Main

The gastrointestinal (GI) tract undergoes substantial changes during the early postnatal period. Profound growth, morphologic changes, and functional maturation of the small intestine are observed during this developmental period. The introduction of enteral nutrition initiates significant changes in mucosal structure and function required for the utilization of milk. Most of these changes are concluded by the weaning phase when the transition from milk to solid food occurs. Physiologic changes in digestive and absorptive processes are well described (1), but the understanding of the mechanisms by which intestinal growth and epithelial turnover are regulated remains fragmentary. However, it is clear that biologically active peptides such as hormones and growth factors are essential parts of this regulatory system. Two growth-stimulating peptides that are thought to play an important role in this process are EGF and TGF-α (27). Both EGF and TGF-α share the same EGF receptor (8, 9), which is known to be present in developing small intestine (1012).

It is well established that breast milk contains large numbers of biologically active substances (13, 14). Because EGF is detected in several mammalian milks, questions have been raised about its functional significance in milk. Previous studies from our laboratory demonstrated the presence of EGF in the developing gut of suckling rats and showed that intestinal EGF levels are predominantly related to the intake of milk-borne EGF (15, 16). In contrast with EGF, TGF-α is absent in rat milk, and intestinal TGF-α content results from endogenous production of this peptide by the small intestine and from pancreatic secretion (17). In the present study, we tested the hypothesis that milk-borne EGF has the potential to modulate endogenous production of EGF and TGF-α in the developing small intestine of neonatal rats. To clarify this hypothesis, we used a unique animal model, AR neonatal rats (18, 19).

METHODS

Animals and AR protocol.

Sixty-five male and female Sprague-Dawley rats originating from 10 different litters were used in the present study. To promote uniformity of growth, litters were routinely culled to 10 pups by d 2 of postnatal life. Beginning on d 8, rat pups were assigned randomly to be mother-fed controls or to the AR protocol in which they were fed with one of the following two diets: growth factor-free RMS or RMS supplemented with rat EGF (Harlan Bioproducts, Indianapolis, IN) in a dose of 100 ng/mL (RMS + EGF). Previous studies have shown (20, 21) that rat milk contains EGF in biologically significant concentrations (35–40 ng/mL). Rao et al. (22) reported the presence of at least three intrinsic peptidase inhibitors in rat milk that may play a role in the protection of milk-borne peptides including EGF in the GI lumen. Because the peptidase inhibitory activity of our rat milk substitute was not tested, we have used a somewhat higher dose of EGF (100 ng/mL) compared with rat milk. The RMS was prepared as previously described by Dvorak and Stepankova (23). AR rats were anesthetized with halothane, and polyethylene cannulas with hook-shaped ends (PE-20, Clay Adams/Benton Dickinson and Co., Parsippany, NJ) were inserted into the stomach by using a 16-gauge Cathlon IV catheter placement unit (Critikon/Johnson & Johnson, Tampa, FL). Cannulas were affixed to rat pups by means of subcutaneous suturing with the cannula exiting at the thoracic vertebral area.

During and after surgery, pups were kept warm on heating pads. After postsurgery recovery, which included a 2-h fast, pups were weighed and the volume of diet was calculated to deliver 37% of body weight per 24-h period. Pups were placed in plastic cups containing Bed O'Cobs corncob bedding (The Anderson's Management Corp., Maumee, OH) and floated in a 39°C water bath (Precision, Chicago, IL) for the duration of the feeding study. The gastrostomy tubes then were connected to syringes on model 44 Harvard infusion pumps (Harvard Apparatus, South Natick, MA) in a refrigerator by means of Silastic tubing (Dow Corning Corp., Midland, MI). Pumps were set to deliver the calculated volume of diet for 20 min followed by a 40-min pause each hour. Initially, the pups received Pedialyte (Ross Laboratories, Columbus, OH) for 2 h; then syringes were replaced to contain either RMS or RMS + EGF. Tubing distance from refrigerated syringe to gastrostomy allowed for gradual warming of the diet to room temperature. The volume of diet to be delivered was recalculated daily. Body weights and tail lengths were recorded daily. Two times daily, urination and defecation were induced by gentle stimulation of the anogenital region. All experiments were approved by the Institutional Animal Care and Use Committee (A-324801–95108).

After a 96-h infusion of RMS with or without EGF, the rat pups were decapitated, trunk blood was collected, and the GI tract was quickly removed. Serum was obtained from blood incubated at 37°C for 1 h followed by centrifugation at 16 000 ×g for 45 s and was stored at −20°C for later analysis. The stomach was separated from the small intestine, which was then carefully extended and its entire length measured. The small intestine was separated from the duodenum and divided into three equal segments referred to as the jejunum, midjejunum, and ileum. Each intestinal segment was flushed with 2 mL of ice-cold 50 mM PBS (pH 7.4). The stomach was incised along the greater curvature, opened, scraped gently to remove luminal contents, and flushed with 2 mL of PBS and patted dry. Postflushing weights of all organs were recorded. A 100-mg section of tissue was taken from the medial portion of the jejunum, midjejunum, and ileum, weighed, and snap-frozen in liquid nitrogen for DNA and RNA analysis. Flushes and remaining organ segments were kept on ice for the duration of the tissue harvest and were then frozen and stored at −70°C.

Tissue and flush preparation.

PBS (2 mL) was added to each tissue sample before processing with a Polytron tissue homogenizer (Brinkmann, Westburg, NY). Samples were kept on ice during the entire tissue preparation including during homogenization. Aliquots of homogenates were diluted with cold deionized water for total protein measurements and then stored at −20°C until assayed. Remaining homogenates were ultracentrifuged at 106 000 ×g for 30 min at 4°C (Beckman Instruments, Palo Alto, CA). Aliquots of each supernatant were diluted and stored at −20°C until they were assayed for total protein. Remaining supernatants were lyophilized (The Virtis Co., Inc., Gardinerville, NY) and stored at −20°C until EGF and TGF-α RIA were performed. The 2-mL flushes were homogenized and ultracentrifuged as described above. Undiluted aliquots of supernatant from flush samples were stored at −20°C until they were assayed for total protein. Supernatants were lyophilized and stored at −20°C until EGF and TGF-α assays were performed.

RNA extraction.

Total RNA was isolated from tissue by using the RNeasy Mini Kit (Qiagen, Santa Clarita, CA) as described in the manufacturer's protocol; then all samples were incubated with RNase-free DNase enzyme (20 U per reaction) for 10 min at 37°C to eliminate DNA contamination. The RNA concentration was quantified by UV spectrophotometry at 260 nm (A260), and the purity was determined by the A260/A280 ratio (SPECTRAmax PLUS, Molecular Devices, Sunnyvale, CA). The integrity of RNA samples was verified by electrophoresis on 1.2% agarose gel containing formaldehyde (2.2 M) and ethidium bromide in 1X MOPS (3-[N-morpholino]propane sulfonic acid) buffer [40 mM MOPS (pH 7.0), 10 mM sodium acetate, and 1 mM EDTA (pH 8.0)].

Reverse transcription competitive-PCR assay.

Reverse transcription (RT) competitive-PCR assay was used to quantify intestinal EGF and TGF-α mRNA levels. DNA fragments containing either rat EGF or TGF-α primer sequences were constructed using the PCR MIMIC Construction Kit (Clontech, Palo Alto, CA) as previously described (24). Even small amounts of peptide-specific mRNA can be measured accurately using this method, without the need for other validation techniques. All chemicals and enzymes used for RT competitive-PCR were purchased from Perkin Elmer (Norwalk, CT).

RT.

Single-stranded cDNA was reverse-transcribed from 1 μg of total RNA in a 10-μL reaction mixture containing 25 U of murine leukemia virus reverse transcriptase, 2.5 μM random hexamers, 10 U of RNase inhibitor, 1 mM of each dNTP, 5 mM MgCl2, 50 mM KCl, and 10 mM Tris-HCl (pH 8.3). The reaction mixture was incubated for 20 min at room temperature, 15 min at 42°C, and then the reaction was terminated by incubation at 99°C for 5 min. The reaction mixture was then kept at 4°C until the start of PCR amplification. The amounts of total RNA used in the RT reactions were calculated from the absorbency at 260 nm and verified by densitometry of the 28S ribosomal RNA band separated on denaturing agarose gels (by Gel Doc 1000 documentation system with Molecular Analyst/PC software, BIO-RAD, Hercules, CA).

Competitive PCR amplification.

The reverse transcriptase product (10 μL) was mixed with a 40-μL PCR reaction mixture containing 0.2 μM of each of the sense and antisense primer, 1 U of Taq DNA polymerase, 1 mM MgCl2, 50 mM KCl, and 10 mM Tris-HCl (pH 8.3). One microliter of a defined concentration of competitor DNA was added to each tube. The mixture was subjected to 30–35 cycles of PCR amplification on a Perkin Elmer DNA thermal cycler 480. The PCR cycle conditions included melting for 1 min at 94°C, annealing for 1 min at 64°C, and primer extension for 2 min at 72°C. Twenty microliters of the amplification products was separated on 2% agarose gel in 40 mM Tris-acetate, 1 mM EDTA (pH 8.0) buffer, and stained with ethidium bromide. The band densities were evaluated by using the Gel Doc 1000 system with Molecular Analyst/PC software (BIO-RAD). To assess relative amounts of natural PCR product, the density ratios between the competitor and target bands were calculated. All PCR products were sequenced on both strands by using the same primers used for the PCR amplification and an automated sequencing system with fluorescent dye terminators (DNA sequencing service, University of Arizona, Tucson, AZ). Control reactions without reverse transcriptase or without total RNA yielded no PCR product bands.

RIA.

The iodination of TGF-α was performed using a modification of the chloramine-T method as described previously (17). Rat TGF-α peptide used for iodination and reference standards was purchased from Bachem Bioscience, Inc. (King of Prussia, PA) and Na 125I from ICN Pharmaceuticals, Inc. (Irvine, CA). The immunoreactive (ir) TGF-α (irTGF-α) in all samples was determined by homologous RIA as described (17). Rabbit anti-rat TGF-α serum used as a primary antibody was purchased from Peninsula Laboratories (Belmont, CA). Radiolabeling of EGF peptide used rat EGF (Biomedical Technologies, Inc., Stoughton, MA) and Na 125I (Amersham Corp., Arlington Heights, IL) in a modification of the chloramine-T method (16). The protocol for the EGF RIA called for rabbit anti-rat EGF serum (Peninsula Laboratories) as primary antibody. Goat anti-rabbit IgG secondary antibody (Antibodies, Inc., Davis, CA) and normal rabbit serum (Sigma Chemical Co., St. Louis, MO) were used for both TGF-α and EGF RIA.

DNA and protein measurements.

Total DNA content in tissue was assayed by the diphenylamine method of Burton (25) and determined by spectrophotometry. Assays for total protein content (26) were performed on homogenates and supernatants; concentration was determined by spectrophotometry (SpectraMAX Plus, Molecular Devices, Sunnyvale, CA).

GI transit.

Poly R-478 (an acetylated anthrapyridone chromophore linked to a polyamino-ethylene-sodium ethylene sulfonate copolymer backbone purchased from Sigma Chemical Co. Corp., St. Louis, MO) was used as a nonabsorbable GI transit marker (27). Neonatal rats were fasted for 4 h; then 100 μL of a 5% solution of Poly R-478 in double distilled water was administered by gastric intubation. Animals were decapitated at 45 min after Poly R-478 feeding, and the stomach and small intestine were then removed. The small intestine was cut into six segments of equal length, numbered from 1 (proximal) to 6 (distal), and all segments were flushed with 6 mL of 0.6% saline. Individual samples from the wall and lumen were collected and analyzed separately for Poly R-478 by a colorimetric assay (28). Results were expressed as a percentage of the sum of total amounts found in the GI tract.

Statistics.

Statistical analysis of the results was performed by 1-way ANOVA followed by Fisher protected least significant difference (PLSD) using the statistical program Statview for Macintosh computers (Abacus Concepts, Inc., Berkeley, CA). A value of p < 0.05 was considered significant. All data in the figures are mean ± SEM.

RESULTS

Body weights and intestinal parameters.

Animals were artificially reared from d 8 until d 12. The survival of AR rats 24 h after intragastric cannulation was approximately 80%. No traces of irEGF or irTGF-α were detected in the RMS diet by RIA. Recovery of rat EGF from the RMS + EGF diet was approximately 95%. Initial body weight of suckling rats at the beginning of the experiment was 21.1 ± 1.1 g; AR rats steadily gained between 2.0–2.4 g/d during the experiment. No statistically significant differences were observed in weight gain and tail length between the AR groups and dam-fed controls (Table 1). Stomach and small intestine wet weights of RMS and RMS + EGF-fed animals were not different, whether expressed either as net organ weight (in milligrams) or as a percentage of total body weight. No significant differences were observed between male and female rats with respect to changes in body or organ weights. However, both AR groups (RMS and RMS + EGF) had significantly enlarged stomach and small intestine compared with dam-fed controls. Intestinal protein content, DNA content, and protein-to-DNA ratios were similar between RMS and RMS + EGF experimental groups but significantly higher compared with control animals.

Table 1 Growth parameters, protein, and DNA content in the GI tract

Intestinal EGF and TGF-α mRNA levels.

To evaluate the changes in the endogenous EGF and TGF-α synthesis as a result of nutritional treatment, intestinal mRNA levels of these two growth factors were measured. The RT competitive-PCR assay was used to measure the steady state levels of EGF and TGF-α mRNA in the duodenum and three segments of the small intestine (Fig. 1). In the duodenum, feeding of growth factor-free RMS resulted in a significant increase of EGF mRNA content (approximately 2–3-fold in comparison with dam-fed littermates). Supplementation of the RMS diet with EGF was associated with a statistically significant decrease in duodenal EGF mRNA content, with the mRNA level in this group intermediate between mRNA levels of dam-fed and RMS-fed rats. No statistically significant changes of EGF mRNA levels were observed in the jejunum and midjejunum. Interestingly, EGF mRNA detected in the ileum exhibited a pattern similar to results observed in duodenum.

Figure 1
figure 1

Expression of EGF and TGF-α mRNA in the small intestine of dam-fed (control) and AR (RMS, RMS + EGF) neonatal rats. A, EGF mRNA levels; (B) TGF-α mRNA levels in the duodenum, the jejunum, the midjejunum, and the ileum. The steady state mRNA levels were quantified by using RT competitive-PCR. Columns are mean values (n = 10–15 samples per group);vertical lines are SE. p < 0.01 AR rats vs control. ‡p < 0.01 RMS vs RMS + EGF. The term amol indicates attamoles (10−18 mol).

The gene expression of TGF-α in the entire small intestine of RMS-fed rats was also markedly changed compared with dam-fed control or RMS + EGF animals. In duodenum, midjejunum, and ileum of RMS-fed rats, TGF-α mRNA levels were increased approximately 2–3-fold in comparison with control dam-fed rats. Supplementation of RMS with EGF normalized TGF-α transcript levels in duodenum and ileum significantly, reaching the same level as dam-fed controls. No statistically significant changes of TGF-α mRNA levels were observed in the jejunum of RMS animals compared with either RMS + EGF or dam-fed controls.

Intestinal EGF and TGF-α peptide levels.

Total content of EGF and TGF-α peptides in small intestinal tissue and intestinal lumen from AR rats was determined by RIA and compared with values obtained from dam-fed littermates (Fig. 2). Animals fed growth factor-free RMS for 4 d exhibited markedly low levels of EGF in the small intestinal tissue (EGF content was decreased by 85% in comparison with dam-fed rats, p < 0.01). Feeding of the RMS + EGF diet resulted in an increase of intestinal EGF level, reaching an approximately 3-fold higher level compared with dam-fed rats. The EGF levels detected in the small intestinal lumen exhibited a pattern similar to tissue values. Tissue content of EGF (in RMS or RMS + EGF groups) was approximately 70% (65 and 76%, respectively) of the total EGF content in the small intestine. In contrast with EGF, artificial rearing of rats with RMS resulted in statistically significant increases of TGF-α levels in the small intestinal tissue compared with dam-fed controls (approximately 38% higher compared with dam-fed rats, p < 0.01). Feeding of the RMS + EGF diet normalized total intestinal TGF-α content to levels similar to those of dam-fed suckling rats. TGF-α content in the lumen of the small intestine was the same in all studied groups.

Figure 2
figure 2

Total content of EGF (A) and TGF-α (B) in the small intestinal tissue and lumen of dam-fed (control) and AR (RMS, RMS + EGF) neonatal rats. Columns are mean values;vertical lines are SE from 15 to 20 rats per experimental group. p < 0.01 AR rats vs control. ‡p < 0.01 RMS vs RMS + EGF.

EGF levels in serum.

Serum EGF concentrations were measured in all experimental rats and compared with data obtained from dam-fed suckling rats (Fig. 3). Serum EGF levels of RMS-fed animals were decreased by 40% (81 ± 6 pg/mL) in comparison with those of dam-fed suckling rats (135 ± 16 pg/mL). Supplementation of RMS with EGF resulted in a statistically significant increase of serum EGF concentration (119 ± 4 pg/mL) compared with RMS rats, reaching concentrations similar to these of dam-fed rat pups.

Figure 3
figure 3

Concentrations of EGF in the serum from dam-fed (control) and AR (RMS, RMS + EGF) 12-d-old rats measured by RIA. Columns are mean values;vertical lines are SE from 10 to 15 rats per experimental group. p < 0.01 AR rats vs control. ‡p < 0.01 RMS vs RMS + EGF.

GI transit.

To test the effect of milk-borne EGF on GI motility, measurements of gastric emptying and intestinal transit were performed at 45 min after marker ingestion (Fig. 4). Results from these measurements showed no effect of milk-borne EGF on either gastric emptying or intestinal motility.

Figure 4
figure 4

The effect of milk-borne EGF on GI motility of AR neonatal rats. Total amount of Poly R-478 marker in the wall and flush per individual segments are expressed as percentage of total counts found in the entire GI tract. Each data point represents the mean;vertical lines are SE from 8 to 10 animals per group.

DISCUSSION

The presence of biologically active peptides, including EGF, in the milk of a number of mammals is well established (13). The low proteolytic activities in the GI tract of newborns and higher “permeability” for macromolecules in neonates indicate a possible functional importance of these milk-borne peptides for developing organisms (14). The importance of the role of orogastrically delivered peptides during development is also supported by the fact that amniotic fluid swallowed by the fetus in the late period of gestation (29) contains considerable concentrations of various biologically active peptides including EGF (30). Transepithelial transport of EGF from amniotic fluid has been demonstrated in fetal rat intestine (31). Despite the large number of studies that have been performed in this field, the physiologic role of EGF in the lumen of the GI tract remains unclear (32).

The major purpose of the present study was to examine the effect of milk-borne EGF on endogenous production of EGF and TGF-α in the small intestine of suckling rats. Because EGF is absent in human infant formulas (33), we also examined the effect of feeding a diet devoid of EGF. Feeding of infant rats via gastric cannula has been used in many nutritional studies during the last three decades. The usefulness of the AR approach has been reviewed in detail by Patel et al. (34, 35). Previously, we have shown the effect of milk-borne IGF-I on the development of neonatal rats by using the AR technique (18). Recently, Staley et al. (19) used an AR model to evaluate the role of milk-borne IGF-I on the intestinal morphology of newborn rats. In the present study, the same technique was used to study the effect of milk-borne EGF on local EGF and TGF-α synthesis in neonatal rats.

Overgrowth of the GI tract is a known consequence of AR, but factors responsible for the acceleration of growth are unknown. Recently, we reported a study investigating the role of artificial diet in causing intestinal overgrowth. Results indicate that the artificial formula, not gastrostomy or AR technique, is responsible for intestinal overgrowth (36). The major focus in the present study was on the effect of milk-borne EGF on the developing GI tract. The results indicate that supplementation of milk formula with physiologic concentrations of EGF has no effect on the somatic growth of AR rat pups over the 4-d study period. Moreover, feeding RMS + EGF had no effect on the weight of the stomach and the small intestine or on the length of the small intestine compared with RMS alone.

Results from previous studies regarding EGF and the suckling have been inconsistent. Moore et al. (37) have shown that the addition of 62 ng/mL of mouse EGF to the diet of AR rats had no effect on body weight gain, and they did not observe any changes in growth of the stomach and the small intestine. In contrast, Berseth (38) reported that short-term feeding (39 h) of newborn rats with formula supplemented with EGF enhanced the growth of rat stomach and the small intestine. However, the doses of EGF used in this study exceeded by 15–75 times the reported concentration of EGF in rat milk. O'Loughlin et al. (39) reported that the intraperitoneal but not orogastric administration of EGF (40 μg·kg−1·d−1) to suckling rabbits from 3 to 18 d of age increased wet weight of the small intestine, whereas body weight gain was not affected. Recently, Vinter-Jensen et al. (40) reported that systemic treatment of adult rats with EGF causes an increase in intestinal weight. This effect was time-dependent, as the weight of the small intestine increased with the duration of the EGF treatment. Thus, it is possible that a 4-d exposure to RMS with or without EGF is too short a time period to detect any statistically significant changes in growth of the entire GI tract.

In suckling rats, intestinal content of EGF and TGF-α peptides is similar but the origin is different. The major source of intestinal EGF content is maternal milk (15, 16). In contrast, TGF-α is not detectable in rat milk and intestinal TGF-α content is likely the result of endogenous synthesis by the small intestine and by pancreatic secretion (17). These observations prompted the question of whether endogenous production of TGF-α and EGF in the small intestine could be stimulated by the absence of EGF in the diet. Previously, we have localized the presence of EGF and TGF-α transcripts in the epithelial crypt cells of duodenum and proximal jejunum of suckling rats (41). Results from recent studies have shown that in dam-fed rat pups (10–12 d of age), intestinal TGF-α mRNA content was about 10-fold higher in comparison with EGF mRNA content (24). Moreover, the transcript stability of TGF-α mRNA was remarkably lower than the stability of EGF mRNA. Thus, high levels of TGF-α mRNA accompanied by the high degradation rate of this mRNA suggested a rapid turnover of intestinal TGF-α mRNA, confirming our hypothesis that intestinal TGF-α is produced locally.

In the present study, we have used the RT competitive-PCR technique to evaluate changes in gene expression of EGF and TGF-α in the small intestine of AR neonatal rats. Feeding of growth factor-free diet for 4 d resulted in significant increases of TGF-α mRNA levels in the developing duodenum and the ileum. Interestingly, supplementation of RMS with EGF normalized TGF-α mRNA levels to the values observed in dam-fed control littermates. Levels of duodenal EGF mRNA exhibited a pattern similar to those of TGF-α mRNA. Animals fed the RMS diet had increased EGF mRNA levels compared with dam-fed controls. Thus, feeding RMS + EGF appeared to down-regulate duodenal EGF mRNA expression, although EGF mRNA levels still remained higher compared with dam-fed rats.

Intestinal EGF peptide content did not correlate well with changes in mRNA levels for EGF. Despite the increase in duodenal EGF mRNA production, total intestinal EGF content in RMS-fed animals fell extensively. Thus, it is likely that the increase of endogenous production of EGF in small intestine cannot completely compensate for the deficit of exogenous milk-borne EGF. The significant increase of intestinal EGF peptide levels in RMS + EGF rats (2–3 times higher compared with dam-fed controls) clearly corresponds to the amount of EGF that was added to RMS (2–3 times higher compared with EGF levels in rat milk). In contrast with EGF, changes in intestinal TGF-α levels correlate well with changes in TGF-α gene expression. Thus, local intestinal production of TGF-α is stimulated by a situation in which EGF delivery is decreased, such as during artificial formula feeding. However, it is likely that growth promoting factors other than TGF-α and EGF are involved in the immediate growth enhancement of the small intestine of AR rats.

Previous studies from our laboratory have shown that orogastrically (21) and/or intestinally administered EGF (42) is absorbed by the gastrointestinal epithelium and transferred via the portal vein into systemic organs. In the present study, we have shown that long-term feeding of the growth factor-free diet (RMS) markedly decreases EGF serum levels compared with dam-fed controls. Supplementation of RMS with EGF normalized EGF serum levels to the values of dam-fed controls. These data suggest that milk-borne EGF is an important factor for maintenance of EGF serum levels in the suckling. However, it remains unclear whether normalization of serum EGF levels in this study was due to exogenous (milk-derived) EGF uptake or to endogenous EGF synthesis.

The differences in EGF serum levels between experimental groups also suggested the possibility of changes in GI motility. Previously, we have shown that the parenteral administration of EGF affects GI transit in suckling rats (27). In that study, the subcutaneous injection of EGF significantly delayed gastric emptying, and the changes in stomach evacuation were EGF-dose dependent. A similar phenomenon was observed in the small intestine in which the administration of EGF markedly delayed small intestinal transit (27). In the present study, results from measurements of gastric emptying and intestinal transit have shown that milk-borne EGF had no effect on GI motility. These results are in agreement with our previous observation that short-term orogastrically administered EGF has no immediate effect on intestinal motility (Shinohara, personal communication, 1998).

In conclusion, results of the present study indicate that feeding of milk with EGF helped to maintain EGF and TGF-α levels in the developing small intestine of neonatal rats. Nutritional deficiency of milk-borne EGF for 4 d (i.e. formula feeding) resulted in significant decreases of intestinal and serum EGF levels. Despite the fact that duodenal EGF mRNA production was elevated in RMS-fed rat pups, intestinal EGF peptide content remained very low compared with dam-fed controls. In contrast to EGF, significantly increased TGF-α mRNA levels in duodenum, midjejunum, and ileum in RMS-fed animals resulted in an increase of the total intestinal TGF-α peptide content. Supplementation of RMS with EGF led to the normalization of duodenal and ileal TGF-α mRNA expression and ileal EGF mRNA level, intestinal TGF-α peptide levels, and EGF serum levels. Feeding of milk with EGF maintained EGF in blood and small intestine, whereas removing EGF from the milk was associated with an increase in intestinal tissue TGF-α levels. GI motility was not affected by milk-borne EGF. These findings strongly suggest that EGF normally found in mammalian milk has the ability to modulate intestinal EGF and TGF-α levels in developing small intestine.