Clostridium perfringens epsilon toxin increases the small intestinal permeability in mice and rats

Abstract

Epsilon toxin is a potent neurotoxin produced by Clostridium perfringens types B and D, an anaerobic bacterium that causes enterotoxaemia in ruminants. In the affected animal, it causes oedema of the lungs and brain by damaging the endothelial cells, inducing physiological and morphological changes. Although it is believed to compromise the intestinal barrier, thus entering the gut vasculature, little is known about the mechanism underlying this process. This study characterizes the effects of epsilon toxin on fluid transport and bioelectrical parameters in the small intestine of mice and rats. The enteropooling and the intestinal loop tests, together with the single-pass perfusion assay and in vitro and ex vivo analysis in Ussing's chamber, were all used in combination with histological and ultrastructural analysis of mice and rat small intestine, challenged with or without C. perfringens epsilon toxin. Luminal epsilon toxin induced a time and concentration dependent intestinal fluid accumulation and fall of the transepithelial resistance. Although no evident histological changes were observed, opening of the mucosa tight junction in combination with apoptotic changes in the lamina propria were seen with transmission electron microscopy. These results indicate that C. perfringens epsilon toxin alters the intestinal permeability, predominantly by opening the mucosa tight junction, increasing its permeability to macromolecules, and inducing further degenerative changes in the lamina propria of the bowel.

Figure 1. C. perfringens epsilon toxin alters fluid homeostasis in the small intestine.

(A) The enteropooling assay detected small intestinal fluid accumulation induced by luminal enterotoxin. Groups of 4 mice were orally administered with different doses of C. botulinum C2 toxin and intestinal fluid was determined 6 hours after oral administration. The results shown are the mean±standard error of the mean. (B) Epsilon toxin of C. perfringens altered fluid homeostasis in the mouse small intestine. Enteropooling was measured in groups of 4 mice treated with toxin (1,000 LD₅₀) or vehicle solution 2 and 6 h after oral administration. The results are expressed as the mean±standard error of the mean. (C) Epsilon toxin of C. perfringens produced fluid accumulation in mouse intestinal loops. Vehicle solution with or without 1,000 LD₅₀ of toxin were injected in ligated ileal segments. The loops were excised 3 hours after injection of the toxin and intestinal water was determined gravimetrically. Each point represents the mean±standard error. (D) Basal absorption of water by the mouse small intestine was diminished by C. perfringens epsilon toxin when measured by the single-pass perfusion method. Perfusion began with an equilibration period of 45 min. Then, Ringer solution containing 500 LD₅₀/ml of epsilon toxin or Ringer solution alone was perfused and after 60 minutes four 15-min samples were collected. The volume of each sample was determined. Each mouse was killed at the end of the perfusion, and the segment of jejunum was removed and its length measured. The absorbed or secreted volume was calculated as collected volume minus perfused volume and expressed per centimeter of perfused bowel per minute. Data were relativized to the value obtained at time 15. Changes become statistically significant after 45 minutes of toxin delivery (P<0.05).

Figure 2. In vitro characterization of the effects of C. perfringens epsilon toxin in the electrical parameters of the murine small intestine.

Epsilon toxin was incubated in the mucosal side of ileal sheets mounted in modified Ussing chambers. (A) Short circuit current (I_sc) and (B) resistance (Rt) parameters were recorded each 10 minutes in tissues from 5 mice. (C) Rt values of ileal sheets incubated in the serosal side with 8,000 LD₅₀/ml of epsilon toxin. Each bar represents results for 4 mice. Results are expressed as means±SEM.

Figure 3. Ex vivo characterization of the effects of C. perfringens epsilon toxin in the electrical parameters of the murine small intestine.

Intestinal loops were performed in groups of 4–6 anesthetized mice and injected with different concentrations of epsilon toxin in Ringer solution. In a particular set of experiments, intestines were removed after 3 hours of toxin injection and samples were mounted in Ussing chamber to measure changes in (A) short circuit current (I_sc) and (B) resistance (Rt). Those tissues were exposed to (C) luminal glucose or (D) serosal teophylline and the I_sc changes were recorded. In another set of experiments, ileal loops were injected with 1,000 LD₅₀ of epsilon toxin and incubated during different periods of time. Values of (E) I_sc or (F) Rt were recorded for those intestinal samples. Data are expressed as mean±SEM.

Figure 4. Mucosal ETX binding.

Immunofluorescent detection of ETX (A) was negative in control tissues treated only with the vehicle solution. (B) ETX (2,000 LD₅₀/ml) treated small intestinal segments gave a clear signal in the tips of the villi and (C) lower in the crypts.

Figure 5. Epsilon toxin incubated in the small intestine allowed the passage of macromolecules trough the tight junction.

(A) HRP infiltration in the small intestine was not observed beyond the tight junction in control loops of rats and mice. (B and C) In the luminal zone of the mucosa of ETX treated loops, HRP could be seen as strong electrondense deposits, filtering through the tight junction, from the lumen of the bowel, towards the chorion (arrows). (D) Extravasation of Evans blue bound to plasma protein was higher in ETX treated animals than control mice.

Figure 6. Transmission electron microscopy of control and epsilon toxin treated small intestinal loops displaying predominantly edematous changes.

In the control loop (A), a blood-vessel exhibiting normal looking endothelial cells and well preserved fibroblast (scale bar = 2,5 µm). (B) In epsilon treated loops, endothelial cell displaying early degenerative changes, such as irregular nuclear shape are seen adjacent to apoptotic cells from the lamina propria (scale bar = 2 µm). (C) Platelet aggregation is seen in one vessel, in an area with evident perivascular edema and irregular-shaped cells (scale bar = 5 µm). (D) a erythrocyte in a vessel is observed surrounded by perivascular edema (scale bar = 7 µm) and (E) degenerated fibroblast with scattered collagen fibers in the edematous lamina propria are also seen (scale bar = 7 µm). (F) Plasma cell in the proximity of an apoptotic-like cell (scale bar = 4 µm) (G) and an endothelial cell surrounding a degenerative a degenerative cell.

Figure 7. Transmission electron microscopy of control and epsilon toxin treated small intestinal loops displaying predominantly apoptotic-like changes.

In control loops (A), enterocytes with normal looking nucleus and organelles are seen (scale bar = 15 µm). In epsilon treated loops (B) fibroblast with fragmented nucleus and edema with a polymorfonuclear cell (eosinophils) is seen; note as well, the electron lucent gaps between the epithelial cells (scale bar = 10 µm). (C) Abnormal looking red-blood cells together with degenerating cells, some displaying organelle and nuclear fragmentation (scale bar = 4 µm, or (D) cytoplasmatic (see in the inlet an apoptotic cell surrounded by another cells in process of nuclear fragmentation, probably at an early apoptosis stage) and (E) nuclear condensation (scale bar = 10 and 4 µm). (F) Lymphocyte in contact with a mast-cell (upper-left corner) together with some degenerating fibroblasts (scale bar = 4 µm).