Fruits were collected between June and July 2021 at La Labor del Río, San Luis Potosí (21º46′46′′ N 100°36′37′′ O). Fresh berry cactus juice was extracted as previously reported for similar cactus fruit [11]. The BJC supernatant underwent a proximal analysis before being stored in aliquots at 80°C. Aliquots were lyophilized (277.12g) at 63°C under 3 Pa for 48 h and protected from light at 20°C until use.

The chemical analysis of the BJC composition was performed by the Centro de Investigación y Asistencia en Tecnología y Diseño del Estado de Jalisco A.C. using Mexican Official Standards based on AOAC methods (Official methods of analysis of AOAC International) [11]. The mass (g/100 g) of the sample was determined using the moisture (%) obtained by BJC evaporation from a heat treatment using the following equation: moisture (%) = [(wet sample weight dry sample weight)/sample weight 100]. Ash is the dry residue obtained by incinerating and calcining the BJC sample at 600°C and quantification of total ash. Crude fiber was digested with sulfuric acid (BJC), and the insoluble residue of crude fiber and salts was filtered, dried, weighed, and calcined for weighing again. These analyses were performed gravimetrically. The Soxhlet method was used to determine the fat content. Briefly, the fat content in 2 g of BJC was determined by acid hydrolysis of the protein-fat complex (33% hydrochloric acid). Subsequently, the fat was extracted with petroleum ether at 30°C–60°C using a Soxhlet extractor, and the fat was quantified directly. The protein content was measured using the Kjeldahl method, and the total carbohydrates were determined by the difference of proximal analysis using the following equation: total carbohydrates (%) = [100 (protein + lipids + ash + crude fiber)]. All samples were expressed as g/100 g and analyzed from at least two independent experiments. Supplementary Table 1A shows the results.

Experiments were performed with the approval of the Care Committee and Use of Animals of the Medicine Faculty of the Universidad Autónoma de San Luis Potosí (BGFM UASLP-22-24) according to the ARRIVE 2.0 Guidelines and the National Institutes of Health. The animals were housed in individual cages and maintained in a controlled temperature room (20–22°C), humidity (44–66%), and illumination (12-h light/dark cycle) throughout the research protocol. The animals were fed a standard diet and water ad libitum at the beginning of the research protocol and acclimatized for one week.

Twenty male Wistar rats (120–150 g, 21 days old) were randomly divided into four groups (n = 5) Fig. 1: the control group receiving AIN-93 diet plus 1 mL of sterile water (control); the second group received AIN-93 diet plus 200 mg/kg of BJC lyophilized and hydrated with 1 mL of sterile water (BJC); the third group received a high-fat diet (HFD) based on AIN-93 with 45% calories from lipids based on lard [12] plus 1 mL of sterile water; and the fourth group was treated with HFD plus 200 mg/kg of BJC lyophilized and hydrated with 1 mL of sterile water (HFD + BJC). The AIN 93 diet was paired with the average weekly HFD consumption; the animals that received AIN 93 were supplemented with drinking water, while the HFD groups were supplemented with 2.5 % scrose in drinking water ad libitum during the 20 weeks of dietary regimen. The BJC dose was established based on our group’s studies using a pitaya juice concentrate derived from cactus fruit (Stenocereus huastecorum), which has a similar chemical composition and bioactive metabolites [13]. All sterile water and BJC treatments were dispensed daily by gavage (G18), following reported protocols [11], for 140 days. Body weight and food and water intake were recorded daily. Subsequently, the animals were euthanized after anesthesia with sodium pentobarbital from PISA (Guadalajara, Jalisco, México; 90 mg/kg, i.p.), and blood was collected from the aorta in heparinized tubes (50 U/mL). The samples were centrifuged at 1,470 g at 4°C for 10 min, and the plasma was collected and stored at − 20°C for determination of biochemical parameters. Subcutaneous adipose tissue, right inferior lobule of the liver, and intestine (ileum and jejunum regions) were collected to assess the total fat percentage following the modified Folch method [11]. Fresh fecal samples from rats were collected in sterile tubes 48 h before sacrifice and were stored at 80°C until further analysis.

The clinical criteria for MetS diagnosis in diet-induced rat models were established according to previous studies in rodents and humans, which include an increase of 3 or more clinical parameters as follows: increase in body weight, high body mass index, hyperglycemia, insulin resistance, triglycerides, total cholesterol (TC), decrease in high-density lipoprotein cholesterol (HDL-C), and increase in diastolic and systolic blood pressure. After 140 days of treatment, the following criteria were met: body weight ≥ 432 g, fasting plasma glucose ≥ 151 mg/dL, triglycerides ≥ 109 mg/dL, TC ≥ 121 mg/dL, and an increase in total fat mass (subcutaneous and visceral) of ≥ 20%.

Unhealthy eating habits and a sedentary lifestyle are major risk factors for obesity and metabolic syndrome. Lifestyle interventions often provide pharmacological and surgical treatment with adjuvant results. Thus, addressing these habits, primarily through diet, is essential for managing progressive MetS [14]. In this study, the effects of BJC on diet-induced MetS in rats were investigated. At the beginning of the experiments, no significant differences in body weight, glucose, triglycerides, TC, ALT, and AST were observed between groups. However, after 140 days, the rats fed with HFD exhibited a marked increase in body weight compared with those on an AIN-93 diet (679 ± 110.63 g vs 434.8 ± 16.39 g, p-val. = 2.18E-05), indicating that the HFD induced overweight (Table 1). The BJC treatment significantly reduced body weight gain in the HFD + BJC group (611 g ± 46.82) compared to the HFD group (679 g ± 110.62). The control and BJC groups showed a similar weight gain among themselves (434.8 g ± 16.39 and 440 g ± 36.69, respectively).

^a The mean and standard deviation of each parameter are shown. Control, AIN-93 diet control group. BJC, Berry cactus juice concentrated (200mg/kg), AIN-93 diet plus BJC treatment group. HFD, high-fat diet group (45%). HFD-BJC, high-fat diet with berry cactus treatment. Significant differences in biochemical parameters are marked with *, p-val. < 0.05 (Welch’s T test).

Evidencing the impact of an HFD on metabolic health by significant weight gain and metabolic disturbances, including hyperglycemia, hypertriglyceridemia, hypercholesterolemia, hyperleptinemia, and hyperinsulinemia, which are hallmarks of MetS and align with the guidelines of the National Endocrinology Program ATP-III, the World Health Organization, and the International Diabetes Federation for MetS diagnosis [14]. In this study, HFD group exhibited a significant increase in metabolic markers compared to control group, they presented a hyperglycemia (Δ 62 mg/dL, p-val. = 0.00009), hypertriglyceridemia (Δ 48.25 mg/dL, p-val = 0.001), hypercholesterolemia (Δ 30.59 mg/dL, p-val. = 0.013), hyperleptinemia (Δ 117.34 pg/mL, p-val. = 0.014), and hyperinsulinemia (Δ 10.59, p-val. = 0.002), as shown in Fig. 2a-e, respectively. This group also showed increased serum creatinine concentration (Δ 0.24 mg/mL, p-val. = 0.003 (Table 1). In addition to increasing the total hepatic fat percentage (Δ 2.4 %, -val. = 0.0004), and the total jejunum fat percentage (Δ 11.01 %, -val. = 0.0009) and decreased compared with the total ileum fat percentage (Δ 9.64 %, -val. = 0.051) with the control group (Table 1). In response to a substantial fat load, elevated serum creatinine levels, along with increased hepatic and jejune fat percentages in the HFD group, provide additional evidence of metabolic dysfunction, confirming the detrimental effects on health, pointing out the importance of the liver and intestine in the body’s adaptation to the metabolic changes induced by HFD exposure [15].

In contrast, the HFD + BJC group exhibited a significant reduction in serum TC (Δ -37.16 mg/dL, p-val. = 0.06), insulin (Δ -8.25 ng/mL, p-val. = 0.1), and BUN levels (Δ -6.4 mg/mL, p-val. = 0.06) compared with the HFD group, as shown in Fig. 2c, e, and f, respectively. There was also a decrease in fasting blood glucose (Δ -18.9 mg/dL, p-val. = 0.2), triglycerides (Δ -13.8 mg/dL, p-val. = 0.48), and leptin levels (Δ -86.89 pg/mL, p-val. = 0.27) (Fig. 2a, b, and d). In addition to total hepatic fat, intestinal fat (jejunum), and adipose tissue fat, these changes did not reach statistical significance (Table 1). Furthermore, there were no significant differences in ALT and AST levels between the groups (p-val. = 0.3 and 0.43, respectively).

These findings suggest that BJC supplementation mitigates some of the metabolic alterations induced by HFD, particularly by improving lipid and glucose metabolism (significantly reducing serum TC, insulin, and BUN levels) and reducing total fat accumulation in the ileum. No significant effects on glucose, triglycerides, leptin, ALT, and AST levels, nor on total fat accumulation in the liver, jejunum, and adipose tissue (Table 1). Previous studies with a similar model (49% lard) showed that administering a leaf extract from Cecropia peltata for 90 days significantly reduced serum glucose and lipid levels [16]. Note that the follow-up period in the present study was 140 days and 45%-lard, which may be related to the differences between previous studies (AIN93 diet-based). In this study, the benefits of cactus fruit-based interventions in promoting weight loss, associated with minor total adipose and ileum fat percentage, and normo insulinemia, which have the potential to improve overall health outcomes, considering the strong link between MetS and various diseases.

In addition, in this study investigated the liver and kidneys, which play a critical role in glucose metabolism. High serum TC, insulin, creatinine, and BUN levels confirmed that HFD intervention disrupted liver and kidney function. Notably, treatment with BJC effectively decreased serum insulin, leptin, creatinine, and BUN levels without modulating liver enzymes (ALT and AST), providing evidence of BJC´s ability to regulate liver and kidney function. In this regard, an experiment made with streptozotocin-induced diabetic rats that received 4 g/kg berry cactus juice showed diminished circulating glucose, TG, and TC levels, in addition to improved renal function. It was associated with the re-establishment of glutathione and glutathione S-transferase levels in renal tissue, compared to non-treated diabetic rats. This report concluded that the renoprotective properties can be attributed to dietary bioactive compounds, including flavonoids, phytosterols, phenolic acids, betalains, and antioxidants [15].

In last decade, microbiota has emerged as a new factor associated with MetS pathophysiology [5]. Dysbiosis contributes to MetS development, but its molecular mechanisms remain poorly understood [5]. The critical commensal microbial phyla include Proteobacteria, Actinobacteria, Verrucomicrobia, Firmicutes, and Bacteroidota, with the latter being the most dominant in the gastrointestinal tract [5]. To investigate the impact of BJC intervention on the abundance of specific bacterial taxa, a 16S analysis of community composition at the phylum level was conducted. On average, ~ 10,000 fragments were sequenced per sample, with five replicates per group. The total abundance per group showed no differences, except for HFD + BJC_rep1 (almost 60,000 fragments) and HFD + BJC_rep20 (less than 1000 fragments; Supplementary Fig. 1), which were excluded.

In Fig. 3a shows that the control group showed Firmicutes and Bacteroidota as the most abundant (~ 53% and 38%, respectively), followed by 3.5% Campylobacterotaby, Patescibacteria, and Proteobacteria (~ 2%, respectively) and 0.01% Verrucomicrobiota. The BJC group showed 43.1% Firmicutes and ~ 51% for Bacteriodota, 3% Verrucomicrobiota, 2.1% Proteobacteria, while < 1% Patescibacteria. While, HFD group had ~ 54% Bacteroidota, ~ 41% Firmicutes, 0.06% Proteobacteria, ~ 2% Patescibacteria, and Campylobacterotaby. The < 1% Verrucomicrobiota. The administration of BJC in rats with HFD (HFD + BJC) induced an increase in Firmicutes to ~ 64%, and a reduction in Bacteroidota to 18%. Proteobacteria, Actinobacteriota, and Verrucomicrobiota increased to 11%, 3.5%, and 1.1%, respectively. BJC induce a diversity restoration of Firmicutes and Proteobacteria in this model, which were decreased in the HFD group. So, Firmicutes and Bacteriodota were the most abundant phyla in all groups. However, their abundance differs across the groups. According to the present findings, HFD increased the abundance of the Parabacteriodes genus, and BJC reversed those changes. This result is like those observed by other groups, where its increase was associated with metabolic imbalance, an increase in body weight, and in lipid metabolism, including TC and TGs, among others [17]. In fact, it was described that the presence of Bacteroidetes strain was associated with increased B. fragilis and P. diastolic. The reduction in the metabolism of carbohydrates and decrease short-chain fatty acids in the gut have been related to MetS development, as we can see here, BJC addition restores the diversity; this effect was also seen in MetS patients that were fed with a Mediterranean diet with high concentrations of polyphenolic compounds, increase the abundance of beneficial bacteria like Enterococcus, Prevotella, and Bacteriodetes [18]. Phylum relative abundance in each of the replicates of all groups is shown in Supplementary Fig. 2.

These results shown that the Alpha diversity was higher in the HFD + BJC group, showing higher Shannon and Simpson’s diversity compared with the other groups (4.885 and 0.984, respectively; Fig. 3b, cyan boxes). The Shannon diversity mean was 4.04, 4.19, and 4.15 (Fig. 3b, right panel), and the Simpson's diversity mean was 0.975, 0.977, and 0.973 (Fig. 3b, left panel) for the control, BJC, and HFD, respectively. Supplementary Table 2 presents the alpha diversity values of all the groups. Meanwhile, beta diversity analysis showed that HFD + BJC has higher similarity to the control than to the HFD group (Fig. 3c), likely due to gut microbiota disruption dominated by Bacteroidetes and Firmicutes.

Besides, a sparse partial least-squares Discriminant Analysis (sPLS − DA) permutation test was used to identify relevant taxa between groups. Initially, there was no clear group differentiation in the clustering pattern, including all genera (Fig. 4a). After sPLS-DA considering only the variables with significant changes for each group, a clear differentiation was observed in the clustering pattern (Fig. 4b), identifying relevant genera surrounded by HFD versus HFD + BJC, as shown in Fig. 4c and Fig. 5.

Specifically, Fig. 5a shows that the control, BJC, and HFD groups had similar abundances of Parabacteriodes, but it was significantly reduced in the HFD + BJC group. Further analysis revealed that the HFD + BJC group reduced the Helicobacter genus relative to the control and HFD groups. In addition, it was observed that only BJC administration reduced the abundance of the Helicobacter taxa in both the control and HFD groups. Thus, BJC directly modulates Helicobacter genus reduction, independent of diet composition (Fig. 5b). Similar results were observed by Jin & Zhang (2020), who reported a reduction in Helicobacter, which was associated with a pro-inflammatory gut environment and an increase in carcinogenesis in animals fed with HFD [19]. H. pyllori infection has been associated with insulin resistance, lipotoxicity and liver damage. Interestingly, red wine or grape juice consumption has been associated with reduced Helicobacter pylori symptoms [20]. Here, it was observed that BJC reduces the Helicobacter genus, so BJC may help to minimize dysbiosis due to similarities between the compositions of juices.

A few genera were present in only one group, e.g. the Weissella genus was present in low abundance in the HFD group (Fig. 5c) and the unclassified NK4A214 group, which was present in BCJ-treated rats (Fig. 5d). In particularly, the low abundance of Weissella genus has been previously described and has been associated with reduced dysbiosis in the MetS model. In this respect, Weissella cibaria MG5285 has been related to the reduction of metabolic parameters such as glucose and lipogenic proteins like fatty acid synthase, peroxisome proliferator-activated receptor gamma, and sterol regulatory element-binding protein 1c, inflammatory marker reduction, and an increase in bacteria strains that increased short-chain fatty acid production. This effect was proved in a mouse model showing the same effects and an increase in insulin secretion and reduction of fat mass [21]. Thus, the lower abundance of the Weissella genus in the HFD group may mediate a beneficial effect on rats in the present study. However, this genus has also been associated with the harmful impacts of having a dual function. Some strains have been reported to have pathogenic potential.

Additionally, the genus Lactobacillus did not show substantial changes between the control, BJC, and HFD groups, whereas it showed increased abundance in the HFD + BJC groups (Fig. 5e). Lactobacillus abundance has been linked to lower body weight, glucose, and insulin levels, and improves lipid metabolism [22]. The increase in the treated group (HFD + BJC) may indicate an improvement in lipid metabolism, which correlates with the biochemical parameters.

In the case of Faecalibaculum and Bifidobacterium, the abundance was only present in the HFD + BJC group (Fig. 5f and g, respectively). In this regard, an increase in the abundance of several short-chain fatty acid-producing bacteria, such as Faecalibaculum rodentium and Bifidobacterium, has been reported in mice fed an obesogenic diet co-supplemented with cranberry polyphenol extract [23], among others. In this study, an increase in Bifidobacteria in the HFD + BJC group was associated with reduced metabolic markers, lipids, leptin, and insulin levels, promoting an anti-obesogenic environment [24]. Berry juices have been linked to increased Bifidobacterium abundance, which exhibits anti-inflammatory and cardioprotective effects [25].

Whereas the abundance of HT002 (Limosilactobacillus) was significantly higher in the HFD + BJC group than in the other groups (Fig. 5h). BJC administration in HFD-fed animals showed significant presence of Desulfovibrio, Adlercreutzia, and C. sensu stricto 1 (Figs. 5i, j, and l, respectively). In addition, BJC supplementation showed a relative abundance of Desulfovibrio and Clostridium sensu stricto 1 when compared with other groups in the study. The increase in these two taxes is usually related to pro-inflammatory effects and endotoxemia [26]. In this regard, Jiao et al. (2019) suggested that blueberry polyphenol extract, as a potential prebiotic agent, influences the gut microbiota to positively affect HFD-induced obesity in C57BL/6J mice, highlighting a significant increase in abundance in Desulfovibrio, Adlercreutzia, and Bifidobacterium, which is associated with improved glucose tolerance [27].

In the case of Akkermansia taxa, no changes were observed between the control and HFD groups; however, BJC consumption increased significantly regardless of the dietary pattern (Fig. 5k). Similar results have been reported using a polyphenol-rich cranberry extract in C57BL/6J mice treated with MetS (200 mg/kg/day for 8 weeks) and in healthy humans treated with Prebiocran™ for 4 days [28], showing a proportional abundance increase in Akkermansia spp. These changes were associated with beneficial metabolic effects and anti-obesogenic effects, respectively, through reduced weight gain, visceral obesity, improved insulin sensitivity, decreased liver weight, attenuated serum levels of AST and ALT [28], and triglyceride accumulation associated with blunted hepatic oxidative stress and inflammation. Notably, the metabolic effects observed in the mentioned study are similar to those observed in animals with MetS supplemented with BJC.

Fusicatenibacter, Phascolarctobacterium, and Incertae sedis did not show any changes (Fig. 5m, n, and p). Considering all reports and the results in this study, the lack of changes may reflect interindividual variability linked to specific gut microbial ecologies and bioactive diets, such as those including polyphenols, which are associated with the dietary patterns and probably with the content of bioactive compounds content of BJC [29].

The Caproiciproducens genus showed a significant increase in BJC rats compared with the HFD + BJC group (Fig. 5o), without observing substantial changes in the HFD group. Caproiciproducens is an anaerobic bacterial genus that produces medium-chain fatty acids, particularly caproic acid. Its presence has been associated with fermentative metabolism, lipid metabolism, and gut barrier function. It has been observed that the administration of 200 mg/kg of wild blueberry polyphenol extract in high-fat, high-sucrose-fed C57BL/6J male mice for 8 weeks resulted in a tendency toward a greater abundance of several fermenting genera, including Caproiciproducens [30], associated with improved glucose tolerance and restored intestinal mucus integrity [31].

The Lachnospirceae UGC-008 genus shows a reduction in HFD rats compared with the control group animals, although this difference is not significant. The BJC consumption in rats with HFD significantly increased the presence of this genus, as shown in Fig. 5q. Finally, C. Saccharimonas was abundant in the HFD group, and BJC supplementation potentiated its abundance compared with the BJC group (Fig. 5r). These results align with previous reports showing that in obese Spanish [32] subjects, the abundance of the Lachnospiraceae genus was significantly lower in obese compared with lean subjects, and the abundance of gut microbiota was directly related to body mass index. Furthermore, Sprague-Dawley rats fed a 45% HFD for 81 days showed a significant increase in C. Saccharimonas from the first week of HFD feeding, which became more pronounced after 11 weeks [33], like our case.

Overall, the findings to this point suggests that BJC improves metabolic markers, including less accumulation of body fat in tissues, reductions in serum TC, TG, leptin, and insulin levels, while persevering microbiota diversity in MetS rats, potentially through its dietary bioactive compounds (polyphenols, pectins, and sterols) [15], which might stimulate the proliferation of beneficial bacteria [10] or limit MetS-related phyla to restore the abundance and uniformity of the HFD-induced microbial community to levels similar to the control group. The enhanced alpha diversity, shown as a higher Shannon and Simpson’s diversities in the HFD + BJC group, further reinforces this protective effect, as greater microbial diversity has been linked to better metabolic health [6, 7]. Nevertheless, this study has some limitations that need to be addressed. Further investigation is required to confirm the changes in the microbiome caused by BJC in the human MetS population.

The abundances of various genera and cellular pathways were further analyzed to explore the role of the bacterial microbiota in MetS. Using sPLS-DA, specific pathways associated with different treatments were identified. This pathway analysis offered further insights into the molecular mechanisms underlying the observed changes. Here, enriched functions, signaling, and metabolic pathways are described, with only correlations having a p-value < 0.1 being described. The complete association network between genus abundance and pathways is shown in Fig. 6 and Supplementary Fig. 3.

First, the nodes of relevant taxa (yellow circles) and pathways (green squares) of HFD (blue color) and HFD + BJC (red color) groups are highlighted. In Fig. 6, only relevant pathways for HFD + BJC were base excision repair, starch and sucrose metabolism, glucagon signaling pathway, RNA polymerase, glycolipid metabolism, and tropane, piperidine, and pyridine alkaloid biosynthesis pathways, suggesting that BJC modulates these pathways to restore metabolic balance in MetS. Specifically, in relation to glycolipid metabolism, which correlated positively with beneficial genera such as Lactobacillus, Faecalibaculum, and Bifidobacterium, previous studies have demonstrated that these strains play an essential roles in carbohydrate fermentation, gut permeability, and inflammation [5].

Moreover, seleno-compound metabolism, degradation of aromatic compounds, platinum drug resistance, cysteine and methionine metabolism, and benzoate degradation were also relevant for HFD + BJC compared to HFD (Fig. 6) and control (Suppl. Figure 3). Meanwhile, sphingolipid metabolism, glyoxylate and dicarboxylate metabolism, other glycan degradation, and peroxisome pathways were relevant for all experimental groups compared to HFD + BJC. This suggests that the microbiota becomes enriched in these pathways following BJC administration in the context of HFD.

Furthermore, Fig. 6 shows that in the HFD and HFD + BJC comparison, a positive correlation (red line) was found between the genera UCG-005 and NK4A214, Lactobacillus, HT002, Adlercreutzia, Faecalibaculum, and Bifidobacterium. These genera were positively correlated with glycerolipid metabolism, benzoate degradation, and biosynthesis of tropane, piperidine, and pyridine alkaloids. Furthermore, Lactobacillus was negatively correlated (blue line) with Helicobacter and Parabacteroides, which were also negatively correlated with the aforementioned pathways (green squares). These correlations were similar in the other comparisons with the control and BJC groups (Supplementary Fig. 3).

In contrast, sulfur metabolism, naphthalene, chloroalkene, and aminobenzoate degradation were exclusively relevant for HFD + BJC compared to HFD (Fig. 6). Furthermore, pathways related to lipid accumulation, including the citrate cycle, steroid hormone biosynthesis, lysosome, polyketide sugar unit biosynthesis, and atherosclerosis, were identified as relevant in HFD, further linking HFD to increased lipid accumulation and metabolic dysfunction [34]. These modulations indicate that BJC may influence carbohydrates and lipid metabolism in MetS through microbiota modulation.

Additionally, the identification of pathways involved in ferroptosis, streptomycin biosynthesis, and biofilm formation suggests that gut microbiota in the HFD group may adapt to a lipid-rich environment through the synthesis of antibiotics and biofilm formation as possible mechanisms for the damaged barrier function and reducing intestinal permeability by gut colonization of microbiota associated with MetS, potentially exacerbating MetS symptoms [34]. For the BJC group valine, leucine, and isoleucine degradation, two − component systems and lipopolysaccharide biosynthesis pathways were relevant. In the comparison of the HFD + BJC group with the control group (Suppl. Figure 3), the relevant pathways included fatty acid degradation, thiamine metabolism, MAPK signaling, glycolysis/gluconeogenesis, seleno-compound metabolism, phosphor-transferase system, fructose, and mannose metabolism. These findings suggest that BJC administration revealed significant differences in metabolic pathway activation between the groups, highlighting its potential modulatory effect on MetS. Furthermore, the diversity of pathways associated with the microbiota abundances in the different evaluated treatments, such as amino acid degradation, two-component systems, and MAPK signaling, opens exciting possibilities. These pathways are key to understanding the profound alterations in microbiota that occur after BJC administration in normal and HFD conditions and their impact on metabolic parameters. These findings suggest that the relationship between BJC and dysbiosis-MetS is potentially complex, and further studies are needed to investigate the effects of BJC on the gut microbiome or individual bacteria.

A comprehensive approach was employed to predict the functions and cellular pathways of relevant genera. Biochemical data such as weight or serum TC levels were used to perform correlation analyses, identifying 19 genera, 48 cellular pathways, and 83 functions with significant associations regardless of the treatment and diet (Figs. 7 and 8, and Supplementary Figs. 4 and 5, and Supplementary Tables 3 and 4).

The diversity of pathways associated with the microbiota abundances in the different evaluated treatments and their impact on metabolic parameters, such as amino acid degradation, two-component systems, and MAPK signaling, opens exciting possibilities. Several key associations emerged when focusing on the HFD + BJC group versus HFD alone, as shown in Fig. 7. The genus Bifidobacterium was positively correlated with body weight (ρ = 0.52, p = 0.02) and serum triglycerides (ρ = 0.51, p = 0.02), whereas Akkermansia abundance was positively correlated with serum TC (ρ = 0.58, p = 0.008). In contrast, both Romboutsia and Adlercreutzia were negatively correlated with serum AST levels (ρ = − 0.46, p = 0.04; ρ = − 0.55, p = 0.014, respectively). These findings suggest that the relationship between BJC and dysbiosis-MetS is potentially complex, and further studies are needed to investigate the effects of BJC on the gut microbiome or individual bacteria. The correlation analysis between the abundance of genera and biochemical markers underscores the intricate relationship between gut microbiota and metabolic health.

Additional significant relationships were observed across all treatments and dietary patterns. Bifidobacterium, the genus with the greatest number of associations, was also positively correlated with serum leptin (ρ = 0.56, p-val. = 0.013), total hepatic fat percentage (ρ = 0.62, p-val. = 0.004), and serum creatinine (ρ = 0.57, p-val. = 0.01), suggest that this genus plays a key role in metabolic deregulation. High fecal Bifidobacterium levels have been associated with overweight and obesity in children [35]. This highlights the potential of further research on this genus to advance targeted studies on MetS. While Weissella abundance was positively correlated with serum leptin (ρ = 0.47, p-val. = 0.038) and total hepatic fat percentage (ρ = 0.48, p-val. < 0.03), see Supplementary Fig. 4.

In contrast, Allobaculum, Adlercreutzia, Clostridium sensu stricto 1, and Romboutsia were all negatively associated with AST (ρ= -0.49, -0.46, -0.52, and − 0.55; p-val. = 0.03, 0.04, 0.02, and 0.014, respectively) and with serum ALT levels ρ= -0.5, -0.41, -0.43, -0.56; p-val. = 0.07, 0.02, 0.06, and 0.01, respectively. These four genera, as well as Turicibacter, Faecalibaculum, Ligilactobacillus, and Intestinimonas, were positively associated with serum creatinine (ρ = 0.44–0.61, all p-val. < 0.05). Finally, Lactoacillus inversely correlates with AST (ρ= -0.5 and, p-val. = 0.015) Saccharomyces cerevisiae correlated positively with serum leptin (ρ = 0.49, p-val. = 0.03), Bacteroides with ileum fat area percentage (ρ = 0.46, p-val. = 0.04), Parabacteroides inversely correlated with serum leptin (ρ = − 0.48, p-val. = 0.03), and Caproiciproducens positively correlated with serum total cholesterol (ρ = 0.47, p-val. = 0.04) (Supplementary Tables 3 and 4). The other relevant genera were Akkermansia and Bacteroidetes. These were associated with serum TC levels and the ileum intestinal fat area percentage, respectively. Both phyla have been associated with the producting of short-chain fatty acids [36], which promote adipocyte leptin secretion, insulin sensitivity, and intestinal epithelium metabolism, and inhibit pathogenic bacteria. They also have anti-inflammatory actions, which underscores their potential as probiotics for MetS [34].

Parabacteroides and Bacteroides were relevant for the HFD + BJC and were associated with fatty acid biosynthesis and total ileum fat percentage. These were related to the phospholipid/cholesterol/gamma − HCH transport system ATP − binding protein, which participates in polyunsaturated fatty acid biosynthesis. Microbiota’s polyunsaturated fatty acids can confer host resistance to obesity by reducing triacylglycerol accumulation [36]. In addition, other proteins, such as a magnesium transporter and gluconate 5-dehydrogenase, were negatively associated with fasting plasma glucose and may be related to changes in microbiota abundances that could lead to dysbiosis if these changes persist. For instance, magnesium deficiency modulates the abundance of gut Bifidobacteria and metabolic disorders in mice [37], whereas gluconate 5-dehydrogenase is necessary for bacterial growth and colonization of the large intestine, and it has been proposed as a target in bacterial infections. Another enzyme that was positively associated with Parabacteroides and Bacteroides is the alpha-L-rhamnosidase. This enzyme had a negative correlation with serum TC and insulin levels, both of which are related to MetS. BJC administration likely promotes the proliferation of Parabacteroides, impacting MetS by modulating short-chain and polyunsaturated fatty acids and steroid hormone biosynthesis, as well as producing α-L-rhamnosidase for the bioavailability of flavonoids [9] that have been linked to systemic and intestinal health [10], and potentially can participate in the reduction of high levels of serum TC and insulin, as is negatively correlated with these. Furthermore, the presence of α-L-rhamnosidase in the genomes of Parabacteroides and Bacteroides genera, as previously reported [38], indicates that this enzyme participates in the catabolism of berry cactus flavonoids and stimulates the utilization of L-rhamnose in berry cactus glycosides. Once L-rhamnose is free, it is metabolized to other carbohydrates, which can be used as a substrate throughout the fructose and mannose pathway and redirected to glycolysis or amino sugar and nucleotide sugar metabolism. The fructose and mannose pathways were correlated with fasting plasma glucose, creatinine, BUN, and the total ileum fat percentage. These findings emphasize the complex interactions between gut microbiota, metabolic pathways, and host biochemistry, revealing potential microbial targets for MetS therapeutic interventions. To develop effective strategies for combating MetS, future studies should focus on elucidating the molecular mechanisms by which BJC and other dietary interventions modulate MAB-associated metabolic functions.

Finally, the genomes of various genera were annotated to confirm the presence of genes in pathways correlated with the biochemical parameters assessed. A complex participation of all genera in the pathways was identified. As shown in Fig. 8, the dendrogram on the left side of the heatmap highlights the three main groups of functions correlated with the biochemical data (column dendrogram). At the bottom of the heat map, the first group was positively correlated with total ileum fat percentage and negatively correlated with serum leptin. The genera associated with these metabolic parameters include Parabacteroides and Bacteroides (Fig. 7-bottom of the heatmap), and these genera are relevant for the HFD + BJC group (Fig. 6). Besides, the metabolic pathways related to this group include the citrate cycle, fatty acid metabolism and biosynthesis, phenylalanine, tyrosine, and tryptophan biosynthesis, carbon metabolism, ubiquinone, and terpenoid-quinone biosynthesis, oxidative phosphorylation porphyrin and chlorophyll metabolism, and bacterial chemotaxis. Notably, the total ileum fat percentage was associated with the phospholipid/cholesterol/gamma − HCH transport system ATP − binding protein (ρ = 0.5, p-val. = 0.03), which participates in the biosynthesis of polyunsaturated fatty acids; perosamine synthetase (ρ = 0.47, p-val < 0.04), which synthesizes membrane lipopolysaccharides of Gram-negative bacteria; and alpha − L−rhamnosidase ρ = 0.46, p-val. = 0.05). Alpha − L−rhamnosidase was correlated inversely with leptin (ρ= -0.43, p-val. = 0.06) (Supplementary Fig. 5). These enzymes are negatively correlated with serum TC and insulin levels, both of which are related to MetS.

The center of the dendrogram (Fig. 8) shows the second group that negatively correlates with the total ileum fat percentage and BUN levels and is positive with serum TC, AST, ALT, and insulin levels. Genera such as C. Saccharimonas, Akkermansia, Clostridium sensu stricto 1, and Adlercreutzia show similar associations (Fig. 7-top of the heatmap). These genera are relevant to the HFD (Fig. 6). Pathways included peptidoglycan biosynthesis (ρ= -0.76, p-val. = 0.0001), pyrimidine metabolism (ρ= -0.62, p-val. = 0.004), among other related pathways (p-val. < 0.05) to amino acid biosynthesis, antibiotic resistance, and virulence factors (Fig. 5). Supplementary Table 4 shows the complete list of pathways, Spearman’s coefficients, and p values. Serum TC levels were positively correlated (p-val. < 0.05) to the biosynthesis of amino acids (ρ = 0.48), lysine (ρ = 0.47), phenylalanine, tyrosine and tryptophan biosynthesis (ρ = 0.52), novobiocin biosynthesis (ρ = 0.47, p-val. = 0.04) and negatively associated with propanoate metabolism (ρ= -0.5, p-val. = 0.03). Serum insulin levels were positively correlated with glycerolipid metabolism (ρ = 0.46, p-val. = 0.04) (Fig. 8-middle of the heatmap). The functions related to the total ileum fat percentage were Xaa − Pro dipeptidase (ρ= -0.48, p-val. = 0.03), energy − coupling factor transport system permease protein (ρ= -0.48, p-val. = 0.04), oligopeptide transport system substrate − binding protein (ρ= -0.48, p-val. = 0.04), thioredoxin reductase (NADPH) (ρ= -0.51, p-val. = 0.02), cysteine desulfurase (ρ= -0.54, p-val. = 0.016), and ATP − dependent Clp protease ATP − binding subunit ClpE (ρ= -0.54, p-val. = 0.016) (Supplementary Fig. 5).

On top of the heatmap (Fig. 8), the third group of functions is positively associated with fasting plasma glucose, creatinine, and BUN and negatively associated with serum AST and ALT levels (Fig. 8). Clostridium sensu stricto 1, Adlercreutzia, Allobaculum, Ligilactobacillus, Faecalibacterium, Turicibacter, and Intestinimonas have similar associations (Fig. 7-top of the heatmap). These genera were relevant for the control and BJC groups and irrelevant for HFD + BJC. The pathways included MAPK signaling (ρ = 0.63, p-val. = 0.003), glucagon signaling (ρ = 0.6, p-val. = 0.006), fructose and mannose metabolism (ρ = 0.59, p-val. = 0.008), hypoxia-inducible factor-1 signaling (ρ = 0.55, p-val. = 0.01), ascorbate and aldrete metabolism (ρ = 0.47, p-val. = 0.04), and amino and nucleotide sugar metabolism (ρ = 0.5, p-val. = 0.03). Moreover, functions that show a negative association with serum AST levels include the glucagon signaling pathway (ρ= -0.54, p-val. = 0.01), S. aureus infection (ρ= -0.48, p-val. = 0.03), starch and sucrose metabolism (ρ= -0.46, p-val. = 0.04), sulfur metabolism (ρ= -0.47, p-val. = 0.04), serum ALT (ρ= -0.54, p val. = 0.016; Fig. 7). Protein − tyrosine phosphatase was associated with glucose (ρ = 0.55, p-val. = 0.02) and serum creatinine levels (ρ = 0.65, p-value). < 0.01) and, as well as magnesium transporter, positively associated with serum creatinine (ρ = 0.6, p-val. < 0.002) and negatively correlated with serum AST (ρ= -0.49, p-val. = 0.03). Gluconate 5 − dehydrogenase had an inverse correlation with serum glucose (ρ= -0.62, p-val. = 0.005) and creatinine (ρ= -0.6, p-value = 0.06) (Supplementary Fig. 5).

Other metabolic parameters with significant associations included the percentage of total adipose tissue fat with Oscillospira (ρ = 0.52, p-val. = 0.02; Fig. 7-bottom of the heatmap), serum ALT with Fusicatenibacter (ρ = 0.48, p-val. = 0.038; Fig. 7-bottom of the heatmap), arginine biosynthesis (ρ = 0.49, p-val. = 0.03), oxidative phosphorylation (ρ = 0.55, p-val. = 0.013), and sulfur metabolism (ρ= −0.54, p-val. = 0.016). Serum creatinine with C5 − branched dibasic acid metabolism (ρ = 0.48, p-val. = 0.03), oxidative phosphorylation (ρ= -0.53, p-val. = 0.018), chromate transporter (ρ= -0.53, p-val. = 0.008), and beta galactosidase (ρ= -0.53, p-val. = 0.018). BUN levels were associated with a two − component system (ρ = 0.51, p-val. = 0.02), biofilm formation − E. coli (ρ = 0.64, p-val. = 0.003), and fructose and mannose metabolism (ρ = 0.59, p-val. = 0.008) (Fig. 8-bottom of the heatmap). In addition, BUN levels were associated with different families (OmpR, NarL, and NtrC) of two-component system sensor histidine kinases (p-val. < 0.01). Ammonium transporters in the Amt family were correlated with BUN (ρ = 0.67, p-val. = 0.001). However, specific genera promote the bioavailability of some carbohydrates. For example, all the mentioned genera contain fructose and mannose metabolism genes. However, as previously described, Bacteroides and Parabacteroides preferentially participate in the catabolism of glycosides containing terminal L-rhamnose present in berry cactus flavonoids.