**3. Results**

#### *3.1. Characterization of the Obese Population with Gut Dysbiosis*

A total of twenty patients (14 females, mean age 46.4 ± 5.7 years-old; and 6 males, mean age 41.0 ± 12.1 years-old) with mostly class-I obesity were included in this pilot study (Table 1). At baseline, their body weight and the body mass index (BMI) were 87.8 ± 8.4 kg and 32.2 (5.3) kg/m2, respectively. According with the established pathophysiological parameters of blood pressure and glycemia [26,27], they were in the limit of normotension and normoglycemia. Their lipid profile was also slightly altered [28], showing a minor elevation of triglycerides, total cholesterol, and LDL-cholesterol, whereas HDL-cholesterol was in the normal range.

**Table 1.** Distribution of age, gender, and baseline blood pressure, glycemic and lipid profile between groups.


BMI, A1C, LDL, and HDL, as the body mass index, glycated hemoglobin, and low- and high-density lipoprotein, respectively. T and U values are also shown. Df, as degrees of freedom.

In addition, this obese cohort with a potential low risk of metabolic and cardiovascular disease showed an altered composition and distribution of bacterial gu<sup>t</sup> microbiota. By directed qPCR analysis of fecal samples, we evaluated the presence of muconutritive, immunomodulatory, and proteolytic bacteria. We observed an overall decrease in bacteria number, with a reduction in Firmicutes and mostly Bacteroidetes phylum (8.3 and 7.5 log NGC/g, respectively), compared to reference parameters (Table 2). Among Firmicutes, *Faecalibacterium sp* and *Enterococcus spp* were lessened. In the Bacteroidetes phylum, there was a notorious diminution of *Bacteroides spp*. Thus, the ratio of Firmicutes/Bacteroidetes was 0.5, which was over the reference range. In contrast, these patients showed normal levels of the Proteobacteria and Fusobacteria phyla (Table 2), but a robust decrease in the Actinobacteria and Verrucomicrobia phyla, particularly in the *Bifidobacterium spp* and *Akkermansia muciniphila,* respectively. As expected, these data sugges<sup>t</sup> that obese patients exhibited gu<sup>t</sup> dysbiosis with a significant alteration in the number and distribution of, particularly, muconutritive and immunomodulatory bacteria.


**Table 2.** Gut microbiota in obese patients.

Relevant bacteria phyla; Firmicutes and Bacteriodetes, and its ratio, and Proteobacteria, Fusobacteria, Actinobacteria, and Verrumicrobia, were evaluated in fecal samples before anti-obesity treatments. The reference ranges for the bacteria phyla were obtained from fecal samples of a control population (see Methodology and Supplementary Materials Figure S1). T and U values are shown. Df, as degrees of freedom. In bold, bacteria levels outside the reference ranges. NGC/g: number of gene copies per gram of feces.

#### *3.2. Reduction of the Body Weight and CV Risk Factors by PENS-Diet* +/− *Probiotics*

Since patients displayed an elevated BMI with altered microbiota, we next examined the effect of an anti-obesity strategy based on satiety neurostimulation and intake of a hypocaloric diet (PENS-Diet), with or without administration of multistrain probiotics. Patients were randomly divided into two groups (*n* = 10, each), with no significant differences in age, gender, body weight, BMI, blood pressure, glycemia, and lipid profile (Table 1). Gut microbiota was also akin in both groups (Table 2). Thus, before interventions, anthropomorphic characteristics and microbiota distribution were similar between groups. After ten weeks, patients with a PENS-Diet showed significant reductions in body weight and BMI (Table 3). BMI was reduced by 13% and dropped into the overweight range (28.0 kg/m2). PENS-Diet also decreased systolic and diastolic blood pressure, fasting glucose, triglycerides, and total cholesterol. Interestingly, patients with a PENS-Diet + probiotics exhibited a similar effect by ameliorating 20% of body weight and BMI (26.3 kg/m2), as well as blood pressure, fasting glucose, A1C, and triglycerides. Moreover, no adverse effects were found in both groups of subjects.

By further comparison between both therapeutic approaches (Table 4), PENS-Diet + probiotics unveiled a significantly higher weight loss (16.2 vs. 11.1 kg, respectively; *p* = 0.022) and total weight loss (%TWL) (17.5 vs. 12.9%, respectively; *p* = 0.02) than the PENS-Diet intervention. The excess BMI lost (%EBMIL) was also significantly higher (84.2 vs. 57.0%, respectively; *p* = 0.021) after probiotics. Moreover, plasma A1C, triglycerides and HDL-cholesterol levels were more reduced (−0.46 vs. −0.05 mg/dL, *p* = 0.032; −47.0 vs. −8.5 mg/dL, *p* = 0.002; and 10.5 vs. 0.05 mg/dL, *p* = 0.005, respectively) (Table 4). These data sugges<sup>t</sup> that an administration of multistrain probiotics to a PENS therapy under hypocaloric diets could further decrease the body weight, glycemia, and dyslipidemia in obese patients. In this regard, we tested the potential associations between probiotics and the body weight parameters, A1C, and lipid levels. By univariate linear regression, probiotics administration was significantly associated with the difference of WL, %TWL, %EBMIL, and A1C. Similarly, by quantile regression, Systolic blood pressure (mmHg)

probiotics was associated with the difference of TG and HDL (Figure 1A). Indeed, probiotics showed a positive association with WL, %TWL, %EBMIL, and HDL, while it was negative with A1C and TG (Figure 1B).


 **120.0 (2.5)**  **0.00**  **0.018**

**140.0 (20.0)**



The body weight, BMI, systolic and diastolic blood pressures, fasting glucose, A1C, and lipid profiles were analyzed after ten weeks of anti-obesity approaches. In bold are the statistically significant data. Variables with normal distribution were compared using Student's t test for paired samples, whereas variables with non-normal distribution were compared using the Wilcoxon Signed Rank test. T and W values are shown. Df, as degrees of freedom. *p* < 0.05 was considered significant. BMI, A1C, LDL, and HDL, as the body mass index, glycated hemoglobin, and low- and high-density lipoprotein, respectively.

**Table 4.** Differences in weight loss, blood pressure, and plasma parameters between PENS-Diet and PENS-Diet + probiotics.


The weight loss and percentages of TWL and EBMIL, systolic and diastolic blood pressure, and plasma glucose and lipids (triglycerides, total cholesterol, LDL-c and HDL-c) were compared between both groups of patients. In bold, the statistically significant data. Variables with a normal distribution were compared using Student's t-test for independent samples, while those with a non-normal distribution were compared using the Mann–Whitney U test. T and U values are also shown. Df, as degrees of freedom. *p* < 0.05 was considered significant. WL, %TWL and %EBMIL, as weight loss, percent of total weight loss, and the percentage excess BMI loss, respectively. A1C, LDL, and HDL, as glycated hemoglobin, and low- and high-density lipoprotein, respectively.


**(A)** 

**Figure 1.** Associations for probiotics and the body weight, glycemia, and dyslipidemia in obese patients. By univariate linear regression, the probiotics administration was significantly associated with WL, %TWL, %EBMIL, and A1C, whereas by a quantile regression, probiotics were associated with TG and HDL (**A**). In bold, the statistically significant data. Probiotics exhibited a positive association with WL, %TWL, %EBMIL, and HDL, while it was negative with A1C and TG (**B**). The associations between variables with a normal distribution were studied by a univariate linear regression, while those with non-normal distribution were studied by a quantile regression. WL, %TWL and %EBMIL, as weight loss, percent of total weight loss, and the percentage excess BMI loss, respectively. A1C and HDL, as glycated hemoglobin, and high-density lipoprotein, respectively.

#### *3.3. Microbiota Modifications after PENS-Diet* +/− *Probiotics*

Alterations in human obesity, glycemia, and lipidemia could parallel changes in gu<sup>t</sup> microbiota [29]. In fact, PENS-Diet showed a tendency to enrich some specific bacteria (i.e., *Prevotella spp*, *Bifidobacterium spp*) and to improve the Firmicutes/Bacteroidetes ratio (Table 5). However, PENS-Diet + probiotics was able to increase *Prevotella spp* (+1.3%, *p*=0.05) and further reduce the Firmicutes/Bacteroidetes ratio (0.10). This intervention also stimulated *Bifidobacterium spp* (+51.2%; *p* = 0.005) and *Akkermansia muciniphila* (+41.1%, *p* = 0.033) growth (Table 5). Thus, an addition of probiotics to anti-obesity intervention with a PENS-Diet may help to attenuate the altered Firmicutes/Bacteroidetes ratio in an obese gut, and to enrich its content of *Prevotella spp,* and mostly, Actinobacteria *(Bifidobacterium spp)* and Verrucomicrobia *(Akkermansia muciniphila)* bacteria.


**Table 5.** Bacterial differences after PENS-Diet or PENS-Diet + probiotics interventions.

*Prevotella spp*, *Bifidobacterium spp*, and *Akkermansia muciniphila* levels (log NGC/g) in obese patients after PENS-Diet or PENS-Diet + probiotics. The ratio of Firmicutes/Bacteroidetes is also shown for both strategies. Variables with a normal distribution were compared using Student's t test for independent and paired samples, whereas variables with non-normal distribution were compared using the Mann–Whitney U test for independent samples and Wilcoxon Signed Rank test for related samples. *p* < 0.05 was considered significant. T \* and U \*, as T-value and U-value between PENS-Diet + probiotics and PENS-Diet interventions. Df, as degrees of freedom. *p* \*, as *p* value betweenPENS-Diet+ probioticsandPENS-Dietinterventions.Inbold,thestatisticallysignificantdata.
