Collection and storage of biological samples for future study is a controversial component of several 'cutting-edge' initiatives such as the UK Biobank, the Swedish National Biobank program and the Genetic Alliance Biobank. A biobank is a repository for genetic material, with associated information describing or characterising the people to whom the genetic material (usually DNA) belongs.
Researchers new to biobanks will encounter a wealth of often conflicting information about how to archive genetic material. This article provides an overview of the topic, concentrating on the nitty gritty of molecular archiving.
We now live in an era which follows the Human Genome Project. A wealth of information has been delivered into the hands of clinicians and human health scientists. The limiting step for the translation of this data into clinical practice is the linking of disease phenotype with genotype and understanding the complex interactions occurring at a genomic level. One of the main obstacles to establishing this link in complex polygenic diseases is the difficulty dealing with the vast amount of data coming from clinical studies. Whole genome association studies and gene expression analysis using high density microarrays are examples. These problems, however, are being overcome with the establishment of bioinformatics as a new discipline in statistics. What is becoming clear from research thus far is that large datasets derived from small numbers of patients do not produce the results required to reach statistical significance and the rate of false discoveries is high. The establishment of large Biobanks is the only method to overcome this shortfall. The deCODE and Utah population database are two examples of biobanks that are ahead of their time in realising the potential for large Molecular repositories. The establishment of the United Kingdom Biobank, funded by the Wellcome trust, will dwarf both of these by enrolling 500,000 individuals into a database that hopes to address complex diseases like cancer, heart diseases, diabetes, arthritis, forms of dementia and more. What should not be forgotten in this enthusiasm to address common complex diseases are complex orphan diseases that account for significant morbidity in those that suffer from them. Establishing single biobanks that will have sufficient power to address the issues with these diseases will not be possible. There is therefore a need fore a more global collaborative effort to collect data and biological samples, allowing core labs to analyse associations between phenotype and genotype.
Samples used in biobanks may come from a variety of sources with advantages and disadvantages for each. It is possible to collect complete nucleic acid samples of both DNA and RNA in one collection.
|Hair and skin||
If collected appropriately, e.g. using a prepared medium (such as Whatman® FTA elute paper) blood spots have the most ideal profile for collection. Extensive experience has been obtained using Guthrie cards in the postnatal testing of infants for monogenic diseases and diseases of inborn errors of metabolism.
A number of methods can be used to stabilise samples prior to nucleic acid extraction. For DNA, optimal conditions for sampling from any source is at ambient temperatures of less than 24 degrees C. Samples should not be exposed to direct sunlight. Handling should be minimised to prevent contamination. Stabilisation media exist for immediate preservation of nucleic acids, such as Whatman® FTA elute paper. This paper is embedded with proprietary factors that lyse cells to release nucleic acids, prevent bacterial growth and trap inhibiting substances, such as haemoglobin, in a matrix. This medium also allows one step extraction, using a punch card system, and elutes into DNAse/RNase free water. The presence of the inhibitors in this process means that the extracted nucleic acids are suitable for many downstream applications that require high fidelity samples. The added advantage of this system is that other biological tissues can be stored on the same medium e.g. buffy coat, buccal swab or homogenised organ tissue.
An alternative to this method that allows all-in-one sampling is plasma and buffy coat extraction from whole blood. Ideal sampling of whole blood should be performed using a CPDA (yellow top) vacutube which optimises DNA and RNA retrieval. After centrifugation at ~3,000 g for 5 mins, preferably in a refrigerated rotor, plasma, and buffy coat can be removed from a sample simultaneously. Buffy coat can be preserved in RNAlater® which recently has not only been shown to preserve RNA, but also yields DNA when an extraction process is applied. Buffy coat in RNAlater® is preferably stored in a low temperature freezer i.e. -80°C freezer.
Samples specifically taken for gene expression can be stabilised in RNAlater® which has proven to be an extremely effective method of halting transcription and stabilising RNA for prolonged periods, even at room temperature. An alternative to this approach, in circumstances where immediate processing and extraction are not available, is to use a prepared blood collection tube such as the Qiagen PAXGene system or Ambion’s Ribopure. Using these collection tubes cellular transcription can be halted until a formal extraction process can be performed some hours later.
Traditional methods of DNA extraction use hazardous chemicals and solvents such as chloroform and phenol. Using these substances in a laboratory requires specific safety measures and handling protocols. At the very least a fume hood is necessary when handling these compounds; this may increase the cost for any low budget operation. Other methods of extraction include salting out methods, and matrix elution. Newer technologies such as magnetic bead extraction systems avoid any toxic chemicals and are used in robotic systems that allow high throughput extraction processes without multiple pipetting steps. Robotic systems provide consistent extractions which has benefits for massively paralleled downstream applications. Though initially expensive they become cost effective in a high throughput environment.
The Biocompare website at www.biocompare.com is a good place to start to assess the extensive range of extraction kits and other scientific supplies.
Extraction from solid tissue poses some added challenges. Steps to prepare the sample are required which include homogenization and cellular lysis. Added steps increase the chances of disruption of RNA, particularly by shearing forces from the homogenizer. Increased chances of contamination also become an issue.
Adequate storage must be as secure, effective and robust over time in order to respect the donors who supplied samples, and the effort invested by researchers.
Any archival system must include a systematic storage and retrieval system. Common sense dictates that this requires a marker system with identifiers for each sample e.g. barcoding or numbering. If samples are to be stored at low temperature, possibly with other solvents, then the identifying markers must be resistant to smudging and dissolution etc. The registration of each sample entering and exiting the system must be centrally stored, preferentially on a computer-based system that can be backed up frequently. The physical location of each sample should be registered to allow the rapid location of specimens.
The archival system should de-identify samples to respect the privacy of donors and allow blinding of researchers to analysis. For the same reasons the database including clinical data should be kept separately with a method to link clinical information to tissue samples that cannot be easily accessed. A separate electronic key, held by a curator of the database, would be ideal to prevent tampering.
Software driving these systems should be encrypted or password protected to further ensure privacy and confidentiality to donors.
Room temperature storage
Storage at room temperature is the ideal method for nucleic acid storage but may not be suitable for other biological tissues. Low temperature storage carries a number of disadvantages, with the possibility of freezer failure being the most critical. Costs of running, maintenance and the physical space required for freezers limit their capabilities. Because of these concerns there is a move towards room temperature storage which overcomes a lot of these problems.
'Bank in a Box' by Genvault® offers solid state, room temperature storage with identifiers for specimens contained embedded within samples, in a fully archived system. This comprehensive package comes at a relatively low startup cost. The system features GenPlates, the core of GenVault’s integrated biosample management system. GenPlates are 384 well plates which contain 6mm disc-shaped elements of Whatman FTA® paper in each well, allowing a wide variety of biosample types to be stored at room temperature. The footprint of the Genvault bank is two cubic feet making it compatible with many benchtops. This system is capable of storing 38,400 sample aliquots in about 1/10th the space of a -80°C freezer.
The identifier used with the system is known as GenCode, which is an oligonucleotide based tag that is embedded in each well of a GenPlate and remains with the DNA sample even after that DNA is removed from the FTA paper for permanent sample identification. Other room temperature storage systems exist but most use a similar platform.
Archived samples must be protected physically from accidental or intentional damage. Therefore basic protection against fire, water, and theft must be considered. Minimising access to only a few individuals reduces the chances of unexpected events. Replicates or split samples stored at a separate location also reduce the chances of loss of samples.
In the lifetime of a project researchers may enter and leave. Ownership of samples ultimately should always remain with the donor however the ownership of a collection is a complicated issue that needs to be predefined before banking begins. An independent curator, who controls access, may be necessary to manage the biobank.
As with the collection of any data or biological samples the rights, autonomy, cultural and gender specific issues for donors must be respected. The deCODE database of the Icelandic population is a case example of where donor rights have come into dispute. This database was originally established as a public owned database but after many samples had been collected it transformed into a privately owned entity. The company now collaborates with a number of pharmaceutical and diagnostics companies and is developing lead compounds and diagnostic tests for commercial use. The population of Iceland has retained some intellectual property of any discoveries made as drugs forthcoming from deCODE’s research must be offered to the Icelandic people free of charge.
For criticism of the UK Biobank, see this BBC article.