SCIENTIFIC PAPERS

Influence of Epidermal Growth Factor on Intestinal Function The Rat: Comparison of Systemic Infusion Versus Luminal Perfusion

in

Marshall Z. Schwartz, MD and Rostyk B. Storozuk, MS, Sacramento, California

Over the past two decades, many new peptides have been identified that exert a physiologic or pharma- cologic effect on the gastrointestinal tract. One of these substances is epidermal growth factor, a pep- tide first isolated by Cohen [1] in 1962. This peptide was initially extracted from the submandibular gland of mice but Subsequently has been identified in the Brunner's glands of the duodenum and in various body fluids includingsaliva, urine, amniot ic fluid, and breast milk [1-7]. Epidermal growth fac- tor contains 53 amino acids, has a molecular weight of 6,000 daltons, is acid stable, and is resistant to trypsin degradation [1,8,9]. Many recent studies have shown that epidermal growth factor is capable of influencing proliferation of several types of cells including those of the epithelium of gastrointestinal tract in cell culture and in situ [10-15].

Because our laboratory has been interested in studying peptides suggested to exert a trophic effect on the small intestine, we began investigating the effect of epidermal growth factor on small intestine function. We provided the first report that systemic administration of epidermal growth factor could enhance small intestine absorption [16]. We also presented preliminary findings that luminally ad- ministered epidermal growth factor could enhance small intestine function [17]. This study was de- signed to evaluate the influence of long-term con- tinuous administration of epidermal growth factor on small intestine function by determining DNA concentration and substrate absorption and to mea- sure the difference in response between systemic infusion and luminal perfusion,

From the Department of Surgery, Division of Pediatric Surgery, University of California, Davis Medical Center, Sacramento, California.

Requests for reprints should be addressed to Marshall Z. Schwartz, MO, Department of Surgery, University of California, Davis Medical Center, 4301 X Street, Room 2310, Sacramento, California 95817.

Presented at the 28th Annual Meeting of the Society for Surgery of the Alimentary Tract, Chicago, Illinois, May 12 and 13, 1987.

Material and Methods

Fisher strain rats weighing between 150 and 175 g were utilized for both the Systemic infusion and luminal perfu- sion studies. In preparation for the luminal study, Thiry- Vella loops were created. After general anesthesia with ketamine, a midline abdominal incision was made and a 10 cm segment of mid small intestine was isolated with its mesentery. Both ends of the intestinal segment were brought out to the anterior abdominal wall as stomas. Intestinal continuity was reestablished by a single-layer end-to-end anastomosis. A I week recovery period before initiation of the study was sufficient.

The systemic infusions were accomplished using os- motic minipumps (model 2002, Alza, Palo Alto, CA). These pumps are designed to deliver their contents at a constant rate for 14 days. The pumps were filled with saline solution or epidermal growth factor to deliver 150 ~g/kg/day. After the pumps were filled, they were placed in a subcutaneous paravertebral location.

Osmotic minipumps were also employed for the lumi- nal perfusions: Two pumps with 22 gauge polyethylene catheters attached were used for each intestinal segment to be perfused in an effort to establish a more homoge- neous distribution of the perfusate. The tips of the cathe- ters were passed through the proximal stoma and placed at 3 and 6 cm from the proximal end of the isolated intestinal segment. A purse-string suture was placed around the proximal stoma to form a watertight seal. Both tubings were then tunneled subcutaneously to the back of the rat where they were connected to the osmotic pumps filled with either saline solution or epidermal growth fac- tor, for a combined dose of 150 #g/kg/day.

The technique for carbohydrate and amino acid ab- sorption employs a closed recirculation and perfusion system. In the case of the luminal perfusions, the intesti- nal segment studied was the previously isolated Thiry- Vella loop (Figure 1). In the systemically infused rats, a 10 cm segment of mid small intestine was isolated specifical- ly for the absorption studies. Twelve gauge polyethylene catheters were placed in the proximal and distal ends of

18 The American Journal of Surgery

Epidermal Growth Factor and Intestinal Function

Figure 1. Technique for measuring substrate absorption in the Thiry-Vella loop after luminal perfusion of saline solution or epidermal growth factor.

Figure 2. Technique for measuring subslrate absorption in an isolated mid small Intestine segment after systemic saline solution or epidermal growth factor infusion.

the intestinal segment to be studied (Figure 2). They were secured at both ends using purse-string sutures to form watertight seals. The other ends of the catheters were attached to a roller pump (Holter model 903, Extracor- poreal Medical Specialties, King of Prussia, PA). A con- tinuous perfusion at 2 ml/hour for 4 hours was begun utilizing the closed recirculation technique. The perfu- sate consisted of either carbon-14 galactose (1 ~Ci/10 ml) or carbon-14 glycine (1 ~Ci/10 ml) added to their respec- tive unlabeled 5 mM stock solutions prepared in phos- phate-buffered saline solution. The perfusate also con- tained hydrogen-3 inulin (5 ~Ci/10 ml). The inulin was used as a nonabsorbable marker to function as an internal control for loss of substrate by processes other than ab- sorption. During perfusion, duplicate samples of 0.1 ml were obtained from the reservoir at 0, 0.5,1, 1.5, 2, 3, and 4 hours. Samples were placed in a fluor and counted in a liquid scintillation counter. The amount of substrate ab- sorbed was determined by the disappearance of counts from the reservoir expressed as micromoles of substrate absorbed per square centimeter of small intestine accord- ing to the following formula: (1-dpmf/dpmi])(InR) (Ci)(Ve)/Cm 2 intestine, where dpmf indicates the final disintegrations per minute, dpmi indicates the initial dis- integrations per minute of test substance, InR indicates

dpmf/dpmi for inulin, Ci indicates the concentration of test material in micromoles per milliliter, and Ve indicates the circulated volume in milliliters corrected for sam- pling.

After completion of the absorption studies, biopsy specimens of the small intestine mucosa were obtained for determination of DNA concentration. A standard diphe- nylamine procedure described by Burton [18] and later modified by Giles and Myers [19] was used. This tech- nique employs a colorimetric method in which a standard curve is determined using calf thymus DNA diluted in 0.5 normal perchloric acid. The DNA concentration is ex- pressed as micrograms DNA per milligram of intestinal mucosa. The Student's t test for unpaired data was used for statistical analysis.

Results

After systemic infusion of saline solution in the control group (12 rats), the mid small intest ine mu- cosal D N A concentra t ion was 370 • 33 #g/mg com- pared with 438 • 37 ug/mg for the ep idermal growth factor infused rats (15 rats). After luminal perfusion with saline solution in the control group (15 rats), the mucosa l D N A concen t ra t ion was 394 • 43

Volume 155, January 1985 19

Schwartz and Storozuk

TABLE I DNA Concentration and Substrate Absorption After Epidermal Growth Factor Infusion

DNA Concentration Administration n (#g/mg mucosa)

Absorption (p,M/cm 2 small intestine) Galactose Glycine

Systemic Infusion Control 12 370 4- 33 Epidermal growth factor 15 438 4- 37

Luminal Perfusion Control 15 394 4- 43 Epidermal growth factor 14 440 -I- 31

1.39 4- 0.17 1.78 4- 0.38 2.60 4- 0.55* 7.81 4- 0.951"

1.44 4- 0.18 1.63 -4- 0.31 3.51 4- 0.59* 6.71 =:E 0.571

* p <0.05. t p <0.01.

w 9 z

8 u~ uJ 7 I,- z 6 m

.u 5

. - I

4 ~ 3

0

:~ 0

[ ~ ] Control

EGF * *

Galactose Glycine Galactose Glycine

SYSTEMIC LUMINAL

Figure 3. Epidermal growth factor and substrate absorption. Substrate absorption after 14 day continuous exposure, Comparison of systemic infusion with luminal perfusion. Single asterisk Indicates p <0.05. Double asterisk indicates p <0.01.

~g/mg. Luminal perfusion of epidermal growth fac- tor (14 rats) resulted in a mucosal DNA concentra- tion of 440 4- 31 ~g/mg. Statistical analysis showed no significant change in DNA concentration after either systemic infusion or luminal perfusion of epi- dermal growth factor (Table I).

Absorption of galactose was 1.39 4- 0.17 #M/cm 2 small intestine after saline infusion in the control group (12 rats), whereas after epidermal growth factor (15 rats), it was 2.60 4- 0.55 uM/cm 2 small intestine (p <0.05). Glycine absorption was 1.78 4- 0.38 ~M/cm 2 small intestine for the control group and 7.81 4- 0.95 ~M/cm 2 small intestine for the epidermal growth factor-infused rats (p <0.01).

Luminal perfusion of epidermal growth factor �9 also resulted in a marked increase in substrate ab- sorption. Galactose absorption was 1.44 4- 0.18 uM/ cm 2 small intestine for the control group and 3.51 4- 0.59 ~M/cm 2 small intestine after epidermal growth factor perfusion in 14 rats (p <0.05). Glycine ab- sorption was 6.71 4- 0.57 #M/cm 2 small intestine after epidermal growth factor perfusion compared with 1.63 4- 0.31 ~M/cm 2 small intestine for the control group (p <0.01). Although there was a dra- matic response to epidermal growth factor when compared with the control values, there was no sta- tistically significant difference between the luminal

or systemic routes of administration. These data are summarized in Table I and Figure 3.

Comments

The physiologic role that epidermal growth factor may play in the development, adaptation, and func- tion of the gastrointestinal tract is unknown. How- ever, data from several studies suggest that epider- mal growth factor may be involved in all of these phenomena. A significant influence by epidermal growth factor on fetal and newborn rat and mouse smal l i n t e s t i n e has been d e m o n s t r a t e d [11,12,14,15,20,21]. Recently, Read et al [21] report- ed an increase in intestinal weight, protein content, and sucrase activity per unit length after 50 percent small intestine resection in weanling rats when epi- dermal growth factor was added to the formula com- pared with a similar group in which no epidermal growth factor was present in the formula. They sug- gested that this may be an indication that epidermal growth factor plays a role in intestinal adaptation.

Most studies that have evaluated the trophic ef- fects of epidermal growth factor measured mucosal DNA synthesis or ornithine decarboxylase activity. Both of these parameters are thought to be indica- tors of mucosal proliferation. The only previous study that evaluated substrate absorption after ad- ministration of epidermal growth factor was report- ed by us in 1985 [16]. We showed that systemic epidermal growth factor infusion produced a statis- tically significant twofold to fivefold increase in ga- lactose and glycine absorption.

In previous reports, administration of epidermal growth factor has usually been intermittent and continued for a short period of time, most often, for less than 72 hours. In contrast, the current study incorporated several unique features. Epidermal growth factor was administered continuously for a prolonged period of time (14 days). In addition to measuring changes in DNA concentration, absorp- tion of substrates was evaluated as a more relevant parameter of intestinal function, and systemic infu- sion was compared with luminal perfusion. The data from this study show that, although there was a slight increase in DNA concentration after systemic and luminal exposure to epidermal growth factor, it

20 The American Journal of Surgery

was not statistically significant. However, galactose absorption increased 1.5-fold after systemic infu- sion and 2.4-fold after luminal perfusion of epider- mal growth factor. The glycine absorption response to epidermal growth factor was even more signifi- cant with a 4.4-fold increase after systemic infusion and a 4.1-fold increase after luminal perfusion. This enhancement of substrate absorption after luminal epidermal growth factor exposure has not been re- ported previously.

The mechanism for the enhancement of intestinal absorption after epidermal growth factor exposure is unknown. Some investigators have shown an in- crease in DNA and ornithine decarboxylase after epidermal growth factor exposure, suggesting an increase in mucosal turnover or mass [11,22]. Our findings did not support a statistically significant increase in mucosal mass despite the impressive increase in substrate absorption. Also, it is not clear whether epidermal growth factor has a direct effect on small intestine epithelium or whether it stimu- lates the elaboration of other growth factors. Under organ explant and cell culture conditions, epidermal growth factor has produced atrophic response, sug- gesting that it acts directly on the cells [13]. How- ever, under in vivo conditions, it is likely that the influence of epidermal growth factor is much more complex, perhaps acting in an endocrine or para- crine fashion or both. Other reports, as well as the data from this study, demonstrate that epidermal growth factor can stimulate the small intestine mu- cosa in vivo when given systemically or luminally. Ulshen et al [22] reported an increase in jejunal activity after luminal perfusion of the ileum, sug- gesting that either epidermal growth factor was ab- sorbed and acted systemically or it can stimulate a distant mucosal response by an unknown mecha- nism.

By whatever mechanism, epidermal growth fac- tor appears to be a potent stimulator of small intes- tine function. This effect may be physiologic in that epidermal growth factor might be an endogenous growth factor for intestinal development, matura- tion, or adaptation. The data from the pres- ent study demonstrating enhancement of intestinal absorption after administration of epidermal growth factor suggest that there may be a therapeu- tic role for this peptide in children and adults with short bowel syndrome. In addition, epidermal growth factor may have a therapeutic role by accel- erating intestinal maturation in premature infants with poor gastrointestinal development.

Summary Epidermal growth factor may be a trophic sub-

stance for the small intestine. Previous studies had not evaluated intestinal absorption after long-term continuous administration of epidermal growth fac- tor or compared luminal perfusion with systemic infusion. Epidermal growth factor (150 gg/kg/day)

Epidermal Growth Factor and Intestinal Function

"was continuously infused either systemically or lu- minally for 14 days into young adult Fisher strain rats. Luminal studies were performed by creating 10 cm Thiry-Vella loops. At the conclusion of the 14 day infusions, mucosal DNA concentration and ab- sorption of carbon-14 galactose and carbon-14 gly- cine were determined. The increase in DNA concen- tration after systemic or luminal infusion of epidermal growth factor was not statistically signifi- cant. However, galactose absorption increased 1.9- fold (p <0.05) after systemic epidermal growth fac- tor infusion and glycine absorption increased 4.4-fold (p <0.01). Luminal epidermal growth factor perfusion increased galactose absorption 2.4-fold (p <0.05) and glycine absorption 4.1-fold (p <0.01). Thus, both systemic and luminal administration of epidermal growth factor can significantly increase substrate absorption. Additional studies on the physiologic implications and clinical usefulness of these data are warranted.

References 1. Cohen S. Isolation of a mouse submaxillary gland protein

accelerating incisor eruption and eyelid opening in the newborn animal. J Biol Chem 1962; 237: 155-62.

2. Kirkegaard P, Olsen PS, Nexo E, Hoist JJ, Poulsen SS. Effect of vasoactive intestinal polypeptide and somatostatin on secretion of epidermal growth factor and bicarbonate from Brunner's glands. Gut 1984; 25: 1225-9.

3. Li AKG, Schattenkerk ME, Hoffman RG, Ross JS, Malt RA. Hypersecretion of submandibular saliva in male mice: tro- phic response in small intestine. Gastroenterology 1983; 84: 949-55.

4. Gregory H, Holmes JE, Willshire IR. Urogastrone levels in the urine of normal adult humans. J Clin Endocrinol Metab 1977; 45: 668-72.

5. Barka T, Van der Noen H, Gresik EW, Kerenji T. Immunoreac- tive epidermal growth factor in human amniotic fluid. Mt Sinai J Med (NY) 1978; 45: 679-84.

6. Beardmore JM, Lewis-Jones DI, Richards RC. Urogastrone and lactose concentrations in precolostrum, colostrum, and milk. Pediatr Res 1983; 17: 825-8.

7. Read LC, Upton FM, Francis GL, Wallace JC, Dahlenburg GW, Ballard NJ. Changes in the growth promoting activity of human milk during lactation. Pediatr Res 1984; 18: 133-9.

8. Savage CR, Cohen S. Epidermal growth factor and a new derivative: rapid isolation procedures and biological and chemical characterizations. J Biol Chem 1972; 247: 7609- 11.

9. Gregory H. Isolation and structure of urogastrone and its relationships to epidermal growth factor. Nature 1975; 257: 325-7.

10. Cohen S. Epidermal growth factor (EGF). Cancer 1983; 51- 1787-91.

11. Malo C, Manard D. Influence of epidermal growth factor on the development of suckling mouse intestine mucosa. Gastro- enterology 1982; 83: 28-35.

12. Oka Y, Ghishan FK, Green EHL, Orth DN. Effect of mouse epidermal growth factor/urogastrone on the functional maturation of rat intestine. Endocrinology 1983; 112: 940- 4.

13. Chabot JG, Payet N, Hugon JS. Effects of epidermal growth factor (EGF) on adult mouse small intestine in vivo and in organ culture. Comp Biochem Physiol [B] 1983; 748: 247- 52.

14. Dembinski A, Gregory H, Konturek SJ, Polanski M. Trophic action of epidermal growth factor on the pancreas and gastroduodenal mucosa in rats. J Physiol 1982; 325: 35-

Volume 155, January 1988 21

Schwartz and Storozuk

42. 15. AI-Nafussi AI, Wright NA. The effect of epidermal growth

factor (EGF) on cell proliferation of the gastrointestinal mucosa in rodents. Virchows Arch [Cell Pathol] 1982; 40: 63-9.

16. Schwartz MZ, Storozuk RB. Epidermal growth factor en- hances small intestine function. Gastroenterology 1985; 88: 1578.

17. Schwartz MZ, Storozuk RB. Influence of epidermal growth factor on intestinal function in the rat: comparison of sys- temic infusion versus luminal perfusion. Presented at the 6th International Symposium on Gastrointestinal Hor- mones. Vancouver, BC 1986.

18. Burton K. A study of the conditions and mechanisms of the diphenylamine reaction for the colorimetric estimation of deoxyribonucleic acid. Biochem J 1956; 62: 315-23.

19. Giles KW, Myers A. An improved diphenylamine method for the estimation of deoxyribonucleic acid. Nature 1965; 206: 93.

20. Beaulieu JF, Calvert R. The effect of epidermal growth factor (EGF) on the differentiation of the rough endoplastic reticu- lure in fetal mouse small intestine in organ culture. J Histo- chem Cytochem 1981; 29: 765-70.

21. Read LC, Ford WDA, Filsell OH, McNeil J, Ballard FJ. Is orally derived epidermal growth factor beneficial following pre- mature birth or intestinal resection? Endocrinol Exp (Bratisl) 1986; 20: 199-207.

22. Ulshen MH, Lyn-Cook LE, Raasch RA. Effects of intraluminal epidermal growth factor on mucosal proliferation on the small intestine of adult rats. Gastroenterology 1986; 91: 1134-40.

D iscuss ion

Courtney M. Townsend, Jr . (Galveston, TX): Dr. Schwartz, I think your study provides a basis for further studies that ought to delineate the development of the gut with respect to cell numbers and function. Because the Thiry-Vella loop is known to result in hypoplasia of the mucosa, it seems significant that you were able to restore function, although the DNA content didn't increase. This was just a single-dose study. Have you tried increased doses to see whether you can restore the cellular hypopla- sia that results from creating the loop?

James C. Thompson (Galveston, TX): Dr. Schwartz, since I have never been able to figure out why one would ever want a peptide to act in the gut, would you care to speculate on the mechanism of action of the luminally administered epidermal growth factor. Was it simply ab- sorbed and, therefore, acted systemically, or do you think it had some action? Would this local action occur in the presence of normal chyme?

Gerald M. Larson (Louisville, KY): Epidermal growth factor has many effects that you have alluded to, Dr. Schwartz. The eye surgeons at our institution are using it very effectively to treat corneal abrasions, and the plastic surgeons are starting to use it experimentally for therapy of burn wounds. Both groups believe that epidermal growth factor has topical activity and that if you adminis-

ter epidermal growth factor systemically, it would proba- bly not achieve the same results. Therefore, I find it interesting that you studied both luminal and systemic delivery of epidermal growth factor and observed the same results.

Since epidermal growth factor is known to have trophic effects on epithelial surfaces as well as the epithelium of the gastrointestinal tract, have you looked at other prolif- erative indices such as ornithine decarboxylase, weight, or thymidine uptake? This, of course, has a direct bearing on the possible mechanism of action. It may be that your increase of 11 to 15 percent in DNA content might be significant. I recall another study by Johnson and Guthrie in Houston who demonstrated a 20 percent increase in DNA content of oxyntic gland mucosa which, in their experiment, was significant. So perhaps that 11 or 12 percent increase is, in fact, a measure of some trophic activity.

Marshall Z. Schwartz (closing): Dr. Townsend, we did not study different dosages. The same dose by weight was infused continuously for 14 days in all of the rats. At the present time, we have initiated studies to determine a dose-response curve for epidermal growth factor in this model.

Dr. Thompson, you posed the same question to me in 1979. It was a relevant question then, and it is a very relevant question now. The mechanism of action of epi- dermal growth factor and other so-called growth peptides is unknown. Previous studies have shown an increase in DNA concentration or ornithine decarboxylase activity after epidermal growth factor administration in in vivo and cell culture models. This suggests that epidermal growth factor can increase mucosal mass. Data from the cell culture studies suggest that epidermal growth factor can exert a local effect on small intestine epithelial cells. Finally, a study in rats in which epidermal growth factor was infused intraluminally into the ileum resulted in an increase in DNA synthesis and ornithine decarboxylase activity more proximal in the jejunum. The mechanism of this phenomenon is unknown as is the mechanism of action of epidermal growth factor. It is my impression that epidermal growth factor has a complex effect on small intestine epithelium and includes stimulation of cell growth and cell function.

Dr. Larson, we have measured DNA but have not mea- sured weight of the intestine or ornithine decarboxylase. Although we did not show a statistically significant in- crease in DNA concentration after administration of epi- dermal growth factor, we did show a statistically signifi- cant increase in DNA concentration when we studied gastrin in a similar model. You may be correct in suggest- ing that more control animals and more data points may result in achieving statistical significance.

22 The American Journal of Surgery