Introduction
Mixed linked b (1→3, 1→4)-d-glucan (b-glucans) are cell wall nonstarch polysaccharides (NSPs) of barley endosperm and are classified as dietary fibre. Barley b-glucans resist hydrolysis by mammalian alimentary enzymes but can be hydrolysed and fermented by gastro-intestinal microflora through the secretion of hydrolases (Roberfroid, Reference Roberfroid1993) in the front and hindgut (Bach Knudsen and Hansen, Reference Bach Knudsen and Hansen1991). b-glucans are readily fermented by bifidobacteria and lactobacilli species, promoting cell proliferation and production of short-chain fatty acids (SCFAs) (Charalampopoulos et al., Reference Charalampopoulos, Wang, Pandiella and Webb2002; O’Connell et al., Reference O’Connell, Callan, Byrne, Sweeney and O’Doherty2005). A rise in SCFA concentration through fermentation lowers digesta pH, thus creating an unsustainable environment for pH-sensitive pathogenic bacteria. A low carbohydrate-to-protein ratio in the large intestine leads to an increase in proteolytic fermentation (Aumaitre et al., Reference Aumaitre, Peiniau and Made1995). Ammonia and odorous branched-chain fatty acids (BCFAs) are produced from the fermentation of amino acids (Macfarlane and Macfarlane, Reference Macfarlane and Macfarlane2003). Isobutyric and isovaleric are metabolic end products from the fermentation of valine and leucine (Mackie, Reference Mackie1995). However, if the level of NSPs available for microbial fermentation are increased, BCFAs, odour and ammonia levels are reduced (O’Connell et al., Reference O’Connell, Callan, Byrne, Sweeney and O’Doherty2005; Garry et al., Reference Garry, Fogarty, Curran and O’Doherty2007). By increasing the carbohydrate to protein ratio entering the large intestine, nitrogen (N) excretion shifts from the more volatile urinary N to faecal N (Mroz et al., Reference Mroz, Moeser, Vreman, van Diepen, Van Kempen, Canh and Jongbloed2000; O’Connell et al., Reference O’Connell, Callan and O’Doherty2006), which is more stable.
However, barley b-glucans also exhibit anti-nutritive properties. The physical enclosure of dietary nutrients by cell wall NSPs prevents the diffusion of nutrients and digestive enzymes (De Lange, Reference De Lange2000) resulting in a reduction in digestibility (Johnson and Gee, Reference Johnson and Gee1981) and average daily gain of pigs offered barley-based diets compared with wheat-based diets (Nielsen et al., Reference Nielsen and Ingvartsen2000). O’Connell et al. (Reference O’Connell, Callan, Byrne, Sweeney and O’Doherty2005) showed that total tract apparent digestibility is improved with exogenous b-glucanase supplementation; however, the beneficial properties of b-glucans in barley are removed. Therefore, the objective of this study was to investigate the effects of increasing the intake of dietary b-glucans by exchanging wheat for barley on diet digestibility, N retention, N excretion, intestinal microflora, volatile fatty acid (VFA) concentration and manure ammonia emissions in finishing pigs.
Material and methods
All procedures described in this experiment were conducted under experimental licence from the Irish Department of Health in accordance with the Cruelty to Animals Act (1876) and the European Communities (Amendments of the Cruelty to Animals Act, 1976) Regulations (1994).
Experimental design and diets
The experiment was designed as a complete randomised design comprising four dietary treatments. All diets were formulated to have identical concentrations of digestible energy (DE) (13.7 MJ/kg) and ileal digestible lysine (8.5 g/kg) (Sauvant et al., Reference Sauvant, Perez and Tran2004). Barley replaced wheat and soya oil was adjusted to maintain DE. The ileal digestible amino acid requirements were met relative to lysine (Close, Reference Close1994). High protein diets (200 g/kg) were used in the current study so that the effects of b-glucan on protein degradation could be examined. The experimental treatments were as follows: (T1) 0 g/kg barley (control diet) (5.6 g/kg b-glucan), (T2) 222 g/kg barley (12.1 g/kg b-glucan), (T3) 444 g/kg barley (18.9 g/kg b-glucan) and (T4) 666 g/kg barley (25 g/kg b-glucan). All diets were fed in meal form. The dietary composition and analysis are presented in Table 1.
Table 1 Composition and analysis of experimental diets
†Provided per kg of complete diet: 3 mg retinol, 0.05 mg cholecalciferol, 40 mg alpha-tocopherol, 90 mg copper as copper II sulphate, 100 mg iron as iron II sulphate, 100 mg zinc as zinc oxide, 0.3 mg selenium as sodium selenite, 25 mg manganese as manganous oxide and 0.2 mg iodine as calcium iodate on a calcium sulphate/calcium carbonate carrier.
‡Sauvant et al. (2004).
Animals and management
Twenty-four finishing boars (progeny of Meat line boars × (Large White × Landrace sow)) with an initial live weight of 66 (s.d. = 3.4) kg were used in this experiment. The pigs were blocked on the basis of live weight and were randomly allocated to one of four dietary treatments. The pigs were allowed a 14-day dietary adaptation period after which time they were weighed and 16 pigs were selected according to a uniform weight and transferred to individual metabolism crates. The pigs were given a further 5 days to adapt to the metabolism crates before collections began. The collection period was subdivided into two parts to facilitate studies on ammonia emission (days 1 to 2) and apparent digestibility and N balance (days 3 to 7). The daily feed ad libitum allowance (DE intake (MJ/day) = 3.44 × (live weight)0.54 (Close, Reference Close1994) was divided over two meals and was based on an anticipated DE concentration of 13.7 MJ/kg. The mean feed intake was 2.00 (±0.05) kg. Water was provided with meals in a 1:1 ratio. Between meals, fresh water was provided ad libitum. The metabolism crates were located in a temperature-controlled room, maintained at 22°C (±1.5°C).
Ammonia emissions
Four separate collections of total faeces and urine were taken at 12-h intervals during collection days 1 to 2. Following collection, the excreta were stored separately in sealed containers at 4°C. After the last collection, the urine and faeces samples were mixed together (w/w) according to the original excretion ratio. Samples (2 kg) of the manure homogenate from each pig were placed in duplicate, in sealed containers within a climate-controlled room maintained at 20°C. Ammonia emission from the manure was measured over 240 h from the first container, in a laboratory scale set-up according to the method of Derikx and Aarnink (Reference Derikx and Aarnink1993). The equipment consisted of a vessel containing 2 kg slurry, vacuum pump and three impingers in series per sample. The first two impingers contained 1 mol/l nitric acid and the third impinger contained water. The ventilation rate in the container was 4.2 l/min. The first impinger was replaced at 48, 96 and 144 h and the second impinger was replaced at 96 h. Samples were taken from all three impingers at 240 h. The concentration of ammonia-nitrogen (NH3-N) in the impingers was determined by the micro-diffusion technique of Conway (Reference Conway1957). Ammonia production (mg) from manure is compared between the different dietary treatments using the quantity volatilised from 0 to 240 h/g of N intake. The sample in the second ventilated container was used to conduct pH analysis of the slurry whenever the first impinger was replaced.
Total tract apparent digestibility and nitrogen balance study
During collections, urine was collected in a plastic container, via a funnel below the crate, containing 20 ml of sulphuric acid (25% H2SO4). To avoid N volatilisation, the funnel was sprayed four times daily with a weak solution of sulphuric acid (2% H2SO4). The urine volume was recorded daily and a 50-ml sample was collected and frozen for laboratory analysis. Total faeces weight was recorded daily and oven dried at 100°C. A sample of freshly voided faeces was collected daily and then frozen for N analysis. At the end of the collection period, the faeces samples were pooled and a subsample retained for laboratory analysis. Feed samples were collected each day and retained for chemical analysis.
Microbiology
All 24 pigs remained on their respective dietary treatments until slaughter. Digesta samples (approximately 10 (±1) g) were aseptically removed in aerobic conditions from the caecum and distal colon of each animal immediately after slaughter and stored in sterile containers (Sarstedt, Wexford, Ireland) on ice and transported to the laboratory within 7 h. Lactobacillus spp. and Enterobacteria were isolated and counted according to the method described by O’Connell et al. (Reference O’Connell, Callan, Byrne, Sweeney and O’Doherty2005). Lactobaccilus spp. were chosen because of their health-promoting properties (Gibson and Roberfroid, Reference Gibson and Roberfroid1995) while Enterobacteria were chosen because harmful effects of some species in the gastro-intestinal tract (Gibson and Roberfroid, Reference Gibson and Roberfroid1995).
pH measurements
Samples of digesta from the caecum and distal colon were taken and placed in universal containers. The pH of the digesta was taken on site, immediately after collection. All pH measurements were made on a Mettler Toledo MP 220 pH meter, which was calibrated with certified pH 4 and 7 buffer solutions. Distilled water was added to some very viscous samples to enable their pH to be read.
Volatile fatty acid sampling and analysis
Samples of digesta from the caecum and the colon of individual pigs (n = 24) were taken for VFA analysis. VFA concentrations in the digesta were determined using a modified method of Porter and Murray (Reference Porter and Murray2001). One gram of sample was diluted with distilled water (2.5 × weight of sample) and centrifuged at 1400 × g for 4 min (Sorvall GLC – 2B laboratory centrifuge). One ml of the subsequent supernatant and 1 ml of the internal standard (0.5 g 3-methyl-n-valeric acid in 1 l of 0.15 mol/l oxalic acid) were mixed with 3 ml of distilled water. Following centrifugation to remove the precipitate, the sample was filtered through Whatman 0.45 μm polyethersulphone membrane filters into a chromatographic sample vial. One μl of the sample was injected into a model 3800 Varian gas chromatograph with a 25 m × 0.53 mm i.d. megabore column (coating CP-Wax 58 (FFAP) – CB (no. CP7614)) (Varian, Middelburg, The Netherlands).
Laboratory analysis of samples
Proximate analysis of diets for dry matter (DM) and ash were carried out according to Association of Official Analytical Chemists (1995). The DM of the feed and faeces was determined after drying for 24 h at 103°C. Ash was determined after ignition of a known weight of concentrates or faeces in a muffle furnace (Nabertherm, Bremen, Germany) at 500°C for 4 h. The gross energy (GE) of feed and faeces samples was measured using an adiabatic bomb calorimeter (Parr Instruments, IL, USA). The neutral-detergent fibre (NDF) and acid-detergent fibre (ADF) content of feed and faeces was determined using a Fibertec extraction unit (Tecator, Sweden) according to the method of Van Soest et al. (Reference Van Soest, Robertson and Lewis1991). The soluble and insoluble b-glucan content was determined according to McCleary and Glennie-Holmes (Reference McCleary and Glennie-Holmes1985). The crude protein content of the feed was determined as N × 6.25 using the LECO FP 528 instrument (Leco Instruments, UK Ltd, Newby Road, Hazel Grove, Stockport, SK7 5DA, Cheshire). The nitrogen content of faeces was analysed by the macro-Kjeldahl technique using a Buchii apparatus.
Statistical analysis
Linear and quadratic effects of dietary b-glucan concentration were tested using the general linear model procedure of Statistical Analysis System Institute (SAS, 1985). Metabolic live weight (live weight0.75) and daily feed intake were included as covariates in the model. The model is
where Yi is the dependent variable, μ is the overall mean, b1Xi is the linear effect of b-glucan inclusion, b2X i2 is the quadratic effect of b-glucan inclusion and ei is the residual error. The manure pH data measured over 10 days was performed by repeated measures analysis using the Proc Mixed procedure of SAS 6.14 (Littell et al., Reference Littell, Milliken, Stroup and Wolfinger1996). The individual pig served as the experimental unit. The probability level that denotes significance is P < 0.05. The data in the tables are presented as least-square means±standard error of the mean, which was calculated as a pooled value across treatments.
Results
Coefficient of total tract apparent digestibility and nitrogen balance study
The effect of dietary treatment on the coefficient of total tract apparent digestibility of nutrients and N balance are presented in Table 2. Dietary treatment had no significant effect on food or N intake. There was a linear decrease in the digestibility of DM (b1 = −0.00005; s.e. 0.00001; P < 0.001), organic matter (OM) (b1 = −0.00006; s.e. 0.000009; P < 0.001), GE (b1 = −0.00005; s.e. 0.00001; P < 0.001), NDF (b1 = −0.0002; s.e. 0.00004; P < 0.001), ADF (b1 = −0.0005; s.e. 0.00005; P < 0.001) and N (b1 = −0.00004; s.e. 0.00002; P < 0.05) as the level of barley in the diet increased.
Table 2 Effect of dietary barley β-glucan level on total tract nutrient apparent digestibility and nitrogen utilisation (least-square means ± s.e.)
† Linear = linear response to dietary barley, quadratic = quadratic response to barley. *P < 0.05, **P < 0.01, ***P < 0.001, ns = non-significant (P > 0.05).
Increasing levels of dietary barley resulted in a linear increase in faecal DM content (b1 = 0.004; s.e. 0.002; P < 0.05) and faeces output (b1 = 0.0003; s.e. 0.0002; P < 0.05). Faecal N excretion was affected by dietary barley level (quadratic, P < 0.05). Dietary treatments had no significant effect on N retention (P > 0.05).
Microbiology study
The effect of dietary treatment on selected microbial populations and pH in the caecum and colon are presented in Table 3. There was a linear decrease in Enterobacteria concentrations (b1 = −0.001; s.e. 0.0005; P < 0.05) with increasing dietary barley level in the caecum. Colonic (b1 = −0.0003; s.e. 0.0001; P < 0.01) and caecal digesta pH (b1 = −0.0003; s.e. 0.0001; P < 0.001) linearly decreased as the level of barley in the diet increased. Dietary treatments had no significant effect on Lactobacilli concentrations in the caecum or colon (P > 0.05).
Table 3 Effect of dietary barley β-glucan level on microbial ecology and pH in the caecum and colon (least-square means ± s.e.)
†Linear: linear response to dietary barley. There was no quadratic response to dietary barley (P > 0.05). *P < 0.05, **P < 0.01, ***P < 0.001, ns = non-significant (P > 0.05).
Volatile fatty acid study
The effect of dietary treatment on the concentration and profile of caecal and colonic VFA is shown in Table 4. Total VFA concentrations and proportion of propionic acid were affected by dietary barley level (quadratic, P < 0.05) in the caecum. There was a linear decrease in acetic acid (b1 = −0.0001; s.e. 0.00002; P < 0.001), isobutyric acid (b1 = −0.000003; s.e. 0.000001; P < 0.05), isovaleric acid (b1 = −0.00001; s.e. 0.000002; P < 0.05), acetic acid: propionic acid ratio (b1 = −0.001; s.e. 0.0002; P < 0.001) and a linear increase in propionic acid (b1 = 0.0001; s.e. = 0.00002; P < 0.001) with increasing levels of dietary barley in the caecum.
Table 4 The effect of dietary barley β-glucan level on total volatile fatty acids (VFA) concentration in digesta and molar propotions of VFA in the caecum and colon (least-square means ± s.e.)
†Linear = linear response to dietary barley, quadratic = quadratic response to barley. *P < 0.05, **P < 0.01, ***P < 0.001, ns = non-significant (P > 0.05).
There was a linear increase in the total concentration of VFA (b1 = 0.066; s.e. 0.023; P < 0.01) and in the molar proportions of butyric acid (b1 = 0.00003; s.e. 0.00001; P < 0.05) and propionic acid (b1 = 0.0001; s.e. 0.00002; P < 0.001) with an increasing barley inclusion level in the colon. There was a linear decrease in the proportion of acetic acid (b1 = −0.0001; s.e. 0.00002; P < 0.001), isobutyric acid (b1 = −0.00001; s.e. 0.000001; P < 0.01), isovaleric acid (b1 = −0.00001; s.e. 0.000003; P < 0.05) and acetic acid: propionic acid ratio (b1 = −0.001; s.e. 0.0002; P < 0.001) with increasing dietary barley level in the colon.
Ammonia emission study
The effects of dietary treatment on manure ammonia emissions and slurry pH during storage are presented in Table 5. There was no significant linear or quadratic effect of dietary barley on ammonia or slurry pH (P > 0.05).
Table 5 The effect of dietary barley level on ammonia production and slurry pH (least-square means ± s.e.)
†Linear = linear response to dietary barley. There was no quadratic response to dietary barley (P > 0.05).
Discussion
Cereals are one of the major sources of dietary fibre in monogastric nutrition; however, the composition of the fibre may vary depending on its source (De Lange, Reference De Lange2000). Both wheat and barley contain anti-nutritional factors such as arabinoxylans and b-glucans. Elevated levels of these polysaccharides, in particular the soluble fraction, can cause an increase in digesta viscosity, preventing interaction between nutrients and digestive enzymes (Campbell and Bedford, Reference Campbell and Bedford1992) and reducing nutrient digestibility (O’Connell et al., Reference O’Connell, Callan, Byrne, Sweeney and O’Doherty2005) in pigs. However, decreases in ammonia emission and odour have been reported when sources of fermentable fibre are added to the diet (Garry et al., Reference Garry, Fogarty, Curran and O’Doherty2007) and b-glucans have been shown to have prebiotic properties (O’Connell et al., Reference O’Connell, Callan, Byrne, Sweeney and O’Doherty2005). The objective of this study was to investigate the effects of increased levels of dietary b-glucans on diet digestibility, N metabolism, intestinal microflora, VFA concentration and manure ammonia emissions in finishing pigs.
Bach Knudsen and Hansen (Reference Bach Knudsen and Hansen1991) reported substantial bacterial fermentation of b-glucans in the small intestine of the pig. This was attributed to the high solubility and easy degradation of b-glucans. However, previous work from our laboratory has shown that the beneficial effects of barley b-glucans such as a reduction in odour and ammonia emissions were removed as a result of b-glucanase and xylanase addition, while enzyme inclusion had no effect on wheat-based diets (O’Connell et al., Reference O’Connell, Callan, Byrne, Sweeney and O’Doherty2005; Garry et al., Reference Garry, Fogarty, Curran and O’Doherty2007). This would indicate that not all of the b-glucans are broken down in the small intestine. If all the b-glucans were broken down in the small intestine, there would have been no response to the enzyme inclusion. Similarily, Högberg and Lindberg (Reference Högberg and Lindberg2004) found that digestion of b-glucans was more pronounced in the total tract than at the ileum indicating that some fermentation of b-glucans occurred in the large intestine. It is probably the soluble b-glucans that are broken down in the small intestine while the insoluble b-glucans pass into the large intestine, where they are fermented by bacteria. Barley b-glucans have been shown to be 70% soluble and 30% insoluble (Högberg, Reference Högberg2003). Secondly, this also would indicate that the response benefits obtained in the current study were due to an increasing barley content and not due to a decreasing wheat content in the diet.
There was a linear decrease in the apparent digestibility of DM, OM, NDF, ADF and GE with increasing levels of dietary barley. The linear decrease in digestibility coefficients corresponds to the elevated NSP concentrations. Analysis of the total b-glucan content of the diets was 5.6, 12.1, 18.9 and 25.4 g/kg and the insoluble b-glucan content was 1.0, 4.3, 7.8 and 10.7 g/kg for the control, low, medium and high barley diets, respectively. NSPs from cell walls can physically hinder the access of digestive enzymes to nutrients that are enclosed inside the cell wall. As a consequence, degradation and utilisation of nutrients is limited, resulting in a depression of digestibility (De Lange, Reference De Lange2000). The decrease in diet digestibility recorded in the current experiment would indicate that pig performance could be compromised. However, dietary treatment had no effect on N retention in the current study. Previous work at this laboratory indicates no significant reduction in performance of grower-finisher pigs offered barley v. wheat-based diets (Garry et al., Reference Garry, Fogarty, Curran and O’Doherty2007). The lack of response maybe due to the fact that the diets were formulated to have similar DE concentrations and were offered to grower finisher pigs.
Although there is clear evidence in this study to demonstrate the anti-nutritive properties of barley b-glucan, its addition to the diet also resulted in reductions of BCFAs in the caecum and colon and digesta Enterobacteria in the caecum. As the b-glucan: protein ratio increased there was a linear decrease in the concentrations of isobutyric, isovaleric acid and the acetic: propionic acid ratio in digesta, which corroborates with the reductions in odour emissions reported by Garry et al. (Reference Garry, Fogarty, Curran and O’Doherty2007) and Leek et al. (2007). By limiting protein fermentation and increasing carbohydrate fermentation in the large intestine, the production of some of the most offensive smelling compounds emanating from pig production can be reduced (Mackie et al., Reference Mackie, Stroot and Varel1998). Both protein and carbohydrate fermentation contribute to the production of acetic acid; however, the BCFAs i.e. isobutyric and isovaleric acid are produced exclusively from protein fermentation and have a more offensive odour than the straight-chain VFAs (Mackie et al., Reference Mackie, Stroot and Varel1998). Leek et al. (2007) found that there was a linear relationship between odour emission rate and the acetic : propionic acid ratio in the slurry, while Garry et al. (Reference Garry, Fogarty, Curran and O’Doherty2007) reported that barley-based diets reduced odour by 47% compared with wheat-based diets when fed to grower finisher pigs. The linear decrease in BCFAs and acetic : propionic acid ratio in the colon would suggest that in the current study 666 g/kg barley or 25 g/kg b-glucan would emit the least odourous compounds from pig houses.
There was a linear increase in total VFA concentration and in the molar proportions of propionic acid and butyric acid with increasing levels of dietary barley in the colon. SCFAs are the principal luminal anions, thus an increase in SCFA production lowers digesta pH (Bach Knudsen et al., Reference Bach Knudsen, Jensen, Andersen and Hansen1991). Digesta caecum and colon pH was significantly reduced due to increases in total VFA as a consequence of increasing dietary barley level. The luminal environment created by the fermentation of barley b-glucans, increased concentrations of SCFAs and a low digesta pH is conducive to the production of beneficial bacteria at the expense of acid-sensitive Enterobacteria spp. (O’Connell et al., Reference O’Connell, Callan, Byrne, Sweeney and O’Doherty2005). The linear reduction in Enterobacteria spp. concentrations with increasing dietary barley is probably due to the significant reduction in digesta pH. Mikkelsen et al. (Reference Mikkelsen, Naughton, Hedemann and Jensen2004) reported that a low pH, as a result of high levels of lactic acid, killed Salmonella and Escherichia coli bacteria in the gastro-intestinal tract. Although many species of E. coli are commensal, Smith and Halls (Reference Smith and Halls1968) found that barley hulls reduced the incidence of enterotoxigenic haemolytic E. coli, which are responsible for post-weaning colibacillosis. Some E. coli species are acid intolerant and are not able to compete at a low pH < 5.5 (Prescott et al., Reference Prescott, Harley and Klein2002). When the pH drops below 5.5 to 6, E. coli synthesise an array of new proteins as part of their ‘acidic tolerance response’ in an effort to cope with the pH change (Prescott et al., Reference Prescott, Harley and Klein2002). The average caecal pH of pigs offered the 666 g/kg barley diet was 5.3, which may have been sufficient to upset the pH homeostasis of the Enterobacteria. The reduction in Enterobacteria concentrations due to the inclusion of barley in the diet is reflected in a higher faecal DM content and production of butyric acid in the colon compared with the pure wheat-based diet, indicating an improved gut function with efficient absorption of colonic water. Butyrate is rapidly metabolised by colonocytes supplying 70% of respiratory fuel (Roediger, Reference Roediger1995), therefore it is critical for the maintenance and metabolic integrity of the colonic epithelial cells. Given that the salvage function of the colon for absorption of unabsorbed sodium and water from the small intestine depends on the metabolic integrity of the colonic epithelial cells (Roediger, Reference Roediger1994), it is not unreasonable to expect that barley-based diets would result in higher faecal DM partly due to increased butyrate production (Williams et al., Reference Williams, Verstegen and Tamminga2001).
There was no linear or quadratic effect of dietary barley on ammonia emissions. However, the pure barley-based diet had numerically reduced ammonia emissions compared with the pure wheat-based diet. This coincides with a linear increase in faecal N excretion and a numerical decrease in the urine : faeces N ratio as the level of dietary barley b-glucan increased. The effect of this is a shift in N excretion from its more volatile state as urea in urine to bacterial N in faeces (Mroz et al., Reference Mroz, Moeser, Vreman, van Diepen, Van Kempen, Canh and Jongbloed2000), thus limiting the capacity for rapid ammonia volatilisation.
In conclusion, there was a linear decrease in diet digestibility as the level of barley in the diet increased. However, this reduction in digestibility did not affect N retention. There was a linear decrease in BCFAs, acetic : propionic acid ratio, caecal Enterobacteria spp. and colonic butyrate proportion as the level of barley increased. This would suggest that 50 g/day of barley b-glucans is necessary in order to reduce odourous compounds in pig houses and to improve gut health.
