| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ANIMAL NUTRITION |
* Department of Animal and Food Science, Facultat de Veterinària, Universitat Autònoma de Barcelona, 08193 Bellaterra, Spain; and
and
Danish Institute of Agricultural Sciences, Department of Animal Nutrition and Physiology, Research Centre Foulum, Tjele, Denmark
 |
Abstract |
|---|
 Top Abstract INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
Key Words: energy • fermentation • in vivo-in vitro methodology • nonstarch polysaccharide • pig
|
INTRODUCTION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
). Moreover, fiber may improve intestinal health because it is necessary for the stimulation of the intestinal compartments (Longland et al., 1994), and it is usually associated with a reduction of potentially harmful products from protein fermentation (McBurney et al., 1987). However, the inclusion of fiber in the diet offered to pigs is related to reductions in foregut and whole-tract digestibility of OM (Graham et al., 1986; Bach Knudsen and Hansen, 1991) leading to a lower absorption of nutrients and energy.
From a physiological point of view, dietary fiber is composed of dietary components resistant to degradation by mammalian enzymes but degradable by microbial fermentation (Bach Knudsen, 2001); nonstarch polysaccharides (NSP) and resistant starch are the most quantitatively important components of this fraction (Macfarlane and Cummings, 1991). Short-chain fatty acids (SCFA) produced during fermentation can be taken up by the epithelial cells and metabolized, supplying energy to the host (Bergman, 1990). The mean supply of energy from SCFA to the net energy for maintenance using different methodologies has been reported to be about 15 to 24% for growing and finish-ing pigs (Dierick et al., 1989; Yen et al., 1991; McBurney and Sauer, 1993).
The supply of energy from hindgut fermentation can be studied by different methodologies, in vivo, in vitro, or using an integration of both methodologies. The integration of in vivo-in vitro methods, where SCFA are not absorbed or metabolized, has been shown to be a valuable technique to measure the production of these fermentation products (McBurney and Sauer, 1993).
The aim of the present work was to investigate, using an in vivo-in vitro method, the available energy from hindgut fermentation to pigs fed diets differing in NSP concentrations.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
Experimental Diets, Animals, and Feeding
Three diets differing in the amount of NSP were formulated. The ingredients and the chemical composition of each diet are shown in Table 1. A standard Danish diet, considered as standard fiber diet (SFD), based on barley, wheat, and toasted soybean meal, was modified to obtain a low fiber diet (LFD) by replacing the cereal grains with wheat flour, and a high fiber diet (HFD) in which sugar beet pulp replaced a portion of the barley. The chemical analysis of the diets revealed that the amount of NSP were 77.4 g/kg of DM in LFD, 160.5 g/kg of DM in SFD, and 240.0 g/kg of DM in HFD. All diets contained 0.2% chromic oxide to be used as a digestibility marker.
|
The pigs were fed 3 times daily at 0700, 1500, and 2300. The daily amount of feed offered to the animals was adjusted weekly to 3% of their BW. The initial BW of the animals was 48.6 ± 1.32 kg, and at the end of the experimental period the BW was 85.2 ± 1.91 kg. Fecal samples were collected beginning on d 11, for 8 h and for 3 d, in plastic bags attached to the back of the pigs, and were changed 3 to 4 times per day. The collected samples were immediately frozen and stored at –20°C. Ileal samples were collected on 2 consecutive days, beginning on d 14, by attaching a tube-shaped plastic bag (length approx. 20 cm) to the cannula. On d 14, samples were collected from 0700 to 0900 and from 1100 to 1300, and on d 15 from 0900 to 1100 and from 1300 to 1500.
At the end of the collections, ileal and fecal samples were thawed by careful heating and stirring for minimal time to restrict microbial degradation, and pooled by pig and period, resulting in a total of 6 samples per diet and collection site. The samples were freeze-dried before analysis with the exception of samples taken for SCFA determinations. Ileal and fecal samples were analyzed for DM, chromic oxide, NSP, CP, and GE; feces were also analyzed for SCFA content.
In Vitro Incubations
After 7 d of adaptation to the experimental diets, feces were collected. Fecal slurries were prepared as described by Christensen et al. (1999), with a final concentration of feces in the fermentor of 5% (wt/vol).
The 18 ileal samples collected, 6 per diet, were subjected to in vitro fermentation in a batch system. Six fermentation series were carried out, each consisting of 3 fermentors containing fecal slurry and ileal digesta, and 3 containing only fecal slurry to be used as a blank. A volume of 700 mL of fecal slurry was added to each of the 6 fermentors, and 7 g of freeze-dried ileal digesta was added to 3 fermentors, one from each of the 3 diets. The total incubation volume contained 5% (wt/vol) fecal slurry and 1% (wt/vol) freeze-dried ileal digesta.
Three ileal samples, one from each of the 3 experimental diets, were tested at one time. The pH in the fermentors was automatically adjusted to 6.0 with 5 M NaOH and 5 M HCl, and the incubation temperature was kept constant at 38°C. Anoxic conditions in the fermentor were maintained with high purity N2. To homogenize the content, the first sample was taken after 1 min of stirring, which was considered as time 0. Samples from the fermentors were aseptically taken at 0, 3, 6, 12, 24, and 48 h of incubation. Short-chain fatty acid measurements were carried out for all samples, and microbial counts were performed on samples taken at times 0, 6, 24, and 48 h. Nonstarch polysaccharides were analyzed in freeze-dried samples at time 0 and at the end of fermentation (48 h).
Analytical Methods
Dry matter content was determined by drying the samples at 103°C to constant weight (Commission Directive 71/393/EEC), and ash content was determined by ashing the samples in a muffle oven at 525°C for 7 h (923.03; AOAC, 1990). Total N was determined by the Dumas method using an elemental analyzer, model CNS-2000 (Leco, St. Joseph, MI), as described by Hansen (1989). Gross energy was determined in a Leco AC 300 automated calorimeter system 789-500 (Leco, St. Joseph, MI). Neutral NSP were analyzed as alditol acetates by gas chromatography, and uronic acids were analyzed by a colorimetric assay, as previously described (Bach Knudsen, 1997). Chromic oxide was determined colorimetrically using a Lambda 900 spectrophotometer (Bodenseewerk, Perkin-Elmer GmbH, D-Überlingen), after oxidation to chromate with sodium peroxide following the procedure of Schurch et al. (1950). Analysis of SCFA and lactic acid was performed by gas chromatography (HP-6890 Series Gas Chromatograph, Hewlett Packard Palo Alto, CA) using an HP-5 column (30 m x 0.32 mm x 0.25 µm) with 5% phenylpolysiloxane and 95% dimethylpolysiloxane and a flame ionization detector, after submitting the samples to an acid-base treatment followed by ether extraction and derivatization, as described by Jensen et al. (1995).
Samples from the fermentors (10 mL) were transferred immediately after collection under a flow of CO2 into flasks containing 90 mL of a salt medium (Holdeman et al., 1977) containing: 902 mg/L of NaCl, 451 mg/L of KH2PO4, 27 mg/L of CaCl2·2H2O, 20 mg/L of MgCl2·6H2O, 12 mg/L of MnSO4·4H2O, 10 mg/L of CaCl2·6 H2O, 902 mg/L of (NH4)2SO4, 8 mg/L of FeSO4·7H2O, 451 mg/L of K2HPO4·3H2O, 2.5 g/L of yeast extract (Merck 1.03753), 2.5 g/L of peptone from casein (Merck, 1.07213), 1 mL/L of resazurin, 2.5 mL/L of hemin, 1 mL/L of tween-80, and 10 mL of a reducing solution. The reducing solution contained 50 g/L of NaHCO3, 2.5 g/L of cysteine hydrochloride, and 2.5 g/L of NaS. The suspension was then transferred to a CO2-flushed plastic bag and homogenized in a stomacher blender (model number BA6021, Seward Medical, London, UK) for 2 min. Then, 10-fold dilutions were prepared in prereduced salt medium by the technique of Miller and Wolin (1974).
Samples (0.1 mL) were plated on or inoculated to selective and nonselective media. Total anaerobic bacteria were enumerated by culturing the samples in roller tubes containing colon fluid-glucose-cellobiose agar (Holdeman et al., 1977) and incubating them anaerobically at 38°C for 7 d. Lactic acid bacteria were enumerated on the Man, Rogosa, and Sharp agar (Merck 1.10660, Darmstadt, Germany), after anaerobic incubation at 38°C for 2 d. Enterobacteria were enumerated on McConkey agar (Merck 1.05465), after aerobic incubation at 37°C for 1 d. Yeasts were enumerated on malt chloramphenicol agar with 10 g/L of glucose (Merck 1.08337.1000), 3 g/L of malt extract (Merck 1.05397), 3 g/L of yeast extract (Merck 1.03753), 5 g/L of peptone (Merck 1.07224), 50 mg/L of chloramphenicol (Sigma-Aldrich Chemie Gmbh C-0378, Steinheim, Germany), and 15 g/L of agar (Merck 1.01614), after aerobic incubation at 37°C for 2 d.
Calculations
In vivo ileal flow of dietary components, the amount of dietary components excreted in the feces, and net disappearance of NSP in the hindgut were calculated relative to the insoluble marker (Cr2O3). Results are expressed per kilogram of DM of feed rather than by day because of the weight and feed intake differences between the pigs.
Samples taken from the fermentors were used to calculate in vitro disappearance of NSP and SCFA production. Before calculations of the in vitro data, the concentration of NSP and SCFA determined in the fermentors without ileal substrate were subtracted from the corresponding samples incubated at the same time. Net disappearance of NSP per kilogram of DM of feed in the in vitro fermentation was calculated as
 |
where Dt48 is the net disappearance of NSP after 48 h of fermentation, Xt0 is the concentration of NSP at time 0, and Xt48 is the concentration of NSP at the end of fermentation. The energy equivalent of NSP was calculated as 17.6 kJ/g of NSP disappeared, DM basis.
In vitro production of SCFA at different sampling times was expressed as millimoles per gram of DM in the ileal effluent. Estimated total production of SCFA per kilogram of feed DM, which in the present work includes acetic (A), propionic (P), butyric (B), and valeric (V) acids, was calculated as follows:
 |
The energy equivalent of each SCFA (kJ/mol) was assumed as follows: acetic acid = 875, propionic acid = 1,528, butyric acid = 2,185, and valeric acid = 2,839 (CRC, 1977).
Statistical Analyses
The statistical analyses were performed with SAS (version 8, SAS Inst. Inc., Cary, NC). The in vivo data and the in vitro disappearance of NSP and total production of SCFA after 48 h of fermentation were analyzed using the following GLM:
 |
where Yijk is the dependent variable; µ is the overall mean; 
i is the effect of diet; i = 1, 2, 3; ßj is the effect of period, j = 1, 2, 3; 
is the effect of animal, k = 1,...,6; and 
~ N(0,
2) represents the unexplained random error.
Production of SCFA and microbial counts measured at different fermentation times were analyzed following the mixed model of SAS:
 |
where Yijk is the dependent variable; µ is the overall mean; i is the effect of diet, i = 1, 2, 3; ßj is the effect of fermentation time; j = 1,..., 6 for SCFA production and j = 1,..., 4 for microbial counts; (ß)ij is the interaction between diet and sampling time; Uk is the variance component that accounts for the repeated measurements made on the same fermentor with 1 diet; and ~ N(0,
) represents the unexplained random error.
Comparisons and regressions of the data from the in vivo and in vitro methodologies were made using t-test and PROC REG of SAS.
The alpha level used for determination of significance for all the analysis was 0.05. Differences between means were compared by Tukey’s least significant difference. Data from the in vitro experiment is presented as least squares means because values from 2 fermentors were determined to be outliers using the Univariate Procedure of SAS, 1 from SFD and HFD, giving a sample size of 5; thus, SEM is specified for n = 5 or n = 6.
|
RESULTS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
). Increasing the amount of NSP in the diet resulted in an increase (P < 0.001) in the amount of DM entering the cecum per kilogram DM of feed, ranging from 199 g for the pigs fed LFD to 468 g for the pigs fed HFD. Differences in the amount of ileal digesta were partly explained by the significant differences in the ileal flow of protein and of NSP, which increased as the amount of dietary fiber increased.
|
In Vivo and In Vitro Disappearance of Nonstarch Polysaccharides
The disappearance of NSP in the hindgut of the cannulated animals (g/kg of feed DM) and in vitro are presented in Table 3. No significant differences between the in vivo and in vitro measurements were observed, except for a lower in vitro disappearance of the NSP contained in the LFD diet (P = 0.006). This is due to the lower in vitro disappearance of glucose (P = 0.021), which was only 40% of the in vivo disappearance.
|
), the greatest correlations were found for uronic acids and arabinose and the poorest for glucose and xylose.
In Vitro Production of Short-Chain Fatty Acids
The amount of SCFA produced after 48 h of fermentation in vitro is presented in Table 4. No differences in the total SCFA produced per gram of ileal digesta were observed among diets. The amount of SCFA produced per kilogram of feed DM increased (P < 0.001) with increasing amounts of NSP in the diets due to the greater ileal flow of DM with increasing concentrations of dietary NSP. Molar ratios of SCFA showed that acetic acid had the greatest ratio at the end point in all diets, followed by propionic, butyric, and valeric acid. The proportion of acetic acid found after fermentation of ileal digesta from the HFD-fed pigs differed significantly from that of pigs fed the SFD diet and the LFD diet, whereas the proportion of butyric acid tended to be affected by the diet (P = 0.081), with the molar ratio in the LFD treatment being greater than in the HFD treatment.
|
. Production of SCFA with the different diets during fermentation increased rapidly from 0 to 12 h, showing thereafter a lower rate of production, but not reaching a plateau after 48 h of fermentation. Although no differences were seen in the total production of SCFA at the end point, an interaction between diet and time of fermentation (P = 0.016) was observed, due to lower SCFA production with the LFD diet during the first hours of fermentation. After 3 h of fermentation, the LFD diet showed a lower production of total SCFA compared with the other 2 diets; 0.98 mmol/g of DM in the ileal effluent for LFD, 2.75 mmol/g of DM in the ileal effluent for SFD, and 2.00 mmol/g of DM in the ileal effluent for HFD, not reaching the same values as the other 2 diets until 12 h of fermentation. Lower production of total SCFA with the LFD at the beginning of fermentation was associated with a lower production of the SCFA measured in this study and a greater concentration of lactic acid (up to 1 mmol/g of DM in the ileal effluent after 3 h of incubation). The concentration of lactic acid with the LFD diet was greatest from 3 to 6 h, and it was no longer detected after 24 h of incubation in any diet. Production of branched chain fatty acids with the HFD diet was lower throughout the fermentation, the differences between diets being bigger with increasing time of fermentation. After 48 h of incubation, the greatest value was found with the LFD diet (0.09 mmol/g of DM in the ileal effluent) being greater than the SFD diet (0.06 mmol/g of DM in the ileal effluent, P = 0.014) and the HFD diet (0.03 mmol/g of DM in the ileal effluent, P < 0.001).
|
. As expected, microbial counts at the beginning of the fermentation were not different in the fermentors with ileal effluent compared with the corresponding blank fermentors (data not shown).
|
0.037). Microbial groups showed different behaviors during the time course of fermentation. In all diets, enterobacteria and yeast counts decreased with time of fermentation, the counts after 48 h being lower than the initial counts (P 0.020). On the contrary, total anaerobic bacteria showed greater counts after 12 and 24 h compared with the initial values (P 0.001), and the values found after 48 h were not different from the initial values. Changes in lactic acid bacteria counts with time of fermentation depended on the diet as shown by the significant interaction between diet and time (P = 0.029). Lactic acid bacteria counts in the pigs fed LFD and SFD were lower after 48 h of fermentation compared with initial values (P 0.036), and no effect of time was measured in the HFD-fed pigs.
Energy
The amount of available energy from foregut and hindgut digestion is presented in Table 6. The largest amount of energy (MJ/kg of feed DM) was digested in the foregut and was greatest for the diets with lower content of NSP (P < 0.001). In contrast, the energy digested in the hindgut increased (P < 0.001) with the increasing amounts of NSP in the diet. The energy digested in the foregut and in the hindgut was also affected by the source of dietary NSP (P = 0.007 and P = 0.010, respectively). Relative to LFD, the energy that disappeared in the foregut was reduced by 0.036 MJ/g of NSP added in the SFD and 0.027 MJ/g of NSP added in the HFD. Energy digested in the hindgut increased 0.016 MJ/g of NSP when added in the SFD and 0.012 MJ/g of NSP when added in the HFD.
|
0.02). From these in vitro results, it could be calculated that, of the total amount of energy digested in the hindgut of the pigs, 67% was available to the animal as SCFA when feeding HFD and SFD, and 74% when feeding LFD. The contribution of absorbed SCFA to the total available energy of the animal was affected by diet (P < 0.001) with 7.1% of the total when LFD was fed, 13.6% with SFD, and 17.6% with HFD.
|
DISCUSSION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
; Morales et al., 2002) and in vitro (Stephan, 1983; Barry et al., 1995; Auffret et al., 2004) fermentability of NSP has been related to chemical structure and composition, as well as with its physical characteristics. The difference in fermentability between the HFD and SFD diet reflects the differences in the fibrous ingredients present in the diets (i.e., sugar beet pulp in the HFD diet and barley in SFD diet). Cell walls from sugar beet pulp consist of arabinogalactans and cellulose embedded in a pectic matrix, conforming a homogeneous network of polysaccharides (Bertin et al., 1988), which confers high macroporosity and water retention capacity (Auffret et al., 2004). The composition and structural characteristics result in high fermentability of sugar beet pulp by colonic bacteria (Graham et al., 1986; Stevens and Selvendran, 1988). In contrast, NSP from the husk of barley are glucuronoarabinoxylans and crystalline cellulose, conforming a heterogeneous structure, which makes the fiber less available for fermentation (Fardet et al., 1997).
Disappearance of NSP in the hindgut of pigs was reasonably well estimated by the in vitro procedure used in the current study (r = 0.960, P < 0.001). However, the in vitro procedure underestimated the in vivo disappearance of NSP from the LFD diet, specifically of NSP glucose. Using the present methodology, different results have been found when comparing the disappearance of NSP in vitro and in vivo. Christensen et al. (1999), in an experiment with pigs fed 2 wheat-based diets or an oat bran-based diet, did not find differences between in vivo and in vitro disappearance of total NSP or their component saccharides when fermenting ileal effluents for 72 h. However, Glitsø et al. (2000), in an experiment where pigs were fed rye bread diets differing in the amount of milling fractions, found differences between the in vivo and in vitro values when fermenting ileal effluents for 48 h. They suggested that the disagreement between their results and those of Christensen et al. (1999) could be related to the fact that pigs used to obtain the feces to prepare the inoculums were not adapted to the experimental diets, to differences in the amount of NSP in the ileal effluents used as substrate, or to a shorter length of in vitro fermentation. At the same time, in the study of Glitsø et al. (2000), the in vivo and in vitro disappearance of arabinoxylans was not different, suggesting that the environment in the fermentor could have enhanced the activity of arabinoxylans degrading bacteria. In the current study, inoculums were prepared with feces collected from animals adapted to the experimental diets. Furthermore, differences between in vivo and in vitro NSP disappearance, mainly due to glucose, were found in the diet with the lowest percent of soluble NSP glucose. The solubility of the fiber components is known to affect its fermentability. For example, cellulose, an insoluble component of the NSP, is known to be fermented in vitro at a lower rate than other NSP, only reaching a significant extent of fermentation when more fermentable components, such as acidic monomers, are almost completely fermented (Salvador et al., 1993). Moreover, Morales et al. (2002) found that disappearance of NSP glucose in the hind-gut of pigs can be compromised when transit time in the gastrointestinal tract is reduced. Thus, the fact that glucose was digested less in the diet containing a more insoluble NSP glucose suggests that the length of in vitro fermentation has an important role in the differences found in the current study between in vitro and in vivo disappearances of glucose. On the other hand, possible modifications of the microbiota during preparation of the inoculum or during in vitro fermentation could be detrimental to the bacteria that ferment this component.
Production of Short Chain Fatty Acids
Measurement of SCFA production in vivo is difficult due to their rapid absorption from the gut lumen (Argenzio and Southworth, 1975). In contrast, in vitro fermentation of ileal digesta provides a reliable technique to estimate SCFA production because these products are not absorbed (McBurney and Sauer, 1993). The nature and yield of fermentation products is known to be dependent on the type and amount of available substrate (Barry et al., 1995), as well as on the microorganisms involved (Bernalier et al., 1999). Several studies have shown that the different fiber sources are fermented at different rates and release different amounts of SCFA (Barry et al., 1995). However, the amount of SCFA per kilogram of dry feed in the current study was clearly dependent on the amount of ileal effluent rather than on the origin of ileal digesta. Moreover, the presence of rapidly fermentable substrates, like sugar beet pulp, did not result in a faster production of SCFA during fermentation. The fact that the ileal digesta fermented contained NSP, protein, and other organic compounds, rather than fiber alone, could partially explain the absence of major differences in the SCFA production.
The molar ratio of the main SCFA released during in vitro fermentation was affected by diet. Different SCFA ratios have been reported in several studies after fermentation of raw ingredients (Englyst and Macfarlane, 1987; Salvador et al., 1993; Barry et al., 1995) and ileal effluents (Wang et al., 2004). Englyst and Macfarlane (1987) showed that fermentation of starch results in a greater proportion of butyrate, whereas xylans and pectins promote the production of acetate. Similarly, Wang et al. (2004) reported that fermentation of ileal digesta from pigs fed a diet containing 12% sugar beet pulp produced greater proportions of acetic acid compared with other diets, which is in agreement with the results observed in the current study. Lactic acid concentration increased during the first hours when fermenting ileal effluent from pigs fed the LFD diet, and disappeared with time of fermentation, reflecting the intermediate character of this product (Macfarlane and Englyst, 1986). Lactic acid can be utilized by several bacteria, like Propionibacterium acnes, Veillonella spp., Megasphaera elsdenii, Clostridium spp., and dissimilatory sulphate-reducers, and it is converted into acetic, propionic, butyric, and longer chain fatty acids (Bernalier et al., 1999). Production of lactic acid is favored by the fermentation of soluble carbohydrates (Bergman, 1990) and starch (Macfarlane and Englyst, 1986; Macfarlane and Gibson, 1994). The high concentration of lactic acid and starch in the LFD diet could indicate some starch is escaping foregut digestion in animals fed this diet.
Microbial Counts
The amount and type of substrate available to the microbiota is probably the most important control for microbial fermentation in the gastrointestinal tract of monogastric animals (Jensen, 2001). In terms of amount of available substrates to the microbiota in the hindgut, the carbohydrate fraction represents the main dietary component (Bach Knudsen, 2001). Different results are found in the literature with respect to the influence of fiber in the growth of the different bacteria. McDonald et al. (1999) and Hopwood et al. (2004) reported an increase in postweaning colibacillosis when feeding piglets diets with high amounts of fermentable NSP. Similarly, Pluske et al. (1996) demonstrated that swine dysentery can be exacerbated with the presence of a fermentable substrate, mainly soluble nonstarch polysaccharides but with a possible involvement of resistant starch. However, the relationship between soluble NSP and swine dysentery was not confirmed by other works (Kirkwood et al., 2000; Lindecrona et al., 2003). Furthermore, Reid and Hillman (1999) reported a reduction of coliforms and an increment in the production of SCFA in the hindgut of the animals when retrograded starch was added to the diet of pigs.
In the current study, microbial counts in the inoculums prepared from feces of animals fed the 3 experimental diets showed that LFD diet resulted in greater enterobacteria counts compared with SFD and greater yeast counts compared with HFD. Fermentation of dietary fiber in the gastrointestinal tract results in decreased digesta pH (Jensen and Jorgensen, 1994) and increased production of organic acids, which, in turn, can have an effect on pathogenic bacteria. Wang et al. (2003) found a relationship between low pH and low presence of enterobacteria in the stomach of gestating sows.
The increasing counts of anaerobic bacteria during the first hours of in vitro fermentation of ileal effluent indicate that this bacterial group was more active, irrespective of the diet, than the other microbial groups measured. At the same time, lactic acid bacteria counts in the HFD diet did not decrease with time of fermentation, and the value after 48 h was greater in this diet compared with the other 2 diets. The presence of fermentable carbohydrates can stimulate the growth of certain microorganisms, some compounds having a more specific effect on particular microbial species, such as lactobacilli (Rycroft et al., 2001; Rastall and Gibson, 2002). Little information is available on the effects of sugar beet pulp components on the growth of lactic acid bacteria, but Konstantinov et al., (2004) reported a greater prevalence of Lactobacillus reuteri and Lactobacillus amylovorus-like populations in the ileum and the colon in piglets fed a diet containing inulin, wheat starch, lactulose, and sugar beet pulp.
Disappearance and Availability of Energy
Increasing the concentration of NSP in the diet led to a diminution of the energy digested in the foregut and in the whole gastrointestinal tract. Thus, the increased fermentation in the hindgut of the animals fed high concentrations of NSP compared with those fed the LFD diet was not able to compensate for the significant reduction of energy digested in the foregut. The observed decreased foregut and total energy digestibility with increasing concentration of dietary NSP is in accordance with data reported in the literature (Just et al., 1983).
Irrespective of diet, NSP constituted the biggest fraction being fermented in the hindgut. However, energy yield estimated from in vivo NSP fermentation did not account for the energy disappearance from the hindgut of the animals (53% for LFD, 48% for SFD, 77% HFD). In an experiment carried out by Christensen et al. (1999), the energy yield from NSP fermentation accounted for 40 to 47% of the energy that disappeared within the hindgut. Differences between the 2 experiments could be explained by the greater content of NSP in the present diets (77 to 240 g of NSP/kg of feed DM) compared with the diets used by Christensen et al. (56 to 102 g of NSP/kg of feed DM). At the same time, the energy yield from SCFA produced in vitro was greater than the energy yield from NSP that disappeared during in vitro fermentation. Assuming a fixed energy yield from NSP fermentation of 75% (McNeil, 1984), the contribution of NSP to the production of SCFA ranged from 40% with the LFD diet to 85% with the HFD diet. Thus, in the current study, the fermentation of other substrates was quantitatively important, especially when fermenting ileal effluent from diets containing low concentrations of NSP. These fermented substrates may include protein (Macfarlane et al., 1992), resistant starch (Wang et al., 2004), and endogenous material (Bergman, 1990). In the current study, the amount of NSP fermented in the hindgut increased with the amount of dietary NSP. However, the amount of protein fermented at this site was rather similar for all dietary groups (13 to 16 g/kg of feed DM). This implies that protein accounted for a relatively small amount of the total substrates fermented in the hindgut as the dietary NSP increased.
The SCFA released after incubation of ileal effluents in the in vitro system allows for the estimation of the energy available to the animal from hindgut fermentation. As mentioned above, end products of fermentation varied with the composition of the ileal effluents, affecting the energy available to the animal per gram of ileal DM. This was clearly seen when comparing HFD and SFD, while the production of SCFA increased by 33% with the HFD diet with respect to the SFD diet, the energy yield from SCFA increased only by 20%, as a consequence of a greater contribution of acetic acid to the total production of SCFA in HFD. The energy released as SCFA accounted for approximately 67 to 74% of the energy disappearance in the hindgut, which is in agreement with the results reported by Müller and Kirchgessner (1985). Differences between energy digested in the hindgut and the energy produced in vitro as SCFA may be a result of the lower in vitro disappearance of NSP, especially in LFD, as well as a simultaneous energy flow to methane, hydrogen, heat, and build-up of microbial biomass (Noblet and Le Goff, 2001).
The contribution of fermentation products in the large intestine to available energy increased with increasing dietary amounts of NSP, ranging from 7.1% with the LFD to 17.6% with HFD. In the literature, the energy obtained from microbial fermentation accounts for 2.4 to 29.5% of the total digestible energy (Jensen, 2001). Differences in the amount and origin of NSP used in the various trials as well as in the methodologies may account for the differences in the results (Jensen, 2001). Jensen et al. (1997), using the same method as used in the current study, reported that up to 16.4% of the total energy supply may be absorbed in the hind-gut of pigs fed a diet containing 20.1% NSP, which is in line with the results obtained in the current study.
|
IMPLICATIONS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
|
Footnotes |
|---|
2 Corresponding author: montserrat.anguita{at}uab.es
Received for publication April 20, 2005. Accepted for publication May 24, 2006.
|
LITERATURE CITED |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, n = 6), a standard fiber diet (
, n = 5), or a high fiber diet (
, n = 5). Results are expressed as least squares means ± SEM. a–cMeans within fermentation time without a common superscript letter differ (P < 0.05).