Bile Acids Induce Ileal Damage During Experimental Necrotizing Enterocolitis

Development of Experimental NEC

This protocol was approved by the Animal Care and Use Committee of the University of Arizona (A-324801-95081). Sprague–Dawley rats (Charles River Labs, Pontage, MI) were collected by caesarian section 1 day before scheduled birth. Pups were hand fed with cow’s milk– based rat milk substitute formula (NEC, n = 20),²⁶ and stressed twice daily with asphyxia (breathing 100% nitrogen gas for 60 seconds) followed by cold (4°C for 10 minutes) to induce experimental NEC. For comparison, dam-fed (DF, n = 25) littermates also were asphyxia and cold stressed. By using this method, 70%–80% of the formula-fed rats developed NEC vs 0% of the DF rats.^8,27 After 96 hours, all surviving animals were killed via decapitation.^8,27,28

Epidermal Growth Factor Treatment

Rats collected via caesarian section 1 day before scheduled birth were hand fed with cow’s milk– based rat milk substitute formula supplemented with 500 ng/mL epidermal growth factor (EGF) (NEC + EGF, n = 20). In previous publications, we have found that the incidence of NEC was reduced by approximately 45%–55% in pups supplemented with EGF.^8,28

BA Gavage

Two separate protocols for introducing exogenous BA were used. In the first, bile acid gavage– high (BAG-high), 2-day-old rats were fasted for 18 hours before a single gavaged dose of 50 mmol/L sodium deoxycholic acid (NaDOC). Two hours after gavage, ileal tissue was removed and processed for histology and RNA extraction. To induce full necrosis, a second bolus of 50 mmol/L NaDOC was gavaged 1 hour after the first in a separate group of animals. In the second protocol, newborn rats were given 20 mmol/L NaDOC twice per day for 2 days (BAG-low). Control rats were gavaged with vehicle (phosphate-buffered saline). Pups were dam fed during the 2-day study, but fasted for 8 hours before NaDOC gavage.

BA Sequestration

To examine the effects of BA sequestration on the development of NEC, animals were hand fed with cow’s milk–based rat milk substitute formula, asphyxia and cold stressed twice per day, and gavaged with either 120 mg/kg/day cholestyramine (Eon Labs, Laurelton, NY) or vehicle (phosphate-buffered saline) for 4 days. These groups were designated NEC + Chol (n = 15) and NEC (n = 16), respectively.

NEC Evaluation

Pathologic changes in intestinal architecture were evaluated using our previously published NEC scoring system.^8,11,27–29 Histologic changes in the ileum were scored by a blinded evaluator and graded as follows: 0 (normal), no damage; 1 (mild), slight submucosal and/or lamina propria separation; 2 (moderate), moderate separation of submucosa and/or lamina propria, and/or edema in submucosal and muscular layers; 3 (severe), severe separation of submucosa and/or lamina propria, and/or severe edema in submucosa and muscular layers, region villous sloughing; 4 (necrosis), loss of villi and necrosis. Intermediate scores of .5, 1.5, 2.5, and 3.5 also were used to assess more accurately the levels of ileal damage when necessary.^8,11,27–29 To determine the incidence of NEC, animals with histologic scores of less than 2 have not developed NEC and animals with histologic scores of 2 or greater have developed NEC.^8,11,27–29

Total BA Levels

Intestinal luminal BA levels were determined by flushing a 3-cm section of either distal ileum or proximal jejunum with .4 mL sterile cold phosphate-buffered saline.¹¹ Samples were frozen at −70°C until assayed. Portal blood was removed from the portal vein of anesthetized animals before death. Because of the technical difficulty, usable quantities of portal blood could not be obtained from all animals. Trunk (systemic) blood was collected after decapitation. Serum from portal and systemic blood was collected and frozen at −70°C until assayed. Total BA levels were determined from ileal flushes and from portal and systemic blood using the Total Bile Acids Assay Kit (Diazyme, San Diego, CA) according to the manufacturer’s protocol.

BA Composition

BA composition from ileal luminal contents was determined by the Arizona Cancer Center Analytical Core Shared Service (Tucson, AZ) using a TSQ Quantum triple quadruple LC/MS/MS system with a Surveyor MS pump and autosampler (Thermo Finnigan, San Jose, CA). A mass spectrometer was run in negative ion mode using electrospray ionization. Detection was through selective ion monitoring. Chromatographic separation was obtained using a Beckman Ultrasphere-XL C-18 column (Beckman, Palo Alto, CA). Isocratic solvent flow was used, consisting of 75% methanol and 25% ammonium acetate at a flow rate of .3 mL/min. Cholic acid, deoxycholic acid, lithocholic acid, chenodeoxycholic acid, and ursodeoxycholic acid levels were monitored.

RNA Preparation

Total RNA was isolated from ileal tissue using the RNeasy Mini Kit (Qiagen, Santa Clarita, CA) as described in the manufacturer’s protocol. All samples were incubated with RNase-free DNase (20 U per reaction) for 10 minutes at 37°C to eliminate DNA contamination. RNA concentration was quantified by ultraviolet spectrophotometry at 260 nm and the purity was determined by the A₂₆₀/A₂₈₀ ratio (SPECTRAmax PLUS; Molecular Devices, Sunnyvale, CA). The integrity of RNA was verified by electrophoresis on a 1.2% agarose gel containing formaldehyde (2.2 mol/L) and ethidium bromide in 1 × 3-(N-morpholino)propanesulfonic acid buffer (40 mmol/L 3-(N-morpholino)propanesulfonic acid, pH 7.0, 10 mmol/L sodium acetate, and 1 mmol/L ethylenediaminetetraacetic acid, pH 8.0). Only tissue from animals without full necrosis were extracted and analyzed.

Reverse-Transcription and Real-Time Polymerase Chain Reaction

Real-time polymerase chain reaction assays were performed specifically to quantify steady-state messenger RNA (mRNA) levels. Single-stranded complementary DNA was reverse-transcribed from 1 µg of total RNA in a 10 µL reaction mixture, as previously described in detail.³⁰ The amounts of total RNA used in the reverse-transcribed reactions were calculated from the absorbency at 260 nm, and verified by densitometry of the 28S ribosomal RNA band separated on denaturing agarose gels (by Gel Doc 1000 Documentation System with Molecular Analyst/PC software; Bio-Rad, Hercules, CA). Real-time polymerase chain reaction amplification^31,32 was performed using Assay By Design (Applied Biosystems, Foster City, CA) sequenced primers and probes for rat ASBT, IBABP, and OstαOstβ. ASBT primer, forward probe, and reverse probe sequences: TGGACCTCAGTGTTAGC, GGCTCCAATATCCTGGCCTAT, and AGCAAGCAGTGTGGAGCAAGT. IBABAP primer, forward probe, and reverse probe sequences: CACGGTTGCCTTGAA, CAAAGAATGTGAAATGCAGACCAT, and ACCTTGCCACCCTCCATCTT. Ostα primer, forward probe, and reverse probe sequences: CTGCCCACCCCTCATAC, CTGCTGCTGCTGTCCCT, and GCAACTGTAGCTTCTTCCTGGTAA. Ostβ primer, forward probe, and reverse probe sequences: TGGCTCCAATATCCTGGCCTAT, TATTCCATCCTGGTT, and AAGAAGCAGCCACAAGACAACG. Predeveloped TaqMan primers and probe were used for detection of farnesoid X receptor (FXR) (Applied Biosystems). Samples were subjected to 40 cycles of amplification at 95° for 15 seconds followed by 1 minute at 60° using a GeneAmp 5700 Sequence Detection System (Applied Biosystems) according to the manufacturer’s instructions. Water controls were included to ensure specificity. All mRNA levels were normalized according to the level of 18S within each sample, which was used as an endogenous control to ensure the accuracy and integrity of the tissue samples. Cleavage of the sequence-specific probe by Taq polymerase created an increase in fluorescent signal, which was observed during the exponential phase of amplification and allowed the determination of a threshold value for all samples. This threshold value, once normalized, was expressed as a fold change of gene expression relative to control samples.^11,27

Immunohistology

A portion of ileal tissue was removed and fixed overnight in 70% ethanol, paraffin embedded, and serial sectioned at 4–6 µm. Briefly, after deparaffinizing, sections were blocked with 1.5% appropriate serum (Vector Laboratories, Burlingame, CA), incubated with either anti-rat ASBT (kindly provided by Dr Stelzner, University of California at Los Angeles, Los Angeles, CA), anti-rat IBABP (kindly provided by Dr Crossman, Children’s Hospital Medical Center, Cincinnati, OH), anti-rat Ostα or anti-rat Ostβ followed by biotinylated secondary antibody (Vector Laboratories), Vectastain Elite ABC reagent (Vector Laboratories), diaminobenzidine, and counterstained with hematoxylin. Ost antibodies were raised in rabbits using synthetic peptides corresponding to amino acids 315–329 of rat Ostα (MYYRKKDNKVGYEAC), and 90–103 of rat Ostβ (ILRETLISEKADLA). Peptides were coupled to keyhole limpet hemocyanin and used to immunize New Zealand white rabbits (AnaSpec, San Jose, CA). Each of the antibodies was affinity purified using their respective peptide antigen (AnaSpec). No immunostaining was observed in the controls. Sections from experimental groups were stained for a specific primary antibody at the same time so that comparisons between groups could be assessed.

Western Blot

Individual frozen ileum samples were homogenized with a hand-held homogenizer (Pellet Pestle; Kimble/Kontes, Vineland, NJ) in a 5× volume of ice-cold homogenization buffer (50 mmol/L Tris HCl, pH 7.4, 150 mmol/L NaCl, 1 mmol/L ethylenediaminetetraacetic acid, .1% sodium dodecyl sulfate, 1% Na-deoxycholic acid, 1% Triton X-100, 50 mmol/L dithiothreitol, 50 µg/mL aprotinin, 50 µg/mL leupeptin, and 5 mmol/L phenylmethylsulfonyl fluoride). The homogenates were centrifuged at 10,000 rpm for 5 minutes at 4°C and the supernatant was collected. The total protein concentration was quantified using the Bradford protein assay. ⁴ For protein analysis, 50 g of protein was added to an equal volume of 2× Laemmli sample buffer and boiled for 5 minutes. The samples were run on a 10%–20% gradient polyacrylamide gel (Bio-Rad) at 95 V for 1 hour. Protein was transferred to Immuno-Blot polyvinylidene difluoride membranes (Bio-Rad) at 15 V for 1 hour. Membranes were blocked with 5% nonfat milk in Tris-buffered saline with .1% Tween 20 (Sigma, St. Louis, MO) for 1 hour at room temperature and then incubated with anti-ASBT, anti-IBABP, anti-Ostα, or anti-Ostβ antibody overnight at 4°C. After being washed, the membranes were incubated for 1 hour at room temperature with the appropriate horseradish-peroxidase–conjugated secondary antibody. Proteins were visualized with a chemiluminescent system (Pierce, Rockford, IL) and exposed to radiographic film. Membranes were stripped and probed with anti–β-actin to determine equal loading of protein. Densitometry was performed to compare protein expression between groups using QuantityOne software (Bio-Rad).

Statistics

Statistical analyses between groups were performed using analysis of variance followed by Fisher protected least significant difference. Correlation analyses were performed using Spearman rank correlation. Analysis of NEC score was performed using the Mann–Whitney test for nonparametric values. The χ² test was used to analyze differences in disease incidence. All statistical analyses were determined using the statistical program StatView (Abacus Concepts, Berkeley, CA). All numeric data are expressed as mean ± SEM.

Increased BAs in the Luminal Contents of Rats With NEC

To determine if BAs play a role in NEC pathogenesis, we evaluated total BA levels in ileal luminal flushes from DF and NEC groups. Total BA levels were increased significantly in the ileal luminal contents of neonatal rats with NEC (Figure 1). Serum BA levels were similar between groups (data not shown). Animals with NEC had significantly higher levels of cholic acid compared with control and DF animals, deoxycholic acid was detected only in the NEC group (Figure 2). No other toxic BAs were detected in either group.

Ileal/Jejunal and Portal/Ileal BA Ratios in NEC

Levels of ileal BAs can be affected by both the amount of BA produced by the liver and the transport of ileal luminal BAs out of the ileal lumen via enterohepatic circulation. Because direct cannulation of the bile duct is not possible in 4-day-old rats, we evaluated total BA levels in the proximal jejunal contents. The overall levels of jejunal BAs were similar between DF animals and those with NEC. However, the ileal/jejunal total BA ratio from the NEC group was significantly higher than the DF group (Table 1). Under normal circumstances, increased ileal BA levels should correspond to increased portal blood BA levels in an effort to clear excess BA from the ileum. However, animals with NEC had significantly decreased portal/luminal BA ratios compared with DF controls despite their increased ileal luminal BA levels (Table 1).

Sequestration of Ileal BAs Decreases the Incidence and Severity of NEC

If intestinal BAs are required for the development of NEC injury, then reduction of BAs should result in decreased incidence and severity of NEC. Chol binds BAs in the intestine and the Chol–BA complex passes out of the body without being absorbed.^33–35 In a separate set of experiments, neonatal rats artificially were fed cow’s milk formula with or without Chol (NEC + Chol and NEC, respectively) and asphyxia/cold stressed for 4 days. Results are summarized in Table 2. Pharmacologic sequestration of intestinal BAs significantly decreased NEC injury in the terminal ileum and increased survival in the neonatal rat model of NEC.

The accumulation of BAs is known to produce intestinal damage in adult systems. We introduced exogenous BAs to neonatal rats via intragastric gavage to determine if increased levels of ileal BAs would result in ileal damage similar to what is seen in the neonatal rat model of NEC. Neonatal rats given high-dose NaDOC (BAG-high) showed histologic damage to the ileal architecture similar to what is observed in animals with NEC (Figure 3). Animals given lower doses of NaDOC over a 2-day period (BAG-low) did not show significant ileal damage (data not shown).

To determine further if BA levels or composition are reduced when disease is reduced, we examined neonatal rats treated for 4 days with EGF. We previously showed that enteral administration of EGF significantly decreased the incidence and severity of NEC in the neonatal rat model.⁸ Total BA levels and the levels of both cholic acid and deoxycholic acid were reduced significantly after treatment with EGF (Table 3).

Reduction of NEC and BAs

Luminal total BA levels were correlated positively (r = .744, P ≤ .001) with progression of ileal injury, as determined via histologic NEC score (Figure 4).

Ileal BA Transporter mRNA Expression in NEC

The transport of BAs from the ileal lumen to portal circulation is crucial for BA homeostasis. We first evaluated mRNA levels from the BA transporters ASBT, IBABP, Ostα, and Ostβ in the distal ileum during the development of experimental NEC. ASBT mRNA was up-regulated in NEC ileal tissue compared with DF controls. In contrast, there was no change in IBABP mRNA expression between the DF and NEC groups. To determine if ileal BA sequestration would decrease the expression of ASBT, we found ASBT mRNA was decreased in the NEC + Chol group after BA sequestration. In contrast, BAG-low increased the expression of IBABP, but not ASBT. Further, ASBT mRNA was decreased significantly and IBABP mRNA was increased significantly in the NEC + EGF group compared with the NEC group. The mRNA levels for Ostβ were low and without statistical differences in all groups (Figure 5). No significant changes were observed for Ostα between groups (data not shown).

Ileal BA Transporter Protein in NEC

We evaluated ASBT, IBABP, Ostα, and Ostβ protein levels from the distal ileum during the development of experimental NEC. ASBT was not present in most 4-day-old DF neonates. In the NEC group, ASBT was localized on the apical membrane, and EGF treatment significantly reduced the increased ASBT seen in the NEC group (Figure 6A). To quantify changes in ASBT further, we evaluated ileal protein levels using Western blot analysis. Protein extracted from ileal tissue from a 28-day-old weanling rat, diluted 1:10, was run on the same gel for comparison. ASBT was detected in all groups, however, levels in the NEC group were higher than in the DF or NEC + EGF groups (Figure 6B). In addition, ASBT protein levels were decreased significantly in the NEC + Chol group compared with the NEC group (Figure 6C).

In contrast to ASBT, immunohistologic evaluation showed that IBABP is localized in the cytoplasm of ileal enterocytes in the DF and NEC + EGF groups only. IBABP is observed rarely in ileal tissue from the NEC group (Figure 7A). To confirm that IBABP is produced only minimally in animals with NEC, we evaluated ileal protein levels of IBABP using Western blot (Figure 7B). Protein extracted from ileal tissue from a 28-day-old weanling rat, diluted 1:100, was run on the same gel for comparison. IBABP is produced in both DF and NEC + EGF animals, albeit at significantly lower levels than in weanling rats, but is not detectable in most of the animals in the NEC group. IBABP protein levels were decreased in the NEC + Chol group compared with the DF group and increased in the BAG group compared with the NEC group, but not significantly different from the DF animals (Figure 7C).

Ostα and Ostβ protein localization and quantification also were examined. Immunohistologic staining and Western blot showed very low levels of Ostα in all neonatal rat groups (data not shown). For Ostβ, similar localization and intensity of staining was observed between DF, NEC, and NEC + EGF groups. However, in neonatal rats, the localization of Ostβ was found primarily in the crypts and not in the upper villi as seen in weanling animals (Figure 8A). Western blotting showed variability, particularly among DF animals, but confirmed that there are no significant differences in Ostβ protein between the NEC and NEC + EGF groups (Figure 8B). Ostβ protein levels were decreased in the NEC + Chol group compared with the NEC group, and increased in the BAG group compared with DF animals (Figure 8C).

BA responsiveness can be regulated by positive and negative feedback mechanisms via interactions between BAs and the FXR.^36,37 Because BA transporters can be regulated by the FXR, we examined the FXR mRNA levels in all experimental groups. There were no differences between the neonatal groups, however, the FXR expression in DF, NEC, NEC + EGF, NEC + Chol, and BAG-low groups was significantly lower than what was observed in weanling rats (Figure 9).