The Early History of Metabonomics
Introduction
Over the past decade, there has been a revolution in the techniques and approaches used in molecular biology, following on from the decoding of the human and other genomes. The emphasis in biomedical studies rapidly switched to simultaneous determination of gene expression changes mainly carried out using micro-array technology, and this type of study has been given the name of transcriptomics. An equivalent impetus to map out all cellular or tissue protein expression has evolved and has been termed proteomics. Nowadays, there are nearly two hundred different named “omics”, but many of these terms will not survive because of their very specialist application, and indeed many of these terms are not necessary since they serve only to describe a methodology that already has a perfectly valid name.
Early Metabolic Profiling Work
However, the seeds for metabonomics were born long before any of the omics were thought about. During the early and mid 1980s, the simultaneous detection of the plethora of metabolites seen in biological fluids had been carried out largely using 1H NMR spectroscopy, as pioneered by Jeremy Nicholson then at Birkbeck College, University of London (Nichoson et al. 1983, 1984; Bales et al. 1984). A little later, in early 1988, John Lindon (then at the Wellcome Research Laboratories), after hearing a talk by Jeremy at the 1987 Christmas NMRDG meeting, and realising how the pattern recognition methods being used by his group at Wellcome could potentially be very useful, initiated a collaboration with Jeremy to investigate the use of such multivariate statistics to classify samples according to their biological status. When the approach was found to be successful through the initial studies of Kevan Gartland and Maria Anthony, the concept of metabonomics was born (Gartland et al. 1990a, 1990b, 1991), even though the name was not coined then.
One of the early PhD students at Birkbeck College carrying out metabolic profiling using NMR spectroscopy was Elaine Holmes (working on renal toxicity in collaboration with Sterling-Winthrop, later Sanofi-Aventis) and she later took various postdoctoral positions working in metabonomics at both Birkbeck and Wellcome when the concept of metabolic trajectories was developed in 1992 (Holmes et al. 1992). The computing technology used was primitive by today’s standards, relying on a mainframe type DEC-VAX-750 computer to which batch jobs had to be submitted and a now long-defunct software package called ARTHUR; graph plotting was also rather tedious. Much of the user interface software was written in R by Liz Rahr at Wellcome.
Metabonomics is Coined
The term metabonomics was actually derived in 1996 during a brainstorming session in the Chemistry Department seminar room with Jeremy, John and Jeremy Everett, then of Pfizer, present, but the term was not actually used in a publication until 1999 (Nicholson et al. 1999). Thus, metabonomics encompasses the comprehensive and simultaneous systematic profiling of multiple metabolite levels and their systematic and temporal changes caused by factors such as diet, lifestyle, environment, genetic effects, and pharmaceutical effects both beneficial and adverse, in whole organisms. A parallel approach mainly from plant science and from the study of in vitro and microbiological systems and using largely chromatographic-mass spectrometry techniques led to the term metabolomics also being coined and defined and the methods and approaches used in the two disciplines are now highly convergent.
Metabonomics results are mainly achieved by the study of biofluids and tissues with the data being interpreted using chemometrics techniques. The first papers on NMR of biofluids coupled with pattern recognition came out in 1990 with the first published studies concentrating on xenobiotic toxicity in animal models (Gartland et al. 1990a, 1990b, 1991). However even in the early 1990s, pattern recognition studies of human biofluids in health and disease were already being produced by the collaboration between Jeremy’s group at Birkbeck and John’s group at Wellcome, such as NMR of CSF in Alzheimers disease patients (Ghauri et al. 1993) and of urine in renal transplant patients (Foxall et al. 1993a), and new techniques were also being developed. One of the first was the idea of integrating spectral regions ("binning") to overcome the problem of pH-induced chemical shift variation published in 1994 (Holmes et al. 1994). The use of 2-dimensional J-resolved spectra was also exemplified in 1993 (Foxall et al. 1993b) and the measurement of diffusion coefficients for studies of molecular dynamics and compartmentation in 1996 (Liu et al. 1996). The period during the early and mid-1990s was highly productive and many metabonomics studies were published through the collaboration based on NMR spectra of urine, CSF, bile, seminal fluid, cyst fluid, plasma and serum from humans and from animal models. The photograph below shows John Lindon and Jeremy Nicholson with their PhD student Maria Anthony in front of the Bruker 600 MHz NMR spectrometer at Wellcome in 1992.
L-R: John Lindon, Maria Anthony, Jeremy Nicholson - Wellcome - 1992
Developing Concepts and Technology for Metabonomics
With the advent of 750 MHz NMR spectrometers, a definitive and highly cited paper on metabolite assignments in human plasma, written by Jeremy and John was published in 1995 (Nicholson et al. 1995), based on data acquired at Bruker in Germany with the considerable help of Manfred Spraul. At around the same time, it was appreciated that high resolution NMR spectra of intact tissues could be obtained by the magic-angle-spinning technique (Moka et al. 1997) and that these data would be highly complementary to those from biofluids and this would lead to the idea of integrated metabonomics piecing together information from different biological compartments. At the same time, Jeremy and John (together with Ian Wilson at AstraZeneca and Manfred Spraul at Bruker) also worked extensively on developing directly-coupled HPLC-NMR and later HPLC-NMR-MS, applying the approach to problems in drug metabolism, including reactive acyl glucuronide metabolites, leading to studies in quantitative structure-metabolism relationships, all later reviewed (Lindon et al. 2000).
A Move to Imperial College London
Metabonomics came to Imperial College in 1998 when Jeremy was invited to form a new Section under John Caldwell in the Division of Biomedical Sciences, part of the Faculty of Medicine, together with the Section of Molecular Toxicology that had migrated from St Mary’s Paddington. At this time as well as Jeremy, Elaine Holmes and John Lindon also moved over from Birkbeck (John having joined Birkbeck at the end of 1995 when the Wellcome Laboratories were closed down following the Glaxo take-over of that organisation). At Imperial there were no NMR facilities available to the group and it was an urgent requirement to purchase the first of the Bruker 600 MHz systems. The space occupied by the group, later the section, and then the Department of Biomolecular Medicine, gradually expanded on the 6th floor of the SAF building to house the increasingly comprehensive technology. Later, together with Profs Paul Freemont and Steve Matthews, an 800 MHz NMR system was purchased with SRIF funding for joint use and this is housed in the Chemistry RCS1 building along with the Waters laboratory for mass spectrometry. The new suite was formally opened by the then Rector, Sir Richard Sykes, in September 2006.
Recent Developments
Other more recent developments include the concept of correlation spectroscopy that resulted in a technique called STOCSY in a paper in 2005 (Cloarec et al. 2005); although Jeremy and John actually drafted a patent application for this in 2000, but it was not exemplified until Olivier Cloarec wrote the STOCSY algorithm and got the method working – this has now become one of the mainstream techniques in the group for assignment of NMR peaks to specific metabolites. This was further later developed by Derek Crockford to correlate NMR and MS data in a technique called SHY (Crockford et al. 2006).
One major development has been the concept of pharmacometabonomics, the prediction of drug effects prior to administration of the xenobiotic as this opens up the possibility of metabonomics-driven personalised medicine. This was first exemplified by Andy Clayton with funding from Pfizer in animal models of toxicity (Clayton et al. 2006) and then in a human study of drug metabolism (Clayton et al. 2009). The photograph below shows some of the people involved in the development of pharmacometabonomics.
L-R: Jeremy Nicholson, Andy Clayton, John Lindon, Olivier Cloarec - Imperial 2007.
Photo: Jake Pearce.
Many other major projects and consortia have had metabonomics as a critical element, including both COMET and COMET-2 sponsored by pharmaceutical companies, the Wellcome Trust funded BAIR project, and a series of EU-funded projects including FGENTCARD. The industrially-funded postgraduate training scheme called METAGRAD also funded a number of PhD students on metabonomics based projects.
Current Status
Following the reorganisation of the Faculty of Medicine in 2009, the new Department of Surgery and Cancer was formed with Jeremy Nicholson as Head of Department and Biomolecular Medicine became a major Section led by Elaine Holmes in the Division of Surgery.
by Prof John Lindon. August 2010.
Update: The Section of Biomolecular Medicine changed name to The Section of Computational and Systems Medicine in February 2013 to better reflect the breadth of research conducted.
References
Bales JR, Higham DP, Howe I, Nicholson JK, Sadler PJ. 1984. Use of high-resolution proton nuclear magnetic resonance spectroscopy for rapid multi-component analysis of urine. Clin Chem 30(3):426-32.
Clayton TA, Baker D, Lindon JC, Everett JR, Nicholson JK. 2009. Pharmacometabonomic identification of a significant host-microbiome metabolic interaction affecting human drug metabolism. Proc Natl Acad Sci U S A 106(34):14728-33.
Clayton TA, Lindon JC, Cloarec O, Antti H, Charuel C, Hanton G, Provost JP, Le Net JL, Baker D, Walley RJ and others. 2006. Pharmaco-metabonomic phenotyping and personalized drug treatment. Nature 440(7087):1073-7.
Cloarec O, Dumas ME, Craig A, Barton RH, Trygg J, Hudson J, Blancher C, Gauguier D, Lindon JC, Holmes E and others. 2005. Statistical total correlation spectroscopy: an exploratory approach for latent biomarker identification from metabolic 1H NMR data sets. Anal Chem 77(5):1282-9.
Crockford DJ, Holmes E, Lindon JC, Plumb RS, Zirah S, Bruce SJ, Rainville P, Stumpf CL, Nicholson JK. 2006. Statistical heterospectroscopy, an approach to the integrated analysis of NMR and UPLC-MS data sets: application in metabonomic toxicology studies. Anal Chem 78(2):363-71.
Foxall PJ, Mellotte GJ, Bending MR, Lindon JC, Nicholson JK. 1993a. NMR spectroscopy as a novel approach to the monitoring of renal transplant function. Kidney Int 43(1):234-45.
Foxall PJ, Parkinson JA, Sadler IH, Lindon JC, Nicholson JK. 1993b. Analysis of biological fluids using 600 MHz proton NMR spectroscopy: application of homonuclear two-dimensional J-resolved spectroscopy to urine and blood plasma for spectral simplification and assignment. J Pharm Biomed Anal 11(1):21-31.
Gartland KP, Beddell CR, Lindon JC, Nicholson JK. 1990a. A pattern recognition approach to the comparison of PMR and clinical chemical data for classification of nephrotoxicity. J Pharm Biomed Anal 8(8-12):963-8.
Gartland KP, Beddell CR, Lindon JC, Nicholson JK. 1991. Application of pattern recognition methods to the analysis and classification of toxicological data derived from proton nuclear magnetic resonance spectroscopy of urine. Mol Pharmacol 39(5):629-42.
Gartland KP, Sanins SM, Nicholson JK, Sweatman BC, Beddell CR, Lindon JC. 1990b. Pattern recognition analysis of high resolution 1H NMR spectra of urine. A nonlinear mapping approach to the classification of toxicological data. NMR Biomed 3(4):166-72.
Ghauri FY, Nicholson JK, Sweatman BC, Wood J, Beddell CR, Lindon JC, Cairns NJ. 1993. NMR spectroscopy of human post mortem cerebrospinal fluid: distinction of Alzheimer's disease from control using pattern recognition and statistics. NMR Biomed 6(2):163-7.
Holmes E, Bonner FW, Sweatman BC, Lindon JC, Beddell CR, Rahr E, Nicholson JK. 1992. Nuclear magnetic resonance spectroscopy and pattern recognition analysis of the biochemical processes associated with the progression of and recovery from nephrotoxic lesions in the rat induced by mercury(II) chloride and 2-bromoethanamine. Mol Pharmacol 42(5):922-30.
Holmes E, Foxall PJ, Nicholson JK, Neild GH, Brown SM, Beddell CR, Sweatman BC, Rahr E, Lindon JC, Spraul M and others. 1994. Automatic data reduction and pattern recognition methods for analysis of 1H nuclear magnetic resonance spectra of human urine from normal and pathological states. Anal Biochem 220(2):284-96.
Lindon JC, Nicholson JK, Wilson ID. 2000. Directly coupled HPLC-NMR and HPLC-NMR-MS in pharmaceutical research and development. J Chromatogr B Biomed Sci Appl 748(1):233-58.
Moka D, Vorreuther R, Schicha H, Spraul M, Humpfer E, Lipinski M, Foxall PJD, Nicholson JK, Lindon JC. 1997. Magic angle spinning 1H nuclear magnetic resonance spectroscopy of intact kidney tissue samples. Anal. Commun. 34, 107-109.
Nicholson JK, Buckingham MJ, Sadler PJ. 1983. High resolution 1H n.m.r. studies of vertebrate blood and plasma. Biochem J 211(3):605-15.
Nicholson JK, Lindon JC, 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):1181-9.
Nicholson JK, O'Flynn MP, Sadler PJ, Macleod AF, Juul SM, Sonksen PH. 1984. Proton-nuclear-magnetic-resonance studies of serum, plasma and urine from fasting normal and diabetic subjects. Biochem J 217(2):365-75.
