The administration of PFM or its BFS as a re-nutrition supplementation in a non-severe PEM model caused a weight gain in the animals, being the BFS which showed the fastest effect (after 15 days, Figure 2A). Considering that the intestinal architecture is altered during malnutrition state  and that PFM or its BFS administration improved the gut histology faster than milk alone (Figure 3), the microbial translocation to liver was analyzed. It was observed that increased bacterial translocation in the MC group did not occur after 5 days of re-nutrition, and this effect was similar in the animals that received milk, PFM or BFS. This study was important to verify the safety of the re-nutrition diet used in our model, mainly the administration of the complete fermented milk due to the high number of bacteria (108 CFU/ml) it contains (probiotic bacteria and yoghurt starter cultures).
According to previous results where the administration of PFM to healthy mice increased the immune function and conferred beneficial changes to the intestinal microbiota , different bacterial populations were analyzed in our PEM model. It was reported that protein malnutrition disrupts the normal ecology of the microbiota with excessive growth of anaerobic microorganisms in the upper gastrointestinal tract [16–18]. The results obtained in the present work, after 5 days of re-nutrition showed significant increases for bifidobacteria when the mice received PFM or its BFS but not for milk re-nutrition (Figure 2C). This finding was important considering that bifidobacteria could confer a beneficial effect on the stability of the intestinal microbiota . Allori et al  used an experimental murine model of severe malnutrition and showed that the equilibrium between aerobic and anaerobic strict bacteria in the gut microbiota was altered and that the ingestion of L. casei CRL 431 or conventional yoghurt was able to restore the gut microbiota. In contrast to these results, in our work the malnourished mice increased the number of total lactobacilli and even this increase was always significant compared to the well-nourished mice. After the re-nutrition period, malnourished mice re-nourished with any supplementary diet maintained the increases of total lactobacilli, which is important considering the beneficial properties attributes to Lactobacillus genus in the gastrointestinal tract. This difference and the lack of changes in the other bacterial populations after malnutrition period could be due to the fact that our model differs compared to the severe PEM model used previously by others.
Many beneficial effects especially on the immune system have been observed for bifidobacteria. They include: increased mucosal IgA production ; stimulation of phagocytic activity of mononuclear cells ; stimulation of natural killer cells activity ; increased lymphocyte responsiveness to oral and systemic challenge antigen [23, 24].
In this sense, considering that the consumption of the PFM and its BFS increased this bacterial population, some immunological parameters were measured to analyze their influence on the regulation of the immune system and to demonstrate the usefulness of the PFM as oral adjuvant when it is administered to immune-deficient hosts.
The study of different immune cells showed that PFM administration was the only diet that increased IgA+ cells in the lamina propria of the small intestine after 5 days of re-nutrition, similar to those reported for the administration of the PFM in well nourished adult mice where IgA+ cells and IgA secretion to the intestinal fluid was increased after 5 days of PFM administration .
The gastrointestinal epithelium is covered by a protective mucus synthesized and secreted by goblet cells. The concept of the mucus layer functioning as a dynamic defensive barrier is suggested by studies showing altered mucus-layer in germ-free animals [25, 26] and for consistent evidence of enhanced mucus secretion in response to intestinal microbes . The number of goblet cells was not significantly affected by the non-severe PEM or the different re-nutrition conditions (Table 1). It is important to note that the period of consumption of the PFM re-nutrition did not increase goblet cells as was previously reported in well-nourished mice ; the effect of the PFM in our model was more effective on other immune cells which were severely affected by the restricted diet administered to induce malnutrition.
Dendritic cells are known to be essential immune cells in innate immunity and in the initiation of adaptive immunity. These cells capture and transfer information from the outside the body to the cells of the adaptive immune system. Macrophages are an important cell population for the innate immune response and might also be involved in the regulation of acquired immune responses . It was reported that probiotic bacteria can exert their beneficial properties on the host immune system by activating these cells . The marker F4/80 is present on the surface of a family of cells member of the mononuclear phagocyte system of mice. The expression of this antigen can be considered a specialized adaptive state rather than a separate lineage, which is higher in mature macrophages and its expression is required for regulatory T cell development .
With regards to the antigen recognized by 33D1 antibody, it is an inhibitory receptor and is present on a subpopulation of DC. The lack of this receptor might suggest a gain in function; however, DC recognized by 33D1 are more effective for antigen presentation on the class II major histocompatibility complex, than DC without this marker ; thus the antigen that binds 33D1 antibody on DC, may reflect their maturation state.
Macrophages and DC numbers decreased in the MC mice. After 5 days of re-nutrition, the three diets used increased the number of macrophages but only PFM and BFS administration increased DC in the lamina propria of the small intestine compared to MC group (Table 2). The results obtained for macrophages and DC were similar to those reported for well-nourished mice in a model where PFM was evaluated during nursing and after weaning until immune maturity .
Another immune cells studied in our model were the T lymphocytes. The results obtained did not show significant differences in the re-nourished mice compared to the MC group. The lack of effect on the number of these cells could be explained because it is know that the thymus suffer atrophy after malnutrition . Maybe the period of time used for the re-nutrition was not enough to recover this organ or to stimulate a thymus independent production of T lymphocytes at the intestinal level from the cryptopatches .
To evaluate the functionality of the studied immune cells, even T cells which did not show numerical changes after re-nutrition, the number of various cytokines producing cells was analyzed (Table 3). IL-2 is a cytokine involved in the progression of T lymphocytes as a growth factor . It is produced by Th1 lymphocytes and can also be produced by DC . It was observed that re-nutrition increased the number of IL-2+ cells to values similar to the WC. This finding was important because IL-2 is an important nexus for the adaptive immunity.
TNF-α and IFN-γ are known pro-inflammatory cytokines. They are produced by activated cells and are able to activate other cells during inflammatory responses. However, it was recently demonstrated that these cytokines are more important in the crosstalk between immune cells than in the inflammatory response where IL-17 is involved , and these can be induced at the intestinal level after probiotic stimulation . In our non-severe PEM model TNF-α+ and IFN-γ+ cell numbers were decreased in MC mice; the re-nutrition with milk and PFM increased their numbers in the small intestine. No effects were observed for these cytokines in mice that received BFS during re-nutrition period, showing the importance of the whole fermented product to activate the immune cells to produce cytokines. Similar results were obtained for IL-12+ cells. This cytokine plays a pivotal role in Th1 T cell differentiation and induces naive T cells to produce IFN-γ.
The study of IL-6 producing cells showed the importance of the soluble fraction (BFS), which was more effective to stimulate the production of this cytokine. This cytokine is also involved in the increased clonal expansion of IgA+ cells. PFM and milk administration did not increase the production of this cytokine which was diminished after malnutrition.
Considering the importance of maintaining the gut immune homeostasis by regulatory cytokines such as IL-10 and IL-4, these two cytokines were also analyzed. The results showed that IL-4 and IL-10 secreting cells were decreased in MC mice. The re-nutrition with PFM increased the production of these cytokines, being significant IL-4 increases. It is also important to remark that the re-nutrition with milk significantly increased the number of both IL-4+ and IL-10+ cells showing a higher down regulation than the response obtained with the PFM or its supernatant.
The increases observed for the number of macrophages in the intestinal tissues after re-nutrition led us to analyze the phagocytic activity of macrophages isolated from other sites such as spleen and peritoneum and due to the increased phagocytic activity observed after 5 days of re-nutrition with PFM and BFS (Figure 4), the adjuvant effect of these products on the systemic immune response was evaluated.
It was observed that malnutrition affected the systemic immune response against the OVA antigen (Figure 5). The three re-nutrition diets increased this specific response in the serum of the mice when they continued receiving the respective diet after antigen immunization, although the concentration of anti-OVA IgG was even high when the PFM was stopped. These results lead us to determine also the phagocytic activity of macrophages isolated from spleen and peritoneum after immunizations way of explaining the antibody increases (Figure 4). The results showed that the macrophage activity increased only previous to the immunization in the re-nourished mice. This observation indicates that the increase on the antigen presentation was important at this time point, the day of the immunization, before the adaptive response continued. However, the PFM administration was the only one that increased anti-OVA IgG in the serum of re-nourished mice even when they did not receive a specific diet after immunization. For milk and BFS, the continuous administration, after antigen immunization, was necessary to maintain a good antibody response.
The importance of the host immune system to confront different enteropathogenic infection and how probiotic administration can enhance this immune response exerting a protective effect was previously reported using animal models [14, 39]. The changes observed in the immune response of mice re-nourished with PFM or its BFS led us to study their appliance in the defence against a common enterophatogenic infection. The study of S. Typhimurium infection in our model of malnutrition showed that only the animals that received PFM as a re-nutrition diet prior and after pathogen challenged decreased significantly the translocation of this bacterium to the different organs studied, compared to the others test groups. This finding agrees with previous results where the administration of the same PFM to well nourished adult or newborn mice (after weaning) exerted a protective effect against this pathogenic infection with improvement of the gut and systemic immune response in the mice administered PFM .