Potential of plant polyphenols to combat oxidative stress and inflammatory processes in farm animals

Hepatic acute-phase response

One important secondary response to inflammation is the acute-phase response (APR) which is mainly triggered by proinflammatory cytokines such as interleukin (IL)-6, tumour necrosis factor-α (TNF-α) or IL-1β. APR is characterized by an increased production of more than 200 acute-phase proteins (APP), mainly in the liver (Venteclef et al., 2011). APP play major roles in several aspects of systemic reaction to inflammation, such as opsonization of pathogens, scavenging of toxic substances and the overall regulation of different stages of inflammation (Ceciliani et al., 2012). While the concentrations of APP such as haptoglobin, ceruloplasmin, serum amyloid A and C-reactive protein in the blood are very low under healthy condition, their concentrations are greatly elevated during systemic inflammation (Bradford et al., 2015). Therefore, concentrations of APP have gained widespread acceptance as markers of inflammation (Ceciliani et al., 2012). While APP play a central role in restoring tissue homoeostasis during the early phase of inflammation, the production of several other proteins typically secreted by the liver such as albumin, apolipoproteins, transferrin or retinol-binding protein is reduced during inflammation. The downregulation of the hepatic synthesis of these proteins, which are denominated as negative APP, increases the pool of free amino acids which are utilized partially for gluconeogenesis and the production of APP (Ceciliani et al., 2012).

Hypothalamic inflammation

The production of cytokines is not only important for the regulation of immunological responses against pathogens or toxins. Cytokines have also profound effects on several metabolic pathways that are directly linked with performance of farm animals. Recent studies have shown that circulating cytokines and inflammatory mediators increase production and release of cytokines within the mediobasal hypothalamus, creating a paracrine inflammatory milieu which initiates alterations in the activity of neuronal populations that regulate appetite, body temperature and metabolic programs controlling body mass and energy homeostasis (Burfeind et al., 2015). Metabolic alterations that are regulated in the hypothalamus during a proinflammatory condition mainly aim to shift energy and amino acids away from stores towards metabolic responses that support the immune system. In this respect, accelerated muscle protein degradation is one key event. Increased proteolysis in skeletal muscle during inflammation is mediated by an activation of the ubiquitin proteasome system (UPS). Recent studies suggest that the activation of the UPS is mainly due to increased concentrations of glucocorticoids as a result of the activation of the hypothalamic–pituitary axis in the brain during proinflammatory conditions (Karrow, 2006; Burfeind et al., 2015). Besides glucocorticoids, cytokines such IL-1, IL-6 or TNF-α are able to directly activate the UPS in skeletal muscle (Bowen et al., 2015). Cytokines such as IL-1 moreover have been shown to inhibit the anabolic effects of insulin on skeletal muscle (Klasing and Johnstone, 1991). The aim of both muscle protein degradation and inhibition of muscle protein synthesis is to supply the liver with amino acids available for the synthesis of APP. During inflammation, the uptake of amino acids into the liver is increased, and it has been suggested that at least 60% of the amino acids required for synthesis of APP are deriving from proteolysis in skeletal muscle (Andus et al., 1991; Johnson, 1997). Amino acids deriving from muscle are moreover utilized for gluconeogenesis – a process which is stimulated by increased glucocorticoid secretion – to supply immune cells with fuels. Many immune cells, such as neutrophils and macrophages, rely heavily on glucose to meet their increased metabolic demands during inflammation (Bradford et al., 2015). An increased rate of lipolysis, induced by an increased secretion of catecholamines – a process that is also controlled in the hypothalamus –, is another metabolic alteration which aims to supply the body with metabolic fuels to meet the extraordinary energetic costs of increasing body temperature (fever) and expanding the immune responses (Thaler et al., 2010).

One key event occurring during inflammation, which is negatively linked with animal performance, is the induction of anorexia. Systemic inflammation, triggered for example by injection of lipopolysaccharides (LPS), causes an activation of neurons that produce the anorexigenic peptides pro-opiomelanocortin and cocaine- and amphetamine-regulated transcript, while the secretion of the orexigenic agouti-related peptide and neuropeptide Y is reduced, overall leading to a reduced appetite (Burfeind et al., 2015). Besides effects on the release of orexigenic and anorexigenic peptides in the brain, central cytokines moreover have profound effects on gastrointestinal secretions and motility. Changes in this respect resemble those of satiated animals, such as a decrease in gastric secretion or an inhibition of gastric motility (Johnson, 1997). The loss of appetite in sick animals is not a maladaptive response or the effect of debilitation, but rather an organized, evolved strategy that facilitates recovery. Systemic inflammation causes a shift of the anabolic–catabolic balance towards catabolism meaning that animals have a lower requirement of energy and nutrients for anabolic processes. Thus, anorexia in sick animals can be considered as a means to avoid intake of nutrients that are unusable in their current physiological state (Johnson, 1998). Indeed, it has been shown that survival of infected animals is positively related to anorexia and weight loss, at least in the short term (Murray and Murray, 1979). An overview of hormonal and metabolic changes triggered by inflammation is given in Fig. 1.

Effects of inflammation on performance of pigs

Several studies have shown that typical responses to LPS challenge in pigs are reduced feed intake, increase in body temperature (fever), strong increases in the concentrations of proinflammatory cytokines, APP and cortisol and reduction in the plasma concentration of insulin-like growth factor (IGF)-1. Reduced feed intake, increased energy requirement for the production of fever and hormonal changes shift the metabolism into a more catabolic state which results in reduced body weight gains and an impairment of the gain-to-feed ratio (Webel et al., 1997; Liu et al., 2003; Campos et al., 2014; Wyns et al., 2015). Increases of plasma non-esterified fatty acids (NEFA) and urea concentrations after LPS challenge are indicative of an increased lipolysis in adipose tissue and an increased protein breakdown in muscle (Johnson, 1997).

In pigs, besides several diseases associated with systemic inflammation, the weaning phase represents one important stage in which a local inflammation in the small intestine is occurring, which is called weaning-associated intestinal inflammation. This inflammation is characterized by an increased production of cytokines and adverse changes in intestinal morphology, including reduced villus height, increased villus width, increased crypt depth and reduced absorptive capacity and brush border enzyme activity (McCracken et al., 1995, 1999; Pié et al., 2004). It has been originally suggested that these effects are mainly due to a depressed feed intake (McCracken et al., 1995, 1999). Recent studies, however, have shown that gut function is adversely affected by an activation of the hypothalamic–pituitary–adrenal axis which is caused by weaning stress conditions, including psychosocial (maternal deprivation), immunological, infectious, metabolic and nutritional stress. Downstream events of hypothalamic–pituitary–adrenal axis activation include an increased secretion of corticotrophin-releasing factor and glucocorticoids, and an activation of intestinal mast cells, all of which might be involved in disruption of the intestinal barrier (Smith et al., 2010; Pohl et al., 2015). The compromised intestinal barrier allows the passage of luminal antigens into the lamina propria where an inflammatory response can be initiated. Villus atrophy in this stage might be caused by an activation of matrix metalloproteinases which are induced by proinflammatory cytokines (McCracken et al., 1999). A disturbance of the barrier integrity moreover favours the transepithelial passage of micro-organisms and thus promotes bacterial infections and diarrhoea. Noteworthy, adverse effects of weaning stress on gut function are not limited to the weaning phase. Rather, weaning stress triggers long-term defects in intestinal barrier function (Smith et al., 2010). Therefore, the weaning-associated intestinal inflammation process is negatively linked with health and growth performance of piglets, and controlling this process is a challenge in managing post-weaning health and optimum growth performance (Pié et al., 2004).

Recent studies have shown that lactation is another metabolic condition which is associated with the development of inflammation in the liver of pigs (Rosenbaum et al., 2012a,b). It is likely that the induction of inflammation in the liver of lactating sows – which was also associated with the development of stress of the endoplasmic reticulum (ER stress), a condition triggered by inflammation – might be due to oxidative stress caused by the high metabolic activity or the occurrence of infectious diseases such as mastitis or metritis which are commonly observed in sows after parturition (Gessner et al., 2015a,b). Physiological consequences of inflammatory conditions in lactating sows have been less investigated so far. However, with respect to animal welfare and performance, prevention of proinflammatory conditions in lactation sows should be given more consideration.

Effects of inflammation on performance of poultry

Like in pigs, LPS-induced systemic inflammation in broilers commonly leads to a strong increase in the production of various proinflammatory cytokines and to an increase in body temperature (De Boever et al., 2009, 2010; Zhang et al., 2010; Shen et al., 2010). With respect to animal performance, a decrease in feed intake, a reduction in body weight gain and an increase in the feed conversion ratio were typically observed in broilers subjected to LPS injection (Korver et al., 1998; Shen et al., 2010; Tan et al., 2014). LPS injection in broilers caused also a strong increase in plasma concentration of cortisol, shifting the metabolism in a more catabolic state (Shini et al., 2008). Broilers injected with LPS moreover had increased relative weights (organ weight/body weight) of liver, spleen and intestine (Korver et al., 1998).

Effects of inflammation on performance of dairy cattle

Several studies in the last years have shown that most of the high-yielding dairy cows experience a proinflammatory condition during the peripartal phase (Bertoni et al., 2008; Bradford et al., 2015). The proinflammatory condition is caused by various stimuli, such as LPS, cytokines and ROS released in the course of infectious diseases, such as mastitis or metritis (Bradford et al., 2015). Although this inflammation is mostly of subclinical nature, it is of impact for health and performance of cows during early lactation. Studies with dairy cows, which were subjected to systemic inflammation induced by injection of either LPS or proinflammatory cytokines, typically found a decrease in food intake (Yuan et al., 2013) and a strong reduction in milk yield (Kushibiki et al., 2003; Trevisi et al., 2009; Yuan et al., 2013). In addition to the impairment of animal′s performance, it was reported that the induction of inflammation was associated with an increase in the concentration of NEFA as a result of increased rate of lipolysis in adipose tissue and increased concentration of β-hydroxybutyrate as a result of a stimulation of ketogenesis in the liver and the development of a fatty liver (Trevisi et al., 2009; Bradford et al., 2015). Moreover, cows subjected to TNF-α-induced inflammation had an increased incidence of various diseases, such as ketosis, mastitis, respiratory diseases, metritis and milk fever (Yuan et al., 2013). Conversely, inhibition of the proinflammatory condition during the peripartal phase in dairy cows by early post-partum treatment with non-steroidal anti-inflammatory drugs led to a significant increase in milk yield over the whole lactation period and caused a reduction in somatic cell count in milk being indicative of an improved udder health (Carpenter et al., 2016). According to these findings, investigations aiming to prevent the proinflammatory condition during the peripartal phase, either by optimization of feeding regimes pre-partum (Khan et al., 2015; Zhou et al., 2015; Vailati-Riboni et al., 2016) or by administration of potential antiinflammatory feed supplements (Gessner et al., 2015a,b; Hashemzadeh-Cigari et al., 2015; Winkler et al., 2015), are of great relevance.

Structure and classes of polyphenols

Polyphenols are secondary plant metabolites synthesized by plants to provide protection against invasive pathogens, such as bacteria, viruses and fungi, and to prevent damage to DNA and photosynthetic apparatus due to ultraviolet radiation. The polyphenols comprise a large group of more than 8000 different compounds with the phenol ring being the common structural feature. According to the number of phenol rings and the structural elements that bind these rings to one another, the polyphenols can be divided into the flavonoid-type and the non-flavonoid-type polyphenols (Tangney and Rasmussen, 2013).

The flavonoid-type polyphenols comprise the largest group of polyphenols with more than 4000 compounds identified and share as a common structure two benzene rings connected by three carbon atoms forming an oxygenated heterocycle. Based on the type of the heterocycle, the following flavonoid subclasses can be distinguished: flavonols, flavones, flavan-3-ols, flavanones, anthocyanins and isoflavones (Beecher, 2003). Examples of frequently studied flavonoids are the flavonols quercetin and myricetin, the flavones orientin, vitexin and homoorientin, the flavan-3-ols catechin, epicatechin, epigallocatechin and their gallate esters, the flavanone naringenin, the anthocyanin cyanidin and the isoflavones genistein and daidzein. Isoflavones have an oestrogen-like structure and thus are able to bind to oestrogen receptors (Uifălean et al., 2015). Almost all dietary flavonoids exist in glycosidic form, mostly in O-glycosidic form, but C-glycosylated flavonoids, particularly flavones, are also abundant in plants (Xiao and Jiang, 2015).

Phenolic acids, which comprise one group of non-flavonoid-type polyphenols, are frequent in cereals and can be subdivided into cinnamic acids, such as ferulic, caffeic, coumaric and sinapic acid, and the less abundant hydroxyl benzoic acids, such as gallic and vanillic acid. Typically, the cinnamic acids are bound to arabinoxylans of the plant cell wall which impairs their bioavailability (Andreasen et al., 2001). A second group of non-flavonoid-type polyphenols are the lignans, which are also diphenolic compounds with a 1,4-diarylbutane structure formed by dimerization of two cinnamic acid residues, and several representatives, such as secoisolariciresinol, pinoresinol and syringaresinol, are considered as phytoestrogens due to their oestrogen agonist and antagonist properties (Kitts et al., 1999). The last group of non-flavonoid-type polyphenols are the stilbenoids, from which resveratrol is the most famous compound due to its lifespan prolonging effects (Valenzano et al., 2006) and which have two phenol rings connected by a two-carbon methylene bridge.

Occurrence of polyphenols

Polyphenolic compounds occur in almost all plants, but their distribution at the tissue level and the cellular or subcellular level is not uniform. For instance, the outer layers of plants contain higher polyphenol levels than the inner layers or insoluble polyphenolic compounds are associated with the cell wall, whereas soluble ones are found in the cell vacuoles (Wink, 1997). Moreover, the concentrations and the proportions of polyphenolic compounds in plants are greatly affected by ripeness at the time of harvest, edaphic factors such as soil type, sun exposure, air temperature and rainfall, processing and storage (Manach et al., 2004). With regard to the latter, it can be generally stated that the concentration of polyphenolic compounds in plant material decreases with increasing storage time and when high temperatures are present during storage and processing because of the high susceptibility of polyphenols to oxidation. Collectively, these factors explain that the concentrations and the type of polyphenols vary greatly in plants and the plant products (e.g. extracts) manufactured from.

Typical polyphenol-rich plant materials are fruits, vegetables, leaves, legume seeds and beans. Correspondingly, humans regularly consuming black and green tea, tropical fruits and berries, spinach, onion, soybeans, fruit juices and wine have a high intake of polyphenols (Tangney and Rasmussen, 2013). Due to economic considerations, relevant sources of polyphenols for farm animals are agroindustrial by-products from juice, wine and beer making, such as pomace, peels, seeds, stems and brewery waste, and from processing of grains, seeds and nuts, such as hulls (rice, buckwheat, almond) and husks (coconut).

Bioavailability of polyphenols in monogastric animals

The bioavailability of dietary compounds describes the degree to which they become available in intact form to the target tissues following ingestion into the gastrointestinal tract. Generally it can be assumed that the bioavailability of polyphenolic compounds from the small intestine is very low. Based on estimations from Faria et al. (2014), less than 5% to 10% of plant polyphenols are absorbed in the small intestine. This is explained mainly by the fact that almost all polyphenols in plants exist as esters, glycosides or polymers, and many polyphenols are associated with cell wall constituents, such as arabinoxylans, with proteins or with other organic compounds such as organic acids and lipids (Andreasen et al., 2001; Bunzel et al., 2004). On the other hand, the capacity for hydrolysis and liberation from the plant matrix, respectively, by either digestive or microbial enzymes in the small intestine is generally limited due to either lack of appropriate enzyme activity or the low microbial colonization. Enzymatic hydrolysis of esters and glycosides and the release from cell wall constituents is a prerequisite step for the absorption of polyphenols because only the aglycones and few glycosides (e.g. mono- and diglucosides), but not more complex, polymeric polyphenols can be absorbed in the small intestine (Murota et al., 2002; Cermak et al., 2003; Zubik and Meydani, 2003; Manach et al., 2004). A further reason for the low bioavailability of many polyphenols in target tissues following absorption is that polyphenolic compounds are recognized as xenobiotics by the body′s biotransformation system. Thus, polyphenols are extensively biotransformed by classic detoxification pathways (methylation, glucuronidation, sulphation) in enterocytes and liver to a large number of hydrophilic conjugated (methylated, glucuronidated and sulphated) metabolites (Kuhnle et al., 2000; Vaidyanathan and Walle, 2002), which can partially reach blood and tissues, but are also rapidly excreted via urine and bile (Hackman et al., 2008). As the result of low absorption, extensive biotransformation and excretion of metabolites, concentrations of polyphenols such as quercetin in plasma and tissues after supplementation remain relatively low. Even long-term supplementation of polyphenols does not lead to an accumulation in plasma or tissues (Bieger et al., 2008; Reinboth et al., 2010). Moreover, the diverse potential health benefits of polyphenols, such as antiinflammatory, antioxidative, anticarcinogenic, antiatherogenic and immunomodulatory effects, observed in studies with laboratory animals in response to isolated polyphenols and/or polyphenol mixtures cannot be simply ascribed to the polyphenols themselves, but rather to their metabolites originating from biotransformation. This is a significant limitation for the ability to translate results from cell culture studies, in which incubations were performed only with the parent compounds, to the in vivo situation. Moreover, it has to be pointed that there are large sex-related (Mugford and Kedderis, 1998; Renaud et al., 2011; Liu et al., 2013; Ruiz et al., 2013; Dellinger et al., 2014; Prokopec et al., 2015) and species-dependent (Matal et al., 2008; Yamazaki et al., 2011; Helke and Swindle, 2013; Saengtienchai et al., 2014) differences in xenobiotic metabolism which largely explains that the spectrum of polyphenol metabolites, their tissue distribution and concentrations in blood following ingestion of polyphenols or polyphenol mixtures can differ markedly between species and between males and females (Weinert et al., 2012; Margalef et al., 2016). This is a clear hindrance for the ability to transfer data from male-to-female animals and from one species to another one.

The large proportion of polyphenols, which passes through the small intestine without being absorbed (approximately 90% to 95% of ingested polyphenols; Faria et al., 2014), together with conjugated polyphenol metabolites excreted from the liver via the bile enter the colon and are biotransformed with the help of enzymatic activities of the colon microbiota to various polyphenol metabolites. For instance, the conjugated polyphenol metabolites excreted via the bile are deconjugated by microbial glucuronidases and sulphatases into their aglycones, which can be further degraded by microbial enzymes leading to various metabolites (Hein et al., 2008). In the case of degradation of flavonols, flavones, flavanones and flavan-3-ols, for instance, a large number of different metabolites, such as dihydroxyphenylacetic acids, hydroxyphenylacetic acids, hydroxyphenylpropionic acids, dihydroxybenzoic acid, trihydroxybenzoic acid, dihydroxyphenylethanol, dihydroxybenzaldehyde phenylvalerolactones, phloroglucinol, hydrocaffeic acid, phloretic acid, benzoic acid, phenyllactic acid and vanillic acid, are formed (Meselhy et al., 1997; Déprez et al., 2000; Rechner et al., 2004; Hein et al., 2008).

The polyphenol metabolites in the colon can have different fates: one part is absorbed into the circulation after being conjugated once again in the enterocyte and the liver. Once absorbed into the circulation, the bioavailability of polyphenol metabolites in target tissues depends on their binding affinity to albumin, which was shown to differ according to the chemical structure of polyphenolic compounds (Dangles et al., 2001) and thus influences delivery of polyphenol metabolites to tissues and the rate of clearance from the circulation. Another part of polyphenol metabolites can serve as either growth-promoting substrates or antimicrobial substances for bacteria of the colon microbiota. In this context it is worth mentioning that several studies have shown that polyphenol metabolites can modulate the composition of the colon microbiota in a desirable manner due to simultaneously promoting the growth of beneficial bacteria in a pre-biotic-like manner, while inhibiting that of pathogenic bacteria (Axling et al., 2012; China et al., 2012; Queipo-Ortuño et al., 2012; Etxeberria et al., 2013, 2015; Moreno-Indias et al., 2016). This modulation of the colon microbiota is considered a key mechanism underlying the overall health-promoting effects of polyphenols, because the resulting eubiotic microbiota contributes to gut health, intact gut barrier function, and a properly working intestinal immune system, all of which are important to protect the animal from translocation of pathogens or other infectious noxious into the circulation. The remaining part of non-absorbed and non-metabolized polyphenol metabolites is excreted via the faeces. A schematic illustration of absorption and metabolism of plant polyphenols in farm animals is shown in Fig. 2.

Summarizing the above, it can be stated that the bioavailability of plant polyphenols is critically influenced by their chemical structure (e.g. glycosylation, esterification and polymerization) and associations with other plant cell constituents (fibre, protein). Correspondingly, the most abundant polyphenols in plants or plant extracts are not necessarily those with the highest bioavailability. A second important message is that the two-way interaction between polyphenols and colon microbiota, meaning the biotransformation of polyphenols by the gut microbiota to polyphenol metabolites and the modulation of microbiota composition by polyphenol metabolites, is likely a key mechanism explaining many health-promoting effects of polyphenols.

Bioavailability of polyphenols in ruminants

In contrast to the great number of studies with monogastric animals, studies on the bioavailability of polyphenols in ruminants are very scarce. There are, however, few studies which indicate that at least a small part of polyphenols are absorbed from the diet in ruminants. Gohlke et al. (2013a) have recently shown that quercetin aglycon intraduodenally administered in lactating dairy cows is absorbed at a rate similar to that observed in pigs. In dairy cows, it has moreover been shown that the administration of feedstuffs rich in isoflavones (genistein, daidzein) such as soybean meal or red glover silage causes an increase in the concentrations of those isoflavones in blood and milk (Höjer et al., 2012; Cools et al., 2014). In sheep, it was observed that concentrations of different polyphenols, such as epicatechin, in plasma are increasing following the administration of four different polyphenol-rich plant extracts (rosemary, grape, citrus, marigold) via a rumen cannula (Gladine et al., 2007). Those studies indicate that at least a part of the isoflavones present in the diet is protected against degradation in the rumen and is available in the small intestine, the site of absorption of polyphenols. Besides, few reports exist demonstrating that feeding polyphenol-enriched feeding rations to heifers modulates several parameters of rumen fermentation (pH and concentration of fermentation products in rumen fluid) and rumen bacterial community (Balcells et al., 2012; De Nardi et al., 2016). With regard to the latter, it was reported in one study that feeding a flavonoid-rich plant extract increases the numbers of lactate-consuming and propionate-producing bacteria (Balcells et al., 2012) and in another study that a daily dose of a commercial polyphenol-essential oil mixture as part of a total mixed ration enhances the abundance of many taxa belonging to Bacteroidetes, Firmicutes and Tenericutes phyla (De Nardi et al., 2016). These reports indicate that, such as in monogastrics, the composition of the microbiota in the rumen of ruminants is modulated by plant polyphenols.

Effects of polyphenols on oxidative stress and inflammation in studies with pigs

There are several studies which considered the effects of polyphenols on pathophysiological processes of various diseases in pigs as an animal model for humans. For instance, it has been shown that the stilbenoid resveratrol or resveratrol-rich grape extract lowers fat deposition, improves the lipid profile and glucose metabolism and prevents the development of atherosclerotic lesions in the aorta of pigs (Burgess et al., 2011; Azorín-Ortuño et al., 2012). Other studies observed that resveratrol favourably influences risk factors for coronary heart disease and improves myocardial function in porcine models of metabolic syndrome and myocardial ischaemia (Robich et al., 2010; Burgess et al., 2011; Chu et al., 2011).

In contrast to studies with biomedical background, the number of studies dealing with effects of polyphenols on the antioxidant status and inflammation in pigs as farm animal species is relatively limited, with more studies focusing on potential antioxidant effects of polyphenols. One of these studies has shown that supplementation of the flavonol quercetin (at a level of 10 mg/kg body weight/day) is able to increase plasma and liver concentrations of α-tocopherol and reduce plasma lipid peroxidation products in growing pigs fed fish oil in combination with a low dietary vitamin E supply (7 mg/kg diet) (Luehring et al., 2011) indicating that quercetin exerts a tocopherol-sparing effect at a low vitamin E intake. In pigs treated with subcutaneous LPS injection, supplementation of the diet with ampelopsin (at a level of 400 mg/kg diet), which is also known as dihydromyricetin, the most common flavonoid in dry tender stems and leaves, ameliorated the antioxidative capacity in plasma and lowered concentrations of malondialdehyde (MDA) in the liver (Hou et al., 2014). This study suggested that the flavonoid ampelopsin improves the antioxidative status in pigs subjected to inflammation and oxidative stress. In a study with gilts which were treated with zearalenone, a mycotoxin known to induce oxidative stress, supplementation of isoflavones lowered MDA concentration in plasma (Wang et al., 2012). This finding suggests that isoflavones are able to counteract oxidative stress induced by mycotoxin treatment. In contrast to pigs which were either fed diets with insufficient dietary vitamin E levels or subjected to prooxidative treatments, plant polyphenols had less effect on the antioxidative status in pigs without prooxidative treatment. In the study of Augustin et al. (2008), supplementation of a green tea extract rich in catechins did not increase tocopherol concentrations in plasma and various tissues and did not increase the antioxidative capacity in pigs fed a diet with an adequate supply of vitamin E. In another study, supplementation of grape seed and grape marc extract with a total polyphenol content of 60 mg/g at a level of 10 g/kg diet did also not influence concentrations of tocopherols and thiobarbituric acid-reactive substances (TBARS) in plasma and liver and the antioxidative capacity of plasma in piglets (Gessner et al., 2013). In the study of Zhang et al. (2014), supplementation of a basal diet with a mixture of polyphenols extracted from apple, grapes, green tea and olive leaves (at a level of 1 g/kg diet) reduced concentration of MDA in plasma of weaned piglets but had no effect on the total antioxidative capacity and activities of several antioxidant enzymes in plasma. The overall conclusion of the few studies performed so far is that supplementation of polyphenols improves the antioxidative status and ameliorates oxidative stress in pigs subjected to prooxidative treatment. In pigs without prooxidative treatment, polyphenols have either no or only marginal effects on the antioxidant status.

Regarding potential effects of polyphenols on inflammation in pigs, very few studies are available considering the effects of polyphenol-rich plant extracts on the expression of proinflammatory genes in the intestine. In the study from Gessner et al. (2013), the diet containing grape seed and grape marc meal extract caused a downregulation of various proinflammatory genes in the duodenum of growing pigs. In that study, supplementation of grape seed and grape marc meal extract also caused an increase in the villus height-to-crypt depth ratio, suggesting that plant polyphenols could influence intestinal microarchitecture in a beneficial manner. In another study, two polyphenol-rich plant extracts were used, either grape seed and grape marc meal extract or hop extract, both at dietary levels of 10 g/kg diet. Both plant extracts were able to lower the expression of several proinflammatory genes in various parts of the intestine (duodenum, ileum, colon) (Fiesel et al., 2014). The apparent digestibility of nutrients was not changed by the treatment in that study. However, supplementation of both grape seed and grape marc meal extract and hop extract lowered the abundance of some potentially pathogenic bacteria (Streptococcus spp., Clostridium cluster XIVa) and concentrations of volatile fatty acids in faecal samples. These findings suggest that polyphenols exerted an antimicrobial effect on pathogenic bacteria in the intestine of pigs. In both of the above-mentioned studies, feeding polyphenol-rich plant extracts caused an improvement of the gain-to-feed ratio in pigs, an effect which could be due to an inhibition of proinflammatory processes in the intestine or to antimicrobial effects. A very recent study from Jang et al. (2016) reported that feeding different doses of cocoa powder (2.5, 10 and 20 g) containing 51, 205 and 410 mg flavanols, respectively, to pigs for 4 weeks decreased gene expression of TNF-α and Toll-like receptors -2, -4 and -9 in the ileal Peyer′s patches, mesenteric lymph nodes and the proximal colon. Interestingly, it has been shown in in vitro studies that the antiinflammatory potential of cocoa polyphenols is attributed mainly to the monomeric compounds, such as catechin and epicatechin, whereas the intermediate or large-sized oligomers, the procyanidins, which are also abundant in cocoa powder, have proinflammatory activities as evidenced by an increased secretion of TNF-α from macrophages (Mao et al., 2002; Park et al., 2000). In the context of the above-mentioned study from Jang et al. (2016), this possibly indicates that the anti-inflammatory effects of the monomeric compounds in cocoa powder dominated over the proinflammatory actions of procyanidins. Performance data were not shown in this study, but the authors stated that body weights were not influenced. Besides inhibiting proinflammatory gene expression in the intestine, this study revealed that feeding the cocoa powder increases the abundance of some beneficial bacterial strains, such as Lactobacillus and Bifidobacterium spp., in the proximal colon and faeces. These bacteria are known to inhibit the growth of pathogenic bacteria by the production of lactic acid by fermentation of sugars, reducing the intestinal pH and thus making the environment unfavourable for the growth of pathogens, competition for receptors on the intestinal epithelium and production of antimicrobial compounds (Alakomi et al., 2000; Servin, 2004). In line with findings in pigs, it was reported in a study with healthy humans that consumption of cocoa-derived flavanols has a pre-biotic effect stimulating the growth of beneficial bacterial strains while inhibiting that of potentially pathogenic ones (Tzounis et al., 2011). Thus, these findings indicate that flavanol-rich cocoa powder contributes to gut health through modulation of the colon microbiota composition and reducing expression of inflammatory genes in the intestine. In agreement with the modulation of intestinal microbial populations by polyphenol-rich diets in pigs (Fiesel et al., 2014; Jang et al., 2016), a very recent study from Magistrelli et al. (2016) found that feeding cocoa husk meal (75 g/kg diet), a polyphenol-rich waste from agroindustry representing the integuments of the cocoa beans, to male pigs for 3 weeks increased faecal abundance of Bacteroides–Prevotella and Faecalibacterium prausnitzii. Both bacterial populations are considered beneficial for the intestine due to the production of large amounts of butyrate, which stimulates growth and differentiation of enterocytes and exerts antiinflammatory effects. Unfortunately, these authors did not report on the effect of the cocoa husk meal-enriched diet on inflammatory parameters in the intestine. Performance data of the pigs such as feed intake, weight gain and gain-to-feed ratio were not influenced by feeding of the cocoa husk meal. One limitation of the study form Magistrelli et al. (2016) is that the inclusion of the large amount of cocoa husk meal into the cereal-based test diet not only increase the polyphenol content but also nearly doubled the percentage of acid detergent lignin. Thus, it is unclear to which extent the change of the intestinal microbial populations can be ascribed to cocoa polyphenols. Studies dealing with the effects of plant polyphenols on the expression of proinflammatory genes in tissues other than intestine have not yet been published. However, we have recently observed that supplementation of grape seed and grape marc meal extract or hop extract does not influence the expression of proinflammatory genes in the liver of pigs (Gessner et al., unpublished data).

In conclusion, the few studies published so far suggest that polyphenols are able to suppress inflammation in the small intestine of pigs. Considering similar effects of polyphenol-enriched diets in laboratory animals (Larrosa et al., 2009; Qiao et al., 2014; Etxeberria et al., 2015), it can be postulated that a key mechanism of this beneficial effect of polyphenol-rich diets in pigs is their contribution to gut health by exerting a pre-biotic activity favouring the growth of beneficial bacteria while inhibiting the growth of pathogenic bacteria. This beneficial modulation of the gut microbiota is known to enhance gut barrier function, decrease the translocation of bacterial components into the circulation and stimulate the host immune system (Mosele et al., 2015). In line with this, it was reported that the modulation of the colon microbiota by plant polyphenols correlated with reduced levels of systemic inflammatory markers, such as C-reactive protein, and decreased expression of inflammatory genes in tissues (Axling et al., 2012; Queipo-Ortuño et al., 2012; Moreno-Indias et al., 2016). With regard to the improvements of systemic antioxidative status and the reduction in ROS levels by polyphenol-rich diets observed in some studies with pigs (Luehring et al., 2011; Hou et al., 2014; Wang et al., 2012), we postulate that these effects are unlikely to be mediated by direct antioxidant effects of polyphenols, because of their poor bioavailability, but are secondary due to the improvements of gut health and consequently less translocation of proinflammatory and prooxidative stimuli. However, undoubtedly, further studies are warranted to get more evidence of potential beneficial effects of plant polyphenols in pigs.

Effects of polyphenols on oxidative stress and inflammation in studies with poultry

Similar to pigs, the number of studies dealing with effects of polyphenols in broilers and laying hens is limited and only a few of these studies focused on the antioxidant system or inflammation. In the study of Goni et al. (2007), supplementation of 5, 15 or 30 g grape pomace per kg diet had no effect on growth performance and digestibility of amino acids and also did not influence the antioxidant capacity in plasma and MDA concentration in breast and thigh muscle. However, the susceptibility of lipids in breast and thigh muscle during refrigerated storage was reduced by grape pomace supplementation in that study. In a follow-up study from the same group, doses of 15, 30 and 60 g of grape pomace concentrate per kg diet were tested in broilers (Brenes et al., 2008). In this experiment, doses of 30 and 60 g/kg diet caused a moderate increase in the antioxidant capacity in serum. In agreement with the study of Goni et al. (2007), supplementation of grape pomace concentrate caused a reduced formation of MDA in breast meat during refrigerated storage. In the study of Wang et al. (2008), supplementation with grape seed proanthocyanidin extract improved weight gains and lowered the mortality of broilers infected with Eimeria tenella, but had less effect on antioxidative parameters (MDA concentration, activity of SOD) in plasma. An improved performance and also reduced oxidative stress in broilers fed different polyphenols was reported in a study from Starčević et al. (2015). Theses authors found that the inclusion of different phenolic compounds into the broiler diet, either thymol (200 mg/kg diet), tannic acid (5 g/kg diet) or gallic acid (5 g/kg diet), improved feed utilization and/or final body weight (only tannic acid) and improved oxidative stability (decreased concentration of TBARS) in liver and breast muscle. In a study of Viveros et al. (2011), broiler diets were supplemented with either 60 g/kg diet of grape pomace concentrate or grape seed extract. Supplementation with grape pomace concentrate improved the feed efficiency in broilers, an effect that could be explained by beneficial effects on the intestinal microbial population (increased abundance of Enterococcus, decreased abundance of Clostridium) and an increase in the villus height-to-crypt depth ratio. Interestingly, these effects were not observed in broilers supplemented with grape seed extract, suggesting that polyphenols from different sources exert different biological effects. In the study of Eid et al. (2003), supplementation of green tea polyphenols improved the feed conversion ratio and lowered TBARS concentrations in muscle and liver of broilers treated with corticosterone. In contrast, green tea polyphenols did not lower TBARS concentrations in muscle and liver and even impaired the feed efficiency in broilers which were not treated with corticosterone. In the study of Sridhar et al. (2015), supplementation of resveratrol (at a level of 1% of diet) improved the antioxidative status of serum in broilers, both without and with further treatment with aflatoxin B1. In that study, dietary treatment with resveratrol, however, impaired body weight gains and feed conversion ratio. In a study with Hessian laying hens, the effect of different doses of dietary quercetin (0.2, 0.4 and 0.6 g quercetin/kg diet) on performance, egg quality, caecal microflora populations and antioxidant status was investigated (Liu et al., 2014). This study demonstrated an increased performance (increased laying rate, decreased feed-to-egg ratio) and improvements of caecal microflora status (decrease in the abundance of total aerobes and coliforms, increased abundance of Bifidobacteria) in laying hens fed quercetin. However, no consistent effect of quercetin on the antioxidant status was found; while the activity of SOD in the liver was increased, the concentration of MDA, Trolox equivalent antioxidative capacity and the activity of GPX were not influenced by quercetin supplementation. In a very recent study from Yuan et al. (2016), the effect of different doses of tea polyphenols (600 and 1000 mg/kg diet) on production performance, egg quality and hepatic antioxidant status was investigated in laying hens treated with different concentrations of dietary vanadium (5, 10 and 15 mg vanadium/kg diet) for 8 weeks. This study found that the high concentrations of vanadium caused reduced albumen quality, bleaching effect on eggshell colour and oxidative stress in the liver, while the tea polyphenols prevented the adverse effect of vanadium on egg quality and increased the activity of GPX in the liver being indicative of induction of the antioxidant defence system. To sum up, most of the studies with broilers and laying hens revealed increases of performance and moderate improvements of antioxidant status or oxidative stability of tissues by plant polyphenols. Nevertheless, two studies even reported an impairment of performance of broilers fed polyphenol-enriched diets.

With regard to potential effects of polyphenols on inflammation in broilers, only one study is available so far. That study investigated the effect of tea polyphenols (0.03–0.09 g/kg body weight) on the expression of various proinflammatory cytokines (IL-1B, IL-4, IL-6, IL-10, TNF, IFN-γ) in the intestine of broilers (Li et al., 2015). It was found that most of these genes were downregulated by tea polyphenols. Unfortunately, no performance data were reported in that study. Therefore, it remains unclear whether downregulation of proinflammatory genes by tea polyphenols was associated with an increase in animal performance. In the study of Huang et al. (2013), broilers were fed diets supplemented with either 50 or 100 mg of green tea polyphenols per kg body weight. This study did not consider pathways of inflammation. However, it was found that broilers supplemented with green tea polyphenols have a reduced deposition of body fat, probably as a result of a reduced rate of de novo-fatty acid synthesis and an increased rate of β-oxidation. In overall, the few studies performed so far indicate that polyphenols might exert beneficial effects in the intestine of broilers, like in pigs. It is likely that effects of polyphenols in this respect could be mediated by favourable effect on the microbial composition. However, future studies are warranted to get more evidence of potential beneficial effect of polyphenols in poultry.

Effects of polyphenols on oxidative stress and inflammation in studies with cattle

The focus of most of the studies dealing with effects of polyphenols in dairy cattle was on the production of methane. According to these studies, certain types of polyphenols, particularly tannins, are able to suppress ruminal methanogenesis, both under in vitro and in vivo conditions. The suppression of ruminal methanogenesis was found to be mainly due to a reduction in the number of protozoa and toxic effects of tannins on methanogens (Patra and Saxena, 2009; Cieslak et al., 2013). With regard to an effect of polyphenols on the antioxidant status of dairy cattle, only few studies with divergent results exist. In an earlier study from Colitti and Stefanon (2006), feeding grape polyphenols at a dose of 10 g/day to heifers during the transition period, it was observed that the leucocyte mRNA concentration of the Nrf2 target gene SOD increased after calving. This indicated that grape polyphenols may induce cytoprotective mechanisms after calving to cope with oxidative and inflammatory stress. In the study of Shabtay et al. (2008), supplementation of pomegranate peel as a source of flavonoids increased the concentration of α-tocopherol in plasma of Holstein–Friesian bull calves. In that study, however, no other indices of the antioxidant system were reported. In dairy cows, feeding of anthocyanin-rich corn silage did not influence the antioxidative capacity of plasma (Hosoda et al., 2012), despite a marked increase in the activity of SOD in plasma. Similarly, supplementation of a polyphenol-rich plant extract did not improve concentrations of MDA, α-tocopherol and total antioxidative status of plasma in dairy cows (Gobert et al., 2009). In the study of Gohlke et al. (2013b), 4-week intraduodenal supplementation of quercetin did not influence mRNA concentrations of various antioxidant enzymes in the liver of dairy cows. In a recent study of our group, supplementation of polyphenol-rich grape seed and grape marc meal extract at a dietary level of 1% also did not improve the plasma antioxidant status of dairy cows (as assessed by determination of α-tocopherol, β-carotene, TBARS and total antioxidative capacity in plasma) (Gessner et al., 2015a). In another study, feeding a polyphenol-rich plant product consisting of green tea and curcuma extract also did not exert antioxidant effects in dairy cows (Winkler et al., 2015). Collectively, it can be stated that, like in most studies with monogastric animals, the antioxidant status of dairy cattle is less influenced by supplementation of feed rations with polyphenols from plant sources.

Regarding potential antiinflammatory effects of polyphenols in cattle, there are only very few studies available so far. In a study from Oliveira et al. (2010), the effect of polyphenols from pomegranate extract on health, nutrient digestion and immune competence was investigated in Holstein calves. The calves received either 5 or 10 g pomegranate extract per day throughout the first 70 days of age. Feeding the pomegranate extract polyphenols impaired digestibility of crude protein and fat, increased the secretion of IFN-γ and IL-4 by peripheral blood mononuclear cells and improved total IgG responses to ovalbumin vaccination. These results led the authors to suggest that feeding pomegranate extract polyphenols suppresses nutrient digestibility, likely because of the high tannin content, but enhances mitogen-induced cytokine production and response to vaccination, which might be beneficial for immune competence and health of calves. In the study from Gessner et al. (2015a), feeding a polyphenol-rich grape seed and grape marc meal extract in dairy cows caused a consistent reduction (25–65%) of the mRNA concentrations of a large set of genes involved in inflammation and ER stress in the liver, even though the effects were not statistically significant. Moreover, feeding grape seed and grape marc meal extract to dairy cows in that study caused a significant downregulation of fibroblast growth factor (FGF)-21, a key marker of ER stress and a surrogate marker of liver fat accumulation. In this study, supplementation of the polyphenol-rich product also improved milk performance, which could be – at least in part – the result of a suppression of inflammation and ER stress in the liver. In the study from Winkler et al. (2015), the effect of feeding a polyphenol-rich plant product was also investigated on genes of inflammation and ER stress in the liver of dairy cows. Feeding the polyphenol-rich plant product caused a downregulation of a set of genes involved in inflammation and ER stress, which, however, failed to be significant for most genes. However, in the group of cows supplemented with the plant product, there was also a significant downregulation of FGF21, a reduced fat content in the liver and an increased milk performance. As dry matter intake was not different between cows supplemented with the plant product and control cows, it has been suggested that the plant product improved the utilization of energy for milk production. One plausible explanation for this might be a potential reduction in ER, inflammatory and metabolic stress in the liver by the polyphenols from the plant product.

Overall, the few studies in dairy cows published so far show that plant sources rich in polyphenols could have the potential to lower inflammation and ER stress in the liver during the transition period and even improve milk performance. An interesting observation of the few studies is the downregulation of hepatic FGF21, because of its role for liver fat accumulation, a frequent phenomenon in high-yielding dairy cows. Future studies have to clarify whether a modulation of the rumen microbiota by plant polyphenols (Patra and Saxena, 2009; Cieslak et al., 2013) might contribute to the improvements of the inflammatory status in dairy cows and the significant improvements of milk performance. However, given that rumen disorders, such as subacute rumen acidosis and abomasal displacement (Plaizier et al., 2008; Zebeli and Metzler-Zebeli, 2012; Guzelbektes et al., 2010), which frequently occur during parturition and contribute to the systemic inflammatory condition during early condition, are accompanied by a marked impairment of the rumen microbial community and an increased translocation of inflammatory stimuli, such as LPS and cytokines, this is not unlikely.