Introduction
The UK Adult Twin Registry (TwinsUK) is a cohort of volunteer adult twins from all over the United Kingdom. The Department of Twin Research and Genetic Epidemiology at St. Thomas’ Hospital, King's College London (KCL) hosts the registry, which started in 1992 via media campaigns targeted at middle-aged women. The success of early studies led to rapid evolution of the registry and it now incorporates twins, both male and female, from other sources such as the Aberdeen Twin Registry and Institute of Psychiatry Adult Registry. The primary focus of study has been the genetic basis of complex diseases (cardiovascular, metabolic, musculoskeletal, and ophthalmologic diseases), which has broadened to include the complex healthy ageing process. The third health check of the volunteer twins has provided longitudinal data that, alongside with the state-of-the-art ‘omics’ technologies data, can significantly advance the field of genetic and clinical epidemiology of ageing. We have previously described the design of the original twin registry, facilities and procedures for data collection, clinical and biological assessments, and main findings in 2006 (Spector & Williams, Reference Spector and Williams2006). The current article presents a brief update on the study procedures and recent technological applications in our cohort. More detailed description of phenotypes, research projects and collaborations, papers, and study findings can be accessed through our updated study Web site (http://www.twinsuk.ac.uk).
The Collection
The TwinsUK registry now consists of about 12,000 monozygotic (MZ) and dizygotic (DZ) twins aged 18 to 103 years (Table 1). About 83% of the registry is female (mean age of 55 years). The registry now contains 51% MZ and 49% DZ twins. Between 1992 and 2004, twins were invited for a full comprehensive visit and several project-led studies. More than 7,000 twins responded to some of the annual questionnaires and 5,725 attended a comprehensive visit. Apart from a lifelong lower weight in MZ twins of around 1 kg, all other age-matched characteristics of these volunteer twins were found not to differ from a singleton population-based cohort of British women (Chingford study; Andrew et al., Reference Andrew, Hart, Snieder, de, Spector and MacGregor2001). Between April 2004 and May 2007, all the 6,740 active twins on the registry were invited for a 1-day clinical visit, of whom 3,725 twins attended and 1,299 twins posted their blood DNA samples via their general practitioners. The age of participants ranged between 18 and 82 years (mean 52.5 ± 13 years) and 3,299 of the clinic attendants (89%) were female.
TABLE 1 The UK Adult Twin Registry Update
FF = female–female, MM = male–male, FM = female–male.
The second follow-up visit, also known as the HATS (Healthy Ageing Twin Study) visit, started in August 2007. Only women aged ≥40 years with at least one previous clinical visit (n = 4,610) were invited for this visit. In total, 3,125 women (mean age 59.6 ± 9 years, 48% MZ twins) attended the clinic (response rate = 68%). Follow-up time between first and last visits ranged between 6.1 and 17.4 years (mean 11.2 ± 2 years). Six hundred of the participants in this visit had four or more previous clinical visits. Compared to the baseline and first follow-up visits, participants in the HATS visit appeared to have higher socio-economic status, lower self-rated health status, and be more health aware given their level of alcohol intake and smoking. Comparison of the respondents and non-respondents to the HATS invitation showed that the non-respondents were generally younger and of lower socio-economic status. However, no significant clinical differences were observed between attendants and non-attendants to the HATS visit (Moayyeri et al., Reference Moayyeri, Hammond, Valdes and Spector2012b), suggesting that the participants in this visit can be considered as representative of the original population in the study. Zygosity status was assessed for all twins at the time of registration by a ‘peas in a pod’ questionnaire and confirmed via subsequent genotyping or genome-wide association studies.
Longitudinal Data
Longitudinal data is available for a wide range of phenotypes based on different questionnaires, clinical tests, and biochemical assays performed at different stages of the study (Moayyeri et al., Reference Moayyeri, Hammond, Valdes and Spector2012b). Table 2 summarizes some of these clinical measures showing the breadth of the available data and the potential for studies on various aspects of healthy ageing in this population. For instance, longitudinal changes in bone mineral density in more than 4,000 twins over an average of 17.5 years serves as a valuable endpoint for clinical and genetic epidemiological studies (Moayyeri et al., Reference Moayyeri, Hammond, Hart and Spector2012a). During the study follow-up, care has been taken to perform key clinical tests with similar protocols across all visits. Incident clinical endpoints (e.g., cardiovascular events, stroke, fractures, osteoarthritis, and different cancers) have been assessed over the course of study using questionnaires. Twins are all now registered using their National Health Services (NHS) numbers with the Office for National Statistics of England for retrospective analysis and future follow-up regarding their cancer and mortality status.
TABLE 2 Longitudinal Data Available in the TwinsUK Registry Participants
BMD = bone mineral density, DXA = dual-energy X-ray absorptiometry, QUS = quantitative ultrasound.
Novel Molecular and Genetic Phenotypes
Alongside with the conventional epidemiological phenotypes assessed by questionnaires and clinical visits, the TwinsUK registry benefits from generous and continued donation of biological samples by its volunteering participants. The methods for collection of these biological samples have been described previously, and the updated figures are presented in Table 1. Recently, a wide array of the latest ‘omics’ advances has been applied to subsections of these samples that makes TwinsUK one of the most uniquely phenotyped and deeply genotyped populations in the world. Here we describe some of these advances.
Genome-Wide Association Studies
TwinsUK has contributed to many international consortia for genome-wide association analysis of various phenotypes. Genome-wide scan data using two chips (Illumina HumanHap300 BeadChip and Illumina HumanHap610 QuadChip) are available for 5,710 twins. The data have been fully imputed using ‘HapMap II’ and ‘1000 Genomes’ reference panels (containing ~2.5 and ~16 million single nucleotide polymorphisms, respectively). TwinsUK is a member of many ongoing international consortia for meta-analysis of various traits, such as GIANT, CHARGE, ENGAGE, GEFOS, SUNLIGHT, MolPAGE, VisiGEN, TreatOA, and SpiroMeta (Aulchenko et al., Reference Aulchenko, Ripatti and Lindqvist2009; Dehghan et al., Reference Dehghan, Dupuis, Barbalic, Bis, Eiriksdottir, Lu, Pellikka and Chasman2011; Duffy et al., Reference Duffy, Iles, Glass, Zhu, Barrett, Höiom and Montgomery2010; Dupuis et al., Reference Dupuis, Langenberg, Prokopenko, Saxena, Soranzo, Jackson and Barroso2010; Elks et al., Reference Elks, Perry, Sulem, Chasman, Franceschini, He and Murray2010; Evangelou et al., Reference Evangelou, Valdes and Kerkhof2011; Ganesh et al., Reference Ganesh, Zakai and van Rooij2009; Hysi et al., Reference Hysi, Young and Mackey2010; Kolz et al., Reference Kolz, Johnson and Sanna2009; Lango et al., Reference Lango, Estrada and Lettre2010; Lindgren et al., Reference Lindgren, Heid and Randall2009; Newton-Cheh et al., Reference Newton-Cheh, Johnson, Gateva, Tobin, Bochud, Coin and Munroe2009; Nolte et al., Reference Nolte, Wallace, Newhouse, Waggott, Fu, Soranzo and Bradbury2009; Padmanabhan et al., Reference Padmanabhan, Melander, Johnson, Di Blasio, Lee, Gentilini and Dominiczak2010; Panoutsopoulou et al., Reference Panoutsopoulou, Southam, Elliott, Wrayner, Zhai, Beazley and Loughlin2011; Repapi et al., Reference Repapi, Sayers, Wain, Burton, Johnson, Obeidat and Tobin2010; Richards et al., Reference Richards, Kawoura, Rivadeneira, Styrkársdó;ttir, Estrada, Halldórsson and Spector2009a, Reference Richards, Waterworth, O'Rahilly, Hivert, Loos, Perry and Spector2009b; Smith et al., Reference Smith, Chen, Dehghan, Strachan, Basu, Soranzo and O'Donnell2010; Speliotes et al., Reference Speliotes, Willer and Berndt2010; Willer et al., Reference Willer, Speliotes, Loos, Li, Lindgren, Heid and Hirschhorn2009; Zhai et al., Reference Zhai, van Meurs, Livshits, Meulenbelt, Valdes, Soranzo and Spector2009, Reference Zhai, Teumer, Stolk, Perry, Vandenput, Coviello and Wallaschofski2011).
Next-Generation Sequencing
The UK10K study is an ongoing collaboration between TwinsUK study, the Wellcome Trust Sanger Institute, and several other collaborators for using the state-of-the-art, next-generation sequencing methods to uncover rare genetic variants associated with health and disease. The study involves whole-genome sequencing of 4,000 healthy people with well-documented physical characteristics (2,000 twins from TwinsUK and 2,000 children from ALSPAC study) and whole-exome (protein-coding regions of DNA) sequencing of 6,000 people with extreme health problems (obesity, neurological problems, and rare diseases). At present, all twin samples have been sequenced (6 × depth) and both phenotype and sequence data will be publicly available soon. More details about the study can be accessed at: www.uk10k.org. Moreover, about 1,000 exome sequences at 30–60 × depth have been performed for twin participants as part of projects with Pfizer and the GoT2D consortium. Over 2,000 Exome chips are currently being performed, mainly for control purposes in other consortia.
Epigenetic Markers
The first epigenetic assessment in TwinsUK was performed on DNA methylation patterns using Illumina HumanMethylation27 BeadChip in a sample of 172 female twins. This array examines 27,578 promoter CpG-sites that map uniquely across the genome, and some of these sites were found to be associated with age and age-related phenotypes (Bell et al., Reference Bell, Tsai, Yang, Pidsley, Nisbet, Glass and Deloukas2012). Currently, the Infinium HumanMethylation450 BeadChip (Illumina) is being applied to 500 additional MZ and DZ twin pairs to generate higher-resolution genome-wide DNA methylation profiles. This array includes 485,764 cytosine positions (CpG dinucleotides and CNG sites) across the human genome. Meanwhile, the major ongoing epigenetic project using the TwinsUK population is the EpiTwin study (http://www.epitwin.eu), which uses MeDIP (Methylated DNA immunoprecipitation) sequencing in whole blood samples (Bell & Spector, Reference Bell and Spector2011). This is the largest epigenetic project of its kind, in collaboration with the Beijing Genomics Institute, aiming to assay epigenomic differences in 5,000 adult UK twins aged 16–85 years, discordant and concordant for a wide variety of diseases and environments. Next-generation sequencing has the potential to prove powerful in detecting disease-related methylation differences at a high level of resolution in a sample of this size. The initial targets of the study include obesity, diabetes, allergy, heart disease, osteoporosis, depression, and longevity, but the method can be applied to every common trait or disease.
Gene Expression Measures
Eight hundred fifty-six twins with detailed clinical profiles have been biopsied during the HATS clinical visit. This has been done in the context of the MuTHER (Multiple Tissue Human Expression Resource) project, which is a Wellcome Trust funded study designed to understand the mechanisms involved in common trait susceptibility via gene expression across multiple tissues (Nica et al., Reference Nica, Parts, Glass, Nisbet, Barrett, Sekowska and Spector2011). Gene expression in three tissues of skin, fat, and lymphoblastoid cell lines have been measured using Illumina's whole genome expression array (HumanHT-12 version 3) containing 48,803 probes in three technical replicates. Results for expression quantitative trait loci analysis in 856 twins are freely available in the Web site (http://www.muther.ac.uk), and the main paper has recently been published (Grundberg et al., Reference Grundberg, Small, Hedman, Nica, Buil, Keildson and Spector2012). All of these tissues are now being ribonucleic acid (RNA) sequenced as part of the EuroBATS project (Biomarkers of Ageing using whole Transcriptome Sequencing), which is a 3-year European (EU-FP7) project started in January 2011. EuroBATS aims to discover novel biomarkers of ageing by incorporating novel RNA sequencing, telomere measurement, and bioinformatics techniques (http://www.eurobats.eu). Sequencing is being performed using Illumina GAIIx platform and the standard protocol of RNAseq (expecting 10–20 million reads per sample).
Telomere Length
Telomere length, as a marker of cellular senescence and subsequent cell death, was first measured in 3,256 twins with available genome-wide scans. These measures were derived from the mean of the terminal restriction fragment length by using the Southern blot method on DNA extracted from peripheral leukocytes. This data has contributed to detection of several genes implicated to affect biological age (Codd et al., Reference Codd, Mangino, van der Harst, Braund, Kaiser, Beveridge and Samani2010; Mangino et al., Reference Mangino, Richards and Soranzo2009). Recently, a larger sample (4,899 twins aged 16–99 years) has been assessed for telomere length using an established and validated quantitative polymerase chain reaction technique. Quality control has now been finalized and the data is available for collaborations.
Metabolomic Profiles
In 2009, fasting serum concentrations of 163 metabolites were measured for 1,270 twins using electrospray ionization tandem mass spectrometry (Biocrates AbsoluteIDQ technology). This targeted panel of metabolites covers a wide range of known lipids, amino acids, sugars, acylcarnitines, and phospholipids. This data has been used in several outstanding studies (Zhai et al., Reference Zhai, Wang-Sattler, Hart, Arden, Hakim, Illig and Spector2010). More recently, a larger sample of 6,055 twins has been assessed using a new method of non-targeted metabolomic analysis. This new platform (Metabolon Inc., Durham, USA) incorporates two separate ultra-high performance liquid chromatography/tandem mass spectrometry injections (optimized for basic and acidic species) and one gas chromatography/mass spectrometry injection per sample. The platform has detected and quantified concentration of 510 small molecules (299 known and 211 unknown molecules) including amino-acids, lipids, carbohydrates, vitamins, nucleotides, peptides, xenobiotics, and steroids. Genome-wide association studies of ~37,000 traits from 60 biochemical pathways in a subsample of this population has identified several genes involved in metabolic individuality in humans and promises significant advances in future functional studies (Suhre et al., Reference Suhre, Shin, Petersen, Monda, Thorleifsson, Jackson and Loos2011).
Microbiomics
We have recently started a collaborative National Institutes of Health-funded study with Cornell University, aiming to collect gut flora DNA for analysis with 16S sequencing technology in 5,000 twins. In addition, pilot data assaying microbiome diversity from other human body sites, such as skin, oral, and nasal cavities, as well as gut flora metagenomics are currently underway.
Future Directions and Collaborations
Frequent data collection with detailed clinical, biochemical, behavioral, and socio-economic characterization of participants for about two decades provides the opportunity to look at the prospective single-measure and repeated-measure associations for complex diseases and domains of healthy ageing in TwinsUK population. This also offers a unique opportunity to explore personalized medicine. Data collection, database management, biological sample storage, and statistical quality control have been carried out to a high standard. Blood, urine, and DNA sample aliquots from all visits are available for future measurements. We currently use online questionnaires and are actively engaging with our twin participants via e-mail and social networking Web sites. Our ‘Volunteer Advisory Panel’ helps informed decisions about the ethics, practicalities, and appropriateness of potential studies.
The TwinsUK registry has a history of numerous successful scientific collaborations, and we remain committed to providing the scientific community with access to the phenotype data from the ‘TwinsUK Resource’. A recent Biomedical Resource Grant from the Wellcome Trust is continuing to fund the core functions of TwinsUK. This will enable access to our publicly funded data from the wider scientific community (the ‘resource’), which will be separate from the individual research projects performed by academics within the Department of Twin Research at KCL. The TwinsUK Resource is opening up access of data, currently harmonizing and standardizing phenotypic data collected over the last 20 years; a subset of the data is already available for full open access via our Web site (http://www.twinsuk.ac.uk/data-access) and a search engine for the available phenotypes has been provided. We have an access committee, which meets weekly and reviews about 20 requests a month (www.twinsuk.ac.uk/data-access/submission-procedure). Researchers are encouraged to find out if TwinsUK resource can help them answer their research questions and get in contact for future collaborations.
Acknowledgments
The authors acknowledge financial support from the Wellcome Trust, the Department of Health via the National Institute for Health Research (NIHR) comprehensive Biomedical Research Centre award to Guy's and St Thomas’ NHS Foundation Trust in partnership with King's College London/Arthritis Research Campaign/EC FP6 Programme Grant-512066 (LSHG-CT-2004); MOLPAGE/EC Framework 7 Programme Grant 200800; Treat OA/EC Framework 7 Health-2007-A; ENGAGE/EC Framework 7 Health-2007-2.4.5-4; GEFOS/EC FP6 MRTN-CT-2006-034021; MyEuropia Research Training Network/Chronic Disease Research Foundation (CDRF)/Pfizer Pharmaceuticals/National Health and Medical Research Council (NHMRC)/National Institute of Aging (NIA)/Guide Dogs for the blind Association(GDBA)/Biotechnology and Biological Sciences Research Council (BBSRC). The authors acknowledge the funding and support of the National Eye Institute via an NIH/CIDR genotyping project (R01EY018246-01-1 PI: Terri Young). The authors would also like to thank all the twins who participated and supported this cohort and staff in the Department of Twin Research and Genetic Epidemiology, St. Thomas’ Hospital, King's College London.
