Changes in intestinal glucocorticoid sensitivity in early life shape the risk of epithelial barrier defect in maternal-deprived rats

Abstract

Glucocorticoids (GC) contribute to human intestine ontogeny and accelerate gut barrier development in preparation to birth. Rat gut is immature at birth, and high intestinal GC sensitivity during the first two weeks of life resembles that of premature infants. This makes suckling rats a model to investigate postpartum impact of maternal separation (MS)-associated GC release in preterm babies, and whether GC sensitivity may shape MS effects in immature gut. A 4 hours-MS applied once at postnatal day (PND)10 enhanced plasma corticosterone in male and female pups, increased by two times the total in vivo intestinal permeability (IP) to oral FITC-Dextran 4 kDa (FD4) immediately after the end of MS, and induced bacterial translocation (BT) to liver and spleen. Ussing chamber experiments demonstrated a 2-fold increase of permeability to FD4 in the colon immediately after the end of MS, but not in the ileum. Colonic permeability was not only increased for FD4 but also to intact horseradish peroxidase 44 kDa in MS pups. In vivo, the glucocorticoid receptor (GR) antagonist RU486 or ML7 blockade of myosin light chain kinase controlling epithelial cytoskeleton contraction prevented MS-induced IP increase to oral FD4 and BT. In addition, the GR agonist dexamethasone dose-dependently mimicked MS-increase of IP to oral FD4. In contrast, MS effects on IP to oral FD4 and BT were absent at PND20, a model for full-term infant, characterized by a marked drop of IP to FD4 in response to dexamethasone, and decreased GR expression in the colon only compared to PND10 pups. These results show that high intestinal GC responsiveness in a rat model of prematurity defines a vulnerable window for a post-delivery MS, evoking immediate disruption of epithelial integrity in the large intestine, and increasing susceptibility to macromolecule passage and bacteremia.

Figure 1. In vivo intestinal permeability to oral FD4 during development.

Data show the progressive decrease of IP to FITC-Dextran 4 kDa (FD4) every 10 days from the mild-lactation period (postnatal day (PND) 10) to adulthood (PND50). Data are expressed as the mean of plasma FITC-Dextran concentration (µg/ml)±SEM. Numbers of animals per group and sex: PND10 (n = 22–26), PND20 (27–28), PND30 (n = 12–16), PND40 (n = 10–12) and PND50 (n = 7–12).

Figure 2. Effect of a single MS for 4 h on IP and plasma CORT in PND10 and PND20 rats.

(A) Data show in vivo IP to FD4 immediately after MS (T4 h), and after the pups were returned to their dams (T8 h, T12 h and T24 h). Note that only PND10 rats displayed increased IP in response to MS procedure. Data are mean±SEM (7–10 animals per group). **P<0.01 ***P<0.001 compared to corresponding sham controls. (B) Blood samples from MS and sham pups were obtained in rats throughout the MS procedure lasting for 4 h, then every 4 h for 8 to 12 h after the pups were returned to their dams. Note that basal plasma CORT levels in male and female PND10 were lower than in their PND20 counterparts. In both PND10 and PND20 rats, circulating CORT increased in MS rats soon after they were removed from their mother, and peaked at 4 h. Data are expressed as the mean±SEM in 3–10 pups per time-point. **P<0.01***P<0.001 compared to basal.

Figure 3. Effect of a single MS on colonic and ileal permeability to FD4 and intact HRP in PND10 rats.

Ussing chambers measurements of mucosal-to-serosal permeability to Dextran 4 kDa and HRP 44 kDa in (A) colonic and (B) ileal segments of PND10 pups immediately after MS (T4 h). Note that MS increased FD4 and intact HRP permeability in the colon, but not in the ileum. Pooled data of both genders are shown and are expressed as the mean of permeability to FD4 (cm/s x10⁻⁶)±SEM in 9–16 animals per group, and HRP (cm/s × 10⁻⁷)±SEM in 6–17 animals per group. *P<0.05 **P<0.01 compared to corresponding sham controls.

Figure 4. Effects of RU486 and ML7 on MS-induced increase of IP to FD4 in PDN10 rats.

Treatment with (A) ML7 (1 mg/kg/d in 0.9% NaCl i.p. at 24, 12 and 1 h before IP measurement), and (B) the GR antagonist RU486 (2 mg/kg/d in olive oil s.c. at 12 and 1 h before IP measurement) prior to MS prevented the IP increase in response to MS. Values are mean±SEM (n = 7–17 and n = 3–7 pups per group for experiment A and B respectively), and *P<0.05 **P<0.01, ***P<0.001 compared to their respective controls.

Figure 5. Dose-response study of DEX on IP to FD4 and GR expression in the small intestine and the colon.

(A) qPCR results for GR mRNA using total RNA from colon and ileum lysates of PND10 and PND20 female rat pups. (B) Total in vivo IP to Dextran 4 kDa (FD4) was measured following subcutaneous injections of Dex (0.01 mg to 2.5 mg (kg BW)⁻¹ for 1 day) at PND10 or PND20. In both sexes, note the decreased sensitivity to GC receptor stimulation evoking IP increase to FD4 at PND20 compared to PND10. Data are expressed as the mean±SEM (n = 3–10 pups per group).