*3.3. Microbiome Analysis*

A total of 86 samples were included in the microbiome analysis (44 samples from participants in the probiotic group before test #1 (*n* = 22) and at 6 weeks before test #2 (*n* = 22), and 42 samples from participants in the placebo group before test #1 (n = 21) and at 6 weeks before test #2 (*n* = 21)). The local contributions to beta diversity (LCBD) at family and genus levels from taxonomic identification of the samples sequenced is shown in Figure 2.

Bacterial composition of samples was grouped, and both groups (placebo and probiotic) were compared at baseline and at 6 weeks at the end of the study. Richness, Simpson diversity index, and Shannon diversity index did not change after probiotic consumption (ANOVA test, *p* > 0.05 for all comparisons) (Figure 3).

Differences in bacterial population were measured with a Wald test using DESeq2 analysis. After 6 weeks of ingestion of the probiotic product or placebo (end of study), families Rhodospirillaceae (placebo vs. probiotic, log2 fold = 2.71, adjusted *p* value = 0.019) and Streptococcaceae (placebo vs. probiotic, log2 fold = 2.20, adjusted *p* value = 0.019) showed lower values in the probiotic group (Figure 4, left panel), considering a minimum threshold value of 10 counts (total average). There were statistically significant changes in seven genera, *Rhodospirillum* and *Streptococcus* being higher in the placebo group (Figure 4, right panel). However, within-group differences in the probiotic group showed an increase in specific genera, *Methanobrevibacter* (*M. smithii*), *Holdemanella* (*H. biformis*), and *Blautia* being the most remarkable, although *Lactobacillus* and *Lachnospira* decreased at the end of the study. Within-group differences in the placebo group revealed increases in *Bifidobacterium* and *Blautia*, among others, and decreases in *Shigella* and *Klebsiella* (in this case with low mean at baseline). Detailed data are shown in the Supplementary Materials, with Table S1

showing sequence distribution as well as sample metadata; Table S2 includes microbiome profiles at phylum, family, genus, and species levels, and Table S3 summarizes different populations at the genus level by Deseq2 analysis.

**Figure 2.** Local contributions to beta diversity (LCBD) analysis at family level (**right**) and genus level (**left**) from taxonomic identification of the samples sequenced (42 samples in the placebo group and 44 samples in the probiotic group; T1: before test #1, T2: at 6 weeks before test #2).

**Figure 3.** Richness, Simpson diversity index, and Shannon diversity index (from **left** to **right**) in the placebo and probiotic group at baseline (T1) and at 6 weeks (end of study) (T2).

**Figure 4.** Differences between the placebo and probiotic groups at the end of the study (6 weeks) at the level of families (**A**) and genera (**B**).

The probiotic product was well tolerated, and no adverse effects were observed. Additionally, laboratory tests at the end of the study did not show any abnormalities.

#### **4. Discussion**

In an oxidative stress model of high-intensity and duration physical exercise in male cyclists, daily intake of a probiotic product based on a mixture of *B. longum*, *L. casei* and *L. rhamnosus* for 6 weeks was associated with a significant reduction of lipid-related oxidative stress biomarkers, such as serum MDA, serum Ox-LDL, and DNA-related oxidative stress biomarker, such as urinary 8-OHdG. Several studies have shown that high-intensity and duration physical exercise results in oxidative stress, due to ROS being generated excessively by enhanced oxygen consumption, as well as in changes in muscle antioxidant enzyme activity [24–27]. Additionally, physical exercise models in endurance-trained competitive and non-competitive athletes have been used to assess the benefits of different supplements with antioxidant capacity [28–32].

Probiotic supplements are nutraceuticals with wide applications in different aspects of human health and have recently gained increasing interest for their potential effects as antioxidants due to anti-oxidative enzyme upregulation, stimulation of the production of a variety of bioactive peptides, and gut flora re-establishment [33]. However, there is limited evidence of the influence of probiotic supplementation on oxidative markers in athletes, and as far as we are aware there are only four studies examining antioxidant potential of probiotics in athletes. In a randomized double-blind, placebo-controlled study, 22 elite athletes received *Lactobacillus helveticus* (*n* = 10) or placebo (*n* = 12) for 3 months, and it a significant decrease of MDA and advanced oxidation protein products (AOPP) was found, without modifications in antioxidant enzyme SOD activity [34]. In a comparative study of two groups of 12 athletes each, probiotic supplementation with a combination of *Lactobacillus rhamnosus* IMC 501 and *Lactobacillus paracasei* IMC 502 administered for 4 weeks vs. no supplementation (controls) was associated with an increase in plasma antioxidant levels, thus neutralizing ROS [35]. A randomized, double-blinded, placebo controlled trial conducted in 23 trained men who received multi-species probiotics (*n* = 11) or placebo (*n* = 12) over 14 weeks, was designed to evaluate changes of markers of intestinal barrier, oxidation, and inflammation associated with the use of probiotic supplementation at rest and after intense exercise [36]. Participants performed a 90-min intense cycle ergometry at baseline and after 14 weeks. In this study, supplementation had no effect on protein carbonyl and MDA but decreased zonulin in feces as a marker, indicating enhanced gut permeability [36]. Finally, in a study of marathon runners, *Lactobacillus rhamnosus* GG (probiotic group) or placebo drink (placebo group) were given during the 3-month training period, 6-day preparation period, and marathon run, but probiotics did not show any effect on serum total antioxidant potential Ox-LDL [37]. However, studies requiring larger samples of athletes are needed to assess the beneficial role of probiotic supplementation on markers of oxidative stress damage.

On the other hand, other studies have examined the association between gut microbiota and oxidative stress in diseases in which oxidative stress plays a well-known pathogenetic role, such as type 2 diabetes mellitus. In a systematic review and metaanalysis of 13 randomized clinical trials involving 840 subjects, probiotics intake resulted in significant improvement in serum levels of total antioxidant status, MDA, and total glutathione (GSH), but there was a modest effect on serum glucose levels and glycated hemoglobin (HbA1c) [38]. Wang et al. [15] reported an in-depth review of the antioxidant mechanisms of probiotics, summarizing their involvement in decreasing radical generation and improving the antioxidant system based on modulation of the redox status of the host via their metal ion chelating ability, regulation of signaling pathways, antioxidant systems, ROS-producing enzymes, and gut microbiota.

A diversity of exogenous and endogenous stimuli are involved in complex molecular and cellular changes, including oxidative DNA damage and participation in cancer development [39], and different studies have explored the potential of probiotics (*L. casei* and *L. rhamnosus*) as cell-free supernatants to inhibit colon cancer cell invasion [40], the antiproliferative and apoptotic effects driven by *L. casei* ATCC 393 against experimental colon cancer [41], or *Lactobacilli* strains as modulators of *Fiaf* gene expression in human epithelial intestinal cells [42].

8-hydroxy-2- deoxiguanosine (8OHdG) is usually measured as an index of oxidative DNA damage [43,44] with oxidative modification of DNA that causes mutations during replication [45]. In recent years, there has been an increasing interest in the impact of exercise on epigenetic events; in particular, ROS-mediated methylation patterns are being investigated. The understanding of the mechanisms leading to ROS-associated epigenetic modifications may contribute to a better knowledge of carcinogenesis and its progression, together with discovering of implicated biomarkers [46,47].

An interesting aspect of the present study was the assessment of changes in microbiome besides improvement of biomarkers of oxidative damage induced by a model of high-intensity and duration physical exercise in response to supplementation with the probiotic product. The microbiota can be considered as a true endocrine organ, and the interactions between exercise and its adaptations, probiotics, and the microbiota itself could help athletes by producing beneficial metabolic, antioxidant, or anti-inflammatory effects that improve training. *Methanobrevibacter*, *Holdemanella*, and *Blautia* increased in

participants consuming probiotics, whereas *Lactobacillus* and *Lachnospira* were within the taxa that decreased at the final point. *M. smithii* is a prominent microbe with methanogenic properties. In a humanized gnotobiotic mouse model of host–archaeal–bacterial mutualism, it was shown that *M. smithii* removed H2, which was related with more effective bacterial fermentation and subsequently more efficient short-chain fatty acids (SCFAs) production, increasing energy absorption [48,49]. *Holdemanella* is considered a butyrate producer. In a study of fecal microbiota collected from obese adults aimed to assess the effect of a pectin extracted from lemon and the probiotic strain *B. longum* BB-46, given in combination or alone, there was a positive correlation of *Holdemanella* with acetic and butyric acid, and a negative correlation with ammonium ions [50]. In an experimental high fat-induced oxidative stress, polyphenol supplementation affected different taxonomic levels of the gut microbiome by improving the proportion of *Blautia* (a butyrate producer) [51]. *Blautia* is one of the major taxonomic groups of the human gut microbiota (a genus in the *Lachnospiraceae* bacterial family, degrading complex polysaccharides to acetate, butyrate, and propionate (short chain fatty acids) that can be used by the host for energy and as a source of butyrate [52]. In a study of subjects who completed a 6-week endurance-based exercise intervention, there was an increase in butyrate concentrations induced by the exercise as a result of an increase in *Lachnospira* spp. [53]. This increase was independent of the BMI and decreased after return to sedentary activity. Surprisingly, *Lactobacillus* was found to be decreased at the end of the study, even being part of the probiotic. The reason for this finding is unknown. A point to be considered is how these strains could be able to resist the digestive system and arrive in sufficient amounts to detect enrichment of this genus. Conversely, both were detected by species-specific PCR in preliminary acute ingestion assays in feces (data not shown), although these results cannot be directly extrapolated to humans. Discussing a possible explanation for the functional effect of the formulation, even with a decrease in lactobacilli relative levels, is the potential capacity of extracellular metabolites of lactic acid bacteria to act as a prebiotic for key bacteria, influencing not only their growth and cell death, but also the expression of genes related to cell protection [54]. However, it seems that changes in microbiome do not directly correlate with the strains consumed, pointing that other mechanisms not necessary based on simple colonization might have a role on the results obtained.

The mechanisms by which the microbiome can impact upon oxidative stress and its effects are diverse. Among these, the production by the microbiota of toxic compounds can have a key impact on the health of the individual. Within this group, tryptophan catabolism by tryptophanase of certain bacterial groups produces indole, which is metabolized further to indoxyl-sulfate or indole-3 acetic acid. The latter toxins are secreted into the urine and are accumulated in the case of renal failure. These toxins decrease glutathione levels in renal tubular epithelial cells ren-dering them more vulnerable to oxidative stress [55]. Also, by activating ar-yl-hydrocarbon receptor (AhR) they can exert various deleterious effects [56,57].

Short-chain fatty acids, products of bacterial metabolism, have also been identified as an oxidative stress control mechanism. In a model of apoptosis in β-cells, butyrate and acetate attenuated the overproduction of ROS and NO and prevented cell apoptosis, and reduced viability and mitochondrial dysfunction [58]. Moreover, a bidirectional connection between mitochondrial genotype, ROS production, and gut microbiome has been recently established [59].

The present findings should be interpreted taking into account the limitations of the study, such as the small study population and the short duration of the intervention of only 6 weeks. Therefore, further studies with a larger sample size and duration of consumption of the probiotic product are warranted. It should be noted that in the present study, SOD and GPx were measured in serum samples, and significant differences between the study groups were not observed. However, it may be possible that significant differences could have been obtained by measurement of SOD and GPx in red blood cells.
