Identifying novel genes for late-onset Alzheimer's disease
Alzheimer's disease is a disease that causes the most common cause of dementia, accounting for about two thirds of cases in older people (ARUK). There are two types of Alzheimer's disease, rare early onset cases that generally occur in families (EOAD) and late onset cases that occur in individuals over 65 years of age (LOAD)
​
Late-onset Alzheimer’s disease (LOAD) is a complex genetic disorder and it is likely that multiple gene variants influence risk of developing this disease, with the main genetic risk factor being APOE ε4. In recent years a number of additional genes have been identified, for example CLU and PICALM.
The identification of risk factors for any disease contributes towards a greater understanding of the mechanisms underlying disease and may, in time, lead to the development of novel therapeutics.
​
From 2001 - 2005, we were involved in a collaborative study between the Institute of Psychiatry (Kings College London) and Cardiff University to identify novel risk factors for LOAD. As part of this study, we collected a range of samples (blood, urine, etc) from a large cohort of age and gender-matched Alzheimer's patients and healthy samples. These samples were then used to carry out a range of genetic and proteomic studies to identify novel risk factors and biomarkers for Alzheimer's disease.
Identifying the gene(s) responsible for the chromosome 10 linkage signal.
Previously, a region of chromosome 10 had been identified as likely harbouring a risk factor(s) for LOAD and a number of different studies were initiated to try and identify this gene(s).
​
The initial approach used bioinformatics to identify all genes in the region and to select candidates based on known function. The first gene we evaluated was VR22 (alpha T-catenin), which we showed was expressed in the brain and also interacted with the Wnt signalling pathway, but which ultimately did not account for the chromosome 10 linkage signal (Busby et al., 2004). The second candidate gene chosen, PTEN, was known to be a negative regulator of the PI3-kinase pathway, which targets GSK-3. However, association analysis of eight variants in this gene suggest that PTEN is unlikely to be the chromosome 10 risk variant (Hamilton et al., 2006). In addition to this work, as part of the larger study, genetic variants in a further 23 candidate genes were analysed, including SIRT1, ZWINT and SGPL1. Two SNPs in SGPL1 demonstrated marginal evidence of association with us concluding that these variants may confer susceptibility to LOAD (Morgan et al., 2007).
​
Later, we were involved in two large studies to investigate all genetic variation located in the chromosome 10 linkage region, regardless of gene function. The results from this work showed that variants in RPS3A (Grupe et al., 2006) and DKK1, ANK3, CTNNA3, CXXC6 (Morgan et al., 2008) may have a role in determining risk of LOAD.
​
​
A common genetic link between risk of late-onset AD and type II diabetes.
Epidemiological evidence supports the existence of a possible link between type II diabetes mellitus (T2D) and late-onset AD. Therefore, genes involved in the insulin signalling pathway are possible candidate susceptibility genes for LOAD. Following publication of a study to identify new risk genes for T2D (Barroso et al ., 2003), we performed a two-stage association study to investigate the role of these genes in the risk of developing LOAD.
​
In the first stage of our study, we genotyped 152 SNPs in LOAD (n=258) and healthy control (n=275) individuals. Eight polymorphisms showing statistical significance were then genotyped in a larger independent sample set (LOAD, n=696; control, n=838). Overall in this study, we identified genetic variants from three genes (SOS2, PPARgamma and PCK1) that showed significant association with LOAD (Hamilton et al., 2007b).
​
​
Identification of potential AD biomarkers.
We were also involved in a study to identify readily available biomarkers for AD to aid diagnosis or to monitor disease progression. Using cerebral spinal fluid (CSF), we carried out a comparative analysis of protein expression from initially 50 patients with AD and 50 healthy individuals using 2-D gel electrophoresis coupled with mass spectrometry. Initial observations were followed up in a larger validation sample set (n=511 total) and we showed that two proteins, complement factor H (CFH) and alpha 2 macroglobulin (a-2M), had increased expression that was specific to AD and also correlated with disease severity (Hye et al., 2006).
​
As a follow on to this work, we carried out an association study to determine whether genetic variation in the CFH gene increased risk of developing LOAD. We genotyped the Y402H polymorphism in 617 individuals with AD and 735 healthy controls, and our results demonstrate that the Y402H variant is not associated with LOAD (Hamilton et al., 2007a).
​
​
​
Developing a functional genomics approach.
From 2005 - 2011 Dr Hunter was awarded two personal fellowships, the first from the Alzheimer's Research Trust (now Alzheimer's Research UK) and the second from the Alzheimer's Society, to investigate the use of novel genomic methodology in identifying functional genomic variation for Alzheimer's disease.
​
One contributing cause to the development of AD is thought to be an increase in cellular Aβ species. Aβ species are increased in all forms of AD and it is possible that proteins involved in the generation and/or degradation of this peptide may influence risk of LOAD, particularly if they contain mutations that affect function. There are a number of proteins known to have such roles; for example, nicastrin (NCT) is a component of the γ-secretase complex which has a role in generating Aβ, while ECE-1, ACE and IDE are known degrading Aβ enzymes. Genetic studies of polymorphisms in these genes have produced results that been, as indeed they have been with many other genes, conflicting and, significantly, they do not take into consideration interaction(s) with other genes. Crucially, genetic association studies also neglect the functional effect of polymorphisms.
​
The development of a method called the infectious Bacterial Artificial Chromosome (iBAC) technique (Wade-Martins et al., 2000)) provides a potential way to functionally investigate genetic polymorphisms within the context of the whole gene. This is particularly important as not all polymorphisms positively associated with a disease are coding. The iBAC method is based on the packaging capacity of the Herpes Simplex Virus (HSV). The HSV capsid has a packaging capacity of 152kb and this can accommodate approximately 95% of human genomic loci (Hibbit and Wade-Martins, 2006). The principle of this method is to identify a gene of interest that is known to contain disease associated polymorphisms, then to identify a BAC containing the full genomic sequence of this gene (including all promoter and enhancer sequences), modify the BAC to create a series of BACs expressing different polymorphisms of interest, then to retrofit the BAC so it can be packaged to create viral particles that will deliver the gene of interest to your cell type of interest. Clonal cell lines can then be made for analysis and comparison of different polymorphisms.
​
Nicastrin. The first project that we carried out using this technology was the evaluation of a haplotype in the nicastrin gene. As a component of the gamma secretase complex, nicastrin is an extremely attractive candidate gene for Alzheimer's disease. There were also a number of conflicting association studies published regarding variation in this gene, with a number suggesting that nicastrin variants contribute to LOAD risk (Deary et al., 2005; Lupton et al., 2011).
Our 2012 paper (Hamilton et al., 2012) describes how we identified a number of BACs encompassing the nicastrin genomic loci and how we then sequenced the region to ensure a full comprehension of the haplotype region. A suitably sized BAC was selected and then sequentially modified, by homologous recombination, to engineer a series of five BACs expressing a different nicastrin haplotype. We created clonal cell lines by delivering nicastrin-BACs to nicastrin knockout cell lines and that cell lines expressed nicastrin, and that they had both rescued gamma secretase activity and amyloid beta production. Overall, we were not able to identify any robust functional differences between the nicastrin clonal cell lines and therefore conclude that it is unlikely that common variation at the nicastrin locus is a risk factor for LOAD.
​
Abeta degrading enzymes. The next project we carried out used the iBAC methodology was to investigate the functional effects of LOAD-associated sequence variation in Aβ degrading enzymes on Aβ levels, specifically ECE-1, NEP, IDE and PLAU. As with the nicastrin project, the first step of this project was to identify BACs of a suitable size containing the complete genomic region for each candidate gene, followed by sequential modification of BACs to generate each gene variants of interest. Due to technical difficulties creating and analysing clonal cell lines it was not possible to complete this project as expected. However, we did analyse promoter variation in the ECE-1 gene by cloning the promoter upstream of a luciferase reporter gene, modifying the promoter sequence and carrying out luciferase assays. Our results showed that an ECE-1 promoter polymorphism demonstrates tissue specific expression which may contribute to disease risk by decreasing ECE-1 expression (Hamilton et al., 2012b). Further, by carrying out a genetic association analysis of variants from ECE-1 in the Genetic and Environmental Risk in Alzheimer's Disease Consortium 1 (GERAD1) cohorts, we provide further support that ECE-1 variation contributes to AD risk (Hamilton et al., 2012b).
Current projects
With our GCU colleague (and PhD peer) Dr Fiona Kerr, we have recently (2019) been awarded a pilot grant from the Alzheimer's Research UK Scotland Network to investigate expression of UBA1 in Alzheimer's disease. In this project we will determine whether expression of UBA1 is altered in AD, using both patient samples and tissue obtained from a mouse model of AD. In the same samples, we will also investigate whether histone ubiquitination is altered in AD, with the aim being to determine whether common pathways might underlie at least two (AD and SMA) neurodegenerative conditions (Groen and Gillingwater, 2015).
Publications
Barroso I, Luan J, Middelberg RPS, Harding AH, Franks PW, Jakes RW, Clayton D, Schafer AJ, O’Rahilly S, Wareham NJ. (2003) Candidate gene association study in type 2 diabetes indicates a role for genes involved in beta-cell function as well as insulin action. Plos Biology 1(1):41–55.
​
Busby V, Goossens S, Nowotny P, Hamilton G, Smemo S, Harold D, Turic D, Jehu L, Myers A, Womick M, Woo D, Compton D, Doil LM, Tacey KM, Lau K-F, Al-Saraj S, Killick R, Pickering-Brown S, Moore P, Hollingworth P, Archer N, Foy C, Walter S, Lendon C, Iwatsubo T, Morris JC, Norton J, Mann D, Janssen B, Hardy J, O’Donovan M, Jones L, Williams J, Holmans P, Owen MJ, Grupe A, Powell J, van Hengel J, Goate A, Van Roy F, Lovestone S (2004) Alpha-T-catenin is expressed in human brain and interacts with the Wnt signalling pathway but is not responsible for linkage to chromosome 10 in Alzheimer’s disease. Neuromolecular Medicine, 5(2):133-46
​
Deary IJ, Hamilton G, Hayward C, Whalley LJ, Powell J, Starr JM, Lovestone S (2005) Nicastrin gene polymorphisms, cognitive ability level and cognitive ageing. Neuroscience Letters, 373(2):110-4.
​
Groen EJN and Gillingwater TH (2015) UBA1: At the crossroads of ubiquitin homeostasis and neurodegeneration. Trends in Molecular Medicine 21(10):622-632
​
Grupe A, Li Y, Rowland C, Nowotny P, Hinrichs AL, Smemo S, Kauwe JSK, Maxwell TJ, Cherny S, Doil L, Tacey K, van Luchene R, Myers A, Wavrant-De Vrièze F, Kaleem M, Hollingworth P, Jehu L, Foy C, Archer N, Hamilton G, Holmans P, Morris CM, Catanese J, Sninsky J, White TJ, Powell J, Hardy J, O’Donovan M, Lovestone S, Jones L, Morris JC, Thal L, Owen M, Williams J, Goate A (2006) A scan of chromosome 10 identifies a novel locus showing strong association with late-onset Alzheimer disease. American Journal Human Genetics, Jan;78(1):78-88.
​
Hamilton G, Samedi F, Knight J, Archer N, Foy C, Walter S, Turic D, Jehu L, Moore P, Hollingworth P, O’Donovan MC, Williams J, Owen MJ, Lovestone S, Powell JF (2006) Polymorphisms in the phosphate and tensin homolog gene are not associated with late-onset Alzheimer’s disease. Neuroscience Letters, Jun 19:401 (1-2):77-80
​
Hamilton G, Proitsi P, Williams J, O'Donovan M, Owen M, Powell J, Lovestone S (2007a) Complement Factor H Y402H polymorphism is not associated with late-onset Alzheimer's disease. Neuromolecular Medicine, 9(4):331-4.
​
Hamilton G, Proitsi P, Jehu L, Morgan A, Williams J, O’Donovan MC, Owen MJ, Powell JF, Lovestone S (2007b) Candidate Gene Association Study of Insulin Signalling Genes and Alzheimer’s Disease: Evidence for SOS2, PCK1 and PPARγ as Susceptibility Loci. AJMG part B: Neuropsychiatric Genetics, Jun 5;144(4):508-16
​
Hamilton G, Killick R, GERAD1 consortium, TGen consortium, Lambert J-C, Amouyel P, EADI, Carrasquillo MM, Pankratz VS, Graff-Radford NR, Dickson DW, Petersen RC, Younkin SG, Powell JF, Wade-Martins R (2012a) Functional and genetic analysis of haplotypic sequence variation at the nicastrin genomic locus. Neurobiology of Aging, 33(8):1848e.1-1848.e13
Hamilton G, Harris SE, Davies G, Liewald D, Tenesa A, Payton A, Horan MA, Ollier WER, Pendleton N, GERAD1 consortium, Starr JM, Porteous D, Deary IJ (2012b) The role of ECE-1 variants in cognition and Alzheimer’s disease. AJMG part B: Neuropsychiatric Genetics, 159B(6):696-709
​
Hibbit OC and Wade-Martins R (2006) Delivery of Large Genomic DNA Inserts >100 kb Using HSV-1 Amplicons. Current Gene Therapy, 6, 325-336 325
​
Lupton MK, Proitsi P, Danillidou M, Tsolaki M, Hamilton G, Wroe R, Pritchard M, Lord K, Martin BM, Kloszewska I, Soininen H, Mecocci P, Velas B, Lovestone S, Powell JF (2011) Deep sequencing of the nicastrin gene in pooled DNA, the identification of genetic variants that affect risk of Alzheimer’s disease. PLoS ONE, 5(2):e17298
​
Morgan AR, Turic D, Jehu L, Hamilton G, Hollingworth P, Moskvina V, Jones L, Lovestone S, Brayne C, Rubinsztein DC, Lynch A, Lawlor B, Gill M, O’Donovan MC, Owen MJ, Williams J (2007) Association analysis of 23 positional/functional candidate genes on chromosome 10 with Late Onset Alzheimer’s disease. AJMG part B: Neuropsychiatric Genetics, Mar 20;144B(6):762-770
​
Morgan AR, Hamilton G, Turic D, Jehu L, Harold D, Abraham R, Hollingworth P, Moskvina V, Brayne C, Rubinsztein DC, Lynch A, Lawlor B, Gill M, O’Donovan M, Powell J, Lovestone S, Williams J, Owen MJ (2008) Association analysis of 528 intra-genic SNPs in a region of chromosome 10 linked to late onset Alzheimer's disease. AJMG part B: Neuropsychiatric Genetics, Sep 5;147B(6):727-31
​
Wade-Martins R, White RE, Kimura H, Cook PR, James MR (2000) Stable correction of a genetic deficiency in human cells by an episome carrying a 115 kb genomic transgene. Nature Biotechnology, 18(12):1311-4