Myelomeningocele (MMC), or spina bifida, results from incomplete closure of the spinal cord and spinal canal during gestation, leaving the spinal cord unprotected. Intrauterine chemical and mechanical trauma to the exposed spinal cord results in lifelong lower limb paralysis, bowel and bladder dysfunction, musculoskeletal deformities, and severe cognitive disabilities due to hindbrain herniation in afflicted children. It is the most common congenital cause of lifelong paralysis in the United States, and approximately four children a day are born with this devastating congenital defect. 

Circulating miRNAs the in blood are promising biomarkers for predicting pregnancy complications and adverse birth outcomes. Previous work identified 11 gestationally elevated maternal circulating miRNAs (HEamiRNAs) that predicted infant growth deficits following prenatal alcohol exposure and regulated epithelial-mesenchymal transition in the placenta. Here we show that a single intravascular administration of pooled murine-conserved HEamiRNAs to pregnant mice on gestational day 10 (GD10) attenuates umbilical cord blood flow during gestation, explaining the observed intrauterine growth restriction (IUGR), specifically decreased fetal weight, and morphometric indices of cranial growth. Moreover, RNAseq of the fetal portion of the placenta demonstrated that this single exposure has lasting transcriptomic changes, including upregulation of members of the Notch pathway (Dll4, Rfng, Hey1), which is a pathway important for trophoblast migration and differentiation. Weighted gene co-expression network analysis also identified chemokine signaling, which is responsible for regulating immune cell-mediated angiogenesis in the placenta, as an important predictor of fetal growth and head size. Our data suggest that HEamiRNAs perturb the expression of placental genes relevant for angiogenesis, resulting in impaired umbilical cord blood flow and subsequently, IUGR.

Purpose

The purpose of this study was to determine whether seeding density of placental mesenchymal stromal cells (PMSCs) on extracellular matrix (ECM) during in utero repair of myelomeningocele (MMC) affects motor function and neuronal preservation in the ovine model.

Methods

MMC defects were surgically created in 33 fetuses and repaired following randomization into four treatment groups: ECM only (n = 10), PMSC-ECM (42 K cells/cm²) (n = 8), PMSC-ECM (167 K cells/cm²) (n = 7), or PMSC-ECM (250-300 K cells/cm²) (n = 8). Motor function was evaluated using the Sheep Locomotor Rating Scale (SLR). Serial sections of the lumbar spinal cord were analyzed by measuring their cross-sectional areas which were then normalized to normal lambs. Large neurons (LN, diameter 30-70 μm) were counted manually and density calculated per mm² gray matter.

Results

Lambs treated with PMSCs at any density had a higher median SLR score (15 [IQR 13.5-15]) than ECM alone (6.5 [IQR 4-12.75], p = 0.036). Cross-sectional areas of spinal cord and gray matter were highest in the PMSC-ECM (167 K/cm²) group (p = 0.002 and 0.006, respectively). LN density was highest in the greatest density PMSC-ECM (250-300 K/cm²) group (p = 0.045) which positively correlated with SLR score (r = 0.807, p < 0.0001).

Conclusions

Fetal repair of myelomeningocele with high density PMSC-ECM resulted in increased large neuron density, which strongly correlated with improved motor function.

Type of study

Basic science.

Level of evidence

N/A.

Purpose

The purpose of this study was to determine whether seeding density of placental mesenchymal stromal cells (PMSCs) on extracellular matrix (ECM) during in utero repair of myelomeningocele (MMC) affects motor function and neuronal preservation in the ovine model.

Methods

MMC defects were surgically created in 33 fetuses and repaired following randomization into four treatment groups: ECM only (n = 10), PMSC-ECM (42 K cells/cm²) (n = 8), PMSC-ECM (167 K cells/cm²) (n = 7), or PMSC-ECM (250-300 K cells/cm²) (n = 8). Motor function was evaluated using the Sheep Locomotor Rating Scale (SLR). Serial sections of the lumbar spinal cord were analyzed by measuring their cross-sectional areas which were then normalized to normal lambs. Large neurons (LN, diameter 30-70 μm) were counted manually and density calculated per mm² gray matter.

Results

Lambs treated with PMSCs at any density had a higher median SLR score (15 [IQR 13.5-15]) than ECM alone (6.5 [IQR 4-12.75], p = 0.036). Cross-sectional areas of spinal cord and gray matter were highest in the PMSC-ECM (167 K/cm²) group (p = 0.002 and 0.006, respectively). LN density was highest in the greatest density PMSC-ECM (250-300 K/cm²) group (p = 0.045) which positively correlated with SLR score (r = 0.807, p < 0.0001).

Conclusions

Fetal repair of myelomeningocele with high density PMSC-ECM resulted in increased large neuron density, which strongly correlated with improved motor function.

Type of study

Basic science.

Level of evidence

N/A.

Background

We determined whether in vitro potency assays inform which placental mesenchymal stromal cell (PMSC) lines produce high rates of ambulation following in utero treatment of myelomeningocele in an ovine model.

Methods

PMSC lines were created following explant culture of three early-gestation human placentas. In vitro neuroprotection was assessed with a neuronal apoptosis model. In vivo, myelomeningocele defects were created in 28 fetuses and repaired with PMSCs at 3 × 10⁵ cells/cm² of scaffold from Line A (n = 6), Line B (n = 7) and Line C (n = 5) and compared to no PMSCs (n = 10). Ambulation was scored as ≥13 on the Sheep Locomotor Rating Scale.

Results

In vitro, Line A and B had higher neuroprotective capability than no PMSCs (1.7 and 1.8 respectively vs 1, p = 0.02, ANOVA). In vivo, Line A and B had higher large neuron densities than no PMSCs (25.2 and 27.9 respectively vs 4.8, p = 0.03, ANOVA). Line C did not have higher neuroprotection or larger neuron density than no PMSCs. In vivo, Line A and B had ambulation rates of 83% and 71%, respectively, compared to 60% with Line C and 20% with no PMSCs.

Conclusion

The in vitro neuroprotection assay will facilitate selection of optimal PMSC lines for clinical use.

Level of evidence

n/a.

Type of study

Basic science.

Background

Prenatal alcohol exposure (PAE) has been shown to alter fetal blood flow in utero and is also associated with placental insufficiency and intrauterine growth restriction (IUGR), suggesting an underlying connection between perturbed circulation and pregnancy outcomes.

Methods

Timed-pregnant C57/BL6NHsd mice, bred in-house, were exposed by gavage on gestational day 10 (GD10) to ethanol (3 g/kg) or purified water, as a control. Pulse-wave Doppler ultrasound measurements for umbilical arteries and ascending aorta were obtained post-gavage (GD12, GD14, GD18) on 2 fetuses/litter. RNA from the non-decidual (labyrinthine and junctional zone) portion of placentas was isolated and processed for RNA-seq and subsequent bioinformatic analyses, and the association between transcriptomic changes and fetal phenotypes assessed.

Results

Exposure to ethanol in pregnant mice on GD10 attenuates umbilical cord blood flow transiently during gestation, and is associated with indices of IUGR, specifically decreased fetal weight and morphometric indices of cranial growth. Moreover, RNA-seq of the fetal portion of the placenta demonstrated that this single exposure has lasting transcriptomic changes, including upregulation of Tet3, which is associated with spontaneous abortion. Weighted gene co-expression network analysis (WGCNA) identified erythrocyte differentiation and homeostasis as important pathways associated with improved umbilical cord blood flow as gestation progresses. WGCNA also identified sensory perception of chemical stimulus/odorant and receptor activity as important pathways associated with cranial growth.

Conclusion

Our data suggest that PAE perturbs the expression of placental genes relevant for placental hematopoiesis and environmental sensing, resulting in transient impairment of umbilical cord blood flow and, subsequently, IUGR.