Bovine colostrum in oral treatment of enterogenic endotoxaemia in rats.

Introduction Under conditions of shock, bacteria and endotoxins in the intestines can traverse the mucosal barrier by translocation and enter the blood and lymphatic system. Immunoglobulins and lactoferrin have been reported to neutralize endotoxins and bacteria. We studied the essential therapeutic factors of colostrum products in an animal experiment. Method We simulated endotoxaemia by per-oral administration of a suspension of Escherichia coli and antibiotics into the duodenum of anaesthetized rats after giving intraperitoneal carrageenan. At the same time, pure bovine colostrum or lactoferrin-enriched bovine colostrum was given. Therapeutic effects were studied by examining plasma endotoxin activity and bacterial contamination of mesenterial lymph nodes and peritoneal lavages. Albumin was used in a control group. Results The most effective bovine colostrum was able to reduce the maximum plasma endotoxin value by 67% as compared with the albumin group. The combination of this colostrum with lactoferrin brought about a reduction by 80%. The reduction in bacterial contamination of lymph nodes and peritoneal lavages was also evident. Conclusion Both gammaglobulin and lactoferrin may help to eliminate endotoxins when bovine colostrum is administered into the gut in conditions of septic shock.


Introduction
Septic shock is a frequent cause of death in intensive care medicine. Possible translocation of bacteria and endotoxins renders the gastrointestinal tract a crucial factor in this condition [1]. Reduced perfusion of the splanchnic region because of centralization results in hypoxia and oedema of the intestinal mucosa. The protective function of the gastrointestinal tract breaks down and bacteria from the gut enter the blood and lymph system [2].
A total of 35 male Wistar rats (250-350 g) were anaesthetized with ketamine. In order to achieve a degree of immunosuppression (i.e. an inflammatory state [6]) in the gut and a higher initial level of plasma endotoxin activity, 80 mg of type IV carrageenan/kg (Sigma-Aldrich Corp., St Louis, MO, USA) was injected into the peritoneal cavity. The animals were randomly assigned to five groups, each comprising seven animals (Table 1): groups 1, 2 and 3 received 400 mg iron-saturated bovine colostrum/kg of different compositions (i.e. types 1-3 bovine colostra); group 4 received a combination of 400 mg type 2 bovine colostrum/kg plus 80 mg bovine iron-saturated lactoferrin/kg (Sigma); and group 5 received 400 mg human albumin/kg (control substance).
As albumin, like other proteins, has an unspecific endotoxinbinding capacity, we wanted to have comparable protein quantities as referred to a protein commonly used in intensive care medicine. At the beginning of the 5-hour period of observation, the first blood samples were taken from the exposed external jugular vein. Neomycin and bacitracin are bactericidal and stimulate release of lipopolysaccharide from E. coli, and their administration results in increasing plasma endotoxin levels. Therefore, a combination of 5 × 10 10 colony-forming units of E. coli/kg (strain O:NT H16 clinical isolate; University of Kiel, Germany); 10 mg neomycin and bacitracin/kg; and the group-specific colostrum or albumin was administered through a per-oral tube (diameter 2 mm). The tube was fixed in the duodenum by laparotomy (L.N.), carefully avoiding damage to the efferent biliary tract. Preliminary investigations without carrageenan had shown that the maximum endotoxin level in plasma (45 ± 4 EU/dl) was reached 5 hours after administration of the bacteria/ neomycin-bacitracin suspension. Therefore, the plasma endotoxin activity was measured hourly for 5 hours. At the same time points, 2 ml blood was taken from the jugular vein for culture. The last assessment was followed by laparotomy (L.N.) and peritoneal lavages with 10 ml endotoxin-free 0.9% saline solution.
Furthermore, mesenteric lymph nodes from four different areas of the mesenterium (duodenum and colon) were resected and homogenized. Bacterial contamination of the lymph nodes and the hourly blood cultures were examined using smears on sheep blood agar plates incubated for 48 hours at 37°C. Each lymph node area was examined on one agar plate and the peritoneal lavage on three agar plates. Quantity and specification of the bacterial contamination were not assessed and no anaerobic cultures were performed.
In order to measure the biological endotoxin activity in serum, we used the modified Limulus amoebocyte/lysate test with a chromogenic substrate [7]. First, 100 µl of a 1:40 diluted (0.9% NaCl) plasma sample was heated for 5 min at 80°C. Then 50 µl of Limulus lysate (Pyroquant 50, ChB: 42-109-551; Pyroquant Diagnostik GmbH, Mörfelden-Walldorf, Germany) was added and incubated for 45 min at 37°C. inally 100 µl of chromogenic substrate (S-2423; Chromogenic Company, Mölndal, Sweden) was added and incubated for 4 min at 37°C. The reaction was stopped with 50 µl acetic acid (100%) and the extinction was measured by a photometer at a wavelength of 405 nm. The lower limit of detection is 3 EU/dl. The baseline plasma endotoxin activity without application of the bacteria/neomycin-bacitracin suspension was below the lower limit of detection.
Statistical analysis of plasma endotoxin activity was based on the arithmetic means and standard deviations. Statistical significance was evaluated using the nonparametric (distribution-free method) U test (Wilcoxon and Mann-Whitney). P < 0.05 was considered statistically significant.

Results
Starting from a control value just above the limit of detection, there was an approximately linear rise in plasma endotoxin up to the 4-hour value of 132 ± 29 EU/dl in group 5. The 5-hour value of 135 ± 24 EU/dl was only slightly higher (Fig. 1). The bovine colostra administered in groups 1 and 3 significantly lowered biological activity from the 1-hour value onward ( Table 2). The most effective suppression of biological activity was observed in group 2, in which the maximum plasma endotoxin value was 60 ± 20 EU/dl after 3 hours. At the end of the observation period the value was just 44 ± 10 EU/dl. This amounts to a reduction of 67.3% (Fig. 1). In group 4 endotoxin activity was reduced to 27 ± 7 EU/dl (i.e. a maximum reduction of 80%; Fig. 1).
Bacterial contamination of the peritoneal lavages and lymph nodes was found to be lowest in experimental group 4 ( Table 3). There was no bacterial contamination in any blood culture after incubation for 48 hours at 37°C.

Discussion
The gastrointestinal tract is of great importance for the development and prognosis of septicaemia [1,2]. The course of intensive care patients could be influenced favourably by selective decontamination of the intestine with local antibiotic therapy [8]. However, such bactericidal preparations can liberate lipid A fragments from the bacterial cell wall and thus increase the translocation of endotoxin [9]. It would therefore appear rational to combine antibiotic with a substance that inactivates both bacteria and endotoxins [10,11]. An oral dietetic would be of particular importance in this regard because plain parenteral nutrition lowers the concentration of secretory IgA in bile. This weakens immunological resistance and thus diminishes the barrier function of the intestinal mucosa [12]. We attempted to demonstrate that bovine colostrum is better able to inactivate lipopolysaccharide than albumin.
Administration of bovine colostrum has already proven efficacious in treating bacterial and viral enteritis in babies and infants [13]. Enteral administration of bovine colostrum with a high immunoglobulin content has been found to reduce perioperative translocation of endotoxin from the gastrointestinal tract [14]. However, it is still unclear which constituents of bovine colostrum are the crucial biological factors in this therapeutic effect. Neutralization of endotoxins and bacteria has been reported for immunoglobulins and lactoferrin [5,14].
In our animal experiments, group 2 exhibited the greatest suppression of plasma endotoxin level. This may be due to the high immunoglobulin content. The reduction in endotoxin Mean values and standard deviations of plasma endotoxin activity in experimental group 2 (test 2) and 4 (test 4), and in control group 5 (test 5). *P < 0.05, versus control group; **P < 0.05, versus group 2.   activity was also significant in group 1. Because that colostrum contains only half as much immunoglobulin as the colostra in groups 2 and 3, a distinct effect of lactoferrin in group 1 is possible. The iron-saturation of the preparations rules out the possibility that the bacteriostatic effect of lactoferrin derives from removal of iron from the bacterial cell wall. Therefore, lactoferrin also confers a specific defence mechanism. Intensified elimination of the endotoxin by 'natural killer cells' is conceivable because iron-saturated lactoferrin can activate these cells. The positive therapeutic effect of combining type 2 colostrum with 80 mg lactoferrin/kg supports this.
Corresponding effects were evident in the lymph nodes and peritoneal lavages. The role of systemic and local proinflammatory cytokine levels and the significance of immunoglobulins and lactoferrin [4,15,16] in such animal models should be studied in further research. Because measurement of endotoxin in biological fluids is difficult, new techniques such as the endotoxin activity assay should be considered [17].