学术报告厅

From Genomics to Metabonomics

In 2003, the human genome was completely mapped and sequenced under the global effort in the Human Genome Project (HGP). However, how a single gene may impact on the phenotypic expression is not fully understood. The new and high throughput technologies developed from the HGP are important technological legacy that have served the basic tools for the new era of ‘omics’ research. ‘Omics’ research includes genomics (DNA), transcriptomics (RNA), proteomics (protein) and recently metabonomics (metabolites), which are integrated under the ‘systems biology’ domain to study collectively (Fig. 1).

Fig 1The “Omics” Cascade

Metabonomics is a relatively new “omics” technology, which is defined as “the measurement of the dynamic multiparametric metabolic response of living systems to pathophysiological stimuli or genetic modifications” (1). Metabonomics should not be confused with metabolomics which is often used interchangeably in the literature. The subtle difference between the two terms is that metabolomics deals with the “comprehensive analysis of all measurable metabolites under a given set of conditions” (2). Thus, the metabolome is “the complete complement of all small molecule (＜1500 Da) metabolites as found in a specific cell, organ or organism” (3). We may view the metabolome as “the comprehensive dynamic ‘fingerprint’ of all metabolites found in a living system”, which is similar to the human genome is a “fingerprint” of all genes in the human body.

Phenotypic expression is not simply predicted by the genetic and protein makeup because our body is not merely made up of genes and proteins. Hence, inheriting a cancer disease gene does not necessary develop the cancer phenotype. Similarly, having an abnormal protein expression does not always lead to morbidity. Genes and proteins can be seen as pieces of jigsaw forming an integral part of a puzzle (i.e. phenotype). In fact, biochemical pathways and metabolites interact with genes and proteins in the “omic” cascade to modulate phenotypic expression. For instance, an abnormally high fasting blood glucose level could be an indicator of disordered carbohydrate metabolism in diabetic patients. Therefore, by taking genetics, proteomics and metabolites into account, a complete picture of phenotypic expression can be revealed.

Where are we now?

The work of Human Metabolome Project (HMP) was commenced in 2004 aiming to identify, quantify and categorize all detectable metabolites (＞1μmol) within the human tissues and biofluids. The metabolome contains more than 2180 metabolite entries to date (3). In addition, the database also contains an extensive collection of metabolite concentrations from blood, urine and cerebrospinal fluids generated by Mass Spectrometry (MS) and Nuclear Magnetic Resonance (NMR) techniques. This database is an important tool for metabolite profiling. Metabolomics Society (www.metabolomicssociety.org), an international society to promote research and development in metabolomics was established in 2005 (4). One of its initiatives was to implement a set of reporting standards for metabonomics research to standardize the reporting format in metabolomics research (5). The Society also launched its official journal “Metabolomics” in January 2005.

Metabonomics in Health Research

Metabonomics has been successfully applied in toxicology, pharmaceutical drug discovery and identification of biomarkers in disease development and diagnosis. Metabonomics studies in toxicology have revealed metabolic effects of rat urine profiles after drug administration such as paracetamol and bromobenzene (6), valproic acid (7), cyclosporin A (8) as well as heavy metals toxicity (9). In disease markers discovery, potential disease markers of chronic hepatitis B have been identified in serum of infected patients using metabonomics (10). Furthermore, metabonomics have been used in the detection and diagnosis of inborn errors of metabolism in neonates, namely, phenylketonuria and maple syrup urine disease (11), and also in distinguishing between breast cancer patients from healthy individuals (12). Gender differences in metabolite profiles in humans (13) and rats (14) have also been evaluated using metabonomics. Differentiation of different rat strains by analyzing urine metabolite profiles was also reported (15).

Metabonomics in Nutrition Research

Until recently, the metabonomics approach has been introduced to nutritional research, known as nutritional metabonomics. The relationship between nutrition and human health is a complex interaction between genes, proteins, metabolites and the environment. The metabolic phenotype of an individual is the interaction between genotypic and environmental (diet, lifestyle, co-existing organisms) factors (16). The nutritional environment can have an impact on gene expression and that the genotype of an individual can also influence nutrient metabolism. Genetic polymorphism in genes involving in lipid metabolism has been shown to influence nutrient metabolism (17). Nutritional metabonomics thus aims to provide an insight into how dietary nutrients can affect health and disease. Although metabonomics research in toxicology and pharmacology has been well developed, its application in nutritional research is still in its infant stage of development.

Investigation on the effects of nutrient intake on plasma and urine profiles is actively under research in nutritional metabonomics. Known doses of bioactive food supplements or nutrients have been administered to subjects to evaluate biochemical responses to these substances in intervention studies (18). Consumption of green and black tea in human has been shown to modify the concentration of urinary metabolites and may result in changing oxidative energy metabolism and biosynthetic pathways (19). Metabolic effects of a vegetarian, low meat and high meat diet have been studied in healthy male volunteers. Urinary metabolite changes were detected and distinct separation of urinary profiles were found in the three diets with the application of O-PLS-DA (20). Increases in urinary hippurate and glycine, and a decrease in urinary creatinine concentration after chamomile tea consumption has been observed in a 1H NMR based metabonomics study (21-22).

The use of metabonomics has also been extended to the study of traditional Chinese medicine. The effects of Gingko were investigated using 1H NMR analysis of urine in rats and human volunteers (21-22). The metabolic effects of soy consumption were also studied (23). A 1H NMR metabonomics approach was used in studying the biochemical effects of soy isoflavones on plasma profiles of healthy pre-menopausal women, after consumption of 60g/day of soy protein (45 mg of isoflavones) consistent decreases in plasma sugars and lactate in all subjects as well as subject specific changes in glucogenic amino acids (isoleucine and valine), triacyglycerols and choline were noted. The authors attributed theses changes in carbohydrate and energy metabolism to soy intervention. Furthermore, changes in lipoproteins and increases in metabolites related to lipid metabolism and synthesis, such as choline, acetate, acetoacetate, alanine and glycoproteins following soy consumption were also reported. The same group (24) using a similar approach to investigate the in vivo biochemical effects of conjugated and unconjugated dietary isoflavones in pre-menopausal women reported an increase in urinary trimethylamine-N-oxide (TMAO) excretion after conjugated and unconjugated soy isoflavones consumption. In addition, changes in concentrations of methylamine, dimethylamine, trimethylamines, trimethylamine-N-oxide, choline, glucogenic amino acids (glutamate and glutamine), creatine, hippurate, benzoate and increases in creatinine levels were also noted. The authors concluded from the plasma and urine metabonomic studies that isoflavones had an inhibitory effect on glycolysis resulting in a general shift from carbohydrate metabolism to lipid metabolism. Therefore, metabonomics is a useful approach to gain a new insight in in vivo biochemical mechanisms in human nutrition.

Limitations in nutritional metabonomics

Despite the advances in analytical technologies and statistical programming, nutritional metabonomics faces a great challenge in standardizing both intrinsic and extrinsic factors. Intrinsic factors such as age, gender, biological rhythms (circadian rhythms, estrus cycle) and extrinsic factors such as diet, physical activity, stress and drugs can all modulate the human nutritional metabolome (25). These confounding factors can trigger metabolic changes which will complicate or mask a true change in metabolic effects. A recent study has shown that recent dietary intake and timing of sample collection could confound the findings in nutritional metabonomics study (26). Therefore, well controlled dietary intervention studies must be carried out to minimize inter- and intra-subject variation (18). Furthermore, non-nutrient molecules can be present in foodstuffs which may be metabolized together with nutrient compounds and thus could also complicate the investigation (16). In addition, the metabolome is also affected by metabolites produced from the gut microflora, and these metabolites can contribute significantly to the metabolite profile.

Another challenge in nutritional metabonomics is the daunting task of sorting the avalanche of datasets produced from the metabolite profiling experiments. NMR spectroscopy has been used for much of the work in metabonomics (18). MS is the other technology which has gained much popularity in metabolite profiling, especially when it is coupled with chromatographic separation techniques such as liquid chromatography (LC), gas chromatography (GC) or capillary electrophoresis (CE) (27). Metabolite profiling using these high throughput and sensitive technologies could result in vast quantities of datasets produced and if the data is not properly managed it will become nonsensical and complicated. In metabonomics studies, it is not sufficient just to have the state-of-the-art technologies for biofluid analysis. The traditional way of visual comparison of datasets, or the analysis of single variable using One-Way ANOVA for example, is impossible in metabonomics. Hence, analysis of metabonomic data must be coupled with advanced bioinformatics and statistical tools to effectively manage these data and transformed the data into meaningful and comprehensible information. Different multivariate data analysis (MVDA) tools are available for clustering and classifying datasets but the most commonly used in metabonomics are the principal components analysis (PCA) and partial least square discriminant analysis (PLS-DA). MVDA tools are pattern recognition tools which facilitate the clustering and hence identification of similarities or differences between groups of datasets. This allows the investigator to determine the differences, if any, between normal and treated individuals.

To conclude, metabonomics is increasingly a potential tool in nutritional research. Each individual has his own metabolomic profile. The ultimate end-point of nutritional metabonomics is to identify how nutrients have an impact on human health and metabolism. However, hurdles in nutritional metabonomics research have to be overcome in the near future before personalized nutrition may be prescribed to individuals for maintenance of health and prevention of disease.

References:

1.Nicholson J.K, Lindon J.C and Holmes E (1999) Metabonomics:understanding the metabolic responses of living systems to pathophysiological stimuli via multivariate statistical analysis of biological NMR spectroscopic data. Xenobiotica 29(11), pp1181-1189.

2.Goodacre R. (2007) Metabolomics of a superorganism. The Journal of Nutrition 137, pp259S-266S.

3.Wishart D.S, Tzur D., Knox. C. et al (2007) HMDB: the Human Metabolome Database. Nucleic Acids Research 35, pp521-526.

4.Goodacre R. (2005) Metabolomics the way forward. Metabolomics 1(1), pp1-2.

5.Fiehn O, Kristal B., Ommen B.V et al (2006) Establishing reporting standards for metabolomic and metabolomic studies: a call for participation. Journal of Integrative Biology 10(2), pp158-162.

6.Schoonen W.G.E.J., Kloks C.P.A.M, Ploemen J.P.H.T.M et al (2007) Sensitivity of 1H NMR analysis of rat urine in relation to toxico-metabonomics. Part I: dose dependent toxic effects of bromobenzene and paracetamol. Toxicology Science Epublication ahead of print.

7.Schnackenberg L.K., Jones R.C., Thyparambil S. et al (2006) An integrated study of acute effects of valproic acid in the liver using metabonomics, proteomics and transcriptomics platforms. Journal of Integrative Biology 10(1), pp1-14.

8.Lenz E.M, Bright J., Knight R., Wilson I.D and Major H. (2004) Cyclosporin A-induced changes in endogenous metabolites in rat urine: a metabonomics investigation using high field 1H NMR spectroscopy, HPLC-TOF/MS and chemometrics. Journal of Pharmaceutical and Biomedical Analysis 35, pp599-608.

9.Lafaye A, Junot C, Gall B.R, Fritsch P, Tabet J.C and Ezan E (2003) Metabolite profiling in rat urine by liquid chromatography/electrospray ion trap mass spectrometry.Application to the study of heavy metal toxicity. Rapid Communications in Mass Spectroscopy 17,pp2541-2549.

10.Yang J., Zhao X.J., Liu X.L et al (2006) High Performance Liquid Chromatography-Mass Spectrometry: potential biomarkers for acute deterioration of liver function in chronic hepatitis B. Journal of Proteome Research 5, pp554-561.

11.Constantinou M.A, Papakonstantinou E., Spraul M. et al (2005) 1H NMR-based metabonomics for the diagnosis of inborn errors of metabolism in urine. Analytical Chimica Acta 542, pp169-177.

12.Fan X.H., Bai J.Q, Shen P. (2005) Diagnosis of breast cancer using HPLC metabonomics fingerprints coupled with computational methods. Engineering in Medicine and Biology Society pp6081-6084.

13.Lu G, Wang J.S, Zhao X.J, Kong H.W, Xu G.W (2006) Study on gender differences based on metabolites in urine by ultra high performance liquid chromatography/time of flight mass spectroscopy. Chinese Journal of Chromatography 24 (2), pp109-113.

14.Williams R.E, Lenz E.M, Evans J.A et al (2005) A combines 1H NMR and HPLC-MS-based metabonomics study of urine from obese (fa/fa) Zucker and normal Wistar-derived rats. Journal of Pharmaceutical and Biomedical Analysis 38, pp465-471.

15.Plumb R.S, Granger J.H, Stumpf C.L et al (2005) A rapid screening approach to metabonomics using UPLC and oa-TOF mass spectrometry: application to age, gender and diurnal variation in normal/Zucker obese rats and black, white and nude mice. The Analyst 130, pp844-849.

16.German J.B., Watkins S.M. and Fay L.B. (2005) Metabolomics in practices: emerging knowledge to guide future dietic advice toward individualized health. Journal of American Dietetic Association 105, pp1425-1432.

17.Ye S.Q. and Kwiterovich P.O (2000) Influence of genetic polymorphisms on responsiveness to dietary fat and cholesterol. American Journal of Nutrition 72, pp1275S-1284S.

18.Rezzi S., Ramadan Z., Fay L.B. and Kochhar S. (2007) Nutritional metabonomics: applications and perspectives. Journal of proteome research  6, pp513-525.

19.Dorsten F.A.V., Daykin C.A., Mulder T.P.J., Duynhoven J.P.M.V. (2006) Metabonomics approach to determine metabolic differences between green tea and black tea consumption. Journal of Agricultural and Food Chemistry 54, 6929-6938.

20.Stella C., Beckwith-Hall B., Cloarec O et al (2006) Susceptability of human phenotypes to dietary modulation. Journal of Proteome Research 5, 2780-2788.

21.Wang M., Lamers R.A.N., Korthout H.A.A.J. et al (2005) Metabolomics in the context of systems biology: bridging traditional Chinese medicine and molecular pharmacology. Phytotherapy research 19, pp173-182.

22.Wang Y., Tang H., Nicholson J.K., Hylands P.J., Sampson J.,Holmes E (2005) A metabonomic strategy for the detection of metabolic effects of Chamomile (Matricaria recutita L.) ingestion. Journal of Agricultural and Food Chemistry 53,191-196.

23.Solanky K.S, Bailey N.J.C, Beckwith-Hall B.M, et al (2003) Application of biofluid 1H nuclear magnetic resonance-based metabonomics techniques for the analysis of the biochemical effects of dietary isoflavones on human plasma profile. Analytical Biochemistry 323, pp197-204.

24.Solanky K.S, Bailey N.J.C, Beckwith-Hall B.M, et al (2005) Biofluid 1H NMR-based metabonomics techniques in nutrition research-metabolic effects of dietary isoflavones in humans. Journal of Nutritional Biochemistry 16, pp236-244.

25.Gibney M.J., Walsh M., Brennan L., Roche H.M., German B. and Ommen B.V. (2005) Metabolomics in human nutrition: opportunities and challenges. American Journal of Nutrition 82, pp497-503.

26.Walsh M.C, Brennan L., Malthouse J.P.G, Roche H.M and Gibney M.J (2006) Effect of acute dietary standardization on the urinary, plasma and salivary metabolomic profiles of healthy humans. American Journal of Clinical Nutrition 84, pp531-539.