2.1. Materials, reagents, and milk samples
The Pseudomonas spp. used in this study were isolated from cold raw milk samples from five dairy farms, with counts which ranged from 2.3 × 105 to 1.7 × 106 CFUmL-1 (Almeida et al., 2017). The milk samples were stored in cooling tanks, collected under aseptic conditions, and were kept in a Styrofoam box with reusable ice packs, until analysis. For Pseudomonas spp. isolation, CFC agar-base was used (30 °C for 48 h) (Almeida et al., 2017) and the isolated strains were stored at −20 °C in BHI broth containing 40% glycerol.
The reagents and materials used in this study are described below. (a) CFC agar-base (Himedia, Mumbai, India), Tributyrin Agar (Himedia, Mumbai, India), Brain Heart Infusion (BHI) broth (Himedia, Mumbai, India), and skimmed milk powder (Molico, Nestlé, São Paulo, Brazil) were used for the determination of P. fluorescens and P. putida populations. (b) For PCR, the following reagents were used: (b1) Wizard Genomic DNA Extraction Kit (Promega Corporation, Madison, USA), buffer 10X (Invitrogen, CA, USA), MgCl2 (Invitrogen, CA, USA), dNTPs set kit (Invitrogen, CA, USA), oligonucleotides PA-GS-F, PA-GS-R, 16 SPSER, 16 SPSEfluF, P734, P1455r (Invitrogen, CA, USA), GoTaq DNA polymerase (Invitrogen, CA, USA). (b2) For the positive control in PCR, ATCC 27853, ATCC 13525, and ATCC 31483 (Coleção de Culturas Tropical – Fundação André Tosello, São Paulo, Brazil) were used. (b3) Ultrapure agarose, ultrapure tris, EDTA, and Sybr Safe (Invitrogen, CA, USA) were used for electrophoresis. (c) isopropanol (Syth, Diadema, Brazil) petroleum ether (Syth, Diadema, Brazil), sulfuric acid (Syth, Diadema, Brazil), thymol blue (Sigma, Sao Paulo, Brazil), butanol (Syth, Diadema, Brazil), potassium hydroxide (Syth, Sao Paulo, Brazil) and ethanol (Syth, Diadema, Brazil) for the Lipo R methods.
2.2. <italic>Pseudomonas</italic> spp. isolation, confirmation of <italic>P. fluorescens</italic> and <italic>P. putida</italic>, and lipolytic indexes
2.2.1. PCR
Bacterial genomic DNA was extracted using the Wizard Genomic DNA Extraction Kit following the manufacturer’s instructions. The isolated DNA was stored at −80 °C. The extracted genetic material was subjected to PCR for the identification of the gender Pseudomonas spp. with forward PA-GS-F (5’-GACGGGTGAGTAATGCCTA-3’) and reverse PA-GS-R (5’-CACTGGTGTTCCTTCCTATA-3’) primers (Spilker et al., 2004), which amplified the 16S rRNA (618 bp) gene region. For the identification of the species P. fluorescens, the bacterial isolates were subjected to PCR reactions with oligonucleotides 16 SPSEfluF (5’-TGCATTCAAAACTGACTG-3’) and 16SPSER (5’-AATCACACCGTGGTAACCG-3’) (Scarpellini et al., 2004) for the 16S rRNA gene, which is characteristic of P. fluorescens. For the identification of P. putida, were followed the protocol with the oligonucleotides P734 (5’-CAA CTCGGGCGTTGGCATTCTGCT-3’) and P1455r (5’-CAAGATCGCCTGGGTACGACGGTT-3’) was followed, for amplifying a fragment of 744 bp from the gyrB gene (Yamamoto and Harayama, 1995) which is of P. putida. Ultrapure water was used as the negative control and DNA from the strains P. fluorescens (ATCC 13525) and P. putida (ATCC 31483) were used as the positive control.
All the PCR products were subjected to agarose gel (1.5%) electrophoresis, and the gel was stained with Sybr Safe, followed by photo-documentation under ultraviolet light.
2.2.2. Determination of P. fluorescens and P. putida populations
From eight and six strains confirmed to be P. fluorescens and P. putida, respectively, one strain of each bacterium was used for the experiment. The strains stored at −20 °C were inoculated in 200 mL of 12% milk reconstituted from skimmed milk powder and incubated at 21 °C for 48 h. The lipolytic capacity of the strains was measured by surface plating on Tributyrin Agar (21 °C, 72 h), considering the strains which showed a transparent halo (Frank et al., 1992) as positive.
Subsequently, the number of viable cells of Pseudomonas was determined by surface plating on CFC (cefaloridine, fusidic acid, cetrimide) agar-base (Himedia, Mumbai, India), at 30 °C for 48 h (Dogan and Boor, 2003). After reaching the required bacterial cell count, decimal dilutions were made in 0.85% saline solution to obtain final concentrations of 2 log, 5 log, and 6 log CFU/mL. Each selected dilution was immediately used to set up cultures.
2.2.3. Inoculation with Pseudomonas and measurement of the lipolytic index
For the research, aliquots of 200 mL of 12% reconstituted milk from whole milk powder (sterilized at 121 °C, 15 min) (Ninho, Nestlé, Brazil) were used. Separately, the milk samples were inoculated with 2 mL of Pseudomonas inocula previously prepared (2, 5, and 6 log CFU/mL) and incubated at 2 °C, 4 °C, and 8 °C for 96 h. The temperature of 4 °C was selected to simulate the temperature stipulated by Brazilian regulations for raw milk storage at dairy farms. The storage of raw milk at 2 °C and 8 °C were determined to be common in the Brazilian dairy farm (Almeida et al., 2017) and lipolytic activity, even in the presence of a small number of psychrotrophs, was detected at these two temperatures (Wiking et al., 2003).
The levels of free fatty acids (FFA) (mEq/L) in the milk samples with Pseudomonas spp. inoculum were quantified by the Lipo R method (MAHIEU, 1984) in three steps. In an initial extraction step, milk (4mL) was added to 16 mL of Lipo R reagent (441 mL of isopropanol, 447.75 mL of petroleum ether, and 11.25 mL of 4 N sulfuric acid) and 5 mL of distilled water into test tubes with screw threads. The tubes were shaken by inversion 15 times, followed by resting for 5 min and the supernatant (supernatant-1) was obtained. Then the rinsing process was carried out by transferring 8 mL of the supernatant-1 to test tubes, followed by the addition of 4 mL of rinsing solution (sulfuric acid at 0.05 % in aqueous medium) and the mixture was shaken (by inversion) 15 times. After resting for 5 min, the supernatant-2 was obtained. Finally, 4 mL of the supernatant-2 were mixed with 5 drops of thymol blue indicator (1 g of thymol blue in 1.6 L of Butanol-2), and titrated with potassium hydroxide in an ethanolic solution 0.002 N.
The final results were obtained using the formula:
FFA mEq/L = (X − B). Fc
where X is the volume of base required to neutralize the supernatant-2; B is the volume of supernatant-2, and Fc is the correction factor for potassium hydroxide.
The Pseudomonas spp. enumeration (CFU/mL) and quantification of FFA levels (mEq/L) were carried out every 24 h in duplicate and triplicate, respectively, and time zero was considered the negative control. Two independent experiments were carried out with each of the strains.
2.2.4. Statistical analysis
The influence of storage time, temperature, and Pseudomonas spp. population over the lipolytic index was assessed through a multiple regression analysis with three explanatory variables: storage timeat four levels (24, 48, 72, and 96 h); temperature, at three levels (2 °C, 4 °C, and 8 °C) and bacterial population, with three levels (2, 5, and 6 log CFU/mL). A total of 72 experimental runs were conducted, since plating was performed in duplicate. The response data was plotted as a response surface to obtain a better understanding of the underlying mechanism behind the lipolytic ability of Pseudomonas. The data were analyzed using the software Statistica, release 13.2 (Quest Software Inc., California, USA).
We also accounted for the existing differences in the lipolytic index between the strains of P. fluorescens and P. putida. Thus, differences were evaluated by the Mann-Whitney U test and the significance level was set at α = 0.05.