The pig industry has advanced in the development of precocious genetic lines with better production traits and the weaning of piglets at younger ages (7 - 21 days) (Touchette et al., 2002; Gómez, 2006). As a result, piglets are lighter at weaning and have a less developed digestive system, which makes them more susceptible to digestive problems (Reis et al., 2007a).
Early weaning results in a short period of fasting right after weaning, and the disappearance of the lactobacilli population that was predominant in stomach and intestine. This creates an imbalance that favors the increase of the E.coli population, which cause the release of lipopolysaccharide (LPS) from the cell wall of these bacteria (Amador et al., 2007).
LPSs are pathogenic compounds that increase indiscriminate paracellular transport of molecules within the intestine, causing diarrhea, structural and functional alterations (Zhenfeng et al., 2008), and in consequence, deficient absorption and utilization of nutrients (Pitman and Blumberg, 2000; Fan, 2002; GarcÌa-Herrera et al., 2003).
Previous studies have described pre and post weaning intestinal morphology in pigs (Pluske et al., 1991; Reis et al., 2007b; Gomez et al., 2008), but little information exists on the combined effect of weaning and LPS on the morphology of the intestine in this species (Albin et al., 2007).
The prevention of post-weaning diarrhea syndrome has traditionally been based on the dietary inclusion of antibiotics, copper, and zinc. However, increased bacterial resistance and environmental concerns have led to an international trend to ban the use of antibiotics in animal diets and to reduce mineral inclusion levels. Such policy changes require alternative methods that allow the control of post weaning disorders. Therefore, it is necessary to understand the detailed mechanisms causing structural and functional alterations in the gut during this period in order to develop comprehensive management and control alternatives.
This study aimed at evaluating intestine structural changes, specifically the effect of supplying LPS from E. coli on the morphology of the villi and glands of weaned pigs. The assessment of these changes is important to understand the influence of nutrition on intestinal development after weaning, to identify therapeutic targets and to test therapeutic strategies aimed to efficiently treat post-weaning diarrhea.
Materials and methods
All experimental procedures were conducted according to guidelines suggested by ìThe International Guiding Principles for Biomedical Research Involving Animalsî (CIOMS, 1985). This research was approved by the Animal Experimentation Ethics Committee of the Universidad Nacional de Colombia, MedellÌn (CEMED 001 from January 26, 2009).
Fieldwork was conducted in the Centro San Pablo of the Universidad Nacional de Colombia, MedellÌn, located in the municipality of RÌo Negro, at an altitude of 2100 meters above the sea level, with average temperatures between 12 and 18ºC, corresponding to an area of very humid low montane forest (bmh-MB).
Type of study
Four experimental diets were evaluated: a control diet (basal diet), and three others containing LPS from E. coli, serotype 0111: B4 (SigmaAldrich, Sigma-Aldrich, St Louis, MO, USA), as follows:
- Basal Diet (BD): without LPS.
- Diet 1 (D1): BD plus 0.3 µg of LPS / mg of food.
- Diet 2 (D2): BD plus 0.5 µg of LPS / mg of food.
- Diet 3 (D3): BD plus 1.0 µg of LPS / mg of food.
Animals and diet
The experiment used 52 pigs obtained by alternate crossing of Duroc x Landrace. Piglets were weaned exactly on their twenty first day after birth, weighing 6.5 ± 0.5 kg. The weaned pigs were housed in groups of eight with ad libitum water, in a controlled-temperature room at 26 ± 3 °C. The basal diet consisted of milk and milk by-products enriched with vitamins, minerals, and lysine-HCL. Diets were formulated to meet the minimum nutritional requirements proposed by the NRC (1998) (Tables 1 and 2).
The amount of food offered per cage was 300 g/day; however, additional food was supplied when required. The experimental diets were offered from days 1 to 10 post weaning. During lactation no solid food was offered to the piglets.
Throughout the experimental phase all 52 pigs were sequentially slaughtered, as follows: the first day (day of weaning; day 1), four pigs representing the reference group were slaughtered to check the overall health and to evaluate the macro and microscopic condition of the organs before the beginning of the experiment. On days 5, 7 and 10 post weaning four pigs were slaughtered in each treatment. Diets were provided from the time of weaning up to 2.5 hours before slaughter.
Sampling of the small intestine
The animals were sedated by inhalation of carbon dioxide for 3 minutes, and then slaughtered by exsanguination through a section on the jugular vein. After slaughter, the pigs were placed in supine position to remove the small intestine (from the pyloric junction to the ileocecal valve) through an abdominal incision. The intestine was aligned on a table, measured without any tension, divided into three sections (duodenum, jejunum, and ileum) of equal size, and 20 cm sections were taken from each segmentís center. Once the portions were cut, the content from each one was removed by washing with cold saline infusion as previously described by Makkink et al. (1994), and Reis et al. (2005). Then, 1cm long sub-samples were obtained from each segment. Samples were preserved in 10% buffered formalin and subsequently stored until performing laboratory determinations.
Samples from three regions of the small intestine were processed and analyzed in the Laboratory of Animal Pathology at the University of Antioquia. The tissues were sliced in 4 µm thick cuts, and stained with hematoxylin-eosin according to the methodology reported by Nabuurs et al. (1993). Three transverse cuts were mounted per slide.
Microscopic evaluation and morphometric analysis of images
The histological sections were analyzed quantitatively by computerized digital image processing, as follows: An optical microscope (Leica DMLB, Meyer Instruments, Houston, TX, USA) was used to identify tissue areas; then, the corresponding images were captured with a three-megapixel 200X zoom camera for instant digital microscopy (Moticam 2300, Motic, Hong Kong, China). The images were analyzed with Motic® Images Plus 2.0 image treatment software (Motic, Hong Kong, China).
The morphometric variables measured in each tissue section were: villus height, width and area. Intestinal glands' depth and width were also determined, as previously described by Nabuurs et al. (1993) and Marion et al. (2002). The average value for each variable was calculated after performing measurements in eight villi and their corresponding intestinal glands. Due to the fact that villus height may vary in each intestinal fold, being shorter at the apex, it was required that each region was equally represented in the assessment. In consequence, a circular fold of the mucosa was chosen, measuring two villous from the bottom, two on the right, two from the left side and two from the vertex. This procedure was repeated in each section of the small intestine (duodenum, jejunum and ileum) allowing to verify the effect of different diets on the villi according to their location. As far as we are aware, this analysis has not been performed previously.
The experiment was conducted as a randomized block design (two blocks) in a 4x4 factorial arrangement (four experimental diets and four periods after weaning) (Steel and Torrie, 1985). The animals were blocked by initial weight. Each animal was assigned one of 16 treatments, and each treatment had four repetitions. Statistical data analysis was conducted using the General Linear Models procedure (GLM) of SAS program (2006). A Duncan test was used to compare treatment means (p<0.05).
Pigs fed the basal diet showed good health condition and behaved normally, whereas those receiving LPS showed increases in rectal temperature (above 38ºC) throughout the experiment. However, they didn¥t show any signs of illness that would force their retirement or immediate slaughter. No food leftovers were observed during the experiment.
This study compared the data obtained with the BD in each of the post weaning periods and intestinal sections, in order to determine the effect of weaning on intestinal glands and villi morphology (Tables 3 and 4). Villi's height and area decreased from day one after weaning (p<0.01; Table 3). Animals with at least five days post weaning had the lowest values for height and area (324.7 µm and 35042 µm2, respectively), while villi's width had the highest value (112.2 m). For glandís width and depth (Table 3) significant differences were observed (p<0.01) from day one, reaching their peak on day five after weaning (43.14 µm2 and 207.1 µm2, respectively). In each of the variables under study a partial recovery is shown as time goes by, specifically on day 10 after weaning. However, there were no statistically significant differences between days one and 10 post weaning (p<0.01).
Significant differences were found among sections of intestine (p<0.01) for the variables studied, where the proximal section (duodenum) showed the highest values for villi height and area (354.4 and 34789 µm2, respectively) and the lowest values for gland depth and width (112.4 and 89.9 µm, respectively) (Table 4). Jejunum had the highest values for villi and glands width, (112.3 and 114.5 µm, respectively). Nevertheless, there were no statistically significant differences between the middle section (jejunum) and distal (ileum) for the studied variables (p>0.01).
The average values of the intestinal variables are presented in Tables 5, 6, and 7. There was no statistical interaction between the factors involved (LPS concentrations and post weaning slaughtering periods) for any of the variables studied. Significant decreases were observed (p<0.01) among diets regarding villi's height (Figure 1) and area (Table 5). Animals on D3 had the lowest values for these traits (256.4 m, and 29,309 m2 respectively) compared to samples from those on BD (347.1 m and 36232 m2, respectively). Conversely, villi's width for animals on D3 (128.4 m) showed a significant increase (p<0.01) in comparison with those on BD (106.4 m). For depth and width of the glands, animals on D3 showed (p<0.01) higher values (246.1 m and 121.5 m, respectively) compared to samples from animals on DB (111.2 and 108.4 m, respectively).
During the post-weaning period (Table 6), villi's height and area decreased (p<0.01) from day one, with animals slaughtered 10 days after weaning presenting the lowest values (259.1 µm and 30579 µm2, respectively), while villi's width had the highest values (130.9 µm). For depth and breadth of glands (Table 6), significant increases were observed (p<0.01) since day one, reaching the highest level on day 10 post weaning (197.1 and 132.4 µm).
Significant changes (p<0.01) were observed among intestinal portions for the variables under study (Table 7). The proximal portion (duodenum) had the highest values for villi's height and area (337.7 µm and 39084 µm2, respectively) and the lowest values for glands depth and width (180.3 and 88.7 µm, respectively). For width, both in villus and glands, jejunum had the highest values (126.5 and 123.9 µm, respectively). However, there were no statistically significant differences between the small intestineís mid (jejunum) and distal (ileum) sections for the variables studied (p>0.01)
At weaning, the observed values for villi's height and width, as well as those for depth and width of glands, were similar to the ones reported by other authors (Nabuurs et al., 1993; Marion et al., 2002; Reis et al., 2005). It was further verified that weaning decreases villi's height and area, and induces an increase in villi's width. Weaning also increases intestinal glands' depth and breadth, which is in agreement with previous reports (Hedemann et al., 2003). It was also confirmed that recovery of these parameters occurs around day five, which agrees with previous work (Hedemann et al. 2003; Vente-Spreeuwenberg et al. 2004ab) reporting that villi's growth recovery occurs between 5-8 days post weaning, returning to normal levels between 9-14 days after weaning, under normal conditions (Nabuurs et al.1993; Hedemann et al. 2003; VenteSpreeuwenberg et al., 2004a).
The ratio between villi's height and glands' depth is of great importance and should be at a maximum, due to the fact that minimum values are associated with a decrease in digestion and absorption during the first week post weaning. After weaning, the ratio between villi and glands is affected by a change in the microbial population, the intake of solid food, and allergic reactions (Rodrigues et al., 2007).
Although the mechanisms underlying the changes observed in this study are not fully understood at present, some researchers postulate that such changes are due to the fact that piglets are subjected to environmental, social and nutritional changes at weaning, which favor the appearance of various stress symptoms (LallËs et al., 2004a). Due to this, the post weaning period is characterized by an immediate but transient reduction in food intake, leading to a state of malnutrition and stunted growth. This affects several aspects of the small intestineís architecture and functions, ultimately causing diarrhea. Among others, the following changes have been reported: villi atrophy (Van Beers-Schreurs et al., 1998), deepening of intestinal glands, and decreased digestive enzyme activity (Vente-Spreeuwemberg et al., 2001; Hedemann et al., 2006).
Some research has shown early signs of inflammation after weaning, including leukocyte infiltration, increased expression of several proinflammatory cytokines, enhanced cytoprotection (by over-expression of heat shock proteins), tissue alterations caused by proteases, and epithelial function disorders related to absorption, mineral secretion and intestinal permeability. Nonetheless, after these disorders, there is an intestinal regeneration phase, probably stimulated by the return of feed consumption, which leads to a restored normality in these parameters (LallËs et al., 2004a).
Supplying LPS from E. coli had an additive effect on these variables, altered as a result of weaning. Such effect was dose-dependent, since the greatest changes occurred with the highest dose (D3). The observed shortening of villusí height is probably due to a decrease in villus cells numbers, from the adverse effects of LPS, and alteration in cell turnover rate, which were significantly lower in the animals on DB. Similar effects have been observed after experimental infection of gnotobiotic pigs with E. coli (Willing et al., 2007).
Changes in villi' and glands' morphology (McCracken et al., 1999; Li et al., 2001) represent a balance between gland cells production (Nabuurs et al., 1993; Jin et al., 1994) and loss of villus cells. The decrease in villus height reduces the area for digestion and absorption of nutrients during this period (Rodrigues et al., 2007). The food that is not digested and absorbed in the small intestine ends up in the cecum and colon, generating intense activity and proliferation of microbial population, mainly enteropathogenic, which triggers diarrhea processes that can cause death (Lallès et al., 2004b).
The increased villi's width observed during the post weaning period agrees with previous descriptions (Cranwell, 1995; Yen, 2002). Such change suggest a compensatory response to decreased height (villous atrophy), which occurs in the days after weaning, as described above (Pluske et al.1997; Vente-Spreeuwenberg et al., 2001).
The observed changes in the glands' width and height in jejunum and ileum have been reported by some authors as a normal process related to adaptation, which takes place in the post weaning phase, indicating an accelerated mitotic activity (Pluske et al. 1997; Vente-Spreeuwenberg et al., 2004a). Hedemann et al. (2003) reported that the jejunum's villi are between 350 to 450 µm long at weaning and reduce their size from then, returning to their normal size 14 days after weaning. In consequence, villous atrophy and cell renewal, which determines their recovery and growth, vary according to the intestinal section (Reis et al., 2005). In this study, we observed that these changes were increased in jejunum and ileum due to LPS intake, so these sections are probably more susceptible to the effect of this bacterial component, probably due to an increased specificity for its receptors. This hypothesis must be tested in future research.
In the present study, the effect of E. coli's LPS on intestinal morphology could be largely due to the inflammatory response induced by this component. LPS is known to have the ability to activate a variety of signaling pathways (Amador et al., 2007; Foot et al., 2004) through proinflammatory cytokines such as IL-8, IL-18 and TNF-α (Johnson et al., 2005). These cytokines affect normal cell growth and induce changes in the intestineís structure and functional capacity (Garcia-Herrera et al., 2008; Yoo et al., 2000). In general, this may be attributed to two factors: 1) inflammation causes changes in the intestinal cells architecture and normal functions, promoting the release of inflammatory mediators (Johnson et al., 2005); 2) enterocytesí normal physiology can be profoundly affected by factors derived from coagulation and blood complement system (Yuji et al., 2003).
In addition to the above, some research has shown that Fas receptors co-stimulation with TNFα or INFγ on human enterocytes can induce apoptosis (Ruemmele, 1999). Studies performed on gnotobiotic piglets indicate that conventional E. coli bacteria, contrary to the L. fermentum bacteria, increases general cell turnover by stimulating apoptosis through the expression of FasL, TNFα and increased cell proliferation. However, in pigs, it has not yet been established whether a similar mechanism causes the decrease in intestinal villi cells, and consequently their atrophy after LPS action.
Another mechanism that could be involved in intestinal disorders arising from LPS-induced inflammation is the disruption of intestinal permeability caused by TNF-α. This cytokine alters intestinal permeability through its effect on tight junctionsí structure between epithelial cells (Fengjun et al., 2005), mainly in jejunum (Foot et al., 2004; LallËs et al.2004a). The above mentioned, combined with E. coli's LPS effect, which causes the shortening of actin filaments within the enterocyte, could permeate tight junctions structure to infection-promoting bacteria and compounds, leading to inflammation (Berkes et al., 2003).
From this study we conclude that E. coli's LPS has an effect on intestinal morphological parameters, specifically decreasing villi's height and area, and increasing glands' depth and width. Such effect probably contributes to lowering intestinal nutrient absorption, co-infection with other pathogens, and onset of the post-weaning diarrhea syndrome.
This study shows the usefulness of computerized morphometric analysis to reliably, objectively and reproductively evaluate LPS effects on intestinal morphological parameters. In consequence, it could be used in future research to investigate the physiopathology of this and other enteritis-causing agents, and for evaluating therapeutic strategies.
From these findings it is suggested the need for more research on the digestive physiology associated with pathology, microbiology and immunology, in order to improve the understanding of the mechanisms responsible for digestive problems during the critical post-weaning period.
The authors thank the Research Committee (CODI) of the University of Antioquia for cofunding this research (code 20096000), and the Medellin Research Direction (DIME) of the National University of Colombia for financing it (project codes QUIPU 20301007862 and 20101007950), which made this research possible.
This article was originally published in Revista Colombiana de Ciencias Pecuarias 2011; 24:598-608. This is an Open Access article published under a Creative Commons Atribución-NoComercial-CompartirIgual 4.0 Internacional License.