Main

Whereas it is known that EGF affects gastrointestinal motility in adult animals [guinea pigs in vivo(1) and in vitro in guinea pigs(1) and in rats(2–5)], nothing is known about the effect of EGF in developing mammals. Because EGF receptors were found in the stomach wall of suckling rats(6) and in cultured myocytes from newborn rabbit gastric fundus(7), we hypothesized that EGF will affect gastrointestinal motility in suckling rats. Experiments described below demonstrated that in suckling rats s.c. administered EGF in vivo delays gastric emptying and intestinal transit. Anti-EGF antiserum had an opposite effect on intestinal transit; it caused its acceleration. Preliminary data were published in abstract form(8, 9).

METHODS

Animals. Sprague-Dawley rats 12 d old were obtained from our breeding colony, and the size of litters was adjusted to 10 pups on d 2 postnatally. Gastrointestinal motility was evaluated in most experiments with51 Cr-EDTA(10) and in some experiments with Poly R-478(11) as markers. Pups were removed from the mother 15 h before the experiment and kept in cages with half of the cage resting on a heating pad to maintain body temperature. The experiments were performed between 0900 and 1100 h.

51Cr-EDTA. 51Cr-EDTA (DuPont NEN, Boston, MA; 60,000 cpm in 100 μL of saline/rat) was administered by gastric intubation. Using this volume no counts were found in the duodenum of animals killed immediately after administration. Experimental animals were decapitated 30 or 45 min after 51Cr-EDTA feeding, and the stomach and small intestine then removed. The small intestine was cut into 12 segments of equal length. Each segment was numbered from 1 (proximal) to 12 (distal), and all samples were flushed with 4 mL of saline. Because, in our preliminary experiments, we observed “association” of 51Cr-EDTA with the wall of intestinal segments (for details, see “Results”), we collected and measured samples from the wall and lumen separately. Both the stomach flush and wall and the intestinal flush and wall were placed into test tubes, and radioactivity was determined by a γ counter (Packard Autogamma model 5005, Packard Instrument Co.; Meriden, CT) for 1 min. The recovery of administered counts was in the range of 95-105%; no counts were detected in the urine. Data for each segment were expressed as a percentage of the sum of total counts found in the gastrointestinal tract. Total segmental counts (s), counts in the lumen (f), or counts in the wall(w) were related to the sum of counts found in the appropriate“compartment” of the gastrointestinal tract (i.e. sum of wall and luminal content = S or sum of the total fluid content =F) of the gastrointestinal tract.

Poly R-478. Poly R-478 (purchased from Sigma Chemical Co., St. Louis, MO) was used as another motility marker(11). One hundred microliters of a 5% solution of Poly R-478 in double distilled water were administered by gastric intubation. Animals were decapitated at 45 min after Poly R-478 feeding, and the stomach and small and large intestine were then removed. The small intestine was cut into six segments of equal length. Each segment was numbered from 1 (proximal) to 6 (distal), and all segments were flushed with saline up to the final volume of 6 mL. Samples from the wall and lumen were collected and measured separately; Poly R-478 was determined by colorimetric assay(12). Data were expressed similarly as those obtained with 51Cr-EDTA (see above).

Administration of EGF. Suckling rats were injected s.c. (above the gluteal region; needle directed caudally) with rEGF (Bioproducts for Science Inc., Indianapolis, IN) in varying doses (0.1, 0.5, and 1.0 μg/rat in 0.1% BSA) at different time intervals; the volume used was always 100μL/rat. Control animals received 0.1% BSA only.

RIA. RIA of EGF in the blood was performed on the trunk blood of decapitated sucklings. To release EGF present in platelets(13) fresh blood was incubated at 37°C for 60 min, then spun at 1300 × g for 15 min; supernatants were frozen at-70°C before RIA. Primary antibodies against rEGF(14) were produced in our laboratory as described elsewhere(15) and used as in our previous studies(16); secondary antibodies were obtained from Antibodies, Inc., Davis, CA. In all assays a reference control serum sample was run.

Administration of EGF antiserum. Suckling rats were administered s.c. 50 μL of BSA (0.1%) containing anti-EGF antiserum in a dose neutralizing 125 ng of EGF (anti-mouse EGF serum lot no. 902794, Collaborative Biomedical Products, Becton Dickinson Labware, Bedford, MA); controls received 50 μL of 0.1% BSA only. Rats received simultaneously orogastrically 51Cr-EDTA or in some experiments Poly R-478. A similar dose of antiserum was used in different experiments(17).

Statistics. Statistical analysis of the results was performed by one-way ANOVA followed by Fisher protected least statistical difference 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 means ± SEM.

RESULTS

Verification of the 51Cr-EDTA as a Marker of Gastrointestinal Motility in Suckling Rats

Because in our preliminary experiments we observed an association of51 Cr-EDTA with the gastrointestinal wall, we determined51 Cr-EDTA in suckling rats at 45, 90, and 180 min after its administration separately in the wall and luminal flushes. Results summarized in Figure 1 show that the association of counts increases with the time between 45 and 180 min; it is noteworthy that higher values are seen in segment 6 and beyond. Figure 2,A-C, shows that the peaks of activity expressed as s/S, f/s, andf/F move distally with time in a parallel fashion. Furthermore, data expressed as s/S and f/F were “closer”(especially at 45 min) than when expressed as f/S. This is in agreement with the higher association of 51Cr-EDTA in the distal segments.

Figure 1
figure 1

Association of 51Cr-EDTA in various segments of the gastrointestinal tract. Data are expressed as counts present in the wall(w) divided by counts present in the given segment (s)i.e. (w + f). Means ± SEM are given. St = stomach; numbers denote segment number. Values at 45 min are depicted withfull squares, at 90 min with full circles, and at 180 min with small open squares. n/group = 64, 19, and 3, respectively.

Full size image
Figure 2
figure 2

Presence of 51Cr-EDTA in compartments of the gastrointestinal tract. Means ± SEM are given. Data expressed as total counts per segment as percentage of total counts in the gastrointestinal tract(s/S; open squares), counts in the lumen of a given segments as percentage of total counts in the given segment (f/s;full squares), and as counts in the lumen of a given segment as percentage of counts in the total luminal content of the gastrointestinal tract (f/F; full diamonds). (A) At 45 min,n = 64; (B) at 90 min, n = 19; (C) at 180 min, n = 3.

Full size image

Effect of EGF

51Cr-EDTA studies. Based on the experiments described above we decided to evaluate the EGF effect after 45 min.

Gastric emptying Figure 3 demonstrates the delaying effect of rEGF, given together with 51Cr-EDTA, on stomach evacuation. The lowest dose (0.1 μg) had no effect; however, the 0.5- and 1.0-μg doses were both effective. The same conclusions were obtained if data were expressed as luminal contents only.

Figure 3
figure 3

The dose-dependent effect of rEGF given s.c. on gastric emptying. rEGF was administered at the same time as 51Cr-EDTA. Total counts per stomach (mean ± SEM) expressed as percentage of total counts found in all segments of the entire gastrointestinal tract. n/group= 5-31. Horizontal axis, dose of rEGF in micrograms given per rat;vertical axis, percent of counts found in the stomach.*Significantly different from 0 dose.

Full size image

Figure 4 demonstrates that the effect of the high EGF dose (1.0 μg) is present when EGF was given in the period 10 min preceding and 20 min after 51Cr-EDTA. This effect was absent in those rats that received EGF 60 or 30 min before 51Cr-EDTA.

Figure 4
figure 4

The time dependency of rEGF effect administered on gastric emptying. Arrow denotes administration of 51Cr-EDTA.Horizontal axis, time in minutes. Other arrangement same as inFigure 1. Controls received BSA s.c. (n = 23) at various time periods; because no differences were seen, data were combined. Control data are not shown, because they were identical to values of rats that received rEGF at time -60 min.

Full size image

Intestinal transit. Figures 5 and6 demonstrate that EGF (1.0 μg/rat) delayed the intestinal transit. In agreement with our preliminary observations, we have seen that51 Cr-EDTA is taken up by the intestinal wall. Nevertheless, this phenomenon did not mask the EGF effect, because the EGF-caused delay was seen when data were expressed as counts in the luminal content(Fig. 5), counts in the wall (Fig. 6), or as total counts per segment (data not shown). To verify that the effect was not a result of delayed gastric emptying, we also expressed the intestinal data as the percentage of counts available for the small intestine(i.e. when the counts found in the stomach were subtracted); similar results were obtained (Fig. 7).

Figure 5
figure 5

The effect of EGF on gastrointestinal motility; marker51 Cr-EDTA. Total counts in lumen per individual segments are expressed as percentage of total counts found in lumen of all segments of the entire gastrointestinal tract. Horizontal axis: Stom = stomach, numbers denotes small intestinal segment number. Because in segments 11 and 12 no counts were detected, data for these segments were not included in the figure. rEGF was administered in 0.1% BSA at the same time as 51Cr-EDTA(triangles). Controls (= without rEGF) received 0.1% BSA only(full circles). EGF vs BSA: significant in all segments, except 2, 9, and 10.

Full size image
Figure 6
figure 6

The effect of EGF on gastrointestinal motility; marker51 Cr-EDTA. Total counts in the wall per individual segments expressed as percentage of total counts found in the wall of all segments of the entire gastrointestinal tract. Same arrangement as in Figure 3. EGF vs BSA: significant in stomach and segments 1 and 4-8.

Full size image
Figure 7
figure 7

The effect of EGF on intestinal transit; marker51 Cr-EDTA. Same data and arrangements as in Figure 3, but stomach counts were excluded from total counts. EGF vs BSA: all significant except segments 4 and 8-10.

Full size image

To characterize time and dose dependency of EGF in the small intestine, we have analyzed in detail the values of the combined data for segments 6 and 7. Data (not shown) demonstrated (similarly as shown for the stomach) that in the small intestine both the higher EGF doses had a significant effect and that EGF affects gastrointestinal motility when given in the period 10 min preceding and 20 min after administration of 51Cr-EDTA.

Poly R-478 studies. To verify that the observed effect of EGF was not an artifact caused by the use of 51Cr-EDTA, we performed an additional experiment using Poly R-478 as a motility marker. Suckling rats were given s.c. 1 μg of EGF together with the marker and were killed 45 min later. Results summarized in Figure 8 confirmed the delaying effect of EGF on gastrointestinal motility; in this case, data do not allow us to conclude whether the small intestinal transit was affected by EGF directly or was due to the delayed gastric evacuation.

Figure 8
figure 8

The effect of EGF on gastrointestinal motility; marker Poly R-478. Total amount of Poly R-478 in segmental flush (EGF treated:open circle; controls: full circles) or segmental wall(EGF-treated: shaded columns; controls: black columns).Short vertical lines denote 2 SEM; n/group: EGF = 5, controls = 4). Significant differences: EGF vs BSA (flush: stomach, segments 2-4; wall: segments 3 and 4); flush vs wall (EGF: stomach; controls: segments 2, 3, and 5).

Full size image

Effect of EGF Antiserum

Gastric emptying. These studies used 51Cr-EDTA as a motility marker. There was no effect on the gastric emptying. After 30 min the stomach of control rats contained 25.9% ± 2.0 (n = 18) (mean± SEM), and that of EGF antiserum-treated rats 25.4 ± 2.2(n = 22) of the total counts given.

Intestinal transit. Intestinal transit was accelerated in suckling rats treated with EGF antiserum (Fig. 9). The results were similar when data were expressed as total counts or wall counts per segment.

Figure 9
figure 9

Effect of anti-EGF antiserum on intestinal transit; marker 51Cr-EDTA. Total counts in lumen per individual segments are expressed as percentage of total counts found in lumen of all intestinal segments. Only segments 1-8 contained counts. Controls: full circles, EGF antiserum: squares. EGF antiserum vs BSA, significant differences: segments 4 and 5 and the ratio segment 5/segment 7.

Full size image

Blood levels of EGF in EGF-treated rats. These blood levels were determined in an additional group of suckling rats. Because the effect of EGF on the gastrointestinal motility was maximal when the marker51 Cr-EDTA was given simultaneously with the EGF and the effect of EGF decreased within 10 min, we followed the changes of blood EGF levels 5 min after its administration. The blood EGF levels in controls receiving only BSA were 0.72 ± 0.14 ng/mL (n = 9); after administration of albumin with EGF they increased considerably (4.68 ± 0.71 ng/mL;n = 7).

DISCUSSION

Introductory experiments have shown that, despite the temporary association of 51Cr-EDTA with the gastrointestinal wall on suckling rats, this marker can be used to study the effect of EGF on their gastrointestinal motility, especially if the fate of the 51Cr-EDTA is followed separately in the lumen and wall of the gastrointestinal tract. The use of another motility marker (Poly R-478) substantiated this conclusion.

Present studies demonstrate that EGF influences the gastrointestinal motility of suckling rats in vivo. This thus extends previous observations by others performed in adult guinea pigs(1–5) showing that EGF in vitro affects gastrointestinal motility. The EGF in vivo effect on gastrointestinal motility thus can be added to the list of other gastrointestinal functions that are affected by parental administration of EGF such as gastric secretion(18, 19), pH of the intestinal microclimate(20), and gastric DNA synthesis(21–23).

The conclusion that EGF plays a physiologic role in regulation of gastrointestinal motility in vivo is supported by two basic observations, namely 1) the delaying effect of administered EGF on gastrointestinal motility and 2) the opposite (accelerating) effect of EGF antiserum on intestinal transit. Whereas the dose of EGF used can be considered as a “low” pharmacologic (see below), the inhibitory capability of the amount of the antiserum used corresponds to the amount of EGF calculated to be present in the fasting suckling animal based on published data(16, 24, 25). Furthermore the dose is similar to that used by Berseth(17) who showed its delaying effect on the growth of stomach and small intestine in newborn suckling rats. The EGF dose used (2-4 μg/100 g of body weight) is similar or considerably lower than doses used by others in various experiments exploring the EGF effect on suckling rats, mice, and rabbits. Doses used per 100 g of body weight by others were 4 μg(26), 20μg(27, 28), 40-50 μg(29, 30), 80 μg(31), and 400 μg(32, 33).

The amount of immunoreactive rEGF found in blood (which is about 7% of the body weight, i.e. about 2 mL) represents about 1% of the dose of EGF injected; this is in good agreement with our previous study(34). In this experiment, where 125I-EGF was administered i.v., about 3% of the total radioactivity was found immunoreactive in the blood between 5 and 30 min later.

It is not probable that the EGF effect on gastric emptying is via altering gastric acid production, because gastric acid secretion in suckling rats is negligible and not influenced by s.c. administration of EGF in a large dose range (30-100 μg/100 g of body weight of suckling rats) (Rao et al.(6).

Although it is well established that gastrointestinal motility increases during perinatal development in various experimental mammals(35–38), as well as in humans(39), only a few studies have dealt with the endocrine effects on gastrointestinal motility in the developing gastrointestinal tract. Malloy et al.(40) have shown in dogs that, during the first 2 wk of life, pentagastrin had no effect on antral contraction, but this substance did exhibit an effect at 2-5 wk. Spedaleet al.(41) later noted that pentagastrin decreased the peak duodenal pressure and contraction rate in newborn dogs without affecting the stomach. Although cholecystokinin (octapeptide) administered intraperitoneally was shown to decrease milk intake in suckling rats, it did not slow gastric emptying(42). The effect of neurotensin, bombesin, somatostatin, and vasoactive intestinal peptide administered orogastrically was studied in suckling rats(43); only neurotensin accelerated gastric emptying and small intestinal transit. In the study of Tomomasa et al.(43), bombesin, administered orogastrically in doses of 1 pg, 100 pg, and 10 ng/g of body weight, had no effect, whereas Jianget al.(44) found that gastric emptying was inhibited by higher doses (20 to 1000 ng/g of body weight). Interestingly,in vitro studies using rabbit gastric fundus circular muscle strips(45) have shown that bombesin-stimulated contraction is present in newborns, but loses 80% of its efficacy at the age of weaning. Thein vitro studies of Hyman et al.(46) have shown a relaxing effect of vasoactive intestinal peptide on newborn rabbit gastric muscle strips. Substance P was a less effective agonist in the newborn rabbit than in the weanling in a similar study(47). Our experiments thus include in the list of active agents another factor, namely EGF. EGF, according to Hollenberget al.(5), “appears to be acting directly on the gastrointestinal motility on the smooth muscle elements rather than via the release of neurotransmitters.”