storage time. Overall, these results not only suggest that fermented whey is indeed as effective as chlorine, but also corroborates that our technology of whey fermentation is effective in maintaining the quality of lettuce throughout storage.