| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ANIMAL NUTRITION |
,
,2
* INRA, UR1213 Herbivores, Site de Theix, 63122 Saint Genes Champanelle, France;
and
Alltech-France, 14 Place Marie-Jeanne Bassot, 92593 Levallois-Perret, France; and
Etablissement National d’Enseignement Supérieur Agronomique de Dijon (ENESAD), 26 Boulevard Dr Petitjean, BP 87999, 21079 Dijon, France
 |
Abstract |
|---|
 Top Abstract INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: dietary effect • digestibility • fiber digestion • horse • live yeast • transit time
|
INTRODUCTION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
reported that only limited amounts of starch are digested in the small intestine and that decreasing the dietary hay:grain ratio will increase the amount of undigested starch entering the hindgut, thus disturbing its microbial balance. The accumulation of lactate, and the subsequent drop in pH, impairs the activity of lower-gut cellulolytic bacterial populations and increases the horse’s susceptibility to colic pain or laminitis (Kronfeld and Harris, 1997).
Several strategies have been tested to reduce these dysfunctions of the equine intestinal ecosystem. One strategy consists of adding probiotics such as Saccharomyces cerevisiae, as recommended for ruminants fed high-starch diets (Wallace and Newbold, 1992; Nagaraja et al., 1997). Previous studies indicated that live yeasts can improve the microbial balance in the hindgut of horses, stimulating the population of cellulolytic bacteria and their activity (Medina et al., 2002), and increasing the digestibility of dietary nutrients (Glade, 1991a,b). However, the diet composition-related ability of live yeasts to modify microbial digestion and fiber degradation in horses has not been extensively studied.
Therefore, a study was conducted to evaluate the effect of a live yeast culture of S. cerevisiae CBS 493.94 on in vivo digestibility of dietary components and on the rate of cell wall digestion in horses fed either a high-starch or high-fiber diet. Mean retention times of digesta in the whole digestive tract and the hindgut were measured to explain the observed changes in digestibility.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Animals
Eight crossbred male horses (12 ± 5 yr; mean BW 305 ± 18 kg) were allocated into pairs consisting of a cecum and right ventral colon-fistulated animal and a second animal that was only cecally fistulated. The cannula (polyvinyl chloride, i.d. 30 mm) were fitted surgically by a certified veterinary surgeon at least 6 mo before the beginning of the trial (Drogoul et al., 2000).
Horses were wormed with a double dose of Pyrantel (Strongid, Laboratoire Pfizer, Orsay, France) followed 1 wk later by a single dose of Ivermectin (Eqvalan, Laboratoire Merial, Lyon, France) given 15 d before the experiment began. Indoor housing consisted of concrete-floored, individual loose boxes bedded with flax shavings (Ecolit, Croissanville, France). During the diet-adaptation period (21 d), animals were given access to a sandy paddock for 10 h/wk. During the following 10 d they were individually tethered in stalls and fitted with a fecal collection harness.
Diets
Feedstuffs as well as their physical form were chosen to mimic normal French feeding practices conducted in riding schools. Two diets were used (Table 1): a pelleted high-fiber (HF) concentrate or a pelleted high-starch (HS) concentrate, both given in mixture with long wheat straw to supply a minimum fiber content. Ingredients of the HF and HS concentrates were ground through a 1.5-mm sieve and pressed into 3-mm-diameter pellets. The 2 diets were supplemented (HF+SC; HS+SC) or not (HF+0; HS+0) with a lyophilized flocculent culture of Saccharomyces cerevisiae (SC) strain CBS 493.94, including the growth medium (Yea-Sacc, Alltech Inc., Lexington, KY). The live yeast culture supplement containing 4.5 x 109 cfu/g was given at a rate of 10 g/d per horse.
|
All diets were given at the same rate (i.e., 18.0 g of pelleted feeds + 3.5 g of long wheat straw = 21.5 g of DM/kg of BW per d) to avoid any effect of meal size on the rate of passage of digesta (Drogoul et al., 2001; Pearson et al., 2001). At this intake level the HF diet provided 100% of energy requirements, whereas the HS diet provided 130% of the energy requirements (Martin-Rosset et al., 1994). The daily rations (6.5 kg/horse per d, on average) were divided into 2 equal meals given in individual troughs at 0800 and 1700. The pelleted concentrates were offered first with the SC culture top-dressed (5 g/meal), and the wheat straw fraction was provided 30 min later. Mean daily ingestion of starch was equal to 3.1 and 1.2 g/kg of BW per meal for the HS and HF diets, respectively (Table 2).
|
Experimental Design and Treatments
The 8 horses were randomly assigned by pairs in a double 4 x 4 Latin square design at the beginning of the experiment. Each pair included a cecum and right ventral colon-fistulated animal and a cecal-fistulated animal, and remained unchanged for the duration of the entire study. The 4 pairs of animals received 4 dietary treatments over 4 periods. The 4 dietary treatments described previously were applied as follows: 1) the high-fiber diet (HF+0 treatment); 2) the high-starch diet (HS+0 treatment) ; 3) the high-fiber diet supplemented with 10 g/d of a Saccharomyces cerevisiae CBS 493.94 preparation (HF+SC); and 4) the high-starch diet supplemented with 10 g/d of the same SC preparation (HS + SC).
Each of the 4 experimental periods lasted 31 d, split into a 21-d period that allowed for adaptation to experimental diets, followed by a 10-d period of total fecal collection, which comprised a 3-d period for animals to adapt to the fecal bags followed by a 7-d period for measurements on apparent digestibility and rate of passage of digesta.
Indigestible Markers used for Measurement of the Rate of Passage of Digesta
Three indigestible markers were used to calculate rate of passage of feed particles through the hindgut (cecum + colon) and whole digestive tract, as recommended by Pagan et al. (1998) and Drogoul et al. (2000). Ytterbium was bound to the pelleted feeds and was fed to horses to estimate mean retention time (MRTYb) of small particles (
2 mm) in the whole digestive tract. Europium (Eu) was bound to pelleted feeds and introduced manually through the cecal cannula in order to estimate the mean retention time (MRTEu) of small particles (2 mm) in the hindgut (cecum + colon). To measure mean retention time (MRTDy) of large particles (>2 mm) in the hindgut, dysprosium (Dy) was bound to the undigested fecal particles and introduced manually through the cecal cannula. Fecal particles intended to be labeled were collected during each period of animal adaptation to harnesses and were then water-washed and filtered through a 2.5-mm sieve. The particles retained on the sieve were labeled using a Dy oxide solution (Dy2O3; 50 mg of Dy/g of DM; pH 2.5). The 2 pelleted feeds (HF and HS) were labeled using either a Yb oxide solution (Yb2O3; 35 mg of Yb/g of DM; pH 2.5) or a Eu oxide solution (Eu2O3; 30 mg of Eu/g of DM; pH 2.5) according to the method described by Drogoul et al. (2000). To prevent losses of small particles, pelleted feeds were placed into nylon bags with a 100-µm pore size (Blutex T120, Saati-France, Sailly Saillisel, France) during the labeling procedure. The labeled feeds were hand-pelleted with a syringe so that a precise dose of markers could be administered.
A single dose of Yb-labeled pellets (40 g) was fed to each animal mixed into one-third of the morning meal. After either complete ingestion, or at least a 45-min delay, refusals, if any, were weighed and removed from the trough. The rest of the pelleted concentrate meal was then given. During meal ingestion, all collection harnesses were cleaned of feces. The Eu-labeled pellets (30 g; particles 2 mm) and the Dy-labeled fecal particles (30 g; particles >2 mm) were introduced through the cecal cannula 4 h after the oral dose of Yb-labeled pellets (Drogoul et al., 2000). After ingestion of the Yb-labeled meal, excreted feces were collected from each horse at 4, 7, 10, 13, 16, 19, 22, 25, 28, 32, 36, 40, 46, 52, 58, 64, 72, 84, 96, 120, 144, and 168 h postingestion. Individual fecal collections were weighed and thoroughly mixed, and a subsample (300 g of fresh matter) was taken to measure DM and marker concentrations. Markers were solubilized according to Siddons et al. (1985). Ytterbium, Eu, and Dy concentrations were assayed by atomic absorption spectrophotometry (SpectrAA 300, Zeeman–Varian, Les Ulis, France) with wavelengths set at 398.8, 421.2, and 459.4 nm for Yb, Dy, and Eu, respectively.
Collection of Samples for In Vivo Apparent Digestibility and Rate of Passage of Digesta Measurements
In vivo apparent digestibility and rate of passage of digesta were determined by the total fecal collection method using harnesses and collection bags for 7 d. Amount of feed offered, refused, and feces were weighed every day at 0800. Aliquots of individual feces (10% of total excreted fresh feces), offered feeds (500 g of fresh matter) and all refusals, if any, were taken daily and dried in a forced-air oven until constant weight at 65°C for DM determination. At the end of each period, aliquots were ground through a 0.8-mm mesh, and representative samples from each collection (offered feeds, refusals, and feces) were pooled for each animal over the 7-d period for subsequent analysis. Acid detergent fiber, NDF, and ADL were determined according to the procedure described by Van Soest and Wine (1967). Organic matter was determined after ashing in a muffle furnace (550°C for 6 h). Crude protein determination was based on nitrogen content measured using a semiautomated micro-Kjeldahl method using a VDK 126A distillation unit (Velp Scientifica, Milan, Italy). Starch content was determined according to the method of Thivend et al. (1965). All analyses were performed in duplicate.
Calculations and Statistical Analyses
Apparent digestibility of DM was calculated as the amount of digested DM (ingested DM – excreted DM in feces) per unit of ingested DM. Total DM intakes and fecal DM outputs were measured over the 7-d collection period. In vitro apparent disappearance rate (IVAD) of NDF was calculated as the amount of digested NDF (ingested NDF – excreted NDF in feces) per unit of retention time (h) of solid particles in the hindgut. The same relationship was applied to ADF, and to cellulose (ADF – ADL) and hemicellulose (NDF – ADF) cell wall fractions.
The MRT of small particles in the whole digestive tract and small and large particles in the hindgut compartment were calculated using the model established by Faichney (1975) and applied to the kinetic excretions of Yb, Eu, and Dy in feces as:
 |
where 
ti is the time elapsed between introduction of the markers (time zero) and the middle of the ith collection intervals and calculated as 
with ti being the time to the end of the ith interval; and Mi is the amount of marker excreted in the ith interval with all markers being excreted by the nth interval. The MRT of small particles from mouth to cecum was calculated by the difference MRTYb – MRTEu.
All data were processed by ANOVA using the GLM procedure (SAS Inst. Inc., Cary, NC). The model included the effects of horse (which was considered random), period, diet, yeast supplementation, and the interactive effects of diet type and yeast supplementation. No effect of period was detected; therefore, period was removed from the final statistical model. Least squares means were calculated for each variable and separated using the pairwise t-tests (PDIFF option of SAS).
A repeated-measurements ANOVA was performed to compare differences in marker excretion (cumulative Yb excretion and Eu excretion) at differing time points using the repeated time option of SAS. The model used included the effects of horse, diet, and yeast supplementation. Statistical significance was set at P < 0.05.
|
RESULTS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
All animals gained approximately 10% of their initial BW and there were no obvious signs of digestive or metabolic disorders during the duration of the study.
Nutrient Intake
There were no diet effects on DM, CP, and OM intakes (P > 0.068; Table 2
). The greater proportion of barley in the HS diets led to an increase in starch intake (P < 0.001) and a decrease in NDF, ADF, cellulose, and hemicellulose intakes (P < 0.001). Yeast addition had a positive effect on DM (P = 0.03), NDF (P = 0.038), and hemicellulose (P = 0.023) intakes. A significant diet x SC interaction (P = 0.042) was observed on OM intake.
Rate of Passage of Digesta
There was no effect of SC supplementation on the MRT of solid particles in either the whole digestive tract or the hindgut (Table 3). Conversely, the MRT of small particles in the hindgut was 3.5 h longer with the HS diet than the HF diet (P = 0.036). A similar pattern was observed for the MRT of large particles (+3 h), but the difference was not significant (P = 0.08). The MRT of feed particles in the foregut section (stomach + small intestine), as estimated by the difference between the MRTYb in the whole digestive tract and the MRTEu in the hindgut compartments, was not different (P = 0.972) between the 2 diets. No diet x SC interaction was detected on rates of passage of digesta through the whole digestive tract, the foregut, or the hindgut.
|
). The between-diet difference in peak shift of Eu excretion noted (6.7 h longer with HS) indicated differential rates of passage of dietary solid particles through the hindgut (Figure 2).
|
|
Apparent digestibility of DM, OM, and CP were greater (P < 0.001; Table 4) in HS diets than in HF diets. In contrast, there were no diet effects on the digestibility of total dietary cell wall fractions (NDF; P = 0.995), whereas the ADF fraction was better digested (P = 0.035) and the hemicellulose fraction tended to be less digested (P = 0.080) in HF than in HS diets. No significant diet x SC interaction was detected on in vivo apparent digestibilities.
|
In Vivo Apparent Disappearance Rate of Dietary Cell Wall Fractions
Because the hindgut is the only site of cell wall digestion in the digestive tract of horses, the disappearance rate of dietary cell wall fractions (i.e., IVAD) will be considered as an indicator of the microbial cellulolytic activity. A large diet effect was noted on IVADNDF, IVA-DADF, and IVADcellulose, which were much greater (P < 0.001) in HF than in HS diets. However, the diet had no effect (P = 0.341) on IVADhemicellulose. There was no diet x SC interaction on estimated IVAD parameters.
Yeast supplementation had no effect on IVADNDF, (P = 0.141), IVADhemicellulose (P = 0.453), or IVADcellulose (P = 0.124) in either diet but tended to improve IVADADF in both HF and HS diets (P = 0.077).
|
DISCUSSION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Feed Intake
Although animals received similar amounts of feed, there was a general positive effect of SC on DM intake (P = 0.030). Data on ingestion indicated that non-yeast-supplemented animals tended to refuse a portion of their ration; whereas, yeast-supplemented animals ingested the whole ration. The positive effect of SC supplementation on mean DM, OM, and NDF intakes were greater for HF than HS diets, which explains both the overall positive effect of SC on NDF (P = 0.038) and hemicellulose (P = 0.023) intakes, and the significant diet x SC interaction on OM intake (Table 2).
Effect of Starch Supplementation on Digestion in Horses
A negative effect of starch on microbial cellulolysis in the hindgut of horses has been reported in previous experiments in which native grains were incorporated into the diet (>50% of DM intake) at the expense of forage (Thompson et al., 1984; Martin-Rosset and Dulphy, 1987; Palmgren-Karlsson et al., 2000; Drogoul et al., 2001). The amount of starch intake per se is one factor that could explain such negative interactions with cellulolysis and the degree of cellulolysis inhibition (Ott, 1981; Martin-Rosset and Dulphy, 1987). Furthermore, technological processes applied to grains (grinding, pelleting) can influence the enzymatic digestion of starch in the small intestine, and therefore, alter the amount of starch reaching the large intestine and interacting with fiber digestion (Julliand et al., 2006).
In the current experiment the amount of starch fed in the HS diet (3.1 g/kg of BW per meal) was similar to the 3.5 to 4.0 g/kg of BW per meal limit defined by Potter et al. (1992) and Kienzle (1994) as the maximum capacity of the small intestine for starch digestion. Therefore, it can be hypothesized that intestinal undigested starch can enter the hindgut of horses fed the HS diet. It is likely that the transit time of finely ground and pelleted diets is shortened when compared with unground diets, thus allowing more starch to escape small intestine digestion, reach the hindgut, and interact with the local microbial environment (Julliand et al., 2001). Also, we assumed that starch was more easily hydrolyzed by endogenous amylases in ground cereals, and the resultant effect of grinding on starch digestibility in the small intestine was probably minimal.
Medina et al. (2002) analyzed the concentrations and activities of hindgut microflora of horses fed diets similar to those used in the current study. They observed that the microbial profiles were altered for a high-starch diet. The number of cellulolytic bacteria was decreased benefiting lactobacilli and streptococci; thus, the lactic acid concentration was increased and pH of the hindgut content decreased, which impaired microbial fibrolytic activity. These results are consistent with the decreased disappearance rate of cell wall components (IVADNDF and IVADADF) observed here in the HS diet (Table 5), which explains the subsequent reduction of ADF digestibility in horses fed the HS diet compared with the HF diet (Table 4). As observed in the rumen (Hoover et al., 2006), it is likely that negative interactions on microbial fiber digestion originating from starch fermentation occurred in the hindgut of horses fed the HS diet. In addition, the lower IVADNDF and IVADADF are likely due to the decreased amount of cell wall material supplied to the hindgut and used as substrate by cellulolytic bacteria when animals were fed the HS compared with HF diet.
|
, Yoder et al. (1997), and Palmgren-Karlsson et al. (2000). The same pattern has been observed when forage-based diets were supplemented with raw cereals (Hintz et al., 1971b; Ott, 1981; Thompson et al., 1984; Pagan et al., 1998; Palmgren-Karlsson et al., 2000; Drogoul et al., 2001) or pelleted concentrates (Hintz et al., 1971a; Martin-Rosset and Dulphy, 1987; Vermorel et al., 1997). The greater DM and OM digestibility in the HS diet compared with the HF diet is due to diet composition, because starch is digested to a greater extent than the cell wall fractions of forages.
The apparent digestibility of CP was greater in the HS diet than in the HF diet and independent of SC supplementation. This is explained by the greater availability of soybean proteins to the endogenous enzymes in the small intestine of horses than the proteins from alfalfa. This could be due to a protective effect of cell wall components on intracellular proteins, and to differences in the secondary or tertiary structure of proteins between alfalfa and soybean meal as indicated by Wallace and Cotta (1988). Also, greater proteolytic activity of hindgut digestive microbial ecosystem with HS diet could explain such a result, given that starch favors the growth of amylolytic bacteria (Medina et al., 2002) and that most amylolytic bacteria in the digestive tract are also proteolytic (Wallace et al., 1997).
Because the dietary ingredients of both diets were finely ground and pelleted, the MRT of small particles was considered a good indicator of the MRT of whole diets. The mean whole-tract digestive MRT in horses fed the HF diet (32.3 h) was identical to that reported by Pagan et al. (1998) for thoroughbred geldings fed a forage diet. Also, the mean MRT of feed particles in the hindgut of animals fed the HF diet (25.0 h) was in the same range as the MRT in the large intestines (23.3 h) of ponies fed timothy hay (Udén et al., 1982). Furthermore, the mean MRT of small particles in the prececal tract of horses fed the HF diet (8.0 h) was similar to that (8.7 h) reported for ponies (Cabrera, 1995).
The greater MRT of small particles and the trend for greater MRT of large particles in the hindgut of horses fed the HS diet, which are in agreement with the results obtained by Yoder et al. (1997) on ponies, could contribute to the improvement in DM and OM digestibility that was noted. Contrary to Nuss et al. (1982), who observed that supplying long-form forages accelerated the rate of passage of digesta in the proximal part of digestive tract located before the large intestine, a similar transit time of solid digesta (8 h) was observed in both the HS and HF diets. The observation that animals were fed a limited amount of finely ground and pelleted feed may have reduced the strict fiber effect on transit that was observed with long forages.
Effect of Saccharomyces cerevisiae CBS 493.94 Supplementation on Digestion in Horses
Yeast supplementation had a positive overall effect on ADF apparent digestibility independently of diet. Because microbial digestion of the cell wall fraction occurs exclusively in the hindgut, the extent of microbial digestion is the result of the product of [microbial cellulolytic activity expressed as the amount of degraded cellulose per h = IVADADF] x [Time available for digestion expressed as the MRT in the hindgut, in h]. Because SC had no effect on the MRT of solid particles in the hindgut, the increase in the digestion of ADF material following SC supplementation is likely to be due to the stimulation of microbial cellulolytic activity in the hindgut, as confirmed by the IVADADF values (Table 5). Positive effects of SC on cellulose digestion in horses have previously been reported by Glade and Sist (1988), Glade (1991a,b, 1992), Kim et al. (1991), and Medina et al. (2002), whereas other studies showed no effect of yeasts (Glade and Biesik, 1986; Hall et al., 1990). These discrepancies stemmed from a yeast strain effect and the concentrations of viable cells in the added preparations, as well as the experimental conditions, especially feeding conditions.
Because no literature or data on the effect of SC supplementation on the MRT of solid particles through the hindgut and the whole digestive tract of horses are available, the authors used data on ruminants to validate their results. The absence of an effect of SC supplementation on MRT of solid particles shown here has previously been reported by Plata et al. (1994) in beef cattle fed an oat straw-based diet and by Jouany et al. (1998) in sheep fed a mixed hay and barley diet dosed with 10 g of a preparation containing S. cerevisiae CBS 493.94 and 109 cfu/d of S. cerevisiae CNCM-1096, respectively.
The mode of action of live yeasts on digestive microbes has been extensively studied in ruminants (Dawson, 1990; Wallace and Newbold, 1992; Jouany, 2006) and horses (Glade and Biesik, 1986; Glade and Sist, 1988; Hall et al., 1990; Glade, 1991a,b; Kim et al., 1991; de Vaux and Julliand, 1994; Rowe et al., 1994; Medina et al., 2002). The microbial digestive disorders due to a negative digestive interaction between starch and cellulose that are expected to be corrected by yeasts would likely be less pronounced in the hindgut of horses than in the rumen, because most of the starch ingested is normally digested in the small intestine and only a small fraction reaches the hindgut. In addition, the positions of the rumen and the hindgut in the digestive tract make it likely that a lower concentration of live yeasts would reach the hindgut compared with the rumen for the same orally administered dose. However, the SC concentrations detected in the cecum (4.3 x 106 cfu/g) and in the colon (between 103 and 105 cfu/g) of horses given a 10-g oral dose of an SC preparation containing 4.5 x 109 cfu/g (Medina et al., 2002) were similar to those found in the rumen of sheep receiving a similar dose of yeasts (Durand-Chaucheyras et al., 1998).
In conclusion, the current study indicated that SC significantly improved the digestibility of the cellulose fraction in horses, regardless of diet. The results suggest that SC supplementation improved chiefly the activity of the complex microflora involved in digestion of the dietary ADF fraction.
|
Footnotes |
|---|
2 Corresponding author: v.julliand{at}enesad.fr
Received for publication December 5, 2006. Accepted for publication August 30, 2007.
|
LITERATURE CITED |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
This article has been cited by other articles:
|
|
 |
|
  J.-P. Jouany, B. Medina, G. Bertin, and V. Julliand Effect of live yeast culture supplementation on hindgut microbial communities and their polysaccharidase and glycoside hydrolase activities in horses fed a high-fiber or high-starch diet J Anim Sci, September 1, 2009; 87(9): 2844 - 2852. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
