Hirai, Chie*; Ichiba, Hiroyuki*; Saito, Mika*; Shintaku, Haruo*; Yamano, Tsunekazu*; Kusuda, Satoshi†
Growth factors such as epidermal growth factor (EGF), insulin-like growth factor-1 (IGF-1), transforming growth factor-α (TGF-α), and hepatocyte growth factor (HGF) have a trophic effect on the fetal and neonatal gastrointestinal tract (1), promoting proliferation and differentiation of fetal and neonatal gastrointestinal cells. In our previous study, we established a bioassay for growth factors in human milk using a human fetal small intestinal cell line, reporting that human milk had a growth-promoting effect on the cells and also demonstrating that EGF in human milk contributed to this activity (2). These growth factors also have been identified in amniotic fluid (3–5). Other investigations that used various cell lines indicated that growth factors in amniotic fluid might promote proliferation of fetal intestinal cells (6,7). However, the single-factor models used in these studies represented a limited range of in vivo interactions (2,6–8). Thus little is known about the combined actions of multiple growth factors in human milk or in amniotic fluid on the immature gastrointestinal tract (9).
Very low-birth-weight (VLBW) infants who receive early enteral feeding with human milk have been reported to require fewer days to reach full enteral feeding, to have greater gains in weight and head circumference, and to experience fewer gastrointestinal complications than formula-fed infants (10–14). Although gut hormones such as gastrin probably are involved, precise mechanisms are unclear. Suspecting that growth factors in human milk could contribute to these beneficial effects, we tested amniotic fluid, human milk, and various growth factors contained in these fluids on a human fetal small intestinal cell line that represented a suitable model of intestinal cells in the fetus and VLBW infant.
A human fetal small intestinal cell line, FHs 74 Int (15–17), was purchased from the American Type Culture Collection (line number CCL241 ATCC, Bethesda, MD, U.S.A.). Its morphology is epithelial-like, the viability is approximately 93%, and EGF (30 ng/ml) reduces the doubling time from 168 to 65 hours. Dulbecco minimal essential medium and fetal calf serum were obtained from Nissui (Tokyo, Japan). [3H]-thymidine (4.0 Ci/mmol) was purchased from ICN Biomedicals Inc. (Costa Mesa, Ca, U.S.A.). A radioimmunoassay kit for EGF was purchased from Amersham (Amersham Place, United Kingdom). Genistein and dimethyl sulfoxide were purchased from Calbiochem (Tokyo, Japan). Recombinant human TGF-α was purchased from Wako Pure Chemical (Osaka, Japan), whereas recombinant human EGF, HGF, fibroblast growth factor (FGF), anti-human EGF, HGF, and IGF-2 were purchased from Genzyme/Techne (Minneapolis, MN, U.S.A.). Recombinant human IGF-1 was purchased from SIGMA (S. Louis, MO, U.S.A.). Antibodies against human IGF-1, TGF-α, FGF, and vascular endothelial growth factor (VEGF) were purchased from Pepro Tech (London, United Kingdom).
All donors of amniotic fluid and human milk samples gave informed consent. Thirty samples of amniotic fluid were obtained at delivery. Twelve samples were from the mothers of full-term infants (37 to 42 weeks of gestation), and 18 were from mothers of preterm infants (25 to 36 weeks). Five samples of manually expressed breast milk were obtained from five mothers of full-term infants. No maternal disease was documented during pregnancy. Bloody or stained amniotic fluid was excluded. Samples were centrifuged at 18,000 g for 10 minutes at 4°C, and supernatant was frozen and stored at −80°C until subsequent thawing for assays. Epidermal growth factor concentrations were measured using the RIA kit from Amersham (Amersham Place, United Kingdom).
For bioassay of growth-promoting activity, a uniform number of human fetal small intestinal cells were added to Costar 96-well plates and cultured with 200 μL of Dulbecco minimal essential medium containing 10% (vol/vol) fetal calf serum, for 10 to 14 days. All measurements were performed on confluent monolayers. Cells were cultured with 200 μL of serum-free Dulbecco minimal essential medium for 10 hours and then with 20 μL of control fluid, sample fluids, or various concentrations of growth factors added to each well for 24 hours, followed by 2 hours of incubation with [3H]-thymidine (1 μCi/well). Cells were collected on a cell harvester and deposited on discs of filter paper. Radioactivity then was measured with a scintillation counter. Radioactivity was observed and defined as the measure of cell proliferation (growth).
Genistein, an inhibitor of tyrosine kinase (18), was dissolved in dimethyl sulfoxide and used at a final concentration of 50 μg/mL. Cells were preincubated with this inhibitor for 1 hour and then washed free of inhibitor before adding test samples. To study blocking of several antibodies against growth factors, cells were incubated for 24 hours with antibodies against EGF, IGF-1, IGF-2, FGF, HGF, TGF-α, or VEGF (5 μg/mL, respectively) together with fluid samples.
The significance of differences between means was determined using the Student t test, and correlations were determined using the Pearson test. Statistical significance was set at P less than 0.05.
Figures 1 and 2 show [3H]-thymidine incorporation into cultured small intestinal cells incubated with various concentrations of amniotic fluid or human milk. With both fluids, the growth increment beyond control cultures was dose dependent, reaching a maximum at 40% concentration for amniotic fluid and 5% for milk. Accordingly, we used 40% amniotic fluid and 5% human milk in subsequent assays.
Figure 3 shows EGF concentration in amniotic fluid at each gestational age. The concentration of EGF in preterm amniotic fluid obtained before 30 weeks of gestation was significantly lower than in fluid obtained subsequently. However, no significant difference was observed among gestational ages in growth-promoting activity in amniotic fluid (Fig. 4).
Genistein inhibited growth-promoting activity in amniotic fluid by 96% (mean) in cultured small intestinal cells (P = 0.002). This activity in amniotic fluid also was partially inhibited by antibodies against EGF (40%, P = 0.047), IGF-1 (38%, P = 0.047), or FGF (58%, P = 0.014, Fig. 5). Genistein inhibited cell culture growth-promoting activity in human milk by a mean of 98% (P h 0.0001), and activity was partially inhibited by antibodies against EGF (22%, P = 0.036), IGF-1 (40%, P = 0.009), FGF (75%, P = 0.004), HGF (56%, P = 0.001), or TGF-α (73%, P = 0.001, Fig. 6).
When cultured human small intestinal cells were incubated with EGF, IGF-1, FGF, HGF, or TGF-α, [3H]-thymidine incorporation into cells increased in a dose-dependent manner (Fig. 7). Although EGF, IGF-1, FGF, HGF, or TGF-α had a trophic response on the cells, the response was much lower than that for amniotic fluid or human milk. When EGF, IGF-1 and FGF; or EGF, IGF-1, FGF, HGF, and TGF-α were added in combination, the response was greater than that for each growth factor alone. However, the synergistic response of these growth factors was lower than that for amniotic fluid or human milk (Table 1).
To clarify the role of growth factors in amniotic fluid and human milk in the gastrointestinal adaptation of the fetus and the VLBW infant, effects of these fluids and multiple growth factors were investigated in a human fetal small intestinal cell line (FHs 74). Because this cell line is derived from the normal human fetus, because its morphology is epithelial-like, and because EGF reduces the doubling time, this cell line is a suitable model of intestinal cells in the fetus and VLBW infant.
After incubation of cultured human fetal small intestinal cells with amniotic fluid, [3H]-thymidine incorporation showed a dose-dependent increase. This result indicated that amniotic fluid possessed growth-promoting activity in cultured human fetal small intestinal cells. Mulvilli et al. (19,20) reported that esophageal ligation of the fetal rabbit resulted in significant inhibition of gastrointestinal tract growth and gastric function that was reversed by intragastric infusion of amniotic fluid. These results suggest that amniotic fluid may promote gastrointestinal tract growth in the human fetus.
After incubation of fetal intestinal cells with human milk, [3H]-thymidine incorporation into the cells showed dose-dependent increases, indicating that human milk had growth-promoting activity in these cells. Evidence suggests that the fat compartment of human milk provides growth factors and receptors through its milk fat globule/membrane (21). We studied the effects of the aqueous, nonfat, and noncellular components of human milk, and investigated whether, if these components were studied as whole milk, perhaps greater cellular proliferation and, therefore, trophic effects would be attributable to human milk. Small intestinal weight and protein and DNA content in puppies fed canine breast milk were significantly greater than in puppies fed formula (22). Permeability of the small intestine was lower and gastrointestinal tolerance and intestinal lactase activity were greater in preterm infants fed human milk than in those fed formula (13,14). These results suggest that human milk may promote gastrointestinal tract growth and function in newborns.
No significant differences in growth-promoting activity of amniotic fluid in cultured small intestinal cells were seen among a wide range of gestational ages. Because EGF concentration in amniotic fluid obtained before 30 weeks of gestation was lower than the concentration in fluid obtained subsequently, additional factors may contribute to growth-promoting activity in amniotic fluid. Incubation with a tyrosine kinase inhibitor, genistein, almost completely abolished growth-promoting activities in amniotic fluid or human milk, suggesting that these activities are mediated by receptors functionally linked to tyrosine kinase. Blocking activity by antibodies against growth factors suggested that EGF, IGF-1, and FGF contribute to activity in amniotic fluid, whereas EGF, IGF-1, FGF, HGF, and TGF-α contribute to activity in human milk. Although EGF has been considered the main growth factor in amniotic fluid and human milk (20,23), our study suggests that other growth factors such as IGF-1, FGF, HGF, and TGF-α are comparably active. Because the trophic effect of each recombinant growth factor or the synergistic effect of these growth factors were much lower than the effects of amniotic fluid or of human milk, unknown factors in these fluids may be important. Therefore, intestinal cell growth-promoting activity equal to that of amniotic fluid or human milk cannot yet be duplicated artificially, even with recombinant growth factors.
This is the first investigation of the combined action of multiple growth factors in amniotic fluid and human milk on cultured human small intestinal cells in which the two fluids were compared. Growth-promoting activity of amniotic fluid on these cells proved equal to that of human milk. The growth factors in amniotic fluid apparently promote proliferation of intestinal cells in utero, whereas the growth factors in human milk promote proliferation of neonatal intestinal cells. These responses are important for the gastrointestinal tract as it adapts to postnatal requirements.
Very low-birth-weight infants, born in the second trimester, have had reduced exposure to the growth factors in amniotic fluid. This may cause gastrointestinal immaturity and lead to complications. However, VLBW infants fed human milk had fewer gastrointestinal complications (24). More specifically, early initiation of enteral feeding with human milk reduced such complications (10). Growth factors in human milk may act on the gastrointestinal tract of VLBW infants to favor structural and functional maturation in lieu of the growth factors no longer supplied by amniotic fluid. Therefore, enteral feeding of VLBW infants with human milk should be initiated as soon as possible after birth.
In conclusion, EGF, IGF-1, FGF, HGF, and TGF-α in amniotic fluid or in human milk have a trophic effect on immature intestinal cells and may be important in perinatal adaptation of the gastrointestinal tract. Unknown factors in these fluids also may participate, and this issue requires further investigation.
The authors thank Professor S. Ogita and Dr. Y. Nakai for collecting amniotic fluid samples.
1. Carver JD, Barness LA. Trophic factors for the gastrointestinal tract. Clin Perinatol 1996; 23:265–85.
2. Ichiba H, Kusuda S, Itagane Y, et al. Measurement of growth promoting activity in human milk using a fetal small intestinal cell line. Biol Neonate 1992; 61:47–53.
3. Scott SM, Buenaflor GG, Orth DN. Immunoreactive human epidermal growth factor concentrations in amniotic fluid, umbilical artery and vein serum, and placenta in full-term and preterm infants. Biol Neonate 1989; 56:246–51.
4. Merimee TJ, Grant M, Tyson JE. Insulin-like growth factors in amniotic fluid. J Clin Endocrinol Metab 1984; 59:752–5.
5. Kurauchi O, Itakura A, Ando H, et al. The concentration of hepatocyte growth factor (HGF) in human amniotic fluid at second trimester: relation to fetal birth weight. Horm Metab Res 1995; 27:335–8.
6. Mulvihill SJ, Hallden G, Debas HT. Trophic effect of amniotic fluid on cultured fetal gastric mucosal cells. J Surg Res 1989; 46:327–9.
7. Kong W, Yee LF, Mulvihill SJ. Hepatocyte growth factor stimulates fetal gastric epithelial cell growth in vitro. J Surg Res 1998; 78:161–8.
8. Wagner CL, Forsythe DW, Wagner MT. The effect of recombinant TGFα, human milk, and human milk macrophage media on gut epithelial proliferation is decreased in the presence of a neutralizing TGFα antibody. Biol Neonate 1998; 74:363–71.
9. Wagner CL, Forsythe DW. Effect of human milk and recombinant EGF, TGF-α, and IGF-I on small intestinal cell proliferation. Adv Exp Med Biol 2000; 478:373–4.
10. Ichihashi H, Nagasawa H, Kuwabara N, et al. Early enteral feeding for the neonates less than 1000 gram birth weight. Acta Neonat Jpn 1998; 34:589–94.
11. Berseth CL. Effect of early feeding on maturation of the preterm infant's small intestine. J Pediatr 1992; 120:947–53.
12. Troche B, Harvey-Wilkes K, Engle WD, et al. Early minimal feedings promote growth in critically ill preterm infants. Biol Neonate 1995; 67:172–81.
13. Shulman RJ, Schanler RJ, Lau C, et al. Early feeding, antenatal glucocorticoids, and human milk decrease intestinal permeability in preterm infants. Pediatr Res 1998; 44:519–23.
14. Shulman RJ, Schanler RJ, Lau C, et al. Early feeding, feeding tolerance, and lactase activity in preterm infants. J Pediatr 1998; 133:645–9.
15. Smith HS, Springer EL, Hackett AJ. Nuclear ultrastructure of epithelial cell lines derived from human carcinomas and nonmalignant tissues. Cancer Res 1979; 39:332–44.
16. Owens RB, Smith HS, Nelson-Rees WA, et al. Brief communication: epithelial cell cultures from normal and cancerous human tissues. J Natl Cancer Inst 1976; 56:843–9.
17. Smith HS. In vitro properties of epithelial cell lines established from human carcinomas and nonmalignant tissue. J Natl Cancer Inst 1979; 62:225–30.
18. Akiyama T, Ishida J, Nakagawa S, et al. Genistein, a specific inhibitor of tyrosine-specific protein kinases. J Biol Chem 1987; 262:5592–5.
19. Mulvihill SJ, Stone MM, Forkalsrud EW, et al. Trophic effect of amniotic fluid on fetal gastrointestinal development. J Surg Res 1986; 40:291–6.
20. Mulvihill SJ, Stone MM, Debas HT, et al. The role of amniotic fluid in fetal nutrition. J Pediatr Surg 1985; 20:668–72.
21. Mather IH, Banghart LR, Lane WS. The major fat-globule membrane proteins, bovine components 15/16 and guinea-pig GP 55, are homologous to MFG-E8, a murine glycoprotein containing epidermal growth factor-like and factor V/VIII-like sequences. Biochem Mol Biol Int 1993; 29:545–54.
22. Schwartz SM, Heird WC. Further studies of colostrum stimulated enteric mucosal growth. Pediatr Res 1981; 15:546.
23. Carpenter G. Epidermal growth factor is a major growth-promoting agent in human milk. Science 1980; 210:198–9.
24. Lucas A, Cole TJ. Breast milk and neonatal necrotising enterocolitis. Lancet 1990; 336:1519–23.
© 2002 Lippincott Williams & Wilkins, Inc.