Effects of glutamine on gut structure and function in endotoxemic rats

Abstract

AIM: To study the effects of glutamine-supplemented parenteral nutrition on protein metabolism, small intestinal mucosal morphology, and barrier function in endotoxin-treated rats.

METHODS: Thirty-five male Wistar rats were divided randomly into four groups: group A, normal control; group B, enteral nutrition (EN); group C, non-glutamine total parenteral nutrition (TPN); group D, glutamine TPN. Endotoxemia was induced by continuous intravenous infusion of endotoxin at a dose of 2 mg/kg per day throughout the 5-d study period. The small intestinal bacterial translocation rate and contents of mucosal protein, DNA, superoxide dismutase (SOD), malondialdehyde (MDA), adenosine triphosphate (ATP) and secretory immunoglobulin A (sIgA) were determined in mucosal homogenates. Mucosal thickness and villous height were measured with light microscope, and thickness of microvillus was measured with electron microscope.

RESULTS: The bacterial translocation rates of group B and group D were lower than that of group C (P < 0.01). Group D showed increased protein content and DNA content in the small bowel (P < 0.01), and, unlike the other groups, maintained the height of intestinal villi, the thickness of mucosa and the whole small intestine (P < 0.01). Group D showed increased intestinal mucosal contents of ATP and sIgA and decreased contents of SOD and MDA, compared with group C (P < 0.01). All parameters returned to normal levels in group B with EN, which also showed higher villous height and mucosa thickness than group A (P < 0.01).

CONCLUSION: Glutamine can improve gut metabolism, decrease the extent of mucosal atrophy and assist in the maintenance of the mucosal barrier function. Early use of TPN plays an important physiological role in the small bowel.

Key Words: Glutamine, Septicemia, Parenteral nutrition, Intestine, small, Bacteria

INTRODUCTION

Increasing attention has been focused on the role of the gut as a reservoir of bacteria and their endotoxins (ET), both of which may enter the circulation by crossing over a compromised gut mucosal barrier and thereby initiating a septic process and multi-organ failure. Functional defects of the gut mucosal barrier occur under various conditions, such as hemorrhagic shock, thermal injury, major operation, or sepsis. Maintenance of a healthy gut requires a variety of therapeutic measures.

The development of total parenteral nutrition (TPN) has offered a means to provide adequate calories and protein to these patients. In spite of its many advantages, TPN has been shown to lead to intestinal mucosal atrophy in both animals and man[1,2]. This atrophy is considered to be due in large part to a deficiency of glutamine (GLN), the principal energy substrate for intestinal mucosal cells, which is absent from standard intravenous amino acid solutions[3,4]. Several studies have, in addition, suggested that a deficiency of GLN in TPN may cause a breakdown of the intestinal mucosal barrier and subsequent bacterial translocation[5,6]. The addition of GLN to TPN solutions reduces atrophy of the intestinal mucosa[2,7] and helps maintain the mucosal barrier[8].

The purpose of this study was to investigate whether the provision of GLN could stimulate mucosal metabolism, reduce mucosal atrophy and maintain the gut barrier function during endotoxemia in rats.

MATERIALS AND METHODS

Animals and study procedure

Thirty-five male Wistar rats, weighing 200-220 g, were allowed ad libitum intake of water and standard rat food. After 16 h of fasting, all rats were anesthetized with intraperitoneal injections of sodium pentobarbital (40 mg/kg). The neck and interscapular region of the rats were shaved and prepared in a sterile manner for catheterization. A silastic catheter (inner diameter of 0.6 mm, outer diameter of 1.0 mm) was inserted through the external jugular vein to reach the superior vena cava. The catheter was tunneled subcutaneously to the mid-scapular region. A flexible spring guarded the catheter, which was then hooked up to an infusion pump. The rats were maintained in individual metabolic cages. The nutrient solution was infused by pump at a constant rate of 2 mL/h.

Endotoxemia in the rats was induced by intravenous infusion of endotoxin (Lipopolysaccharide, Escherichia coli O 127: B8, Sigma Chem Co, St. Louis, MO, United States) at a dose of 2 mg/kg·d in TPN solution or, for controls infusion of normal saline, from the morning of day 1 and throughout the 5-d study period.

Experimental groups

Following catheterization, all rats received intravenously 0.9% saline solution at 1.0 mL/h and ad libitum access to food and water for 24-30 h to allow for complete recovery from the effects of anesthesia before randomization to the experimental treatment groups. Rats were allocated randomly to four groups: group A (n = 6), normal control; group B (n = 9), enteral nutrition (EN); group C (n = 10), non-GLN solution; group D (n = 10), GLN-TPN solution (2.5 g/100 mL) group. The PN solution used in groups C and D was isonitrogenous and isocaloric. The daily dose of amino acids infused was 2.5 g of nitrogen per kilogram body weight, and the amount of non-protein calories given was 1045 KJ/kg per day as 50% glucose and 10% intralipid (energy index 1:1). Multivitamins and electrolytes were also included in the TPN solutions.

Endotoxemia was induced in groups B, C, and D by intravenous infusion of ET at 2 mg/kg·d in 0.9% saline solution or in TPN for 5 d. At the end of the period, the animals were sacrificed by exsanguination, and samples of the small intestine and mesenteric lymph nodes were properly treated.

General observations

All rats were weighed before or after surgery and at the time of tissue sampling. Rats with any catheter complications or infusion pump malfunction were excluded from the study. At the time of death, the intra-abdominal viscera were closely inspected before exsanguination.

Bacterial translocation rate (BTR)

Immediately after anesthesia, the rats were placed on a sterile field and the ventral abdominal wall was cleaned with 70% isopropyl alcohol and opened with sterile instruments. Several mesenteric lymph nodes (MLNs) were isolated from the ileocolic junction and put into sterilized glass tissue grinders. After manual grinding, 200 μL of the homogenate was plated onto general agar plates and agar plates with 5% sheep′s blood. These plates were then incubated at 37 °C for 48 h. Bacterial colonies were stained by Gram′s method and identification of bacterial species was performed using standard microbiologic methods.

Intestinal measurements

Following MLN excision, the rats were sacrificed by cardiac exsanguination. Ten-centimeter segments of proximal jejunum and distal ileum were then rapidly excised, opened along the antimesenteric border, and the mucosa was blotted with tissue paper and then weighed. All samples were placed at -70 °C. The mucosae of the jejunal and ileal segments were scraped off from the specimens using a glass slide and homogenized in a blender at 30000 rpm for 30 s. The homogenates were centrifuged for 15 min at 3000 ×g and the supernatants assayed spectrophotometrically for protein[9], DNA[10] and adenosine triphosphate (ATP)[11]. The contents of superoxide dismutase (SOD), malondialdehyde (MDA) (the kits were supplied by Nanjing Railway Medical College) and secretory immunoglobulin a (sIgA) were measured by simple agar diffusion test. Values are expressed per gram of intestinal tissue.

Histologic study of intestines

Following intestinal sampling for intestinal measurements, 10-cm sections of the proximal jejunum and ileum were cut, opened along the antimesenteric border, pined flat onto wax, and fixed in Boin′s solution. Tissues were subsequently embedded in paraffin, and 5-μm sections were taken and stained with hematoxylin-eosin. All slides were coded and morphologically interpreted in a “blind” fashion.

Statistical analysis

Analysis of the bacterial translocation data from rats was performed using χ² analysis. Intestinal measures are expressed as x^-± s of the sample and comparisons between treatment groups were made using one-way analysis of variance. A P-value below 0.05 was considered statistically significant.

RESULTS

General observations

All rats lost weight over the 5-d study period (Table 1), but rats in groups B and D did not differ in body weight loss, although these two groups had less weight loss compared with rats in group C. The mortality rate was 33% (3 of 9) in group B, 30% (3 of 10) in group C and 20% (2 of 10) in group D; the between-group differences did not reach statistical significance.

Table 1 General conditions.

	Group A	Group B	Group C	Group D
Weight loss, g	-	20.50 ± 4.10	29.00 ± 6.45	15.30 ± 4.12b
Death, n	-	2	3	2
BTR, % (n/total)	-	33.3 (3/9)	77.8 (7/9)	40 (4/10)b

^bP < 0.01 vs group C.

BTR

Bacterial translocation data are presented in Table 1. There was a significant increase of the BTR in the MLN of group C (7 of 9) (P < 0.01). There was no difference in the BTR between groups B (3 of 9) and D (4 of 10).

DNA and protein contents

The intestinal mucosal DNA and protein values are listed in Table 2. Group D rats had higher mucosal DNA and protein contents of all segments of gut compared with group C (P < 0.01).

Table 2 Intestinal protein and DNA contents (mg/g).

	Group A	Group B	Group C	Group D
Jejunum protein	161.79 ± 36.75	159.92 ± 43.20	142.79 ± 26.14	155.51 ± 15.88b
Ileun protein	141.62 ± 37.90	150.82 ± 23.68	140.20 ± 26.48	173.99 ± 1.77b
Jejunum DNA	14.16 ± 3.60	13.10 ± 2.83	9.55 ± 2.97	15.02 ± 3.57b
Ileum DNA	14.53 ± 3.87	12.15 ± 2.95	8.47 ± 2.39	14.32 ± 4.15b

^bP < 0.01 vs group C.

ATP, SIgA, SOD and MDA

The results of intestinal mucosal measurements are listed in Table 3. Group D had higher mucosal ATP and sIgA contents of the jejunum and ileum, but lower intestinal SOD and MDA, as compared with group C (P < 0.01). The differences between groups B and D had no statistical significance.

Table 3 Intestinal mucosal adenosine triphosphate, secretory immunoglobulin A, superoxide dismutase and malondialdehyde contents (/g).

	Group A	Group B	Group C	Group D
Jejunum ATP, mmol	2.78 ± 0.16	2.03 ± 0.17	1.78 ± 0.13	2.23 ± 0.20b
Ileum ATP, mmol	2.54 ± 0.1	2.09 ± 0.31	1.56 ± 0.22	2.14 ± 0.12b
Jejunum sIgA, μg	597 ± 79	439 ± 97	311 ± 73	460 ± 81b
Ileum sIgA, μg	546 ± 87	387 ± 65	289 ± 56	408 ± 84b
Jejunum SOD, Nu	92 ± 12	327 ± 21	365 ± 32	302 ± 29b
Ileum SOD, Nu	101 ± 14	307 ± 24	347 ± 27	286 ± 34b
Jejunum MDA, nmol	6.07 ± 1.1	13.09 ± 5.33	17.11 ± 5.21	11.67 ± 4.70b
Ileum MDA, nmol	5.13 ± 1.3	10.44 ± 2.48	14.54 ± 3.82	9.80 ± 2.52b

^bP < 0.01 vs group C. ATP: Adenosine triphosphate; sIgA: Secretory immunoglobulin A; SOD: Superoxide dismutase; MDA: Malondialdehyde.

Histologic observations

Data of the histologic cross-sections of the jejunum and ileum of all groups are presented in Table 4. Bowel wall thickness, mucosa thickness and villus height were all increased compared with those in group C (P < 0.01). Between groups B and D there was no statistically significant differences. Group D had higher microvilli than group C, and group B had the highest microvilli.

Table 4 Intestinal histologic changes (μm).

	Group A	Group B	Group C	Group D
Jejunum layer	692.25 ± 11.37	715.26 ± 17.43	605.77 ± 17.63	751.86 ± 14.88b
Ileum layer	652.20 ± 37.20	679.39 ± 7.12	502.30 ± 10.06	551.31 ± 12.32b
Jejunum mucosa	568.20 ± 69.72	674.55 ± 10.02	519.60 ± 9.31	579.08 ± 11.26b
Ileum mucosa	474.40 ± 27.31	513.52 ± 25.72	376.34 ± 9.25	423.79 ± 7.24b
Jejunum villi	364.12 ± 6.42	471.50 ± 6.71	322.57 ± 3.09	379.28 ± 8.42b
Ileum villi	241.37 ± 8.25	304.20 ± 6.1	202.37 ± 674	249.26 ± 3.78b

^bP < 0.01 vs group C.

DISCUSSION

The present experimental model of endotoxemia induced in rats by constant infusion of lipopolysaccharide with TPN was designed to mimic the clinical condition. A 20% mortality rate during the 5-d course of study indicated a moderate septic insult in this experimental model. In terms of the results of BTR and the gut morphologic observations, this model is suitable for observing the effects of GLN on the small intestinal structure and function in endotoxemia rats.

Many studies have confirmed atrophy of the intestinal mucosa in patients with long-term intravenous nutrition. In a series of experiments, investigators have tried to optimize parenteral nutritional formulas by supplementation with GLN to support the intestinal mucosa. O′Dwyer et al[2] reported that when a standard amino acid solution was supplemented with 2 g of GLN per 100 mL, jejunal cellularity increased significantly compared with the same solution supplemented with 2 g of glycine per 100 mL. In addition to PN, exogenous endotoxin further impairs gut mucosal metabolism[12]. In our study, group D (with GLN-TPN) showed increased small intestinal protein and DNA contents, and significantly maintained the level of intestinal villi, the thickness of mucosa and of the whole small bowel wall. These findings suggested that GLN administration may exert a nutritional effect on the gut mucosa, even in endotoxemia. The exact mechanism remains unclear; however, the following explanations may be made. First, GLN serves to promote cellular differentiation of the intestine[13], and increased GLN uptake and consumption by the intestine supply fuel for oxidation and also nitrogen; the latter may support nucleotide biosynthesis. Second, GLN may act as a secretogogue to stimulate the release of gut peptides, which have nutritional effects on the mucosa[14]. Both of these processes would support cellular replication and maintain gut structure.

GLN is the principal fuel for maintenance of gut metabolism, structure and function, particularly during critical illness[15]. GLN provides nitrogen for the synthesis of protein, purines, pyrimidines, and ATP, through which it improves gut mucosal metabolism and transport. Our study showed that group D (with GLN) had increased mucosal ATP content in all segments of the gut. In addition, group D had significantly increased mucosal sIgA content of the intestine, as compared with group C. SIgA is considered the major effector of the gut-associated lymphoid tissue. Endotoxin infusion may destroy the tight junction barriers of the intestinal epithelium and induce bacterial translocation[5]. SIgA can bind to bacteria, prevent their adherence to the epithelium and subsequent translocation. Therefore, GLN is thought to preserve immune barrier function[16], as shown by a decreased BTR in group D.

Moreover, rats in group D had a significantly decreased mucosal SOD and MDA contents, compared with those in group C. SOD and MDA are related to the content of free oxygen radicals and the degree of damage to cells caused by free oxygen radicals in the body. Endotoxin damaged the mucosa barrier of gut directly, it also activated xanthine oxidase to release free oxygen radicals, which further destroy gut mucosa. Provision of exogenous GLN can reduce the injury induced by free oxygen radicals and can promote the healing of damaged mucosa. It is important to guide clinical treatment of small bowel ischemia and to prevent enteric perforation in endotoxemia.

To our knowledge, this is the first report in the literature documenting the influence of GLN-supplemented TPN on intestinal ATP, sIgA, SOD and MDA. GLN can preserve not only the morphology and structure by increasing mucosa DNA and protein of the jejunum and ileum but also the mucosal barrier function by increasing mucosa ATP and SIgA and decreasing SOD and MDA contents.

It is very interesting that Group B (with EN) had the highest villi, thickest mucosa and small bowel wall among all the groups, together with increased mucosa contents of ATP and SIgA, decreased SOD and MDA contents, and significantly decreased BTR; these findings are similar to those of group D (with GLN). It has been previously shown that there is abundant GLN in the general food rich in protein and EN provides sufficient GLN. What is more, EN is the primary stimulus for gastrointestinal cell growth. Although our experimental rats were under the endotoxemic state, the small bowel still had considerable absorption function. EN can improve the structure and function of gut, and protect mucosa barrier function in endotoxemic rats. Our study confirms that use of EN plays an important physiological role in the small bowel.