Production of Conjugated Linoleic and Conjugated α-Linolenic Acid in a Reconstituted Skim Milk-Based Medium by Bifidobacterial Strains Isolated from Human Breast Milk

Eight bifidobacterial strains isolated from human breast milk have been tested for their abilities to convert linoleic acid (LA) and α-linolenic acid (LNA) to conjugated linoleic acid (CLA) and conjugated α-linolenic acid (CLNA), respectively. These bioactive lipids display important properties that may contribute to the maintenance and improvement human health. Three selected Bifidobacterium breve strains produced CLA from LA and CLNA from LNA in MRS (160–170 and 210–230 μg mL−1, resp.) and, also, in reconstituted skim milk (75–95 and 210–244 μg mL−1, resp.). These bifidobacterial strains were also able to simultaneously produce both CLA (90–105 μg mL−1) and CLNA (290–320 μg mL−1) in reconstituted skim milk. Globally, our findings suggest that these bifidobacterial strains are potential candidates for the design of new fermented dairy products naturally containing very high concentrations of these bioactive lipids. To our knowledge, this is the first study describing CLNA production and coproduction of CLA and CLNA by Bifidobacterium breve strains isolated from human milk in reconstituted skim milk.


Introduction
Conjugated linoleic acid (CLA) and conjugated -linolenic acid (CLNA) are bioactive lipids with potentially relevant benefits to human health. They have been shown to have in vitro and in vivo anticarcinogenic, antiatherogenic, antiinflammatory, and antidiabetic activities and ability to reduce body fat [1,2].
In relation to CLA, the bioactive isomer cis 9, trans 11 CLA is the most abundant in the diet, constituting more than 90% of the total CLA content in milk fat [2,3]. On the other hand, the CLNA isomers resulting from the metabolism of intestinal and rumen bacteria are cis 9, trans 11, cis 15 CLNA and trans 9, trans 11, cis 15 CLNA [1]. The predominant isomer is cis 9, trans 11, cis 15 CLNA, which has been detected at low concentrations in milk fat [4,5]. Presence of CLA and CLNA isomers in ruminant milk fat are the result of microbial partial biohydrogenation of dietary linoleic (LA) or -linolenic (LNA) acid to stearic acid metabolism pathway in the rumen by the action of the linoleic acid isomerase [6].
Nevertheless, as the current nutritional recommendations for whole fat dairy products are that their consumption should be limited, the CLA and CLNA content of human diet is too low for obtaining health beneficial effects. Therefore, a promising strategy to increase human intake of these bioactive lipids would be to include CLA and/or CLNAproducer bacteria in fermented dairy products. In the last years, several studies have reported that some lactic acid bacteria and bifidobacterial strains are able to efficiently convert LA to CLA in milk, milk-based media, and dairy products [10][11][12]. Moreover, other study demonstrated that a CLA-producing Bifidobacterium breve strain can be applied for the development of functional dairy products when used as a started culture [13]. In contrast, at present, we have not found studies showing CLNA production by bacteria in milk and dairy products. In this context, the aim of the present work was to evaluate the ability of some bifidobacterial strains isolated from human breast milk to produce CLA and/or CLNA when growing not only in MRS broth but also in reconstituted skim milk. In this study, we demonstrated that some Bifidobacterium breve strains are able to (co)produce CLA and CLNA in both media.

Bacterial Strains, Growth Media, and Conditions.
Eight bifidobacterial strains previously isolated from human milk [14,15] were used in the study ( Table 1). The bacterial strains were grown overnight at 37 ∘ C in MRS broth supplemented with 0.05% (w/v) L-cysteine-HCL (Sigma) and 0.1% (w/v) Tween-80 (MRS-Cys broth) under anaerobic conditions in an anaerobic station (Bactron II, Shellab, Cornelius, Oregon, USA). Three percent (v/v) of these cultures were transferred to fresh MRS-Cys broth (10 mL) containing free LA (500 g mL −1 ) and/or free LNA (500 g mL −1 ) and incubated at 37 ∘ C for 24 h under anaerobic conditions. The samples were analyzed when the bifidobacterial strains reach the early stationary phase, obtaining concentrations of ∼1 × 10 9 cfu mL −1 . Only the strains that showed CLA production in MRS-Cys broth after an initial qualitative screening were subsequently tested for CLA and/or CLNA production in 10% skim milk (Scharlau, Sentmenat, Barcelona, Spain) supplemented with 0.05% (w/v) L-cysteine and 0.8% (w/v) casamino acids (milkbased medium), as described above.
Since the production of CLA and CLNA by the three selected Bifidobacterium strains in MRS-Cys was similar, B. breve M7-70 was chosen as the model strain for the subsequent assays. First, the sensitivity of B. breve M7-70 to different concentrations of LA or LNA (0, 250, 500, 1000, 1500, and 2000 g/mL) was evaluated in MRS-Cys broth since LA and LNA have antimicrobial properties. Then, this strain was submitted to a comparative analysis of CLA versus CLNA production at different times (0, 1, 2, 3, 4, 6, 8, 24, and 48 h) in MRS-Cys broth.

Qualitative Screening of CLA Producers by UV Spectroscopy Method.
Lipid isolation from culture media was carried out using a chloroform/methanol (2 : 1, v/v) solution according to Folch method modified by [16]. The lipid residues obtained were subjected to a N 2 flow and remained dissolved in chloroform at −20 ∘ C until spectrophotometric analysis. For this analysis, lipid extracts (200 L) from each sample were placed on a quartz 96-well plate (Hellma GmbH & Co. KG, Müllhein, Germany) and total CLA was quantified at a wavelength of 233 in a spectrophotometer (Varioskan Flash, Thermo Fisher Scientific, Waltham, MA, USA) according to [10]. Measurements were obtained in triplicate.

Quantitative Analysis of CLA and CLNA Production.
The concentrations of CLA and CLNA in the culture media were determined using a direct transesterification method [17]. Heptadecanoic acid (C17:0; Sigma) was added to the samples as an internal standard. The fatty acid methyl esters (FAMEs) were dissolved in n-hexane and determined by gas liquid chromatography (GLC) in a chromatograph (Clarus 500, Pelkin Elmer, Beaconsfield, UK) equipped with a VF-23 column (30 m × 0.25 nm × 0.25 m, Varian, Middelburg, Netherlands). For GLC analysis, the initial temperature was 80 ∘ C. Then, the temperature was increased to 170 ∘ C at 30 ∘ C min −1 , held at 170 ∘ C for 3 min, increased to 230 ∘ C at 30 ∘ C min −1 and, and finally held at 230 ∘ C for 7 min. Helium was used as the carrier gas at a pressure of 15 psig and with a split ratio of 1 : 50. The injection volume was 0.5 L and the analysis time was 15 min. Peaks were identified by comparing the retention times of CLA methylated standards (Nucheck, USA) and by gas chromatography-mass spectrometry (GC/MS). CLA and CLNA concentrations were expressed as g mL −1 and their conversion rates from LA and LNA were calculated using the formula [CLA/(CLA + LA)] × 100 and [CLNA/(CLNA + LNA)] × 100, respectively.

Results
The eight bifidobacterial strains assayed in this work were screened spectrophotometrically at A 233 for CLA production from the LA added to the growth media, following the rapid method described by [10]. With this approach, a total of 3 strains (Bifidobacterium breve ZL12-28, B. breve 29M2, and B. breve M7-70) were identified as able to transform LA into CLA (Table 1).
Subsequently, each of the three selected CLA-producing strains was assayed for CLA and CLNA production by GLC determination after incubation for 24 h at 37 ∘ C in MRS-Cys and skim milk (Tables 2 and 4). The concentration of CLA produced by the selected Bifidobacterium strains reached values above 158.7 g mL −1 , indicating that the minimal conversion rate from the added LA was approximately 74%. The bacterial strains were able to produce different CLA isomers, such as cis 9, trans 11 CLA, trans 10, cis 12 CLA and trans 9, trans 11 CLA ( Figure 1); among them, cis 9, trans 11 CLA (rumenic acid) was the most abundant isomer, accounting for more than 80% of the total CLA in all cases ( Table 2). As it has been reported, many bacteria are inhibited by free long-chain fatty acids in the media [18]; however, B. breve M7-70 was able to grow in the presence of LA and LNA at concentrations up to 1500 and 500 g mL −1 , respectively ( Table 3).
Production of CLNA in MRS-Cys by B. breve strains was higher than that of CLA since the concentrations found in the respective culture media were higher than 200 g mL −1 , and the LNA to CLNA conversion rate was close to 100% ( Table 2). Two CLNA isomers (cis 9, trans 11, cis 15 CLNA and trans 9, trans 11, cis 15 CLNA) could be detected in the chromatogram profiles (Figure 2), and cis 9, trans 11, cis 15 CLNA (rumenic acid) accounted for approximately 80% of the total amount of CLNA in the cultures of the three strains ( Table 2). The conversion of LNA to CLNA by B. breve M7-70 in MRS-Cys was faster than that of LA to CLA since CLNA production began after 2 h incubation at 37 ∘ C while CLA formation required at least 6 h of incubation to be detected ( Figure 3).
Subsequently, the three selected bifidobacterial strains were assayed for CLA and/or CLNA production in the milkbased medium after 24 h incubation at 37 ∘ C. CLA production was above 75 g mL −1 for the three B. breve strains (Table 4). Both CLA production and LA conversion rate were lower in milk than in MRS-Cys. As expected, the predominant CLA isomer produced in reconstituted skim milk was cis 9, trans 11 CLA, accounting for more than 80% of total CLA (Table 4).
All the selected strains were able to produce CLNA when growing in the milk-based medium. In this case, the concentration of CLNA produced by these bifidobacteria  (200 g mL −1 ) and the conversion rates (∼100%) were similar to those observed in MRS-Cys broth (Table 4). Rumenic acid was also the dominant isomer, accounting for approximately 90% of the CLNA total amount. Finally, the three selected bifidobacterial strains were tested for their ability to produce CLA and CLNA when both substrates (LA and LNA) were added to the milkbased medium. The three B. breve strains were able to simultaneously produce CLA and CLNA in these growth conditions (Figure 4). The CLA and CLNA concentrations produced were about 100 g mL −1 of CLA and 300 g mL −1 of CLNA (Figure 4). Interestingly, the production of both bioactive lipids was much higher when both substrates, LA and LNA, were present at the same time in the growing medium than when they were individually added (Table 4 and Figure 4).

Discussion
Bifidobacteria are numerically important members of the human gut microbiota and are believed to play a beneficial role in maintaining the health of the host. Some studies have suggested that infants with delayed bifidobacterial colonization and/or decreased bifidobacterial numbers may be more susceptible to a variety of gastrointestinal or allergic conditions [19,20]. In these cases, the exogenous administration of selected bifidobacterial strains, alone or in combination with lactic acid bacteria, can reduce the incidence of such conditions [21][22][23]. Therefore, bifidobacteria are generally regarded as potentially probiotic microorganisms. Recently, it has been shown that human milk is a source of live lactic acid bacteria and bifidobacteria to the infant gastrointestinal tract [24,25]. If health benefits could be associated with bifidobacterial strains isolated from such biological fluid, then they would be immediately regarded as particularly attractive microorganisms since they would fulfil some of the main criteria generally recommended for human probiotics, such as human origin and adaptation to mucosal and dairy substrates [26,27].
Since production of conjugated fatty acids by some bifidobacteria and lactic acid bacteria has been reported in the last years [1,2,12,28], it is considered of interest to have human milk strains available with such capability. In this work, we describe the assayed conditions to select and characterize three B. breve strains with the ability to produce high amounts of CLA and CLNA during their growth in skim milk. It is not strange that the selection process ended with three strains belonging to the species B. breve, since this species seems to be particularly suited for production of these bioactive lipids [28][29][30][31]. The bioactive lipid production is a strain-specific attribute and, in fact, a non-CLA-producing B. breve strain (B. breve ZL12-22) was also found in this study.
It has been hypothesized that some bacteria can convert LA to CLA and LNA to CLNA as a detoxification mechanism  [18,28,29,32]. LA and LNA have antimicrobial activity and may alter the permeability of the plasmatic membrane of some Gram-positive bacteria [33]. It has been shown that both substrates have inhibitory effects on the growth of CLA and CLNA producing strains [28,34]. Generally, free LNA is more toxic to bacteria than free LA, which is coincident with the results obtained in this work with B. breve M7-70. The hypothesis of the detoxification mechanism may explain why CLNA production by the three B. breve strains was more efficient than CLA production. It also could explain why the production of both CLA and CLNA was higher when both substrates were added together than when the strains were grown on each substrate separately (Table 4 and Figure 4). At present, CLA is better characterised than CLNA. There are some research works that report the health-promoting properties of CLNA [1]. A recent study demonstrated the ability of a B. breve strain to produce CLNA and other conjugated fatty acids, such as conjugated -linolenic acid (CGLA) or conjugated stearidonic acid (CSA), fromlinolenic acid and stearidonic acid, respectively [35]. Another work reported that B. breve NCIMB 702258 displayed a high conversion rate (79%) of -linolenic acid into the cis 9 trans 11 cis 15 CLNA isomer. This isomer can inhibit the growth of SW480 colon cancer cells [31]. All these mentioned works have described the CLNA production during growth of the strains in MRS broth. In contrast, the results of our study indicate that the selected B. breve strains were able to produce significant amounts of CLNA (200 g mL −1 ) in skim milk, with a conversion rate of LNA close to 100%. In addition, this study has demonstrated that B. breve strains were able to coproduce high levels of CLA and CLNA during growth in skim milk. Coproduction of CLA and CLNA in the described concentrations would be of relevance to increase the amount of these bioactive lipids in fermented milks. Globally, these results suggest that the three selected B. breve strains have a strong potential to be used as probiotics in fermented milks in order to increase human intake of CLA and CLNA. Further studies to optimize the culture conditions for increasing CLA and CLNA production at large-scale by fermentation processes would be needed. Additional studies will be also required to further elucidate the relevance of consumption of these bioactive lipids in human health.