- Email: [email protected]
Database of Complement Gene Variants: A comprehensive database providing insights on function, structure and allele frequency for genetic variants identified in complement-mediated diseases
Abstracts / Immunobiology 221 (2016) 1131–1225
205
206
The solution structure of the human complement regulator CFHR5 reveals a compact dimeric structure by X-ray scattering and analytical ultracentrifugation
Database of Complement Gene Variants: A comprehensive database providing insights on function, structure and allele frequency for genetic variants identified in complement-mediated diseases
Nilufar Kadkhodayi-Kholghi 1,∗ , Jayesh Gor 1 , Anna Ferlin 2 , Lindsay C. McDermott 3 , Daniel P. Gale 2 , Stephen J. Perkins 1 1
Department of Structural and Molecular Biology, University College London, London, United Kingdom 2 UCL Centre for Nephrology, Royal Free Hospital, University College London, London, United Kingdom 3 Department of Life Sciences, University of Bedfordshire, Luton, United Kingdom Human Complement Factor H Related 5 (CFHR5) belongs to the same complement regulator family as Factor H. The CFHR5 protein comprises a linear sequence of nine short complement regulator (SCR) domains. A mutation of CFHR5 that results in duplication of the N-terminal SCR-1/2 domain pair causes CFHR5 nephropathy, a common cause of kidney failure in Cypriots. To clarify the molecular basis of CFHR5 nephropathy, the domain arrangement of full-length CFHR5 was studied by analytical ultracentrifugation and X-ray scattering. Sedimentation velocity reported a molecular mass of 134 kDa, indicating that CFHR5 is dimeric with 18 SCR domains and not nine. The sedimentation coefficient of 5-6 S for CFHR5 decreased with increase in NaCl concentration, showing that its domain conformation had become more extended in high ionic strength. X-ray scattering also showed that CFHR5 was dimeric to confirm the ultracentrifugation result. The X-ray mean radius of gyration RG was 5.5 ± 0.2 nm, and its maximum length was 20 nm. This length of 20 nm for a protein with 18 SCR domains is low compared to that of 32 nm for Factor H with 20 SCR domains, indicating that CFHR5 possesses a more compact SCR arrangement than that seen for Factor H. The scattering curve modelling of CFHR5 was performed using a new modelling procedure called SASSIE that involved molecular dynamic simulations to generate physicallyrealistic atomistic SCR structures for CFHR5. The best-fit modelling structures confirmed that CFHR5 possessed a folded-back compact domain structure. We expressed the single SCR-1 domain and the SCR-1/2 domain pair of CFHR5 to show from analytical ultracentrifugation that SCR-1 was monomeric, while SCR-1/2 was dimeric, thus locating the CFHR5 dimerization site to its N-terminus. We are currently performing functional experiments with the SCR-1 and SCR-1/2 domains to clarify their role in CFHR5 activity. In summary, our results indicate a markedly more compact solution structure of CFHR5 than previously thought, and we have located its dimerization site to SCR-1/2 which may have a major role in causing CFHR5 nephropathy. http://dx.doi.org/10.1016/j.imbio.2016.06.220
1223
Amy J. Osborne 1,∗ , Santiago Rodriguez de Cordoba 2 , Veronique Fremeaux-Bacchi 3 , Marina Noris 4 , Richard J. Smith 5 , Bert van den Heuvel 6 , Timothy H.J. Goodship 7 , Stephen J. Perkins 1 1 Department of Structural & Molecular Biology, University College London, UK 2 Departamento de Inmunologia, Centro de Investigaciones Biologicas (CSIC), Madrid, Spain 3 Service d’Immunologie Biologique, Hopital Europeen Georges Pompidou, Paris, France 4 Laboratory of Immunology and Genetics of Rare Diseases and Transplantation, Mario Negri Institute for Pharmacological Research, Italy 5 Iowa Institute of Human Genetics, University of Iowa, Iowa City, USA 6 Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands 7 Institute of Genetic Medicine, Newcastle University, UK
Atypical haemolytic uremic syndrome (aHUS) and C3 glomerulopathy (C3G) are associated with dysregulation of the alternative pathway (AP) of complement. Recently, we described an updated compilation of 324 genetic variants identified in four AP genes encoding the proteins factor H (FH), factor I (FI), membrane cofactor protein (MCP, CD46) and C3 in our interactive FH-HUS web-database (www.fh-hus.org). Our database has been updated further to a new version, the ‘Database of Complement Gene Variants’ (www.complement-db.org), with variant data contributed by six centres (Newcastle, Paris, Bergamo, Madrid, Iowa City, and Nijmegen) from >1000 patients with aHUS and C3G. This laboratory dataset comprises 601 variants (allele frequency <1%) in 13 genes (CFH, CFI, CD46, C3, CFB, CFHR1, CFHR3, CFHR5, CFP, PLG, DGKE, THBD, ADAMTS13). The database provides enhanced search tools and the ability to assess variant pathogenicity using: • Comparison of the estimated disease-associated variant allele frequencies to those of matched variants in 3 control datasets, namely “The 1000 Genomes Project,” the “ExAC” and the “Exome Variant Server (EVS)” using Chi-square analyses. • The mapping of missense variants onto protein structural models. • Determination of the pathogenicity of each variant from its residue position, amino acid property, binding sites and residue conservation using PolyPhen-2, PROVEAN and SIFT. • Measurement of the degree to which amino acid properties at any residue position are conserved across evolution, both within humans, and across other species, by multiple sequence alignment methods.
Thus, the tools in this new version of the database enable us to identify rare variants of the complement proteins enriched in disease which are worthy of further functional analysis. We will describe their occurrence in these 13 proteins, and discuss the utility of this new database for characterising complement-mediated diseases. http://dx.doi.org/10.1016/j.imbio.2016.06.221
205
206
The solution structure of the human complement regulator CFHR5 reveals a compact dimeric structure by X-ray scattering and analytical ultracentrifugation
Database of Complement Gene Variants: A comprehensive database providing insights on function, structure and allele frequency for genetic variants identified in complement-mediated diseases
Nilufar Kadkhodayi-Kholghi 1,∗ , Jayesh Gor 1 , Anna Ferlin 2 , Lindsay C. McDermott 3 , Daniel P. Gale 2 , Stephen J. Perkins 1 1
Department of Structural and Molecular Biology, University College London, London, United Kingdom 2 UCL Centre for Nephrology, Royal Free Hospital, University College London, London, United Kingdom 3 Department of Life Sciences, University of Bedfordshire, Luton, United Kingdom Human Complement Factor H Related 5 (CFHR5) belongs to the same complement regulator family as Factor H. The CFHR5 protein comprises a linear sequence of nine short complement regulator (SCR) domains. A mutation of CFHR5 that results in duplication of the N-terminal SCR-1/2 domain pair causes CFHR5 nephropathy, a common cause of kidney failure in Cypriots. To clarify the molecular basis of CFHR5 nephropathy, the domain arrangement of full-length CFHR5 was studied by analytical ultracentrifugation and X-ray scattering. Sedimentation velocity reported a molecular mass of 134 kDa, indicating that CFHR5 is dimeric with 18 SCR domains and not nine. The sedimentation coefficient of 5-6 S for CFHR5 decreased with increase in NaCl concentration, showing that its domain conformation had become more extended in high ionic strength. X-ray scattering also showed that CFHR5 was dimeric to confirm the ultracentrifugation result. The X-ray mean radius of gyration RG was 5.5 ± 0.2 nm, and its maximum length was 20 nm. This length of 20 nm for a protein with 18 SCR domains is low compared to that of 32 nm for Factor H with 20 SCR domains, indicating that CFHR5 possesses a more compact SCR arrangement than that seen for Factor H. The scattering curve modelling of CFHR5 was performed using a new modelling procedure called SASSIE that involved molecular dynamic simulations to generate physicallyrealistic atomistic SCR structures for CFHR5. The best-fit modelling structures confirmed that CFHR5 possessed a folded-back compact domain structure. We expressed the single SCR-1 domain and the SCR-1/2 domain pair of CFHR5 to show from analytical ultracentrifugation that SCR-1 was monomeric, while SCR-1/2 was dimeric, thus locating the CFHR5 dimerization site to its N-terminus. We are currently performing functional experiments with the SCR-1 and SCR-1/2 domains to clarify their role in CFHR5 activity. In summary, our results indicate a markedly more compact solution structure of CFHR5 than previously thought, and we have located its dimerization site to SCR-1/2 which may have a major role in causing CFHR5 nephropathy. http://dx.doi.org/10.1016/j.imbio.2016.06.220
1223
Amy J. Osborne 1,∗ , Santiago Rodriguez de Cordoba 2 , Veronique Fremeaux-Bacchi 3 , Marina Noris 4 , Richard J. Smith 5 , Bert van den Heuvel 6 , Timothy H.J. Goodship 7 , Stephen J. Perkins 1 1 Department of Structural & Molecular Biology, University College London, UK 2 Departamento de Inmunologia, Centro de Investigaciones Biologicas (CSIC), Madrid, Spain 3 Service d’Immunologie Biologique, Hopital Europeen Georges Pompidou, Paris, France 4 Laboratory of Immunology and Genetics of Rare Diseases and Transplantation, Mario Negri Institute for Pharmacological Research, Italy 5 Iowa Institute of Human Genetics, University of Iowa, Iowa City, USA 6 Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands 7 Institute of Genetic Medicine, Newcastle University, UK
Atypical haemolytic uremic syndrome (aHUS) and C3 glomerulopathy (C3G) are associated with dysregulation of the alternative pathway (AP) of complement. Recently, we described an updated compilation of 324 genetic variants identified in four AP genes encoding the proteins factor H (FH), factor I (FI), membrane cofactor protein (MCP, CD46) and C3 in our interactive FH-HUS web-database (www.fh-hus.org). Our database has been updated further to a new version, the ‘Database of Complement Gene Variants’ (www.complement-db.org), with variant data contributed by six centres (Newcastle, Paris, Bergamo, Madrid, Iowa City, and Nijmegen) from >1000 patients with aHUS and C3G. This laboratory dataset comprises 601 variants (allele frequency <1%) in 13 genes (CFH, CFI, CD46, C3, CFB, CFHR1, CFHR3, CFHR5, CFP, PLG, DGKE, THBD, ADAMTS13). The database provides enhanced search tools and the ability to assess variant pathogenicity using: • Comparison of the estimated disease-associated variant allele frequencies to those of matched variants in 3 control datasets, namely “The 1000 Genomes Project,” the “ExAC” and the “Exome Variant Server (EVS)” using Chi-square analyses. • The mapping of missense variants onto protein structural models. • Determination of the pathogenicity of each variant from its residue position, amino acid property, binding sites and residue conservation using PolyPhen-2, PROVEAN and SIFT. • Measurement of the degree to which amino acid properties at any residue position are conserved across evolution, both within humans, and across other species, by multiple sequence alignment methods.
Thus, the tools in this new version of the database enable us to identify rare variants of the complement proteins enriched in disease which are worthy of further functional analysis. We will describe their occurrence in these 13 proteins, and discuss the utility of this new database for characterising complement-mediated diseases. http://dx.doi.org/10.1016/j.imbio.2016.06.221