Heritable spina bifida in sheep: A potential model for fetal repair of myelomeningocele

Abstract

Background/purpose: In 2004, a heritable occurrence of spina bifida was reported in sheep on a farm in the United States. We maintained and characterized the spina bifida phenotype in this flock to assess its potential as an alternative surgical model.

Methods: A breeding strategy was developed in which the sheep were crossed to maintain or increase the occurrence of spina bifida. Measurements and observations were recorded regarding lesion size, birthweight, ambulatory capacity, or urological function, and necropsies were performed on spina bifida afflicted lambs in conjunction with magnetic resonance imaging to determine the character of the spina bifida defects and assess the presence of Chiari-like malformations or hydrocephalus.

Results: The defects were observed to be more prevalent in ram lambs, and the rate of spina bifida per litter could be increased through backcrossing or by selection of a productive ewe breed. The lambs displayed a range of ambulatory and urological deficits which could be used to evaluate new fetal repair methodologies. Finally, affected lambs were shown to demonstrate severe Chiari malformations and hydrocephalus.

Conclusions: We have determined that use of these sheep as a natural source for spina bifida fetuses is feasible and could supplement the deficits of current sheep models for myelomeningocele repair.

Level of evidence: Level IV.

Keywords: Fetal sheep model; In utero surgery; Myelomeningocele; Spina bifida.

Figure 1.. Heritable Occurrence of Spina Bifida at the Bayliss Sheep Farm.

A) Representative cases of spina bifida in two lambs, 984-14 (left) and 472-09 (right). B) Distribution of spina bifida and lambs with tail defect by year during the course of this study (2001-2018). The sires used each year are numbered and represented in a graphical timeline at the base of the chart. C) Pie charts showing the gender distribution of all lambs used in the study (left, n=1356), lambs affected by spina bifida aperta (middle, n =111), and lambs affected with a tail defect (right, n = 335). D) A pie chart showing the distribution of the observed spinal level (lumbar, sacral, or lumbosacral) at which the spina bifida lesion occurred in 84 lambs.

Figure 2.. Characterization of Tail Defects in the Bayliss Sheep Population.

A) Distribution of tail defect types as observed in all affected lambs (left, n=313) and in lambs affected with spina bifida aperta (right, n=98). B) Distribution of tail lengths in normal lambs (n=957) compared to lambs with spina bifida aperta (n=104) and lambs that had tail defects but not spina bifida (n=215). C) Comparison of tail lengths by tail type: straight (n=964), crooked (n=118), curly (n=103), bunny (n=58), absent (n=31). *Significance determined by one-way ANOVA with Dunnett’s multiple comparisons correction, and error bars represent the standard deviation around the mean.

Figure 3.. Characterization of Survival Outcomes for Lambs with Spina Bifida.

A) Kaplan-Meier survival curves for unaffected lambs (n=969), tail defect lambs (n=221), and SBA affected lambs (n=82) during the first post-natal month. Stillborn or euthanized lambs were not included in the analysis. *Significance determined by a Logrank test. B) Scatter plot demonstrating a correlation between anterior-posterior length and lateral width for each spina bifida lesion (n=93). *Significance of correlation determined by the Pearson method. C) Comparison of lesion size, expressed as an approximate area, between SBA affected lambs that died (n=70) and those that survived beyond the first post-natal month (n=7). D) Comparison of lesion size, expressed as an approximate area, between SBA affected lambs that were paralyzed (n=53) and those that were ambulatory (n=22), as defined by the ability to stand. *For C and D, significance was determined by Welch’s t test, error bars represent the standard deviation around the mean.

Figure 4.. Lamb Necropsies, Histology, and Magnetic Resonance Imaging.

A) Posterior spinal cords from unaffected (left), tail defect (middle), and SBA affected lambs (right). B) Transverse sections through lumbar level spinal cords of unaffected (left) and SBA affected lambs (right) stained with hematoxylin and eosin. C) Neonate brains from unaffected (left) and SBA affected lambs (right) demonstrating a Chiari-like deformation of the cerebellum and posterior cerebrum. D) Transverse MRI images through post-natal brains of unaffected (above) and SBA affected lambs (below) demonstrating a hydrocephalus phenotype with cerebrospinal fluid accumulation in the lateral ventricles of affected individuals (red arrow). Transverse sections appear from most anterior (left) to posterior (right).