Final Report Summary - SUGARYBONE (The Role of Protein N-Glycosylation in Bone and Energy Homeostasis)
Project Objectives
Osteoporotic bone loss due to aging or disease is a major health issue that affects hundreds of millions of people worldwide and costs tens of billions of euros each year. Since current osteoporosis medication only reduces fracture risk by 25-50%, there is an urgent need to define new pathways that control bone remodelling and strength in order to identify new therapeutic targets. Via the ‘Mouse Genetics Project’ consortium, we identified a mutation in one of the enzymes involved in the N-linked protein glycosylation pathway that leads to profound osteoporosis and obesity in mice; this observation accorded with the outcome of comparable human pathologies (congenital disorders of glycosylation). Though bone contains a large amount of glycoproteins, the importance of protein N-glycosylation to skeletal homeostasis remains to be characterised. This project aims to:
1. Identify the role of protein N-glycosylation in bone homeostasis,
2. Assess mechanistically how bone cell functioning is regulated by this process, and
3. Characterise the metabolic abnormalities associated with defects in N-glycosylation and to address its linkage to bone.
We hypothesise that defective N-glycosylation impairs bone formation by osteoblasts, leading to the observed osteoporosis, and, likely, reduced bone-derived osteocalcin levels, which will in turn result in hampered insulin release and insulin resistance, with the observed obesity as a consequence. By combining the applicant’s and host’s skills in mouse genetics, skeletal phenotyping and high-throughput techniques, with the expertise in energy homeostasis, mouse embryonic stem cell technologies and (glyco)proteomics at the Wellcome Trust Sanger Institute, we are confident to gain insight in a process that most possibly regulates osteoblast functioning and bone matrix production; this knowledge will contribute to the development of novel, urgently warranted anabolic medication to treat osteoporosis.
Work performed
We aimed to gain mechanistic insight into the role of protein N-glycosylation in energy and bone homeostasis by using mice with a mutation in the key enzyme for initiating N-glycosylation as a model (Alg13 ,UDP-N-acetylglucosaminyltransferase 13, Alg13mut mice). These mice had been generated through the Mouse Genetics Project (MGP) at the host institute [Wellcome Trust Sanger Institute (WTSI)]. Bioinformatic and transcriptomic analyses were carried out to assess whether the Alg13Δex7 line had, in fact, reduced enzymatic activity. Experiments were performed in collaboration with Dr. Stuart Haslam (Imperial College London) to assess the glycoprofile of kidney and liver samples of WT and Alg13Δex7 mice. A vector targeted to exon 2 of Alg13 was generated and electroporated into mouse ES cells, to generate a new mutant mouse line. Over 300 ES cell clones were picked over two independent experiments and screened for recombination by short and long-range PCR.
Results
Alg13 has two isoforms: a short one (exon 1-4) containing the enzymatic activity, and a long one of which the function is unknown. Closer analysis of the targeting strategy used by the Mouse Genetics Project (WTSI) revealed that the Alg13Δex7 mice did not have a deletion in a critical exon, yet in an inactive exon located in the large isoform of Alg13. Consequently, mice, unlike humans, do not express the long isoform of Alg13. Based on these data, we decided to first confirm the defects in N-glycosylation before continuing the animal experiments. Detailed characterisation of permethylated N glycans by MALDI-TOF MS show a high level of similarity between WT and Alg13Δex7 kidney samples. The variety of terminal antennae structures are conserved between WT and KO and display no defects in trimming or modifications performed by glycosyltransfersases involved in the N-terminal glycosylation pathway. Also in liver, the presence of glycan structures were comparable between WT and Alg13Δex7 samples and no prominent structural alterations were observed. Also, no abnormalities were observed in O-glycosylation in the kidney, nor in the liver samples. In conclusion, the deletion of exon 7 of Alg13 did not lead to any abnormalities in posttranslational glycosylation. As a consequence, the available Alg13Δex7 mice cannot be used to study the role of N-glycosylation in bone or energy metabolism.
Since the available Alg13Δex7 mESC and mice could not be used, we decided to generate an appropriate model by targeting exon 2 of Alg13 (Alg13Δex2). Importantly, Alg13 is an X-linked gene and thus has only one gene copy in the classically used mESC, which are all male. Since these mESC are widely used at the institute and are, in contrast to the female cells, known to give good germline transmission, we chose to first try them. Since we foresee Alg13 to be very critical and consequently that inactivation may influence survival, in vitro recombination to eliminate the gene-disrupting lacZ/Neo cassette was planned (conditional allele generation) by introducing FLP recombinase in vitro. In this regard, we planned two types of targeting strategies: the first in WT mESC followed by a transient electroportation with FLP recombinase expressing vector; the second in mESC already expressing an antibiotic-activated FLP recombinase vector. Short-range PCR with primers designed on exon 2 and the homology arm showed that from the 300 cells that were picked, none were correctly targeted. To confirm that this was not due to handling errors of the researcher, a parallel targeting has been done through the MGP pipeline by a highly experienced researcher (Dr Barry Rosen, bespoke MGP pipeline). Consistent with previous data, none of the 100 picked clones showed correct 5’ and 3’ targeting.
Conclusion and impact
Generation of Alg13Δex2 mouse ES cells has been unsuccessful despite picking over 300 clones in two separate attempts. Most likely, the correctly targeted Alg13Δex2 ES cells do not survive the short period in which Alg13 function is disrupted. This period is however needed to get enough cells to activate the in vitro elimination of the gene-disrupting cassette. The project was terminated at this point, 8 months into the fellowship, therefore the socio-economic impact of the objectives towards developing new treatments for osteoporosis was not realized.
Osteoporotic bone loss due to aging or disease is a major health issue that affects hundreds of millions of people worldwide and costs tens of billions of euros each year. Since current osteoporosis medication only reduces fracture risk by 25-50%, there is an urgent need to define new pathways that control bone remodelling and strength in order to identify new therapeutic targets. Via the ‘Mouse Genetics Project’ consortium, we identified a mutation in one of the enzymes involved in the N-linked protein glycosylation pathway that leads to profound osteoporosis and obesity in mice; this observation accorded with the outcome of comparable human pathologies (congenital disorders of glycosylation). Though bone contains a large amount of glycoproteins, the importance of protein N-glycosylation to skeletal homeostasis remains to be characterised. This project aims to:
1. Identify the role of protein N-glycosylation in bone homeostasis,
2. Assess mechanistically how bone cell functioning is regulated by this process, and
3. Characterise the metabolic abnormalities associated with defects in N-glycosylation and to address its linkage to bone.
We hypothesise that defective N-glycosylation impairs bone formation by osteoblasts, leading to the observed osteoporosis, and, likely, reduced bone-derived osteocalcin levels, which will in turn result in hampered insulin release and insulin resistance, with the observed obesity as a consequence. By combining the applicant’s and host’s skills in mouse genetics, skeletal phenotyping and high-throughput techniques, with the expertise in energy homeostasis, mouse embryonic stem cell technologies and (glyco)proteomics at the Wellcome Trust Sanger Institute, we are confident to gain insight in a process that most possibly regulates osteoblast functioning and bone matrix production; this knowledge will contribute to the development of novel, urgently warranted anabolic medication to treat osteoporosis.
Work performed
We aimed to gain mechanistic insight into the role of protein N-glycosylation in energy and bone homeostasis by using mice with a mutation in the key enzyme for initiating N-glycosylation as a model (Alg13 ,UDP-N-acetylglucosaminyltransferase 13, Alg13mut mice). These mice had been generated through the Mouse Genetics Project (MGP) at the host institute [Wellcome Trust Sanger Institute (WTSI)]. Bioinformatic and transcriptomic analyses were carried out to assess whether the Alg13Δex7 line had, in fact, reduced enzymatic activity. Experiments were performed in collaboration with Dr. Stuart Haslam (Imperial College London) to assess the glycoprofile of kidney and liver samples of WT and Alg13Δex7 mice. A vector targeted to exon 2 of Alg13 was generated and electroporated into mouse ES cells, to generate a new mutant mouse line. Over 300 ES cell clones were picked over two independent experiments and screened for recombination by short and long-range PCR.
Results
Alg13 has two isoforms: a short one (exon 1-4) containing the enzymatic activity, and a long one of which the function is unknown. Closer analysis of the targeting strategy used by the Mouse Genetics Project (WTSI) revealed that the Alg13Δex7 mice did not have a deletion in a critical exon, yet in an inactive exon located in the large isoform of Alg13. Consequently, mice, unlike humans, do not express the long isoform of Alg13. Based on these data, we decided to first confirm the defects in N-glycosylation before continuing the animal experiments. Detailed characterisation of permethylated N glycans by MALDI-TOF MS show a high level of similarity between WT and Alg13Δex7 kidney samples. The variety of terminal antennae structures are conserved between WT and KO and display no defects in trimming or modifications performed by glycosyltransfersases involved in the N-terminal glycosylation pathway. Also in liver, the presence of glycan structures were comparable between WT and Alg13Δex7 samples and no prominent structural alterations were observed. Also, no abnormalities were observed in O-glycosylation in the kidney, nor in the liver samples. In conclusion, the deletion of exon 7 of Alg13 did not lead to any abnormalities in posttranslational glycosylation. As a consequence, the available Alg13Δex7 mice cannot be used to study the role of N-glycosylation in bone or energy metabolism.
Since the available Alg13Δex7 mESC and mice could not be used, we decided to generate an appropriate model by targeting exon 2 of Alg13 (Alg13Δex2). Importantly, Alg13 is an X-linked gene and thus has only one gene copy in the classically used mESC, which are all male. Since these mESC are widely used at the institute and are, in contrast to the female cells, known to give good germline transmission, we chose to first try them. Since we foresee Alg13 to be very critical and consequently that inactivation may influence survival, in vitro recombination to eliminate the gene-disrupting lacZ/Neo cassette was planned (conditional allele generation) by introducing FLP recombinase in vitro. In this regard, we planned two types of targeting strategies: the first in WT mESC followed by a transient electroportation with FLP recombinase expressing vector; the second in mESC already expressing an antibiotic-activated FLP recombinase vector. Short-range PCR with primers designed on exon 2 and the homology arm showed that from the 300 cells that were picked, none were correctly targeted. To confirm that this was not due to handling errors of the researcher, a parallel targeting has been done through the MGP pipeline by a highly experienced researcher (Dr Barry Rosen, bespoke MGP pipeline). Consistent with previous data, none of the 100 picked clones showed correct 5’ and 3’ targeting.
Conclusion and impact
Generation of Alg13Δex2 mouse ES cells has been unsuccessful despite picking over 300 clones in two separate attempts. Most likely, the correctly targeted Alg13Δex2 ES cells do not survive the short period in which Alg13 function is disrupted. This period is however needed to get enough cells to activate the in vitro elimination of the gene-disrupting cassette. The project was terminated at this point, 8 months into the fellowship, therefore the socio-economic impact of the objectives towards developing new treatments for osteoporosis was not realized.