Lachlan Smith, PhD
University of Pennsylvania
“Pathogenesis of Bone Disease in Mucopolysaccharidosis Disorders”
Background and Objectives
The mucopolysaccharidoses (MPS) are lysosomal storage disorders characterized by deficiencies in enzymes that degrade glycosaminoglycans (GAGs), resulting in multi-systemic disease manifestations. Skeletal abnormalities are prevalent, often require surgery, and significantly impact patients’ quality of life. MPS VII (Sly Syndrome) is characterized by deficient beta-glucuronidase (GUSB) activity, leading to systemic accumulation of incompletely degraded chondroitin, heparan and dermatan sulfate GAGs. Spine disease is particularly severe and includes morphological abnormalities in the vertebral bones. Resulting spinal cord compression and kypho-scoliotic deformity impact directly on patient mortality and quality of life. The objective of this project was to establish the molecular mechanisms of failed vertebral bone formation in MPS, focusing primarily on MPS VII. Our overall hypothesis was that abnormal GAG accumulation in developing MPS VII vertebrae impairs bone formation by disrupting key regulatory signaling pathways.
Summary of Key Findings
A defining feature of spine disease in MPS VII is the presence of cartilaginous lesions in the vertebral bodies, indicative of failed cartilage-to-bone conversion during postnatal development; however, the underlying molecular mechanisms are poorly understood. During normal vertebral bone formation, cartilaginous rudiments form a template where resident chondrocytes undergo distinct stages of differentiation regulated by a highly orchestrated pattern of growth factor signaling, culminating in ossification by osteoblasts. GAGs perform critical roles in regulating the activity of these growth factors. We hypothesized that abnormal GAG accumulation in MPS VII disrupts chondrocyte differentiation by interfering with growth factor signaling, preventing normal cartilage-to-bone conversion. We collected thoracic vertebrae post mortem from 9 and 14 day old control (normal) and MPS VII affected dogs for qPCR and micro-computed tomography (microCT) analyses. Comparison of mRNA expression of chondrocyte differentiation and osteoblast markers, and microCT visualization of vertebrae, demonstrated that secondary ossification commenced in controls between 9 and 14 days, but in MPS VII did not. Further, these results suggested that resident chondrocytes in MPS VII epiphyseal cartilage fail to successfully transition from proliferation to hypertrophy at the appropriate developmental age – a critical prerequisite to subsequent bone formation.
Having established this key developmental window when failed bone formation first manifests in MPS VII dogs, we conducted studies to establish the underlying molecular mechanisms. Specifically, we sought to establish which signaling pathways fail to activate in MPS VII epiphyseal cartilage, and to examine cellular responses to related secreted growth factors using a cartilage explant model. Vertebral epiphyseal cartilage from control and MPS VII dogs was collected postmortem at 9 and 14 days. Whole transcriptome sequencing (RNA-Seq) was performed and differential gene expression for pathways that regulate endochondral ossification was determined. RNA-Seq revealed 411 and 1104 genes significantly differentially expressed (fold-change>2) between control and MPS VII at 9 and 14 days respectively, with Wnt/β-catenin signaling the top dysregulated bone formation pathway. Nuclear β-catenin protein levels were measured, revealing significantly higher levels at 14 compared to 9 days in controls but no change in MPS VII. Vertebral epiphyseal cartilage from 9 day control and MPS VII animals was cultured for 1, 3, or 7 days with or without Wnt3a. Controls exhibited a significant decrease in SOX9 with Wnt3a treatment after just 1 day, however MPS VII explants did not exhibit a significant decrease until 3 days. These results suggest that Wnt/β-catenin signaling does not activate at the appropriate developmental stage to initiate chondrocyte differentiation in MPS VII, and that MPS VII chondrocytes have impaired capacity to respond to Wnts. In parallel studies, we demonstrated that extracellular GAG composition in MPS VII epiphyseal cartilage is significantly altered as early as 9 days-of-age. This is consistent with our overall hypothesis that abnormal GAG accumulation contributes to altered growth factor signaling and failed bone formation.
Through this work, we have established the earliest developmental stage when failed vertebral bone formation manifests in MPS VII dogs, and demonstrated that this can be traced to the inability of epiphyseal chondrocytes to undergo hypertrophic differentiation. Further, our results suggest that impaired chondrocyte differentiation is associated with abnormal GAG accumulation in epiphyseal cartilage and the failure of the Wnt/β-catenin signaling to activate at the appropriate developmental stage. Our ultimate goal is to identify new therapeutic targets to promote healthy bone formation in MPS patients. We anticipate that GUSB ERT, currently undergoing clinical trials for MPS VII, will not fully normalize bone formation, and that other adjuvant therapeutics such as modulators of Wnt signaling or other pathways will also be required. This work lays the foundation for development of such therapies.
Publications Arising from this Work:
• Peck SH, O’Donnell PJ, Kang J, Shore EM, Pacifici M, Dodge GR, Haskins ME, Malhotra NR, Smith LJ. (2015) Delayed Hypertrophic Differentiation of Epiphyseal Chondrocytes Contributes to Failed Bone Formation in Mucopolysaccharidosis VII Dogs. Molecular Genetics and Metabolism. 116:195-203.
Conference Abstracts Arising from this Work:
• Peck SH, Kang JL, Chiaro JA, Malhotra NR, Haskins ME, Casal ML, Smith LJ. (2016) Progression of Vertebral Bone Disease in Mucopolysaccharidosis VII Dogs from Birth to Skeletal Maturity. Orthopaedic Research Society Annual Meeting, Orlando, USA.
• Peck SH, Shore EM, Malhotra NR, Tobias JW, Pacifici M, Haskins ME, Smith LJ. (2016) Impaired Wnt Signaling Contributes to Delayed Chondrocyte Differentiation in Mucopolysaccharidosis VII Dogs. Orthopaedic Research Society Annual Meeting, Orlando, USA.
• Peck SH, Kang JL, Dodge GR, Malhotra NR, Haskins ME, Smith LJ. (2016) Aberrant Glycosaminoglycan Accumulation and Sulfation in Epiphyseal Cartilage in Mucopolysaccharidosis VII. Orthopaedic Research Society Annual Meeting, Orlando, USA.
• Peck SH, O’Donnell PJ, Kang J, Casal M, Shore EM, Pacifici M, Haskins ME, Malhotra NR, Smith LJ. (2015) Failed Vertebral Bone Formation in Mucopolysaccharidosis VII is Associated with Aberrant Sox9 Regulation and Altered Wnt Signaling. American Society for Bone and Mineral Research Annual Meeting, Seattle, USA.
• Peck SH, O’Donnell PJ, Chiaro JA, Shore EM, Pacifici M, Haskins ME, Malhotra NR, Smith LJ. (2015) Impaired Chondrocyte Hypertrophic Differentiation is Associated with Failed Vertebral Bone Formation in Mucopolysaccharidosis VII Dogs. World Symposium on Lysosomal Diseases, Orlando, USA
• Peck SH, O’Donnell PJ, Chiaro JA, Shore EM, Pacifici M, Haskins ME, Malhotra NR, Smith LJ. (2015) Pinpointing the Earliest Manifestations of Bone Disease in Mucopolysaccharidosis Disorders. Orthopaedic Research Society Annual Meeting, Las Vegas, USA
Awards Arising from this Work:
• First prize, poster competition, ORS/PSRS 3rd International Spine Research Symposium for “Inactivation of Wnt/β-Catenin Signaling Contributes to Failed Vertebral Bone Formation in MPS VII”, Philadelphia, USA, 2015.
Richard Steet, PhD Dr. Dwight Koeberl
University of Georgia Duke University
Athens, GA Durham, NC
“Adjunctive therapy for Hurler syndrome.”
Second Year Progress Report: “Adjunctive Therapy for Hurler Syndrome”
Progress Report Year 2 MPS/ISMRD Grant to Heather Flanagan-Steet – Investigating the
Role of Cathepsin Proteases in MLII Cardiac Pathology
In the last year we have made progress on several aspects of this grant. In addition to
progress on the proposed studies, we have made two important advances that will have a
major impact on our ongoing pursuit of molecular mediators of MLII cardiac pathology.
These include the generation of several unique TALEN-mediated and TILLING based
MLII mutant lines. During this year we not only established pure stable lines, but we also
confirmed the genetic and biochemical characteristics of these lines. These animals are
essential to confirm all of our morpholino-based findings, for analyses of later stage
aspects of disease, and for small molecule screens. Importantly, several of the lines
exhibit the same craniofacial and cardiac anomalies noted in morpholino-generated
animals. Using both TALEN-generated and morpholino based animals, we continue to
assess the impact of cathepsin proteases in disease pathology. Toward this goal genetic
and pharmacological inhibition studies are beginning to yield promising results regarding
the impact of cathepsin during pathogenesis.
Jeffrey Esko, PhD – due September 2015
University of California, San Diego
La Jolla, CA
“Delivery of sulfamidase to the brain.”
This proposal had three aims: (1) Generate GNeo-conjugated forms of murine sulfamidase; (2)
Compare intravenous or intranasal enzyme delivery to the CNS; (3) Evaluate GNeo-sulfamidase
for reduction of pathological markers in MPS IIIA mice. Over the last two years we successfully
produced hihgly purified recombinant sulfamidase (SGSH) under serum free conditions and
optimized conditions for conjugating the enzyme with our carrier system (guanidinylated
neomycin, GNeo). Although the conjugated enzyme cleared very rapidly from the circulation (t1/2
~ 8 min), efficient delivery to the brain by intranasal administration was achieved. GNeo does
not appear to have adverse side effects based on the lack of obvious changes in gross behavior,
feeding behavior, and growth of the mice after administering the conjugate every other day for
MPSIIIA is characterized by accumulation of heparan sulfate and the pathological carbohydrate
biomarker N-sulfoglucosamine on the non-reducing end (NRE) of the chain. Analysis of the
NRE in the glycosaminoglycans extracted from whole brain after intranasal treatment of mice
every other day for 1 month with GNeo-SGSH showed a ~25% reduction in the biomarker.
GNeo-SGSH was ~5-fold more effective than SGSH. An initial trial in which 3-month old mice
were treated for 5 months with 1 mg/kg of GNeo-SGSH did not show significant reduction of
neuropathological markers. Current studies are focused on whether intranasal administration
can be initiated at 1,2, or 4 weeks after birth, prior to the development of any neuropathology.
A natural history study showed that heparan sulfate storage is present in the brains of newborn
MPSIIIA mice and increased in postnatal mice in proportion to brain size. We noted that the
NRE biomarker was significantly higher in the early postnatal cortex than in whole brain
samples, suggesting that the cortex is more seriously affected in the developing postnatal brain.
A preliminary histological analysis at different embryonic and postnatal time points showed that
differentiation of neural progenitors, axon specification, neuronal migration and glial cell
generation occurs normally. However studies of the neural circuitry showed an increase in the
number of excitatory synapses at the end of synaptogenesis in the somatosensory cortex, due
to enhancement in postsynaptic sites. This work shows a developmental phenotype in MPSIII
and emphasizes the importance of early therapeutic intervention. Current work is focused on
determining the cellular mechanisms leading to increased excitatory synapses using in vitro
models of primary neural cells from the cortex of Sgsh mice and whether the synaptic changes
result in altered electrophysiology.
We also generated a conditional Sgsh-deficient mouse (Sgshf/f) and generated a germ line null
allele, whose phenotype is virtually identical to the standard strain, which contains a
hypomorphic allele and produces 3-4% residual enzyme activity. To ascertain the in vivo
contribution of neurons and astrocytes to early disease pathology, we intercrossed the
conditional Sgshf/f with mice bearing specific Cre expression in different cell types. Restricting
Sgsh deletion to cells of neuroepithelial origin, neurons and glia resulted in attenuated heparan
sulfate storage and secondary histopathology, suggesting that other cell types in the brain
contribute to storage or can act as a source of enzyme for cross-feeding. Deletion of Sgsh in a
single cell type (neurons, astrocytes or endothelial cells) did not induce storage, consistent with
Support from the MPS Foundation provided the preliminary data needed to apply for grants
though the NIH and a private foundation.
ML II/II(Partnership Grant with ISMRD)
Heather Flanagan-Steet, PhD
Complex Carbohydrate Research Center
University of Georgia
“Investigating the role of cathepsin proteases in ML II cardiac pathology”
Progress Report Year 2 MPS/ISMRD Grant to Heather Flanagan-Steet – Investigating the Role of Cathepsin Proteases in MLII Cardiac Pathology
In the last year we have made progress (detailed below) on several aspects of the aims outlined in this grant. In addition to progress on the proposed studies, we have made two important advances that will have a major impact on our ongoing pursuit of molecular mediators of MLII cardiac pathology. These include 1) the generation of five unique TALEN-mediated MLII mutant lines and 2) the isolation of a separate MLII mutant line from a sperm TILLING screen. During this year we not only established pure – outcrossed stable lines, but we also confirmed the genetic and biochemical characteristics of these lines. These animals are essential to confirm all of our morpholino-based findings, for analyses of later stage aspects of disease, and for small molecule screens. Importantly, several of the lines exhibit the same craniofacial and cardiac anomalies noted in morpholino-generated animals. When published we will credit MPS/ISMRD for contributing to the establishment of these vital tools.
Aim 1: Assess individual contribution of cathepsin K and L toward cardiac defects in ML zebrafish.
1) Cathepsin expression and generation of transgenic animals. Immunohistochemical analyses of cathepsin K and L demonstrate that both proteases are expressed in multiple heart tissues, including the myocardium, endocardium, and epicardium. These analyses have guided the choice of tissue specific promoters for transgenic constructs that will be used to generate zebrafish expressing fluorophor tagged WT and mutant versions of Cts K and L. Toward this goal, we have now successfully shown that following mRNA injection, the CtsK-mCherry fusion protein is both expressed and active in developing embryos. Establishing these parameters was critical before proceeding with transgenesis. These animals, as well as multiple ctsk variants/mutants, particularly those that mimick MLII, are in production.
2) Inhibition of cathespin K in MLII. To assess CtsK’s contribution toward ML cardiac pathology, its expression or activity was inhibited in the MLII background. CtsK expression was genetically inhibited using one of two gene specific morpholinos and its activity reduced pharmacologically using the FDA-approved cathepsin K inhibitor Odanacatib. All aspects of cardiac morphology and function, including formation of the AV valve, unidirectional blood flow, and heart contraction were substantially improved by these treatments.
Aim 2 – Determine whether increased cathepsin activity impacts TGFß signaling in MLII hearts.
1) Are there differences in either TGFß or BMP signaling in WT and MLII hearts? In combination with TGFß and BMP-reporter transgenic animals, immunohistochemical analyses of pSmad levels was used to assess differences in signaling between WT and MLII hearts. These analyses revealed important changes in these pathways that were restored by inhibition of cathepsin K.
Aim 3 – Determine which patient mutations are pathogenic for cardiac dysfunction.
To identify which patient mutations are associated with altered cardiac morphology and function, we have introduced mRNA bearing specific lesions in the morphant background and compared their ability to rescue heart phenotypes with ML animals co-injected with WT mRNA. We have analyzed three different sets of mutations including several within the DMAP domain (including K732N), three mutations within Notch domain 1 (C442Y, C461G, and C468S), and one mutation in Notch domain 2. Thus far analyses have been limited to gross morphology. Unlike WT gnptab mRNA, which reduces cardiac edema, restores normal cardiac morphology, and increases blood flow in 85% of the animals analyzed, mRNA bearing the K732N mutation did not significantly restore any of the tested parameters. Therefore, as previously noted in the cartilage, the DMAP domain is essential for normal phosphotransferase function – and mutations within it are highly likely to be pathogenic in both the cartilage/bone and the heart. In contrast, preliminary analyses of the three Notch domain 1 mutations suggest that the C442Y lesion, which has been associated with MLIII, is less pathogenic in the heart than the ML-intermediate lesion C468S. This is supported by the fact that mRNA“rescue” experiments show improved cardiac morphology and reduced edema in 85% of the MLII animals injected with the C442Y-containing mRNA versus only 40% recovery with the mRNA carrying the C468S lesion. mRNAs bearing either the C461G (Notch 1/MLIII) or C505Y (Notch2/MLIII) rescued gross cardiac pathology ~60% of the time, suggesting that in addition to differences in pathogenic severity between MLII and MLIII, different mutations within a single ML class may have different degrees of pathogenicity. More in depth experiments are currently underway. Collectively these data support the idea that genotype-phenotype correlations can be established in this system, and as such may eventually combine with patient data to predict the severity and course of mutation specific tissue pathologies.
Summary of progress to date:
Although complete understanding of the complicated nature of MLII cardiac pathology still demands more work, we are highly encouraged by the findings to date. First, the fact that both cathepsin K knockdown and Ctsk inhibitors improved MLII heart development is very promising. These data not only support a central role for cathepsin(s) in primary pathogenesis but point to FDA-approved drugs for future consideration. Second, the fact that both cathepsin K and TGFß signaling are involved in cardiac pathology suggests common mechanisms underlie ML skeletal and heart dysfunction. If true, this would not only be an important advance for future scientific investigation, but also mean that an individual modality may therapeutically improve both systems. Third, the tools generated thus far and those underway will serve as novel platforms to both further investigate these and other emerging mechanisms and to rapidly screen the efficacy of potential drugs.