MORPHOPHYSIOLOGICAL ADAPTATIONS OF THE GASTROINTESTINAL TRACT IN PIGLETS FED A SESAME MEAL OR SOYBEAN MEAL DIET

An important indicator to recommend a protein sourc e for piglet nutrition is the absence of intestinal damage. The effects of sesame or soybean meal based diets on Gastrointestinal Tract (GIT) morphophysiology, diarrhea incidence and severity i n piglets were studied during the first two weeks p o tweaning. Thirty-six piglets weaned at 21 days of ag e were fed one of three diets: A control casein die t (C), a Sesame Meal diet (SM), or a Soybean Meal Diet (SBM) . Diarrhea incidence and fecal score were determined once daily over a 14-day period. After 1 4 days, 3 piglets in each experimental group were f asted 12 h, fed for 1 hour and then slaughtered at 3, 6, 9 and 12 h after feeding, at which times relative G IT weights (g·kg), intestinal morphology, digesta pH, trypsin and c hymotrypsin activities were evaluated. Stomach and small intestine weight were higher (p<0 .05) in piglets fed SBM and SM than in piglets fed C. The gastric pH was higher in piglets fed C and SM a nd lower in piglets fed SBM (p<0.05). The pH of the different segments of the GIT was not affected (p>0 .05) by dietary protein source. The specific activi ty of chymotrypsin was higher (p<0.05) in animals fed C t han in those fed SM and SBM and did not vary with after feeding time (p>0.05). Trypsin activity was h igher (p<0.05) in piglets fed SBM than those fed C and SM. The dietary protein source had no impact (p<0.0 5) on villus height or duodenal and ileal crypt dep th. Jejunal villi in piglets fed SBM were shorter (p<0. 05) than in piglets fed C and SM. Dietary treatment had no effect on diarrhea incidence and severity. These findings show that sesame meal can replace soybean meal as a protein source in starter diets for weane d piglets.


INTRODUCTION
At weaning, piglets experience a period of underfeeding and diarrhea caused by the stress of separation from the sow, changing facilities and diet changes (Vente-Spreeuwenberg et al., 2001). The lowered feed intake and gastrointestinal tract development observed during weaning have been associated with diet ingredients (Lallès et al., 2007), particularly the presence of Anti-Nutritional Factors (ANFs), as well as the quantity and type of fiber in vegetable protein sources (Hermes et al., 2009). Soybean meal is the main vegetable protein source used in piglet diets because of its amino acid profile; however, it contains ANFs that may limit its use (Makkink et al., 1994;Palacios et al., 2004). Sesame meal is an alternative to soybean meal that can be used in starter diets; its high protein content and low ANF levels are Science Publications AJAVS attractive, but its phytic acid concentration may prove problematic (Jimoh et al., 2011) and there is little information about its use in piglet diets. The present study evaluates the effect of sesame and soybean meal starter diets in weaned piglets on gastrointestinal tract weight, villi height, intestinal crypt depth, gastric and intestinal digesta pH (duodenum, jejunum and ileum) and trypsin and chymotrypsin activities.

MATERIALS AND METHODS
Procedures were conducted at the CENID-Physiology experimental farm, according to the guidelines established in the Mexican Official Norm NOM-062-ZOO-1999 for production, care and use of laboratory animals (DOF, 2001) and the guidelines of the International Guiding Principles for Biomedical Research Involving Animals (CIOMS, 1985).

Ingredients and Chemical Composition of Experimental Diets
The chemical composition of the sesame and soybean meals used in experimental diets is described in Table 1. Sesame meal showed a higher content of the following nutrients compared to soybean meal: Crude protein, ether extract, neutral detergent fiber, combined methionine and cysteine, arginine and phytic acid. Soybean meal showed a higher lysine content and a higher Trypsin Inhibitor Activity (TIA) compared to sesame meal.
Three diets were formulated: Casein diet (C) (used as a reference protein), Sesame Meal diet (SM) and Soybean Meal diet (SBM) ( Table 2). Antibiotics were not used in any of the diets offered. The SBM diet contained 20% soybean meal (Friesen et al., 1993). In the SM diet, adequate sesame meal was added to ensure a protein level equal to that in the SBM diet. In the SBM and SM diets, casein was supplemented to provide an adequate amount of protein and a complete amino acid profile, as recommended by the NRC (1998). Vitamins and minerals were supplemented in all diets based on requirements recommended by the NRC (1998).

Animals and Experimental Management
A total of 36 Fertilis 20×G Performance piglets (Genetiporc) (Body Weight [BW], 7.1±0.5 kg) weaned at 21±0.8 days were used. Animals were randomly assigned to one of three experimental diets, comprising 12 piglets per experimental group. Within each experimental group, piglets were allocated into two pens based on BW, comprising 6 piglets each.
Each elevated pen was equipped with 6 manual feeding spaces, a nipple water dispenser and a plasticcovered expanded metal floor. Diets and water were available ad libitum. All piglets were weighed individually at weaning and at day 14. Fecal consistency was visually examined daily for 14 days after weaning to determine the fecal score and diarrhea incidence. Fecal score was determined using the following scoring criteria: 0 (normal), 1 (soft feces), 2 (mild diarrhea), or 3 (severe diarrhea) (Opapeju et al., 2009). The mean fecal consistency was calculated for each experimental group based on the measured fecal scores (Reis de Souza et al., 2010). The diarrhea incidence was calculated based on the mean proportion of days that diarrhea was observed relative to the total experimental period.
On day 13 at 1900 h feed was removed. On day 14 at 0700 h all piglets were fed during one hour. Then three piglets per treatment were slaughtered at 3, 6, 9 and 12 h after feeding. Piglets of all groups were stunned using CO 2 and euthanized by exanguination by severing the jugular vein.
A midline incision was made in the abdomen to expose the digestive tract. The stomach and small intestine were tied at the proximal and distal ends, removed from the abdominal cavity, emptied and weighed. The small intestine was removed from the abdominal cavity and divided into duodenum, jejunum and ileum. The pancreas was then excised, dissected from connective tissue, weighed, frozen immediately in liquid nitrogen and stored at -80°C. The organ weight was reported as a proportion of BW (relative weight, g·kg -1 ).

Laboratory Analysis
Dry Matter (DM), Crude Protein (CP) and Ether Extract (EE) were determined for each individual feed ingredient (casein, sesame meal and soybean meal) and for each diet (C, SM and SBM), according to AOAC (2000) methods 934.01, 976.05 and 920.39 respectively. The Neutral Detergent Fiber (NDF) level was measured according to the method described by Van Soest et al. (1991). Samples were prepared for amino acid determination according to method 994.12 in the AOAC (2000). Samples were hydrolyzed at 110°C for 24 h in 6M HCl. To measure the methionine and cysteine component, oxidation using performic acid was performed, followed by amino acid analysis using reverse phase chromatography on a Hewlett Packard, model 1100 HPLC apparatus, as described by Henderson et al. (2000). Trypsin Inhibitor Activity (TIA) was measured as described by Kakade et al. (1974) and phytic acid concentration was determined as recommended by Vaintraub and Lapteva (1988).
Intestinal villus height and Lieberkühn crypt depth were measured in piglets slaughtered at 12 h after feeding, using 10 cm portions of the duodenum, jejunum and ileum. The samples were fixed in neutral buffered formaldehyde, embedded in paraffin and sectioned into 4 µm slices. The sections were stained with hematoxylin and eosin (Nabuurs et al., 1993).

Statistical Analysis
Statistical analyses were performed using the GLM procedures of SAS (2008) for a completely randomized design. Enzymatic activity and pH data were subjected to a 3×4 factorial analysis (3 diets and 4 h after feeding). Relative organ weights, villus height, intestinal crypt depth, diarrhea incidence and fecal score data were analyzed as a completely randomized design with type of diet as the main effect. Differences between means were determined using a Tukey test at a 5% significance level (Steel et al., 1997). Table 3, protein source did not affect pancreatic weight (p>0.05), but the stomach and small intestine weights were higher in piglets fed SM and SBM than in those fed a C diet (p<0.05).

As shown in
Gastric digesta pH was higher (p<0.05) in piglets fed C than in those fed SBM, but no difference was observed among animals fed SM and SBM (p>0.05) ( Table 4). Small intestine digesta pH was not affected by protein source. Gastric and ileal digesta pH did not vary over time in the hours after feeding (p>0.05), but duodenal and jejunal digestas pH increase between 3 and 12 h after feeding (p<0.05).
Chymotrypsin activity was higher (p<0.05) in animals fed the C diet than in those fed the SM and SBM diets ( Table 4), regardless of time after feeding (p>0.05). Trypsin activity was higher p<0.05) in animals fed SBM than in those fed C and SM diets. Trypsin activity decreased (p<0.05) between 3 and 9 h after feeding and then increased at 12 h after feeding.  Table 4. Digestive content pH and specific activity of pancreatic trypsin and chymotrypsin of piglets fed with C, SM and SBM diets in different hours after feedings Experimental diets* Hours after feeding p< .90 *C = Casein diet (reference diet), SM = sesame meal diet, SBM = soybean meal diet. D = Diet effect, H = Hour after feeding effect, D*H = Diet*hour after feeding interaction. SEM = Standard Error of Mean. a,b Values with different superscripts between diets in the same raw are significantly different (p<0.05). AB Values with different superscripts between hours after feeding in the same raw are significantly different (p<0.05). NS = Non Significant  Intestinal morphology results are described in Table  5. Villus height and duodenal and ileal crypt depths were not affected by dietary protein source (p>0.05). The average jejunal villus length was longer in piglets receiving the SM diet (p<0.05) than in piglets receiving the SBM and C diets. Jejunal crypt depth was equal (p> 0.05) among all experimental groups.

AJAVS
The dietary protein source did not affect the diarrhea incidence or the fecal score (p>0.05).

DISCUSSION
Stomach and small intestine weight was greater in piglets fed a vegetable protein diet than in those fed the casein diet. This may be caused by the higher fiber content and the presence of the trypsin inhibitor in vegetable proteins, as previously reported by Csaky and Fekete (2004). Furthermore, the experimental diets did not cause any detrimental effect on visceral weight of the experimental animals, which is similar to findings in other studies (Csaky and Fekete, 2004;Reis de Souza et al., 2007;Opapeju et al., 2008).
Stomach digesta pH values found, indicate that the diet modulates the chemical characteristics of gastric secretion. The optimal stomach pH range for pepsin protein digestion (Makkink et al., 1994;Morales et al., 2012;Heo et al., 2013) is 2 to 4. The lowest gastric acidic content was observed in piglets fed the C diet and it is likely caused by casein's higher buffering capacity due its greater number of amino acids with carboxylic radicals, as well as the presence of phosphate radicals (Salaun et al., 2005). Sesame meal also showed a high buffering capacity, which is derived from its high concentration of arginine and cysteine and prevented a drastic pH decrease in the stomach. Soybean meal showed comparatively less buffering capacity, probably due to its high concentration of acidic amino acids (glutamic and aspartic), as it has been demonstrated in other legumes (Al-Dabbas et al., 2010). Duodenal digesta pH values observed in piglets in the present study (average pH = 5.7), showed that the acidic chyme was neutralized to protect the intestinal mucosa and prevent denaturation from digestive enzymes. The pancreas also contributes to duodenal pH by secreting bicarbonate ions into the duodenum under the influence of the hormones Cholecystokinin (CCK) and secretin (Clemente and Domoney, 2006). In the present study, the progressive increase of pH from the stomach to the ileum allowed the proper conditions for enzymatic activity in the small intestine (Makkink et al., 1994;Morales et al., 2012).
The decreased activity of trypsin in the first 9 h after feeding was a consequence of decrease in the quantity of protein present in the gastrointestinal tract. Makkink et al. (1994) observed that the secretion of proteolytic enzymes from the pancreas is directly proportional to the presence of protein in the gastrointestinal tract, which decreases over time after feeding. However, the increase in specific activity of trypsin (expressed as IU·mg protein −1 ) observed at 12 h after feeding, probably due the decrease of pancreatic protein concentration in absence of feed (fasting) in the gastrointestinal tract (Nagy et al., 1989). Enzyme and hormone secretion from the digestive tract in weaned piglets depends on the degree of gastrointestinal stimulation from the feed ingested (Cranwell, 1995).

AJAVS
Other factors modulating digestive tract function are the protein source quality, protein digestibility, feed processing and the presence of ANF (Makkink et al., 1994;Hermes et al., 2009). Findings in the present study indicate that ANF is one of the most important factors that may explain the influence of the protein source on digestive enzyme activity and morphology of intestinal mucosa. We suspect that the higher trypsin specific activity in piglets fed SBM is caused by the trypsin inhibitor factor in soybean meal, which in turn may stimulate CCK hormone secretion from intestinal cells, resulting in greater pancreatic trypsin synthesis and secretion through negative feedback mechanisms (Owyang et al., 1994). This negative feedback loop mediated by CCK has been reported in rats, pigs, calves and humans (Clemente and Domoney, 2006;Woyengo et al., 2009).
The chymotrypsin specific activity in piglets fed SM and SBM was lower than in piglets fed C, a finding possibly resulting from the presence of phytic acid. Phytic acid may induce ionic bonding to the basic amino acids in proteins such as chymotrypsin (lysine and arginine) at the pH present in the small intestine (5.7 to 7.6) (Adeola and Sands, 2003).
Previous studies have revealed the presence of certain allergenic proteins (glycinin and β-conglycinin), lectins and trypsin inhibitor in soybean meal (Csaky and Fekete, 2004;Purushotham et al., 2007), all of which can decrease jejunal villus height, as observed in piglets receiving SBM diet. Li et al. (1991) and Csaky and Fekete (2004) have observed similar effects in weaned piglets fed soybean meal diets. Zhan et al. (2008) and Shan et al. (2012) observed that dietary supplementation of arginine improved intestinal mucosa development in weaning piglets. Sesame meal contributes to 48% more arginine than soybean meal, which may explain the larger jejunal villi observed in piglets fed SM diet. Arginine is theorized to increase enterocyte metabolism by increasing nitrogen transport in tissue proteins; it is also a substrate in multiple enzymatic pathways, including arginase, nitric oxide synthase, ARG-glycine aminotransferase and arginyl-tRNA synthetase and is a precursor for the synthesis of creatinine, proline, glutamate, polyamines and nitric oxide synthase (Wu et al., 2007).
Diarrhea incidence and associated fecal score in all piglets, regardless of diet offered, were consistent with what is commonly observed at weaning. Fecal score was similar to values observed by Opapeju et al. (2008) and Bhandari et al. (2009) at 14 days post-weaning. Diarrhea incidence in the present study was higher than that reported by Hermes et al. (2009), who used diets containing antibiotics, but fecal score did not show any relative increase. Opapeju et al. (2008) and Bhandari et al. (2009) observed fecal scores during the second week postweaning in animals fed SBM diets without antibiotics, which is similar to fecal scores reported in the present study. Zhan et al. (2008) and Grimble (2007) have reported that arginine induces nitric oxide-mediated water and electrolyte secretion, which at low levels (0.7%) acts as an absorber and at high levels (>1.2%) as a secretagogue; however, in the present study, the arginine level in the SM diet (0.86%) was not high enough to induce diarrhea or increase diarrhea incidence.

CONCLUSION
A sesame meal diet fed to piglets during the weaning period did not have a negative effect on gastrointestinal tract development, digesta pH, diarrhea incidence, fecal score, trypsin activity, or small intestine villus morphological characteristics. In addition, the higher content of crude protein, sulfur amino acids and arginine, together with the lower trypsin inhibitory activity in sesame meal compared to that of soybean meal, make it an important alternative protein source for piglets feeding during the weaning period. However, further studies about nitrogen digestibility and growth performance in piglets are needed to recommend the use of sesame meal in starter diets.

ACKNOWLEDGEMENT
Partially funded by SEP-CONACYT Grant No. 179898 CB-2012-01. The authors gratefully acknowledge to Dipasa Internacional de México, S.A. de C.V. for providing the sesame meal for the experiment.