Bacterial translocation (BT) has been reported since 1950 and was defined by Berg and Garlington in 1979 (1) as the passage of viable bacteria from the gastrointestinal tract to the mesenteric lymph nodes (MLNs) and other systemic organs. Bacterial translocation occurs directly at the site of injured tissues after surgery and trauma. (2) The mechanisms of bacterial invasion across the intact epithelial layers have not yet been elucidated (3–5). The nature of the intestinal mucosal barrier of the host influences the incidence of BT (6). Because the bacteria in the intestine are the exclusive source of BT (1), the numbers and species of the intestinal microflora are probably an important factor in inducing BT (7).
Infants and neonatal animals appear to be vulnerable to intestinal bacterial invasion (8) because of the immaturity of their gut mucosal barrier, their systemic immune defense system (9–12), and possibly, aberrations in the nature of their gastrointestinal microflora (13–15). Steinwender et al noted that BT to MLNs occurred spontaneously in healthy mother-reared rat pups (16), although it was not detected in adult rats. Incomplete colonization dominated by aerobic bacteria appears more often in preterm or formula-fed infants than in full-term infants (13,17–20). In neonates, especially preterm neonates, abnormal intestinal bacterial flora may in fact promote serious conditions such as necrotizing enterocolitis (21–24).
In the antiseptic environment of the newborn intensive care unit, the incidence of infants with pure cultures of aerobic gram-negative bacteria in the stool was higher in newborn intensive care unit infants than in normal full-term infants (25). This observation suggests that newborn intensive care unit infants may be at increased risk for BT (21). Formula fed infants in intensive care may also have an increased risk for morbidity resulting from septicemia than those fed breast milk (26). It has been reported that breast milk contains bioactive substances that fortify the epithelial barrier function, defend against necrotizing enterocolitis (26), and facilitate the establishment of an anaerobic bacterial flora in the intestine (15). Orrhage and Nord showed that formula fed infants had higher numbers and isolation frequencies of enterococci and clostridia than breast-fed children, and that Enterobacteria were also increased (27).
Additionally, it has been reported that formula feeding in both human and animal neonates enhanced BT to systemic organs (16,28). There are few studies which have evaluated the protective role of breast milk against systemic BT in neonatal animals, which have simultaneously assessed the association between the colonization patterns of intestinal microflora and invading bacterial species.
Our aim in this study was to confirm the protective role of breast milk against intestinal BT by comparing breast-fed neonatal rats with formula-fed rats during almost the entire suckling period. We performed artificial rearing during the suckling period with a refined formula that was made to resemble rat milk (29). We compared in detail the development of the intestinal bacterial colonization during suckling and the bacterial species cultured from the MLNs and systemic organs between breast-fed and formula-fed rat pups.
MATERIALS AND METHODS
The experimental protocol was approved by our institute's Committee for Research on Experimental Animals and was conducted in accordance with the NRC Guide for the Care and Use of Laboratory Animals (NRC, 1985).
Pregnant time-dated Sprague-Dawley rats were purchased from Japan SLC (Shizuoka, Japan). All animals were kept individually under environmentally controlled temperature (25 ± 2°C), humidity (55 ± 2% relative humidity), and light (12-hour light–dark cycle) conditions. They had free access to water and chow (CA-1; Japan Clea, Tokyo, Japan) and were allowed to deliver spontaneously. Three days after birth, the rat pups were weighed and assigned to three groups in random order to eliminate the influence of interlitter variation. In the first group, litters of 10 pups were reared by one of their mothers throughout the study (MR). In the second group, an intragastric cannula was esophageally implanted in each pup 2, 3, or 7 days after birth using the method of Hall (30), and the rats were then artificially reared (AR). In the third group, an intragastric cannulation was performed esophageally in the rat pups on the same days as for the AR group, and the cannula was removed immediately. Then, the pups were returned to one of their mothers 3 hours after cannulation and reared with breast milk throughout the study (Sham).
The artificial rearing method used in this study involved implanting a cannula in the stomach. The milk-pumping side of the cannula tube passed from inside to outside of the walls of the stomach and abdomen by a guide of piano wire, and the inner gastric side of the cannula was stopped by a rim with a thin plastic cover to avoid coming out from the stomach wall. The use of a sham operated control group allowed us to evaluate the impact of the surgical procedure on intestinal microflora and BT separate from the impact of formula feeding. Suckling rats begin to nibble their mother's food at approximately 15 days of age. The dams and pups in the MR and Sham groups were placed into a special cage that we designed from 10 days of age, so that the pups had no access to the mother's food.
Artificial rearing of rats
The milk formula for artificial rearing of rat pups was prepared under aseptic conditions. Its composition closely resembled that of rat mother's milk (29). The composition of the formula is shown in Table 1. The artificial milk was divided into 50-mL sterilized polypropylene bottles and stored frozen at −40°C. The osmotic pressure and pH of the formula was 343 and 6.48, respectively. There were no Enterobacteriaceae in the milk. The frozen formula was sterilized by gamma-ray irradiation (30 kGy).
The volume of formula administered was adjusted daily according to the growth of the AR pups in comparison with MR pups and was increased from 2.7 mL per day on day 4 to 8.2 mL per day on day 21. The syringe on the pump was placed in a refrigerator (5°C) to prevent bacterial growth in the milk formula inside the tubes. If the cannula fell off during artificial rearing, the rat was killed.
Fecal samples were harvested by stroking the anal–genital area of each rat with a cotton swab in the morning. The content of the cecum was obtained after organ removal for the BT examination. The samples were immediately placed in preweighed sterile tubes filled with carbon dioxide gas and weighed. Then the tubes were placed in an anaerobic chamber and maintained in an anaerobic atmosphere of 10% carbon dioxide, 10% hydrogen, and 80% nitrogen. The specimens were homogenized with a prereduced diluent buffer (31). After homogenization, serial 10-fold dilutions of the homogenates were prepared from each sample. One hundred-microliter aliquots of each diluent were applied to prereduced BL agar (Eiken Chemical Co. Ltd., Tokyo, Japan) as a nonselective medium for the anaerobes, LBS agar (BBL; Becton Dickinson and Company, Cockeysville, MD) for Lactobacillus, and modified NBGT agar (32) for Bacteroides. The plates were incubated in anaerobic conditions at 37°C for 48 to 72 hours. These dilutions were also plated onto aerobic agar plates and incubated aerobically at 37°C for 24 to 48 hours. The aerobic media were as follows: TS agar (BBL) as a nonselective medium for the aerobes, TATAC agar (33) (Bacto-Pepton; Difco Laboratories, Detroit, MI) 15g; tryptone (DIFCO) 10 g; yeast extract (DIFCO) 10 g; sucrose (WAKO Pure Chemical Laboratories, Osaka, Japan) 1 g; esculin (WAKO) 1 g; Bacto Agar (DIFCO) 16 g; horse serum (GIBCO BRL, Life Technologies, Rockville, MD) 50 mL; sodium azide (WAKO) 99 mg; sodium glutamate (WAKO) 6.6 g; acridine orange (WAKO) 2 mg; triphenyl tetrazolium chloride (MERCK, Darmstadt, Germany) 20 mg; crystal violet (Nacalai Tesque Inc., Kyoto, Japan) 0.0065 mg; thallous sulfate (WAKO) 0.33 g; distilled water 1,000 mL for Enterococcus and Streptococcus; MacConkey agar (Eiken Chemical Co., Ltd., Tokyo, Japan) for Enterobacteriaceae; and Staphylococcus medium 110 agar (Eiken Chemical Co., Ltd., Tokyo, Japan) for Staphylococcus. After incubation, identification of the bacteria was performed by observation colony shape, cellular morphologic features, gram staining, and a regrowth test (aerobic) of the colonies that grew on the BL nonselective anaerobic agar according to the method of Mitsuoka (15). Colonies of anaerobic rods or cocci growing on on BL agar minus those specifically identified by growth onanaerobic selective agar media (LBS and NBGT) were counted as “other” anaerobic gram-positive (or -negative) rods (or cocci) after regrowth test on aerobic TS agar. Colonies of aerobic rods or cocci that grew on TS nonselective aerobic agar minus those identified by selective aerobic agar plates (MacConkey and TATAC) were counted as “other” aerobic gram-positive (or -negative) rods (or cocci).
Analysis for bacterial translocation
The animals were anesthetized with ether and placed in a bath with 75% alcohol. Then, the abdomen was opened and the MLNs, liver, and spleen were harvested for culture of translocated organisms under aseptic conditions. Each tissue was placed in a sterile preweighed glass homogenizer and weighed. It was then homogenized with twofold or fourfold volume of saline containing 0.2% peptone. Then, depending on the amount of available material, either two or three plates (BL agar) were inoculated with the resulting homogenate (100 μL/plate) and cultured anaerobically for 72 hours at 37°C. After incubation, bacteria were identified from the shape of the colonies and the cellular morphologic features by Gram staining, the results of growth tests performed aerobically, and growth on selective media if needed. The bacterial counts in the MLNs and the organs were expressed as the mean ± standard deviation of the log count.
The intestinal bacterial counts were compared between two groups using the Student t test and between three groups using a one-way analysis of variance, then by applying the Fisher post hoc test (P < 0.05). The incidence of BT was examined for significance using the chi-square test and then by applying the Fisher exact test (P < 0.05). All statistical analyses were performed using STATVIEW 4.1 (Abacus Concepts, Inc., Berkeley, CA).
Flora in breast-fed rats
Table 2 shows bacterial counts and the incidence of viable organisms in feces of the three mothers and their neonatal pups (10 days of age) which were housed separately by family. The total number of bacteria found in mothers and pups was not significantly different, however, the ratio of each species was significantly different. Anaerobic bacteria were the dominant bacteria in the mother rats, whereas aerobic bacteria were dominant and anaerobic bacteria were not detected in the neonatal rats. The numbers of Enterobacteriaceae, Enterococcus, and Streptococcus were significantly higher in the neonates than in the mother rats. The numbers of anaerobes and Lactobacillus were lower in the neonates than the mothers.
Bacterial translocation in breast-fed rats
Table 3 shows the incidence and number of positive cultures in the MLNs, liver, spleen, and systemic blood in normal breast-fed rats during the suckling period and after weaning at 5 weeks. BT to MLNs and systemic organs was not found in the rats at 5 weeks of age. In the neonates, however, BT to the MLNs was observed throughout the suckling period, although positive cultures were almost exclusively found in MLN's and not elsewhere. The bacterial counts in MLNs peaked at 7 days of age followed by a gradual decline.
Bacterial species which translocated to MLNs, such as Enterobacteriaceae, Lactobacillus, Staphylococcus, and Enterococcus, were found in the fecal flora. However, the ratios of their numbers differed between the feces and MLNs (Fig. 1). A marked difference between the % of Staphylococcus between fecal and MLN's was seen. being less than 0.1% in the feces, and more than 15% in the MLNs.
Flora in formula-fed rats
There were no significant differences in the weight gain among breast-fed rats reared by their mothers (MR), formula-fed rats reared artificially (AR), and breast-fed Sham rats (Sham) throughout the suckling period (Fig. 2). Table 4 shows the microflora in the feces of the rats in the AR, Sham, and MR groups. The numbers of Enterobacteriaceae and Clostridium were higher in the AR group than in the MR and Sham groups at 14 days of age. No significant differences were observed in the fecal flora between the Sham and MR groups.
Bacterial translocation in the formula-fed group
We compared the time required for the disappearance of BT in breast-fed and formula-fed rat pups after cannulation. Table 5 shows the incidence of BT to the MLNs and liver in AR, Sham, and MR rat pups from 7 to 14 days of age. The occurrence of systemic BT was detected in both the AR and Sham groups at 7 and 10 days of age (4 and 7 days after cannulation), respectively. At 14 days of age (11 days after cannulation), systemic BT was not observed in the Sham group but was still observed in half of the rats in the AR group. The bacteria translocated to the liver were Enterobacteriaceae and Lactobacillus.
Table 6 shows the incidences of BT in further experiments, in which rat pups were cannulated or sham operated at various days of age. In the Sham group, operated at 2 or 3 days of age, systemic BT disappeared within 10 to 11 days of the operation. In the AR rats, systemic BT was detected up to 18 days after the operation. In the Sham group operated on at 7 days of age, systemic BT disappeared within 7 days of the operation. However, systemic BT was still detected 7 days after the operation in the AR rats.
Steinwender et al. (16) reported that BT to MLN spontaneously occurred in neonatal rats at 1 to 5 days of age. They did not examine pups at any other ages. In the present study, we confirmed their results in rat pups through the entire suckling period and also demonstrated that BT to systemic tissues did not occur during almost the entire suckling period in breast-fed rat pups. Furthermore, we have demonstrated that breast-milk feeding but not formula feeding shortened the period at risk for systemic BT in pups that were stressed by surgery. We also demonstrated that the bacterial species translocated to the MLNs or liver were not directly related to the numbers of colonies found in the intestinal lumen.
The present study showed that the bacterial composition of the intestinal microflora of the neonatal rats was less complex in comparison with that in the mother rat. Although Lactobacillus was found in both mothers and pups, there may be a difference between mother and neonatal rats as the Lactobacillus isloated from mothers and pups (Table 2) were often different in shape, colony size, and cellular morphologic features, suggesting that there may be unidentified subspecies differences between the two groups We also confirmed not only higher intestinal colonization levels of Enterobacteriaceae and Clostridium, but also lower levels of anaerobic bacteria in formula-fed rat pups than in breast-fed rat pups. These results are comparable with those in human neonates (13–15,27,34). The different colonization of the fecal flora in breast-fed and formula-fed infants is not yet fully understood but appears to relate to: 1) the type of milk protein, 2) the availability of iron, which is determined by the amount of iron in the infant formula and the presence of high concentrations of lactoferrin in human milk, 3) oligosaccharides in breast milk, which are shown to be “growth factors” for Bifidobacterium, and 4) the buffering activity of pH—breast milk has a lower buffering activity than formula milk (21). The above speculations in humans may also be adopted to rat pups except for number 3, because Bifidobacterium is not dominant in the intestine of rat pups.
In adult rats, we saw little spontaneous BT to MLN's. In neonatal rats, however, spontaneous BT to MLN occurred throughout the suckling period peaking at 7 days of age (Table 3). These findings indicate that neonatal rats may be highly sensitive to spontaneous BT into MLN. A similar spontaneous BT to MLN that peaked at 6 days of age has been demonstrated in the neonatal rabbit (8). As for the vulnerability to spontaneous BT in rat neonates, three factors are considered: 1) bacterial colonization peculiar to the intestine of a neonate, 2) immaturity of the immune functions, especially in the gut, and 3) immaturity of the intestinal epithelial barrier. In a mouse model, Berg and Garlington (1) suggested that high bacterial population levels promote translocation of certain bacteria from the gastrointestinal lumen to the MLNs and that the population levels of particular bacterial species like Escherichia coli that have colonized the gastrointestinal tract may be a critical factor in inducing BT. In addition, we found that the pattern of bacterial colonization in the intestine during the early suckling period was very simple and that aerobic gram-negative bacteria were detected in high numbers (more than 10 9 ), with only small numbers of anaerobic bacteria.
In the present study, the dominant bacteria translocated to the MLNs were Enterobacteriaceae, Lactobacillus, and Staphylococcus. Among these species, the former two were also the dominant species in the intestine, whereas Staphylococcus was a nondominant bacterium in the intestines and feces of suckling rats. These findings indicate that the composition of translocating bacterial species to MLNs does not directly mirror that of the feces in suckling rats. Steffen et al. (7) reported, based on studies in germ-free mice colonized with a single strain of bacterium, that not all bacteria were able to translocate equally well and that aerobic bacteria could translocate more easily inside the tissues than anaerobic bacterial species from the intestine. These facts suggest the presence of a mucosal selectivity against particular species.
As for the recognition of invading bacteria, it is reported that the cell wall component of gram-negative bacteria, such as lipopolysaccharide and O antigens present in lipopolysaccharide, and capsular polysaccharides (35) effectively regulate phagocytosis in the cell-mediated immune defense system. Enterobacteriaceae have those polysaccharides on their cell walls (36), but most strains of Lactobacillus do not. The numbers of both bacteria in the gastrointestinal flora were extremely high in the rat pups in our experiment. Therefore, if the rate of contact with, or of crossing, the intestinal mucosa is high, then the number of invading bacteria may increase beyond the capacity of the defense systems.
It is interesting that Staphylococcus frequently translocated to the MLNs despite its very low levels in the intestinal contents of the suckling rat. Staphylococcus does not have lipopolysaccharide on its cell wall, but some strains have a capsular polysaccharide. Staphylococcus aureus strains that have a capsular polysaccharide are more resistant to bactericidal lipids produced in abscesses than strains that have no capsule polysaccharide (37). Also, Staphylococcus has a strong ability to adhere to fibrinogen and easily forms bacterial clumps that are likely resistant to phagocytosis (38). This may be one reason why Staphylococcus is resistant to phagocytes and is detected at high levels in MLNs. This speculation may support the observation that human newborns and infants up to the age of 1.5 to 2 years are unable to produce antibodies to bacterial capsular polysaccharides (39). As a consequence, children up to the age of 2 years have an increased susceptibility to infection by capsulated bacteria.
In the present study, we demonstrated that the numbers of Clostridium and Enterobacteriaceae were higher in the artificially reared rats than the breast-fed rats. These findings are similar to results reported in human studies (13,14,17,27,34).
Artificial feeding of infant rats by continuous gastric infusion using an infusion pump has been examined by Messer et al. (40) and Lebenthal et al. (41). In both of their experiments, the gastric cannula was implanted surgically, which may cause stress to the animal. In our experiments, to minimize surgical invasion, the cannula, stiffened with a silicone covered piano wire stylet was inserted via the mouth by the method of Hall (30). To determine the effects of the stress resulting from the cannulation on BT in the neonatal rats, the time-dependent occurrence of BT was examined after a sham operation in which the gastrostomy was placed and removed at 7 days of age followed by breast fed and mother reared conditions.. Even 3 hours after sham operation, systemic BT to the liver, spleen, and systemic blood were detected (data were not shown) in all pups in both rearing conditions. The higher pH in the neonatal than adult stomach may explain the tolerance to higher levels of bacterial colonization. Thus, gross invasion by gastric bacteria could occur directly via residue of milk after cannulation.
We demonstrated in the present study that systemic BT rarely occurred in normal suckling rats, but in the AR and Sham rats, systemic BT to the liver was induced after an intragastric cannulation. The dominant bacteria translocated to the liver after the cannulation were Enterobacteriaceae, Lactobacillus, and Staphylococcus, as shown in Table 5, whereas no Clostridium was detected selectively. It is interesting that there were no differences between the experimental groups in the bacterial species translocated to MLNs, except to the liver. Staphylococcus was not detected in the liver at all in breast-fed sham-operated rats although, it was detected in AR rats (Table 5). We have no data to explain why this difference occurred. More studies are required to clarify the mechanisms of selection in translocating bacteria.
We demonstrated in the present study that levels of spontaneous BT in MR rats were detected with a peak at 7 days of age and gradually decreased with aging (Table 3). Urao et al. (42) suggested that spontaneous BT in newborn rabbits was related to the immaturity of gut-associated lymphoid tissue. Other studies have suggested that maturation of systemic T cell function in neonatal rats and germ free mice may also depend on intestinal colonization and bacterial translocation (43,44). Decreased adhesion of neutrophils to endothelial cells and a delayed transendothelial cell migration of neutrophils have both been seen consistently in neonatal animals and humans (45).
The decrease in BT in neonates fed breast milk was were reported by Steinwender et al. (16) in rats and by Go et al. (28) in rabbits. Steinwender et al. showed that systemic BT did not occur because of either a decrease in caloric intake or separation from the mother, but occurred only in response to the feeding of formula (16). They reared pups using a manual feeding method without surgical stress, but the animals might have suffered invasive stress at the time of administration of formula by intermittant gavage several times daily. We have no data to compare the rate of BT in breast milk and formula fed rat pups without gastrostomy. In our experiments, BT did not occur because of a change of mother or physical exposure to other (maternal) rat pups. However, the extent of systemic BT that occurred after the cannulation was reduced by subsequent breast feeding. More experiments on artificial rearing without any invasion should be performed to compare the effects of composition of milk, intestinal colonization, and BT on the immunologic and physiologic development of animals.
In conclusion, in the present study, we confirmed that mother's milk has the ability to inhibit systemic BT in the suckling rat, probably by the promotion of a normal population of intestinal flora and the activation the immune system of the pup, which acts to decrease the survival of systemically invading bacteria. In addition, we confirmed that the composition of the translocated bacteria differed from the intestinal composition of the flora. Staphylococcus may be especially difficult for phagocytes or host antibodies to recognize and destroy.
The present study demonstrated that artificially fed rat pups may be a useful model for examining the barrier function against BT in neonatal rats. The present findings support the concept that elements present in breast milk somehow enhance the intestinal barrier function and protect neonates from BT. Further study is necessary to identify and elucidate the mechanisms involved.
The authors thank Emeritus Prof. Tomotari Mitsuoka, Tokyo University, for helpful advice about the identification of the intestinal microflora of rats and Mrs. Kayo Wake and Mrs. Keiko Kawanishi for skillful technical assistance.
1. Berg RD, Garlington AW. Translocation of certain indigenous bacteria from the gastrointestinal tract to the mesenteric lymph nodes and other organs in a gnotobiotic mouse model. Infect Immunol 1979; 23: 403–11.
2. Brathwaite CEM, Ross FSE, Nagele FR, et al. Bacterial translocation
occurs in humans after traumatic injury: Evidence using immunofluorescence. J Trauma 1993; 34: 586–90.
3. Alexander JW, Boyce ST, Babcock GF, et al. The process of microbial translocation. Ann Surg 1990; 212: 496–512.
4. Lichtman SM. Bacterial translocation
in humans. J Pediatr Gastroenterol Nutr 2001; 33: 1–10.
5. Volkheimer G. Passage of particles through the wall of the gastrointestinal tract. Environ Health Perspect 1974; 9: 215–25.
6. Wang X, Pärsson H, Soltesz V, et al. Bacterial translocation
and intestinal capillary permeability following major liver resection in the rat
. J Surgical Res 1995; 58: 351–8.
7. Steffen EK, Berg RD, Deitch EA. Comparison of translocation rates of various indigenous bacteria from the gastrointestinal tract to the mesenteric lymph node. J Infect Dis 1988; 157: 1032–8.
8. Van Camp JM, Tomaselli V, Coran AG. Bacterial translocation
in the neonate
. Curr Opin Pediatr 1994; 6: 327–33.
9. Dickinson EC, Gorga JC, Garrett M, et al. Immunoglobulin A supplementation abrogates bacterial translocation
and preserves the architecture of the intestinal epithelium. Surgery 1998; 124: 284–90.
10. Roy MJ, Varvayanis M. Development of dome epithelium in gut-associated lymphoid tissues: Association of IgA with M cells. Cell Tissue Res 1987; 248: 645–51.
11. Go LL, Ford HR, Watkins SC, et al. Quantitative and morphologic analysis of bacterial translocation
in neonates. Arch Surg 1994; 129: 1184–90.
12. Bakker JM, Broug-Holub E, Koroes H, et al. Functional immaturity of rat
alveolar macrophages during postnatal development. Immunology 1998; 94: 304–9.
13. Yoshioka H, Iseki K, Fujita K. Development and difference of intestinal flora
in the neonatal period in breast-fed and bottle-fed infants. Pediatrics 1983; 72: 317–21.
14. Mackie RI, Sghir A, Gaskins HR. Developmental microbial ecology of the neonatal gastrointestinal tract. Am J Clin Nutr 1999; 69: 1035s–45s.
15. Mitsuoka T. The human gastrointestinal tract. In: Wood BJB, ed. The Lactic Acid Bacteria. Essex, UK: Elsevier Science Pubishers, 1992; 69–114.
16. Steinwender G, Schimpl G, Sixl B, et al. Effect of early nutritional deprivation and diet on translocation of bacteria from the gastrointestinal tract in the newborn rat
. Pediatr Res 1996; 39: 415–20.
17. Kleessen B, Bunke H, Tovar K, et al. Influence of two infant formulas and human milk on the development of the faecal flora in newborn infants. Acta Pediatr 1995; 84: 1347–56.
18. Sakata H, Yoshioka H, Fujita K. Development of the intestinal flora
in very low birth weight infants compared to normal full-term newborns. Eur J Pediatr 1985; 144: 186–90.
19. Rubaltelli FF, Biadaioli R, Pecile P, et al. Intestinal flora
in breast- and bottle-fed infants. J Perinat Med 1998; 26: 186–91.
20. Blakey JL, Lubitz L, Barnes GL, et al. Development of gut colonization in pre-term neonate
. J Med Microbiol 1982; 15: 519–29.
21. Dai D, Walker WA. Role of bacterial colonization in neonatal necrotizing enterocolitis and its prevention. Acta Paed Sin 1998; 39: 357–65.
22. Caplan MC, MacKendric W. Necrotyzing enterocolitis: A review of pathogenetic mechanisms and implication for prevention. Pediatr Pathol 1993; 13: 357–69.
23. Hoy C, Millar MR, MacKay P, et al. Quantitative changes in faecal microflora preceding necrotizing enterocolitis in premature neonates. Arch Dis Child 1990; 65: 1057–9.
24. Deitch EA. Role of bacterial translocation
in necrotizing enterocolitis. Acta Pediatr 1994; 369: 33–6.
25. Bell MJ, Rudinsky M, Brotherton T, et al. Gastrointestinal microecology in the critically ill neonate
. J Pediatr Surg 1984; 19: 745–51.
26. Duffy LC, Zielenzny MA, Carrion V, et al. Bacterial toxins and enteral feeding of premature infants at risk for necrotizing enterocolitis. Am J Hum Biol 1998; 10: 211–9.
27. Orrhage K, Nord CE. Factors controlling the bacterial colonization of the intestine in breastfed infants. Acta Pediatr Suppl 1999; 430: 47–57.
28. Go LL, Albanese CT, Watkins SC, et al. Breast milk protects the neonate
from bacterial translocation
. J Pediatr Surg 1994; 29: 1059–64.
29. Kanno T, Koyanagi N, Katoku Y, et al. Simplified preparation of a refined milk formula comparable to rat
's milk: Influence of the formula on development of the gut and brain in artificially reared rat
pups. J Pediatr Gastroenterol Nutr 1997; 24: 242–52.
30. Hall WG. Weaning and growth of artificially reared rats. Science 1975; 190: 1313–5.
31. Mitsuoka T, Ohno K, Benno Y, et al. Vergleich des neu entwickelten verfahrens mit dem bisheringen üblichen verfahren zur darmfloraanalyse. Zbl Bakt Hyg 1976; 234: 219–33.
32. Yasui H, Yasutake N, Ohwaki M. Immunogenicity of bacteroides isolated from mice: Relationship between immunogenicity and cell wall antigen. Infect Immunol 1979; 24: 39–46.
33. Mitsuoka T, Sega T, Yamamoto S. Eine verbesserte methodik der qualitativen und quantitativen analyse der darmflora von menschen unt tieren. Zentralbl Bakteriol Parasitenkunde Infektionskrankheiten Hyg 1964; 195: 455–69.
34. Lundequist B, Nord CE, Winberg J. The composition of the faecal microflora in breastfed and bottlefed infants from birth to eight weeks. Acta Pediatr Scand 1985; 74: 45–51.
35. Kim JO, Romero-Steiner S, Sorensen UB, et al. Relationship between cell surface carbohydrates and intrastrain variation on opsonophagocytosis of Streptococcus pneumoniae
. Infect Immunol 1999; 67: 2327–33.
36. Whitfiels C, Roberts IS. Structure, assembly and regulation of expression of capsules in Escherichia coli
. Mol Microbiol 1999; 31: 1307–19.
37. Mortensen JE, Kapral FA. Effect of capsulation on the resistance of Staphylococcus aureus
to the bactericidal lipids produced in abscesses. J Med Microbiol 1992; 36: 337–40.
38. Dominiecki ME, Weiss J. Antibacterial action of extracellular mammalian group: A phospholipase A2 against grossly clumped Staphylococcus aureus
. Infect Immun 1999; 67: 2299–305.
39. Rijkers GT, Sanders EAM, Breukels MA, et al. Infant B cell responses to polysaccharide determinants. Vaccine 1998; 16: 1396–400.
40. Messer MTE, Galofre A, Dallman T, et al. Artificial feeding of infant rats by continuous gastric infusion. J Nutr 1969; 98: 404–10.
41. Lebenthal E, Sunshine P, Kretchmer N. Effect of carbohydrate and corticosteroids on activity of α-glucosidases in intestine of the infant rat
. J Clin Invest 1972; 51: 1244–50.
42. Urao M, Teitelbaum DH, Drongowski RA, et al. The association of gut-associated lymphoid tissue and bacterial translocation
in the newborn rabbit. J Pediatr Surg 1996; 31: 1482–7.
43. Helgeland L, Brandtzaeg P, Rolstad B, et al. Sequential development of intraperitoneal gamma delta and alpha beta T lymphocytes expressing CD8 alpha beta in neonatal rat
intestine: Requirement for the thymus. Immunology 1997; 92: 447–56.
44. Kawaguchi-Miyashita M, Nanno M, Shimada S, et al. A step-wise expansion of intestinal intraperitoneal T lymphocytes in association with microbial colonization is defined by sensitivity to cyclosporin A. Immunology 1997; 91: 628–34.
45. Lorant DE, Li W, Tabatabaei N, et al. P-selectin expression by endothelial cells is decreased in neonatal rats and human premature infants. Blood 1999; 94: 600–9.