HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: jas.2007-0145v1 86/5/1156    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Meunier, J. P. Articles by Alric, M. Search for Related Content PubMed PubMed Citation Articles by Meunier, J. P. Articles by Alric, M.

Evaluation of a dynamic in vitro model to simulate the porcine ileal digestion of diets differing in carbohydrate composition¹

J. P. Meunier^*,2, E. G. Manzanilla, M. Anguita, S. Denis^*, J. F. Pérez, J. Gasa, J.-M. Cardot^*, F. Garcia, X. Moll and M. Alric^*

^* Equipe de Recherche Technologique Conception, Ingénierie et Développement de l’Aliment et du Médicament, Centre de Recherche en Nutrition Humaine, Faculté de Pharmacie, Université d’Auvergne, 28 place H. Dunant, 63001 Clermont-Ferrand, France; and and Animal Nutrition, Management and Welfare Research Group, and Departament de Medicina i Cirurgia Animal, Universitat Autònoma de Barcelona, 08193 Bellaterra, Spain

	Abstract

Top Abstract INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

 
The aim of the study was to assess the ability of a dynamic in vitro model to determine the digestibility of OM, CP, and starch compared with a validated, static, in vitro method and in vivo ileal digestibility obtained from growing pigs fitted with a T-cannula. Five experimental diets with different carbohydrate types and level were assessed: a standard corn-based diet (ST) or the same diet with coarse ground corn (CC), 8% sugar beet pulp (BP), 10% wheat bran (WB), or 8% sugar beet pulp and 10% wheat bran (HF). In the in vivo experiment, diets CC and HF reduced (P = 0.015) ileal digestibility of OM compared with the ST diet. The inclusion of sugar beet pulp reduced (P = 0.049) ileal CP digestibility of the BP diet. This reduction was not statistically significant when sugar beet pulp was combined with the wheat bran in the HF diet. No differences were shown for in vivo starch digestibility among diets. With the static in vitro method, the OM disappearance was greater than that observed in the in vivo experiment. In this static method, the BP and HF diets reduced (P = 0.004 and < 0.001, respectively) the disappearance of the OM compared with the ST diet. The coarse grinding of corn did not alter OM digestibility but decreased (P = 0.005) the starch digestibility. The R² between the in vivo results and the static in vitro methods for OM and starch digestibility was 0.99 when the CC diet was not considered. The dynamic in vitro model yielded OM and CP digestibility coefficients comparable with those obtained in vivo for the ST and CC diets. However, the values were considerably affected by the incorporation of the fibrous ingredients. Diets BP, WB, and HF had decreased (P = 0.009, 0.058, and 0.004, respectively) OM digestibility compared with the ST diet. Protein digestibility was also decreased (P < 0.001, P = 0.019, and P = 0.003, respectively) with the BP, WB, and HF diets compared with the ST diet. However, digestibility was decreased to a greater extent in the BP diet than in the WB and HF diets, both of which contained wheat bran. The R² between the dynamic in vitro model and the in vivo results for CP digestibility was 0.99 when the CC diet was not considered. No differences were detected for starch digestibility among the diets with the dynamic in vitro model. This dynamic in vitro model yielded ileal digestibility results comparable with those obtained in vivo for CP and OM with a corn-soybean diet, or with a diet including coarse corn, but it underestimated digestibility when fibrous ingredients were included in the diet.

Key Words: carbohydrate • digestibility • in vitro model • pig • sugar beet pulp • wheat bran

	INTRODUCTION

Top Abstract INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

 
Numerous in vitro models simulating digestion of the pig have been developed (Boisen and Fernández, 1997; Knarreborg et al., 2002; Bollinger et al., 2004). Most of these models are static, and the gradually changing conditions of digestion are not reflected in the in vitro process. The Toegepast-Natuurwetenschappelijk Onderzoek (TNO) in vitro model (TIM) offers a dynamic digestion process where a major feature is the use of an absorption system by incorporation of a dialysis membrane that constantly removes the nutrients already digested from the lumen and avoids saturation of the media. Thus, digestibility estimates can be obtained from measurements of nutrients absorbed instead of undigested residues.

The TIM has been used successfully to simulate the digestion process in humans (Marteau et al., 1997), pigs (Minekus, 1998), and dogs (Smeets-Peeters et al., 1998), and has been used in different scientific studies to assess the effects of enzymes on digestion (Minekus, 1998), the luminal availability of different compounds and formulations (Minekus et al., 2005; Meunier et al., 2007), prebiotic and probiotic survival (Marteau et al., 1997; van Nuenen et al., 2000), and the behavior of oral dosage forms (Blanquet et al., 2004; Souliman et al., 2006). The TIM has been used to study the digestibility of protein from different sources in pig (Minekus, 1998). However, the model has never been challenged using nonpurified pig diets, and it has never been used with diets including fibrous carbohydrates as nonstarch polysaccharides (NSP).

In this study, we assessed the TIM using practical diets with different carbohydrate composition, and we compared the results with those obtained in vivo with T-cannulated pigs and in vitro with the static method developed by Boisen (1991) to predict energy digestibility. We conducted this study to assess the ability of TIM to determine digestibility of diets differing in carbohydrate composition compared with a validated static in vitro method (Boisen, 1991) and in vivo ileal digestibility obtained with growing pigs fitted with a T-cannula.

	MATERIALS AND METHODS

Top Abstract INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

 
Dietary Treatments

All experiments were performed with 5 dry mash diets (Table 1). The diets were formulated according to the nutritive value of the ingredients and the requirements of growing pigs, as described by the NRC (1998). Diets were balanced for ME (13.47 MJ/kg of feed) and total lysine (1.11%). The standard diet (ST) consisted of corn, barley, and soybean meal ground to pass through a 2.5-mm screen. The ST diet was modified to obtain another 4 experimental diets: the coarse corn diet (CC) obtained by grinding corn to a coarser particle size (4.0-mm screen), the sugar beet pulp diet (BP) obtained by replacing some of the corn with sugar beet pulp (8%), the wheat bran diet (WB) obtained by replacing some of the corn with wheat bran (10%), and the high-fibrous diet (HF) obtained by replacing some of the corn with sugar beet pulp (8%) and wheat bran (10%). Mean particle size of experimental diets, as determined with the method described by Pfost and Headley (1976), was 472, 557, 528, 578, and 634 µm for the ST, CC, BP, WB, and HF diets, respectively. Standard geometric deviations for each diet were 2.28, 2.25, 2.22, 2.19, and 2.20, respectively. All diets included 0.15% chromium dioxide as an indigestible marker.

The treatment, housing, husbandry, and slaughtering conditions conformed to the European Union Guidelines (The Council of the European Communities, 1986).

The in vivo experiment was performed at the Experimental Unit of the Universitat Autònoma de Barcelona and received prior approval from the Animal Protocol Review Committee of the institution.

Following the procedures described by Zhang et al. (2004), 5 pigs ([Landrace x Large white] x Pietrain), 20 to 22 kg of BW and 65- to 67-d-old, were surgically fitted with a simple T-shaped cannula (16-mm i.d.) at the terminal ileum, approximately 15 cm cranial to the ileocecal junction. After surgery, the animals were housed in individual pens and allowed to recover for 20 d before the beginning of the experiment. During this period, they were fed a commercial diet ad libitum.

The experiment used a 5 diet x 5 period Latin square design. During each period, the animals were fed the corresponding experimental diet ad libitum. Each experimental period lasted 10 d, with a 7-d diet acclimation period followed by 2 d of digesta collection for two 8-h (0800 to 1600) periods on d 8 and 10. Each 8-h collection period consisted of four 1-h periods of digesta collection followed by 1-h periods of resting. Plastic collapsible bags (70-mm i.d.) were attached to the cannula, with digesta collected and frozen (–20°C) at 10- to 20-min intervals. Care was taken to ensure that the digesta flow into the bag was unobstructed during collection. Digesta samples were subsequently thawed and pooled by pig for the 2-d period, mixed, lyophilized, and stored at –20°C before analysis.

In Vitro Static Method

The 5 diets were digested following an in vitro method (Boisen, 1991), which simulates gastric and small intestinal digestion in the pig. Each diet was digested in duplicate. Briefly, 0.5 g of each diet was introduced into a 10-mL screw cap tube. To simulate stomach digestion, diets were incubated with 5 mL of 0.1 M sodium phosphate buffer (pH 6.0), 2 mL of 0.2 M HCl, 0.07 mg of chloramphenicol/mL to prevent bacterial development, and 1 mg of porcine pepsin/mL (Sigma, St. Louis, MO). The final pH of the mixture was 2.0. The tubes were kept at 39°C for 4 h in a horizontal shaking water bath. To simulate small intestinal digestion, the pH of the medium was adjusted to 6.8 by addition of 2 mL of 0.2 M sodium phosphate buffer (pH 6.8) and 1 mL of 0.6 M NaOH. Porcine pancreatin (10 mg/mL, Sigma) and amyloglucosidase (13 IU/mL, Sigma) were added and tubes were kept at 39°C for 4 h in a horizontal shaking water bath. The undigested residues were collected in a filtration unit (Fibertec system M, Tecator, Sweden) by using dried and preweighed glass filter crucibles (Foss number 2, Foss, Barcelona, Spain) to determine OM. Samples (200 µL) from the mixture were taken at the end of the digestion process.

TNO In Vitro Model

The multicompartmental, dynamic, computer-controlled system developed at TNO Nutrition and Food Research (Zeist, the Netherlands) has been described in detail by Minekus et al. (1995). The model comprises 4 serial glass compartments simulating stomach, duodenum, jejunum, and ileum, which are connected by computer-controlled valve pumps. The main variables of digestion, such as pH, body temperature, peristaltic mixing and transit, salivary, gastric, biliary, and pancreatic secretions, and absorption of small molecules (e.g., nutrients and drugs) and water are simulated. The model was programmed to mimic physiological conditions in the lumen of the growing pig gut based on information obtained from the literature.

Meal. The diets were used in the same form (not pelleted) as used in vivo and in the static in vitro model. The test meals consisted of 40 g of the corresponding diet mixed with 258 mL of distilled water and 2 mL of an aqueous solution of alpha-amylase (32500 TAU/L, Spezyme AA Genencor International B.V., Leiden, the Netherlands) simulating saliva (Young and Schneyer, 1981; Zebrowska et al., 1983).

Gastric Conditions. To control the delivery of the meal from the stomach and ileal compartments, the equation described by Elashoff et al. (1982) and modified by Decuypere et al. (1986) was used: f = 1 – 2^–(t/t)β, where f, t, t_1/2, and β represent the fraction of meal delivered, the time of delivery, the half-life of delivery, and a coefficient describing the shape of the curve, respectively. For mixed feeds, factors affecting gastric emptying can be integrated in this general model (Bastianelli et al., 1996). The values set for t_1/2 and β to reproduce gastric delivery, which are based on an average of in vivo data (Braude et al., 1976; Cuber et al., 1980; Laplace, 1981; Laplace and Cuber, 1984; Low et al., 1985; Rainbird and Low, 1986), were 150 min and 1, respectively.

Following the procedure reported by Minekus (1998), the stomach pH followed a preset curve programmed to mimic the average physiological decrease of the pH in the stomach after a meal (pH 6.0, 3.5, 3.0, 2.5, and 2.0 at 5, 30, 120, 180, and 240 min, respectively, by addition of 1 M HCl). A lipase solution of 37,500 U/L (Rhizopus lipase F-AP 15, Amano Enzyme, Chipping Norton, UK) and a solution containing 441,000 U of pepsin/L (P7012, Sigma, Saint-Quentin Fallavier, France) were adjusted to pH 4 with 1 M HCl and infused during the digestion process (0.25 mL/min each) in the stomach compartment. Both enzyme solutions were prepared with a gastric salt solution containing 3 g of NaCl/L, 1.1 g of KCl/L, 0.15 g of CaCl₂ dihydrate/L, and 0.6 g of NaHCO₃/L. A solution composed of 5 mL of each enzyme solution and adjusted to a pH of 1.5 with HCl was initially introduced into the gastric compartment with the meal to mimic an initial fasting state in the stomach.

Small Intestine Conditions. Small intestinal conditions were adjusted following the procedure reported by Minekus (1998). The half-life of small intestinal delivery of the meal was fixed at 650 min and β was fixed at 2.17. The temperature of all compartments was set at 39°C. The pH in the different compartments of the small intestine was set at an average physiological level of 5, 6.5, and 6.5 for the duodenal, the jejunal, and the ileal compartments, respectively. Pancreatic output was simulated by infusing a 10% pancreatin solution (Sigma) into the duodenum at 0.25 mL/min. Bile output was simulated by infusing a 4% bile extract solution (Sigma) into the duodenum at 0.5 mL/min. Before the experiment, the duodenal compartment was filled with 1 mL of a trypsin solution (2 mg/mL, Sigma), 14 mL of the bile extract solution, 7.5 mL of the pancreatin solution, and 7.5 mL of a small intestine salt solution (NaCl, 5 g/L; KCl, 0.6 g/L; and CaCl₂ dihydrate, 0.23 g/L). The jejunum and ileum compartments were filled with 115 mL of the same small intestine salt solution. This solution was also used as dialysis fluid, circulating in the hollow fibers at 10 mL/min.

Digestions and Sampling. The TIM differs from the in vivo and static models of digestion in that dialysis fluids are collected and analyzed to model absorbed nutrients as opposed to analyzing residual materials and estimating absorbed nutrients. Dialysis fluids from the jejunum and ileum compartments were collected during 2-h periods for 8 h. Dialysis volume was measured and weighed, and two 100-mL aliquots were collected for each period for jejunum and ileum. All aliquots were stored at –20°C and analyzed for all nutrients separately. The 5 diets were tested in duplicate. Blank runs, which replaced the diet with the gastric salt solution, were performed to determine the amount of DM, CP, and glucose contributed by infused fluids during digestion, and these data were used to correct the amounts of absorbed nutrients.

Analytical Procedures

Chemical analysis of the diet for DM, OM, and CP were performed according to standard procedures (AOAC, 1995). The GE was determined by an adiabatic calorimeter, and the Cr concentration in the diet and ileal digesta was analyzed by an atomic absorption spectrophotometer according to the procedure described by Williams et al. (1962). Total starch of feed and digesta samples and nonstarch polysaccharides (NSP) of feed were measured by the method described by Theander (1991). Water retention capacity was determined by centrifugation, as explained by Anguita et al. (2007).

Calcuations and Statistical Analysis

Ileal apparent digestibility of each nutrient fraction (Nf) of the in vivo experiment was calculated by the marker (Cr) ratio method between diet (D) and digestive content (dc) and using the equation:

In the static in vitro method, starch hydrolysis was calculated as the ratio between the amount of glucose released from each sampling time and the total amount of starch in the feed. Digestibility of OM was calculated by dividing the difference between OM in the diet and the undigested residue by the OM in the diet. Digestibility of CP was not calculated because this method was used only to validate carbohydrate digestibility.

For TIM, in vitro OM, CP, and starch digestibility coefficients were expressed as the amount of OM, nitrogen, or glucose present in the dialysis fluids from the small intestine as a percentage of the total intakes. Digestibility coefficients were calculated and reported for 6 and 8 h of digestion in the TIM.

The results of all the experiments were analyzed by ANOVA with the GLM procedure (SAS Inst. Inc., Cary, NC). For the in vivo experiment, a Latin square analysis was used, including diet, animal, and period as factors. For the in vitro experiments (static and TIM), every sampling time was analyzed separately and diet was included as a factor. Calculated digestibility coefficients were compared among models by 2-way ANOVA using diet, method, and their interaction as sources of variation. All mean separations were done using Tukey’s test. Regressions between the results in different experiments were calculated with the REG procedure of SAS. The alpha level used for determination of significance for all analyses was set at 0.05. When the P-value was less than 0.10, statistical tendencies were reported in the text.

	RESULTS

Top Abstract INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

 
In Vivo, Static In Vitro Model, and TIM Digestibility

The chemical analyses are presented in Table 2. The results obtained from the experiment on growing pigs are presented in Table 3. The inclusion of coarse-ground corn or high fiber level reduced (P = 0.015) ileal digestibility of OM compared with the ST diet. The inclusion of sugar beet pulp reduced (P = 0.049) ileal digestibility of CP (76.6% for ST vs. 65.9% for BP), but this reduction was not present when sugar beet pulp was combined with the wheat bran in the HF diet. No differences were observed for starch digestibility, which reached almost 90% in all diets.

The results of total digestibility at 6 and 8 h obtained with TIM are presented in Table 3. The amount of nutrients digested at 8 h was greater than those at 6 h. However differences among diets were similar for both times. The OM digestibility in the BP, WB, and HF diets, was decreased (P = 0.009, 0.058, and 0.004, respectively) compared with the ST diet at 8 h. Digestibility of CP in the BP, WB, and HF diets was also lower (P < 0.001, P = 0.019 and 0.003, respectively) compared with the ST diet at 8 h. However, CP digestibility in the BP diet was even decreased (P = 0.008) more than that observed with HF. Starch digestibility was, on average, numerically decreased for all diets compared with the ST diet, even though differences were not statistically significant.

Table 4 shows digestibility of CP and starch in the TIM for each 2-h period of digestion. Differences in the CP digestibility were observed mainly in periods 0 to 2 h and 2 to 4 h. In period 0 to 2 h, CP digestibility for ST diet was greater (P = 0.003, 0.009, and 0.008, respectively) than for BP, WB, and HF diets, and in period 2 to 4 h, it was higher (P = 0.037) than for BP diet. Starch digestibility was not different among the ST and the rest of the diets at any period. However, starch digestibility in the HF diet tended to be greater (P = 0.054) than in the BP diet in the last period. The amount of glucose released by the ST diet in the period 2 to 4 h was greater than the amount released by the HF diet (P = 0.018) and tended to be greater than the amount released by the diets CC, BP, and WB (P = 0.092, 0.057, and 0.086, respectively). In the period 4 to 6 h, only the HF diet tended to release a lower amount of glucose than the ST diet (P = 0.093). In this period, the CC diet released the greatest amount of glucose and was higher than the HF diet (P = 0.035).

When the 3 methods and all the diets were combined in the same analysis, the interaction was always significant (P < 0.001). Digestibility values of BP and HF diets in the TIM were always lower than the ones obtained in vivo, and only diets ST, CC, and WB presented values comparable to those obtained in vivo. Digestibility of OM was not different for diets ST and CC between TIM and in vivo methods. However, CP digestibility was greater for the ST diet (P = 0.049) and tended to be greater for the CC diet (P = 0.059) in the TIM than in vivo at 8 h. In contrast, starch digestibility was decreased (P = 0.001) in the TIM compared with the in vivo for diet CC at 8 h, but not for diets ST and WB.

The static in vitro method showed greater OM digestibility than the TIM and the in vivo for all the diets except for ST diet in the TIM. Starch digestibility in the static in vitro method was only different (P < 0.001) from in vivo for diet CC.

The static in vitro method used is known to have a high R² with in vivo digestibility. However, the R² of this method with the in vivo data were non significant when all diets were included. Diet CC did not fit in the regression. Diets SB, WB, and HF had different NSP compositions, but diet CC was the only one varying in resistant starch. Thus, we decided to exclude the CC data and observed a significant R² between the in vivo and the static in vitro method for OM (Y = 16.70 + 0.65x; R² = 0.99) and starch digestion (Y = 1.079x; R² = 0.99). Comparing the dynamic in vitro system to the in vivo, OM and CP had greater R² values (0.81 and 0.99, respectively) when the CC diet was not considered.

	DISCUSSION

Top Abstract INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

 
The TIM has been evaluated to study the digestibility of protein from different sources in pig (Minekus, 1998), and it has been recently used to study human digestibility of resistant starch types 2 (native starch granules) and 3 (retrograde starch; Fässler et al., 2006). However, those studies have been carried out with single ingredients ground for the TIM and not using practical diets where physicochemical interactions between ingredients could be expected. The aim of this study was to assess the ability of the TIM to simulate digestion and absorption of pig diets varying in the carbohydrate content. We incorporated different fibrous ingredients into a commercial diet to increase total amount of NSP, and coarse ground corn to increase the particle size (resistant starch type 1, which is physically protected from hydrolysis as defined by Englyst et al., 1996), and thus study the ability of the TIM to simulate porcine digestion of high-fiber diets. Because of the large amount of fiber present in the diet, we increased the t_1/2 in Elashoff equation (Elashoff et al., 1982), slowing gastric emptying, and extended the digestions from 6 h, as previously reported, to 8 h, permitting calculation of digestibility coefficients for both time points. These were the only variations introduced to the protocol used by Minekus (1998). For the ST diet, digestibility of OM and CP after 6 h was not different to that obtained in vivo. Thus, 6 h is enough time to mimic in vivo digestion. However, digestibility values were greater after the 8-h digestion, indicating that the TIM needs more than 6 h to completely digest nonpurified pig diets.

In the static model, OM disappearance was greater than in vivo. This difference between in vivo and in vitro estimates could result from an underestimation of real values in in vivo digestibility because of the presence of products of endogenous excretion in ileum digesta. It is accepted that more than 50% of the protein reaching cecum is of endogenous origin and this would increase the OM and could reduce in vivo digestibility 6 to 19% (Danfaer and Fernandez, 1999). Accordingly, CP digestibility of diets ST and CC in the TIM was increased compared with the in vivo values. The TIM did not show greater OM digestibility probably because starch digestibility was decreased compared with the in vivo, even though the difference was only statistically significant for CC diet (probably due to the low replicate number). Values for starch digestibility in the static model yielded intermediate results. One possible explanation for these results is that the amylolytic activity of the in vitro models is partially restricted. However, previous studies in our lab with purified diets (casein-starch) in the TIM and the static model showed that greater amounts of starch were totally digested. The known role of mastication for the reduction of the particle size of digesta and the possible contribution of the foregut microbiota to starch digestion are both factors that may be involved in the greater digestibility of starch in vivo compared with simulated in vitro conditions. The role of mastication and reducing particle size for a complete digestion is clear by comparing diets ST and CC. Mastication and the inclusion of microbial populations are technical modifications that could be incorporated to make the model more realistic.

Digestibility of OM and CP was dramatically reduced in the TIM by the incorporation of fibrous ingredients, differences being especially pronounced with diets containing sugar beet pulp (BP and HF diets), which was in agreement with the static incubation values. The TIM had never been used with high fibrous diets before, and the initial amounts of ingesta have been calculated for less bulky diets. Intestinal wall is simulated in TIM by a flexible membrane inside each glass compartment, and peristaltic movements are simulated by alternate pumping of water between the glass wall and the flexible membrane. The total volume of the TIM glass walls is fixed, and it may have created a volume limitation for high-fiber diets. This limitation may have been critical when sugar beet pulp was part of the diet because the high amount of soluble NSP of this ingredients result in an elevated water retention capacity of the diet. This volume limitation could be solved by using a smaller initial meal. This problem does not arise in the in vitro static method because the ratio between liquid and solids is clearly greater. Moreover, Anguita et al. (2007) showed in vivo that pigs ingested smaller amounts of CC, BP, and WB diet compared with ST diet because of the amount of fiber added. They suggested that decreased transit rate was one way in which the animals adapted to the diet. Thus, another option to improve TIM performance is to adjust the transit time for the diet used. Thus, for commercial diets (as the ST diet), 6 h of digestion could be the appropriate time, but it should be prolonged for diets containing high fiber levels if the initial amount of diet is not reduced.

In vivo, the inclusion of sugar beet pulp in the BP diet also produced a decrease of the ileal digestibility of CP compared with ST diet. Graham et al. (1986) previously observed that ileal digestibility of ash, protein, and fat, but not starch, was impaired when sugar beet pulp was included in the pig diet. These authors pointed to an increase in endogenous CP excretion from the epithelium as a cause of the reduced ileum protein digestibility. However, a similar impairment of protein digestibility was observed in experiments with TIM, a situation to which this hypothesis does not apply. The decrease in protein digestion in vivo is probably due, at least partially, to the physical effects of fiber on the enzymatic kinetics of digestion. It is also relevant that supplementation with wheat bran in the HF diet reduced the CP digestibility impairment associated with the sugar beet pulp observed in vivo and in vitro. This indicates that, in some cases, including mixed fiber sources in the diet may have some positive effects on digestion compared with including simple sources, which is probably due to combined changes in the physicochemical properties of digesta that can be detected both in vivo and in vitro.

In addition to estimating digestibility at a defined time, the in vitro models provide an inexpensive, rapid, and simple means of studying the kinetic of release and absorption of nutrients. In particular, TIM allows the continuous monitoring of nutrients absorbed in dialysis fluid. In this experiment we showed that differences in CP digestibility occurred in the initial 4 h of digestion, and thus were probably related to differences in enzymatic protein digestion in the stomach. Moreover, despite the absence of differences in starch digestibility, TIM allowed us to study the amount of glucose absorbed for each diet, which was different for each diet, and especially low in HF diet. Diets containing sugar beet pulp also maintained a more constant glucose absorption by avoiding acute increases in a particular period. Glucose absorption is an important variable in human nutrition, and its estimation by the TIM could be a realistic indicator of the actual in vivo glycemic index of a diet. The glycemic index of foods has been previously estimated with other in vitro methods (Goñi et al., 1997; Frei et al., 2003).

Despite the decreased digestibility achieved by TIM with high-fiber diets, results correlated with in vivo data for OM and CP digestibility. Starch digestibility did not show a significant R² between in vivo data and TIM, which could be due to the lack of a wide range of values in both methods. These regressions are based only in 4 points and replication is low; thus, these results should be carefully used and interpreted with caution. In particular, the TIM data show more variability than other procedures, and so more replications should be used in future digestibility studies.

These experiments show that TIM simulated the ileal digestibility of CP for a corn-soybean diet. Although OM digestibility correlated well with in vivo, the model appeared to markedly underestimate starch digestibility questioning the OM values. The TIM can be used to represent effects of finesse of grinding (resistant starch type 1) on in vivo DM and protein digestibility, whereas the static model does not. However, TIM did not predict the response to added fiber sources well and was poorer than a simple static in vitro model that is more realistic in representing the effects of added fiber on in vivo starch and OM digestibility than TIM. The TIM is an interesting model to study the interactions among different nutrients and digestive variables because it can study the modification of a single variable of digestion. It is also well suited to the study of the kinetics of the digestion and absorption of nutrients or additives. The development of this prognostic in vitro test should lead not only to a reduction in the work needed for formulating diets, but also in the number and size of in vivo studies required. However, some future modifications need to be introduced in this system to optimize results.

	Footnotes

 
¹ The authors are grateful to AXISS SAS, France, and to the Departament d’Universitats, Recerca i Societat de la Informació (DURSI) of the Generalitat de Catalunya for providing financial support for this research.

² Corresponding author: j-philippe.meunier{at}u-clermont1.fr

Received for publication March 7, 2007. Accepted for publication January 11, 2008.

	LITERATURE CITED

Top Abstract INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

 


Anguita, M., J. Gasa, M. Nofrarias, S. M. Martin-Orúe, and J. F. Perez. 2007. Effect of coarse ground corn, sugar beet pulp and wheat bran on the voluntary intake and physiological characteristics of digesta of growing pigs. Livest. Sci. 107:182–191.[CrossRef]

AOAC. 1995. Official Methods of Analysis, 16th ed. Assoc. Off. Anal. Chem., Arlington, VA.

Bastianelli, D., D. Sauvant, and A. Rerat. 1996. Mathematical modelling of digestion and nutrient absorption in pigs. J. Anim. Sci. 74:1873–1887.[Abstract]

Blanquet, S., E. Zeijdner, E. Beyssac, J. P. Meunier, S. Denis, R. Havenaar, and M. Alric. 2004. A dynamic artificial gastrointestinal system for studying the behavior of orally administered drug dosage forms under various physiological conditions. Pharm. Res. 21:585–591.[CrossRef][Medline]

Boisen, S. 1991. A model for feed evaluation based on in vitro digestible dry matter and protein. Pages 135–145 in In Vitro Digestion for Pigs and Poultry. M. F. Fuller, ed. CABI International, Wallingford, UK.

Boisen, S., and J. A. Fernández. 1997. Prediction of the total tract digestibility of energy in feedstuffs and pig diets by in vitro analyses. Anim. Feed Sci. Technol. 68:277–286.[CrossRef]

Bollinger, D. W., A. Tsunoda, D. R. Ledoux, M. R. Ellersieck, and T. L. Veum. 2004. A simple in vitro test tube method for estimating the bioavailability of phosphorus in feed ingredients for swine. J. Agric. Food Chem. 52:1804–1809.[CrossRef][Medline]

Braude, R., R. J. Fulford, and A. G. Low. 1976. Studies on digestion and absorption in the intestines of growing pigs. Measurements of the flow of digesta and pH. Br. J. Nutr. 36:497–510.[CrossRef][Medline]

The Council of the European Communities. 1986. Council Directive 86/609/EEC of 24 November 1986 on the approximation of laws, regulations and administrative provisions of the Member States regarding the protection of Animals used for Experimental and Scientific purposes. http://europa.eu.int/comm/environment/chemicals/lab_animals/legislation_en.htm Accessed Jan. 24, 2008.

Cuber, J. C., J. P. Laplace, and P. A. Villiers. 1980. Fistulation of the stomach and residual gastric contents after ingestion of a semi-purified diet of maize starch base in the pig. Reprod. Nutr. Dev. 20:1161–1172.[Medline]

Danfaer, A., and J. A. Fernandez. 1999. Develoments in the prediction of nutrient availability in pigs: A review. Acta Agric. Scand. Sect. A. 49:73–82.[CrossRef]

Decuypere, J. A., R. M. Denhooven, and H. K. Henderickx. 1986. Stomach emptying of milk diets in pigs. A mathematical model allowing description and comparison of the emptying pattern. Arch. Anim. Nutr. 36:679–696.

Elashoff, J. D., T. J. Reedy, and J. H. Meyer. 1982. Analysis of gastric emptying data. Gastroenterol. 83:1306–1312.[Medline]

Englyst, H. N., S. M. Kingman, G. J. Hudson, and J. H. Cummings. 1996. Measurement of resistant starch in vitro and in vivo. Br. J. Nutr. 75:749–755.[CrossRef][Medline]

Fässler, C., E. Arrigoni, K. Venema, V. Hafner, F. Brouns, and R. Amadò. 2006. Digestibility of resistant starch containing preparations using two in vitro models. Eur. J. Nutr. 45:445–453.[CrossRef][Medline]

Frei, M., P. Siddhuraju, and K. Becker. 2003. Studies on the in vitro starch digestibility and the glycemic index of six different indigenous rice cultivars from the Phillippines. Food Chem. 83:395–402.[CrossRef]

Goñi, I., A. García-Alonso, and F. Saura-Calixto. 1997. A starch hydrolysis procedure to estimate glycemic index. Nutr. Res. 17:427–437.[CrossRef]

Graham, H., K. Hesselman, and P. Aman. 1986. The influence of wheat bran and sugar beet pulp on the digestibility of dietary components in a cereal-based pig diet. J. Nutr. 116:242–251.[Abstract/Free Full Text]

Knarreborg, A., N. Miquel, T. Granli, and B. B. Jensen. 2002. Establishment and application of an in vitro methodology to study the effects of organic acids on coliform and lactic acid bacteria in the proximal part of the gastrointestinal tract of piglets. Anim. Feed Sci. Technol. 99:131–140.[CrossRef]

Laplace, J. P. 1981. The transit of digesta in the different parts of the digestive tract of the pig. Prog. Clin. Biol. Res. 77:847–872.[Medline]

Laplace, J. P., and J. C. Cuber. 1984. Total vagal deafferentation and gastric emptying in swine. Reprod. Nutr. Dev. 24:655–670.[Medline]

Low, A. G., R. J. Pittman, and R. J. Elliott. 1985. Gastric emptying of barley-soya-bean diets in the pig: Effects of feeding level, supplementary maize oil, sucrose or cellulose, and water intake. Br. J. Nutr. 54:437–447.[CrossRef][Medline]

Marteau, P., M. Minekus, R. Havenaar, and J. H. Huis in’t Veld. 1997. Survival of lactic acid bacteria in a dynamic model of the stomach and small intestine: validation and the effects of bile. J. Dairy Sci. 80:1031–1037.[Abstract]

Meunier, J.-P., J.-M. Cardot, E. G. Manzanilla, M. Whisshaar, and M. Alric. 2007. Use of spray cooling technology for development of micro-encapsulated capsicum oleoresin for the growing pig as an alternative to in-feed antibiotics; study of release using in vitro models. J. Anim. Sci. 85:2699–2710.[Abstract/Free Full Text]

Minekus, M. 1998. Development and validation of a dynamic model of the gastrointestinal tract. PhD Diss. Universiteit Utrecht, the Netherlands.

Minekus, M., M. Jelier, J. Z. Xiao, S. Kondo, K. Iwatsuki, S. Kokubo, M. Bos, B. Dunnewind, and R. Havenaar. 2005. Effect of partially hydrolyzed guar gum (PHGG) on the bioaccessibility of fat and cholesterol. Biosci. Biotechnol. Biochem. 69:932–938.[CrossRef][Medline]

Minekus, M., P. Marteau, R. Havenaar, and J. H. J. Huisint Veld. 1995. A multicompartmental dynamic computer-controlled model simulating the stomach and small intestine. ATLA-Altern. Lab. Anim. 23:197–209.

NRC. 1998. Nutrient Requirements of Swine. 10th rev. ed. Natl. Acad. Press, Washington, DC.

Pfost, H., and V. Headley. 1976. Methods of determining and expressing particle size. In Feed Manufacturing Technology II. H. Pfost, ed. Am. Feed Manuf. Assoc., Arlington, VA.

Rainbird, A. L., and A. G. Low. 1986. Effect of various types of dietary fibre on gastric emptying in growing pigs. Br. J. Nutr. 55:111–121.[CrossRef][Medline]

Smeets-Peeters, M., T. Watson, M. Minekus, and R. Havenaar. 1998. A review of the physiology of the canine digestive tract related to the development of in vitro systems. Nutr. Res. Rev. 11:45–69.[CrossRef]

Souliman, S., S. Blanquet, E. Beyssac, and J. M. Cardot. 2006. A level A in vitro/in vivo correlation in fasted and fed states using different methods: Applied to solid immediate release oral dosage form. Eur. J. Pharm. Sci. 27:72–79.[CrossRef][Medline]

Theander, O. 1991. Chemical analysis of lignocellulose materials. Anim. Feed Sci. Technol. 32:35–44.[CrossRef]

van Nuenen, H. M. C., K. Venema, M. Minekus, and R. Havenaar. 2000. Prebiotics and TNO in vitro gastro-intestinal models. Reprod. Nutr. Dev. 40:224.

Williams, C. H., D. J. David, and O. Iismaa. 1962. The determination of chromic oxide in faeces samples by atomic absorption spectrophotometry. J. Agric. Sci. 59:381–385.

Young, J. A., and C. A. Schneyer. 1981. Composition of saliva in mammalia. Aust. J. Exp. Biol. Med. Sci. 59:1–53.[Medline]

Zebrowska, T., A. G. Low, and H. Zebrowska. 1983. Studies on gastric digestion of protein and carbohydrate, gastric secretion and exocrine pancreatic secretion in the growing pig. Br. J. Nutr. 49:401–410.[CrossRef][Medline]

Zhang, Y. C., H. Jorgensen, J. A. Fernandez, and K. E. Bach Knudsen. 2004. Digestibility of carbohydrates in growing pigs: A comparison between the T-cannula and the steered ileo-caecal valve cannula. Arch. Anim. Nutr. 58:219–231.[CrossRef][Medline]

This Article Abstract Full Text (PDF) All Versions of this Article: jas.2007-0145v1 86/5/1156    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Meunier, J. P. Articles by Alric, M. Search for Related Content PubMed PubMed Citation Articles by Meunier, J. P. Articles by Alric, M.