Secondary Logo

Journal Logo


Role of Nutrients and Bacterial Colonization in the Development of Intestinal Host Defense

Walker, W. Allan

Author Information
Journal of Pediatric Gastroenterology and Nutrition: March 2000 - Volume 30 - Issue - p S2-S7
  • Free

As the field of mucosal immunology has developed and mechanisms of intestinal host defense have been better defined during the past 2 decades, we have gained a healthy appreciation for the role of indigenous bacterial flora in the process. Despite strong evidence that the human newborn intestine has the capacity to mount a mucosal immune response (1), the neonate must first enter the extrauterine environment and become exposed to colonizing bacteria before the mucosal immune effector response is operational. Therefore, we must have a better understanding of the environmental factors that affect colonization of the newborn intestine by indigenous flora (2). Several clinical studies have suggested that breast-fed infants have a different luminal bacterial content than infants fed with formula (3). This has led to a resurgence of the use of functional foods (pre-, pro-, and synbiotics) to effect bacterial colonization. This supplement to the Journal of Pediatric Gastroenterology and Nutrition has been organized to examine the role of nutrition and probiotics in pediatric health and disease.

Before considering the role of nutrients and bacteria in intestinal mucosal defense, we must define the terms to be used throughout the supplement. A probiotic is a live microbial food supplement that beneficially affects the host animal by improving its intestinal mucosal balance. In contrast, a prebiotic is a nondigestible food ingredient, usually fucosylated oligosaccharides, that affects the host by selectively targeting the growth and/or activity of a limited number of bacteria in the colon and thus has the potential to improve host health. A new term that has entered the literature is synbiotic, which is a mixture of prebiotic and probiotics that beneficially affects the host by improving the survival and multiplication of live microbial dietary supplements in the gastrointestinal tract (3). Studies using these terms will be referred to throughout the supplement.

The intestinal mucosal barrier to foreign antigens, microbial flora, and their toxins is a combination of epithelial and immunologic luminal and mucosal factors working in concert to control penetration of noxious substances and modulate self-limited inflammation and appropriate immunologic reactions. Although a detailed description of these processes is beyond this review, a brief summary of these defenses will be provided (Fig. 1) as detailed in a review by Insoft et al. (1). Luminal factors that nonspecifically control pathologic bacterial colonization and excessive uptake of foreign antigens include gastric acidity to prevent excessive ingestion of harmful bacteria; mucus to trap microbes, antigens, and toxins; digestive enzymes to hydrolyze antigens; and active peristalsis to facilitate the intestinal expulsion of noxious substances (1). Polymeric immunoglobulin A (pIgA) and IgM (pIgM) are released into the intestine in response to antigenic–microbial stimulation as prototypic luminal antibodies that function to agglutinate foreign substances and prevent their attachment and penetration across the mucosal surface (4). The enterocyte is an active participant in mucosal defense. In response to bacteria, toxin, and/or antigen attachment, the enterocyte, through signal transduction pathways, stimulates the genes necessary to upregulate inflammatory cytokine transcription and translation. In addition, basolateral surface molecules that participate in mucosal immune responses—e.g., class II antigens, pIgA, and cytokine receptors—are upregulated. In addition, growth factors used to stimulate proliferation and differentiation of damaged mucosa in response to inflammation are released into the luminal and mesenchyme milieu to facilitate repair of damaged mucosa. Through this response, the enterocyte acts as a sensor of luminal stimuli to alert the mesenchymal lymphoid tissues to respond. It is important to emphasize that colonizing bacteria participate in each step in the expression of these barriers. This participation will be described in detail later by other authors.

FIG. 1.
FIG. 1.:
Barriers to antigenic absorption in the intestine. Antigen entry is limited by nonimmunologic and immunologic mechanisms and by the structure of the epithelium. If potential antigens cross into the enterocyte as they are being degraded and presented as antigens, signals are produced, causing T-cell activation and subsequent additional cytokine production and release. Reproduced, with permission, from a review by Insoft et al. (1).


As stated, initial bacterial colonization of the intestine with indigenous (nonpathologic) flora is an important component in mucosal host defense in the human newborn. Although very little is known at this point about the specific molecular mechanisms by which indigenous flora activate intestinal host defense, this will be an important area of investigation in years to come. A few studies have suggested that probiotics (particularly Lactobacillus GG) may stimulate an increased secretion of pIgA (5) through upregulation of pIgA receptor expression. In addition, clinical evidence exists to suggest that probiotics can strengthen epithelial tight junctions (6). It has also been suggested that specific probiotics have the potential to preferentially stimulate subsets of T helper cells (Th1 or Th2) and thereby modify intestinal inflammatory or allergic responses (7). Studies to specifically define these mechanisms or the specific organism involved are needed to gain a better sense of the specific use of pre-, pro-, and synbiotics in pediatric health and disease.

As stated, our sense of initial bacterial colonization of the newborn gut comes from studies of breast-and bottle-fed infants. Infants who are exclusively breast-fed develop a specific flora by one week after birth that reaches dominance by 1 month. Initial colonizing bacteria at birth come from the mother's birth canal and large intestine (8). Because of prebiotic factors in breast milk, a flora predominant in lactobacilli and bifidobacteria is established by 1 month (9). Several luminal factors in breast-fed infants such as production of lactic acid causing an acid milieu and the presence of oligosaccharides, which compete for bacterial receptors on the mucosal surface and thereby prevent pathologic colonization, and specific nutrients in breast milk (bifidus factor, lactoferrin, casein, and nucleotides) contribute a luminal milieu that favors proliferation of these indigenous bacteria (10). In contrast, in newborns who receive infant formula at birth an intestinal flora develops that is rich in enterobacteria and gram-negative organisms because of a more alkaline milieu and the absence of the prebiotic modulatory factors present in breast milk. The process of bacterial colonization in the formula-fed infant can potentially be modified by supplementation with pre-and probiotics. However, additional clinical studies are needed before such practices can be recommended as routine in infant feeding.

Yet, based on some published clinical studies, there appears to be a need for supplemental pre-and probiotics to prevent and minimize neonatal infections with pathologic bacteria and viruses. A suggested approach to supplemental pre-and probiotics will be discussed in later reviews.


During the past decade, microbiologists and cell biologists have advanced our knowledge of how micro-organisms, particularly pathogens, use epithelial cells, particularly enterocytes, to gain access to human hosts (11). This interaction between micro-organisms and epithelial cells has been termed bacterial–epithelial cross-talk. To colonize the gastrointestinal tract, micro-organisms must first adhere to microvillus membrane (MVM) glycoconjugates (glycoproteins or lipids lodged in the MVM) through adhesions on the bacterial cell wall (Fig. 2). Frequently, the glycoconjugate in the MVM is a receptor for a physiologic ligand—e.g., growth factors or cytokines—and has an intracellular domain that is linked to a signal transduction pathway that can activate genes through transcriptions factors (12). By co-opting the physiologic receptor, the micro-organism can use epithelial cellular pathways to facilitate its translocation into the host's interstitium or intravascular space. This area of research has yielded important new concepts on the pathogenesis of pathologic bacterial gastroenteritis and sepsis.

FIG. 2.
FIG. 2.:
For bacteria to colonize the human intestine, they must first attach to glycoconjugates (terminal sugars on oligosaccharide side chains on microvillus membrane, proteins, and lipids). Specific bacteria adhere to specific sugars in a lectin-like manner. This illustration depicts the expression of glycoconjugates on the microvillus surface and the putative attachment of specific bacteria to these molecules on that surface.

The gastrointestinal tract is an important site for microbe–host interaction and represents the primary site for gram-negative organisms to gain entry to the host's internal milieu. As stated, the first step in bacterial colonization is adherence to microvillus glycoconjugates (terminal sugars on oligosaccharides, the side chains of MVM proteins, and lipids). Different organisms have a different affinity for terminal sugars. For example, Escherichia coli species adhere to the terminal sugar mannose. Therefore, the carbohydrate composition and glycosylation of MVM proteins and ligands may be a determinant in the type of flora that colonizes the gut. Once adherence occurs, micro-organisms can proliferate to formally colonize the gut surface. With colonization comes bacterial–epithelial cross-talk, in which colonizing bacteria use the epithelial cell machinery to facilitate translocation (Fig. 3). For example, Salmonella typhimurium adheres to the epithelial growth factor receptor and uses its signal-transduction pathway to release molecules that open tight junctions between enterocytes, allowing enhanced paracellular translocation (12). E. coli can attach to a physiologic receptor or insert its own receptor to modify the epithelial surface actin composition and thereby facilitate engulfment by the enterocyte and then enhance translocation by endocytosis (13). These are two examples of pathologic organisms that use enterocyte functions to enhance entry into the host milieu.

FIG. 3.
FIG. 3.:
After adherence of pathologic organisms to glycoconjugates on the microvillus membrane (MVM), bacteria can colonize the gut. Colonization and interaction with physiologic receptors in the MVM through bacterial–epithelial cross-talk result in co-opting of intracellular signal pathways to facilitate the pathologic micro-organisms' translocation across or between epithelial cells, resulting in inflammation and tissue destruction. This illustration depicts these events.

Although most basic research in bacterial epithelial cross-talk involves pathologic organisms, we now know that indigenous (probiotic-type) organisms can also communicate with the intestinal surface to enhance intestinal host defenses against bacteria-induced clinical disease. The precise cellular mechanisms for these responses in most cases have not been identified, however, and represent a fruitful area for future research. Probiotics can protect the host from pathologic colonization by many different events occurring within the intestine. For example, probiotic organisms also have preferential carbohydrate receptors. If their preference for a specific sugar attachment (e.g., mannose for Bifidobacterium profringins is the same as a pathologic organism such as mannose for E. coli) they can competitively inhibit E. coli colonization (Fig. 4) (1). As stated, their attachment to the intestinal epithelium can actively stimulate epithelial and lymphocyte functions (3) enhancing the protective capacity of mucosal defenses. The human neonate at birth has the potential to mount a mucosal immune response but requires the luminal bacterial stimulation of the initial bacterial colonization before active defenses are operational (1). What influences the appropriate colonization of the neonatal gut to produce adequate intestinal host defense is an important area for basic and clinical investigation in the future. In addition, more studies on the actual cellular mechanism of probiotic–indigenous flora–epithelial cross-talk will be necessary as we consider probiotics as an additive for infant formula and toddlers' diet.

FIG. 4.
FIG. 4.:
Illustration of the proposed mechanisms of toxin-induced gastrointestinal diseases. Exotoxin interacts with its microvillus membrane receptor and through activation of signal pathways stimulates increased chloride secretion leading to toxigenic diarrhea. Endotoxin interacts with the enterocyte by an unknown mechanism and stimulates upregulation of inflammatory cytokines leading to cell destruction.

Finally, we now know that not only do bacteria interact with the enterocyte to produce host-induced physiologic–pathologic responses, but components of bacteria (endo-and exotoxins) also communicate with intestinal epithelium. For example, the lipopolysaccharide component of gram-negative bacterial cell walls (endotoxin) can interact with enterocytes to activate the transcription and translation of inflammatory cytokines (tumor necrosis factor) including chemokines (interleukin 8), which participate in the recruitment of neutrophils, leading to inflammation (14). The endotoxin attaches to enterocytes by an unknown process and activates signals leading to the production of the transcription factor NF, which enhances the transcription of mRNAs for these cytokines. A better understanding of this process may lead to ways of controlling gram-negative endotoxin shock in patients in hospitals.

In like manner, secreted bacterial peptides (exotoxins) can have a profound effect on gastrointestinal secretory and inflammatory functions (15). These soluble secreted bacterial by-products interact with epithelial receptors and alter cellular metabolic, secretory, and protective function. Many bacterial diarrheas, heretofore thought to be strictly invasive, are now known to have an exotoxigenic component to their pathogenesis.


As we develop a better understanding of bacterial–intestinal interactions and the pathogenesis of disease, we recognize that certain infectious gastrointestinal diseases occur with increased frequency and severity in early infancy. This observation suggests that immaturities in bacterial–intestinal interactions may contribute to the pathogens of disease in this age group. This hypothesis will be discussed in detail with regard to necrotizing enterocolitis, a disease of premature infants, by Caplan and Jilling later in this supplement. For purposes of this introduction, several examples will be mentioned here.

As stated earlier, the nature of glycosylation in enterocytes, particularly glycosyltransferase activity, determines glycoconjugates' availability for bacterial colonization (Fig. 2). Interaction of specific types of bacteria with their sugar receptor determines colonization and may influence translocation (Fig. 3). A number of studies reported during the past few years as reviewed (15) suggest that glycosylation and glycosyltransferase activity are developmentally regulated and may favor pathologic colonization over indigenous flora colonization in the neonatal intestine (15). Details of this observation are discussed by Dai et al. in another review in this supplement.

As stated, the lipopolysaccharide component of gram-negative organisms (endotoxin) can stimulate the enterocyte to increase its production of inflammatory cytokines (tumor necrosis factor, interleukin-8, and others). This enterocyte response to endotoxin stimulation is considered an appropriate self-limited inflammatory defense against the invasion of gram-negative bacteria. However, when the response has been studied in patients with necrotizing enterocolitis or in the immature human neonatal intestine, immature enterocytes appear to respond inappropriately (excessively) to endotoxin stimulation with elevated and sustained levels of platelet activity factor and interleukin-8 (16). This inappropriate response to microbial endotoxin may help explain the pathologic origin of necrotizing enterocolitis. Again, Caplan and Jilling cover this subject in another review in this supplement.

Finally, several investigators in human and animal studies have suggested that certain toxigenic diarrheas occurring in greater frequency and severity in young infants may do so because of an immature response of the neonatal intestine to exotoxin stimulation (15). For example, Drs. Cohen and Giannella have reported that neonatal rodents and humans have more E. coli heat-stable toxin receptors on the intestinal surface during the neonatal period than at later ages (17). The increased number of receptors undoubtedly accounts for the enhanced diarrheal response to this toxin in infancy. In studies from our laboratory (15), we have shown that E. coli labile toxin and cholera toxin produce a greater secretory response to a specific dose of toxin in animals before weaning than after (18). We subsequently showed that this accentuated response is due to an enhanced signal transduction activation in the immature intestine (15). This inappropriate response to exotoxin in infancy probably explains the increased incidence of these toxigenic diarrheas in this age group. These examples of immature responses to bacteria and their toxins in the newborn gut help explain the increased incidence of certain bacterial gastrointestinal diseases at this time of life.


With changing priorities in the practice of medicine today, a greater interest has developed in the prevention of chronic, debilitating and life-threatening diseases rather than simply improving therapy for these conditions after they have developed. Because evidence exists that diet can prevent disease, a major interest has developed in functional foods (19–22). The use of functional foods is of major interest among large companies that produce infant formulas and among other infant food manufacturers. This supplement, which was part of a Sino-American symposium in Shanghai held on March 24–25, 1999, addresses aspects of these issues.

We look to breast milk as the gold standard for neonatal and infant nutrition. As mentioned in the introduction, infants exclusively breast-fed show development of a specific type of intestinal flora at 1 month of age (8,9). Because indigenous flora are important to gastrointestinal function, studies have been conducted to determine the probiotic effects of these flora. The principal flora studied as probiotics are lactobacilli and bifidobacteria. From the myriad of studies reported, the literature must be read carefully, because different strains of probiotic within the same species have different effects on host defense, disease prevention, and treatment. The secret to the success of probiotics as a protection against pathologic organisms is their varied antibacterial effects (23). These “good” micro-organisms can produce antibiotic molecules that directly influence proliferation of pathologic organisms. They can competitively prevent pathologic bacterial colonization by competing for the same sugar glycoconjugate on the epithelial surface (3). Alternatively, by virtue of the intestinal colonization by probiotics, the host's mucosal defenses are strengthened through the enhancement of the secretory antibody response, through a tightening of the mucosal physical barrier to micro-organism translocation, and by a balance in the T helper cell response (5–7). In fact, recent studies have shown that given in conjunction with either oral or systemic vaccines, probiotics can enhance the immunologic response to the vaccine (24). These important properties may be very helpful in the initial orientation of the newborn to the extrauterine environment and may help prevent infectious diseases common to that age group.

In like manner, attention has been paid to specific nutrients in breast milk that may not only provide optimum nutrition for neonates but may also act as protective substances or prebiotics (19–23,25). For example, many recent studies have suggested that nucleotides may be a “conditionally essential nutrient” for the gastrointestinal tract under conditions of stress (e.g., infection, postintestinal resection, trauma). The indigenous de novo pathway for synthesis of nucleotides cannot keep up with increased demand under these conditions, thus requiring exogenous supplement (26). Supplemental nucleotides have also been shown to be helpful in amplifying immune responses and in helping in the repair of damaged gut mucosa (27). Recently, nucleotide supplemental formula in normal human infants was shown to amplify the response to routine vaccinations given in the first 6 months of life (28).

In like manner, casein has been shown to function as a prebiotic (9) and lactoferrin as an anti-inflammatory agent (28). Furthermore, omega 3 fatty acids have also been shown to be effective in the prevention and treatment of inflammation in necrotizing enterocolitis (29). These studies suggest that nutrients may also be effective as anti-inflammatory agents and stimulants for mucosal epithelial and lymphocyte host defense.


In this introduction to the supplement on the use of pre-and probiotics in the health and disease of pediatric patients, I have summarized factors affecting the initial colonization of the neonatal intestine. The term bacterial–epithelial cross-talk was defined, and examples of the enterocyte response to both pathologic and indigenous flora stimulation illustrated. Immaturities in the human neonatal intestinal response to bacteria and their toxins were reviewed in the context of the pathogenesis of age-specific, bacterial gastrointestinal infectious diseases. Finally, the importance of pre-and probiotics as measures to strengthen the neonate's intestinal host defenses in the prevention and treatment of specific age-related disease were considered.


1. Insoft RM, Sanderson IR, Walker WA. Development of immune function in the intestine and its role in neonatal diseases. Pediatr Clin North Am 1996; 43:551–71.
2. Hopper LV, Bry L, Falk PG, et al. Host-microbial symbiosis in the mammalian intestine: Exploring an internal ecosystem. Bioessays 1998; 20:336–43.
3. Dai D, Walker WA. Protective nutrients and bacterial colonization in the immature human gut. Adv Pediatr 1999; 46:353–82.
4. Sanderson IR, Walker WA. Uptake and transport of macromolecules by the intestine: Possible role in clinical disorders (an update). Gastroenterology 1993; 104:916–25.
5. Majamas H, Isolauri E. Probiotics: A novel approach in the management of food allergy. J Allergy Clin Immunol 1997; 99:178–85.
6. Isolauri E, Majamas H, Hrvola T, et al. Lactobacillus casei strain reverses increased intestinal permeability induced by cow's milk in suckling rats. Gastroenterology 1993; 105:1643–50.
7. Sutas Y, Hurme M, Isolauri E. Down-regulation of anti-CD3 antibody-induced IL-4 production by bovine caseins hydrolyzed with Lactobacillus GG-derived enzymes. Scand J Immunol 1996; 43:687–9.
8. Langhendries JP, Detry J, Van Hees J, et al. Effect of fermented infant formula containing viable bifidobacteria on the fecal flora composition and pH of healthy full-term infants. J Pediatr Gastroenterol Nutr 1995; 21:177–81.
9. Balmer SE, Wharton BA. Diet and fecal flora of newborn: Breast milk and infant formula. Arch Dis Child 1989; 64:1672–7.
10. Bernt KM, Walker WA. Human milk as a carrier of biochemical messages. Acta Paediatr Scand 1999; 88:27–41.
11. Wick MJ, Madara JL, Fields BN, et al. Molecular crosstalk between epithelial cells and pathologic microorganisms. Cell 1991; 67:651–9.
12. Galan JE, Pace J, Hayman MJ. Salmonella typhimurium enter epithelial cells via the epidermal growth factor receptor. Nature 1992; 94:588–9.
13. Krogfelt KA. Bacterial adhesion: Genetics, biogenesis and role in pathogenesis of fimbrial adhesions of Escherichia coli. Rev Infect Dis 1991; 13:721–35.
14. Kagnoff MF, Eckmann L. Epithelial cells as sensors for microbial infection. J Clin Invest 1997; 100:551–5.
15. Chu SW, Walker WA. Bacterial toxin interaction with the developing intestine: A possible explanation for toxigenic diarrhea of infancy. Gastroenterology 1993; 104:916–25.
16. Caplan M, MacKendrick W. Inflammatory mediators and intestinal injury. Clin Perinatol 1994; 21:235–43.
17. Laney DW, Mann E, Dellon SC, et al. Novel sites for expression of Escherichia coli heat-stable enterotoxin receptor in the developing rat. Am J Physiol 1992; 263:G826–G1.
18. Chu SW, Ely IG, Walker WA. Age and cortisone alter host responsiveness to cholera toxin in the developing gut. Am J Physiol 1989; 256:G220–G6.
19. Simopoulos AP. The role of fatty acids in gene expression: Health implications. Ann Nutr Metab 1996; 40:303–11.
20. Lin B-F, Huang C-C, Chiang B-L. Dietary fat influences Ia antigen expression, cytokines and prostaglandin E2 production of immune cells in autoimmune-prone NZB X NZW F1 mice. Br J Nutr 1996; 75:711–22.
21. Solis–Pereyra B, Aattouri N, Lemonnier D. Role of food in the stimulation of cytokine production. Am J Clin Nutr 1997; 66:521S–5S.
22. Grimble RF. Interaction between nutrients, pro-inflammatory cytokines and inflammation. Clin Sci 1996; 91:121–30.
23. Silva M, Jacobus NV, Deneke C. Antimicrobial substance from a human Lactobacillus strain. Antimicrob Agents Chemother 1987; 31:1231–3.
24. Bengmark S. Ecological control of the gastrointestinal tract: The role of probiotic flora. Gut 1998; 42:2–7.
25. Cynober L. Can arginine and ornithine support gut functions? Gut 1994;:S42–5.
26. Carver JD, Walker WA. The role of nucleotides in human nutrition. Nutr Biochem 1995; 6:58–72.
27. Pickering L, Granolt DM, Erickson JR, et al. Modulation of the immune system by human milk and infant formula containing nucleotides. Pediatrics 1998; 101:242–9.
28. Goldman AS. The immune system of human milk: Antimicrobial, anti-inflammatory and immunomodulating properties. Pediatr Infect Dis J 1993; 12:664–71.
29. Akisu M, Baka M, Coker I. Effect of dietary n-3 fatty acids on hypoxia-induced necrotizing enterocolitis in young mice. Biol Neonate 1998; 74:31–8.
© 2000 Lippincott Williams & Wilkins, Inc.