Growth, Organic Acids Profile and Sugar Metabolism Of

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Food Research International 48 (2012) 21–27

Contents lists available at SciVerse ScienceDirect

Food Research International

journal homepage: www.elsevier.com/locate/foodres

Growth, organic acids proﬁle and sugar metabolism of Biﬁdobacterium lactis in co-culture with Streptococcus thermophilus: The inulin effect

Ricardo Pinheiro de Souza Oliveira a,⁎, Patrizia Perego b, Maricê Nogueira de Oliveira a, Attilio Converti b

a Biochemical and Pharmaceutical Technology Department, São Paulo University, São Paulo 05508-900, Brazil b Department of Chemical and Process Engineering, Genoa University, Genoa University, Genoa I-16145, Italy

article info abstract

Article history: The organic acids profile, sugar metabolism and biomass growth of Streptococcus thermophilus (St) and Received 18 August 2011 Bifidobacterium lactis (Bl) have been studied in pure cultures or binary co-culture (St–Bl) in skim milk either Accepted 14 February 2012 containing 40 mg/g of inulin or not. With inulin, the time required by St, Bl and St–Bl to complete fermentation (i.e., when the pH reached 4.5) was about 14, 8 and 49% shorter than without inulin, respectively. This prebiotic Keywords: also enhanced the levels of lactic and acetic acids and volatile compounds, showing a positive synbiotic effect be- Inulin tween pre- and probiotics. In particular, the St–Bl co-culture showed final concentrations of both microorganisms Streptococcus thermophilus Bifidobacterium lactis about 15 and 38% higher than in their respective pure cultures, thus highlighting a clear synergistic effect between Biomass these microorganisms due to mutual interactions. In addition, the well-known bifidogenic effect of inulin was Metabolic end-products confirmed. © 2012 Elsevier Ltd. All rights reserved.

1. Introduction Donkor et al. (2007) reported that, in set-type yogurt, prebiotics exert a protective effect and potentiate probiotics activity. The capability In recent years, the number of functional dairy products enriched of these microorganisms to ferment prebiotic ingredients may in fact be with live probiotic strains belonging mainly to the Bifidobacterium an especially important characteristic, because the availability of carbo- and Lactobacillus genera has remarkably increased. According to hydrates that escape metabolism and adsorption in the small intestine FAO/WHO (2002), probiotics are live microorganisms which, when has a major influence on the colon microbiota (Mattila-Sandholm et administered in adequate amounts, confer health benefits on the host. al., 2002). As a significant part of the world population suffers gastrointestinal On the other hand, although the homofermentative and heterofer- diseases caused by pathogenic bacteria, numerous studies are currently mentative pathways involved in the metabolism of lactic acid bacteria in progress (Donkor, Nilmini, Stolic, Vasiljevic, & Shah, 2007). (LABs) and bifidobacteria are well documented (Axelsson, 1998; Available evidence indicates that ingestion of probiotic bacteria Cronin, Ventura, Fitzgerald, & van Sinderen, 2011; Mayo et al., 2010), may reduce the severity and frequency of diarrheal diseases and there is limited published information on the microbial metabolic improve lactose digestibility among lactose-intolerant individuals interactions in probiotic co-cultures used to ferment non-fat milk. (Mattila-Sandholm et al., 2002). These therapeutic effects are the reason From a technological standpoint, these interactions are very for which most probiotics are used in yogurts, fermented milks, ice important for the correct development of flavor and texture of functional creams and nutraceutical products (Cruz et al., 2010; Oliveira, Florence, fermented dairy product (Cruz et al., 2010; Martin et al., 2011; Østlie, et al., 2009). Helland, & Narvhus, 2003; Salazar et al., 2009), being responsible for Nowadays, to enhance their therapeutic effects, it has recently been variations in the amounts of organic acids, volatile compounds and studied the ability of probiotics to partially ferment oligosaccharides exopolysaccharides released during manufacture. Metabolically, they and inulin (prebiotics) in fermented milk (Oliveira, Perego, Oliveira, & are relevant so as to allow microorganisms to better obtain energy and Converti, 2011; Rodrigues et al., 2011; van den Broek & Voragen, maintain their NAD+/NADH ratio (Axelsson, 1998; Mayo et al., 2010). 2008). In fact, these compounds may be also added to foods in attempts Species and strains of Bifidobacterium genus are classified as Gram- to increase fiber ingestion and obtain low-calorie dairy products positive, non-spore forming, non-motile and catalase-negative anaer- (Oliveira, Florence, et al., 2009; Roberfroid, 2005). obes, which may assume various shapes. Bifidobacteria differ from other colonic genera in the mechanism employed to ferment carbohydrates. Lacking the enzymes aldolase and glucose-6-phosphate NADP+ oxidoreductase (Mayo et al., 2010), they are in fact unable to carry out either the usual glycolytic pathway or the hexose monophosphate fi ⁎ Corresponding author. Tel.: +55 390 10 3532593; fax: +55 390 10 3532586. shunt. In particular, the bi dus pathway, which is characterized by the E-mail address: [email protected] (R.P.S. Oliveira). presence of fructose-6-phosphate phosphoketolase (Fandi, Ghazali,

Yazid, & Raha, 2001), allows bifidobacteria to produce more ATP from 2.4. Fermentations carbohydrates (2.5 mol ATP/mol glucose) than conventional hetero- and homofermentative pathways, leading to a theoretical yield of After inoculation, the flask content was transferred to a 3.0 L- 1.5 mol acetic acid and 1 mol lactic acid (Cronin et al., 2011). fermenter, model Z61103CT04 (Applikon, Schiedam, The Nether- These bacteria bring about several benefits to their human hosts, such lands), with 2.0 L-working volume and provided with an electron- as vitamin production, anticarcinogenic activity, immunostimulating ic device, model ADI1030 (Applikon). Dissolved oxygen (DO) effects, hypocholesterolemic power and pathogen inhibition (Ejtahed concentration was measured by a sterilized galvanic electrode, et al., 2011). In particular, Bifidobacterium animalis subsp. lactis has InPro6000 Series (Mettler-Toledo, Novate Milanese, Italy). Batch been found in children's intestine, where it promotes production of IgA fermentations were carried out with single and mixed cultures, that is important in their immune system; it also possesses a number in triplicate, without any agitation, at 42 °C, and stopped when of desirable technological features, i.e. tolerance to oxygen, acid and the pH reached 4.5, according to the common practice of yogurt bile resistance and ability to grow on milk-based media (Chen, Ruan, preparation. Zhu, & He, 2010; Janer, Arigoni, Lee, Pelaez, & Requena, 2005). In the present study, the associative behaviors of Streptococcus thermophilus TA040 and B. animalis subsp. lactis BL 04 have been 2.5. Analytical methods investigated in skim milk through a metabolic study based on the a) lactic and acetic acids formation from glucose and partly from galactose, Determination of organic acids was carried out using the method b) release of unmetabolized galactose, c) diacetyl and acetoin formations, described by Donkor et al. (2007). Five 3.0 mL-fermented milk and d) biomass growth. Finally, we examined the effect of inulin, as one samples were taken in different values of pH (6.5, 6.0, 5.5, 5.0 and of the most attracting prebiotics in functional food preparation, on the 4.5) and then mixed with 80 μL of 15.5 M nitric acid. Subsequently, fermentation patterns either of pure cultures of S. thermophilus and B. the samples were diluted with 1.0 mL of the mobile phase (0.01 M lactis or of their co-culture. sulfuric acid). The resulting mixture was centrifuged at 15,000×g for 20 min using an Eppendorf 5415R centrifuge (Eppendorf, Milan, 2. Materials and methods Italy) for removal of proteins. The supernatant was filtered through membrane filters with 0.20 μm-pore diameter (MilliporeTM, Milan, 2.1. Microorganisms Italy) in a HPLC vial. A high-performance liquid chromatograph, model 1100 (Hewlett Packard, Palo Alto, CA), was used to analyze Two commercial starter freeze-dried strains (Danisco, Sassenage, lactose, glucose, galactose, acetic acid, formic acid, ethanol, diacetyl, France) were used in this study, specifically S. thermophilus TA040 acetoin, and lactic acid. The system consisted of a HP-1050 Intelligent (St) and B. animalis subsp. lactis BL 04 (Bl). Auto Sampler, a HP-1047A Refractive Index Detector, a HP-1050 UV Detector and a HP-1050 pump. Separation was achieved using a 2.2. Milk preparation Supelcogel H59304-U column (Sigma Aldrich, Bellefonte, PA) at 50 °C with 0.01 M sulfuric acid as eluent at 0.4 mL/min flow rate. Milk was prepared adding 13 g of skim powder milk (Castroni, The column was stabilized for at least 3 h before use utilizing the same Reggio Emilia, Italy) in 100 g of distilled water. Skim milk base was either solution under the same conditions as the separation. Quantification of used as such (control) or supplemented with 40 mg of inulin/g (trade organic acids was performed from standard curves obtained using name: Beneo TM) (Orafti Active Food Ingredients, Oreye, Belgium). The solutions of pre-determined concentrations (Donkor et al., 2007). Each above solid content of milk corresponds to the average value for integral analysis was done in triplicate. cow milk (Restle, Pacheco, & Moletta, 2003), while the concentration of After dilution of samples and casein precipitation by acidification inulin is in the range admitted by the Brazilian legislation for yogurts to pH 4.5 with HCl, except the sample which was taken at pH 4.5 (ANVISA, 2002). Both milks were thermally treated at 90 °C for 5 min (Hipp, Groves, Custer, & McMeekin, 1950), biomass concentration in water bath, model Y14 (Grant, Cambridge, United Kingdom), and was determined in pure cultures by optical density (OD) measure- transferred to 1.0 L-sterile flasks, cooled in ice bath, distributed into ments at 640 nm using a UV–vis spectrophotometer (Model Lambda 250 mL-sterile Schott flasks inside a laminar flow chamber, and stored 25, Perkin Elmer, Wellesley, MA) and an OD versus dry weight cali- at 4 °C for 24 h before use. bration curve. For dry weight determinations, cells were harvested by centrifugation cycles (16,000×g for 10 min) in Eppendorfs, then 2.3. Inoculum preparation separated from casein by filtration on membranes with 0.20-μm- pore diameter, washed with 2 volumes of distilled water and dried B. lactis pre-culture was prepared by dissolving 45 mg of freeze-dried to constant weight at 101 °C. In co-cultures, biomass was determined culture in 50 mL of milk (10% of total solids; autoclaved at 121 °C for by the same selective cell counts procedure earlier described for pre- 20 min). After blending and activation at 42 °C for 30 min, 1.0 mL of cultures, and the results were expressed in g/L dry weight using the pre-culture was inoculated in 500 mL-Erlenmeyer flasks containing LogCFU/mL versus g/L dry weight calibration curves previously- 250 mL of skim milk. S. thermophilus pre-culture was prepared in the prepared for each microorganism. same way by adding 90 mg of its freeze-dried culture to 50 mL of milk. Cell counts of these pre-cultures were made in triplicate, as previously described by Oliveira, Perego, Converti, and Oliveira (2009), and ranged 2.6. Statistical analysis from 6.0 to 6.3 LogCFU/mL. In particular, samples (1.0 mL) were added to 9.0 mL of 0.1% sterile peptonated water; then, appropriate dilutions The experimental data of biomass concentration collected in pure or were made. After verifying the inability of the S. thermophilus strain binary co-cultures of S. thermophilus (St) and B. lactis (Bl) at the end of employed in this study to grow on MRS medium, it was plated into fermentations were presented as mean values. Variations with respect M17 Agar (Oxoid, Basingstoke, UK) and then submitted to aerobic to the mean values were presented as standard deviations. Mean values incubation at 37 °C for 48 h (Liu & Tsao, 2009). B. lactis was counted in of concentrations were submitted to analysis of variance (ANOVA) by MRS agar medium supplemented with 50 g/L cysteine (Sigma Chemical the Statistica Software 6.0 (Padova, Italy). They were compared using Co., St. Louis, MO, USA) and incubated anaerobically in anaerobic jar at the Tukey's test at significance level (P)b0.05, and different letters 37 °C for 72 h (Oliveira, Perego, et al., 2009). Anaerobic conditions were were used to label values with statistically significant differences ensured by the use of AnaeroGen (Oxoid). among them. R.P.S. Oliveira et al. / Food Research International 48 (2012) 21–27 23

3. Results and discussion Bl (+26%) or in St–Bl (+8%), respectively. On the other hand, the concentration of acetic acid in St–Bl (1.2±0.0 g/L) was less than 3.1. Acidification profiles 20% of that in Bl, likely due to a slight inhibition of the heterofermentative features of B. lactis like that proposed by Rodrigues et al. Fig. 1 shows the acidification curves of pure cultures of S. thermophilus (2011). According to the same authors, this behavior contributes to (St) and B. lactis (Bl) and a binary co-culture of S. thermophilus and B. lactis a better stability and organoleptic quality of the functional foods. (St–Bl), either in milk as such or in milk supplemented with 40 mg of Metabolically, members of the Bifidobacterium genus produce inulin/g at 42 °C until reaching pH 4.5. Important differences were acetic acid and lactic acid from lactose via the fructose-6-phosphate observed in the acidification profiles of both mono-cultures and the St– shunt, leading to the formation of 3 mol of acetic acid and 2 mol of Bl co-culture. lactic acid per 2 mol of consumed glucose (Scardovi & Trovatelli, It should be noted that the time to complete fermentation ranged 1965). However, in the present work, the molar ratio referred to lac- from 6.5 h (St–Bl in the presence of inulin) to 16.4 h (Bl in the absence tose was little lower (2.54:2) than the theoretical one (3:2), which of inulin), which means that the acidification profile depended not cannot be justified by the formation of little amounts of acetoin and only on the interactions between St and Bl, but also on the inulin supple- diacetyl (see the next section). Østlie et al. (2003) ascribed a more mentation. In the presence of inulin, the time to complete fermentations significant reduction of molar ratio (1.7:2) in B. animalis BB12 to a by St, Bl and St–Bl was in fact 13.8, 7.9 and 49.2% shorter than without temperature increase up to 45 °C and reported that, organoleptically, inulin, respectively (Fig. 1). Such an effect demonstrates that inulin it is important that the level of acetic acid is not too high in fermented stimulated the metabolism of both microorganisms, thus confirming dairy products. Therefore, taking into account that no acetaldehyde its prebiotic effect already reported by other authors (Akalın&Unal, was detected in the medium, we can infer that the use in the present 2010; Donkor et al., 2007; Oliveira, Florence, et al., 2009). work of a temperature (42 °C) significantly higher than the optimum for Bl (37 °C) could have reduced the aldehyde dehydrogenase activity 3.2. Sugar metabolism so as to decrease the rate of acetic acid formation. As expected the acetic/lactic acids ratio dramatically decreased (to 0.4:2) in the St–Bl Fig. 2 illustrates the metabolic behaviors of St and Bl, either in pure co-culture, because of the well-known inability of St to produce acetic cultures or binary co-cultures, which were monitored in fermented acid. Therefore, supposing a metabolic behavior of Bl in co-culture skim milk by determination of the main metabolic products (lactic with St similar to that in its own mono-culture, we can estimate from acid and acetic acid), consumed lactose and released galactose. The this result that 84% of produced lactic acid was linked to the homofer- concentrations of all the metabolic products of these cultures increased mentative St metabolism. Wide differences in end-products formation during the fermentations mainly due to metabolization of the glucose were actually reported in literature, depending on the strain, the carbon moiety of lactose, while a relevant portion of the galactose one was source, the cultural medium, and the growth conditions (Palframan, excreted in the medium (Fig. 2). Gibson, & Rastall, 2003; van der Meulen, Verbrugghe, & de Vuyst, Without inulin, Bl released significantly (Pb0.05) less galactose 2006; Vasiljevic, Kealy, & Mishra, 2007). (30.2±0.6 g/L) than St (+4%) and St–Bl (+9%) (Table 1), likely because The presence of inulin (Fig. 2B) significantly enhanced (Pb0.05) both S. thermophilus has weak transcription from gal promoters or mutations lactic and acetic acid levels. In fact, the concentrations of lactic acid in the Leloir genes (de Vin, Rådström, Herman, & de Vuyst, 2005). Bifido- released by St (11.0±0.4 g/L), Bl (8.1±0.4 g/L) and St–Bl (9.5±0.5 g/ bacteria are in fact saccharolytic organisms that have been shown to L) were about 15, 7 and 7% higher than those obtained in the absence possess the ability to ferment this sugar (Cronin et al., 2011)through of inulin (Table 1), and the ones of acetic acid about 17% (St–Bl) and the activity of the enzymes lacto-N-biose phosphorylase and gal 1-P 23% (Bl) higher. As far as the acetic/lactic acids ratio is concerned, it uridylyl transferase (Urashima, Kitaoka, Asakuma, & Messer, 2009). was increased only a little (to 2.70:2) by the addition of inulin, while As shown by the ANOVA results presented in Table 1, consistently the fraction of lactic acid produced by St in the St–Bl co-culture practically with the homofermentative nature of St and the heterofermentative keptthesame(83%). one of Bl, the concentration of lactic acid in St mono-culture (9.6± These results demonstrate that the fermented skim milk produced in 0.3 g/L) was statistically significantly higher (Pb 0.05) either than in the present study can be considered as a potential synbiotic one, according to the definition proposed by Holzapfel and Schillinger (2002),in that the interaction between pre- and probiotics was positive. In particular, the bifidogenic effect of inulin can be justified by a number of different causes on the basis of literature findings. The most likely hypothesis is that this prebiotic can stimulate the metabolism of bifidobacteria like it does in LABs, as the likely result of the increased level of fructose released from its partial hydrolysis mediated by inulinase, which may have been metabolized as an additional carbon and energy source (Mayo et al., 2010). In the presence of external oligosaccharides and polysaccharides like inulin, bifidobacteria are in fact able to uptake the monomers from their hydrolysis through the fructose-6-phosphate shunt (van der Meulen et al., 2006). Another possible cause could be the production and release of exopolysaccharides (EPSs). According to Ruas-Madiedo et al. (2007),LABs synthesize EPSs from glucose, galactose or other monosaccharides by the combined action of different types of glycosyl transferases, and the same effect was demonstrated in bifidobacteria (Audy, Labrie, Roy, & Lapointe, 2009; Ruas-Madiedo et al., 2007; Salazar et al., 2009). Four major consecutive steps of EPS biosynthesis involve sugar transport into the cytoplasm, synthesis of sugar-P, polymerization of repeating Fig. 1. Acidification profiles of skim milk fermentations by pure cultures of S. thermo- unit precursors, and finally EPS export to the outside of the cell philus (●) and B. lactis (■) or a binary co-culture of S. thermophilus with B. lactis (▲), in the absence (dashed lines) or in the presence (solid lines) of 40 mg inulin/g at (Welman & Maddox, 2003). However, it was demonstrated that in 42 °C until reaching pH 4.5. bifidobacteria EPSs can be internally synthesized by β-galactosidase, 24 R.P.S. Oliveira et al. / Food Research International 48 (2012) 21–27

70,0 70,0 St-A St-B 60,0 60,0

50,0 50,0

40,0 40,0

30,0 30,0

Lactose (g/L) 20,0 20,0 Lactose (g/L)

10,0 10,0 Galactose , Lactic acid (g/L) Galactose , Lactic acid (g/L) 0,0 0,0 6,5 6,0 5,5 5,0 4,5 6,5 6,0 5,5 5,0 4,5 pH pH

70,0 Bl-A 70,0 Bl-B 60,0 60,0 50,0 50,0 40,0 40,0 30,0 30,0

Lactose (g/L) 20,0 20,0 Lactose (g/L) Acetic acid (g/L) Acetic acid (g/L) 10,0 Galactose , Lacti cacid 10,0 Galactose , Lacti cacid 0,0 0,0 6,5 6,0 5,5 5,0 4,5 6,5 6,0 5,5 5,0 4,5 pH pH

70,0 StBl-A 70,0 StBl-B 60,0 60,0

50,0 50,0

40,0 40,0

30,0 30,0 Lactose (g/L) 20,0 Lactose (g/L) 20,0 Acetic acid (g/L) Acetic acid (g/L) Galactose , Lacti cacid 10,0 Galactose , Lacti cacid 10,0

0,0 0,0 6,5 6,0 5,5 5,0 4,5 6,5 6,0 5,5 5,0 4,5 pH pH

Fig. 2. Concentration versus pH profiles of skim milk fermentations by pure cultures of S. thermophilus (St) and B. lactis (Bl) or a binary co-culture of S. thermophilus with B. lactis (St– Bl) in the absence (A) or in the presence (B) of inulin. Concentrations of lactose (■), galactose (●), lactic acid (▲) and acetic acid (⁎). which would hence act either as a hydrolase or a glycosyl transferase, Such an indirect inulin metabolization, likely co-responsible for and are then metabolized more quickly that those externally added as the bifidogenic power, appears to be confirmed by the absence of the only carbon source (van den Broek & Voragen, 2008). So, besides detectable levels of formic acid and ethanol (results not shown). In the early-discussed stimulation of the energetic metabolism, inulin fact, in response to ATP or NAD+ cellular requirements associated to could also have potentiated EPS biosynthesis, a portion of which could the low uptake rate of oligosaccharides or polysaccharides like inulin have been released and the other metabolized faster than the same at the end of fermentations, a portion of them should be addressed to inulin alone. the pyruvate-formate lyase-mediated productions of formic acid and

Table 1 Experimental results of milk fermentations by Streptococcus thermophilus (St) and Biﬁdobacterium lactis (Bl) collected at the end of either pure or binary co-cultures (pH=4.5).

Run Biomass (g/L) Lactose Galactose Lactic acid Acetic acid Diacetyl Acetoin

St Bl (g/L) (g/L) (g/L) (g/L) (mg/L) (mg/L)

Without inulin 1 2.7 ±0.2f – 29.9±0.7a 31.4±0.4b 9.6±0.3c – 15.1±0.3e 0.5±0.1a 2 – 1.3 ±0.2a 31.7±0.6b 30.2±0.6a 7.6±0.5a 6.2±0.4c 1.5±0.1a 7.1±0.1c 3 3.1 ±0.1d 1.7 ±0.2b 29.4±0.5a 33.0±0.8c 8.9±0.6b 1.2±0.0a 6.7±0.2c 9.2±0.2e

With inulin 4 3.1 ±0.2d – 32.3±0.5b 33.2±0.8c 10.4±0.4d – 17.1±0.3f 0.7±0.1b 5 – 1.8 ±0.1c 34.0±0.4c 32.6±0.5c 8.6±0.4b 7.7±0.3d 1.8±0.1b 8.1±0.2d 6 4.1 ±0.1g 2.6 ±0.2e 32.6±0.7b 30.4±0.5a 9.3±0.5c 1.4±0.1b 7.6±0.2d 10.2±0.2f

Runs 1 and 4 = St pure culture; Runs 2 and 5 = Bl pure culture; Runs 3 and 6 = St–Bl co-culture. Different letters in the same column mean statistically significant difference among the values of the same parameter, according to the test of Tukey (Pb0.05). ±Means standard deviations with respect to the mean values of quadruplicate runs. R.P.S. Oliveira et al. / Food Research International 48 (2012) 21–27 25 ethanol (van der Meulen et al., 2006) to yield extra ATP (Palframan et products even at very low concentration (up to 5 mg/L) (Oberman & al., 2003; van der Meulen et al., 2006), contrary to the conventional Libudzisz, 1998). catabolic route leading only to acetic and lactic acids. Amounts of diacetyl and acetoin often higher than this level were In general, the lower the acetic acid level, the lower the levels of produced by St (15.2±0.3 mg/L and 0.5±0.1 mg/L), Bl (1.7±0.1 mg/ formic and succinic acids, and the higher that of lactic acid, and vice- L and 7.2±0.1 mg/L) and St–Bl (6.8±0.2 mg/L and 9.1±0.2 mg/L) versa (Doleyres, Fliss, & Lacroix, 2004). Likewise in the present work, during fermentations without inulin (Fig. 3A). It is possible that a Amaretti et al. (2007) reported that formic acid was not produced in a little fraction of diacetyl and acetoin was further metabolized to 2,3- medium containing galactose, lactose, and galactooligosaccharides butanediol that, however, was not measured in this study. Contrary (GOS) fermented by Bifidobacterium adolescentis MB 239, while Østlie to St mono-culture, in that of Bl, and to a less extent also in St–Bl et al. (2003) observed only 8.2 mg/kg ethanol in UHT supplemented co-culture, diacetyl was significantly reduced to acetoin at the end milk with 0.5% (w/v) filter sterilized tryptone after 48 h incubation of fermentation. with B. animalis BB12 at 30 °C. According to Østlie et al. (2003), B. animalis BB12 produced some acetoin (25 mg/kg as an average) instead of acidic end-products to maintain pH homeostasis at high levels of pyruvate, and the same 3.3. Diacetyl and acetoin contents occurred in Lactobacillus plantarum, Lactobacillus johnsonii LA1 and Lactobacillus rhamnosus GG (Tsau, 1992). It has been reported that in Diacetyl and acetoin are important flavor compounds that impart the LABs like S. thermophilus (Axelsson, 1998)andinB. animalis (Helland, typical aroma to many dairy products (Mayo et al., 2010). In particular, Wicklund, & Narvhus, 2004), concentrations of diacetyl and acetoin diacetyl is responsible for the characteristic “buttery” aroma in milk are normally dependent upon the rate of citrate metabolism, or

70,0 20,0 70,0 20,0 60,0 St-B 60,0 St-A 15,0 15,0 50,0 50,0 40,0 40,0 10,0 10,0 30,0 30,0 Lactose (g/L) 20,0 Lactose (g/L) 20,0 5,0 5,0

10,0 Diacetyl, Acetoin (mg/L) 10,0 Diacetyl, Acetoin (mg/L)

0,0 0,0 0,0 0,0 6,5 6,0 5,5 5,0 4,5 6,5 6,0 5,5 5,0 4,5 pH pH

70,0 20,0 70,0 Bl-A 20,0 Bl-B 60,0 60,0 15,0 15,0 50,0 50,0

40,0 40,0 10,0 10,0 30,0 30,0 Lactose (g/L) Lactose (g/L) 20,0 20,0 5,0 5,0 10,0 10,0 Diacetyl, Acetoin (mg/L) Diacetyl, Acetoin (mg/L) 0,0 0,0 0,0 0,0 6,5 6,0 5,5 5,0 4,5 6,5 6,0 5,5 5,0 4,5 pH pH

70,0 20,0 70,0 20,0 StBl-A StBl-B 60,0 60,0 15,0 15,0 50,0 50,0

40,0 40,0 10,0 10,0 30,0 30,0 Lactose (g/L) Lactose (g/L) 20,0 20,0 5,0 5,0 Diacetyl, Acetoin (mg/L) 10,0 Diacetyl, Acetoin (mg/L) 10,0

0,0 0,0 0,0 0,0 6,5 6,0 5,5 5,0 4,5 6,5 6,0 5,5 5,0 4,5 pH pH

Fig. 3. Concentration versus pH profiles of skim milk fermentations by pure cultures of S. thermophilus (St) and B. lactis (Bl) or a binary co-culture of S. thermophilus with B. lactis (St– Bl) in the absence (A) or in the presence (B) of inulin. Concentrations of lactose (■), diacetyl (□) and acetoin (♦). 26 R.P.S. Oliveira et al. / Food Research International 48 (2012) 21–27 alternative pyruvate metabolism, and are produced in significant health properties, in reasonably shorter time compared to the present amounts only if there is surplus pyruvate in the cell with respect to the common practice. need for NAD+ regeneration. The higher amount of acetoin rather than diacetyl detected in this work in Bl and St–Bl, compared to St, may Acknowledgments have been due to the higher NAD+ regeneration requirements of the bifidus metabolism (a sort of more reduced environment) and/or to This research is supported by FAPESP (São Paulo State Research the activity of diacetyl reductase responsible for diacetyl reduction to Foundation) and CAPES (Coordination of Improvement of Higher acetoin (Mayo et al., 2010). Education). So, the formation of such compounds in this work can be justified, at least partially, also by the presence of citrate in significant amounts in milk of many animals, like cows and goats (2 g/L), which can be co- References metabolized with sugars (Cocaign-Bousquet, Garrigues, Loubiere, & — Lindley, 1996). Agência Nacional de Vigilância Sanitária ANVISA. Legislação. VisaLegis. Resolução RDC n.2, de 7 de janeiro de 2002. Aprova o regulamento técnico de substâncias Fig. 3B shows that the addition of inulin resulted in statistically sig- bioativas e probióticos isolados com alegação de propriedades funcional e ou de nificant (Pb0.05) increases in the levels of both diacetyl (by 12–13%) saúde. Diario Oficial da União, Brasília, January 9th and July 17th, 2002. and acetoin (by 12–20%) in all cultures (Table 1). According to Akalın,A.S.,&Unal,G.(2010).Theinfluence of milk supplementation on the microbiological – fl stability and textural characteristics of fermented milk. Milchwissenschaft, 65,291 294. Oliveira et al. (2011), the increased concentrations of these avoring Akın, M. B., Akın, M. S., & Kirmaci, Z. (2007). Effects of inulin and sugar levels on the viability end-products in co-cultures could be ascribed to substantial metabolic of yogurt and probiotic bacteria and the physical and sensory characteristics in probiotic changes induced by the prebiotic, and may be exploited at an industrial ice-cream. Food Chemistry, 104,93–99. Altieri, C., Bevilacqua, A., D'Amato, D., Del Nobile, M. A., & Sinigaglia, M. (2008). Modelling level to increase their presence in fermented dairy products. the survival of starter lactic acid bacteria and Bifidobacterium bifidum in single and simultaneous cultures. Food Microbiology, 25,729–734. Amaretti, A., Bernardi, T., Tamburini, E., Zanoni, S., Lomma, M., Matteuzzi, D., et al. 3.4. Biomass growth (2007). Kinetics and metabolism of Bifidobacterium adolescentis MB 239 growing on glucose, galactose, lactose, and galactooligosaccharides. Applied and Environmental – Without inulin, the St–Bl co-culture showed final concentrations Microbiology, 73,3637 3644. Audy, J., Labrie, S., Roy, D., & Lapointe, G. (2009). Sugar source modulates exopolysaccharide of both microorganisms (3.1±0.1 g/L of St and 1.7±0.2 g/L of Bl) biosynthesis in Bifidobacterium longum subsp. longum CRC 002. Microbiology, 156, statistically significantly higher (Pb0.05) by about 15 and 38% than 653–664. fi in their respective pure cultures (Table 1), thus highlighting a clear syner- Axelsson, L. T. (1998). Lactic acid bacteria: Classi cation and physiology. In S. Salminen, & A. von Wright (Eds.), Lactic acid bacteria (pp. 1–72). New York: Marcel Dekker. gistic effect between these microorganisms due to mutual interactions Chen, Q., Ruan, H., Zhu, D., & He, G. (2010). Isolation and characterization of an oxygen, (Garro, de Valdez, & de Giori, 2004, Oliveira et al., 2011). Wang, Yu, acid and bile resistant Bifidobacterium animalis subsp. lactis Qq 08. Journal of the – Yang, and Chou (2003) reported that maximum growth of starter dairy Science of Food and Agriculture, 90, 1340 1346. Cocaign-Bousquet, M., Garrigues, C., Loubiere, P., & Lindley, N. D. (1996). Physiology of culturesmaybeobtainedmorequicklywhenS. thermophilus or L. pyruvate metabolism in Lactococcus lactis. Antonie Van Leeuwenhoek, 70, 253–267. acidophilus are used in mixed cultures together with bifidobacteria. Cronin, M., Ventura, M., Fitzgerald, G. F., & van Sinderen, D. (2011). Progress in genomics, me- As far as the possible causes of such a synergism are concerned, sev- tabolism and biotechnology of bifidobacteria. International Journal of Food Microbiology, doi:10.1016/j.ijfoodmicro.2011.01.019. eral authors reported that free amino acids (FAA) formed by some LABs' Cruz, A. G., Faria, J. A. F., Walter, E. H. M., Andrade, R. R., Cavalcanti, R. N., Oliveira, C. A. proteolytic activities can stimulate the growth of both lactobacilli and F., et al. (2010). Optimization of the processing of probiotic yoghurt added with bifidobacteria (Donkor et al., 2007). Another possibility is the increase glucose oxidase using the response surface methodology. Journal of Dairy Science, 93, 1058–1069. in viscosity consequent to the release of EPSs, which could have provid- Cruz, A. G., Walter, E. H. M., Cadena, R. S., Faria, J. A. F., Bolini, H. M. A., Pinheiro, H. P., ed better protection to both microorganisms (Donkor et al., 2007; et al. (2010). Survival analysis methodology to predict the shelf-life of probiotic Martínez-Villaluenga, Frías, Gómez, & Vidal-Valverde, 2006). Moreover, flavored yogurt. Food Research International, 43, 1444–1448. fl Altieri, Bevilacqua, D'Amato, Del Nobile, and Sinigaglia (2008) demon- de Oliveira, L. B., & Jurkiewicz, C. H. (2009). In uence of inulin and acacia gum on the viability of probiotic bacteria in symbiotic fermented milk. Brazilian Journal of strated that bifidobacteria are able to reduce oxygen content in the Food Technology, 12, 138–144. medium, thus reducing a stressing element for streptococci; but, at de Vin, F., Rådström, P., Herman, L., & de Vuyst, L. (2005). Molecular and biochemical analysis the same time, they seemed to get a benefit from the consociation of the galactose phenotype of dairy Streptococcus thermophilus strains reveals four different fermentation profiles. Applied and Environmental Microbiology, 71,3659–3667. with S. thermophilus, as proved by higher acidification of the medium Doleyres, Y., Fliss, I., & Lacroix, C. (2004). Continuous production of mixed lactic and lower viability loss. starters containing probiotics using immobilized cell technology. Biotechnology – In the presence of inulin, the concentrations of St (4.1±0.1 g/L) Progress, 20, 145 150. – Donkor, O. N., Nilmini, S. L. I., Stolic, P., Vasiljevic, T., & Shah, N. P. (2007). Survival and and Bl (2.6±0.2 g/L) in the St Bl co-culture were about 35 and 53% activity of selected probiotic organisms in set-type yoghurt during cold storage. higher (Pb0.05) than in their respective pure cultures. These results International Dairy Journal, 17, 657–665. as a whole confirm the synbiotic and bifidogenic effects already men- Ejtahed, H. S., Mohtadi-Nia, J., Homayouni-Rad, A., Niafar, M., Asghari-Jafarabadi, M., fi ı ı Mo d, V., et al. (2011). Effect of probiotic yogurt containing Lactobacillus acidophilus tioned by several authors (Ak n, Ak n, & Kirmaci, 2007; de Oliveira & and Bifidobacterium lactis on lipid profile in individuals with type 2 diabetes mellitus. Jurkiewicz, 2009; Šimunek & Evačić, 2009). Journal of Dairy Science, 94,3288–3294. Fandi, K. G., Ghazali, H. M., Yazid, A. M., & Raha, A. R. (2001). Purification and N-terminal amino acid sequence of fructose-6-phosphate phosphoketolase from 4. Conclusion Bifidobacterium longum BB536. Letters in Applied Microbiology, 32, 235–239. FAO/WHO (2002). Guidelines for the evaluation of probiotics in food, Joint FAO/WHO working group report on drafting guidelines for the evaluation of probiotics in Pure cultures and binary co-cultures of S. thermophilus and Bifido- food. London, ON, Canada, April 30th May 1st. ftp://ftp.fao.org/es/esn/food/ bacterium lactis are investigated for their ability to ferment skim milk wgreport2.pdf Accessed August 4th, 2011 supplemented with inulin or not. The addition of inulin significantly Garro, M. S., de Valdez, G. F., & de Giori, G. S. (2004). Temperature effect on the biological activity of Bifidobacterium longum CRL849 and Lactobacillus fermentum CRL 251 in reduces the time to complete both pure and binary co-cultures, pure and mixed cultures grown in soy milk. Food Microbiology, 21,511–518. enhances biomass growth and increases the levels of lactic and acetic Helland, M. H., Wicklund, T., & Narvhus, J. A. (2004). Growth and metabolism of selected acids. Also, the addition of inulin results in a statistically significant strains of probiotic bacteria in milk- and water-based cereal puddings. International Dairy Journal, 14,957–965. increase in the levels of both diacetyl and acetoin in all pure and co- Hipp, N. J., Groves, M. L., Custer, J. H., & McMeekin, T. L. (1950). Separation of γ-casein. cultures. Journal of the American Chemical Society, 72, 4928–4931. These results as a whole suggest that a S. thermophilus–B. lactis co- Holzapfel, W. H., & Schillinger, U. (2002). Introduction to pre- and probiotics. Food Research International, 35, 109–116. culture enriched with inulin could be satisfactorily employed in the Janer, C., Arigoni, F., Lee, B. H., Pelaez, C., & Requena, T. (2005). Enzymatic ability of dairy industry to obtain probiotic fermented milks with interesting Bifidobacterium animalis subsp. lactis to hydrolyze milk proteins: Identification R.P.S. Oliveira et al. / Food Research International 48 (2012) 21–27 27

and characterization of endopeptidase O. Applied and Environmental Microbiology, Roberfroid, M. B. (2005). Introducing inulin-type fructans. The British Journal of Nutrition, 71, 8460–8465. 93,S13–S25. Liu, S. Q., & Tsao, M. (2009). Enhancement of survival of probiotic and non-probiotic Rodrigues, D., Rocha-Santos, T. A. P., Pereira, C. I., Gomes, A. M., Malcata, F. X., & Freitas, lactic acid bacteria by yeasts in fermented milk under non-refrigerated conditions. A. C. (2011). The potential effect of FOS and inulin upon probiotic bacterium per- International Journal of Food Microbiology, 135,34–38. formance in curdled milk matrices. LWT — Food Science and Technology, 44, Martin, F., Cachon, R., Pernin, K., De Coninck, J., Gevais, P., Guichard, E., et al. (2011). Effect 100–108. of oxidoreduction potential on aroma biosynthesis by lactic bacteria in nonfat yogurt. Ruas-Madiedo, P., Moreno, J. A., Salazar, N., Delgado, S., Mayo, B., Margolles, A., et al. Journal of Dairy Science, 94,614–622. (2007). Screening of exopolysaccharide-producing Lactobacillus and Bifidobacterium Martínez-Villaluenga, C., Frías, J., Gómez, R., & Vidal-Valverde, C. (2006). Influence of strains isolated from the human intestinal microbiota. Applied and Environmental addition of raffinose family oligosaccharides on probiotic survival in fermented Microbiology, 73,4385–4388. milk during refrigerated storage. International Dairy Journal, 16, 768–774. Salazar, N., Ruas-Madiedo, P., Kolida, S., Collins, M., Rastall, R., Gibson, G., et al. (2009). Mattila-Sandholm, T., Myllärinen, P., Crittenden, R., Mogensen, G., Fondén, R., & Saarela, Exopolysaccharides produced by Bifidobacterium longum IPLA E44 and Bifidobacterium M. (2002). Technological challenges for future probiotic foods. International Dairy animalis subsp. lactis IPLA R1 modify the composition and metabolic activity of human Journal, 12, 173–182. faecal microbiota in pH-controlled batch cultures. International Journal of Food Mayo, B., Aleksandrzak-Piekarczyk, T., Fernández, M., Kowalczyk, M., Pablo Álvarez-Martín, Microbiology, 135,260–267. P., & Bardowski, J. (2010). Updates in the metabolism of lactic acid bacteria. In F. Mozzi, Scardovi, V., & Trovatelli, L. D. (1965). The fructose-6-phosphate shunt as a peculiar R. R. Raya, & G. M. Vignolo (Eds.), Biotechnology of lactic acid bacteria: Novel applications pattern of hexose degradation in the genus Bifidobacterium. Analytical Microbiology (pp. 3–33). Iowa, USA: Wiley-Blackwell. Enzymology, 15,19–27. Oberman, H., & Libudzisz, Z. (1998). Fermented milks. In B. J. B. Wood (Ed.), Microbiol- Šimunek, M., & Evačić, S. (2009). Effect of inulin on the growth and survival of Bifidobacterium ogy of fermented foods (pp. 309–350). London: Blakie Academic & Professional, longum BB536 in fermented goat's and cow's milk. Mljekarstvo, 59, 209–216. London, UK. Tsau, J. L. (1992). Conversion of pyruvate to acetoin helps to maintain pH homeostasis Oliveira, R. P. S., Florence, A. C. R., Silva, R. C., Perego, P., Converti, A., Gioielli, L. A., et al. in Lactobacillus plantarum. Applied and Environmental Microbiology, 58, 891–894. (2009). Effect of different prebiotics on the fermentation kinetics, probiotic survival Urashima, T., Kitaoka, M., Asakuma, S., & Messer, M. (2009). Milk oligosaccharides. In and fatty acid profiles in nonfat symbiotic fermented milk. International Journal of P. L. H. McSweeney, & P. F. Fox (Eds.), Advanced dairy chemistry: Lactose, water, Food Microbiology, 128,467–472. salts and minor constituents (pp. 295–340). New York, USA: Springer Science. Oliveira, R. P. S., Perego, P., Converti, A., & Oliveira, M. N. (2009). Effect of inulin on growth van den Broek, L. A. M., & Voragen, A. G. J. (2008). Bifidobacterium glycoside hydrolases and acidification performance of different probiotic bacteria in co-cultures and mixed and (potential) prebiotics. Innovative Food Science and Emerging Technologies, 9, culture with Streptococcus thermophilus. Journal of Food Engineering, 91,133–139. 401–407. Oliveira, R. P. S., Perego, P., Oliveira, M. N., & Converti, A. (2011). Effect of inulin as a van der Meulen, R., Verbrugghe, T. A. K., & de Vuyst, L. (2006). Kinetic analysis of prebiotic to improve growth and counts of a probiotic cocktail in fermented skim bifidobacterial metabolism reveals a minor role for succinic acid in the regeneration milk. LWT — Food Science and Technology, 44, 520–523. of NAD+ through its growth-associated production. Applied and Environmental Østlie, H. M., Helland, M. H., & Narvhus, J. A. (2003). Growth and metabolism of selected Microbiology, 72,5204–5210. strains of probiotic bacteria in milk. International Journal of Food Microbiology, 87, Vasiljevic, T., Kealy, T., & Mishra, V. K. (2007). Effect of beta-glucan addition to a probiotic 17.–27 containing yogurt. Journal of Food Science, 72,C405–C411. Palframan, R. J., Gibson, G. R., & Rastall, R. A. (2003). Carbohydrate preferences of Wang, Y. C., Yu, R. C., Yang, H. Y., & Chou, C. C. (2003). Sugar and acid contents in soymilk Bifidobacterium species isolated from the human gut. Current Issues in Intestinal fermented with lactic acid bacteria alone or simultaneously with bifidobacteria. Food Microbiology, 4,71–75. Microbiology, 20,333–338. Restle, J., Pacheco, P. S., & Moletta, J. L. (2003). Genetic group and postpartum nutritional Welman, A. D., & Maddox, I. S. (2003). Exopolysaccharides from lactic acid bacteria: level on the milk yield and composition of beef cows. Revista Brasileira de Zootecnia, Perspectives and challenges. Trends in Biotechnology, 21, 269–274. 32,585–597.