DISCUSSION
The treatment with L. fermentum CECT5716 to HFD-fed mice improved HFD-induced obesity, reducing body weight gain, which was associated with an amelioration of glucose and lipid metabolism. The treatment did not significantly modify energy intake, thus discarding any anorexigenic effect, but it showed anti-inflammatory and immunomodulatory properties, as seen before in different experimental models of intestinal inflammation (Rodriguez-Nogales et al., 2017; Rodríguez-Nogales et al., 2015). Indeed, L. fermentum downregulated the expression of pro-inflammatory mediators in HFD-fed mice as well as upregulated the expression of key transcription factors that control adiponectin such as PPARα (Kim et al., 2015). As in human obesity, this was associated with an improvement in glucose and lipid metabolism, most probably derived from the amelioration of the obesity-associated insulin resistance, as shown by the impact of the probiotic treatment on HOMA-IR values. Moreover, L. fermentumameliorated leptin resistance in liver and adipose tissue as well as restored adiponectin expression in fat contents, which has been previously reported for other probiotics, like different strains ofL. plantarum (Choi, Dong, Jeong, Jung, Kim & Kim, 2019; Kwon et al., 2020).
Obesity-associated insulin resistance is also characterized by impairment of intracellular glucose uptake that is mediated by the insulin-dependent receptor GLUT-4 (Vargas, Podder & Carrillo Sepulveda, 2020), whose expression is reduced when insulin resistance appears (Shepherd & Kahn, 1999), similarly to that observed in the present study. L. fermentum also improved insulin resistance, which was associated with an increased expression of Glut-4 , ameliorating glycemic levels and glucose utilization by target tissues, as previously reported for other probiotics (Chen et al., 2018).
Besides, the obesity-associated inflammatory state is closely related to the development of cardiovascular disease and endothelial dysfunction, in part, mediated by leptin (Huby, Otvos & Belin de Chantemele, 2016). Conversely, adiponectin has been shown to correlate inversely with arterial hypertension and endothelial dysfunction, since it facilitates the phosphorylation of endothelial nitric oxide synthase (eNOS) enzyme and the subsequent production of nitric oxide (NO), the most effective acetylcholine-induced endothelium-derived relaxing factor in aorta (Adya, Tan & Randeva, 2015). Additionally, vascular endothelial and smooth muscle cells are also targeting of the pro-inflammatory adipokines TNF-α and IL-6, which, as well, increase expression of pro-inflammatory cytokines, including TNF-α and IL-1β, as observed in the present study, and adhesion molecules. This results in the activation of the renin-angiotensin system that leads to inflammation of the vascular walls and development of pre-atheromatous lesions, impairment of vasodilation in humans, most probably through altering the expression and activity of eNOS and NADPH oxidase (Walther et al., 2015). Thus, it modifies the production of NO and superoxide anion (O2-) facilitating vascular oxidative stress (Didion, 2017). Interestingly,L. fermentum treatment reduced vascular expression of the pro-inflammatory cytokines TNF-α and IL-1β in obese mice, as well as inhibited the increased NADPH activity in the aortic tissue. This suggests a reduction of ROS production and a higher NO bioavailability, which could promote the restoration of the impaired endothelium‐dependent relaxation to acetylcholine, similarly to that reported previously for this probiotic in an experimental model of systemic lupus erythematosus (Toral et al., 2019).
Regarding the involvement of the gut in the pathogenesis of obesity, there is a defect in the intestinal barrier function. This leads to increased gut permeability (Teixeira et al., 2012) that facilitates bacterial components translocation, like LPS, that could reach systemic circulation and provoke metabolic endotoxemia (Cani et al., 2007). LPS may contribute to obesity-associated systemic inflammation upon binding to its main receptor TLR-4, located in immune cells, liver, adipose tissue and skeletal muscle. This promotes the activation of the transcription factor NFκB and subsequent production and release of cytokines, adipokines and ROS, thus altering glucose and lipid homeostasis (Boutagy, McMillan, Frisard & Hulver, 2016). The present study confirms these observations, since obese mice displayed reduced expression of the colonic markers of epithelial integrity as well as increased LPS plasma levels and up-regulated expression of Tlr-4 in liver, fat and aorta. Importantly,L. fermentum treatment significantly increased the colonic expression of the different markers involved in gut integrity in obese mice, thus restoring the intestinal barrier function and preventing bacterial components translocation, since it reduced LPS plasma levels and downregulated Tlr-4 expression. Other probiotics, likeL. sake OK67 (Lim, Jeong, Woo, Han & Kim, 2016), L. gasseri (Kawano, Miyoshi, Ogawa, Sakai & Kadooka, 2016)(and Bifidobacterium adolescentis IM38 (Lim & Kim, 2017) have also been reported to improve intestinal integrity and ameliorate inflammation in obesity in mice. Closely related to the above, the modulation of gut microbiota in obese mice exerted by L. fermentum seems to play a key role. Changes in gut microbial composition, mainly caused by external factors, can result in a dramatic alteration of the symbiotic relationship between gut bacteria and the host and promote the development of metabolic diseases, maybe by facilitating a low-grade inflammation, as mentioned before (Marchesi et al., 2016). In addition, it is well established that gut microbiota composition is altered in obesity (Ley, Backhed, Turnbaugh, Lozupone, Knight & Gordon, 2005), consisting on an enrichment inFirmicutes (F) as well as a reduction in Bacteroidetes(B), both in humans (De Filippo et al., 2010) and mice (Bagarolli et al., 2017). Increased F/B ratio has been associated with a more efficient hydrolysis of non-digestible polysaccharides in the intestinal lumen, so obese individuals extract more calories and fat from food than lean ones (Backhed et al., 2004). Our study agrees with these observations since HFD-fed mice showed an increased F/B ratio when compared with non-obese mice. However, L. fermentumtreatment was able to modulate gut microbiota composition, restoring the main bacteria phyla to the normal values observed in control diet-fed mice. The PCA analysis showed a clear separation between the clusters, indicating a shift in the gut bacterial composition induced by the probiotic. This amelioration of obesity-associated dysbiosis could be associated with the reduction of energy assimilation and potentially contribute to the beneficial effects observed. Additionally, in obese patients different phyla have been reported to be increased, includingFusobacteria and Proteobacteri a, whereasVerrucobacteria is reduced (Crovesy, Masterson & Rosado, 2020). Special attention has been paid to the latter, which includes the bacteria Akkermansia muciniphila , a mucin-degrading bacterium whose abundance is inversely related to body weight and type 2 diabetes in mice (Everard et al., 2013) and humans (Karlsson, Onnerfalt, Xu, Molin, Ahrne & Thorngren-Jerneck, 2012). Actually, it has been reported that A. muciniphila treatment, or those that promote its abundance, could reverse HFD-induced metabolic disorders (Vezza et al., 2019). In the present study, the proportion of the genus Akkermansia was reduced in untreated obese mice, while increased after L. fermentumtreatment, maybe through an increment in colonic mucins production. This probiotic could become a valid tool to acquire a properAkkermansia abundance that could have a therapeutic effect in obesity, as demonstrated before (Chang et al., 2019; El Hage, Hernandez-Sanabria & Van de Wiele, 2017).
Furthermore, it has been reported that Erysipelotrichi , specifically, Clostridium spp , is associated with metabolic syndrome and obesity in humans (Karlsson et al., 2013) and mice (Chakraborti, 2015). Our results confirmed this since Erysipelotrichi class and Clostridium spp abundance were increased in HFD-fed mice while L. fermentum treatment reduced significantly their presence in obese mice.
Regarding Bacteroidetes genus, a significant association between HFD and Bacteroides abundance was found, being higher in lean individuals than in obese ones (Castaner et al., 2018). Thus, some strains of Bacteroides , such as B. acidifaciens , have been reported to prevent obesity and improve insulin sensitivity in mice (Yang et al., 2017). Accordingly, in our study, untreated HFD-fed mice manifested a reducedBacteroides proportion in comparison with control diet-fed mice whereas L. fermentum significantly increased it.
Gut metagenomic analysis, in both control diet and HFD mice, has provided some insights into the mechanism of the microbiome to affect weight gain and obesity. Therefore, within Bacteriodetes , lineages rich in genes involved in amino acid metabolism, translation, and nucleotide metabolism, have been found less abundant in HFD mice, while those genes for membrane transport and replication and repair were increased. Similarly, HFD has been also described to modify gene activation on Firmicutes phylum. Membrane transport (mostly ABC transports), transcription and cell motility pathways were increased in abundance, while those with more genes for carbohydrate and energy metabolisms were decreased (Hildebrandt et al., 2009). Since ABC transporters control the transport of a variety of nutrients such as lipids, sugars, peptides and metals, an increase in the number of these transporters could favor energy intake, as seen in HFD-fed mice. Furthermore, a collection of genes involved in intake and assimilation of sugars are also found to be more abundant in microbiome samples from HFD-fed mice, as well as genes for phosphorus metabolism, mainly phosphotransferase systems involved in the uptake and assimilation of sugars (Greenblum, Turnbaugh & Borenstein, 2012). These results were confirmed in our study, where HFD-fed mice showed enrichment on genes responsible for transport (including ABC transporter), bacterial secretion, motility and sugars assimilation, among others, and were decreased after L. fermentum administration, confirming the potential of this probiotic on the modulation of gut microbiota.
In summary, HFD consumption promotes alterations in gut microbiota that may increase intestinal permeability and LPS translocation, as well as lead to insulin resistance, glucose homeostasis imbalance and systemic low-grade inflammation. However, these obesity-related features were reversed by changes in the gut microbiota profile induced by L. fermentum CECT5716 administration, which suggests a potential use ofL. fermentum in clinical practice. In conclusion, the probiotic treatment can be an important tool to prevent and treat patients with obesity and metabolic syndrome.