Enterocytic Gene Expression Is Altered in Experimental Gastroschisis

JOURNAL OF SURGICAL RESEARCH ARTICLE NO.

68, 1–6 (1997)

JR964986

Enterocytic Gene Expression Is Altered in Experimental Gastroschisis SADEESH K. SRINATHAN, M.D.,* JACOB C. LANGER, M.D.,* JOSEPH L. WANG, M.D.,† AND DEBORAH C. RUBIN, M.D.† Departments of *Surgery and †Medicine, Washington University School of Medicine, St. Louis, Missouri 63110 Submitted for publication June 13, 1996

of human material are lacking because the bowel is rarely resected. However, in the fetal lamb model there is evidence of intestinal smooth muscle hypertrophy and hyperplasia associated with thickening of the submucosa and increased deposition of submucosal collagen [6]. These changes correspond to a decrease in intestinal contractility in this model [4]. Changes in mucosal function have also been documented in the lamb model and consist of villus blunting and decreased mucosal peptidase and lactase activity [7]. Other workers have used a fetal rabbit model of gastroschisis to demonstrate decreased mucosal uptake of proline and glucose [8]. In the present studies, we have used a similar model to characterize the molecular basis of these epithelial changes using immunohistochemical and RNA blot hybridization techniques. In this model we created a large abdominal wall defect in order to focus on the effect of amniotic fluid exposure, without bowel constriction, on epithelial development. We chose to examine expression of lactase, apolipoprotein A-I, and cellular retinol binding protein II. These three enterocytic genes are important in the absorption and processing of different types of nutrients [9–12] and are normally expressed in fetal intestine.

Gastroschisis is a congenital anomaly in which exposure of the intestines to amniotic fluid throughout fetal life results in nutrient malabsorption. To begin to understand the molecular basis underlying epithelial changes in this condition, we investigated enterocytic gene expression during fetal development. Gastroschisis was surgically created at 24 days gestation (term Å 31 days) in fetal rabbits; sham-operated and unoperated fetuses served as controls. Bowel was harvested at 28 and 31 days gestation. Cellular lactase expression was detected using immunohistochemistry, and apolipoprotein A-I and cellular retinol binding protein II (CRBPII) mRNA levels were quantitated using Northern blot analysis. Despite absence of gross histological changes in the mucosa, lactase protein expression and apolipoprotein A-I and CRBPII mRNA expression were decreased in intestine from gastroschisis compared to sham-operated animals. Expression of GAPDH (a housekeeping gene) increased over the same period, suggesting that the changes in enterocytic absorptive gene expression associated with gastroschisis were relatively specific. In conclusion, a decrease in expression of a variety of genes involved in nutrient absorption and trafficking within the enterocyte may contribute to the absorptive defects seen in this gastroschisis. q 1997 Academic Press

METHODS The model. The experimental model of gastroschisis was adapted from Phillips et al. [13]. Time-dated pregnant New Zealand White rabbits (3.5–4.5 kg) were acclimated in the Animal Care Facility at Washington University Medical School at least 4 days prior to surgery, fed high fiber rabbit diet (Purina, St. Louis), and housed with standard 12-hr day/night cycles. At 24 days of gestation (term Å 31 days), does were given intramuscular ketamine (20 mg/kg for sedation), medroxyprogesterone (8.5 mg/kg for tocolysis), and enrofloxacin (5 mg/kg for antibiotic prophylaxis). Anesthesia was induced and maintained using 1.5–3.0% halothane by spontaneous ventilation. After maternal laparotomy, the ovarian end of each uterine horn was exposed, fetal position was determined by palpation, and a purse string suture of 5-O polypropylene suture was placed over the rump of the fetus. The uterus was incised and the fetal hindquarter was exposed, taking care not to expose or kink the umbilical cord. A 5- to 10-mm fetal laparotomy was performed through a paramedian incision and the intestine was extruded through the abdominal wall defect by gently massaging the fetal back and abdomen using a cotton-tipped applicator. The intestine was not handled. Warmed sterile saline was infused into the amniotic cavity, the fetus was returned to the uterus, and the purse string was closed. The sham operation was identical except that a skin suture was placed

INTRODUCTION

Gastroschisis is a congenital anomaly that occurs approximately once every 5000 live births, characterized by an abdominal wall defect through which the bowel eviscerates during fetal life [1]. Children born with gastroschisis face prolonged difficulties with nutrient absorption [2] and poor intestinal motility [3], likely due to exposure of the developing intestine to amniotic fluid and to mechanical obstruction at the abdominal wall [4]. In some cases, evidence of nutrient malabsorption persists long after the initial intestinal dysmotility has resolved, and some children may have long-term sequela such as poor weight gain and delayed development [5]. The intestines in gastroschisis are usually thickened and covered by a fibrinous ‘‘peel.’’ Histological studies

0022-4804/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.

1

AID

JSR 4986

/

6n1a$$$281

03-07-97 01:38:46

srga

2

JOURNAL OF SURGICAL RESEARCH: VOL. 68, NO. 1, FEBRUARY 15, 1997

on the fetus instead of a fetal laparotomy. In each doe, one gastroschisis and one control (sham or unoperated) were used from the two ovarian ends of the uterus. The 24th day of gestation was chosen because of poor survival after fetal surgery at earlier ages [14]. However, 24 days was early enough to permit exposure of the fetal bowel to amniotic fluid during intestinal villus formation [15]. All experiments were approved by the Animal Studies Committee at Washington University. Tissue collection. Gastroschisis, sham-operated, and unoperated fetuses were delivered at 26, 28, and 31 days gestation by cesarean section. Gross appearance was graded (1 Å normal, 2 Å fibrous peel without bowel wall thickening, 3 Å fibrous peel with wall thickening, 4 Å areas of necrosis). The intestine was then removed in continuity from the duodenum to the descending colon and immediately placed in iced 0.9% NaCl solution where all subsequent dissections were carried out. The lumen was flushed with iced 0.9% NaCl to remove digestive enzymes. The intestine was dissected free of the mesentery and the ileum was separated from the duodenum and jejunum. Several sections of ileum were fixed in 4% paraformaldehyde in phosphate-buffered saline and embedded in paraffin for histological sections, and the rest of the ileum was snap-frozen in liquid nitrogen and stored at 0707C for RNA extraction. Morphometry. Sections were stained using hematoxylin and eosin. Villus height was determined by measuring 10 intact villi from 2 cross sections of each piece of bowel. A digitized image at 2001 magnification was obtained with a computerized image analysis system consisting of a microcomputer running an image analysis software (OPTIMAS, Seattle, WA) coupled to a Nikon photomicroscope and a Sony video camera. Villus height, defined as the distance from the crypt villus junction to the villus tip, was measured. Immunohistochemical analysis. Lactase expression was detected in intestine from 31-day gestation fetuses by standard immunogold staining with silver enhancement techniques [16]. Paraffin-embedded tissue sections (6–8 mm) from two cross sections of each piece of bowel were immunostained using a human monoclonal antibody to lactase (mlac 5; 1:40, gift of D. Swallow, University College, London, UK), which has been previously shown to detect lactase expression in the rabbit [17]. Villi were divided into thirds and the distribution of lactase staining was assessed blindly for each slide. RNA isolation and Northern blot analysis. To examine patterns of expression of apolipoprotein A-I, cellular retinal binding protein II (CRBPII), and glyceraldehyde 3-phosphate dehydrogenase (GAPDH), quantitative Northern blot analysis was performed using total RNA. RNA was extracted from 200 mg of ileum, pooled from 5 to 8 animals per group, using guanidium isothiocyanate extraction [18]. Total RNA (20 mg) was electrophoresed through a 1% agarose– formaldehyde gel, transferred to a nylon membrane and hybridized with 32P-labeled cDNA probes for human apolipoprotein A-I (gift of V. Zannis, Boston University), rat CRBP II (M. Levin, Washington University), or human GAPDH (American Type Culture Collection). Both of these human probes have been shown to cross-react with the rabbit. Hybridization was performed at 427C and blots were stringently washed in 0.21 SSC at 507C. This resulted in detection of a single band on Northern blot for each probe. The relative abundance of each mRNA in the different mRNA samples was calculated using scanning laser densitometry. RNA blots were normalized for errors in loading by quantitation of 18S ribosomal RNA. Images of ethidium bromide-stained membrane photographs were captured using a UMAX PS-2400X scanner and UMAX Magicscan, version 1.2 (UMAX Technologies, Fremont, CA) on a Macintosh computer system and quantitated using the NIH Image, version 1.55, densitometric quantitation program. Statistical analysis. Continuous variables which had a normal distribution were compared using a two-factor (time and intervention) analysis of variance (ANOVA). Continuous variables not following a normal distribution or having dissimilar standard deviations were analyzed using Kruskal–Wallis test for ranked data. Statistical analysis of mRNA levels was not done, since pooled samples were used.

RESULTS

Fetal survival. Operations were performed on 40 mothers (77 fetuses). Gastroschisis defects were cre-

AID

JSR 4986

/

6n1a$$$281

03-07-97 01:38:46

ated in 56 fetuses, of which 30 (56%) survived until the delivery. Sham operation was carried out on 21 fetuses, of which 16 (76%) survived. Fetal weight. There were no significant differences in body weight between groups at 26 or 28 days gestation. At 31 days gestation gastroschisis animals weighed 42.4 { 9.60 g (mean { SD), while the sham animals weighed 54.7 { 6.8 g (P õ 0.05, ANOVA). Bowel grade and histology. Sham and unoperated animals all had a median bowel grade of 1 at all gestational ages. There were obvious changes in gross appearance in the gastroschisis animals, which had median grades of 2, 3, and 3 at 26, 28, and 31 days. These changes were variable both within and between animals. Only one gastroschisis rabbit demonstrated dilation of the bowel, but the rest showed no evidence of intestinal obstruction. There were no differences in the amniotic fluid of experimental and control rabbits with respect to volume, color, or odor. Histologically, gastroschisis bowel was characterized by the presence of a fibrous peel and thickening of the smooth muscle and submucosal layers. In the epithelium, focal areas of gross blunting and villus dilatation were present in some specimens. However, epithelial cells in both groups appeared intact, with normal histologic appearance on H & E staining (Fig. 1). Villus height. Villus height was more variable in gastroschisis animals than controls, with focal areas of gross blunting evident in some specimens. No significant differences in villus height were noted between sham-operated and gastroschisis intestine at 26 days (281 { 38 vs 315 { 39 mm), 28 days (336 { 53 vs 252 { 50 mm), or 31 days (332 { 32 vs 343 { 61 mm). Immunohistochemical analysis of lactase expression. Lactase expression was evaluated only in term fetuses (Fig. 2). All animals in both control groups had staining at the villus tips, and some of the animals had staining in the middle and base of the villus. Although most of the gastroschisis animals had staining at the villus tips, only one animal had evidence of lactase expression in the proximal sections of the villus (Fig. 3). The difference in distribution of lactase expression between groups was statistically significant. Expression of mRNA for apolipoprotein A-I and CRBPII. To determine whether the effect of gastroschisis on enterocytic gene expression was a generalized phenomenon affecting many absorptive functions, we examined the expression of apolipoprotein A-I and CRBPII. Antibodies to rabbit apolipoprotein A-I and CRBPII were not available, but cDNA probes were readily accessible. We therefore quantitated apolipoprotein A-I and CRBPII mRNA levels in sham-operated and gastroschisis intestine, using the Northern blot hybridization technique (Fig. 4). At 28 and 31 days of gestation, expression of apolipoprotein A-I and CRBPII was diminished in gastroschisis compared to sham-operated animals (Figs. 5A and 5B). This decrease was particularly striking at 31 days of gestation.

srga

SRINATHAN ET AL.: ENTEROCYTIC GENE EXPRESSION IN GASTROSCHISIS

3

FIG. 1. Histological characterization of normal and gastroschisis intestine at 31 days gestation. Sections of distal intestine from unoperated, sham-operated or gastroschisis rabbit fetuses were stained with hematoxylin and eosin. Tissues were harvested at 7 days after surgery. (A) Unoperated 31-day-gestation rabbit intestine, demonstrating intact villi and nascent crypts. (B) Sham-operated 31-day-gestation rabbit intestine. The normal morphology of the mucosa is preserved. (C) Section of gastroschisis rabbit intestine. Note that the majority of villi demonstrate normal height. (D) Gastroschisis intestine. There is dilatation in the lamina propria of some villi but the epithelium remains intact. (E) High-power view of unoperated mucosa demonstrating normal enterocytes and villus structures. (F) The normal morphology of the 31-day-gestation rabbit intestine is preserved in gastroschisis fetuses (A, 1001; B, 1001; C and D, 1001; E and F, 2001).

Expression of mRNA for GAPDH. Since lactase, apolipoprotein A-I, and CRBPII expression were all diminished in the mucosa from gastroschisis animals, we wished to determine whether this was a generalized effect involving all intestinal genes or was specific to those gene products with an absorptive function in the intestine. We therefore quantitated GAPDH expression in all three groups. Unlike CRBPII and apolipoprotein A-I, GAPDH mRNA levels were unchanged in gastroschisis intestine compared to controls at 26 and 28 days of gestation and were markedly increased at term (Fig. 5C). DISCUSSION

In this study we report that the fetal rabbit model of gastroschisis is characterized by markedly decreased expression of several enterocytic genes with absorptive function. Lactase was chosen because its develop-

AID

JSR 4986

/

6n1a$$$281

03-07-97 01:38:46

mental physiology is well understood and because it is clearly critical for the normal digestion of lactose in breast milk [12]. Cellular retinol-binding protein II is also known to be expressed during fetal life, is involved in the absorption of micronutrients, and is important in vitamin A absorption and trafficking in the enterocyte [9, 10]. Vitamin A malabsorption could play a critical role in preventing recovery of gut function after surgical correction of gastroschisis. Apolipoprotein A-I is expressed during fetal life and is most abundantly synthesized in the gut [11]. It is an important component of intestinal chylomicrons, which are required for the transport of dietary triglyceride and fat-soluble vitamins [19]. Deficient chylomicron synthesis could also contribute to weight loss and malnutrition in gastroschisis patients. Apolipoprotein A-I is also the major lipoprotein in high-density lipoprotein cholesterol [11]. We believe that the changes in gene expression seen in our model were due primarily to amniotic fluid expo-

srga

4

JOURNAL OF SURGICAL RESEARCH: VOL. 68, NO. 1, FEBRUARY 15, 1997

This effect also appeared to be relatively selective, since the expression of GAPDH, a ‘‘housekeeping’’ gene, was not decreased. In fact, GAPDH mRNA levels were increased at the time of maximal diminution in the expression of lactase, apolipoprotein A-I, and CRBPII, suggesting that GAPDH may be developmentally regulated in rabbit intestine, as in other systems such as rat brain and fetal rat intestine [23; D. Rubin, unpublished observation]. It is unclear why there was further enhancement in GAPDH expression in gastroschisis animals. GAPDH mRNA levels can be induced by a variety of substances, including glucose [24], glucocorticoids, and thyroxine [25]. It is possible that systemic factors or substances contained in amniotic fluid induce the expression of this gene in gastroschisis. Since RNA was isolated from full-thickness intestine, it is also possible that this increase could reflect changes in nonepithelial cells. The smaller weights of gastroschisis animals at term were consistent with the growth retardation seen in human infants with this condition. Several lines of evidence indicate that the observed decreases in gene expression were not simply a result of a reduction in enterocyte cell mass due to lower body weight. First, immunohistochemical analysis showed that lactase expression per cell was altered in villi that were otherwise normal in height and cell number. Second, GAPDH expression would be expected to be decreased, not increased, in the gastroschisis animals if enterocytic cell mass were decreased in gastroschisis animals. Third, there were no significant differences in villus height between gastroschisis and control groups, and since the height of the villus is determined by the number of enterocytes per villus, we must conclude that the average number of enterocytes per villus was the same. Our findings support the notion that amniotic fluid exposure exerts its effects by mechanisms other than simple destruction of the mucosa. Although there were focal changes in villus height with occasional blunting noted, the average villus height was not significantly different among groups, and enterocytes appeared histologically normal. In addition, the distribution and exFIG. 2. Lactase expression at 31 days gestation. Sections of distal intestine from unoperated, sham-operated, or gastroschisis rabbit fetuses were incubated with a monoclonal anti-human lactase antibody that also detects rabbit intestinal lactase (see Methods). Antigen–antibody complexes were detected using immunogold staining with silver enhancement techniques. The black color indicated by the arrows depicts the presence of lactase in the brush border. (A) Section of unoperated intestine. Positive immunostaining is apparent from villus base to tip. (B) Section of sham-operated intestine demonstrating lactase expression on the villus tip and mid villus. (C) Section of gastroschisis intestine. Lactase is expressed on the villus tips only (A, 1601; B and C, 1001).

sure, since the use of a large abdominal wall defect resulted in less bowel distension or mesenteric venous and lymphatic dilatation than in previous rabbit models [13, 20]. An adverse effect of amniotic fluid on the intestine has been documented by other authors in a variety of models [4, 21, 22].

AID

JSR 4986

/

6n1a$$$281

03-07-97 01:38:46

FIG. 3. Distribution of lactase expression at 31 days gestation. S, sham operated; G, gastroschisis. Villi were divided into thirds, and presence or absence of lactase staining in each third was noted. The gastroschisis animals demonstrated a significant decrease in staining, particularly in the crypt and mid-villus regions.

srga

SRINATHAN ET AL.: ENTEROCYTIC GENE EXPRESSION IN GASTROSCHISIS

tent of lactase expression was markedly altered in villi of normal height. These results indicate that amniotic fluid can affect the function of intact, normal-appearing enterocytes. The mechanisms by which amniotic fluid exerts this effect are unknown. The persistent expression of lactase at the villus tips, but not at the bases, in gastroschisis intestine suggests that the deleterious effect of serosal amniotic fluid exposure is mediated within the crypts. It may be that ‘‘old’’ cells, already present on the villi and preprogrammed to express lactase, do so despite exposure to amniotic fluid, whereas new cells produced by the crypts that have begun their migration onto the villus at the time of surgery are inhibited in some manner from producing their normal enterocytic products. A variety of cytokines, including epidermal growth factor and interleukin-6, are present in amniotic fluid [26, 27]. These factors have been shown to inhibit the expression of another brush border enzyme, sucraseisomaltase, in Caco-2 cells [28]. It is possible that these or similar substances present in the amniotic fluid may indirectly influence the crypt cell to shut off the genetic program that initiates the expression of a variety of enterocytic genes. However, the ‘‘master genes’’ that regulate the differentiation of the enterocyte have not yet been identified. The cis- and trans-acting elements that regulate the expression of lactase, apolipoprotein A-I, and CRBPII genes have been at least partially characterized [29–32], but the influence of amniotic fluid and its components on their expression is unknown. The effect of serosal amniotic fluid exposure is likely mediated by mechanisms within the smooth muscle or the submucosa. We have shown in other studies using the same model that the submucosa becomes thickened and exhibits a marked increase in collagen mRNA and protein expression [33]. This collagen is produced by unidentified submucosal cells, and not by smooth muscle cells. Mesenchymal–endodermal interactions have long been known to be critical in the initiation and maintenance of a fully differentiated epithelium [12, 34]. Identifying the mechanisms that mediate the mucosal effects of serosal amniotic fluid exposure remains a critical area for future investigation. The contribution of diminished enterocytic gene expression to gastroschisis-related malabsorption is unknown, since the nature of the defect in human neo-

5

FIG. 5. Enterocytic gene expression, detected by Northern blot analysis and quantitated by scanning laser densitometry. (A) Expression of mRNA for apolipoprotein A-I was markedly diminished in the gastroschisis animals at both 28 and 31 days gestation. (B) Expression of mRNA for CRBPII was markedly diminished in the gastroschisis animals at both 28 and 31 days gestation. (C) Expression of mRNA for GAPDH was similar between the two groups at 26 and 28 weeks gestation, but was markedly increased in the gastroschisis animals at term.

nates has not been well documented or characterized. Disturbances in motility likely play a major role. Experimental studies have documented decreased mucosal absorption of glucose and proline [8] as well as decreased brush border enzyme activity [7]. In the present study, we have shown for the first time that exposure to amniotic fluid without bowel constriction inhibits enterocytic gene expression, even without histologic evidence of damage. These studies suggest that correction of defective mucosal function, in addition to treatment of dysmotility, may improve clinical outcome for children with gastroschisis. REFERENCES

FIG. 4. Example of Northern blot analysis. This autoradiograph represents ileum from 31-day-gestation animals (lane 1, unoperated; lane 2, sham operated; lane 3, gastroschisis). Two bands are seen in each group, corresponding to apolipoprotein A-I and 18S standard.

1. Langer, J. C. Omphalocele and gastroschisis. Semin. Pediatr. Surg. 5: 124, 1996. 2. O’Neill, J. A., and Grosfeld, J. L. Intestinal malfunction after antenatal exposure of viscera. Am. J. Surg. 127: 129, 1974. 3. Rubin, S. Z., Martin, D. J., and Ein, S. H. A critical look at

AID

srga

JSR 4986

/

6n1a$$$281

03-07-97 01:38:46

6

JOURNAL OF SURGICAL RESEARCH: VOL. 68, NO. 1, FEBRUARY 15, 1997 delayed intestinal motility in gastroschisis. Can. J. Surg. 21: 414, 1978.

4. Langer, J. C., Longaker, M. T., Crombleholme, T. M., Bond, S. J., Finkbeiner, W., Rudolph, C. A., Verrier, E. D., and Harrison, M. R. Etiology of bowel damage in gastroschisis. I. Effects of amniotic fluid exposure and bowel constriction in a fetal lamb model. J. Pediatr. Surg. 24: 992, 1989. 5. Berseth, C. L., Malachowski, N., Cohn, R. B., and Sunshine, P. Longitudinal growth and late morbidity of survivors of gastroschisis and omphalocele. J. Pediatr. Gastroenterol. Nutr. 1: 375, 1982. 6. Srinathan, S. K., Langer, J. C., Blennerhassett, M. G., Harrison, M. R., Pelletier, G. J., and Lagunoff, D. Etiology of intestinal damage in gastroschisis. III. Morphometric analysis of the smooth muscle and submucosa. J. Pediatr. Surg. 30: 379, 1995. 7. Langer, J. C., Bell, J. G., Castillo, R. O., Crombleholme, T. M., Longaker, M. T., Duncan, B. W., Bradley, S. M., Finkbeiner, W. E., Verrier, E. D., and Harrison, M. R. Etiology of intestinal damage in gastroschisis. II. Timing and reversibility of histologic changes, mucosal function, and contractility. J. Pediatr. Surg. 25: 1122, 1990. 8. Shaw, K., Buchmiller, T. L., Curr, M., Lam, M. M., Habib, R., Chopourian, H. L., Diamond, J. M., and Fonkalsrud, E. W. Impairment of nutrient uptake in a rabbit model of gastroschisis. J. Pediatr. Surg. 29: 376, 1994. 9. Levin, M. S., Li, E., Ong, D. E., and Gordon, J. I. Comparison of the tissue-specific expression and developmental regulation of two closely linked rodent genes encodeing cytosolic retinol binding proteins. J. Biol. Chem. 262: 7118, 1987. 10. Levin, M. S. Cellular retinol binding proteins are determinants of retinol uptake and metabolism in staly transfected Caco-2 cells. J. Biol. Chem. 268: 8267, 1993. 11. Davidson, N. O. Cellular and molecular melchanisms of small intestinal lipid transport. In L. R. Johnson (Ed.), Physiology of the Gastrointestinal Tract, 3rd ed. New York: Raven Press, 1994. Pp. 1909–1934. 12. Henning, S. J., Rubin, D. C., and Shulman, R. J. Ontogeny of the intestinal mucosa. In L. R. Johnson (Ed.), Physiology of the Gastrointestinal Tract, 3rd ed. New York: Raven Press, 1994. Pp. 571–610. 13. Phillips, J. D., Kelly, R. E., Fonkalsrud, E. W., Mirzayan, A., and Kim, C. S. An improved model of experimental gastroschisis in fetal rabbits. J. Pediatr. Surg. 26: 784, 1991. 14. Kizilcan, F., Tanyel, F. C., Buyukpamukcu, N., and Hicsonmez, A. Fetal survival after in utero experimentation in the rabbit. Lab. Anim. Sci. 44: 144, 1994. 15. Toofanian, F., and Targowski, S. P. Morphogenesis of rabbit small intestinal mucosa. Am. J. Vet. Res. 43: 2213, 1982. 16. Rubin, D., Swietlicki, E., Roth, K. A., and Gordon, J. I. Use of fetal intestinal isografts from normal and transgenic mice to study the programming of positional information along the duodenal-to-colonic axis. J. Biol. Chem. 267: 15122, 1992. 17. Maiuri, L., Rossi, M., Raia, V., D’Auria, S., Swallow, D., Quaroni, A., and Auricchio, S. Patchy expression of lactase protein in adult rabbit and rat intestine. Gastroenterology 103: 1739, 1992. 18. Chomezynski, P., and Sacchi, N. Single step method of RNA isolation by acid–guanidium thiocyanate, phenol–chloroform extraction. Anal. Biochem. 162: 156, 1987.

AID

JSR 4986

/

6n1a$$$281

03-07-97 01:38:46

19. Tso, P. Intestinal lipid absorption. In L. R. Johnson (Ed.), Physiology of the Gastrointestinal Tract, 3rd ed. New York: Raven Press, 1994. Pp. 1867–1908. 20. Sherman, N. J., Asch, M. J., Isaacs, H., and Rosenkrantz, J. G. Experimental gastroschisis in the fetal rabbit. J. Pediatr. Surg. 8: 165, 1973. 21. Albert, A., Julia, M. V., Morales, L., and Parri, F. J. Gastroschisis in the partially extraamniotic fetus: Experimental study. J. Pediatr. Surg. 28: 656, 1993. 22. Aktug, T., Erdag, G., Kargi, A., Akgur, F. M., and Tibboel, D. Amnio–allantoic fluid exchange for the prevention of intestinal damage in gastroschisis: An experimental study on chick embryos. J. Pediatr. Surg. 30: 384, 1995. 23. Lustig, R. H., Hua, P., Wilson, M., and Federoff, H. J. Ontogeny, sex dimorphism, and neonatal sex hormone determination of synapse-associated messenger RNAs in rat brain. Brain Res. Mol. Brain Res. 20: 101, 1993. 24. Hauguel-de-Mouzon, S., Miejen, C., Alengrin, F., and Van Obberghen, E. Glucose-induced stimulation of human insulin-receptor mRNA and tyrosine kinase activity in cultured cells. Biochem. J. 305: 119, 1995. 25. Freyschuss, B., Sahlin, L., Masironi, B., and Eriksson, H. The hormonal regulation of the oestrogen receptor in rat liver: An interplay involving growth hormone, thyroid hormones and glucocorticoids. J. Endocrinol. 142: 285, 1994. 26. Miyazawa, K. Role of epidermal growth factor in obstetrics and gynecology. Obstet. Gynecol. 79: 1032, 1992. 27. Yoon, B. H., Romero, R., Kim, C. J., Jun, J. K., Gomez, R., Choi, J. H., and Syn, H. C. Amniotic fluid interleukin-6: A sensitive test for antenatal diagnosis of acute inflammatory lesions of preterm placenta and prediction of perinatal morbidity. Am. J. Obstet. Gynecol. 172: 960, 1995. 28. Molmenti, E. P., Ziambaras, T., and Perlmutter, D. H. Evidence for an acute phase response in human intestinal epithelial cells. J. Biol. Chem. 268: 14116, 1993. 29. Mangelsdorf, D. J., Umesono, K., Kliewer, S. A., Borgmeyer, U., Ong, E. S., and Evans, R. M. A direct repeat in the cellular retinal binding protein type II gene confers differential regulation by RXR and RAR. Cell 66: 555, 1991. 30. Nakshatri, H., and Chambon, P. The directly repeated RG(G/ T) TCA motifs of the rat and mouse cellular retinol-binding protein II genes are promiscuous binding sites for RXR, RAR, HNF-4, and ARP-1 homo- and heterodimers. J. Biol. Chem. 269: 890, 1994. 31. Troelsen, J., Mehlum, A., Olsen, J., Spodsberg, N. H., G. H., Prydz, H., Noren, O., and Sjostrom, H. 1 kb of the lactasephlorizin hydrolase promoter directs post-weaning decline and small intestinal-specific expression in transgenic mice. FEBS Lett. 342: 291, 1994. 32. Bisaha, J. G., Simon, T. C., Gordon, J. I., and Breslow, J. L. Characterization of an enhancer element in the human apolipoprotein C-III gene that regulates human apolipoprotein A-I gene expression in the intestinal epithelium. J. Biol. Chem. 270: 19979, 1995. 33. Srinathan, S. K., Langer, J. C., Botney, M., and Pelletier, G. J. Submucosal collagen in experimental gastroschisis. J. Surg. Res. 65: 25, 1996. 34. Kedinger, M., Simon-Assmann, P. M., Lacroix, B., Marxer, A., Hauri, H. P., and Haffen, K. Fetal gut mesenchyme induces differentiation of cultured intestinal endodermal and crypt cells. Dev. Biol. 113: 474, 1986.

srga