Intrauterine Growth Restriction Alters Mouse Intestinal Architecture during Development

Abstract

Infants with intrauterine growth restriction (IUGR) are at increased risk for neonatal and lifelong morbidities affecting multiple organ systems including the intestinal tract. The underlying mechanisms for the risk to the intestine remain poorly understood. In this study, we tested the hypothesis that IUGR affects the development of goblet and Paneth cell lineages, thus compromising the innate immunity and barrier functions of the epithelium. Using a mouse model of maternal thromboxane A2-analog infusion to elicit maternal hypertension and resultant IUGR, we tested whether IUGR alters ileal maturation and specifically disrupts mucus-producing goblet and antimicrobial-secreting Paneth cell development. We measured body weights, ileal weights and ileal lengths from birth to postnatal day (P) 56. We also determined the abundance of goblet and Paneth cells and their mRNA products, localization of cellular tight junctions, cell proliferation, and apoptosis to interrogate cellular homeostasis. Comparison of the murine findings with human IUGR ileum allowed us to verify observed changes in the mouse were relevant to clinical IUGR. At P14 IUGR mice had decreased ileal lengths, fewer goblet and Paneth cells, reductions in Paneth cell specific mRNAs, and decreased cell proliferation. These findings positively correlated with severity of IUGR. Furthermore, the decrease in murine Paneth cells was also seen in human IUGR ileum. IUGR disrupts the normal trajectory of ileal development, particularly affecting the composition and secretory products of the epithelial surface of the intestine. We speculate that this abnormal intestinal development may constitute an inherent "first hit", rendering IUGR intestine susceptible to further injury, infection, or inflammation.

Fig 1. Maternal exposure to U-46619 decreases fetal body weight at E19 compared to sham pups.

Pregnant mice were exposed to a TXA₂- analog (U-46619) or vehicle via micro-osmotic pump infusion from E12.5 until birth (term ~20 days). Body weights of exposed fetuses remained similar to shams at E15.5 and E17.5, but were statistically decreased by E19 (p < 0.001: e15.5 Sham n = 13, e15.5 TXA n = 21, e17.5 Sham n = 8, e17.5 TXA n = 11, e19 Sham n = 5, e19 TXA n = 5).

Fig 2. IUGR pups have decreased body weight and ileal length compared to shams.

Pregnant mice were exposed to a TXA₂-analog or vehicle via micro-osmotic pump infusion, and pups were delivered and raised by cross-foster dams. (A) Representative picture of a weaned P21 sham and IUGR mouse to illustrate size differences. (B) Body weights of IUGR pups were significantly less than sham pups from birth to P56 (p < 0.001: P0 Sham n = 15, P0 IUGR n = 12, P14 Sham n = 14, P14 IUGR n = 6, P28 Sham n = 12, P28 IUGR n = 8, P56 Sham n = 7, P56 IUGR n = 7). (C) Harvested ileal weights were similar in both sham and IUGR groups, but (D) ileal lengths were significantly shorter in P14 IUGR pups compared to sham pups (p = 0001, n for each treatment group was the same as Fig 2B).

Fig 3. IUGR pups have reduced goblet cell counts at age P14 compared to sham controls.

(A) Mucus-positive goblet cells were quantified per 100 epithelial cells at various ages in IUGR and sham mice (P0 Sham n = 13, P0 IUGR n = 11, P14 Sham n = 14, P14 IUGR n = 6, P28 Sham n = 12, P28 IUGR n = 8, P56 Sham n = 5, P56 IUGR n = 7). IUGR mice at P14 had significantly fewer goblet cell/epithelial cell ratios compared to sham mice (p = 0.001). (B) Despite differences in mucus-positive goblet cell numbers at P14, no differences were noted in MUC2 mRNA levels (p = 0.9: P0 Sham n = 4, P0 IUGR n = 4, P14 Sham n = 9, P14 IUGR n = 5, P28 Sham n = 10, P28 IUGR n = 6, P56 Sham n = 7, P56 IUGR n = 7). (C) Representative micrographs from P14 sham and IUGR mice taken with a 40x objective with goblet cells denoted by arrows.

Fig 4. IUGR pups have reduced Paneth cell counts gene products compared to sham controls.

(A) Granule-positive Paneth cells were quantified per crypt at different ages in IUGR and sham mice (P14 Sham n = 14, P14 IUGR n = 5, P28 Sham n = 12, P28 IUGR n = 6, P56 Sham n = 7, P56 IUGR n = 7). IUGR mice had significantly fewer Paneth cell counts at P14 (p = 0.01) and P28 (p = 0.002) compared to P14 shams. (B) Representative micrographs of P14 mice taken with a 40x objective showing Paneth cells as denoted by arrows in sham and IUGR. (C) IUGR mice also had significantly reduced lysozyme 1 (p<0.0001), cryptdin (p = 0.003), Ang4 (p = 0.017), and Reg3γ (p = 0.024) mRNA levels at P14. In contrast, cryptdin and Ang4 mRNA levels were significantly increased in P56 IUGR mice compared to sham mice (p = 0.02 and 0.003 respectively: P14 Sham n = 9, P14 IUGR n = 5, P28 Sham n = 7, P28 IUGR n = 6, P56 Sham n = 7, P56 IUGR n = 7).

Fig 5. IUGR pups have decreased cell proliferation at age P14 compared to sham controls.

(A) Actively proliferating cells were stained with anti-Ki67 antibody and the number of Ki-67 positive cells per intestinal crypt was quantified in IUGR and sham mice. IUGR decreased the number of Ki-67⁺ cells at P14 only compared to shams (n = 5 for each treatment group, p = 0.04, representative micrographs from P14 mice are shown as the lower panel of A). (B) Apoptotic cells were stained with anti-cleaved caspase-3 antibody and the number was quantified in IUGR and sham mice. IUGR did not affect the number of apoptotic events per 100 vili at any time point (n = 3 for each treatment, p = 0.6, representative micrographs are shown from P28 mice as the lower panel of B).

Fig 6. IUGR pups show altered mRNA levels compared to sham controls.

(A) Expression of mRNA was measured using qRT-PCR. TXA₂ analog-induced IUGR caused no significant change in mRNA levels of either occludin or ZO-1 at any age (Fig 6A, n = 4 for all treatment groups). (B) Immunohistochemistry of ileal samples showed no differences in quantity or localization of the tight junction protein ZO-1. ZO-1 staining is shown in red, and nuclei as marked by DAPI staining are shown in blue (Fig 6B, n = 3 for all treatment groups, samples from P28 sham and IUGR animals are shown). (C) TXA₂ analog-induced IUGR had no effect on mRNA levels of IL-6, IL-8, IL-1β, or TNF at P14, but caused significant decreases in IL-6 at P28 and P56, and in TNF at P56 (p = 0.02, 0.004, and 0.02 respectively, n = 4 for all treatment groups).

Fig 7. The severity of IUGR significantly impacts its effect on Paneth cells.

Data obtained from prior experiments were calculated as a percentage of change from the average control value of the same age. Data were plotted as deviation in weight compared to change in the desired variable and significance was determined using linear regression. There was a significant correlation between severity of IUGR and (A) Paneth cell loss (n = 29, p = 0.0009, r² = .34), (B) decrease in cryptdin mRNA expression (n = 22, p = 0.0041, r² = .34), (C) decrease in lysozyme mRNA (n = 15, p = 0.0005, r² = 0.62), (D) decrease in Wnt3 mRNA (n = 15, p = 0.0094, r² = 0.42), (E) decrease in Ang4 mRNA (n16, p = 0.012, r² = 0.38), (F) and decrease in INF-γ mRNA (n = 15, p = 0.0075, r² = 0.56). (F) While statistically non-significant, there were trends towards increased mRNA levels of TNF, IL-1β, IL-6, and IL-8, which are all pro-inflammatory cytokines seen in inflammatory intestinal diseases.

Fig 8. Human preterm infants diagnosed with IUGR have decreased Paneth cell counts compared to preterm non-IUGR control infants.

Tissues obtained from preterm infants with non-inflammatory surgical small intestinal diseases were stained for presence of goblet and Paneth cells. Infants with a diagnosis of IUGR (n = 9) were compared to age-matched non-IUGR infants (n = 4). (A) IUGR infants had similar numbers of goblet cells per 100 epithelial cells compared to non-IUGR infants (p = 0.5). (B) IUGR infants had decreased numbers of Paneth cells per crypt compared to non-IUGR infants (p = 0.04). (C) Shown is a representative micrograph of human tissue taken with a 20x objective and stained with Alcian Blue-PAS staining.