Main

Protein energy malnutrition is a major complication of chronic cholestasis in infants and children, particularly those suffering from extrahepatic biliary atresia not corrected by the Kasaï portoenterostomy and those awaiting orthotopic liver transplantation. To achieve positive metabolic balance, pediatric hepatologists have supplemented the diet of these patients with dextrine maltose and medium chain triglycerides for their fuel capacity and absorption mechanisms that are independent of endoluminal bile acids and micellar transport. Previous studies have shown that proteins are preferentially used in these children for energy purposes, because low protein intake leads to endogenous protein catabolism(1, 2).

BCAA have been shown to stimulate protein synthesis and to decrease catabolism due to cirrhosis(3). In various human pathologic situations, they may improve or even reverse the negative nitrogen balance. In human cirrhosis, BCAA may improve the nitrogen balance by increasing protein synthesis and by decreasing their catabolism(4, 5).

The aim of the present study was to evaluate the effects of a diet enriched in BCAA on growth, nitrogen balance, and body composition in a rat experimental model of extrahepatic biliary atresia cirrhosis(6).

METHODS

The present study was approved by the ethical committee of animal welfare. Sixty male Wistar rats were used. They were born in the animal house and kept in standard conditions of light and temperature, in sawdust-lined cages; they had free access to food pellets (UAR, Animal Labo, Brussels, Belgium) and water. They were weaned at d 21 and divided into four series of 15 rats, including for each series 12 experimental rats with three spare animals in case of postoperative death.

Group A. Control rats were sham-operated at d 30 under ether anesthesia. They had free access to a normal laboratory diet, containing 17.5% of energy as proteins (casein).

Group B. In group B rats (Chol/BCAA) at d 30, bile duct excision and ligation caused complete obstructive cholestasis progressing to cirrhosis(6). They had free access to a diet enriched in BCAA: 17.5% proteins, plus 8.5% BCAA (leucine/valine/isoleucine, ratio 1:1:1) included in the laboratory pellets during the industrial fabrication procedure(Animal Labo, Brussels, Belgium).

Group C. Group C rats (Chol/HP) had the same operation as in group B, but with free access to a diet supplemented in proteins (17.5% proteins plus 8.5% casein included in the food pellets during the manufacturer's fabrication procedure (Animal Labo, Brussels, Belgium)).

Group D. These cholestatic rats (Chol/Con) received the same operation as in group B and had free access to a normal laboratory diet (17.5% of caloric intake as proteins).

Animal weight and food intakes were recorded daily from the time of surgery(d 30) until sacrifice 32 d later (d 62). Rats were killed under ether anesthesia followed by cervical dislocation.

Nitrogen balance. Thirty days after surgery, nitrogen balance was measured in all animals. Rats were placed in a metabolic cage for separate collections of stools and urine. Urinary nitrogen was measured according to standard laboratory procedures for measurements of ammonium, urea, creatinine, and uric acid.

Muscle mass estimation. The soleus and extensor digitorum longus muscles were dissected from both posteroir legs for assessment of weight and protein concentration. The latter parameter was expressed as milligrams/g of muscle weight. For this analysis, a 10-mg sample was taken, and the remaining muscle was replaced in the animal carcass for carcass analysis.

Carcass analysis (7). After sacrifice and muscle dissection as mentioned above, the rat small intestine was washed out with saline and then replaced in the body. The carcass was weighed(W2) before being frozen at -20 °C in airtight plastic bags. The animal carcass was thereafter placed for 72 h in a laboratory lyophilizator (Edwards High Vacuum, Manor Royal, Crawley, West Sussex, UK) until the internal chamber pressure remained stable. Finally, the carcass was placed at 35 °C for 12 h to prevent further evaporation, and finally weighted (W3). The carcass was then cut in small pieces and mixed in a domestic Kitchen Robo (Magimix 2000, Van den Borre, Brussels, Belgium), until an homogeneous fine powder was obtained. The powder was then stocked in plastic bags at room temperature. The percentage of body water was estimated by the difference of weight before and after lyophilization, divided by the initial carcass weight (W2 -W3/W2).

Body protein composition was obtained by measuring the protein concentration in the carcass powder. Fifteen to 20 mg of powder were diluted 1:500 in 0.5 M NaOH, homogenized for 60 s using a Ultraturax homogenizer(Janke-Kunkel, Van der Heyden, Brussels, Belgium), and incubated for 12 h before the protein concentration was determined by the method of Lowryet al.(8). Results were expressed in grams of protein/100 g of wet body weight.

Lipid content of the carcass was measured according to standard described methods, using methanol/chloroform extraction from a 1-g sample of carcass powder(9). Results were expressed in grams of lipids/100 g of wet body weight.

Ash content of the carcass was measured after complete calcination of a 2-g sample in an oven at 900 °C. The calcinated sample was cooled in a desiccator at room temperature before being weighed. Results were expressed in grams/100 g of wet body weight.

Statistical analysis. Results were expressed as means ± SEM. Differences between experimental groups (B, C, and D) and controls (A) were tested by analysis of variance, followed by a t test. Differences between means were considered significant if p < 0.05. Each group was compared with the control group by a method of multiple comparisons of the means (Dunnett test). Groups were also compared two by two using the Fisher exact test.

Serum amino acid analysis. Serum amino acids were measured using standard chromatography methods. For this purpose, 1 mL of blood was taken before sacrifice at postoperative d 32, under ether anesthesia.

RESULTS

Ten rats died after surgery (two from the control group A, three from the Chol/BCAA group B, two from the Chol/HP group C, and three from the Chol/Con group D. They were replaced by the corresponding spare animals, so that 12 rats were available for study in each group. Table 1 reports the mean daily food and energy intakes in the four groups, as well as the total food and energy intake during the entire 32 postoperative days for each groups. Chol/Con (D) animals ingested a significantly lesser amount of food than the other groups. Compared with the other groups, the decreased food intake in Chol/Con (D) animals did occur from postoperative d 18, whereas no difference in food intake was observed during the early postoperative period from d 1 to 17.

Table 1 Food intake
Full size table

Table 2 reports the course of body weight gain and the final liver weight of the four groups. Final weight of groups B, C, and D animals were 85, 81, and 64% of controls (group A), respectively. No significant difference was observed in weight gain between Chol/HP and Chol/BCAA animals.

Table 2 Body weight evolution and final liver weight
Full size table

Table 3 details urinary nitrogen and nitrogen balance measurements performed at postoperative d 30 in the four groups. In the Chol/BCAA group B, the nitrogen balance was similar to controls (A), whereas it was reduced to 63% of controls in group C and 44% in group D.

Table 3 Total urinary nitrogen and nitrogen balance
Full size table

Table 4 details the results of the carcass analysis in the four groups. Although weight gain was similar in Chol/BCAA group B than in Chol/HP group C (Table 2), the H2O composition of the latter group was significantly higher (71.9%), as in Chol/Con group D(71.4%), whereas for the Chol/BCAA group B, the carcass water content was 62.3%, similar to control group A (61.9%). Body protein content was significantly higher in groups A and B than in Chol/Con animals of group D, and intermediate in Chol/HP group C rats. Lipid content was significantly lower in Chol/HP rats. In the Chol/BCAA rats, ash content was 84% of control group A (NS), but was decreased to 50% in Chol/HP group C and 23% in Chol/Con group D.

Table 4 Carcass analysis
Full size table

Weight and protein composition of the soleus and extensor digitorum is shown in Table 5. The muscle protein content remained unchanged in all four groups. The absolute weight of the extensor digitorum longus muscle was significantly lower in the three cholestatic groups (78% for B, 67% for C, and 63% for D) compared with control group A, although it was significantly higher in the Chol/BCAA group B than in both Chol/HP group C and Chol/Con group D. The soleus mass was similar in BCAA group B than in control group A, but decreased to 80 and 61% in groups C and D, respectively.

Table 5 Soleus and extensor digitorum longus (EDL) weight and protein concentration
Full size table

Serum BCAA (valine, leucine, and isoleucine) were markedly increased in the Chol/BCAA group B compared with the values recorded in the three other groups(Table 6). The increase compared with Chol/Con group D reached 275% for valine, 176% for leucine, and 158% for isoleucine. Compared with control group A, the respective increases were 176, 120 and 147%. Compared with Chol/HP group C, the respective increases were 222, 188, and 208%. Serum BCAA were significantly lower in Chol/Con group D and Chol/HP group C than in control group A (Table 4). The methionine level was significantly higher in Chol/HP group C, and similar to controls in Chol/BCAA group B and Chol/Con group D. The tyrosine level was higher in all three cholestatic groups compared with controls, but the Chol/BCAA group B was closer to controls as compared with Chol/HP group C and Chol/Con group D. Similar observations were made for phenylalanine.

Table 6 Serum amino acids
Full size table

DISCUSSION

Very few controlled studies have so far been conducted in animals to assess the potential for nutritional variation in the composition of the diet to improve malnutrition and growth in childhood cholestasis(2). Although recent clinical studies encourage the use of BCAA supplements in these infants, data from animal studies are still necessary to extend the experimental basis in this field of nutrition(2, 10). The experimental animal model of extrahepatic biliary atresia used in this study was previously shown to reproduce the pathogenic course of extrahepatic biliary atresia in humans, including marked cholestasis, pruritus, absence of bile in the bowel, and progression to a biliary cirrhosis within a month after surgery(6). Liver weight was significantly increased in cholestatic animals, reflecting cholestasis and an inflammatory process. The metabolic organization of the cirrhotic nodules was shown to be similar to that occurring in livers from patients with extrahepatic biliary atresia(11, 12). In one experimental study, the authors found that the addition of 2.5% BCAA to the diet was able to improve nitrogen balance and plasma proteins in an animal model of CCl4 cirrhosis(13). Such a model does not reflect adequately human biliary diseases, in particular extrahepatic biliary atresia, the most common cause of chronic liver disease in childhood. In CCl4-induced cirrhosis, the metabolic zonation pattern is the reverse of that found in human biliary atresia and in experimental biliary atresia cirrhosis(6, 11, 12). The bile flow is not interrupted, and the liver damage is a consequence of discontinuously repeated injuries to the perivenular zone instead of a continuous pathologic process(6, 11, 12, 14).

The data presented in the present study clearly indicate that a 50% increase in nitrogen intake given as oral BCAA is able to improve significantly the nutritional status of cholestatic rats. After supplementation with BCAA, the first effect was a significant increase in serum BCAA concentrations, compared with the cholestatic animals fed a normal diet or a diet enriched in casein. This effect may at least partially correct the plasma amino acid imbalance of cirrhosis characterized by an increase in circulating aromatic amino acids and a decrease in BCAA. Such imbalances have been described in humans with biliary atresia and other types of cirrhosis(15, 16). Furthermore, the higher serum BCAA concentration could improve muscle protein synthesis, and, serving as substrates for gluconeogenesis, avoid excessive endogenous protein catabolism(3–5). Casein supplementation may, to the contrary, be deleterious, because the group supplemented with casein (Con/HP group C) exhibited a significant increase in serum aromatic amino acid concentrations (tyrosine, phenylalanine, and methionine), an effect which was not observed in the BCAA-supplemented group. Although our aim was not to investigate the effects of the diet on encephalopathy, these changes in aromatic amino acids are worth mentioning because these amino acids are known to play an important role in the development of acute and chronic encephalopathies in chronic liver diseases(4). The fact that no significant increase of aromatic amino acid levels were observed in the Chol/BCAA group B rats may be related to either lower intake or better liver function and metabolism. The low BCAA concentration, noted in cirrhotic patients and also observed in our control cholestatic (Chol/Con group D) animals, may be due to a decrease in the release of these amino acids by the liver itself(17). This is thought to play a significant role in muscle wasting and malnutrition. This effect was investigated in our study by carcass analysis, a method considered as the gold standard for the study of body composition in animal studies(7). Body weight was similar in controls and in cholestatic rats supplemented with BCAA or casein, whereas nonsupplemented animals had a significantly lower weight. Although the mean energy intake per unit of body weight was the same in all groups, nonsupplemented cholestatic animals had spontaneously decreased their total daily energy intake from postoperative d 18, which may be the consequence of poorer general condition in these nonsupplemented animals, as is also observed in clinical situations. Although food, energy intake, and body weight were similar between rats supplemented with BCAA or with casein, body composition of the casein group revealed marked alterations including higher water content, lower fat, lower muscle mass, and surprisingly lower ash content. This last finding may reflect a decrease in synthesis rate of bone protein matrix with secondary defects in mineral retention. Higher water content was found in carcass analysis, but not in the skeletal muscles, reflecting probably ascites and interstitial edema. To the contrary, the body composition of cholestatic animals with BCAA reached similar values to those of noncholestatic controls, confirming a biologic effect of the increased serum BCAA concentration discussed above. The higher body water content which is well known in cirrhosis was thus corrected by the 50% supplementation in BCAA and may itself play a role in leucine metabolism(18).

This beneficial effect of BCAA was further confirmed by the 22 and 56% increase in the respective mass of both extensor digitorum longus and soleus muscles. Their weights were similar to that of the sham-operated controls, whereas both casein-supplemented and nonsupplemented cholestatic animals showed a significant decrease in muscle mass.

Finally, a nitrogen retention effect in response to BCAA supplementation could also be evidenced by the study of nitrogen balances. Nitrogen balance was similar in BCAA animals and in controls, but was markedly decreased in casein-supplemented and nonsupplemented cholestatic rats. This experimental study demonstrates that an increased oral intake in nitrogen in the form of BCAA markedly improves nitrogen retention, serum amino acid profile, dry weight gain, and body composition in an experimental model of extra hepatitic biliary atresia. When nitrogen supplements are given as casein, the benefit was milder, improving only bone mineral content and absolute skeletal muscle weight, but not nitrogen balance nor the excessive water content compared with nonsupplemented animals. The beneficial effect of BCAA is therefore not due solely to the increased caloric and nitrogen content of food, but to the biologic quality of the BCAA. These observations should lead to the consideration of human clinical studies to determine whether infants with cholestatic diseases may benefit from a diet supplemented in free BCAA and/or proteins with high BCAA content.