Effects of Macleaya cordata extract supplementation on digestive parameters of ponies

ABSTRACT High amounts of grains in the equine diet led to high starch intake, causing gut alterations. Aimed at reducing harmful effects, Macleaya cordata extract (MCE) is a phytogenic additive that stands out for its antibiotic and anti-inflammatory effects proven in different species. However, there is no useful information for horses. The objective of this study was to evaluate the effects of different levels of the inclusion of commercial MCE on body weight (BW), body condition score (BCS), total apparent digestibility (AD) of nutrients, faecal pH and fermentative products, on ponies fed a high-starch diet. Eight healthy gelding Mini Horse ponies were used. The study design was contemporary double Latin-square 4 × 4 in the experimental unit, with the animal inside each experimental period (n = 8 experimental units per group). The experiment was conducted over four 20-d periods. Basal diet attended 1.75% BW dry matter daily and starch intake was 2.2 g/kg BW/meal. The experimental groups were as follows: control – without food additive; S1–1 mg/kg BW MCE; S1.5–1.5 mg/kg BW MCE and S2–2 mg/kg BW MCE. The data were analysed by PROC MIXED of SAS (p < 0.05). Tendency was considered when 0.05 < p < 0.1. Our results showed higher ether extract (EE) AD for S2 group (63.75%) when compared with the control (54.55%) (p = 0.0377). Lactate was lower (p = 0.0391) in S1 (3.27 mmol/l) and S2 (3.24 mmol/l) groups, although pH was not different between groups. Iso-valerate was greater in S1 group (2.29 mmol/l; p = 0.0289), and a tendency of higher butyrate values was found for S1 and S2 groups (p = 0.0980). We concluded that MCE supplementation probably positively influences equine resident microbiota, improving EE AD and increasing iso-valerate concentration. It can also minimise harmful high-starch impact. We recommend further studies using MCE in horses for a better understanding of its local activity and possible benefits.


Introduction
Diet components, as well as type, quantity and frequency of feeding can cause problems in horses by influencing gut microbiota, fermentative parameters and inflammatory processes (Shirazi-Beechey 2008;Fernandes et al. 2014). Equine domestication led to major changes in their diet, moving from a low-starch diet to one with large amounts of grains and, consequently, a large amount of starch. This drastic change in diet has brought some problems to animals. Due to the large amount of starch ingested, part of this starch will not be digested by the horse since it exceeds the digestion capacity of the small intestine and will be fermented by the microbiota of the large intestine, reducing energy utilisation and causing serious health problems (Harris and Geor 2009;Julliand and Grimm 2017;Stewart et al. 2017).
In this scenario, feed additives, with the ability to reduce the harmful effects of the diet and modulate the intestinal microbiota, have gained great prominence. Thus, phytogenic additives, such as the Macleaya cordata extract (MCE), have drawn a lot of attention. MCE supplementation is related to improvement in body weight gain and feed conversion in broilers (Khadem et al. 2014), pigs (Kantas et al. 2015) and ewes (Estrada-Angulo et al. 2016). This greater productivity seems to be associated with beneficial microorganism selection in gut and anti-inflammatory effect of extract compounds, especially the isoquinoline alkaloid Sanguinarine (SG) (Kubáň et al. 2006;Dvorak and Simanek 2007;Niu et al. 2012;Wang et al. 2017), as observed in rats (Vrublova et al. 2010), dogs (Faehnrich et al. 2019), weaning ewes (Chen et al. 2020) and chicken (Guo et al. 2021).
However, there is no information about the effects of using MCE in horses. Therefore, our hypothesis was that MCE supplementation for horses receiving large amounts of starch could minimise negative impacts on the intestine and benefit nutrient digestibility without altering body weight (BW) and body condition score (BCS). The aim of this study was to evaluate the effects of different levels of a commercial MCE on BW, BCS, total apparent digestibility of nutrients, faecal pH and fermentative products in ponies fed a high-starch diet.

Locale and animals
The experiment was carried out at the Equine Digestive Health and Performance Research Laboratory (LabEqui), from the University of São Paulo -Campus Fernando Costa, in Pirassununga (SP) (21º59'46" S latitude and 47º25'33" W longitude). Eight gelding Mini Horse ponies [mean age 8.59 + 0.17 y, mean BW 150.47 + 16.38 kg and mean BCS 6.78 + 0.89 (1-9) (Henneke et al. 1983)] were used in the experiment. Animals were considered healthy according to normal clinical and laboratory examinations. All ponies were dewormed and vaccinated against rabies, influenza, tetanus and encephalomyelitis. They were maintained in individual stalls during the experimental period, with 1 h social interaction in a pen during the adaptation period. Stall feeders and drinkers were adapted according to each pony's height. During the wash-out period, animals were kept together in a pen. Animals were handled in accordance with the Institutional committee for Ethics in Research from the University of São Paulo (CEUA-FMVZ number: 4587200519).

Experimental design and diets
Experimental design was contemporary double Latin-square 4 × 4 in the experimental unit, with the animal inside each experimental period (n = 8 experimental units per group). Animals were randomly distributed between experimental groups and periods. The experiment was conducted over four 20-d periods (15 d for locale and feed adaptation and 5 d for sample collection), with 15 d of wash out between them ( Figure 1).
Diets were formulated to challenge the gastrointestinal tract of ponies to maintain nutrient requirements. Basal diet requirement 1.75% BW dry matter daily, being 1.05% concentrate and 0.7% roughage feed (concentrate:roughage proportion of 60:40). Concentrated food was a corn-based mash ration (Table 1), and roughage was Cynodon dactylon cv. Tifton-85 hay. Their chemical analyses are presented in Table 2. Both were weighed and offered together twice a day, at 7:00 a.m. and at 4:00 p.m. during the experimental period. Starch intake per meal was 2.2 g/kg BW. During wash out,  animals received only C. dactylon cv. Tifton-85 hay (2% BW dry matter daily). Water and mineral salt were always offered ad libitum. MCE commercial phytogenic Sangrovit® (Phytobiotics, Eltville, Germany) was added to the ration of the supplemented groups by top dressing, over the experimental period. Animals were divided into four groups: control -without food additive; S1-1 mg/kg BW MCE commercial phytogenic (96 ppm); S1.5-1.5 mg/kg BW MCE commercial phytogenic (145 ppm); S2-2 mg/kg BW MCE commercial phytogenic (194 ppm).

Body weight and body condition score
BW and BCS were evaluated at the beginning and at the end of each experimental period. BW was measured on a digital balance (model: Toledo MGR-3000 Júnior (Toledo do Brasil, Brazil)), while BCS was set up by a single technician using a 1 to 9 scale (Henneke et al. 1983).

Total apparent digestibility of diets
Total apparent digestibility of nutrients was performed by total faecal collection (TFC) from d 15 to d 20 in each experimental period. Therefore, the animal's beds were removed during TFC to enable immediate faecal collection after defaecation. Urine, mane and hair were constantly withdrawn to avoid any contamination. Every 12 h, surplus concentrate and roughage were collected and weighed, including, total faeces, which was weighed and homogenised, and a 10% aliquot (simple sample) was separated in plastic bags and stored at −20°C. After 5 d of TFC, 10% aliquots formed composite samples. Parameters evaluated were as follows: dry matter (DM), organic matter (OM), crude protein (CP), nitrogen-free extract (NFE), mineral matter (MM) and ether extract (EE), according to official AOAC international methods; neutral detergent fibre (NDF) and acid detergent fibre (ADF) by fibre partition (Van Soest et al. 1991).

Short chain fatty acids (SCFA) and lactate (L)
SCFA and L samples were collected immediately after the first defaecation on the eighteenth day of each experimental period. For SCFA, 10 g of faeces was diluted in 20 ml of distilled water, homogenised and leached. Then, 4 ml was transferred to glass tubes with 1 ml of formic acid HPLC degree 98-100%. Tubes were centrifuged for 12 min at 2500 × g (centrifuge Excelsa® II 206 MP (Fanem, São Paulo, Brazil)) and 2 ml of supernatant was collected and stored in microtubes at −20°C. SCFA evaluation was done using the Agilent 7890A gas chromatography technique with a flame ionisation detector (7683B) and a fused silica capillary column (J&W19091F-112, Agilent Technologies, Santa Clara, CA, USA) (Ferreira et al. 2016). For L evaluation, 1 g of faeces was collected from each animal in a 10 ml dry glass tube with cap, frozen at −20°C and analysed by spectrophotometric method for biological fluids (Pryce 1969).

Faecal pH
On d 20, faecal samples were collected after the first defaecation, diluted with distilled water in a 1:1 proportion, homogenised and leached. A pH metre was immersed in the leached liquid to determine faecal pH (Jouany 1982;Zeyner et al. 2004;Goachet et al. 2014).

Statistical analysis
As horses receiving a standard diet in a control environment are difficult to obtain, due to ethical and financial concerns, eight animals were recruited in the present study. A Latin square design was chosen so that each animal could act as its own control, thus reducing the individual variability attaining a significance level of 0.05 with a power of 80% for the variables tested. Data were analysed using the Mixed package of the Statistical Analysis System software (version 9.0, SAS Institute, Cary, NC, USA). They were previously submitted by the Shapiro-Wilk normality test (p < 0.05). Subsequently, they were subjected to analysis of variance, and the means were compared by the Tukey test, at a significance level of 5%. The statistical model considered fixed treatment effect and the random effects of period, square, animal within the square, as well as the errors associated with each observation. Means were calculated for all variables using Tukey's multiple comparison procedure. The significance level was set at p < 0.05. Tendency was considered when 0.05 < p < 0.1.

Results
There were no group effects for ponies BW and BCS (Table 3). For the total apparent digestibility coefficients, our results showed higher values of EE AD (p = 0.0377) for S2 group (63.75%) when compared to control group (54.55%), without alterations for AD of DM, OM, CP, NFE, MM, NDF and ADF (Table 4). L was lower (p = 0.0391) in S1 (3.27 mmol/l) and S2 groups (3.24 mmol/l) when compared to S1.5 (5.48 mmol/l) group; however, MCE supplementation had no influence in faecal pH. For SCFA, our study showed that supplementation affected iso-valerate (p = 0.0289), which was higher in S1 group (2.29 mmol/l) when compared to S1.5 group (0.43 mmol/l). Also, a tendency was detected (p = 0.0982) for higher values of butyrate concentration in S1 and S2 groups (Table 5).   In the present study, even with no effect on BW, we found a positive influence of MCE on EE AD for S2 group when compared to the control. The EE AD expected for horses receiving equal proportions of hay and grains is 55% (Kronfeld et al. 2004), which is similar to our results for control, S1 and S1.5 groups, but lower than S2 group mean. In a study that evaluated diets rich in starch (2.4 g/kg BW starch per meal), horses that received a probiotic of Lactobacillus acidophilus, Lactobacillus casei, Bifidobacterium bifidum and Enterococcus faecium showed higher EE digestibility. These bacteria are known to improve digestive health (Swyers et al. 2008). Thus, better EE AD in S2 group could be related to a possible selection of beneficial gut bacteria.

Discussion
There was no influence on faecal pH. Equine gut pH interferes with cellulose and hemicellulose breaking and influences SCFA transportation through intestinal epithelial. High amounts of rapidly fermentative carbohydrates can result in severe consequences for equine health, alterations in resident microbiota, higher incidence of gastritis and more (Cipriano-Salazar et al. 2019). Faecal pH values found in the present study are compatible with previous results found for horses consuming roughage and concentrate (Willing et al. 2009;van den Berg et al. 2013;Kabe et al. 2016) but smaller than others (Zeyner et al. 2004;Hansen et al. 2015;Correa et al. 2016;Harlow et al. 2016;Johnson and Rossow 2019;Muhonen et al. 2021).
Our findings did not characterise acidosis where pH values are below 5.8 (Cipriano-Salazar et al. 2019). However, lactate values were worrisome, being extremely high when compared to values previously described in equine faeces and colon (Zeyner et al. 2004;Daly et al. 2012;Muhonen et al. 2021). The smaller median obtained in our study was more than four times greater than that observed in the faeces of horses fed a 35:65 concentrate:roughage ratio (Muhonen et al. 2021). Bacteria to the detriment of lactateutilising bacteria (Harlow et al. 2016). Despite L high values for all groups, MCE supplementation at doses of 1 and 2 mg/kg led to a decrease in L concentration in faeces. This is another indication that there could be a possible selection of beneficial bacteria or a mitigation of dietary effects on resident microbiota due to MCE compounds.
Another interesting result was faecal SCFA concentrations. Acetate concentration in our study was similar to findings for horses receiving diets with a higher roughage proportion in relation to concentrate (Kabe et al. 2016) and were higher than results for ponies and horses in a body weight gain programme (Langner et al. 2020) and after diet adaptation (Muhonen et al. 2021). However, the values of propionate were very low. Butyrate, iso-butyrate, valerate and iso-valerate were similar to other findings (Zeyner et al. 2004;Kabe et al. 2016;Langner et al. 2020;Muhonen et al. 2021).
Our results could be due to the amount of starch (2.2 g/kg BW starch per meal) and the starch source (corn). Feeding more than 1 g/kg BW starch per meal favours it to reach the large intestine and be fermented by resident microbiota (Garber et al. 2020). In caecum, Bacteroidetes, Proteobacteria and Tenericutes increase and microbiota diversity decreases (Hansen et al. 2015). Additionally, the progressive increase of concentrate in the equine diet leads to Streptococcus spp. population increase, which is related to gut diseases. Lactobacillus spp. and lactate-utilising bacteria also rise in this situation (van den Berg et al. 2013;Grimm et al. 2018).
When corn is the starch source, amylolytic bacteria increases and fibrolytic bacteria decreases in the hindgut. There is also a prevalence of amylolytic bacteria Enterococcus faecalis, Gram-positive coccus favouring and decreasing in lactobacilli numbers and faecal lactate-utilising bacteria. These events are more pronounced when the starch source is corn rather than when it is oat (Harlow et al. 2016).
In the present study, acetate production predominated in all groups when compared to other SCFA, and the propionate concentration was very low. Acetate production is related to higher microbiota diversity and fibre digestion (Garber et al. 2020). This result is curious and controversial since a decrease in fibrinolytic activity, a decrease in acetate production and an increase in propionate were expected due to high starch (Hansen et al. 2015).
There were alterations in iso-valerate concentrations from the addition of MCE in addition to 1 mg per kg BW dose and a tendency for higher values of butyrate for S1 and S2 groups. Iso-valerate is produced from the catabolism of the amino acids valine, isoleucine, leucine and proline and is essential for structural carbohydrate digestion and microbial protein synthesis. Iso-valerate supplementation in cattle was responsible for increased rumen SCFA concentration, establishment of a ruminal pattern of acetate production (Liu et al. 2009), increased microbial enzyme activity responsible for cellulose and hemicellulose degradation, increased ruminal butyrate concentration and increased predominant cellulolytic bacteria populations (Liu et al. 2014(Liu et al. , 2016. In the present study, S1 group did not have an increase in acetate production, but a tendency to higher values of butyrate was found in S1 and S2 groups. Butyrate is a product of fibre digestion, an important energy source for horses, a regulator of intestinal wall pro-inflammatory gene expression and essential for colonic epithelial cell maintenance (Nedjadi et al. 2014). Therefore, our findings could indicate a favouring of fibrolytic bacteria populations and an attenuation of harmful consequences due to high lactate production by large intestine amylolytic bacteria by MCE supplementation.

Conclusions
We can conclude that MCE supplementation of ponies receiving a high starch diet improves EE digestibility without altering BW or BCS. Additionally, MCE is capable of increased iso-valerate concentration and decreased L production, while faecal pH does not change, and it tends to influence butyrate production by resident microbiota, pointing to a possible attenuation of adverse effects due to high starch amounts. The most beneficial MCE commercial supplement daily dose for horses was 1 mg per kg BW. More studies should be performed to understand MCE local activity better and to define whether its use could bring more benefits to equines.