Nutrimetabolomics: A step further into personalized nutrition

Keywords: Metabolomics, nutrimetabolomics, personalized nutrition, Metabolome

Abstract

Metabolomics, which relates to the measurement of small molecules (<1,500 Dalton) within a biological sample, is part of the omics sciences developed in the last 20 years. Traditionally, two different approaches had been developed, the untargeted and targeted investigations, which differ mainly in the identification of the molecules analyzed and the statistical data analysis. The application of metabolomics tools in the nutrition sciences is known as nutrimetabolomics and has shown potential in the discovery of new food intake biomarkers, validation of food frequencies questionnaires and assessment of dietary compliance or dietary patterns. Furthermore, the application of nutrimetabolomics within dietary interventions might help to characterize the molecules responsible for modulating health and identifying their mechanism of action. This paper presents an update of the applications of nutrimetabolomics toward personalized nutrition and discusses the challenges that must be faced to become a reality the use in the clinical entourage.

Introduction

For centuries, food has been considered an essential energy source to develop the multiple activities of the human being, as well as a modulator of health and well-being. Different ancient civilizations such as Egypt, Persia, India, and China already used food to treat and prevent the disease. In fact, ancient Chinese doctors used ants to evaluate if the urine contained high levels of glucose, thus making possible to diagnose diabetes (1).

Historically, nutrition science has had a reductionist approach, focusing on the analysis of single compounds derived from food. However, technological progress has brought the ease of making more specific and useful analyses at the molecular level, opening new horizons in nutrition research. Therefore, a more holistic vision might be useful for understanding the interactions between the variety of chemical compounds present in food and the biochemical networks in complex organisms. Modern nutrition aims to characterize the relationship between diet, lifestyle, and health at the molecular level, identifying the role played by nutrients in the organism. Hence, the application of novel high-throughput technologies such as omics sciences will provide critical answers to these many questions. Here we discuss the role of metabolomics in nutrition toward personalized attention.

Metabolomics

Metabolomics is the study of global changes in the entire set of metabolites (metabolome), all molecules of low and average molecular weight (<1,500 Dalton ), present in cells, tissues, organs and organisms derived from the process known as metabolism (i.e., substrates, and enzymes products) (2). Some approximations indicate that the number of metabolites present in the human being can range between 3,000 and 20,000, whereas the number of genes is estimated at 23,000 and that of proteins in 100,000 (3)

Metabolites are part of the structure of large macromolecules and cell membranes and are classified as endogenous and exogenous (Figure 1). The former includes amino acids, organic acids, nucleic acids, fatty acids, sugars, vitamins, and cofactors, among others and are products of the organism metabolism; the latter comes from the interaction with the external environment and can include drugs, environmental contaminants, toxins and those from the diet. Metabolites from the diet, also known as the food metabolome, are the sum of the metabolites derived from digestion, absorption in the intestine and biotransformation conducted in tissues or the microbiota. Moreover, we should consider that each food has its metabolome, i.e., the metabolome of orange or fatty fish, and therefore the human metabolome is composed of fractions of such metabolomes, partially transformed after digestion, thus forming the human food metabolome (4). Metabolites from the diet may act as both a source of energy or as regulators of the energy metabolism pathways; i.e., as signaling messengers or as antioxidants, protecting the cell from oxidative stress (5).

Because the metabolites reflect the physiological state and provide a better understanding of cellular functioning, they can be a powerful and useful tool to study the metabolism and physiology of living organisms. The characterization of all the metabolites is commonly known as metabolomic profiling or fingerprint. When comparing two or more metabolomic profiles is possible to determine patterns of variation between different groups, i.e., control subjects vs. subjects in a nutritional intervention. Furthermore, we can monitor the result of an intervention, either pharmacological or nutritional, by observing the variation of the fingerprints through the trajectory of time. The elaboration of classifications based on the fingerprint is known as metabotyping.

How to study the metabolome?

One of the major goals of metabolomics is the identification of a large number of metabolites minimizing the losses or alterations in the process. Unlike the proteome, the metabolome is composed of a wide variety of chemical compounds of a complex nature thus it is virtually impossible to determine the entire metabolome simultaneously; consequently, the combination of different methods and configurations for extraction makes its study more feasible. We know that the genome remains static, however, the metabolome is in continuous change and reflects several environmental factors, including drugs, pollutants, the activity of the intestinal microbiota and, notably, the diet. Hence, the metabolomic profile offers a high-level description of biological systems that transcend genetic information and reflect more accurately the human phenotype.

The complexity of the metabolome due to the intra-individual variability, the dynamic nature of the compounds within it and the metabolic flux, make of utmost importance to select an adequate analytical approach. Therefore, an appropriate selection of experimental design with reliable protocols for reproducibility will provide data of biological relevance. Traditionally, study designs on metabolomics have been categorized into two classes: targeted and untargeted.

The untargeted analysis focuses in obtain as much data as possible in a sample; for this, platforms with highly accurate analytical capabilities have been developed. However, the primary bottleneck is the lack of identification of many of the metabolites that will be detected. The chemical identification and structural elucidation of such metabolites require an intensive post-analysis work, necessary to be able to perform a correct biological interpretation.

In a targeted analysis, the chemical identities of the metabolites sought are known before the analysis, due to the use of pure standards and a methodology focused on providing high precision and selectivity. The major advantage of targeted analysis is related to the knowledge of the metabolites investigated that helps to obtain a straightforward biological interpretation. Thus, this approach is advantageous when we want to test a hypothesis.

As of today, it is technologically impossible to determine, quantify and identify each of the metabolites present in a biological sample. Therefore, high-throughput platforms are needed, either separately or in combination for such analysis. Nevertheless, each platform has its advantages and limitations on fundamental aspects such as the specificity and sensibility. The most widely used platforms in metabolomics studies to optimize the results are nuclear magnetic resonance (NMR), and mass spectrometry (MS) coupled to various chromatographic variants (i.e., liquid or gas). Table 1 presents the main advantages of each platform. An analysis of the technical instrumentation is reviewed in depth by Williams et al. (6). Moreover, due to the large amount of data generated, specific statistical approaches are needed and are reviewed elsewhere (7).

Table 1. Strengths and limitations of the most common metabolomics platforms

PLATFORM

STRENGTHS

LIMITATIONS

NMR

High reproducibility and reliability

Minimum sample preparation

Nondestructive

Low sensitivity

GC-MS

High reproducibility and analytical sensitivity

High identification

Only volatile compounds and thermal stable

Destructive

LC-MS

Chromatographic alternatives (i.e. RP-C18, HILIC, etc.)

Wide coverage

Time-consuming bioinformatics processing

Destructive

CE-MS

Minimal sample amount

Polar charged compounds

Limited robustness and low reproducibility

NMR: nuclear magnetic resonance; GC-MS: gas chromatography-mass spectrometry; LC-MS: liquid chromatography-mass spectrometry; CE-MS: capillary electrophoresis-mass spectrometry

Nutrimetabolomics

Nutrimetabolomics is the implementation of metabolomics tools in the nutritional sciences (8) and has been used to identifying metabolic diseases influenced or modulated by the food metabolome (9). Furthermore, the identification of metabolites that serve as food intake biomarkers might help to assess nutritional interventions and validate dietary surveys. Technological advances and the development of more studies in the field of nutrimetabolomics have helped to enhance the understanding of the health-diet relationship. For instance, through the determination of metabolomic changes after a nutritional intervention or after the exposure to a food or a dietary pattern, or with the purpose of verifying the response to a dietary recommendation and distinguishing between responders and non-responders.

Besides, the identification of the endogenous metabolome is paramount to determine the inter- and intra-subject variability and allow the classification of subjects according to their nutritional status or dietary habits. Hence, designing tailored recommendations according to nutritional characteristics or vulnerabilities of the population in which the metabolome is determined. Hereafter, nutrimetabolomics established as a handy tool to bring personalized nutrition into a daily basis and here we discuss some applications.

Biomarkers of consumption to assess the exposure to the diet

Traditionally, the strategies used to assess food consumption have included conventional tools such as food frequency diaries, 24-hour recalls or nutritional diaries, carried out in large cohorts for a few months (10). Nutrimetabolomics might support the data validation derived from extensive studies by providing new biomarkers, and more objective, that could offer additional information outstanding the traditional methods aforementioned (11). Therefore, it can facilitate future research when investigating the associations between diet and disease. Moreover, a proper metabolite characterization provides substantial evidence when applying for nutritional claims and will aid in offering the population better nutritional guidance.

Biomarkers of consumptions should be characterized by their specificity, which means that they are not be influenced by the consumption of similar food or a food pattern. For example, stachydrine, which is the methyl betaine of proline, was reported for the first time as a biomarker of orange intake (12) and, subsequently, validated by other researchers (13). Recently, our research group proposed a new metabolic signature including stachydrine plus betonicine, methyl glucopyranoside (alpha+beta), dihydroferulic acid and galactonate capable of distinguishing between consumption of orange juice with different polyphenol contents (14). Therefore, it seems that stachydrine might be used as a suitable biomarker and its selectivity increases when combined with other metabolites. However, other metabolites have shown their lack of specificity as shown by beta-alanyl-N-methylhistidine and methylhistidine which were associated with oily fish consumption (15), whereas other authors reported them as markers of chicken or meat intake (16).

Specific biomarkers for specific food might allow a better understanding of their impact on metabolic pathways regulation. Similar foods but with different chemical composition may have different effects on the metabolome, and this should be considered, such is the case of the isoflavones derived from soy, which have shown a different impact according to the type of conjugate ingested (17).

Assessing dietary patterns through nutrimetabolomics

Diet diversity, whether due to geographical or cultural reasons, has generated considerable interest in exploring dietary patterns through metabolomics. Understanding the effect of every component of the diet in the metabolome would help to comprehend its potential protective or harmful effects, and furthermore their mechanisms of action. In such regard, the European Prospective Investigation into Cancer (EPIC) cohort studied the profiles of healthy male consumers of meat, fish, vegetarians, and vegans. The metabolic profiling allowed to differentiate between each group, particularly those that belonged to the vegan group in front of the consumers of products of animal origin; mainly, due to the lower concentrations of glycerophospholipids and sphingolipids that were found in the vegan subjects (18). This method opens a door for research in which the hypothesis is whether the differences between metabolic profiles have an impact on health, such as the reduction or the increased risk of chronic diseases like CVD.

As of today, several dietary patterns have been profiled using metabolomics. On the one hand, the Nordic diet has been associated with biomarkers such as trimethylamine N-oxide, hydroquinone-glucuronide, hippuric acid, indole-3-acetic acid and 3,4,5,6-tetrahydro hippurate (19). On the other hand, the Mediterranean diet has been related to 3-hydroxybutyrate, citrate and cis-aconitate; creatine, creatinine; to different amino acids: proline, N-acetyl glutamine, glycine, amino branched-chain acids and their derivatized metabolites; lipids, oleic and suberic acid and some microbial co-metabolites such as phenyl acetyl glutamine and p-cresol (20). Therefore, the assessment of the geographical factor in multi-centric studies should be considered as an independent variable to make an adequate interpretation in subsequent investigations. The LIPGENE study provides a clear example (21), in which the inclusion of subjects with metabolic syndrome from several countries in Europe, revealed variations in urinary and plasma metabolic profiles. These differences made possible to group the individuals according to their geographical region. For example, individuals from the northwest region presented higher concentrations of hippurate and methyl N-nicotinate, whereas the northeast region was characterized by elevated levels of creatinine, citrate, and EPA in plasma. Finally, the southeastern region was defined by a higher concentration of trimethylamine and reduced levels of eicosapentaenoic acid in plasma.

Studying the effect of food and its impact on health

Although nutrimetabolomics is becoming an exciting alternative to characterize the response to foods or dietary, it is essential to bear in mind that there are still technical shortcomings that complicate its clinical use. One major concern is related to the quantitation across studies. Therefore, differences between the same metabolite in different interventions must be carefully interpreted, to avoid wrong conclusions in regards to their mechanisms of action. Furthermore, the enhancement in the identification and characterization of unknown compounds will provide a better understanding of their role in the metabolism and its impact on health.

In recent years, metabolomics has been used to determine the effect of specifics nutrient or foods in RCTs. For example, an RCT of eight weeks of duration including Finnish women compared the use of different types of bread. A regular bread decreased the levels of leucine and isoleucine, whereas the consumption of rye bread led to an increase of betaine and N, N-dimethylglycine concentrations. These metabolic differences point toward a modulation in the metabolism of the branched-chain amino acids and the one-carbon metabolism as the pathways in which rye bread exerts its protective effects against cardiovascular diseases or certain types of cancer (22).

Regardless of dietary patterns, individuals have a basal metabolic trace that might be influenced by various factors, such as metabolism, intestinal microbiota, lifestyle, physical activity, and body composition. Therefore, all these factors should be taken into account when designing nutrimetabolomics studies related to nutritional interventions. The determination of the basal metabolic phenotype, as well as the establishment of the extent to which nutritional interventions can modulate the basal phenotype, should be considered.

Microbiota, diet, and metabolomics

In recent years, interest in the gut microbiota has increased due to the presence of a large number of microorganisms producing different metabolites that have been related to health and disease states in humans (23,24); also, the specificity of some microbiota-related metabolites has shown to be outstanding (4). Evidence has shown the influence of diet on diseases such as colorectal cancer, and this might be related to the microbes’ metabolite production that might serve as modulators. Probably, the most documented data relates to short-chain fatty acids, which are fiber fermentation products, and their protective role impeding the growth arrest and differentiation of colorectal cancer cells (25).

Hence, the role of the intestinal microbiota and its manipulation through food intake could be a potential tool for the prevention and management of diseases (26,27). For example, the combination of phytochemicals and fiber might counterattack the harmful effects of those metabolites with a pro-carcinogen character. Because many phytochemicals, just as polyphenols, are fermented in the colon by specific microbes producing more bioactive compounds, consequently, amplifying their protective properties and acting together with metabolites like the short-chain fatty acids. Therefore, comprehensive profiling of the microbiome metabolome might aid in the conception of the interaction between microbiota, diet, disease, and treatment.

Future perspectives

The field is growing up as reflected by the increasing number of studies ongoing. Nevertheless, several challenges must be acknowledged and solved. The standardization of the methods and protocols is probably one of the major concerns among the community to guarantee the replication across different labs. Fortunately, works such as the report from Wang and colleagues (28) are helping in the identification of potential novel food biomarkers and demonstrating the validity of early reported biomarkers in external cohorts and using different types of biofluids. Also, the use of different types of study designs will allow identifying whether the metabolites or metabolic fingerprints are biomarkers of intake or markers of a response to the food intake.

Finally, yet importantly, it is necessary to pursue the openness of data; there are several initiatives to improve in this regard. Making public the raw metabolomics data will allow testing, validate, replicate or re-analyze previous findings. Currently, there are numerous statistical approaches, and it is time-consuming for the original researchers to apply all of them. Thus the public availability might help to explore in depth and from different perspectives by other researchers. Altogether, filling these gaps will approach the use of metabolomics in the clinical setting and contribute to the development of personalized nutrition.

Acknowledgments

The research leading to this publication has received funding from the European Union Seventh Framework Programme (FP7-PEOPLE-2013-COFUND) under grant agreement n° 609020 - Scientia Fellows

REFERENCES

  1. MacCracken J, Hoel D. From ants to analogues. Puzzles and promises in diabetes management. Postgrad Med. 1997;101:138–50.

  2. Rangel-Huerta ODO, Gil A. Nutrimetabolomics: An Update on Analytical Approaches to Investigate the Role of Plant-Based Foods and Their Bioactive Compounds in Non-Communicable Chronic Diseases. Int J Mol Sci. 2016;17:2072.

  3. 1000 Genomes Project Consortium, Auton A, Brooks LD, Durbin RM, et al. A global reference for human genetic variation. Nature. 2015;526:68–74.

  4. Scalbert A, Brennan L, Manach C, et al. The food metabolome: A window over dietary exposure. Am J Clin Nutr. 2014;99:1286–308.

  5. Bjørklund G, Chirumbolo S. Role of oxidative stress and antioxidants in daily nutrition and human health. Nutrition. 2017. p. 311–21.

  6. Williams MD, Reeves R, Resar LS, Hill HH. Metabolomics of colorectal cancer: past and current analytical platforms. Anal Bioanal Chem. 2013;405:5013–30.

  7. L. Yi, N. Dong, Y. Yun, et al. Chemometric methods in data processing of mass spectrometry-based metabolomics: A review. Anal Chim Acta. 2016;In press:17–34.

  8. Zhang X, Yap Y, Wei D, et al. Novel omics technologies in nutrition research. Biotechnol Adv. 2008;26:169–76.

  9. LeMieux MJ, Aljawadi A, Moustaid-Moussa N. Nutrimetabolomics. Adv Nutr An Int Rev J. 2014;5:792–4.

  10. Hjartåker A, Andersen LF, Lund E. Comparison of diet measures from a food-frequency questionnaire with measures from repeated 24-hour dietary recalls. The Norwegian Women and Cancer Study. Public Health Nutr. 2007;10:1094–103.

  11. Playdon MC, Moore SC, Derkach A, et al. Identifying biomarkers of dietary patterns by using metabolomics. Am J Clin Nutr. 2016;ajcn144501.

  12. Lloyd AJ, Beckmann M, Favé G, et al. Proline betaine and its biotransformation products in fasting urine samples are potential biomarkers of habitual citrus fruit consumption. Br J Nutr. 2011;106:812–24.

  13. Gibbons H, Michielsen CJRR, Rundle M, et al. Demonstration of the utility of biomarkers for dietary intake assessment; proline betaine as an example. Mol Nutr Food Res. 2017;61:1–22.

  14. Rangel-Huerta OD, Aguilera CM, Perez-de-la-Cruz A, et al. A serum metabolomics-driven approach predicts orange juice consumption and its impact on oxidative stress and inflammation in subjects from the BIONAOS study. Mol Nutr Food Res. 2017;61:1600120.

  15. Lloyd AJ, Beckmann M, Haldar S, et al. Data-driven strategy for the discovery of potential urinary biomarkers of habitual dietary exposure. Am J Clin Nutr. 2013;97:377–89.

  16. Cheung W, Keski-Rahkonen P, Assi N, et al. A metabolomic study of biomarkers of meat and fish intake. Am J Clin Nutr. 2017;ajcn146639.

  17. Barnes S. The Biochemistry, Chemistry and Physiology of the Isoflavones in Soybeans and their Food Products. Lymphat Res Biol. 2010;8:89–98.

  18. Schmidt JA, Rinaldi S, Ferrari P, et al. Metabolic profiles of male meat eaters, fish eaters, vegetarians, and vegans from the EPIC-Oxford cohort. Am J Clin Nutr. 2015;1–9.

  19. Khakimov B, Poulsen SK, Savorani F, et al. New Nordic Diet versus Average Danish Diet: A Randomized Controlled Trial Revealed Healthy Long-Term Effects of the New Nordic Diet by GC–MS Blood Plasma Metabolomics. J Proteome Res. 2016;15:1939–54.

  20. Vázquez-Fresno R, Llorach R, Urpi-Sarda M, et al. Metabolomic pattern analysis after mediterranean diet intervention in a nondiabetic population: A 1- and 3-year follow-up in the PREDIMED study. J Proteome Res. 2015;14:531–40.

  21. Walsh MC, McLoughlin GA, Roche HM, et al. Impact of geographical region on urinary metabolomic and plasma fatty acid profiles in subjects with the metabolic syndrome across Europe: the LIPGENE study. Br J Nutr. 2014;111:424–31.

  22. Moazzami A a, Bondia-Pons I, Hanhineva K, et al. Metabolomics reveals the metabolic shifts following an intervention with rye bread in postmenopausal women--a randomized control trial. Nutr J. 2012;11:88.

  23. Boulangé CL, Neves AL, Chilloux J, et al. Impact of the gut microbiota on inflammation, obesity, and metabolic disease. Genome Med. 2016;8:42.

  24. Shoaie S, Ghaffari P, Kovatcheva-Datchary P, et al. Quantifying Diet-Induced Metabolic Changes of the Human Gut Microbiome. Cell Metab. 2015;22:320–31.

  25. O’Keefe SJDD. Diet, microorganisms and their metabolites, and colon cancer. Nat Rev Gastroenterol Hepatol. 2016;13:691–706.

  26. Sheflin AM, Borresen EC, Kirkwood JS, et al. Dietary supplementation with rice bran or navy bean alters gut bacterial metabolism in colorectal cancer survivors. Mol Nutr Food Res. 2017;61:1–11.

  27. Zhang C, Yin A, Li H, Wang R, et al. Dietary Modulation of Gut Microbiota Contributes to Alleviation of Both Genetic and Simple Obesity in Children. EBioMedicine. 2015;2:968–84.

  28. Wang Y, Gapstur SM, Carter BD, et al. Untargeted Metabolomics Identifies Novel Potential Biomarkers of Habitual Food Intake in a Cross-Sectional Study of Postmenopausal Women. J Nutr. 2018;148:932–43.