| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ANIMAL NUTRITION |
ARS, USDA, U.S. Meat Animal Research Center, Clay Center, NE 68933
Abstract
Four Duroc x White composite crossbred (21.8 ± 1.0 kg BW) and four 12-wk-old Meishan purebred (20.7 ± 1.6 kg BW) growing barrows were used to determine the relative breed differences in metabolic and microbial responses to a high-fiber diet. The pigs were trained to consume 700 g of a diet containing 35% (as-fed basis) dehydrated alfalfa meal once daily. The pigs’ daily intakes of DM, N, GE, NDF, and ADF were 610 g, 16.6 g, 2.64 Mcal, 150 g, and 88 g, respectively. On d 12 after surgical catheterization of the portal vein, ileal vein, and carotid artery, a 3-d total urine and feces collection was conducted. On d 24 after surgery, each pig was placed in an open-circuit calorimeter, and its catheters were connected to a system for simultaneous measurements of oxygen consumption by portal-drained viscera and by whole body, and the net portal absorption of VFA after a 24-h fasting and during a 5-h postprandial period. The VFA measured included acetic, propionic, isobutyric, butyric, isovaleric, and valeric acids. A second 3-d total urine and feces collection was conducted on d 30 after surgery. There were no differences (P = 0.13) between the first and second collections in apparent total-tract digestibility coefficients for nutrients and N retention of pigs. Compared with Duroc x White composite crossbred pigs, Meishan pigs had lower (P = 0.05) apparent digestibility coefficients for DM, N, NDF, hemicellulose, and N retention, but their portal-drained viscera used a greater (P = 0.05) fraction of whole-body oxygen consumption. No differences (P = 0.12) were found between Duroc x White composite crossbred and Meishan pigs in total viable bacteria and cellulolytic bacteria from fecal samples, in vitro digestibility of alfalfa NDF fractions by fecal inocula, whole-body oxygen consumption, net portal absorption of VFA, total energy of absorbed VFA, and the potential of absorbed VFA for meeting the energy needs for whole-body heat production. These results indicate that, in contrast to previous beliefs, the ability of Meishan growing pigs to utilize a high-fiber diet is not superior to that of Duroc x White composite crossbred growing pigs.
Key Words: Energetics • High-Fiber Diet • Microbial Fiber Degradation • Nutrient Utilization • Pigs • VFA Absorption
Introduction
Chinese Meishan pigs are well recognized for their high prolificacy (Pond and Mersmann, 2001
). Meishan pigs were also reported to digest high-fiber diets more efficiently than Large White (Fevrier et al., 1988) and Dutch Landrace pigs (Kemp et al., 1991). This difference in high-fiber digestion might be related to size and microbial activity of the gastrointestinal tract because the gastrointestinal tract of Meishan pigs, as a percentage of BW, is greater than that of Large White (Fevrier et al., 1988) or Duroc x White composite (DWc) crossbred pigs (Yen et al., 1991b).
In Western breeds of pigs, the large intestine contains active fiber-degrading microflora, and the populations of these microflora increase in response to the ingestion of diets containing high levels of alfalfa meal (Varel and Yen, 1997). Feeding a high-fiber diet results in more substrate entering the large intestine for fermentation to produce VFA. In White composite growing pigs, the gut VFA absorbed into the portal vein could contribute 24% energy to the whole-body heat production (Yen et al., 1991a), and the portal-drained viscera (gastrointestinal tract, spleen, and pancreas) would use >20% of whole-body oxygen consumption (Yen, 1997). For Meishan pigs, no information is available on the gut fiber-degrading microflora, portal absorption of VFA, and energy expenditure of whole body and portal-drained viscera. The objective of the present study was to determine the relative breed differences between DWc crossbred and Meishan purebred growing pigs in total-tract nutrient digestibility, activity of fecal fiber-degrading bacteria, portal absorption of VFA, and energy expenditure of the whole body and the portal-drained viscera when pigs were fed a high-fiber diet containing 35% dehydrated alfalfa meal (as-fed basis) and 21% NDF (as-fed basis).
Materials and Methods
Animals and Diets
The protocol of this study was approved by the Animal Care and Use Committee of the U.S. Meat Animal Research Center. This study used four DWc crossbred (21.8 ± 1.0 kg BW) and four 12-wk old Meishan purebred (20.7 ± 1.6 kg BW) growing barrows. The White composite pigs were from inter se mating of a crossbred foundation with equal genetic contributions from Chester White, Landrace, Large White, and Yorkshire. The DWc crossbred and Meishan pigs were born in the same farrowing season and housed and managed similarly before being used for the study. The pigs were allowed ad libitum access to feed before the study. During the study, the pigs were housed in individual pens that measured 1.2 m x 1.2 m and equipped with an automatic nipple waterer. The temperature of the rooms housing the pigs was maintained at 21°C, and the lights in the rooms were kept on 24 h a day. Once daily at 0930, the pigs were allowed to enter metabolism cages for 1 h to consume a high-fiber diet (Table 1) mixed with water. The metabolism cages had adjustable sides and backs (81 cm in height) and a 48- x 122-cm flattened expanded-metal floor that was 81 cm above the ground. After feeding, the pigs were returned to their individual pens. The high-fiber, corn-soybean meal diet of the present study contained 35% (as-fed basis) dehydrated alfalfa meal and was supplemented with minerals and vitamins to meet or exceed NRC (1998) nutrient requirements. To standardize daily nutrient intake between genotypes, the same amount of feed was offered to all pigs. Initially, the pigs were offered 1.2 kg of the diet mixed with 1.2 L of water. However, the pigs could not consume the diet completely during the 1-h feeding period because of the bulkiness of the diet. After a week of training, the pigs were able to consume 0.7 kg of feed mixed with 1.0 L of water within an hour. This quantity of feed was therefore chosen as the daily allowance throughout the test. The pig’s daily intake was 610 g of DM, 16.6 g of N, 2.64 Mcal of GE, 150 g of NDF, and 88 g of ADF.
|
). On d 12 ± 1 after surgery, when pigs had regained their presurgery appetite for at least 6 d, the first 3-d total urine and feces collection of pigs was conducted. The day before the initiation of total urine and feces collection, the pigs were weighed 2 h after feeding. At 0930 on the first day of collection, the pigs were placed into metabolism cages and fed 0.7 kg of feed mixed with 1.0 L of water. The pigs were kept continuously in the metabolism cages for 72 h and fed once daily at 0930. Additional water (1.0 L/d) was placed in the feeder after the pig had finished its daily feed allowance. The feces and urine were collected separately for 72 h, using a screen and a funnel placed 5 cm under the expanded-metal floor of the metabolism cage. Urine was filtered through glass wool into a plastic bottle containing 20 mL of 6 N HCl. Urine was collected once daily, measured for total volume and a 10% aliquot was saved. The daily aliquots of urine were pooled within pig and stored at 4°C until analyzed. The daily total feces were collected, pooled within pig, placed in plastic bag, and stored at –20°C. Upon termination of the 72-h collection, the pigs were returned to their home pens and fed once daily their feed allowance in the metabolism cages.
On d 24 ± 2 after surgery, pigs were individually weighed at 1430 (5 h after feeding) and placed into an open-circuit indirect calorimeter. On the next day at 0730, the portal vein and carotid artery catheters were connected, through a two-way peristaltic pump (Gilson Minipuls 2, Gilson Medical Electronics, Middleton, WI), to an arteriovenous O2 concentration difference analyzer (Avox Systems, San Antonio, TX) equipped with three-way stopcocks for blood sampling. The ileal vein catheter was also connected to an infusion pump (Harvard infusion-withdrawal pump, series 940, Harvard Apparatus, Mills, MA) for priming and subsequent 6-h constant infusion of 1% p-aminohippuric acid (PAH) in normal saline solution. The pig was primed with the PAH solution at a rate of 3.82 mL/min for 5 min. The rate of constant infusion of the PAH solution was 0.788 mL/min (7.88 mg of PAH/min). With this prime-constant infusion technique, a steady-state condition of PAH was achieved within 45 min after the onset of constant infusion. A diagram and detailed information regarding the setups of these connections and the calorimeter have been described previously (Yen et al., 1989). No more than two pigs were used each day for the measurements. To eliminate any possible calorimeter effect on the measurements, the same two calorimeters were used alternately for housing pigs of different genotypes.
Pigs were fed at 0930 using a glove sealed to the calorimeter without opening the calorimeter door. Feed (0.7 kg) and water (1.0 L), which were stored in an upper compartment of each calorimeter, were delivered to the feeder. Oxygen consumption by the portal-drained viscera of the pig was determined according to our previously reported procedures (Yen et al., 1989) before feeding (h 0, which was 24 h after the previous feeding), and then hourly for 5 h after feeding. Measurements of whole-body O2 consumption and CO2 production of the pig were obtained by the previously reported procedures (Nienaber and Maddy, 1985) from –0.5 h preprandial through the 5-h postprandial period. The 5-min interval measurements of O2 and CO2 were pooled to derive the 30-min preprandial and hourly postprandial values. The whole-body heat production (Kcal) was calculated as 3.866 x L of O2 consumed + 1.200 x L of CO2 produced (Brouwer, 1965).
Before feeding and then hourly through the 5-h postprandial period, the O2 concentration difference between the carotid artery and portal vein of the pig was measured with the online arteriovenous O2 difference analyzer, and arterial and portal venous blood samples (8 mL each) were also obtained simultaneously. The heparinized blood samples were chilled on ice and subsampled for determination of packed cell volume and for separating plasma from the cells by centrifuging at 4°C and 2,000 x g for 10 min. An aliquot of plasma was refrigerated and assayed within 12 h for PAH concentration as described previously (Yen and Killefer, 1987). Another aliquot of plasma was stored at –20°C until it was analyzed for VFA concentration with our previously described method of gas chromatography following a cleanup by ion-exchange chromatography (Yen et al., 1991b). The VFA measured included acetic, propionic, isobutyric, butyric, isovaleric, and valeric acids. The net portal absorption of VFA was calculated by multiplying the porto-arterial plasma concentration difference of VFA by portal vein plasma flow rate (PVPF). The PVPF was estimated by our previously described indicator-dilution technique employing PAH as the indicator (Yen and Pond, 1990). Energy values (heat of combustion) were assumed to be 209, 365, 521, 521, 678, and 678 kcal/g of molecular weight for absorbed acetic, propionic, isobutyric, butyric, isovaleric, and valeric acids, respectively (CRC, 1985).
Using portal and arterial packed cell volumes and plasma PAH concentrations, the portal vein blood flow rate (PVBF) per unit of time was calculated according to our previously reported equation (Yen and Killefer, 1987; Yen and Nienaber, 1992). The O2 consumption by portal-drained viscera (also termed "portal vein-drained organs") was estimated by multiplying the O2 concentration difference between the carotid artery and portal vein blood by PVBF.
Following the calorimeter and net portal absorption measurements, the pigs were returned to their home pens and fed 0.7 kg of feed once daily. On d 30 ± 2 postsurgery, the second 3-d total urine and feces collection of pigs was conducted with the same procedures as described for the first collection. Upon termination of the second total collection, the pigs were returned to their home pens and again fed once daily until they were slaughtered on d 39 ± 2 after surgery. When the pigs were slaughtered, they had been fed the high-fiber diet for 59 ± 1 d. The pigs were killed by exsanguination after electric stunning. The fresh weights of heart, lungs, liver, kidneys, spleen, and pancreas were recorded. Stomach, cecum, and colon plus rectum were opened, rinsed with tap water to remove the contents, blotted dry, and weighed. Small intestine contents were removed by stripping manually, and the empty small intestine was then weighed.
Fresh fecal samples (100 g) were collected by rectal massage from each pig between 0800 and 0930 on d 7, 35, and 49 after the pig was started on test, corresponding to 13, 17, and 19 wk of age.
Chemical Analysis
The DM content of the feed and feces were determined by drying them in a forced-air drying oven (70°C) until they reached a constant weight. The dried feed and feces were then ground in a Thomas-Wiley mill (model 4, Arthur H. Thomas Co., Philadelphia, PA) and analyzed for GE by a Parr adiabatic oxygen bomb calorimeter (model 1241, Parr Instrument Co., Moline, IL), N by the combustion method (AOAC, 1990) with a LECO model CN-2000 carbon-nitrogen analyzer (LECO, Corp., St. Joseph, MI), NDF (Van Soest and Wine, 1967), ADF, and ADL (Van Soest, 1963). The urine samples were also analyzed for N content.
Microbial Assay
The fresh fecal samples obtained by rectal massage were processed according to the procedures previously described (Varel et al., 1984) for the determination of total viable bacteria and cellulolytic bacteria, as well as in vitro digestibility of alfalfa NDF and its fractions by fecal inocula. The cellulolytic bacteria were quantified by the most probable number procedure and not the cellulase activity. The most probable number procedure used serial dilutions of blend subsample of feces in anaerobic buffer and inoculation of roll tubes and medium for quantifying cellulolytic bacteria.
Statistical Analyses
Repeated measures analyses using mixed model of SAS (SAS Inst., Inc., Cary, NC) were employed for statistical analyses of all data, with the exception of slaughter data. The model included genotype, time, and genotype x time. Collection time was the time variable for the data on apparent total-tract nutrient digestibility coefficients and N retention. Age of pigs was the time variable for the data on total viable bacteria and cellulolytic bacteria from fecal samples obtained at various ages of pigs, and the in vitro digestibility of alfalfa NDF and its fractions by fecal inocula. Blood sampling time was the time variable for the data on arterio-portal oxygen concentration difference, portal vein whole blood flow, oxygen consumption by portal-drained viscera and whole body, fraction of whole-body oxygen consumption used by portal-drained viscera, plasma VFA concentration, portal vein plasma flow, net portal absorption of individual and total VFA, total energy of absorbed VFA, whole-body heat production, and percentage potential of absorbed VFA to meet energy needs of whole-body heat production. Animal was a random effect. Autoregressive order 1 was used for the covariance structure, and pig was the experimental unit.
The GLM procedure of SAS was used to analyze data on the slaughter BW, and visceral organ weights expressed as a percentage of slaughter BW. Pig was the experimental unit. For comparison with previously reported information (Yen et al., 1991a), the five hourly postprandial individual values were averaged to obtain the mean 5-h postprandial value. The GLM procedure of SAS was again used to analyze data on the preprandial and mean 5-h postprandial portal vein whole blood flow, the oxygen consumption by portal-drained viscera and by whole body, the fraction of whole-body oxygen consumption used by portal-drained viscera, portal vein plasma flow, net portal absorption of individual and total VFA, total energy of absorbed VFA, whole-body heat production, and percentage potential of absorbed VFA to meet energy needs of whole-body heat production.
Results
Body Weight, Apparent Total-Tract Nutrient Digestibility, and Nitrogen Retention
As presented in Table 2, the BW of pigs was greater (P = 0.01) for the second collection than the first collection. The DWc crossbred pigs also had greater BW than Meishan pigs at the second collection (P = 0.04) and overall (P = 0.01). Table 2 also shows that no differences were found between the first and second collections in the apparent total-tract digestibility coefficients for DM (P = 0.50), GE (P = 0.46), N (P = 0.51), NDF (P = 0.83), ADF (P = 0.53), hemicellulose (NDF – ADF; P = 0.13), and cellulose (ADF – ADL; P = 0.33), as well as N retention, expressed as % of intake (P = 0.28), or percentage of digested (P = 0.12). Compared with DWc crossbred pigs, Meishan pigs had lower apparent digestibility coefficients for DM (P = 0.05), N (P = 0.01), NDF (P = 0.04), and hemicellulose (P = 0.03), and reduced N retention expressed as a percentage of intake (P = 0.01), or percentage of digested (P = 0.01). The apparent total-tract digestibility coefficients for GE (P = 0.23), ADF (P = 0.14), and cellulose (P = 0.25), however, were not different for DWc crossbred and Meishan pigs.
|
) increased (P = 0.05) as pigs became older (from 13, 17, to 19 wk of age). However, age of pigs had no effect (P = 0.25) on the cellulolytic bacterial count in fecal samples. There were also no differences between DWc crossbred and Meishan pigs in the total viable bacterial (P = 0.28) and cellulolytic bacterial (P = 0.36) counts from fecal samples, whether expressed on the basis of per gram of dry feces or daily total fecal DM output (data not shown). Table 4 shows in vitro digestibility of alfalfa NDF and its fractions by fecal inocula from pigs, when incubation time was 24 h. The in vitro digestibility of alfalfa NDF and its fractions were not different among various age of pigs (P = 0.76 for NDF, P = 0.58 for hemicellulose and P = 0.63 for cellulose) or between DWc crossbred and Meishan pigs (P = 0.64 for NDF, P = 0.15 for hemicellulose and P = 0.43 for cellulose).
|
|
, the arterio-portal oxygen concentration difference was not influenced by genotype x time interaction (P = 0.67) or genotype (P = 0.74). The oxygen concentration difference, however, was affected by time (P = 0.01). There was no genotype x time interaction (P = 0.64) for the portal vein blood flow rate. However, the portal vein blood flow rate was influenced by time (P = 0.01) and genotype (P = 0.05). The oxygen consumption by the portal-drained viscera was not affected by genotype x time interaction (P = 0.71). It was influenced by time (P = 0.01) and tended to be affected by genotype (P = 0.10). The whole-body oxygen consumption was not influenced by a genotype x time interaction (P = 0.18) or genotype (P = 0.45), but was affected by time (P = 0.01). There was a tendency toward a genotype x time interaction (P = 0.09) and a greater genotype effect (P = 0.05) on the fraction of whole-body oxygen consumption used by the portal-drained viscera.
|
. There were no genotype x time interactions for portal plasma concentrations of acetic (P = 0.49), propionic (P = 0.24), or butyric (P = 0.78) acids, or for arterial plasma concentrations of acetic (P = 0.27), propionic (P = 0.06), or butyric acids (P = 0.46). Time had no effects on portal plasma concentrations of acetic (P = 0.35), propionic (P = 0.47), or butyric (P = 0.41) acids, but influenced (P = 0.01) arterial plasma concentrations of acetic, propionic, and butyric acids. Genotype also had no effects on portal plasma concentration of acetic acid (P = 0.15) and arterial plasma acetic acid concentration (P = 0.34). However, plasma concentration of propionic acid was greater in the portal samples (P = 0.05) and tended to be greater in the arterial samples (P = 0.09) for DWc crossbred than Meishan pigs. The DWc crossbred pigs also had greater arterial (P = 0.05) but not portal (P = 0.20) plasma concentrations of butyric acid than Meishan pigs.
|
shows portal and arterial plasma concentrations of isobutyric, isovaleric, and valeric acids in pigs. No measurable isobutyric, isovaleric, and valeric acids were found in arterial plasma of pigs either preprandially or during the 5-h postprandial period. There were no genotype x time interactions for portal plasma concentrations of isobutyric (P = 0.38), isovaleric (P = 0.35), and valeric acids (P = 0.81). A time effect was detected for portal plasma concentration of isovaleric (P = 0.01), but not of isobutyric (P = 0.87) and valeric acids (P = 0.19). Genotype had no effect on portal plasma concentrations of isobutyric (P = 0.90), isovaleric (P = 0.51), and valeric acids (P = 0.19).
|
. No genotype x time interaction occurred for portal vein plasma flow (P = 0.54). Time affected (P = 0.01) portal vein plasma flow rates. Compared with DWc crossbred, Meishan pigs had greater (P = 0.05) portal vein plasma flow rate.
|
and 6, no genotype x time interaction for the net portal absorption of acetic (P = 0.28), propionic (P = 0.10), butyric (P = 0.54), isobutyric (P = 0.56), isovaleric (P = 0.11), or valeric (P = 0.92) acid could be observed. There was a time effect on net portal absorption of acetic (P = 0.01), propionic (P = 0.01), isobutyric (P = 0.04), butyric (P = 0.01), and valeric acids (P = 0.05). Time did not influence (P = 0.13) net portal absorption of isovaleric acid. No differences in the net portal absorption of acetic (P = 0.39), propionic (P = 0.82), butyric (P = 0.90), isobutyric (P = 0.37), isovaleric (P = 0.12), or valeric (P = 0.21) acid were found between DWc crossbred and Meishan pigs.
|
|
); however, the total energy of absorbed VFA was affected (P = 0.01) by time. No genotype x time interaction (P = 0.15) occurred for the whole-body heat production. Nevertheless, the whole-body heat production was influenced (P = 0.01) by time. The whole-body heat production in DWc crossbred pigs did not differ (P = 0.23) from that in Meishan pigs. The percent in the potential of absorbed VFA for whole-body heat production was affected by time (P = 0.01) but not genotype (P = 0.36). It also tended to be influenced by genotype x time interaction (P = 0.06).
|
are the preprandial and the mean 5-h postprandial results from DWc crossbred and Meishan pigs. Both the preprandial (P = 0.01) and the mean postprandial (P = 0.05) portal vein blood flow rates of DWc crossbred pigs were less than those of Meishan pigs. The oxygen consumption by portal-drained viscera was less (P = 0.03) for DWc crossbred pigs than for Meishan pigs at preprandial, but not when averaged over the 5-h postprandial period (P = 0.26). Neither the preprandial (P = 0.44) nor the mean postprandial (P = 0.33) whole-body oxygen consumption was different between DWc crossbred and Meishan pigs. Both the preprandial (P = 0.04) and the mean postprandial (P = 0.05) fraction of whole-body oxygen consumption used by portal-drained viscera were less for DWc crossbred than for Meishan pigs. The preprandial (P = 0.02) and the mean postprandial (P = 0.05) portal plasma flow rates were also less for DWc crossbred than for Meishan pigs. Yet, there were no differences between DWc crossbred and Meishan pigs in pre- or postprandial net portal absorption of individual (P = 0.15 to 0.78 or 0.23 to 0.69) or total VFA (P = 0.57 or 0.86), total energy value of absorbed VFA (P = 0.61 or 0.91), whole-body heat production (P = 0.77 or 0.76), and the potential of absorbed VFA for meeting energy needs of whole-body heat production (P = 0.70 or 0.38).
|
). However, when expressed as a percentage of BW, DWc crossbred had lower weights of lung (P = 0.02), liver (P = 0.01), pancreas (P = 0.01), gastrointestinal tract (P = 0.05), and portal vein-drained organs (P = 0.04) than Meishan pigs. The weights of heart (P = 0.14), kidneys (P = 0.59), and spleen (P = 0.61), as a percentage of body weight, were not different between DWc crossbred and Meishan pigs.
|
Apparent Total-Tract Nutrient Digestibility and Nitrogen Retention
The present study shows clearly that when 700 g of a high-fiber diet containing 35% dehydrated alfalfa meal was fed once daily, Meishan growing pigs had lower apparent total-tract digestibility coefficients for DM, N, NDF, and hemicellulose than did DWc crossbred growing pigs. These lower total-tract nutrient digestibility coefficients for Meishan pigs disagree with the results of Fevrier et al. (1988), who reported that Meishan pigs had greater rather than lesser digestibility coefficients for the above nutrients than Large White pigs. There is no ready explanation for this discrepancy between the two studies. The NDF content of the diet used by Fevrier et al. (1988) was 21.2%, which was similar to the 21.4% NDF concentration in the diet of the present study. However, the fractions of NDF (hemicellulose and cellulose) and the crude fiber content in two diets were different. The wheat-based diet used by Fevrier et al. (1988) consisted of 51.8% wheat bran, whereas the corn-soybean meal-based diet of the present study contained 35% dehydrated alfalfa meal. Compared with the diet used by Fevrier et al. (1988), the diet of the present study contained less hemicellulose (8.9 vs. 13.3%), but more cellulose (9.5 vs. 6.2%). Variations in NDF fractions might be a cause for different digestibility responses of Meishan pigs between the present study and the study of Fevrier et al. (1988).
Although the total-tract N digestibility coefficient for Meishan pigs was shown to be less in our study but greater in the study of Fevrier et al. (1988), both studies showed a lower N retention for Meishan pigs whether it was expressed as percentage of intake, or percentage of digested. A lower N retention for Meishan pigs than for Dutch Landrace gilts was also observed by Kemp et al. (1991), who fed pigs an 11.0% crude fiber diet comprising 9.0% alfalfa meal, 10.0% oat husk meal, and 3.3% straw meal. On the basis of those studies, it can be concluded that N utilization for Meishan pigs is less efficient than for some of western breeds of pigs, such as Large White, Dutch Landrace, and DWc crossbreeds. This less efficient utilization of N for Meishan pigs is apparently due to their low capacity of lean growth compared with western breeds of pigs (Yen et al., 1991b).
Unlike the present study and that of Fevrier et al. (1988), Kemp et al. (1991) detected no difference between Meishan and Dutch Landrace pigs in total-tract N digestibility coefficient. It is unclear why this discrepancy in total-tract N digestibility coefficient occurred among the three studies. Variations in crude fiber content probably is not a contributing factor, because the crude fiber content of our study was 10.3%, which was similar to 11.0% for the study of Kemp et al. (1991). In contrast to our study, Fevrier et al. (1988) showed that Meishan pigs had greater total-tract digestibility coefficients for NDF and crude fiber, and Kemp et al. (1991) reported a tendency of greater crude fiber digestibility coefficient for Meishan pigs. An obvious difference between our study and the studies of Fevrier et al. (1988) and Kemp et al. (1991) was the BW of pigs. The total-tract digestibility coefficients of our study were obtained from pigs weighing <28 kg, whereas those of the studies by Fevrier et al. (1988) and Kemp et al. (1991) were from pigs weighing 40.5 kg and 39 kg, respectively. It is unlikely that this difference in body weight could change the fiber digestibility of Meishan pigs from inferior to superior as compared with western breeds of pigs. Nevertheless, Noblet and Shi (1994) observed a greater digestibility coefficient for nutrients and energy of diets, particularly high-fiber diets, in 100- vs. 45-kg Large White Pietrain barrows.
Microbial Responses in Fecal Samples
In the present study, total viable bacteria in fecal samples rose as the pig’s age increased from 13 to 19 wk and the length on test increased from 7 to 49 d. A positive correlation closely existed between the count of fecal total viable bacteria and the length of adaptation to the high-fiber diet. Yet, there was no age effect on cellulolytic bacteria in fecal samples and in vitro digestibility of alfalfa NDF fractions by fecal inocula. This absence of age effect on cellulolytic activity observed in the present study seems to contradict our previous study (Varel et al., 1984). In our previous study (Varel et al., 1984), a tendency of increased cellulolytic activity in fecal samples between d 5 and 53 of the test was observed when eight White composite growing-finishing pigs were allowed ad libitum access to a high-fiber diet with the same composition as the present study. It is unclear whether this absence of age effect in the present study was caused by the considerable animal-to-animal variation (154% CV for cellulolytic bacteria in fecal samples), limited number of animals used (four pigs per genotype), small feed allowance (700 g once daily), or the procedure for quantifying cellulolytic bacteria. In the present study, cellulolytic bacteria were quantified by the most probable number procedure, which used serial dilutions and inoculations of roll tube and medium. Compared with the cellulase activity analysis based on the quantity of glucose released, the most probable number procedure is less sensitive and less reliable for quantifying cellulolytic bacteria (Varel et al., 1984). If the cellulase activity analysis rather than the most probable number procedure were used in the present study, more cellulolytic bacteria might be detected and a trend for age effect on cellulolytic bacteria might occur. Nevertheless, no differences between DWc crossbred and Meishan pigs could be observed in total viable bacteria and cellulolytic bacteria in fecal samples, as well as the in vitro digestibility of alfalfa NDF and its fractions by fecal inocula in the present study, whether these microbial assays were conducted when pigs were 13, 17, or 19 wk of age. The findings of fecal cellulolytic bacteria may be extrapolated to the cellulolytic bacteria in the colon because the microflora composition in the feces tends to be similar to that of the colon in pigs (Varel and Yen, 1997). The absence of a difference between DWc crossbred and Meishan pigs in the activity of the fecal cellulolytic bacteria observed in the present study could imply a lack of superiority of Meishan pigs over crossbred pigs in the utilization of fibrous diet. This inference is supported by results of the net portal absorption of VFA from large intestine degradation and fermentation of dietary fiber and protein, and the energy expenditures of the portal-drained viscera and the whole body.
Energy Expenditure and Volatile Fatty Acid Absorption
Fasting metabolism reflects the energy expenditure for maintenance and is correlated with the mass and activity of vital organs. The preprandial 24-h fasting energy expenditure, including arterio-portal oxygen concentration difference, portal-drained viscera oxygen consumption, and the fraction of whole-body oxygen consumption used by portal-drained viscera, for DWc crossbred pigs of the present study are all within the ranges shown in our previous studies with White composite pigs (Yen et al., 1989; Yen and Nienaber, 1992, 1993). Compared with the DWc crossbred pigs, Meishan pigs in the present study had greater preprandial portal vein blood flow rate, oxygen consumption by portal-drained viscera, and fraction of whole-body oxygen consumption used by portal-drained viscera. These differences were related to the mass of portal-drained viscera as a percentage of BW. As demonstrated in our previous studies (Yen et al., 1989; Yen and Nienaber, 1992, 1993) and again shown in the present study, the portal-drained viscera of White composite pigs and DWc crossbred pigs are highly metabolically active. They consume >20% of whole-body oxygen consumption while accounting for approximately 5% of BW. The weights of portal-drained viscera as a percentage of BW of Meishan pigs in the present study were 26% greater than that of DWc crossbred pigs. As a result of greater mass, the portal-drained viscera of Meishan pigs had greater blood flow and consumed more oxygen than did that of DWc crossbred pigs. In addition to portal-drained viscera, Meishan pigs in the present study also had greater lung and liver weights as a percentage of BW than did DWc crossbred pigs. The greater organ weights for Meishan pigs in the present study agree with our previous study (Yen et al., 1991b; Hansen et al., 1997). Fevrier et al. (1988) also observed greater weights of small intestine, cecum, and colon in Meishan pigs compared with Large White pigs. The liver weight, as a percentage of BW, of Meishan pigs in the present study was 23% greater than that of DWc crossbred pigs. In pigs, the liver and portal-drained viscera are two principal determinants of whole-body energy requirement and together they can use >40% of whole-body oxygen consumption (Yen, 1997). Because both the liver and portal-drained viscera were greater for Meishan pigs than for DWc crossbred pigs, it is assumed that the preprandial whole-body oxygen consumption would be greater in Meishan pigs than in DWc crossbred pigs. However, no difference was detected in the present study between Meishan pigs and DWc crossbred pigs in whole-body oxygen consumption and heat production, which was calculated from oxygen consumption and CO2 production. This lack of difference was due to the lower lean growth capacity and associated heat production for protein deposition in Meishan pigs compared with DWc crossbred. Although no determination on body composition of pigs was conducted in the present study, we showed previously that Meishan pigs had 45% less muscle (Yen et al., 1991b) and 59% less daily protein accretion (Hansen et al., 1997) than did DWc crossbred. The heat production associated with protein deposition in nonruminants, such as pigs, is approximately twice that for fat accretion (see review by Yen et al., 1991b). In Meishan pigs, the greater oxygen consumption and heat production associated with the portal-drained viscera and the liver is overshadowed by the lesser oxygen consumption and heat production related to protein deposition. Thus, it is not surprising, and in agreement with our previous studies (Yen et al., 1991b; Hansen et al., 1997), that the present study found no differences between DWc crossbred pigs and Meishan pigs in their preprandial whole-body oxygen consumption and heat production.
Feeding produced a postprandial hyperemia and enhanced oxygen consumption by the portal-drained viscera and by the whole body in DWc crossbred pigs, which was expected based on our previous studies (Yen et al., 1989; Yen and Nienaber, 1993). The pattern of postprandial hourly portal vein blood flow and the oxygen consumption by the portal-drained viscera and by the whole body of DWc crossbred in the present study are similar to those shown in our previous studies using White composite pigs (Yen et al., 1989; Yen and Nienaber, 1993). The mean 5-h postprandial portal vein blood flow in the DWc crossbred pigs of the present study is within the ranges reported by our previous studies with White composite pigs (Yen et al., 1989, 1991a; Yen and Nienaber, 1993). However, the mean postprandial fraction of whole-body oxygen consumption used by portal-drained viscera of DWc crossbred pigs in the present study was 35%, which is markedly greater than the value (<25%) observed in our previous studies with White composite pigs (Yen et al., 1989; Yen and Nienaber, 1993). This markedly greater fraction observed in the present study was apparently due to the 18% greater postprandial oxygen consumption by portal-drained viscera coupled with a lower postprandial whole-body oxygen consumption in DWc crossbred pigs compared with White composite pigs of the previous studies. It is unclear what caused the postprandial oxygen consumption by the portal-drained viscera to be greater and that by the whole body to be less in the present study. The mass of the portal-drained viscera probably is not the contributing factor. Although the portal-drained viscera accounted for 5.5% of BW for DWc crossbred pigs in the present study and 4.9% for White composite pigs in our previous study (Yen et al., 1989), there were few differences between the present study and the previous study in the 24-h fasting preprandial oxygen consumption by the portal-drained viscera (1.7 vs. 1.5 mL•min–1•kg BW–1) and by the whole body (5.7 vs. 6.2 mL•min–1•kg BW–1), as well as the fraction of whole-body oxygen consumption used by portal-drained viscera (30 vs. 25%). A high-fiber diet containing 35% dehydrated alfalfa meal was used in the present study, while a typical low-fiber diet was used in the previous studies (Yen et al., 1989; Yen and Nienaber, 1993). Dietary fiber might be the primary determinant for the postprandially increased portal-drained viscera and decreased whole-body oxygen consumption as observed in the present study. To delineate this possibility, further study is needed to compare the energy expenditure of the portal-drained viscera in the same genotype of pigs fed a typical low-fiber diet or a high-fiber alfalfa-containing diet.
Compared with DWc crossbred pigs, Meishan pigs of the current study had greater mean 5-h postprandial portal vein blood flow rate. This greater rate was again due to their greater mass of portal-drained viscera. Yet, mean postprandial oxygen consumption by the portal-drained viscera was only numerically greater for Meishan pigs than DWc crossbred pigs. This absence of a difference apparently resulted from the canceling effect of the numerically lesser arterio-portal oxygen concentration difference on the greater portal vein blood flow rate for calculating the oxygen consumption by the portal-drained viscera in Meishan pigs. As expected from the early discussion about the masking impact of lean capacity on metabolically active viscera and the resulting whole-body energy expenditure, there was also no difference between Meishan and DWc crossbred pigs in mean postprandial whole-body oxygen consumption. However, the whole-body oxygen consumption was numerically less for Meishan than for DWc crossbred pigs. With a numerical greater oxygen consumption by the portal-drained viscera as the numerator and a numerical less oxygen consumption by the whole body as the denominator in Meishan pigs compared with DWc crossbred pigs, a greater postprandial fraction of whole-body oxygen consumption used by the portal-drained viscera resulted for Meishan than DWc crossbred pigs.
Both the pre- and postprandial portal vein plasma flow rates for DWc crossbred pigs of the present study are within the ranges found in our previous studies with White composite pigs (Yen et al., 1991a; Yen and Nienaber, 1993). Compared with that for the White composite pigs in our previous study (Yen et al., 1991a), the portal plasma concentration of isobutyric acid for the DWc crossbred pigs in the present study was greater (approximately 20 vs. <10 µM) during both pre- and postprandial periods. Isobutyric acid is the product of fermentative digestion of valine, a branch-chain AA. The greater portal plasma isobutyric acid concentration observed in the present study probably was due to an increased fermentation of valine contributed by the 35% dietary inclusion of the 17% CP dehydrated alfalfa meal. The valine concentration in the 17% CP dehydrated alfalfa meal is 0.86%, which is more than twice that of corn grain (0.39%; NRC, 1998) used as the primary carbohydrate in the low-fiber diet of our previous study (Yen et al., 1991a). Another branch-chain VFA is isovaleric acid, the fermentation product of leucine, another branch-chain AA. The portal plasma concentration of isovaleric acid was not different (approximately 10 µM) between the DWc crossbred pigs in the present study and the White composite pigs in our previous study (Yen et al., 1991a). This similar portal plasma isovaleric acid concentration between the two studies most likely was the result of a lack of dietary difference in leucine concentration. The concentration of leucine in the 17% CP dehydrated alfalfa meal (1.21%) is only moderately greater than that in the corn grain (0.99%; NRC, 1998). The unmeasurable concentrations of isobutyric, isovaleric, and valeric acids in the arterial plasma at preprandial or during the 5-h postprandial period observed in the present study are in agreement with our previous study (Yen et al., 1991a), indicating a complete removal of these three portal-absorbed VFA by the liver in pigs. The preprandial portal and arterial plasma concentrations of acetic, propionic, and butyric acids, and their net portal absorption in DWc crossbred pigs are also similar to those shown in our previous study with White composite pigs (Yen et al., 1991a). The value of preprandial total portal-absorbed VFA observed for DWc crossbred pigs in the present study (23.0 µM•min–1•kg BW–1) is close to that (24.1 µM•min–1•kg BW–1) for White composite pigs reported in our previous study (Yen et al., 1991a). The preprandial percent potential of absorbed VFA for meeting energy needs of whole-body heat production in DWc crossbred pigs of the present study was 23.8%, which is almost identical to 24.2% for White composite pigs of the previous study (Yen et al., 1991a). Before the 24-h fasting, DWc crossbred pigs of the present study were fed a high-fiber diet, whereas White composite pigs of the previous studies were given a typical low-fiber diet once daily. The similar preprandial total VFA absorption found in the present and the previous (Yen et al., 1991a) studies suggests that a 24-h fasting is adequate to eliminate the possible dietary fiber effect on large intestinal fermentation in growing pigs fed once daily. Feeding produced a 33% mean increase in postprandial portal vein plasma flow compared with the preprandial value in DWc crossbred pigs of the present study, which agrees with the 35% increase observed in White composite pigs of our previous study (Yen et al., 1991a). The percent increase in postprandial total net portal VFA absorption in DWc crossbred pigs of the present study is exactly the same (32%) as that observed in our previous study with White composite pigs (Yen et al., 1991a). The mean 5-h postprandial total VFA absorption (30.3 µM•min–1•kg BW–1) for DWc crossbred pigs of the present study is almost identical to the mean 12-h postprandial value (31.7 µM•min–1•kg BW–1) for White composite pigs of our previous study (Yen et al., 1991a). The postprandial percent potential of absorbed VFA for meeting energy needs of whole-body heat production for DWc crossbred pigs of the present study (24.0%) is almost identical to the 23.8% for White composite pigs of the previous study (Yen et al., 1991a).
Compared with DWc crossbred pigs, Meishan pigs of the present study had greater portal-drained viscera, and both pre- and postprandial portal vein plasma flow rates. Despite having greater large intestinal weight as a percentage of BW, Meishan pigs in the present study apparently did not have greater ability to degrade dietary fiber for VFA production than did DWc crossbred pigs. In fact, Meishan pigs had lesser concentrations of portal plasma propionic, and arterial plasma propionic and butyric acids than DWc crossbred pigs. Even with a greater portal vein plasma flow, Meishan pigs did not have greater net portal absorption of the six VFA over the 5-h postprandial period than DWc crossbred pigs. There were also no differences between Meishan and DWc crossbred pigs in total energy of absorbed VFA, whole-body heat production, and percent potential of absorbed VFA for meeting energy needs of whole-body heat production.
It should be noted that the quantities of total VFA absorption reported in the present study are net portal absorption and not the total transfer or flux of VFA from gut lumen to plasma because a portion of absorbed VFA is metabolized in the gastrointestinal wall in pigs and our portal absorption technique measures only net influx of VFA into plasma in excess of gut mucosa metabolism (Yen et al., 1991a). Nevertheless, the amount of net portal-absorbed VFA determined in the present study represents the quantity available for hepatic uptake. It also should be pointed out that the energetic efficiency of absorbed VFA in the intermediary metabolism of pigs might be only 68% that of absorbed glucose (Kirchgessner and Muller, 1991).
In the present study, the pigs were fed once daily 700 g of the high-fiber, corn-soybean meal diet during the test. The estimated DE requirement for maintenance in pigs is 110 kcal of DE/kg BW0.75 (NRC, 1998). The high-fiber, corn-soybean meal diet used in the present study was predicted to contain 2.97 kcal of DE/g, based on the analyzed chemical contents and the equation of Noblet and Perez (1993). At the beginning of the present study, pigs were given once daily 1.2 kg of the high-fiber diet. Due to bulkiness of the high-fiber diet, pigs could consume only 700 g of the high-fiber diet within a period of one hour when their physical capacity of the stomach was reached. The daily 1-h feeding was chosen to standardize the ingestion period for measurements of digestibility, VFA absorption, and energetic expenditure in pigs. The once-daily feeding of 700 g of the high-fiber diet provided 2,078 kcal of DE, which was 1.9x the DE maintenance requirement for the 21-kg pigs at the onset of the test. For pigs with ad libitum access of feed, the daily DE allowance is close to 2.6x the maintenance requirement. Thus, pigs in the present study did not receive adequate daily energy supply. In fact, the 700 g of the high-fiber diet provided only 1.6x the maintenance DE requirement of pigs during the second collection of feces and urine. It is not surprising that the weight gain during the 27-d period (from first collection of feces and urine to slaughter) was only 2.3 and 1.3 kg for DWc crossbred and Meishan pigs, respectively. Therefore, results of the present study may not be applicable to pigs allowing ad libitum access of feed.
Implications
Chinese Meishan pigs were thought to have greater ability in utilizing fibrous diets than Western breeds of pigs and might serve as a model for developing innovative approaches to increase use of fibrous by-products in pork production. In the present study, Meishan purebred and Duroc x White composite crossbred growing pigs were fed a high-fiber diet containing 35% (as-fed basis) dehydrated alfalfa meal, and measurements were conducted for the apparent total-tract nutrient digestibility, activity of fecal fiber-degrading bacteria, absorption of volatile fatty acids produced in the gut, and energy expenditure by the portal-drained viscera and by the whole body of pigs. Results of the study showed clearly that Meishan growing pigs were not superior to Duroc x White composite crossbred growing pigs in all measurements of dietary fiber utilization. Under situations similar to our experimental conditions, there is no advantage in using Meishan growing pigs in the search for enhanced use of fibrous feed ingredients by swine.
Footnotes
1 The authors thank S. S. Cummins and E. L. Shelter for technical assistance, and J. K. Byrkit for secretarial assistance. 
2 Mention of a trade name, proprietary product, or specific equipment does not constitute a guarantee or warranty by the U.S. Department of Agriculture and does not imply approval to the exclusion of other products that may be suitable.
3 Correspondence: P.O. Box 166 (phone: 402-762-4206; fax: 402-762-4209; e-mail: jtyen{at}email.marc.usda.gov).
Received for publication May 9, 2003. Accepted for publication February 24, 2004.
Literature Cited
AOAC. 1990. Official Methods of Analysis. 15th ed. Assoc. Offic. Anal. Chem., Arlington, VA.
Brouwer, E. 1965. Report of subcommittee on constants and factors. Pages 441–443 in Energy Metabolism. K. L. Blaxter, ed. Academic Press, London, U.K.
CRC. 1985. CRC Handbook of Chemistry and Physics. 65th ed. Robert C. Weast, ed. CRC Press, Boca Raton, FL.
Fevrier, C., D. Bourdon, A. Aumaitre, J. Peiniau., Y. Lebreton, Y. Jaguelin, N. Meziere, and A. Blanchard. 1988. Digestive capacity of the Chinese pig—Effect of dietary fibre on digestibility and intestinal and pancreatic enzymes. Pages 172–179 in Proc. of IV Int. Seminar of Digest. Physiol. in Pigs, Jablona, Poland.
Hansen, J. A., J. T. Yen, J. L. Nelssen, J. A. Nienaber, R. D. Goodband, and T. L. Wheeler. 1997. Effects of somatotropin and salbutamol in three genotypes of finishing barrows: Growth, carcass, and calorimeter criteria. J. Anim. Sci. 75:1798–1809.
Kemp, B., L. A. den Hartog, J. J. Klok, and T. Zandstra. 1991. The digestibility of nutrients, energy and nitrogen in the Meishan and Dutch Landrace pig. J. Anim. Physiol. Anim. Nutr. 65:263–266.
Kirchgessner, M., and H. L. Muller. 1991. Energy utilization via hindgut fermentation in pigs. Pages 41–49 in Digestive Physiology of the Hindgut. M. Kirchgessner, ed. Verlag Paul Parey, Berlin.
Nienaber, J. A., and A. L. Maddy. 1985. Temperature controlled multiple chamber indirect calorimeter-design and operation. Trans. Am. Soc. Agric. Eng. 28:555–560.
Noblet, J., and J. M. Perez. 1993. Prediction of digestibility of nutrients and energy values of pig diets from chemical analysis. J. Anim. Sci. 71:3389–3398.[Abstract]
Noblet, J., and X. S. Shi. 1994. Effect of body weight on digestive utilization of energy and nutrients of ingredients and diets in pigs. Livest. Prod. Sci. 37:323–338.
NRC. 1998. Nutrient Requirements of Swine. 10th ed. Natl. Acad. Press, Washington, DC.
Pond, W. G., and H. J. Mersmann. 2001. General characteristics. Pages 1–40 in Biology of the Domestic Pig. W. G. Pond and H. J. Mersmann, ed. Comstock Publishing Associates, a division of Cornell Univ. Press, Ithaca, NY.
Van Soest, P. J. 1963. Use of detergents in the analysis of fibrous feeds. 2. A rapid method for the determination of fiber and lignin.J. Assoc. Off. Agric. Chem. 48:829–835.
Van Soest, P. J., and R. H. Wine. 1967. Use of detergent in the analysis of fibrous feeds. IV. Determination of plant cell-wall constituents.J. Assoc. Off. Agric. Chem. 50:50–55.
Varel, V. H., W. G. Pond, and J. T. Yen. 1984. Influence of dietary fiber on the performance and cellulase activity of growing-finishing swine. J. Anim. Sci. 59:388–393.
Varel, V. H., and J. T. Yen. 1997. Microbial perspective on fiber utilization by swine. J. Anim. Sci. 75:2715–2722.
Yen, J. T. 1997. Oxygen consumption and energy flux of porcine splanchnic tissues. Pages 260–269 in Digestive Physiology in Pigs. J. P. Laplace, C. Fevrier, and A. Barbeau, ed. Proc. 7th Int. Symp. Digest. Physiol. in Pigs. EAAP Publication No. 88, Saint Malo, France.
Yen, J. T., and J. Killefer. 1987. A method for chronically quantifying net absorption of nutrients and gut metabolites into hepatic portal vein in conscious swine. J. Anim. Sci. 64:923–934.
Yen, J. T., and J. A. Nienaber. 1992. Influence of carbadox on fasting oxygen consumption by portal vein-drained organs and by the whole animal in growing pigs. J. Anim. Sci. 70:478–483.[Abstract]
Yen, J. T., and J. A. Nienaber. 1993. Effects of high-copper feeding on portal ammonia absorption and on oxygen consumption by portal vein-drained organs and by the whole animal in growing pigs. J. Anim. Sci. 71:2157–2163.[Abstract]
Yen, J. T., J. A. Nienaber, D. A. Hill, and W. G. Pond. 1989. Oxygen consumption by portal vein-drained organs and by whole animal in conscious growing swine. Proc. Soc. Exp. Biol. Med. 190:393–398.[Abstract]
Yen, J. T., J. A. Nienaber, D. A. Hill, and W. G. Pond. 1991a. Potential contribution of absorbed volatile fatty acids to whole-animal energy requirement in conscious swine. J. Anim. Sci. 69:2001–2012.[Abstract]
Yen, J. T., J. A. Nienaber, J. Klindt, and J. D. Crouse. 1991b. Effect of ractopamine on growth, carcass traits, and fasting heat production of U. S. contemporary crossbred and Chinese Meishan pure- and crossbred pigs.J. Anim. Sci. 69:4810–4822.[Abstract]
Yen, J. T., and W. G. Pond. 1990. Effect of carbadox on net absorption of ammonia and glucose into hepatic portal vein of growing pigs. J. Anim. Sci. 68:4236–4242.[Abstract]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
