Main

In rodents, maturational changes in intestinal structure and digestive enzyme activity occurs at the time of weaning and involves a dramatic decline in lactase (lactase-phlorizin hydrolase), appearance of sucrase (sucrase-isomaltase), and a rapid increase in maltase, glucoamylase, trehalase, and alkaline phosphatase (1–5). In most neonates, the role of lactase is crucial, because this enzyme is the only small intestinal brush-border hydrolase capable of digesting lactose (4). In adults, sucrase represents the major disaccharidase constituent of the brush-border membrane and forms two active subunits that hydrolyse maltose/sucrose, maltose/isomaltose, some α-amylases, and α-glucosides (5). In immature mammals, such as rats or mice, sucrase activity is absent in fetal life and during suckling but increases at weaning, marking the transition to a diet containing carbohydrates with α-linkages (5). The developmental decline in lactase activity and the appearance of sucrase activity are regulated by a number of factors, including genetic determinants (6,7) and nutritional regulators, particulary in the induction of sucrase activity (8). Glucocorticoids are also key determinants of the rise in sucrase activity at the time of weaning, but a role for glucocorticoids in the coincidental decline in lactase is less clear (3,5). Other regulatory factors include thyroxin (9–11), insulin (12,13), and epidermal growth factor (14–16). Modulation of disaccharidase activity has also been shown for gastrin, enteroglucagon, and polyamines (17,18).

It is now accepted that IGF-I acts in an endocrine, paracrine, or autocrine manner on a variety of tissues, and several lines of evidence have identified IGF-I as a regulator of gastrointestinal growth and maturation during postnatal development (19–23). Although systemically derived IGF-I may be beneficial as therapeutic agent in the treatment of gastrointestinal conditions, such as short bowel syndrome, s.c. delivered IGF-I also exerts selective growth effects on non-gut organs, most notably the kidneys, spleen, and thymus (21). Suggestions that milk-derived IGF-I acts trophically on gut tissues without affecting peripheral organ growth stems from observations that IGF-I is present in mammary secretion in most mammals, with up to 20 times higher levels in early lactation when compared with milk at later postnatal periods (24). However, the few previous studies investigating the effects of orogastrically derived IGF-I on gut growth and enzyme maturation in suckling animals have produced mixed results. In suckling rats, specific enzyme activity of several small intestinal enzymes were increased after orogastric installation of 1 µg/d recombinant IGF-I delivered for 6 d to rat pups between d 10 and 16 pp (25). Similarly, Houle et al. (20) reported that 14 d of oral delivery of 0.2 µg g-1 d-1 IGF-I, to colostrum-deprived piglets born by cesarean section, stimulates small intestinal disaccharidase activity without affecting whole body weight or organ growth. Intestinal enzyme maturation in newborn rats receiving 1 µg/d for a total of 3 d has been observed by Ma and Xu (26), which in an earlier communication reported that, in newborn piglets, addition of 2 µg/mL IGF-I to infant formula fails to stimulate growth of gastrointestinal components but increases crypt cell proliferation (27). In contrast, marked stimulation of mucosal tissue weight, mucosal protein and DNA content, and histomorphometric parameters was observed in newborn piglets after orogastric administration of 3.5 µg g-1 d-1 for 4 d (22). IGF effects on enzyme activity was not investigated in this study.

In a recent study, we have shown that systemically administered IGF-I or the more potent IGF-I analog, LR3IGF-I, stimulates growth of gastrointestinal tissues throughout postnatal development in the suckling rat (21). We now present data to support that systemic but not orogastric infusion of IGF-I peptides stimulates the maturation of intestinal disaccharidases. Furthermore, we examined the effects of orogastrically delivered IGF-I peptides on gut growth and enzyme maturation in normal suckling rats, and for direct comparison, the study design of the orogastric infusion protocol parallels the systemic infusion study, with doses for IGF-I and LR3IGF-I chosen, based on their ability to evoke maximum growth responses when delivered systemically.

METHODS

IGF-I peptides. Recombinant human IGF-I (hIGF-I) and recombinant human LR3IGF-I were supplied by GroPep Pty. Ltd., Adelaide, South Australia. Specifications for LR3IGF-I have been published previously (28). To minimize variations that may occur as a result of differences in peptide batch, a single batch of IGF-I and LR3IGF-I was used in these studies. All concentrations of infused peptides were verified by competitive RIA as described previously (21).

Systemic infusion study. The data presented in the present communication are derived from the same animals described in a previous study (21). Briefly, pregnant Hooded Wistar rats were housed individually and monitored for the time of birth. On the second postnatal day, litters were culled to nine rats per dam to assure similar milk intake across litters. Starting on d 4 pp, all rats and dams were handled and weighed at 24-hourly intervals. On d 6 or 12 pp, rat pups were assigned to their treatment groups receiving IGF-I, LR3IGF-I, or vehicle (0.1 M acetic acid). IGF-I was administered at doses of 30, 75, or 188 µg/d and 45, 114, or 278 µg/d to the 6- and 12-d-old rat pups, respectively. LR3IGF-I was also effective over these dose ranges, but the highest dose was precluded for both age groups due to the appearance of hypoglycemic symptoms. IGF-I peptides or vehicle were infused via mini-osmotic Alzet pumps (model 1007D, Alza, Palo Alto, CA), implanted on d 6 or 12 pp between 1600 and 1700 h. Pump implantation details have been described previously (21). Peptides or vehicle were infused over a 6.5-d treatment period. As the infusion rate was constant over the treatment period, daily peptide dose per g of body weight in animals receiving IGF-I peptides declined gradually with increasing body weight gain but averaged 2, 5, or 12.5 µg g-1 d-1 for both age groups. Each treatment group comprised six to nine animals.

Orogastric infusion protocol. In a separate study, four litters of rat pups were assigned into treatment groups so that two rats from each litter were blindly randomized to receive 12.5 µg g-1 d-1 IGF-I, 5 µg g-1 d-1 LR3IGF-I, or saline vehicle by three times daily gavage at 700, 1200, and 1900 h. There were six to eight animals in each treatment group for each of the two infusion periods. We also included six untreated control animals for each age group to determine whether the gavage procedure would stress the animals. The doses for IGF-I and LR3IGF-I reflect the highest doses of peptides administered in the systemic infusion trial. A schematic representation of the experimental design is shown in Figure 1.

Figure 1
figure 1

Schematic representation of the experimental design for the systemic and orogastric infusion protocols. In the systemic infusion study, vehicle or IGF-I peptides were infused between d 6.5 and 13 pp or d 12.5 and 19 pp. For both age groups, IGF-I was infused at doses of 2, 5, or 12.5 µg g-1 d-1, LR3IGF-I was infused at 2 or 5 µg g-1 d-1, and vehicle-treated rats received 0.1 M acetic acid vehicle. There were six to nine rats per treatment group. In the orogastric infusion study, rat pups received 0.1 mL of saline vehicle, 12.5 µg g-1 d-1 IGF-I, or 5 µg g-1 d-1 LR3IGF-I three times daily by orogastric gavage. IGF-I peptides were administered between d 6 and 12 pp or d 12 and 18 pp with six to eight rats per treatment group.

Full size image

Tissue collection. After the peptide treatment by either orogastric gavage or via miniosmotic pump, rat pups were separated from their mothers and killed. The entire gastrointestinal tract was rapidly excised, and wet tissue weight and length measurements of gastrointestinal components were taken as described previously (29,30). Representative sections were collected for subsequent biochemical and histocytochemical analyses.

Biochemical analysis. The duodenal length in suckling rat pups is inadequate to conduct biochemical as well as histologic and histocytochemical analysis from tissue samples representing the same region, hence all biochemical assays were carried out in tissue samples collected from the proximal jejunum, in a segment immediately adjacent to the duodenum. The 4-cm long segment required for biochemical analysis was collected adjacent to the ligament of Treitz for measurements of mucosal protein and DNA content and biochemical disaccharidase activity. For this purpose, the jejunal specimen was opened longitudinally by cutting along the mesenteric border. The tissue was then rinsed in cold 0.9% wt/vol NaCl, blotted dry, and weighed. Placing the serosal side onto a ice-cold glass slab, the mucosa was separated from the underlying muscularis externa and serosa using a glass slide. Mucosal scrapings were weighed, immediately frozen in liquid nitrogen, and stored at -70°C until subsequent biochemical analysis. To measure the jejunal DNA and protein content, 1.5 mL of 1 M NaOH were added to the thawed samples. Diluted samples were homogenized at 20°C for 30 s using an Ultra Turrax T25 (IKA-Labortechnik, Staufen, Germany) at 15 000 × g. The mucosal DNA content was measured by the method of Burton (31). The protein content was measured by Dulley and Grieve's (32) modification of the Lowry method. Measurements for DNA and protein have been adopted to microplates, and colorimetric reaction products were measured by an MR 7000 ELISA microplate reader (Dynatech, Chantilly, VA). DNA and protein content readings were taken at 620 and 690 nm, respectively.

Lactase (EC 3.2.1.23-62)- and sucrase (EC 3.2.1.10)-specific activities were also measured in homogenates of jejunal tissues by the method described by Dahlqvist (33). Briefly, jejunal homogenates were diluted 1/80 with 50 mM cold phosphate buffer containing 0.002% vol/vol Triton X-100 (pH 6.1). An equal volume of substrate (either 0.2 M sucrose or 0.2 M lactose) in 50 mM phosphate buffer (pH 6.1) was then added to 50 µL of the homogenate and incubated at 37°C for 30 min. The enzyme reaction was stopped by adding 200 µL of 0.625 M Tris (pH 7.0), containing 0.0015% wt/vol horseradish peroxidase, 32 mM o-dianisidine, and 0.375% vol/vol glucose oxidase. Hydrolysis of the substrate by either sucrase or lactase was measured by the production of glucose. The glucose concentration was detected colorimetrically by measuring the samples at 490 nm on the MR 7000 ELISA microplate reader (Dynatech). Enzyme activity units (U) were defined as nanomoles of glucose hydrolyzed per min per cm of tissue at 37°C, and specific enzyme activities were expressed as enzyme activity units (U) per µg of DNA. All biochemical assays were conducted in quadruplicate, corrected for background absorbency, and included relevant controls to establish the inter- and intraassay variability.

Histocytochemical detection of sucrase and lactase activity. To further describe quantitative changes in disaccharidase activity between treatment groups, sucrase and lactase activity was determined histocytochemically in tissue samples from rat pups infused systemically with the highest dose of IGF-I (12.5 µg g-1 d-1) and LR3IGF-I (5 µg g-1 d-1) and compared with vehicle infused rats from the same study. For this purpose, duodenal segments were collected at the time of sacrifice, embedded in OCT tissue medium (Tissue TEK, Miles, Diagnostic Division, Elkhart, IN), and immediately placed in liquid nitrogen before storage at -70°C until further processing. Cryostat sections of 8-µm thickness were prepared and transferred onto gelatin-coated coverslips, rather than glass slides, so that all incubations could be carried out in 5-mL staining jars (Colombia jars). All sections were kept at -18°C during preparation. To determine sucrase and lactase activities from virtually identical sites of the intestine, two coverslips with four sections each were prepared for each animal. Detection of lactase activity was carried out following the general descriptions by Lodja and Kraml (34) and Lodja et al. (35). Briefly, tissue sections were thawed for 10 min at room temperature before 15 min of incubation at 37°C in a medium containing 1.1 mM 5-bromo-4-chloro-3-indolyl-α-frucopyranoside, 3 mM ferricyanide, 3 mM ferrocyanide in 0.1 M citric acid phosphate buffer (pH 6.0). The enzyme reaction was stopped by rinsing the sections in water followed by postfixation in 4% Formal calcium. The coverslips were mounted onto glass slides using Immuno-Mount, an aqueous, nonfading mounting medium (Shandon, Pittsburgh PA). No counterstain was applied. The enzyme reaction product for lactase appeared blue. Sucrase activity was measured as α-glucosidase activity, which represents a combination of sucrase, isomaltase, maltase II and III, and trehalase. The generalized protocol by Gutschmidt et al. (36) was followed, detecting α-glucosidase activity with a saturating substrate concentrations of a solution containing 12 mM 2-naphthyl-α-D-glucopyranoside and 0.6% hexazonium-p-rosaniline in 0.1 M citric acid buffer (pH 6.0). To prevent diffusion of the naphthol substrate enzyme reaction product from its site of expression, the thawed sections were briefly prefixed in 4% Formal calcium for 5 min at 4°C before incubation with the substrate (diffusion was not a problem with the indolyl substrate). Sections were then incubated at 37°C for 18 min, and hydrolysis was stopped by rinsing the sections for 5 min in isotonic saline, followed by postfixation in 4% Formal calcium. The coupling agent hexazonium-p-rosaniline produces an orange to mauve color of the enzyme reaction product. No counterstain was applied. In a series of preliminary experiments, the optimal sample treatment, incubation times, substrate concentrations, and pH of the substrates were determined. For all data presented, incubations were carried out at saturating substrate conditions under initial enzyme reaction rates. Unless specified otherwise, all chemicals were purchased from Sigma Chemical Co. (St. Louis, Mo).

Analysis of enzyme reaction products from histocytochemistry. All sections were examined with an Olympus BH-2 light microscope at a magnification of 250 times. The microscope was mounted with a monochrome, high resolution CCD camera (TM-7, Pulnix, Industrial Products Division, Clayton, Australia). Maximal absorbency for the color reaction product of sucrase occurred over a spectrum range of 480 to 530 nm. Therefore, the microscope was equipped with a neutral density filter (ND-1) and a FS 500 nm circular microscopic field filter for densitometric estimation of the color enzyme reaction product for sucrase. For densitometric scanning of lactase activity, the microscope was fitted with a ND-1 filter and a FS 600 nm circular field filter. For each animal, the absorbency of the enzyme reaction color product was measured along the length of six perfectly oriented villi using an image analysis software program (Video Pro 32, Leading Edge, Adelaide Australia). Absorbency readings were recorded in arbitrary units in values between 0 and 1. Values close to 0 depict color intensities near background of the microscopic slides. Values close to 1 represent maximal detectable color development (black). Thus, after standardization for background staining, absorbency readings were computed along the right-hand side of a villus structure by tracing a 1-pixel thick line (1.67 µm) starting from the base of the crypt villus junction up to the villus tip. Absorbency readings were taken automatically at 4-pixel intervals (6.7 µm), and the corresponding distance along the length of the villus was recorded as shown in Figure 2. On average, 100 measurements were taken for a single villus structure, but for graphic representation, absorbency readings of the enzyme product have been expressed as a function of distance along the villus axis, with the base and tip of the villus representing the starting and end point, respectively. To account for differences in villus length between the six individual measurements for each animal, villus length was standardized to represent the average villus length. Absorbency values were then expressed along the standardized villus height so that, for all animals within the same group, absorbency readings at the 50% mark of the villus length was representative of half the villus distance. This correction was necessary because average villus length varied between individual animals and with treatment. The mean of the standardized villus length was then plotted against the corresponding average absorbency value for each animal to establish the enzyme distribution profile for each group. For statistical analysis, absorbency readings were compared between groups at predetermined distances along the villus axis to represent characteristic enzyme activities at the base of the villi (50 and 100 µm), the midvillus region (250 and 400 µm), and the villus tip (500 and 600 µm).

Figure 2
figure 2

Schematic representation of measurements of enzyme color reaction product for lactase and α-glucosidase staining along the length of duodenal villi by computer-assisted image analysis. Absorbency readings were recorded along the brush border from the base to the tip of duodenal villi.

Full size image

Data analysis. All values are expressed as mean ± SEM. For each experiment, treatment groups were compared by ANOVA and where significance was attained (p < 0.05), a post hoc Fisher's protected least significant difference was applied for identification of between-group differences. All statistical analysis was performed using Super ANOVA (Abacus Concepts, Berkeley, CA).

RESULTS

Induction of duodenal mucosal growth by IGF-I peptides. In the current communication, we have focused on the effects of IGF-I on the maturation of disaccharidases in the proximal small bowel. The growth parameters of this region of the gut are shown in Table 1. The wet tissue weight of the duodenum in the younger rat pups increased by up to 19% after systemic infusion of 2-12.5 µg g-1 d-1 IGF-I and by 28 and 59% after administration of 2 or 5 µg g-1 d-1 of the more potent analog, LR3IGF-I, respectively. In the older rat pups, an increased responsiveness to IGF-I and LR3IGF-I was demonstrated, leading to a maximum increase in duodenal tissue weight after continuous s.c. delivery of 5 µg g-1 d-1 LR3IGF-I, by up to 86% above control values (Table 1). As in the younger animals, LR3IGF-I was substantially more potent than was the native peptide. Increments in wet tissue weight were, in part, due to the lengthening of the duodenum in rat pups infused systemically with IGF-I peptides (Table 1). In contrast, no change in the weight or length of the duodenum was observed in the rat pups treated orogastrically with 12.5 µg g-1 d-1 IGF-I at both age groups. Moreover, LR3IGF-I did not confer any potency advantage over the native IGF-I when delivered via the orogastric route (Table 1).

Table 1 Duodenal wet tissue weight and length in rat pups infused systemically between days 6.5-13 pp or days 6-12 pp (syst) with 2, 5, or 12.5 µg g-1 d-1 of IGF-I, 2, or 5 µg g-1 d-1 of LR3IGF-I or with vehicle. In the orogastric infusion study (orog.), only the highest dose of IGF-I (12.5 µg g-1 d-1) and LR3IGF-I (5 µg g-1 d-1) was administered. Orogastric instillations were administered between days 6 to 12 pp or days 12 to 18 pp. Values represent means ± SEM for six to nine rats per group.
Full size table

To determine the mucosal wet weight, protein, and DNA content, a 4-cm jejunal segment was scraped to separate mucosa from the underlying tissue layers (muscularis externa and serosa). In the younger rats, infused systemically with IGF-I peptides, mucosal weight and protein and DNA contents were similar with no statistically significant difference between treatment groups (Table 2). In the older rat pups, the mucosal weights from the 4-cm jejunal scrapings were significantly greater in rat pups treated with 12.5 µg g-1 d-1 IGF-I and both doses of LR3IGF-I than in controls (Table 2). In the group infused with the highest dose of LR3IGF-I, the increase in mucosal wet tissue weight was accompanied by an increase in mucosal DNA content (Table 2). In the rat pups treated orogastrically with IGF-I or LR3IGF-I, mucosal weight and DNA and protein contents were comparable to values obtained for the saline-treated and untreated control rats at both age groups (data not shown).

Table 2 Weight and protein and DNA content of 4-cm mucosal scrapings of rat pups treated systemically with or without IGF-I peptides over the postnatal periods d 6.5-13 or days 12.5-19 pp
Full size table

Maturation of disaccharidases after systemic IGF-I peptide infusion. For direct comparisons, intestinal lactase and sucrase activity were measured in the same mucosal homogenate as used for the determination of DNA and protein content. In the rat pups treated systemically between d 6.5 and 13 pp, total lactase activity [nanomoles of glucose hydrolyzed per min at 37°C per cm of tissue (U/cm)] was unaffected by systemic infusion of IGF-I, but increased from 287 ± 32 U/cm (vehicle) to 469 ± 61 U/cm in rats treated with 5 µg g-1 d-1 LR3IGF-I (Fig. 3A). This effect presumably reflected changes in tissue mass as the signified response was abolished when lactase activity was expressed per µg of DNA (Fig. 3A, inset). In the same age group, total and specific sucrase activity was minimal across all groups (Fig. 3B, inset). These results suggest that jejunal sucrase activity was not induced by IGF-I peptide administration at this stage of gut development.

Figure 3
figure 3

Duodenal lactase activity per cm of duodenum (U/cm) for rat pups treated systemically for 6.5 d with vehicle (□), 2, 5 or 12.5 µg g-1 d-1 IGF-I (▪), or 2 or 5 µg g-1 d-1 LR3IGF-I (□) starting infusion on either d 6 (A) or d 12 (C), respectively. Duodenal sucrase activity per cm (U/cm) for rat pups treated between d 6.5 and 13 or between d 12.5 and 19 is shown in panels B and D, respectively. Insets represent mean specific enzyme activity in U/µg DNA (nmol glucose min-1 µg-1 DNA). Different superscripts among treatment groups within each age group indicate significant differences at p < 0.05.

Full size image

In the older rat pups, systemic infusion of 12.5 µg g-1 d-1 IGF-I or 2 µg g-1 d-1 LR3IGF-I increased lactase activity per cm of tissue; however, in the rat pups infused with the highest dose of LR3IGF-I (5 µg g-1 d-1) a significant reduction (-49%) in total lactase activity (U/cm) was observed. Lactase activity remained low after corrections for increases in mucosal DNA content in this group (Fig. 3C, inset). Conversely, total sucrase activity (U/cm) was increased 3.6-fold in the same group of rat pups (Fig. 3D), remaining significantly elevated when expressed per µg of mucosal DNA content (Fig. 3D, inset). This suggests that infusion of the highest dose of LR3IGF-I accelerated the reciprocal switch in disaccharidases from lactase to sucrase and that maturation of disaccharidase expression in jejunal enterocytes can be regulated by systemic infusion of LR3IGF-I. In the rat pups infused orogastrically IGF-I or LR3IGF-I did not alter the total or specific enzyme activities for lactase or sucrase in either age group, indicating that orogastrically delivered IGF-I does not mature the disaccharidase profile in suckling rat pups (Fig. 4, A and B).

Figure 4
figure 4

Total duodenal lactase (A) and sucrase activity (B) in U/cm for rat pups infused with vehicle (□), 12.5 µg g-1 d-1 IGF-I (▪), or 5 µg g-1 d-1 LR3IGF-I (□), by three times daily orogastric gavage. IGF-I peptides were administered for 6 d starting treatment on either d 6 or 12 pp with six to eight rat pups per treatment group.

Full size image

Quantitative expression of disaccharidases by rat jejunal enterocytes. To further characterize the effects of systemically administered IGF-I and LR3IGF-I on the expression of jejunal disaccharidases, histocytochemical detection of lactase and sucrase was performed in cryostat sections of duodenum from rat pups infused with the highest dose of IGF-I (12.5 µg g-1 d-1) and LR3IGF-I (5 µg g-1 d-1) and compared with sections from vehicle-treated rat pups.

Histocytochemical staining for lactase activity for the younger rat pups (d 13 pp) showed a distinctive profile. For all treatment groups, expression of lactase in enterocytes near the crypt-villus junction was low, increasing rapidly to achieve maximal activity in the midvillus region and then declining progressively toward the tip of the villi (Fig. 5A). The α-glucosidase (sucrase) activity could be detected only in trace amounts in these animals with values too low for quantitative assessment (results not shown).

Figure 5
figure 5

Histocytochemical staining profile of disaccharidase activity along duodenal villi taken from rat pups treated with vehicle (•), 12.5 µg g-1 g-1 IGF-I (▪), or 5 µg g-1 d-1 LR3IGF-I (▴) by continuous systemic infusion. Absorbency readings were taken starting at the base of the villus to the villus tip in 8-µm cryostat-sectioned duodenum, representing the localization and intensity of the color reaction product for lactase from rat pups treated between d 6.5 and 13 pp (A) or d 12.5 and 19 pp (B). The localization of sucrase activity along duodenal villi in rat pups treated with vehicle or IGF-I peptides between d 12.5 to 19 is shown in panel C. Different superscripts indicate significant differences from vehicle treated group at p < 0.05 as detected by ANOVA.

Full size image

In the older age group, rat pups infused with vehicle showed a lactase profile along the crypt-villus axis similar to the one observed for the younger rats, with low lactase activity at the villus base and tip and maximal expression in the midvillus region. However, the response to systemic IGF-I infusion differed so that treatment with 12.5 µg g-1 d-1 IGF-I resulted in increased lactase expression, particularly in enterocytes near the villus tip (Fig. 5B). Most interestingly, however, the rat pups treated with 5 µg g-1 d-1 LR3IGF-I showed markedly reduced lactase expression in jejunal enterocytes covering the entire length of villi (Fig. 5C). In fact, in three out of the six animals in this treatment group, lactase activity was barley detectable (Fig. 6, A and B). This provides further evidence that systemic infusion of LR3IGF-I stimulated the precocious decline of lactase activity. At the same time, sucrase activity, which was relatively low in the older rats treated with either vehicle or IGF-I, was switched on by the infusion of 5 µg g-1 d-1 LR3IGF-I (Fig. 6, C and D). In addition, these data support the biochemical enzyme data, demonstrating distinctive effects LR3IGF-I on the maturation pattern of lactase and sucrase activity during the late lactation period in suckling rats.

Figure 6
figure 6

Photomicrograph of cryostat-sectioned duodenum taken from rat pups treated for 6.5 d with vehicle (A) or 12.5 µg g-1 d-1 LR3IGF-I between d 12.5 and 19 pp (B). The blue staining in panels A and B indicates the localization and intensity of lactase activity. Lactase activity was determined histocytochemically by the formation of the color reaction product after incubation with 5-bromo-4-chloro-3-indolyl-α-frucopyranoside as the substrate. Sucrase activity was determined in rat pups infused systemically with vehicle (C) or 5 µg g-1 d-1 LR3IGF-I (D) between d 12.5 and 19 pp. The color reaction product for sucrase activity appears mauve to orange after incubation of the cryostat sections using 2-naphthyl-α-D-glucopyranoside as the substrate. Different superscripts in Figure 5 relate to staining intensities between treatment groups at predetermined distances to represent the base of the villus (50 and 100 µm), the midvillus region (250 and 400 µm), and the villus tip (500 and 600 µm). Statistically significance was accepted at p < 0.05, as detected by ANOVA.

Full size image

Effects of orogastric IGF-I and LR3IGF-I on disaccharidase activities. For both age groups, orogastric administration of 12.5 µg g-1 d-1 IGF-I or 5 µg g-1 day-1 LR3IGF-I did not stimulate sucrase- and lactase-specific activity. As shown in Figure 4, total lactase activity was similar in all treatment groups, ranging between 87 and 110 U/cm and 42 and 59 U/cm for the younger and older rat pups, respectively. Sucrase activity was minimal in the younger rat pups (d 12 pp), increasing to approximately 12 U/cm in the older rat pups (d 18 pp). No statistically significant difference in total or specific enzyme activity was detected between treatment groups for both age groups (Fig. 4).

DISCUSSION

The growth-promoting action of IGF-I on tissues of the gastrointestinal tract is well established, and several studies have reported growth responses after systemic infusion of IGF-I and comparable analogs of IGF-I (22,37–41). In a recent publication we have reported that infusion of pharmacologic doses of IGF-I or a more potent analog, LR3IGF-I, markedly stimulates visceral and gastrointestinal growth, throughout postnatal development in suckling rat pups (21). We now show that continuous systemic infusion of LR3IGF-I stimulates the maturation pattern of intestinal disaccharidases, at doses that induce marked mucosal growth. In both age groups examined, systemic administration of pharmacologic doses of IGF-I increased disaccharidase activity proportional to increments of mucosal mass. Furthermore, the more potent analog, LR3IGF-I, was able to accelerate the reciprocal switch from lactase to sucrase as the major brush-border disaccharidase in the older rat pups near weaning but not the younger rat pups.

One other study that reported the maturation of digestive enzymes in suckling rats after systemic administration of IGF-I was conducted by Young et al. (25). These authors found an increase in specific enzyme in the jejunum for maltase, lactase, alkaline phosphatase, and aminopeptidase after both oral or intraperitoneal administration of 1 µg/d IGF-I for a period of 6 d between d 10 and 16 pp. In contrast to the present study, they did not observe a change in specific enzyme activities in the duodenum or ileum, nor was sucrase activity induced at these concentrations. Although it is difficult to directly compare data as the developmental time points are somewhat different, our studies confirm original data by Young et al. (25) in that sucrase activity was not induced during the earlier suckling period between d 13 and 16 pp. However, the present study showed that infusion of the highest dose of LR3IGF-I resulted in the induction of sucrase activity in the older rat pups, closer to weaning (d 19 pp). It appears therefore that the predetermined genetic program that specifies the temporal shift of disaccharidase expression was not overridden by IGF-I administration as indicated by the failure of IGF-I administration to induce the switch from lactase to sucrase activity in the earlier developmental period. Furthermore, in the older rat pups, systemic infusion of LR3IGF-I changed the disaccharidase activities to levels that would be expected normally some 2-3 d later in untreated weaning rat pups, indicating that no aberrant pattern of enzyme expression was induced by LR3IGF-I. Thus the ability of IGF-I peptides to stimulate disaccharidase maturation may confer a selective advantage in conditions where digestive capacity is compromised due to inadequate levels of enzyme expression.

There are several possibilities that may explain these observations. The peptide dose used in the present study was significantly greater, ranging from 30 to 175 µg/d in the younger rats and 50 to 320 µg/d of infusate in the older rats, compared with a dose of 1 µg g-1 d-1 used in the study by Young et al. (25). In addition, in Young's study, IGF-I was administered by intraperitoneal injection rather that continuous s.c. infusion, which may also contribute to the observed differences. Furthermore, only LR3IGF-I, an analog with reduced affinity for IGFBPs, induced dramatic effects on sucrase, indicating that IGFBPs may limit effects of IGF-I on sucrase during postnatal development.

It is known that IGFBPs present in serum or expressed locally can limit the bioavailability of IGFs and their actions on target tissues (41) and may therefore explain the difference in potency seen between IGF-I and LR3IGF-I. At present, the precise expression of IGFBPs in the intestinal mucosa for the developmental periods described in the present study are not known; however, IGFBP-3 may emerge as a possible candidate. A study by Winesett et al. (42) has shown that IGFBP-3 mRNA is abundantly expressed in the jejunum of adult Sprague-Dawley rats. The IGFBP-3 mRNA was localized to cells mainly within the lamina propria throughout the entire villus length and adjacent to the crypt epithelium, suggesting that IGFBP-3 may limit the action of IGF-I in the gastrointestinal tract. Although the in situ analysis did not permit the precise identification of the cell types expressing IGFBP-3 mRNA, subepithelial fibroblasts are potential candidates (43). Consistent with this possibility, histochemical analysis in the present study showed that sucrase activity was increased in cells at the base of the villus in rat pups infused with LR3IGF-I, where expression of IGFBP-3 mRNA was also far less pronounced (42). This hypothesis is further supported by observations that in conditions of adaptive hyperplasia, such as after massive bowel resection (44) or in the regenerative phase after fasting (42), IGFBP-3 is greatly reduced, thus probably increasing the bioavailability of circulating or locally produced IGF-I for interaction with the receptor to stimulate intestinal growth and possible enzyme maturation.

Changes in the circulating IGFBP profile after infusion of the growth factors may also explain the observed differences in disaccharide stimulation between IGF-I and LR3IGF-I. As reported previously (21), infusion of IGF-I and particularly LR3IGF-I between d 12 and 19 pp significantly alters the plasma IGFBP profile, increasing specifically the expression of IGFBP-2 or -1 and IGFBP-3. In addition, circulating GH levels in rodents increase at the time of weaning (d 19-23 pp), and GH-dependent increases of circulating IGFBP-3 are also observed, which may further contribute to the differential effects of IGF-I and LR3IGF-I on the observed changes in disaccharidase expression.

In rodents, glucocorticoids have been convincingly postulated as dual regulators responsible, in part, for the postnatal decline in lactase activity and concomitant increase in sucrase activity (2–5). For instance, administration of glucocorticoids in the third postnatal week leads to a reduction in lactase activity, whereas glucocorticoid administration in the first and second week of postnatal life increases specific lactase activities (45,46). Precocious induction of sucrase activity by corticosterone, the principle glucocorticoid in rodents, has been clearly demonstrated, with greater effectiveness when administered by continuous infusion instead of daily intraperitoneal injections (47). Evidence that glucocorticoids affect enzyme maturation directly stems from observations that effects of glucocorticoids can be demonstrated in intestinal explants and glucocorticoids receptors are present in the intestine at substantially greater levels during the first two postnatal weeks (3).

Although stimulation of disaccharidase activity in intestinal explants by IGF-I have not been demonstrated so far, mitogenic and differentiation response in epithelial cell lines have been reported (48). In addition, receptors for IGF-I have been localized along the gastrointestinal tract in different species, including the rat (49,50), and more importantly, receptor densities are severalfold greater in mucosal membranes from suckling rats compared with adult rat tissues (25,50). This suggests direct action of IGF-I/LR3IGF-I on the intestinal mucosa to stimulate enzyme maturation. Nevertheless, the possibility that IGF-I or LR3IGF-I affects glucocorticoid levels, or direct action of the peptides on the adrenal glands, cannot be excluded. At present there are no data to support this possibility, but it may be speculated that compared with IGF-I, LR3IGF-I, due to the more potent growth effects, selectively stimulated adrenal growth, leading to increased glucocorticoid secretion and hence induction of sucrase activity. However, to clearly dissect glucocorticoid-mediated effects of IGF-I/LR3IGF-I on enzyme maturation from direct actions of the growth factors warrants investigation in the future. Hence, a logical extension of the present study may include a group of preweaning, adrenalectomized rat pups or animals subjected to a pharmacologic adrenal blockage to address if IGF-I effects on digestive enzyme maturation are mediated via the adrenals.

Regulation of disaccharidase activities by insulin has also been reported (12), but it is unlikely to account for the precocious induction of sucrase activity observed in the present study, as insulin levels in peptide-infused rat pups did not differ at the time of sacrifice from levels in vehicle-treated or untreated litter mates (21). In all, the data presented in this study provide clear evidence for a preferential effect of LR3IGF-I compared with IGF-I on neonatal small intestinal enzyme maturation, when delivered systemically. This is relevant to the use of IGF-I or analogs of IGF-I with low IGFBP affinities as therapeutic agents to promote enhanced enzyme maturation of the infant small intestine.

In addition to the importance of systemically derived IGF-I, milk may provide an alternative source of IGF-I for the intestinal mucosa. The ability of enternal IGF-I to stimulate gut maturation is also of interest with regard to the intervention for preterm neonates or neonates with compromised gut function. The few studies conducted so far have produced conflicting results. Young and colleagues (25) found that orogastric gavage of 1 µg g-1 d-1 IGF-I delivered for 6 d to normal rat pups between d 10 and 16 pp significantly stimulated enzyme-specific activities for lactase, maltase, alkaline phosphatase, and aminopeptidase in jejunum, but not in the duodenum or ileum. Similarly, Ma and Xu (26) have demonstrated that orogastric gavage of 1 µg g-1 d-1 IGF-I delivered to newborn rat pups for 3 d enhanced the protein content of the proximal small bowel and the enzyme specific activities of several brush-border enzymes, including lactase, sucrase, maltase and aminopeptidase. Both studies failed to detect growth-promoting activities of non-gut organs after orogastric delivery of IGF-I. More recently, Houle et al. (20) have shown that orogastric supplementation of 0.2 µg g-1 d-1 IGF-I significantly increased enzyme-specific activities for lactase and sucrase in the proximal regions of the small bowel in newborn piglets born by cesarean section. In their study, increased disaccharidase activity was associated with histomorphometric parameters, but no detectable effects on whole body weight gain or peripheral organ growth were reported. This is in contrast to studies by Burrin et al. (22) who have reported that orogastric administration of 3.5 µg g-1 d-1 IGF-I to newborn unsuckled piglets for 4 d significantly increases intestinal weight and protein and DNA content, but not intestinal length. These authors did not examine the effect of orogastric IGF-I on enzyme maturation. Our own study indicates that orogastric instillation of pharmacologic doses of IGF-I or LR3IGF-I, administered to maternally reared, normal suckling rat pups, between d 6 and 18 pp does not stimulate gut growth or enzyme maturation. In agreement with our results, a recent study by Fholenhag et al. (51) showed that s.c. and not oral administration of 2 µg g-1 day-1 IGF-I for a 14-d period to weaned rats selectively stimulates gut weight and histomorphometric parameters of the duodenum. At present it is not clear if the differences observed in the studies described in this communication relates to differences in administration protocol, developmental age, dosage of IGF-I, and/or perhaps species differences.

Finally, although there is evidence that orogastrically absorbed IGF-I can cross the mucosal barrier, it appears that this does not significantly contribute to circulating IGF-I levels. For instance, up to 40% of orogastrically delivered IGF-I remains bioactive in the gastrointestinal tract in rat pups for at least 30 min postingestion (52). In a recent study, the majority of orally administered radiolabeled IGF-I in newborn piglets was detected in tissues of the gastrointestinal tract with only 5% of the infused radiolabeled peptide being limmunoprecipitable from serum (53). Similarly, Baumrucker and Blum (23) have shown that, in newborn calves infused with radiolabeled IGF-I, approximately 12% of the infused label is immunoreactive. In the present study, total plasma IGF-I levels were similar in rat pups with or without orogastrically infused IGF-I, and no growth effects on peripheral organs were observed (data not shown). Taken together, these observations and the fact that orogastrically administered IGF-I does not stimulate growth of peripheral organs suggests that, although the neonatal intestine may absorb small quantities of bioactive IGF-I when administered orally, its contribution to the circulating IGF-I pool appears to be negligible.

In summary, systemically administered LR3IGF-I accelerates the maturational changes of the small intestine that occur at the time of weaning in the rat. Systemically administered LR3IGF-I increases intestinal mass and length in the proximal small bowel and accelerates the natural decline in lactase activity and the switch to sucrase expression. Orogastrically delivered IGF-I peptides at pharmacologic doses do not affect intestinal growth or digestive enzyme maturation in suckling rat pups, indicating the importance of the route of IGF-I delivery. Nevertheless, efficacy of orogastrically delivered IGF-I may be achieved at doses reflecting IGF-I concentrations present in colostrum or milk.