Postoperative Anti-Tgf-β2 Antibody Therapy Improves Intracranial Volume and Craniofacial Growth in Craniosynostotic Rabbits

The clinical manifestations, molecular mechanisms and treatment of craniosynostosis

Craniosynostosis is a major congenital craniofacial disorder characterized by the premature fusion of cranial suture(s). Patients with severe craniosynostosis often have impairments in hearing, vision, intracranial pressure and/or neurocognitive functions. Craniosynostosis can result from mutations, chromosomal abnormalities or adverse environmental effects, and can occur in isolation or in association with numerous syndromes. To date, surgical correction remains the primary treatment for craniosynostosis, but it is associated with complications and with the potential for re-synostosis. There is, therefore, a strong unmet need for new therapies. Here, we provide a comprehensive review of our current understanding of craniosynostosis, including typical craniosynostosis types, their clinical manifestations, cranial suture development, and genetic and environmental causes. Based on studies from animal models, we present a framework for understanding the pathogenesis of craniosynostosis, with an emphasis on the loss of postnatal suture mesenchymal stem cells as an emerging disease-driving mechanism. We evaluate emerging treatment options and highlight the potential of mesenchymal stem cell-based suture regeneration as a therapeutic approach for craniosynostosis.

Rescue of Premature Coronal Suture Fusion with TGF-β2 Neutralizing Antibody in Rabbits with Delayed-Onset Synostosis

OBJECTIVES

An overexpression of Tgf-β2 leads to calvarial hyperostosis and suture fusion in individuals with craniosynostosis. Inhibition of Tgf-β2 may help rescue fusing sutures and restore normal growth. The present study was designed to test this hypothesis.

DESIGN

Twenty-eight New Zealand White rabbits with delayed-onset coronal synostosis had radiopaque markers placed on either side of the coronal sutures at 10 days of age. The rabbits were randomly assigned to: (1) sham control rabbits (n = 10), (2) rabbits with control IgG (100 μg/suture) delivered in a collagen vehicle (n = 9), and (3) rabbits with Tgf-β2 neutralizing antibody (100 μg/suture) delivered in a collagen vehicle (n = 9). Longitudinal growth data were collected at 10, 25, 42, and 84 days of age. Sutures were harvested at 84 days of age for histomorphometry.

RESULTS

Radiographic analysis showed significantly greater ( P < .05) coronal suture marker separation, craniofacial length, cranial vault length, height, shape indices, cranial base length, and more lordotic cranial base angles in rabbits treated with anti-Tgf-β2 antibody than in controls at 42 and 84 days of age. Histologically, rabbits treated with anti-Tgf-β2 antibody at 84 days of age had patent and significantly ( P < .05) wider coronal sutures and greater sutural area compared to controls.

CONCLUSIONS

These data support our hypothesis that antagonism of Tgf-β2 may rescue fusing coronal sutures and facilitate craniofacial growth in this rabbit model. These findings also suggest that cytokine therapy may have clinical significance in infants with progressive postgestational craniosynostosis.

The influence of suturectomy on age-related changes in cerebral blood flow in rabbits with familial bicoronal suture craniosynostosis: A quantitative analysis

Background

Coronal suture synostosis is a condition which can have deleterious physical and cognitive sequelae in humans if not corrected. A well-established animal model has previously demonstrated disruptions in intracranial pressure and developmental abnormalities in rabbits with congenital craniosynostosis compared to wild type rabbits.

Objective

The current study aimed to measure the cerebral blood flow (CBF) in developing rabbits with craniosynostosis who underwent suturectomy compared to those with no intervention and compared to wild type rabbits.

Methods

Rabbits with early onset coronal suture synostosis were assigned to have suturectomy at 10 days of age (EOCS-SU, n = 15) or no intervention (EOCS, n = 18). A subset of each group was randomly selected for measurement at 10 days of age, 25 days of age, and 42 days of age. Wild type rabbits (WT, n = 18) were also randomly assigned to measurement at each time point as controls. Cerebral blood flow at the bilateral hemispheres, cortices, thalami, and superficial cortices was measured in each group using arterial spin-labeling MRI.

Results

At 25 days of age, CBF at the superficial cortex was significantly higher in EOCS rabbits (192.6 ± 10.1 mL/100 mg/min on the left and 195 ± 9.5 mL/100 mg/min on the right) compared to WT rabbits (99.2 ± 29.1 mL/100 mg/min on the left and 96.2 ± 21.4 mL/100 mg/min on the right), but there was no significant difference in CBF between EOCS-SU (97.6 ± 11.3 mL/100 mg/min on the left and 99 ± 7.4 mL/100 mg/min on the right) and WT rabbits. By 42 days of age the CBF in EOCS rabbits was not significantly different than that of WT rabbits.

Conclusion

Suturectomy eliminated the abnormally increased CBF at the superficial cortex seen in EOCS rabbits at 25 days of age. This finding contributes to the evidence that suturectomy limits abnormalities of ICP and CBF associated with craniosynostosis.

Molecular Analysis of Gli3, Ihh, Rab23, and Jag1 in a Rabbit Model of Craniosynostosis: Likely Exclusion as the Loci of Origin

OBJECTIVE

Craniosynostosis (CS) involves the premature fusion of one or more cranial sutures. The etiology of CS is complex and mutations in more than 50 distinct genes have been causally linked to the disorder. Many of the genes that have been associated with CS in humans play an essential role in tissue patterning and early craniofacial development. Among these genes are members of the Hedgehog (HH) and Notch signal transduction pathways, including the GLI family member Gli3, Indian Hedgehog ( Ihh), the RAS oncogene family member Rab23, and the Notch ligand JAGGED1 ( Jag1). We have previously described a colony of rabbits with a heritable pattern of coronal suture synostosis, although the genetic basis for synostosis within this model remains unknown. The present study was performed to determine if coding errors in Gli3, Ihh, Rab23, or Jag1 could be causally linked to craniosynostosis in this unique animal model.

DESIGN

Sequencing of cDNA templates was performed using samples obtained from wild-type and craniosynostotic rabbits.

RESULTS

Several nucleotide polymorphisms were identified in Gli3, Ihh, and Rab23, although these variants failed to segregate by phenotype. No nucleotide polymorphisms were identified in Jag1.

CONCLUSIONS

These data indicate that the causal locus for heritable craniosynostosis in this rabbit model is not located within the protein coding regions of Gli3, Ihh, Rab23, or Jag1.

Molecular Analysis of Ephrin A4 and Ephrin B1 in a Rabbit Model of Craniosynostosis: Likely Exclusion as the Loci of Origin

Craniosynostosis (CS) has a prevalence of approximately 1 in every 2000 live births and is characterized by the premature fusion of one or more cranial sutures. Failure to maintain the cell lineage boundary at the coronal suture is thought to be involved in the pathology of some forms of CS. The Ephrin family of receptor tyrosine kinases consists of membrane-bound receptors and ligands that control cell patterning and the formation of developmental boundaries. Mutations in the ephrin A4 (EFNA4) and ephrin B1 (EFNB1) ligands have been linked to nonsyndromic CS and craniofrontonasal syndrome, respectively, in patient samples. We have previously described a colony of rabbits with a heritable pattern of coronal suture synostosis, although the genetic basis for synostosis within this model remains unknown. The present study was performed to determine if EFNA4 or EFNB1 could be the loci of the causal mutation in this unique animal model. Sequencing of EFNA4 and EFNB1 was performed using templates obtained from wild-type (n = 4) and craniosynostotic (n = 4) rabbits. No structural coding errors were identified in either gene. A single-nucleotide transversion was identified in one wild-type rabbit within the third intron of EFNA4. These data indicate that the causal locus for heritable CS in this rabbit model is not located within the structural coding regions of either EFNA4 or EFNB1.

Craniosynostosis and Resynostosis: Models, Imaging, and Dental Implications

Craniosynostosis occurs in approximately 1 in 2,000 children and results from the premature fusion of ≥1 cranial sutures. If left untreated, craniosynostosis can cause numerous complications as related to an increase in intracranial pressure or as a direct result from cranial deformities, or both. More than 100 known mutations may cause syndromic craniosynostosis, but the majority of cases are nonsyndromic, occurring as isolated defects. Most cases of craniosynostosis require complex cranial vault reconstruction that is associated with a high risk of morbidity. While the first operation typically has few complications, bone rapidly regrows in up to 40% of children who undergo it. This resynostosis typically requires additional surgical intervention, which can be associated with a high incidence of life-threatening complications. This article reviews work related to the dental and maxillofacial implications of craniosynostosis and discusses clinically relevant animal models related to craniosynostosis and resynostosis. In addition, information is provided on the imaging modalities used to study cranial defects in animals and humans.

Genetic Homozygosity and Phenotypic Variability in Craniosynostotic Rabbits

BACKGROUND

Craniosynostosis ranges in severity from single suture involvement with prenatal onset to multiple suture involvement with postnatal onset. The present study was designed to test the hypothesis that increasing homozygosity may be responsible for more severe phenotypic expression by examining the relationship between inbreeding and phenotypic expression in synostotic rabbits.

METHODS

Data were obtained from 173 litters and 209 rabbits with familial craniosynostosis. Five distinct phenotypes were identified (normal n = 62; unicoronal delayed onset synostosis (DOS) n = 47; bicoronal DOS n = 21; unicoronal early onset synostosis (EOS) n = 26, and bicoronal EOS n= 53). Wright's coefficients of inbreeding (CI) were calculated using CompuPed software. Radiographs were taken at 10, 25, 42, 84, and 126 days of age to assess coronal suture, craniofacial, and skeletal growth. The relationship between CI and growth data was assessed using correlation coefficients.

RESULTS

Mean CIs ranged from 15.68 (±2.22) in normal rabbits to 25.89 (±5.03) in bicoronal DOS, to 36.29 (±2.10) in unicoronal EOS to 42.85 (±2.10) in bicoronal EOS rabbits. Significant differences were noted among groups (F = 11.48; P < .001). Significant negative correlations were noted between CI and sutural and craniofacial growth at 25 (r = -.45, P < .001; and r = -.66, P < .001) through 126 (r = -.40, P < .001 and r = -.46, P < .001) days of age.

CONCLUSIONS

While the synostotic phenotype is inherited in an autosomal dominant fashion in these rabbits, increasing homozygosity is associated with more severely affected phenotypes. These findings suggest that an accumulation of additional, modifier genes may determine the severity of the synostotic phenotype in rabbits.

The influence of surgical correction on white matter microstructural integrity in rabbits with familial coronal suture craniosynostosis

OBJECT Craniosynostosis is a condition in which one or more of the calvarial sutures fuses prematurely. In addition to the cosmetic ramifications attributable to premature suture fusion, aberrations in neurophysiological parameters are seen, which may result in more significant damage. This work examines the microstructural integrity of white matter, using diffusion tensor imaging (DTI) in a homogeneous strain of rabbits with simple, familial coronal suture synostosis before and after surgical correction. METHODS After diagnosis, rabbits were assigned to different groups: wild-type (WT), rabbits with early-onset complete fusion of the coronal suture (BC), and rabbits that had undergone surgical correction with suturectomy (BC-SU) at 10 days of age. Fixed rabbit heads were imaged at 12, 25, or 42 days of life using a 4.7-T, 40-cm bore Avance scanner with a 7.2-cm radiofrequency coil. For DTI, a 3D spin echo sequence was used with a diffusion gradient (b = 2000 sec/mm(2)) applied in 6 directions. RESULTS As age increased from 12 to 42 days, the DTI differences between WT and BC groups became more pronounced (p < 0.05, 1-way ANOVA), especially in the corpus callosum, cingulum, and fimbriae. Suturectomy resulted in rabbits with no significant differences compared with WT animals, as assessed by DTI of white matter tracts. Also, it was possible to predict to which group an animal belonged (WT, BC, and BC-SU) with high accuracy based on imaging data alone using a linear support vector machine classifier. The ability to predict to which group the animal belonged improved as the age of the animal increased (71% accurate at 12 days and 100% accurate at 42 days). CONCLUSIONS Craniosynostosis results in characteristic changes of major white matter tracts, with differences becoming more apparent as the age of the rabbits increases. Early suturectomy (at 10 days of life) appears to mitigate these differences.

Insights into the development of molecular therapies for craniosynostosis

For the past 2 decades, clinical and basic science researchers have gained significant insights into the molecular and genetic pathways associated with common forms of craniosynostosis. This has led to invaluable information for families and physicians in their attempts to understand the heterogeneity of craniosynostosis. Genetic mutations have been identified in the fibroblast growth factor receptors (FGFRs) as well as in other targets, including TWIST1, BMP, and RUNX2. Greater understanding of these and other pathways has led to the development of innovative approaches for applying medical therapies to the treatment of craniosynostosis, in particular by maintaining suture patency. In this article, the authors discuss the molecular pathophysiological mechanisms underlying various forms of craniosynostosis. They also highlight recent developments in the field of molecular craniosynostosis research with the hope of identifying targets for medical therapies that might augment the results of surgical intervention.

Transforming Growth Factor-β3 Therapy Delays Postoperative Reossification and Improves Craniofacial Growth in Craniosynostotic Rabbits

Postoperative reossification is a common clinical correlate following surgery. It has been suggested that an underexpression of transforming growth factor-β3 (TGF-β3) may be related to craniosynostosis and postoperative reossification. Adding TGF-β3 may delay reossification and improve postoperative growth. The present study was designed to test this hypothesis. Thirty 10-day-old New Zealand white rabbits with hereditary coronal suture synostosis were divided into three groups: (1) suturectomy controls (n = 14), (2) suturectomy treated with bovine serum albumin (n = 8), and (3) suturectomy treated with TGF-β3 protein (n = 8). At 10 days of age, a 3-mm × 15-mm coronal suturectomy was performed, and serial three-dimensional (3D) computed tomography (CT) scans and cephalographs were taken at 10, 25, 42, and 84 days of age. Calvaria were harvested at 84 days of age for histomorphometric analysis. Mean differences were analyzed using a group by age analysis of variance. Analysis of the 3D CT scan data revealed that sites treated with TGF-β3 had significantly (P < .05) greater defect areas and significantly (P < .05) greater intracranial volumes through 84 days of age compared with controls. Histomorphometry showed that sites treated with TGF-β3 had patent suturectomy sites and significantly (P < .001) less new bone in the suturectomy site compared with controls. Serial radiograph data revealed significant (P < .05) differences in craniofacial growth from 25 to 84 days in TGF-β3-treated rabbits compared with controls. Data show that TGF-β3 administration delayed reossification and improved craniofacial growth in this rabbit model. These findings also suggest that this molecular-based therapy may have potential clinical use.

Models of cranial suture biology

Craniosynostosis is a common congenital defect caused by premature fusion of cranial sutures. The severe morphologic abnormalities and cognitive deficits resulting from craniosynostosis and the potential morbidity of surgical correction espouse the need for a deeper understanding of the complex etiology for this condition. Work in animal models for the past 20 years has been pivotal in advancing our understanding of normal suture biology and elucidating pathologic disease mechanisms. This article provides an overview of milestone studies in suture development, embryonic origins, and signaling mechanisms from an array of animal models including transgenic mice, rats, rabbits, fetal sheep, zebrafish, and frogs. This work contributes to an ongoing effort toward continued development of novel treatment strategies.

Delivery of Transforming Growth Factor-β3 Plasmid in a Collagen Gel Inhibits Cranial Suture Fusion in Rats

Objective

Studies described in this paper were designed to test the hypothesis that an increase in nonviral, plasmid-encoded Tgf-β3 production, localized to the rat posterior frontal suture, prevents programmed suture fusion.

Design

We developed a gene delivery system based on a dense collagen gel to deliver nonviral plasmids that encode for Tgf-β3. Studies were performed to test the ability of this system to rescue rat cranial suture fusion in vitro and in vivo. Immunohistochemical studies were conducted to characterize the possible mechanisms by which increased production and presence of Tgf-β3 protein interferes with suture fusion.

Results

Posterior frontal sutures in the Tgf-β3 plasmid–treated group exhibited 77% to 85% less bony bridging than the collagen control and untreated groups after 15 days in culture. In animals treated with Tgf-β3 plasmid or Tgf-β3 protein, there was a significant reduction in suture fusion in the middle region of the posterior frontal sutures when compared with control groups. In this region the Tgf-β3 plasmid–treated group revealed 70% to 75% less bony bridging than control groups in vivo.

Conclusions

Collagen gel can be formulated to provide release of nonviral plasmid DNA that results in cell transfection and elevated Tgf-β3 protein production. Tgf-β3 is an important regulator of suture fusion, and an increase in plasmid-encoded Tgf-β3 protein is effective in inhibiting programmed suture fusion in rats.

Relaxin Does Not Rescue Coronal Suture Fusion in Craniosynostotic Rabbits

Objectives

Craniosynostosis affects 1 in 2000 to 3000 live births and may result in craniofacial and neural growth disturbances. Histological data have shown that thick collagenous bundles are present in the sutural ligament, which may tether the osteogenic fronts, resulting in premature fusion. The hormone relaxin has been shown to disrupt collagen fiber organization, possibly preventing craniosynostosis by relaxing the sutural ligament and allowing osteogenic fronts to separate normally and stay patent. This study tested this hypothesis with a rabbit model of delayed-onset coronal suture synostosis.

Methods

A total of 18 New Zealand White rabbits with craniosynostosis were randomly assigned to one of three groups: sham control, protein control (BSA), relaxin treatment. After initial diagnosis, sham surgery, BSA, or relaxin was delivered to the fusing coronal suture in a slow-release (56-day) collagen vehicle. Longitudinal radiographs and body weights were collected at 10, 25, 42, and 84 days of age, and sutures were harvested for histology.

Results

Relaxin-treated animals had more disorganized intrasuture content than control groups. These specimens also appeared to have relatively wider sutures ectocranially. There were no significant differences in relaxin-treated animals for all craniofacial growth measures, or suture separation compared with controls.

Conclusions

These data do not support our initial hypothesis that the use of relaxin may rescue sutures destined to undergo premature suture fusion. These findings suggest that collagen fiber arrangement may not be important for suture fusion. This protein therapy would not be clinically useful for craniosynostosis.

The role of vertebrate models in understanding craniosynostosis

BACKGROUND

Craniosynostosis (CS), the premature fusion of cranial sutures, is a relatively common pediatric anomaly, occurring in isolation or as part of a syndrome. A growing number of genes with pathologic mutations have been identified for syndromic and nonsyndromic CS. The study of human sutural material obtained post-operatively is not sufficient to understand the etiology of CS, for which animal models are indispensable.

DISCUSSION

The similarity of the human and murine calvarial structure, our knowledge of mouse genetics and biology, and ability to manipulate the mouse genome make the mouse the most valuable model organism for CS research. A variety of mouse mutants are available that model specific human CS mutations or have CS phenotypes. These allow characterization of the biochemical and morphological events, often embryonic, which precede suture fusion. Other vertebrate organisms have less functional genetic utility than mice, but the rat, rabbit, chick, zebrafish, and frog provide alternative systems in which to validate or contrast molecular functions relevant to CS.

Molecular signaling in pathogenesis of craniosynostosis: the role of fibroblast growth factor and transforming growth factor–β

The interplay of signals between dura mater, suture mesenchyme, and brain is essential in determining the fate of cranial sutures and the pathogenesis of premature suture fusion leading to craniosynostosis. At the forefront of research into suture fusion is the role of fibroblast growth factor and transforming growth factor–β, which have been found to be critical in the cell-signaling cascade involved in aberrant suture fusion. In this review, the authors discuss recent and ongoing research into the role of fibroblast growth factor and transforming growth factor–β in the etiopathogenesis of craniosynostosis.

Genetic basis of potential therapeutic strategies for craniosynostosis

Craniosynostosis, the premature fusion of one or more cranial sutures, is a common malformation of the skull that can result in facial deformity and increased intracranial pressure. Syndromic craniosynostosis is present in ∼15% of craniosynostosis patients and often is clinically diagnosed by neurocranial phenotype as well as various other skeletal abnormalities. The most common genetic mutations identified in syndromic craniosynostosis involve the fibroblast growth factor receptor (FGFR) family with other mutations occurring in genes for transcription factors TWIST, MSX2, and GLI3, and other proteins EFNB1, RAB23, RECQL4, and POR, presumed to be involved either upstream or downstream of the FGFR signaling pathway. Both syndromic and nonsyndromic craniosynostosis patients require early diagnosis and intervention. The premature suture fusion can impose pressure on the growing brain and cause continued abnormal postnatal craniofacial development. Currently, treatment options for craniosynostosis are almost exclusively surgical. Serious complications can occur in infants requiring either open or endoscopic repair and therefore the development of nonsurgical techniques is highly desirable although arguably difficult to design and implement. Genetic studies of aberrant signaling caused by mutations underlying craniosynostosis in in vitro calvarial culture and in vivo animal model systems have provided promising targets in designing genetic and pharmacologic strategies for systemic or adjuvant nonsurgical treatment. Here we will review the current literature and provide insights to future possibilities and limitations of therapeutic applications.

Microarray analysis of the role of regional dura mater in cranial suture fate

BACKGROUND

Craniosynostosis, the premature fusion of cranial sutures, results in serious neurologic and morphologic abnormalities when left untreated. Surgical excision of the fused sutures and remodeling of the skull remains the standard therapy. Development of novel, minimally invasive therapies for craniosynostosis will undoubtedly be dependent on a more thorough understanding of the molecular mechanisms underlying this abnormality. Significant evidence suggests the influence of regional dura mater on the behavior of the overlying suture complex. The mouse model has been instrumental in investigating this observation because of the natural juxtaposition of the posterior frontal suture, which fuses early in life, with the other cranial sutures, which remain patent.

METHODS

The authors used microarray analysis to compare genomic changes in the dura mater underlying the posterior frontal and sagittal sutures of mice. Suture-associated dura mater was harvested from mice before (postnatal day 5), during (postnatal day 10), and after (postnatal day 20) posterior frontal suture fusion (n = 20 mice for each of the three time points).

RESULTS

Microarray results confirmed differential regulation of genes involved in paracrine signaling, extracellular matrix, and bone remodeling between the dura mater underlying the fusing posterior frontal suture and the patent sagittal suture.

CONCLUSIONS

These data confirm global differences in gene expression between regional dura mater underlying fusing and patent sutures. These results provide further insight into potential molecular mechanisms that may play a role in cranial suture biology.

Comparison of craniofacial phenotype in craniosynostotic rabbits treated with anti-Tgf-beta2 at suturectomy site

OBJECTIVE

Overexpression of transforming growth factor-beta 2 has been associated with craniosynostosis and resynostosis following surgery. We examined the effects of localized transforming growth factor-beta 2 inhibition on craniofacial phenotype in rabbits with craniosynostosis.

DESIGN

Twenty-five New Zealand white rabbits with bilateral coronal craniosynostosis were divided into three treatment groups: (1) suturectomy control (n=8); (2) suturectomy with nonspecific, control immunoglobulin G antibody (n=6); and (3) suturectomy with anti-transforming growth factor-beta 2 antibody (n=11). At 10 days of age, a coronal suturectomy was performed on all rabbits. The sites in groups 2 and 3 were immediately filled with a slow-resorbing collagen gel mixed with either immunoglobulin G or anti-transforming growth factor-beta 2 antibody. Computed tomography scans of each rabbit were acquired at ages 10, 25, and 84 days. Craniofacial landmarks were collected from three-dimensional computed tomography reconstructions, and growth and form were compared among the three groups.

RESULTS

Rabbits treated with anti-transforming growth factor-beta 2 antibody differed in form at 84 days of age compared with suturectomy control rabbits, specifically in the snout and posterior neurocranium. Growth in some areas of the skull was greater in rabbits from the anti-transforming growth factor-beta 2 group than in suturectomy control rabbits, but not significantly greater than in IgG control rabbits.

CONCLUSIONS

We find support for the hypothesis that transforming growth factor-beta 2 inhibition alters adult form, but these changes do not appear to be localized to the suturectomy region. Slight differences in form and growth between the two control groups suggest that the presence of the collagen vehicle itself may affect skull growth.