Written by Michael D. O’Neill
The research team from Sweden’s Uppsala University has unveiled an important finding that could change the way we work with biobank blood – that is, the storage time of a blood sample is as significant of a factor as the age of the individual at sampling. Furthermore, the study has also found that the season of sample collection can have effect on the protein levels in the blood plasma.
This interesting research is available as an open-access article titled, “Effects of Long-Term Storage Time and Original Sampling Month on Biobank Plasma Protein Concentrations”. It was published in the October 2016 issue of EBioMedicine journal. The chief conclusion of the Uppsala study is that storage time is a significant factor determining the plasma protein concentration variation in frozen biobank samples from healthy women, and therefore due to its similar effect sizes as individual age it should be included as a covariate parameter in future epidemiological studies. Another major finding of the study is that the protein levels in the blood plasma vary depending on the season or month in which the samples were taken. Some of these differences can be explained by the amount of sunlight subjects were exposed to at the time of sampling.
Before diving into the detail of the study, let’s review the current state of biobank research in general.
Biobanks are collections of human biological tissue specimens and related health data (Hawkins, 2010). Samples of bodily fluid or tissue collected may be kept for several years or even indefinitely, to enable long term future research and advancement in medicine. The stability and quality of biomolecules in clinical biobank is crucial to its applications in life sciences and research. The samples must be stored and maintained in a way that minimize deterioration and damage over time. The ability to maintain the biomolecular composition of the sample at time of collection over the long years of its storage allows important application for longitudinal studies and for retrospective studies, especially in the cases of research on rare-diseases where large sample numbers are difficult to collect at any given time point. It is understandable, then, that research findings on possible quality disruptors and strategies to minimize quality deterioration will help greatly in optimizing quality control of biobank samples. It is well known from previous medical research and publication that age, sex and health factors of the individual of the collected samples are all important factors to consider. There have also been studies showing the effect of the number of freeze-thaw cycles on different proteins (Lee et al., 2015), sample handling and the short-term effects of storage protocols prior to sample freezing (Skogstrand et al., 2008), and other intra-individual variation (Stenemo et al., 2016).
So how was this study by Stefan Enroth (PhD, Associate Professor of Computational Genomics in the Department of Immunology, Genetics and Pathology at Uppsala) special? There has not been a study conducted on large set of proteins for any of the above parameters, including storage time and seasonal variations. Here is a detailed overview of the study.
Researchers Stefan Enroth, Goran Hallmans (Department of Biobank Research, Umea University), Kjell Grankvist (Department of Medical Biosciences, Clinical Chemistry, Umea University) and Ulf Gyllensten (Department of Immunology, Genetics, and Pathology, Biomedical Center, SciLifeLab Uppsala, Uppsala University), analysed 380 different plasma samples from 106 women, which were collected in the Vasterbotten Intervention Program (Hallmans et al, 2003). The age of these women ranged from 29 to 73. The oldest sample was collected in 1988 and the most recent sample was collected in 2014. With the exception of July, the samples were collected over the entire year. In Sweden, July is the month of Summer vacation and that is possibly the reason why there were no samples collected during this month. In order to truly investigate the impact of storage time, the research team then took only the samples from 50-year-old women. A total of 108 proteins were analysed from the samples of 92 individuals. Proximity Extension Assay (PEA) multiplex assays were used to quantify plasma protein abundance levels. After the analysis and the quality control, the plasma samples were pre-processed into Normalized Protein eXpression (NPX). The NPX is an arbitrary unit in which a high value means higher protein expression. Statistical analyses were processed in R (R Development Core Team, 2014). The effect of storage time on protein levels was calculated with a linear model, using the years post original sample collection date as the variable and protein abundance levels as the response.
The results were surprising.
The analysis showed that storage time accounted for up to 35% of variance in a single protein. More specifically, it affected 18 proteins and accounted for 4.8% - 34.9% of the variance observed. Now, that variation is due to storage time alone. From previous studies it is known that age can account for around 27% of the variation in a single protein. In this study results, it was observed that the chronological age at sample collection affected 70 proteins and was responsible for 1.1 – 33.5% of the variance. The season (month of the year, which had varying levels of sun exposure hours) affected 36 proteins and accounted for up to 4.6% of the variation in protein abundance level.
Since this study is the first to investigate the effect of storage time on the levels of plasma protein, the results cannot be compared to others. Previous studies on the effect of UV radiation by Reichert et al. (2015) and Vostalova et al. (2013) provide valuable insight into the possible effect of this parameter on human plasma samples.
"This discovery will change the way the entire world works with biobank blood". Well, it’s a no surprise that Stefan Enroth said so, because the findings that him and his research team has made has important implications and impact on future drug research. It will be an interesting and exciting space to watch as these new findings make changes to the way biobank samples are handled and key parameters are included in future epidemiological studies.
Reference
1. Uppsala Press Release
http://www.uu.se/en/news-media/press-releases/press-release/?id=3392&area=3,8&typ=pm&lang=en
2. EBioMedicine Article
https://www.ebiomedicine.com/article/S2352-3964(16)30394-2/fulltext
3. Hallmans G, Agren A, Johansson G, Johansson A, Stegmayr B, Jansson JH, Lindahl B, Rolandsson O, Soderberg S, Nilsson N, Johansson I, Weinehall L. Cardiovascular disease and diabetes in the Northern Sweden health and disease study cohort — evaluation of risk factors and their interactions. Scand. J. Public Health Suppl. 2003;61:18-24
4. Hawkins AK. Biobanks: Importance, Implications and Opportunities for Genetic Counselors. J Genet Counsel. 2010;19:423-429. doi:10.1007/s10897-010-9305-1
5. Lee JE, Kim SY, Shin SY. Effect of Repeated Freezing and Thawing on Biomarker Stability in Plasma and Serum Samples. Osong Public Health Res Perspect. 2015;6(6):357–362. doi:10.1016/j.phrp.2015.11.005
6. R Development Core Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria (2014)
7. Skogstrand K, Charlotte K. Ekelund CK, Poul Thorsen P, Ida Vogel I, Bo Jacobsson B, Bent Nørgaard-Pedersen B, David M. Hougaard DM. Effects of blood sample handling procedures on measurable inflammatory markers in plasma, serum and dried blood spot samples. J Immunological Methods. (2008); 336(1):78-84. doi:10.1016/j.jim.2008.04.006.
8. Stenemo M, Teleman J, Sjostrom M, Grubb G, Malmstrom E, Malmstrom J, Nimeus E. Cancer associated proteins in blood plasma: determining normal variation. Proteomics. 2016;16(13): 1928-1937, doi:10.1002/pmic.201500204