Role of milk and honey in the tolerance of lactobacilli to oxidative stress

Abstract

In the development of functional probiotic food, the carrier matrices should be carefully selected and optimized to ensure the highest levels of probiotic survival in the symbiotic food along storage. Because milk and honey food matrices are rich in antioxidant substances, the aim of the research was to evaluate their effect in protecting lactobacilli from reactive oxygen species (ROS) generated by the addition of hydrogen peroxide. Viability assays were performed with and without the addition of H₂O₂, in three different matrices: 0.9% peptone saline, 5% honey, or 12% reconstituted skim milk. The milk matrix provided protection for the Lacticaseibacillus paracasei DTA83 and Lacticaseibacillus rhamnosus DTA76. However, this protective effect was not observed in the survival of Lactobacillus acidophilus La 5. Honey solution did not maintain the viability of probiotic microorganisms exposed to hydrogen peroxide and, on the contrary, caused a significant reduction in the population of L. rhamnosus DTA76 (p < 0.001). Lower membrane lipid peroxidation due to H₂O₂ exposure was observed in L. acidophilus La 5 and L. rhamnosus DTA76, but this marker showed no relation with viability. It was concluded: (i) lactobacilli from the Lacticaseibacillus genus were the ones that benefited most from the lactic environment; (ii) the absence of the protective effect of honey was possibly due to the presence of Fe²⁺ which reacts with H₂O₂ to produce hydroxyl radicals; and (iii) cell viability did not correlate with membrane lipid peroxidation, and it is not a good marker to evaluate this type of damage in cells of different microorganisms.

Introduction

Probiotics are live microorganisms that, when administered in adequate amounts, confer a health benefit on the host [1]. The definition was maintained by the International Scientific Association for Probiotics and Prebiotics (ISAPP) in 2014 [2]. Lactobacilli and bifidobacteria species are normal components of the healthy human intestinal microbiota and are the most commonly used as probiotics. Not all strains of lactobacilli and bifidobacteria can be used to obtain probiotic foods, but only those who meet the requirements of being safe and present adequate functional and technological properties such as oxygen sensitivity and storage stability, which is related to their ability to develop as symbiotic cultures [3]. Most probiotic lactobacilli belong to the Lactobacillus acidophilus group and L. casei group, which includes the species L. casei, L. paracasei, and L. rhamnosus [4]. Although species of lactobacilli have been traditionally classified as oxygen-tolerant anaerobes, recent studies have demonstrated that several strains are capable of using oxygen as a substrate in reactions mediated by flavin oxidases to synthesize a minimum respiratory chain [5, 6]. Zotta et al. [5] worked with strains of L. casei and found that most of the 138 strains were able to cope with aerobic conditions and for many strains, aerobic growth and heme or heme/menaquinone supplementation increased biomass production compared to anaerobic cultivation. Only four L. casei strains showed a catalase-like activity under anaerobic, aerobic, and respiratory conditions and were able to survive in the presence of H₂O₂ (1 mM). Indeed, aerobic growth and respiration in the L. casei group and oxygen-tolerant phenotypes may assign technological and physiological advantages that facilitate their use by the industry.

It is, therefore, widely accepted that the growth condition and the type of metabolism significantly affect the stress responses in lactobacilli. Maresca et al. [7] investigated the effect of aerobic (presence of oxygen) and respiratory (presence of oxygen, heme, and menaquinone) cultivation on catalase production and oxidative stress response of Lactobacillus johnsonii/gasseri strains. They concluded that respiratory condition favored their tolerance to oxidative compounds (H₂O₂ and ROS generators).

Although not a free radical (ROS) oxidizing agent, H₂O₂ is extremely deleterious, as it can easily be converted into radicals by participating in reactions producing high reactive hydroxyl radicals. [8, 9] When there is an imbalance between the production of free radicals and/or reactive species and the ability of antioxidants to remove them, oxidation of biomolecules occurs, generating specific metabolites, markers of oxidative stress. Such markers are mainly derived from the oxidation of lipids, proteins, and DNA molecules [10]. Accumulation of ROS can reduce the survival of lactobacillus, in addition to inducing the formation of undesirable compounds in fermented foods [11].

The cellular response process to oxidative stress stimulates antioxidant enzymes or phase II metabolizing enzymes that detoxify xenobiotics by increasing their hydrophilicity, such as superoxide dismutase (SOD), catalase, heme oxygenase-1 (HO-1), and glutathione S-transferases (GSTs), to protect against the damage due to oxidative stress. [12]

Antioxidant substances present in food matrices can aid to protect the cells from oxidative stress in a number of ways. According to Pereira et al. [13], phenolic hydroxyl groups are good hydrogen donors: hydrogen-donating antioxidants can react with reactive oxygen and reactive nitrogen species in a termination reaction, which breaks the cycle of generation of new radicals. The antioxidant effect of phenolic compounds is due to their role as chelators of prooxidant metal ions. However, phenolics can act as prooxidants by chelating metals in a manner that keeps or increases their catalytic activity or by reducing metals, thus favoring the formation of free radicals [14].

Bee honey has aroused the interest of the scientific community as a natural dietary antioxidant whose components responsible for redox properties are probably flavonoids, phenolic acids, enzymes, vitamins, and minerals such as copper and iron [15]. Interestingly, bioactive proteins present in milk such as glutathione peroxidase and peptides from whey and casein protein fractions [16] are believed to be capable of aiding the cell’s endogenous defenses in the elimination of reactive oxygen species (ROS) produced by cellular metabolism, or even through the permeation of oxygen through the packaging, which are detrimental to cellular structures and functions.

Considering the challenge of the food industry in the elaboration of products with viable probiotic cells in adequate concentrations up to the moment of consumption, it is necessary to know about the behavior of such microorganisms against oxidizing agents and in the presence of potentially protective matrices. Such knowledge will provide more accurate information for the selection of probiotic strains and food matrices for the formulation of functional foods.

The aim of the research was to evaluate the effect of milk and honey in protecting lactobacilli from reactive oxygen species (ROS) generated by the addition of H₂O₂. In addition, the effect of this oxidant on membrane lipid peroxidation was investigated in order to identify ROS targets.

Materials and methods

Food matrices

Apis mellifera honey from “assa-peixe” (Vernonia sp.) flowers, of dark amber color, was harvested in the region of Teresópolis, RJ, Brazil, in February 2017. The honey was diluted to 5% (wt/vol) in sterile water at the time of use. Skimmed milk powder (Molico, Nestlé) was reconstituted at 12% (wt/vol) with distilled water and autoclaved at 0.5 ATM for 10 min.

Lactobacillus strains

Lacticaseibacillus paracasei DTA83 and Lacticaseibacillus rhamnosus DTA76 were isolated on LAMVAB agar from newborn’s stools aged up to 21 days and identified by sequencing of the 16S rDNA region as described by Andrighetto et al. [17] and are part of the culture collection of the Laboratory of Microbiology, Department of Food Technology, Federal Rural University of Rio de Janeiro. The safety and probiotic properties of the isolates were previously reported [18]. L. acidophilus La 5 was obtained from Christian Hansen®. The cultures were kept frozen at − 20 °C in MRS broth (Merck), with 20% glycerol as a cryoprotective agent [19]. The working cultures were prepared by activation of the frozen stock cultures through three successive overnight growth in MRS broth (HiMedia) with 0.05% cysteine (Vetec) and aerobic incubation (static growth) at 36 °C. L. acidophilus, however, was incubated at 36 °C in an anaerobic atmosphere enriched with CO₂ obtained by reacting of calcium carbonate with sulfuric acid solution, in addition to burning residual O₂ with a lit candle).

Hydrogen peroxide production by lactobacilli

The capability of H₂O₂ production was evaluated in two matrices, milk and 5% honey solution.

Working cultures were initially inoculated in milk or honey solution and incubated overnight at 36 °C. Next, 2% aliquots were transferred to tubes containing 20 mL of the same matrices and incubated again at 36 °C for 24 h. H₂O₂ production was determined using semiquantitative test strips – Merckoquant Peroxide Test (MERCK, Germany). Strips of the test were dipped into each tube, and different tones of blue were visually compared with a scale provided by the manufacturer.

Survival after exposure to hydrogen peroxide

For this purpose, lactobacilli cultures were subcultured in MRS broth three times as described in the “Lactobacillus strains” section Then the tubes from the third culture were centrifuged at 504g for 3 min, the supernatant discarded, pellet added with peptone saline solution, and centrifugation repeated twice to wash residues from the MRS broth.

Then, an inoculum of each culture was added to tubes with the matrices (milk and honey) and peptone saline solution in order to contain 10⁶ CFU/ml. After 2 h of contact, the volume of each tube was divided into equal amounts (one added with 20 mM H₂O₂ and the other untreated that served as a control). Both were statically kept for 1 h. Aliquots of decimal dilutions were then surface plated on MRS agar. Plates were prepared in duplicate and colonies counted after 48 h of incubation at 36 °C.

Cell survival was determined by the difference between counts obtained from the untreated control samples from the corresponding stressed with H₂O₂. The entire experiment was repeated four times on different occasions.

Determination of iron content in honey

Three grams of the sample was weighed and taken to the heating plate for partial elimination of the organic matter content, until the absence of smoke detachment. To obtain ash, the sample was incinerated in a muffle oven at 550 °C/12 h. After cooling, the ash was dissolved with HNO₃ solution (25%) and transferred with distilled water to a 100 mL volumetric flask. The reading was performed in a photometer (HANNA Instruments HI 83200 multiparameter).

Membrane lipid peroxidation

Tolerance to oxidative stress was performed using a modification of the biological method described by Steels, Learmonth, and Watson [20]. This method measures the response of Saccharomyces cerevisiae cells to oxidative stress based on increased levels of thiobarbituric acid reactive substances (TBARS), which is indicative of the formation of malondialdehyde, a lipid peroxidation product.

For this, a cell suspension obtained as described in the “Survival after exposure to hydrogen peroxide” section was prepared. Suspension volumes containing 50 mg/mL of dry cell mass were centrifuged and washed twice with sterile distilled water. Two tubes were prepared for each culture. In one of the tubes, the pellet was reconstituted with the same volume of oxidant H₂O₂ at a final concentration of 20 mM, and to the other, untreated, the same volume of distilled water, served as control. Both remained at rest for 60 min.

Malondialdehyde was then quantified by adding 500 μL of 10% TCA (trichloroacetic acid) to the cell suspension. The content was transferred to thick-walled tubes containing 1.5 g of 40-μm glass beads to be vortex-agitated (Gehaka® Mark) for 20 s totaling 6 cycles to promote cell wall lysis. The extracts were collected in microtubes, and the glass beads were washed with 500 μl of 10% TCA (Vetec). This washing TCA was added to the same microtube. After lysis, the extracts were centrifuged at 1433.6g. To the collected supernatant, the reagents for the TBARS assay (EDTA, thiobarbituric acid) were added. The assay was performed with the reaction mixture containing 150 μL of extract, 150 μL of H₂O, 100 μL of 0.1 M EDTA (Vetec), and 600 μL of 1% thiobarbituric acid (Merck) in 0.05 M NaOH. A reaction blank containing 0.3 mL of H2O without the cell extract was also prepared. The reaction mixture was incubated at 100 °C for 15 min. The tubes were cooled and the final product of the lipid oxidation (malondialdehyde - MDA) was measured spectrophotometrically at 532 nm.

Statistical analysis

The results were presented as the mean and standard error of at least three independent experiments. The data were analyzed by one-way or two-way analysis of variance (ANOVA). When only a single experimental variable was evaluated, the data were analyzed through one-way ANOVA, followed by the Tukey test. In the cases in which two experimental variables were evaluated, two-way ANOVA and, when necessary, multiple comparisons were made through the Bonferroni test. Differences were considered statistically significant when p < 0.05. Data was computed using Graphpad Prism5 software (La Jolla, California, USA).

Results and discussion

Hydrogen peroxide production by lactobacilli and iron content in honey

Under aerobic static growth conditions, all strains of lactobacilli produced hydrogen peroxide when growing in honey, but L. rhamnosus strains also did when growing in milk (Table 1). L. acidophilus La 5 and L. rhamnosus DTA76 produced higher amounts of peroxide. However, L. acidophilus La 5 and L. paracasei DTA83 did not produce H₂O₂ in milk under the conditions of the experiment. Lactic acid bacteria (LAB) are generally sensitive to H₂O₂, a compound that they can paradoxically produce. Accumulation of ROS can result in irreparable harmful changes to proteins, lipids, and DNA. On the other hand, it is also clear that ROS production serves an essential signaling function within and between cells within tissues [21]. According to Hertzberger et al. [22], accumulation of H₂O₂ mainly occurs in species that lack the main hydrogen peroxide-scavenging enzymes, such as catalase and NADH peroxidase. Species of the L. acidophilus group that accumulate H₂O₂ or other reactive oxygen species (ROS) upon exposure to molecular oxygen typically have an energy metabolism adapted to anaerobic environments. In this study, L. acidophilus was the strain that produced the largest amount of H₂O₂ in honey solution.

Table 1.

Production of H₂O₂ by lactobacilli strains when grown either in milk or honey 5% solution

Lactobacilli	Growing in milk H2O2 (mg/l)	Growing in 5% honey H2O2 (mg/l)
L. paracasei DTA83	Negative	0.5
L. acidophilus La 5 (C Hansen)	Negative	5.0
L. rhamnosus DTA76	0.5	2.0

*Detection limit – 0.5 mg/l H₂O₂

Ricciardi et al. [23] studied the effect of the atmosphere of incubation (not aerated static growth vs aerated shaken growth) and supplementation with Fe²⁺, hemin, Mn²⁺, or their combinations in gene expression and activity of heme-dependent catalases in L. casei strains. They found that the tolerance of oxidative stress was higher in aerated cultures supplemented with hemin and/or Mn, because of high catalase activities. Rochat et al. [24] suggested that associations of. L. casei producing pseudocatalase enzyme-denominated MnKat and L. bulgaricus, L. casei could eliminate H₂O₂ from the culture medium, thereby protecting both L. casei and L. bulgaricus from its deleterious effects. The ability to degrade H₂O₂ by strains of the L. casei group, such as the strains of L. paracasei DTA83 and L. rhamnosus DTA76 used in this study, is likely to be the reason for the lower amount of H₂O₂ detected compared to the L. acidophilus La 5 strain.

The honey used in this study was dark amber in color and contained 1.32 mg/100 g of iron. This value is more than four times higher than that normally found in light-colored honey (Table 2). The data shown correlates to the findings of other authors. The average mineral contents for two types honey classified according to color as light and dark were 0.24 and 0.94 mg/100 g respectively [25, 26].

Table 2.

Iron content in samples of honey of different colors

Coloring of the honey sample	Iron content (mg/100 g)
*Dark amber	1.32
Dark amber	1.13
*Light-colored	0.42
Light-colored	0.34
Light-colored	0.28

*Samples that were used in the experiments

The honey solution proved to be a more suitable matrix for the production of H₂O₂ by all strains of lactobacilli which are possibly related to the presence of Fe²⁺ in the composition of honey. Hategekimana et al. [27] showed that the minerals Ca²⁺ and Mg²⁺ were involved in cream formation in honey, while Fe²⁺ participated in changing color to tar black at pH 4.12 and above. Valdés-Silverio et al. [28] observed a reduction of Fe³⁺ to Fe²⁺ in eucalyptus honey and consequent enhancement of the antibacterial activity which is probably due to free radicals generated by the Fenton reaction [29]. Fenton reaction can generate highly reactive hydroxyl radicals (_●OH) from reactions between Fe²⁺ catalyst and H₂O₂ at acidic or even circumneutral pH. The reaction starts when Fe²⁺ is oxidized by hydrogen peroxide to Fe³⁺, leading to the cascading generation of free radicals.

Survival after exposure to H₂O₂ in different matrices

The results of the tolerance of L. paracasei against the oxidant H₂O₂ and the influence of the honey and milk matrices on its survival are shown in Fig. 1. It was observed a significant reduction in viability (p < 0.001) when cells were exposed to H₂O₂ in the presence of saline and honey solutions. However, when has been treated in the presence of milk, cells maintained viability, and there was no significant difference in the cultures with and without the oxidant, indicating a protective effect of this matrix.

Fig. 1.

Survival of L. paracasei DTA83 subjected to oxidative stress (20 mM H₂O₂) in the presence of peptone saline, honey, or milk. *** represents p < 0.001 and ###p > 0.001

Figure 2 shows the survival number of L. rhamnosus cells treated with the oxidant H₂O₂ in the presence of honey and milk. L. rhamnosus population had a significantly higher reduction in viability when the cells were exposed to hydrogen peroxide in the presence of honey (p < 0.001) than in saline (p < 0.05). In the presence of milk, in the same way as with L. paracasei, this reduction was not significant (p > 0.05) meaning that there was a protective effect of this matrix.

Fig. 2.

Survival of L. rhamnosus DTA76 subjected to oxidative stress (20 mM H₂O₂) in the presence of peptone saline, honey, or milk. *** represents p < 0.001, *p < 0.05, and #p > 0.05

Probiotic bacteria are most frequently included in the composition of dairy products. In fact, the protective effect of milk on maintaining the viability of the L. paracasei and L. rhamnosus cultures corroborates with other reported findings. Lee et al. [16] investigated the potential benefits of consumption of probiotics through dairy foods compared to other food matrices or supplements. Higher survival was observed during passage through the gastrointestinal tract, moreover L. casei BL23 protected against the development of colitis when ingested in milk but not in a nutrient-free buffer, simulating its consumption as a nutritional supplement. This study strengthens the importance of the carrier food matrix choice of the probiotic bacteria to optimize and ensure the highest levels of probiotic survival and influence the effector phase of immune inflammations in the digestive tract.

Viability data of L. acidophilus treated with hydrogen peroxide in the presence of the honey and milk matrices on its survival are shown in Fig. 3. Unlike the bacteria L. paracasei and L. rhamnosus, which maintained viability in the presence of milk, there was no significant difference in the survival of L. acidophilus treated with hydrogen peroxide in the different matrices.

Fig. 3.

Survival of L. acidophilus La 5 subjected to oxidative stress (20 mM H₂O₂) in the presence of peptone saline, honey, and milk. * represents p < 0.05

Out of L. acidophilus, the protector effect of milk was observed with the other lactobacilli. According to Shamala et al. [30] and Stanton [31], bifidobacteria do not adapt well to fermented milk and suffer from the presence of oxygen. As with bifidobacteria, L. acidophilus activity is higher at low oxygen concentrations, because it is microaerophilic or anaerobic and a strict fermenter.

The degradation of H₂O₂ can be catalyzed by catalase and glutathione peroxidase, enzymes present in milk, which can also degrade lipid peroxides. Lactoferrin may play an important role in prooxidant Fe²⁺ binding, but most of these enzymes are inactivated by pasteurization temperatures [32].

On the other hand, it is known that milk casein may also have antioxidant activity. In fact, all the subunits of casein (alpha-casein, beta-casein, and kappa-casein) appear to favor iron auto-oxidation and thus inhibit lipid peroxidation. The mechanisms of antioxidant action are complex, but the strongest effect is achieved by modifying the Fe²⁺/Fe³⁺ equilibrium [33]. The effect of milk alpha-casein on the antioxidant activity of polyphenols present in tea was studied by Bourassa et al. [34] using three complementary oxidation methods: ABTS (+) radical cation elimination, cyclic voltammetry, and inhibition of lipid peroxidation. Using the ABTS (+) assays, the antioxidant activity of all polyphenols was lowered by 11–27% in the presence of caseins. Through cyclic voltammetry, the overall current measured at the electrode was decreased by the presence of the protein, from 21 to 61%. However, using the lipid peroxidation method, the antioxidant activity of all polyphenols changed (from 6 to 75%) after the addition of alpha-casein.

Martin et al. [35] reported a strain of L. jensenii that belongs to the L. acidophilus group and is a strong producer of H₂O₂ that protects itself from the bactericidal effects of this compound through the activity of extracellular peroxidases which is dependent on the presence of Fe3+ of heme source. They conclude that the presence of Fe³⁺ in the growth media, a condition that mimics better the vaginal environment, improves the probiotic characteristics of L. jensenii CECT 4306. Hertzberger et al. [22] identified flavin reductase in L. johnsonii that serves a metabolic purpose in which H₂O₂ is a side product, suggesting that the production of H₂O₂ in itself has a biological function. Martin and Suárez [36] found that hydrogen peroxide starts to be produced by L. jensenii as a consequence of the aeration of the cultures, presumably because the static liquid cultures are virtually anaerobic. Cultures with Fe³⁺, hemin, and hemoglobin did not accumulate H₂O₂. Fe³⁺ activated an extracellular peroxidase that destroyed the H₂O₂ produced by the cultures. On the other hand, in the presence of micromolar amounts of Fe²⁺, Lactobacillus delbrueckii ssp. bulgaricus was able to bind nanomolar quantities of iron. This binding was the result of the oxidation of the Fe²⁺ by H₂O₂ produced by the microorganism and subsequent surface binding of the resulting Fe (III), most likely in the form of Fe (OH)₃.

Differently, we found that the iron content present in honey were toxic to probiotic Lactobacillus from the L. casei group, suggesting that iron present in the honey samples had been reduced from Fe³⁺ to Fe²⁺.

Although rich in phenolic compounds with antioxidant capacity, the honey matrix did not protect the microorganisms and even potentiated the lethal peroxide action for L. rhamnosus. According to Juven and Piearson [29], the ways in which H₂O₂ can be converted to hydroxyl radicals are (i) by transition metal ions (e.g., iron, copper) which have been implicated in the formation of hydroxyl radicals starting from H₂O₂; (ii) by superoxide ions interacting with H₂O₂ in the presence of Fe²⁺ to produce hydroxyl radicals, the iron-catalyzed Haber-Weiss reaction, or the superoxide-driven Fenton mechanism; and (iii) by UV irradiation.

Thus, a possible mechanism to honey toxicity to L. paracasei and L. rhamnosus treated with H₂O₂ is the “Fenton reaction,” where hydrogen peroxide reacts with metal ions such as Fe²⁺ to produce hydroxyl radicals. When combining with an iron atom, receiving an electron, the hydrogen peroxide becomes the most reactive and deleterious of the radicals, responsible for damage to DNA, proteins, and membranes [37]. Trace amounts of metals like Fe and Cu act as promoters of lipid oxidation in the presence of hydroperoxides [38].

A research done by Gül and Pehlivan [39] using ferric reducing/antioxidant power assay (FRAP) provides a direct estimation of reductants in a sample and is based on the ability of the analyte to reduce the Fe³⁺/Fe²⁺couple. According to the results of this study, the darker honey samples, like carob, showed the highest ferric ion reduction capacity, as compared to acacia honey lighter in color.

The effect of light-colored honey was compared to dark-colored honey on the survival of L. paracasei and L. rhamnosus subjected to oxidative stress (20 mM H₂O₂). Light-colored honey did not cause a significant reduction in the population of L. paracasei (p > 0.05), in contrast to exposure to peroxide in the presence of amber honey (p < 0.001). There was a significant reduction in the population of L. rhamnosus in both types of honey, light and amber, when compared to the control (without peroxide). However, a greater reduction was observed in the presence of dark amber honey (p < 0.001), that is, precisely in honey with a higher iron content (1.32 mg/100 g), which would prove the hypothesis that the Fenton reaction would be contributing to less survival of this microorganism (Figs. 4 and 5).

Fig. 4.

Survival of L. paracasei DTA83 subjected to oxidative stress (20 mM H₂O₂) in the presence of dark amber honey (1.32 mg/100 g) and light-colored honey (0.42 mg/100 g). *** represents p < 0.001

Fig. 5.

Survival of L. rhamnosus DTA76 subjected to oxidative stress (20 mM H₂O₂) in the presence of dark amber honey (1.32 mg/100 g) and light-colored honey (0.42 mg/100 g). *** represents p < 0.001 and * p < 0.05

The interest in the possible health benefits of flavonoids has increased because of its antioxidant and free radical sequestering activity observed in vitro. However, the antioxidant effectiveness of flavonoids in vivo is less documented, and because of their prooxidant properties, they are capable of causing oxidative damage by reacting with various biomolecules such as lipids, proteins, and DNA. To what extent flavonoids are capable of acting as an antioxidant or prooxidants in vivo is still poorly understood, and this topic clearly requires further studies [40].

In fact, it has been reported that the addition of honey to fermented dairy beverages promotes an increase in the viability of probiotic microorganisms in these products [41]. The results presented herein were of honey and milk separately and, although apparently contradictory, they were obtained under different conditions. When honey is mixed with milk, the prooxidant effect of its components may be nullified by the milk’s bioactive peptides. Macedo et al. [42] reported that adding 3% Apis mellifera bee honey to L. casei–fermented milk did not interfere in their viability (p > 0.05), but improved viability of L. acidophilus and especially of Bifidobacterium. In addition, the apparent divergences found in the literature regarding the protective effect of this matrix may also be related to the great variety of compounds found in honey. The oligosaccharides and polyphenols of honey vary according to their floral origin, and therefore, the prebiotic effect of different kinds of honey may differ [43].

Membrane lipid peroxidation

L. rhamnosus and L. acidophilus presented similar MDA production in both situations, with no difference in the MDA levels of H₂O₂-treated cells of the untreated controls (Fig. 6). The exposure to the oxidant did not interfere in the membrane lipid degradation of these two microorganisms, which were considered more resistant to the membrane oxidative stress than L. paracasei whose MDA production was significantly higher in cells treated with H₂O₂ (p < 0.001) (Fig. 6). However, L. paracasei was the microorganism that presented higher levels of MDA naturally in the cells, independent of being stressed. In contrast, L. acidophilus and L. rhamnosus presented the lowest MDA content under normal conditions and when treated with H₂O₂. The culture that suffered less degradation in its cell membrane in the presence of hydrogen peroxide was L. acidophilus (p < 0.001), followed by L. rhamnosus (p < 0.01) and L. paracasei (p < 0.05).

Fig. 6.

Lipid peroxidation in L. paracasei DTA83, L. rhamnosus DTA76, and L. acidophilus La 5 assessed by malonaldehyde (MDA) accumulated after treatment with H₂O₂ (20 mM) for 1 h. *** represents p < 0.001

Regarding survival after treatment with H₂O₂ in saline, it was similar for L. paracasei DTA83 and L. rhamnosus. DTA76; L. acidophilus La 5, on the other hand, was the strain that showed the greatest survival and low damage to the membrane, as well as L. rhamnosus (Fig. 7).

Fig. 7.

Log reduction in the number of L. paracasei DTA83, L. rhamnosus DTA76, and L. acidophilus La 5 subjected to oxidative stress (20 mM H₂O₂) in the presence of peptone saline

The higher resistance to lipoperoxidation observed in this study with L. acidophilus and L. rhamnosus indicates that these microorganisms have some antioxidant defense system, which makes them better adapted to the auto-protection against the oxidative damages generated by hydrogen peroxide.

The ability of the lactobacilli to produce hydrogen peroxide and to tolerate high concentrations of this compound may be related to the ability of NADH oxidase and NADH peroxidase synthesis. Many bacteria of this genus synthesize NADH peroxidase, an important enzyme for the detoxification of hydrogen peroxide, and NADH oxidase that consumes the oxygen of the environment. In probiotic bacteria, NADH oxidase activity results in the production of hydrogen peroxide, whereas NADH peroxidase consumes hydrogen peroxide, preventing oxidative stress and consequently cell death [44]. Thus, the activity of NOX/NPR system contributes to the maintenance of the NADH/NAD+ balance by promoting cofactor regeneration [5].

According to Zotta et al. [5], in addition to enzymes directly involved in H₂O₂ production and degradation of reactive oxygen species (ROS), other mechanisms may be involved such as the presence of sequences encoding for thioredoxin (trxA)-thiouronoxin reductase (trxB) and thiol/disulfide leader intracellular balance. Serata et al. [45] recently investigated the role of trxA-trxB on the growth and survival of L. casei Shirota under aerobic conditions, demonstrating its implication in oxygen and H₂O₂ tolerance, and this aspect needs further studies. Researchers have identified the dpr gene responsible for coding a peroxide resistance protein (Dpr), which can prevent the buildup of ROS, reducing Fenton reaction due to the ability to bind iron [46].

Thiols are one important class of compounds engaged in stress protection with emphasis on glutathione. New knowledge about the synthesis and metabolism of glutathione in LAB and Gram-positive bacteria in general suggests that the biosynthesis of this enzyme involves the formation of a peptide bond between glutamate and cysteine catalyzed by γ-glutamylcysteine synthetase (GshA) and subsequent formation of a peptide bond between γ-Glu-Cys and glycine catalyzed by glutathione synthetase (GshB). Alternatively, some Gram-positive bacteria developed a single multi-domain fusion protein (GshF), which catalyzes both synthesis reactions. The glutathione system (c-glutamylcysteine synthetase, GshA; glutathione synthetase, GshB; glutathione reductase, Gro; glutathione peroxidase, Gpo) may directly detoxify H₂O₂, prevent lipid peroxidation, and maintain cellular redox balance [47].

In addition, the lipid composition of its membranes is possibly different; L. acidophilus La 5 possesses a molecular mechanism for remodeling and optimizing the fatty acid composition [48]. The type of lipid present in the cytoplasmic membrane also influences the peroxidation. The oxidative stability of lipids is influenced by the number and nature of the unsaturations present, the type of interface between lipids and oxygen (continuous lipid phase, dispersed or emulsified), exposure to light and heat, oxidants (e.g., transition metal ions), or antioxidants [49]. MDA is formed solely from fatty acids, having at least three double bonds. Sugars, particularly sucrose and glucose, interfere with a strong synergistic effect on the formation of TBARS, thereby overestimating the extent of oxidation. On the other hand, MDA may not react with TBA due to its complexation with proteins, amines, and other compounds [50].

TBARS assays measure the malondialdehyde (MDA) present in the sample as well as the malondialdehyde generated from lipid hydroperoxides by the hydrolytic reaction conditions. MDA is one of several low molecular weight end products formed through the decomposition of certain primary and secondary lipid peroxidation products. However, only a few products of lipid peroxidation generate MDA, and MDA is not the only end product formed from fatty peroxide decomposition, nor a substance generated exclusively through lipid peroxidation [51]. Thus, it is possible that it is not a good marker to evaluate this type of damage in cells of different microorganisms.

Conclusions

Milk matrix protected L. paracasei and L. rhamnosus from the action of peroxide, while honey did not maintain viability of any microorganism and even increased the deleterious effect in L. rhamnosus, which may be possibly associated with the Fe²⁺ present in the matrix reacting with H₂O₂ through the Fenton reaction. Cell viability did not correlate with lipid peroxidation. It is concluded that probiotics have different mechanisms to avoid the toxic effects of the reactive radicals caused by exogenously added H₂O_2.