Nutrition is important to meet the body’s energy and structural needs, as well as to improve the functionality of various body systems. Among these, immune function can be improved by several foods (1). On the other hand, although everybody is aware of the healthy influence of regular moderate-intensity training, physical exercise is often confined to one-time exercise bouts with no prior adequate training. In fact, participation in running races has become increasingly popular during recent decades. Following intensive sports events, a higher rate of mucosal infections, such as upper respiratory tract infections (URTIs) (2), and the appearance of gastrointestinal symptomatology have been reported (3, 4). A single bout of intense exercise is able to alter the counts of circulating leukocytes (5), especially T lymphocytes, probably due to both a redistribution within lymphoid and non-lymphoid organs (6) and an increased apoptosis of highly differentiated T cells (7). While the spleen seems to act as the main donor organ of T lymphocytes in the redistribution induced by exercise, Peyer’s patches (PPs) are one of the main receiving compartments (6, 8). T lymphocyte function can also be impaired following acute exercise, for instance a decreased proliferation ability (9) and a reduced type 1 T helper (Th1) cell cytokine production have been reported (10). Changes in salivary immunoglobulin (Ig) A, as a marker of mucosal immunity, have been widely studied in the field of exercise immunology (10). Although most of the studies point to a decrease in salivary IgA following intense chronic exercise (11), there is no consensus on the impact of acute exercise on this variable (12). Moreover, acute exercise alters intestinal motility, villi structure (13) and increases the gut permeability (14), probably through the alteration of tight junction (TJ) protein expression, redistribution or phosphorylation state (15). Exercise also modifies the intestinal microbiota composition and function (13, 16), as well as its crosstalk with the host, since, among other changes, a decrease in the cell-surface toll like receptor (TLR) expression has also been described (17).

Cocoa consumption has shown antioxidant (18), anti-inflammatory (19) and immunomodulatory (20) properties. These effects are mostly attributed to cocoa flavonoids, mainly (-)-epicatechin, catechin and procyanidins (21), although cocoa also contains other bioactive compounds such as theobromine and a high proportion of fiber (22). Clinical evidence suggests that cocoa and its flavonoids may be useful for the nutritional management of cardiovascular diseases, obesity and diabetes, among other chronic diseases (23), as well as for the prevention of allergies (24). Moreover, although cocoa does not show a clear ergogenic effect in humans, unlike other polyphenol sources (25), its regular intake can be helpful to avoid the intensive exercise-induced oxidative stress (18). On the other hand, preclinical studies have demonstrated cocoa’s ability to modulate both systemic and intestinal immune function (20). In this regard, cocoa intake in rats favors the Th1 response, modulates the mucosal IgA secretion and induces beneficial changes in the composition and function of the gut microbiota, for instance increasing the production of short chain fatty acids (SCFAs) and inducing a differential pattern of TLR gene expression (20, 22, 26). In addition, polymeric cocoa polyphenols are not absorbed in the small intestine and are metabolized in the colon by the gut microbiota, generating secondary bioactive metabolites that are then bioavailable and may exert even greater effects than the original polyphenols (27). Both the cocoa polyphenols and their secondary metabolites can modulate the microbiota composition and function (22, 27). This, together with the high percentage of dietary fiber present in cocoa, can explain the prebiotic properties that have been attributed to cocoa (22, 27).

Considering this background, we hypothesize that cocoa intake can prevent the intestinal and immune alterations induced by a single intense exercise bout. Hence, we aimed to evaluate the effect of a cocoa-enriched diet on the physical performance, the cecal microbiota and the mucosal immune system of rats submitted to high-intensity acute exercise, as well as to elucidate the involvement of cocoa fiber in such effects.

In previous studies, we have demonstrated the immunomodulating (32) and antioxidant properties (31, 39) of cocoa enriched-diets, as well as the prebiotic effects (22) of cocoa fiber in both healthy rats (22, 32) and experimental models of allergy and inflammation (31, 34, 39, 40). However, there are no studies focused on these cocoa properties in a situation of excessive oxidative stress, such as in an acute intensive exercise. In the current study, we aimed to evaluate the influence of a single intense exercise bout and a 25-day cocoa-enriched diet in rats on their cecal microbiota and mucosal immune system. Likewise, we aimed to elucidate the involvement of cocoa fiber in such effects. We found that a single intense exercise bout increased the fecal humidity and, although no change in the cecal microbiota composition was observed, a reduction in some SCFAs, such as propionic acid and a trend to decrease the proportion of valeric acid was found. Moreover, a single bout of intense exercise decreased the IgM content in salivary glands and the proportion of Tγδ CD8αα cells in PPs, whereas increased the percentage of Th lymphocytes in MLNs, and raised the gene expression of the pIgR in the intestinal wall. The cocoa-enriched diet (C10), which did not modify the exercise performance or prevent the exercise-induced increased fecal humidity, was able to increase the small intestine weight and the cecal bacterial content, although it induced a lower proportion of IgA-coated bacteria and Clostridium coccoides/Eubacterium rectale, changes that could be due to other bioactive compounds rather than fiber because the CF diet, which contained the same amount of cocoa fiber and no polyphenols or methylxanthines, did not induce these changes. However, cocoa diet, due to its fiber content, avoided the decrease in cecal propionic acid but not that in valeric acid induced by exercise. Moreover, diets containing cocoa or cocoa fiber alone reduced the production of isobutyric acid. In the mucosal Igs, the cocoa diet decreased the salivary IgM and IgA contents. These changes could be explained again by the effect of other cocoa bioactive compounds rather than fiber, since the CF diet did not produce the same effect. In PPs, 25-day cocoa diet caused some changes in lymphocyte composition, partially due to its fiber content, and prevented the decrease in the proportion of Tγδ CD8αα cells, probably due to other bioactive compounds, induced by exercise. Likewise, the cocoa diet, partially because of its fiber content, reduced the proportion of Th cells but increased that of Tγδ and Tγδ CD8αα cells in MLNs, and prevented the increase in NKT cell percentage induced by exercise in this compartment. With regard to the functionality of MLNLs, reduced the production of the proinflammatory IFN-γ. Finally, after 25 days, cocoa diet, but not cocoa fiber, decreased the FFAR3 and ZO2 gene expression, whereas both diets enhanced two-to threefold that of ZO1. With regard to cocoa fiber, on its own it prevented the increase in fecal humidity, raised the production of SCFAs and the TLR7 gene expression, and avoided the decrease in salivary IgM induced by exercise.

Focusing on exercise performance, the cocoa diet did not improve it, which agrees with most of the human studies using cocoa polyphenols (25). Even so, some preclinical studies have reported that (–)-epicatechin and (–)-epicatechingallate, both flavanols present in cocoa, exert ergogenic effects in rodents (41–43), while others reported no changes (44) or even a decrease in time-to-exhaustion after a 6-month dietary intervention (45). The controversy in these results may be due to the dosage of cocoa or its polyphenol content, as well as the exercise protocol applied.

During the exhaustion test, we measured the BW loss to record the level of dehydration, as well as the total fecal mass produced and its water content, which allowed us to approach changes in gastrointestinal (GI) motility. In line with current literature (4, 46), we found that acute exercise increased fecal humidity which may be due to a faster GI motility. This fact was prevented by the intake of the cocoa fiber diet. The current consensus is to avoid a high intake of fiber 24–72 h before and during intensive exercise to prevent a consequent diarrhea (47), nevertheless, the outcome may depend on the kind of fiber, its source and the ratio of soluble and insoluble fiber consumed. Although a more specific method to evaluate changes in GI motility may be applied in future research, our results suggest that the proportion of cocoa fiber tested exerts protective effects against exercise-induced increased GI motility (48, 49). In addition, the involvement of cocoa components in the crosstalk between neurons, immune cells, and microbes, also known as the intestinal neuro-immune axis (50) also deserves to be studied in depth to ascertain the precise mechanisms involved.

In line with this and with regard to the cecal microbiota composition, acute exercise did not induce changes in the amount of total bacteria, the proportion of those bound to IgA, Clostridium coccoides/Eubacterium rectale and Lactobacillus/Enterococcus percentages or the cecal pH. Surprisingly, these results disagree with the acute exercise-induced decrease in Lactobacillus proportion and the increase of that of Clostridium found by Chaves et al. using a similar experimental design (13). Nevertheless, the available evidence is quite controversial, since other authors reported a higher Lactobacillus proportion in both intensively and moderately exercised rats (51, 52). Further studies may clarify the relationship established between the degree of the exercise’s intensity and its impact on gut microbiota composition in a more exhausting study. On the other hand, the intake of the cocoa-enriched diet tripled the amount of total bacteria present in cecal content, which could contribute to the higher intestinal weight found in these animals. Moreover, cocoa intake lowered the proportion of bacteria bound to IgA as well as the percentage of Clostridium coccoides/Eubacterium rectale, without modifying that of Lactobacillus/Enterococcus, results that are in line with previous studies (26). This influence of cocoa in cecal content was not due to its fiber content because the fiber enriched diet did not produce the same changes. Therefore, these changes may be attributed to the other bioactive compounds present in cocoa, such as polyphenols and methylxanthines. In line with this, we have previously reported a similar increase in the amount of total bacteria and a decrease in the proportion of them bound to IgA due to a 4-week hesperidin supplementation, which is one of the main polyphenols found in citrus fruits (53). Theobromine, the predominant xanthine found in cocoa, also seems to play an important role in lowering the proportion of IgA-coated bacteria (54). The impact of these changes depends on the kind of bacteria that was bound to IgA, since IgA can bind pathogenic bacteria to neutralize them or it can transport commensal bacteria to the intestinal epithelium. Nevertheless, it seems that IgA has a higher affinity to inflammatory bacteria (35, 55), since increased levels of this have been found in inflammatory bowel disease patients (35, 56). However, further studies may elucidate the involvement of cocoa’s polyphenols and other bioactive compounds in cocoa’s impact on cecal microbiota composition.

When assessing the impact on microbiota, it is important to take into account not only the composition but also its function. SCFAs are the main bacterial metabolites of the fermentation of dietary fiber and proteins and they exert an important role as metabolic regulators. Previous research has associated an increased gut-derived SCFA production with a better insulin sensitivity, a reduced inflammation (57) and even an improved exercise performance (58, 59), this last factor being associated with its interaction with skeletal muscle. It has been reported that the regular practice of moderate intensity exercise increases the production of butyric acid (60, 61). Moreover, some authors found an inverse correlation between the fecal concentration of propionic and acetic acids and the exercise performance in a cardiorespiratory fitness test (measured as peak oxygen uptake, VO₂ peak) (60), whereas other studies reported higher propionic and acetic acid levels in elite professionals rugby players (61) and in lean subjects submitted to a 6-weeks endurance-based exercise intervention (62). In spite of these studies focused on long-term exercise, the impact of acute intensive exercise on SCFA production remains unclear. Our present results indicate a lower propionic acid concentration in cecum due to acute exercise, which partially agree with the findings of Estaki et al. (60). Both cocoa- and cocoa fiber- enriched diets attenuated this exercise-induced decrease in propionic acid, and even the CF groups also had higher levels of cecal acetic acid. This agrees with previous studies reporting that a 3-week diet with cocoa fiber increased the cecal concentration of acetic, propionic and butyric acids (22). A recent preclinical study pointed out the importance of acetic acid as an energy substrate during endurance exercise, since a continuous acetic acid infusion was able to restore the lost exercise capacity of mice treated with antibiotics (63). However, we did not observe a higher exercise performance linked to fiber diet. On the other hand, we also found that exercise tended to decrease the cecal content of valeric acid, which together with the branched SCFA isovaleric and isobutyric acids are considered putrefactive SCFAs (pSCFAs). These pSCFAs are products of the bacterial fermentation of proteins and are associated with the generation of other concomitant products that may be detrimental to the colonic epithelium (64). Then, the exercise-induced decrease in valeric acid might be beneficial for the host. Moreover, cocoa-enriched diet, partially because of its fiber content, induced a decrease in the concentration of valeric and isobutyric acids. These results are in line with previous studies that reported a negative correlation between dietary fiber consumption and the levels of pSCFAs (65). Some of the beneficial effects of SCFAs to the host, such as their anti-inflammatory properties, occur through the interaction with the two specific SCFAs receptors FFAR2 (GPR43) and FFAR3 (GPR41) (66, 67). Here, although the activation status of these receptors was not analyzed, a bout of intense exercise did not change the small intestine gene expression of FFAR2 and FFAR3. However, cocoa-enriched diet, but not cocoa-fiber diet, decreased the gene expression of FFAR3, which is mainly activated by propionic and butyric acids and is involved in energy balance maintenance, noradrenaline release from sympathetic neurons and the production of anorectic hormones like the peptide YY, among other functions (68). Then, these changes in the intestinal gene expression of FFAR3 may be somehow associated with the lower BW gain observed in C10 animals (and not in CF rats), here as well as in other studies from our research group (22, 33, 36) and from other authors (69, 70). Moreover, the anti-obesity potential of cocoa has already been discussed in several reviews (21, 71–73). Although we previously attributed the lower BW gain of rats fed with cocoa to the impact of theobromine and flavonoids on lipid and glucose metabolism, fat deposition and even the gut microbiota composition (22, 74), the current results suggest that a decrease in FFAR3 intestinal expression may also contribute to this multifactorial effect.

The intestinal epithelium, by means of the enterocytes, the tight junction proteins, the mucus layer and immune cells, plays an essential role as a barrier that regulates the permeability of nutrients, water and ions, and prevents the invasion of pathogenic bacteria, and together with the intestinal microbiota contributes to the maintenance of the intestinal homeostasis. Numerous studies have shown that intensive exercise impairs the expression and the phosphorylation status of the TJ proteins, disrupting the integrity of the intestinal epithelial barrier and leading to an increased gut permeability (15, 75). This impairment has been associated with the hyperthermia (76), dehydration (77) and the stress (78) induced by intensive exercise. Lambert et al. associated an increased gastric and small intestinal permeability, with a 1.5% of BW loss in trained distance runners after running without fluid for 60 min at 70% VO₂ max (77). The level of dehydration described is like that we found in REF animals. Although we did not assess changes in intestinal permeability per se, we studied the effect of exercise on the small intestine gene expression of TJ proteins, which are essential for the regulation of gut paracellular permeability. The acute exercise protocol applied in the current study did not induce changes in Ocldn, Cldn4, ZO1 or ZO2 small intestine gene expression in REF animals. These results are in line with those of Chaves et al. (13), showing no changes in cldn1 and ZO1 intestinal expression of rats 12 h after performing acute exercise without prior training, although they observed a larger colonic villi interspace in runner rats. Nevertheless, Ducray et al. reported a lower ZO1 intestinal protein expression in rats 4 h after exhaustion running (79). On the other hand, cocoa diet, due to its fiber content, substantially increased the gene expression of ZO1, which agrees with studies recently published (80) and could be somehow associated with the protective effect of cocoa fiber on the increased GI motility induced by acute exercise. Taking together with the results regarding microbiota and intestinal epithelium, our results here suggest that cocoa consumption can improve intestinal health, especially by means of increasing cecal SCFA production and ZO1 expression, which could be useful not only to prevent the alterations induced by intensive exercise, but also in situations with an even more impaired intestinal barrier function, like some gastrointestinal disorders.

Whereas the study of exercise’s impact on gut microbiota has just gained importance in recent years, its effect on IgA production has been widely studied. IgA is the main Ig isotype in the mucosal compartment and plays many essential roles, such as neutralization of pathogens and toxins, antigen sampling and blocking excessive commensal and pathogenic bacteria (55). It has been reported that regular sessions of moderate intensity exercise increase salivary IgA secretion whereas acute intensive exercise or prolonged periods of exhausting exercise might decrease it, explaining the higher rate of mucosal infections observed in athletes (81). Here, our protocol of acute exercise did not induce significant changes in IgA concentration in SMGs or in the intestinal compartment, however, exercised rats had a lower concentration of salivary IgM, which has been described by others to be both decreased and increased after intensive exercise (12, 82). It must be taken into account that IgM is synthetized more quickly than IgA (83) and then it could be faster affected by an acute exercise. Moreover, it is important to clarify that whereas most of the published articles assessed changes in salivary secretory IgA, here we quantified the total concentration of IgA and IgM in submaxillary salivary gland homogenates, which may indicate changes in the ability to produce these antibodies, but not in the transcytosis and exocytosis processes. In further studies, we will ascertain the impact of acute exercise on salivary IgA and IgM concentrations, as well as how it affects their synthesis, exocytosis and transcytosis. Although the cocoa fiber diet, on its own, prevented this decrease in salivary IgM, the cocoa-enriched diet decreased it, both in the control and in exercised animals. These results are in line with results reported in healthy non-exercised rats (32). Moreover, in agreement with previous studies (36), the cocoa-enriched diet decreased the IgA concentration in SMGs. Therefore, cocoa diet, mainly due to bioactive compounds other than fiber, attenuated salivary Ig content. With regards to the intestinal compartment, animals from the CF and C10 groups had a lower concentration of IgA in GW after running.

In addition, to shed some light on the influence on mucosal immunity, we have studied the lymphocyte composition of two compartments of the gut-associated lymphoid tissue (GALT): the PPs and the MLNs. The GALT is the largest lymphoid tissue in the body and comprises the organized or inductor GALT, which includes PPs, MLNs and isolated lymphoid follicles, and the diffuse or effector GALT, which is formed by scattered intraepithelial (IEL) and lamina propria (LPL) lymphocytes. IgA-secreting plasma cells are mainly generated in PPs (84). It has been described that intense exercise, due to adrenergic mechanisms, induces a mobilization of splenic lymphocytes, mainly T cells, into secondary lymphoid organs such as PPs (6). In the current study, although we did not quantify the total number of cells, we did not observe any exercise-induced substantial change in the proportion of T and B cells in PPs, although it modified the proportion of the Tγδ cell subsets, favoring the presence of the Tγδ CD8αβ cells over that of Tγδ CD8αα cells in PPs. Tγδ CD8αβ cells have cytotoxic potential and secrete inflammatory cytokines, such as IFN-γ and TNF-α, that could contribute to the exercise-induced alterations in gut permeability that have been reported in other studies (37). On the other hand, the cocoa-enriched diet prevented the changes in Tγδ cell subsets in PPs, although it increased, as already described (85), the total proportion of Tγδ cells. These changes are not attributed to cocoa fiber content and may be due to other bioactive compounds such as polyphenols and methylxanthines. Nevertheless, cocoa diet, as already reported (85) increased the relative proportion of PPs Tc cells at the expense of Th cells, which could be partially associated with its fiber content.

The MLNs, together with the PPs, play an important role in initiating local immune responses in the gut (86). We found that, in this lymphoid organ, acute exercise induced an increase in NKT and Th cell proportions while it decreased that of Tc cells. These results agree with the reported higher Th/Tc cell ratio in MLNs of rats submitted to a 5-week intensive training (37). The results obtained here demonstrated that a single bout of exercise is able to change MLNLs proportions similarly to a 5-week training, probably due to differential mobilization of Th and Tc cells, but, on the contrary, while a 5-week training decreased the proliferative ability of T cells, the acute exercise applied here did not modify it. To our knowledge, there is no more available evidence on the effect of exercise on MLNs cell composition. Other authors have focused on brachial, axillary and submandibular lymph nodes, among others, but the results are not clear enough to draw any consistent conclusion (87). Cocoa fiber consumption, but not cocoa-enriched diet, prevented the changes induced by acute exercise in Th cells, suggesting the role of cocoa polyphenols or methylxanthines in counteracting the effect of cocoa fiber. On the other hand, both cocoa and cocoa fiber diets increased the proportion of Tγδ cells and decreased that of NKT cells. In the case of cocoa fed control animals, the higher proportion of Tγδ cells can be attributed to an increased proportion of Tγδ CD8αα cells, in agreement with a previous study (34).

Apart from these changes in T cell subset proportions in MLNs, acute exercise also modified their functionality, increasing the secretion of TNF-α, which could be explained by the cortisol released that occurs immediately after intensive exercise (29, 88, 89). Unlike what we previously observed in chronically intensively trained rats (29, 37), acute exercise did not manage to modify IL-2, IFN-γ, IL-4, and IL-6 production. On the other hand, although most of the studies reported a lower TNF-α production immediately after intensive exercise, normally involving a prior intensive training, due to the release of cortisol and noradrenaline (10, 90, 91), other authors have found no changes (13) or even an increased TNF-α production (92). Here, both acute exercise and cocoa fiber-enriched diet increased TNF-α secretion by in vitro stimulated MLNLs, in disagreement with previous studies showing an inhibitory effect on this cytokine production by cocoa consumption (20). Lastly, although our model of acute intensive exercise did not modify the IFN-γ levels released by stimulated MLNLs, the decrease induced by cocoa in the secretion of this proinflammatory cytokine could be useful to attenuate the substantial increase observed after a chronic intensive running training (37).

In conclusion, our results indicate that the intake of cocoa in the days prior to an intense bout of exercise can partially prevent the alterations induced by it. Some of the changes produced by the cocoa diet on the intestine can be attributed to its fiber content, but other changes denote interaction between the phenolic compounds and methylxanthines, which is worth studying depth in the future. Moreover, further research might ascertain whether a more intensive exercise or a sustained exercise can exacerbate the observed alterations, and assess the potential protective effect of cocoa fiber in an even more impaired immune system.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The animal study was reviewed and approved by Ethical Committee for Animal Experimentation of the University of Barcelona.

Under the supervision of MM-C and FP-C, PR-I completed these studies as part of her Ph.D. thesis. PR-I, MM-C, and FP-C contributed to the conception and design of the project. PR-I, MM-C, MR-L, AF, and MC-B performed the experimentation. PR-I conducted the literature search, the acquisition, analysis of data, and wrote the first version of the manuscript. All authors contributed to the revision of the manuscript, read, and approved the final work.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.