Nutrition proteomics and biomarker discovery

Abstract

Baukje de Roos is a principal investigator at the University of Aberdeen, Rowett Institute of Nutrition and Health. She investigates mechanisms through which dietary fats and fatty acids, and also polyphenols, affect parameters involved in the development of heart disease in vivo. This is achieved not only by measuring their effect on conventional risk markers for heart disease but also by assessing their effect on new markers that are being developed through proteomic and mass spectrometry methods. She obtained her PhD in Human Nutrition at Wageningen University, The Netherlands, in January 2000, after which she was appointed as a post-doctoral research fellow at the Department of Vascular Biochemistry, Glasgow Royal Infirmary, in collaboration with GlaxoSmithKline. In June 2001 she joined the Rowett Research Institute in Aberdeen. She is currently working for the University of Aberdeen, where her research is funded by the Scottish Government Rural and Environment Research and Analysis Directorate (RERAD). She is an active member of the European Nutrigenomics Organisation (NuGO), an EU-funded Network of Excellence, which merges the nutrigenomics activities of its 23 partners across Europe.

▪ Please provide the readers of Expert Review of Proteomics with an overview of your work in the field of proteomics

I am interested in processes involved in the development of cardiovascular disease, which is rapidly becoming a primary cause of premature death worldwide. Thus, the identification of dietary compounds that most effectively prevent this chronic disease is critical. Evidence from metabolic studies, prospective cohort studies and clinical trials in the past decades has indicated that intervention with certain dietary fatty acids but also with polyphenols from plant sources could be effective in preventing cardiovascular disease. I aim to assess the mechanisms by which dietary fatty acids, such as omega-3 fatty acids and conjugated linoleic acids, or polyphenols from, for example, cocoa and olives, affect the development of heart disease and related metabolic abnormalities in vivo. We do this not only by measuring their effect on existing risk markers but also by assessing their effect on new risk markers that are currently being identified through proteomics.

▪ What would you say are the most important traits of a biomarker, that increase its clinical value?

Biomarkers may reflect different stages in a biological process, ranging from healthy functioning to a deviation from a ‘healthy’ equilibrium to disease. In clinical practice, biomarkers of interest are diagnostic biomarkers that measure the incidence and progression of a disease process such as hemoglobin A_1C, or prognostic biomarkers that can predict whether a subject will be susceptible to a disorder, such as plasma cholesterol levels. Such biomarkers should have a high sensitivity and specificity for the outcome they are expected to identify. Ideally, they should explain a reasonable proportion of the outcome, independent of other established predictors and be mechanistically involved in the disease process. On a practical level, these biomarkers should be reproducibly obtained in a standardized fashion, acceptable to the patient, and easily interpreted by clinicians [1].

▪ What do you believe to be the greatest challenges when using biomarkers as determinants of disease progression or susceptibility in nutrition sciences?

Nutrition is likely to affect the early stages of disease development, but such mechanisms remain poorly understood. Often the tissues involved in such initial processes are not well defined, which may hamper early detection of chronic diseases and the influence of dietary interventions on these processes. Therefore, we should aim for the development of circulating biomarkers in plasma, urine or blood cells which allow detection of subtle changes in early-stage pathways in chronic diseases upon dietary interventions.

▪ What is the current opinion on the influence of nutrition on the development or prevention of disease?

We believe that 25–40% of all vascular disease and cancer deaths are caused by unhealthy diets and obesity. However, the link between diet and chronic diseases is complex and difficult to unravel. Our diet is made up of many different food compounds and nutrients and most of these may affect risk of developing vascular disease and cancer. In addition, our genes can also affect the way in which diet influences the development of chronic disease. Therefore, our aim should be to identify the mechanisms whereby specific nutrients or dietary regimes can prevent or delay the onset of chronic diseases. Such information could then be used in the formulation of credible recommendations for a healthy diet.

▪ To what extent do you feel that dietary intervention affects the protein complement of a biological system?

We know, from animal and human intervention studies, that diet can affect the protein complement very quickly (i.e., post-prandially). This may occur primarily in the liver, as this is the first port of call for many food compounds, but also may affect those proteins that are present in circulating blood cells, such as peripheral blood mononuclear cells (PBMCs) and platelets. Diet can also cause changes in the protein complement of cells with time, for example, through the incorporation of dietary fatty acids into cell membranes. Using proteomics to identify acute and long-term effects on the protein complement of a biological system has the advantage over transcriptomics because it measures the functional product (protein) of gene expression, and allows the identification of modifications that may relate to the activation or inactivation of proteins through dietary intervention. Such proteins may not only play a major physiological role in a target organ, but could also reflect changes in mechanisms initiated by dietary interventions.

▪ How sensitive are the currently available proteomic techniques in detecting the less abundant proteins? Which proteomic techniques are more widely used for this analysis?

The 2DE technique is probably still one of the most widely utilized approaches in proteomics for the identification of changes in individual proteins of tissues and biofluids, especially those that relate to glucose and fatty acid metabolism as well as pathways relating to oxidative stress, antioxidant defense mechanisms and redox status. Whilst this method is labor-intensive, it actually yields a physical separation of intact polypeptides, providing information on molecular weight and iso-electric points, but also on post-translational modifications. Unfortunately, it is still difficult to visualize and detect differential regulation of low-abundance, very hydrophobic, acidic or basic proteins. For example, proteins involved in inflammatory pathways, such as cytokines, are secreted in the nanogram/milliliter range, and mass spectrometry (MS) has not been sensitive enough to spot relevant changes with such low concentrations. Therefore, 2DE coupled with MS may not represent the most sensitive tool for revealing the effects of nutritional intervention on inflammatory pathways. However, recently, new approaches have been developed which provide a significantly faster and relatively less expensive tool to quantitatively measure low-abundance proteins in blood (present at low nanogram/millileter concentrations) that can be highly ‘multiplexed’. This enables the simultaneous measurement of multiple proteins. Furthermore, the increased use of triple-quadruple-derived technologies combined with linear ion trap in proteomics now enables the quantitative analysis of specific peptides, including those that are post-translationally modified in complex biological mixtures, such as human plasma and serum with high sensitivity and selectivity in the multiple reaction monitoring (MRM) mode [2,3].

▪ How rapidly is the field of plasma biomarker development evolving? Are there any disadvantages to using samples from plasma, serum or circulating cells for biomarker analysis?

Biomarker proteins may not only play a major physiological role in target organs, but could also reflect changes in mechanisms initiated by dietary intervention when circulating in the blood. The most direct approach to take, in biomarker discovery, would be the identification of human plasma protein markers of systemic effects, which could be affected by diet. A problem with plasma proteomics, however, is often the variability in plasma protein levels both within and between subjects, and the variation in abundance between these proteins. Indeed, the plasma represents a systemic ‘pool’ of information originating from different organs, each with distinct physiological functions. Elucidation of changes in platelet and PBMC proteomes upon dietary intervention may, therefore, provide a more sensitive and less variable approach to detecting regulation of, for example, inflammatory and immunological processes. However, despite numerous efforts, it has so far been extremely difficult to identify those proteins that are indeed plasma biomarkers. We are hopeful, though, that improving proteomics technologies will generate novel biomarkers, although it may take longer than people thought [4].

▪ How could the problem of ‘variability’ in the plasma proteome be targeted?

Biomarkers can also be obtained indirectly from the use of animal models, where proteomics of target organ tissues has already provided valuable insights in the effects of several dietary interventions on proteins involved in the regulation of glucose and fatty acid metabolism, oxidative stress, and the redox system [5–7]. Such changes could then be reflected in the regulation of specific (but sometimes low abundance) plasma, platelet or PBMC proteins, and we have found evidence of this on a number of occasions already. More targeted, quantitative and sensitive methods, such as ELISA, or the quantitative analysis of specific peptides that are specific for unique proteins in complex biological mixtures, such as human plasma, in the MRM mode can then be developed and used to evaluate and validate newly discovered candidate biomarkers in human plasma and blood cells in a more accurate and high-throughput fashion.

▪ Has the use of proteomics technology helped to move the nutrition science forward so far?

While proteomics is presented as an emerging and promising tool that enables the elucidation of mechanisms of action, as well as development of relevant biomarkers of health or disease, the actual use of this technique in dietary intervention studies is still rather limited [2]. This may be due to the challenges we are currently facing, occuring as a result of biological and analytical variability, which may mask the subtle effects of dietary intervention. Analytical variability, in contrast to biological variability, however, is something we should be able to control. A number of initiatives are now ongoing to aid the standardization of sample preparation as well as proteomics procedure protocols [8]. Furthermore, most proteomics techniques do not easily detect the differential regulation of low-abundant but often clinically relevant proteins. Although an increase in quality and sensitivity of a new generation of mass spectrometers has hugely accelerated developments in the proteomics field, we are still dealing with MS-related problems, especially those related to misidentification by algorithms used in software programs [9]. Advances in the near future, in MS should make it possible to accurately measure and quantitate proteins from complex biological samples, increasing our understanding of the dynamic changes in the proteome induced by dietary components.

▪ How do you see the field of nutrition proteomics evolving over the next few years?

In the past century, nutrition sciences have mainly focused on the identification of essential nutrients and the understanding of their biological importance in controlling metabolism and in maintaining health. Modern molecular nutritional research is aiming more at health promotion, disease prevention and performance improvement. Some specific dietary patterns or bioactive food ingredients are claimed to either decrease disease risk factors or to improve quality of life by optimizing and maintaining physiological functions. In order to substantiate such claims in a scientific way, new technologies in nutrition research, such as proteomics of human plasma and blood cells, combined with other nutrigenomics technologies such as genomics, transcriptomics, metabolomics and imaging, could provide valuable tools for extensive phenotyping and give access to holistics discovery of efficacy biomarkers. This will help us to fully understand the impact of exposure to nutritional intervention. The nutrition community would clearly benefit from the introduction of a ‘dictionary of validated protein biomarkers’ to assess their efficacy on the mechanistic or functional level.

Financial & competing interests disclosure

The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.