Philip Atherton, Matthew Brook, Ken Smith, & Daniel Wilkinson
MRC – ARUK Centre for Musculoskeletal Ageing Research, School of Medicine, University of Nottingham, Royal Derby Hospital Centre, Derby, UK

https://doi.org/10.36866/pn.105.25

For more than a century, stable isotope tracers (SIT) have been used in physiological research to investigate the regulation of mammalian biological and physiological functions in exquisite detail (for more detail readers are referred to Wilkinson, 2016; Wilkinson et al., 2016). Indeed, much of what we understand of the control of amino acid, lipid and carbohydrate metabolism and the metabolic regulation of growth, development,  and non-communicable diseases have been gleaned from SIT applications. The popularity of SIT extends from the fact that they can provide dynamic measures of the metabolism of a defined biological system (or combination of systems) in vivo (Wolfe & Chinkes, 2005). While still considered a niche technique, SIT remain as informative (if not even more so) to contemporary physiological researchers as they did to the early pioneers (e.g. Ussing, 1941).

Why SIT and not other approaches?

So why use tracers, when there are so many other more accessible and cost-effective analytical techniques available? The key to using tracers is that they provide true dynamic information relating to metabolism, something that few other techniques do. Measurement of the concentration of a metabolite within the blood or tissue provides only a ‘snapshot’ of the metabolic process; even making a measurement following an intervention or repeating this measure over a defined period of time provides limited information as to what is happening to the metabolite i.e. we have no way of knowing if more of the metabolite is being made or if less is being metabolised further or even if there is a combination of these two processes.  In most circumstances, concentrations of metabolites in biofluids such as blood are very tightly regulated and any perturbation from the norm is quickly rectified in order to re-establish homeostasis (e.g. Blood Glucose). However, with the inclusion of a tracer, not only can we measure the concentration, but we can also obtain the flux or rate of change of the metabolite/metabolic pathway of interest, and how other pathways may interact to regulate metabolism, providing a more accurate and complete picture of how an intervention is affecting metabolism. So what inherent properties allow stable isotopes to provide such diverse applications for investigating metabolism in physiological systems? To define this, we must first consider the chemical properties of a stable isotope.

An idiots guide to mass spectrometry and detection of SIT

A stable isotope is a species of an element (e.g. carbon, nitrogen, hydrogen or oxygen) that differs in mass due to the addition of one additional neutron within the atomic nucleus (see figure 1). Importantly stable isotopes, unlike their radio-active counterparts (i.e. ¹⁴C/³H) are ‘stable’ (hence the name) and therefore unlike radioisotopes, are not subject to decay and release of damaging ionizing radiation. These stable isotopes can be substituted for their lighter counterparts in many biological substrates i.e sugars, amino acids, lipids, in either single or multiple positions to make a variety of SIT. Importantly, the extra neutron/s upon substitution into SIT adds mass allowing this to be safely ‘traced’ (hence the name) within the body. This difference in mass of SIT is measurable, using a variety of analytical techniques from mass spectrometry to magnetic resonance spectroscopy, meaning that these SIT (e.g. ¹³C) are distinguishable from their lighter counterpart (¹²C). Therefore oral or infused provision of a SIT leads to their incorporation into compounds and polymers of interest via chemical or enzymatic synthesis, e.g. carbohydrates, proteins or lipids. The incorporated ‘labeled’ compounds are biochemically identical to their endogenous ‘unlabeled’ counterparts within the biological system of interest, thereby permitting the ‘tracing’ of metabolic flux by monitoring the fate of the stable isotope throughout the various stages of the metabolic pathway.

But once we have got these compounds inside our pathways of interest how do we detect the presence of these SIT? The most common method for doing this is through the use of a mass spectrometer (MS). A mass spectrometer is essentially an expensive weighing machine, with many varieties available but maintaining the same essential principles. They are capable of separating and detecting whole molecules (typically using Gas Chromatography-MS) or atoms upon molecule combustion (using GC-IRMS)_based on their mass and/or charge after undergoing some form of ionization as they pass through either an electrical and/or magnetic field under vacuum. For more information about the types, configurations and history of mass spectrometers available please see Wilkinson (2016), this level of information is beyond the scope of this article. Mass spectrometers are traditionally combined with other separation techniques such as gas/liquid chromatography, to permit the separation of complex mixtures of organic compounds prior to ionization and detection in the mass spectrometer from biological samples such as blood, urine or tissues. For example if we were interested in the rate of skeletal muscle protein synthesis we might take sequential samples of muscle over a defined time period of a few hours whilst the subject is being infused with a SIT of an amino acid (i.e. 1,2,¹³C₂ Leucine), we would then analyse each muscle sample for the presence of this SIT within the muscle protein. To do this, we would isolate the protein pool of interest (either mixed, myofibrillar, mitochondrial, sarcoplasmic or collagen) following a process of homogenization and differentiation centrifugation techniques.

These crude fractions can then be chemically hydrolysed (denaturing and breaking of the peptide bonds) by heating in acid overnight, to release the individual AAs. Following a number of purification and derivatization steps these AAs are ready to be separated and analysed for changes in the level of isotopic enrichment (due to the SIT) by mass spectrometry. The amount of the SIT incorporated over time into the muscle protein as detected by the mass spectrometer (Figure 2) provides information on the rate of muscle protein synthesis. For more information on the intricacies of these measurements and other SIT techniques the reader is guided towards Wolfe and Chinkes (2005).

Substrate specific SIT: what have they told us?

SITs are typically substrate specific. That is – stable isotopically labeled amino acids (AA) provide information on amino acid and protein metabolism, labeled fatty acids will inform on fat metabolism while glucose tracers will reflect carbohydrate (CHO) metabolism and storage. These types of tracers have been instrumental in yielding important information relating to the regulation of, flux through and perturbations to each of these pathways (Kim et al., 2016).

In terms of CHO and lipid metabolism, SIT have been able to describe the relative contributions of hepatic and extrahepatic tissues to the regulation of gluconeogenesis, a key factor in understanding the regulation of hypo and hyperglycemia. Moreover, we have been able to determine the rates of glucose uptake and utilization by these tissues under different stresses such as during exercise, as well as the role of disease, ageing and inactivity in development of insulin resistance. In addition, the use of stable isotope tracers has also been key to determining shifts in substrate utilization and fuel use during exercise e.g. highlighting reductions in lipid oxidation at increasing exercise intensities, and the important role of endurance training in increasing utilization of lipids to spare muscle glycogen.

However one of the areas where SITs have had some of their greatest impact over recent years has been in the study of protein metabolism. For example, SITs have helped to determine the key nutritional role of the EAAs, in particular the BCAAs, in driving muscle protein synthesis, alongside the important contribution made by insulin release (provided by CHO or protein ingestion) in suppressing muscle protein breakdown. Moreover, they have helped delineate the protein synthetic and breakdown responses to exercise, and the key interactions between exercise and nutrition for driving muscle adaptation (i.e. when performed in the absence of nutrition, exercise is in fact catabolic). Finally, these techniques have allowed us to start to unravel the intricacies of the major metabolic blunting associated with ageing, inactivity and disease, in an attempt to correct, delay or reverse these consequences through nutritional or lifestyle interventions.

While there are clear benefits to the use of these substrate specific SIT, there are also some significant limitations. To perform a SIT experiment requires preparation of sterile infusates, intravenous (I.V.) cannulation(s) and collection of tissue (usually muscle) biopsies. Such experimental design requires subjects to be studied in a controlled clinical or lab setting – restricting measures to < 24 h – and requires different SIT for each substrate
to be ‘traced’ (e.g. [²H₂]glucose for glucose metabolism, [U-¹³C]palmitate for lipid metabolism and [1,2, ¹³C₂]leucine or [²H₅]phenylalanine for amino acid and protein metabolism) (Figure 3). The time limited and invasive nature of these types of studies narrows their application and as such, may not accurately reflect longer term metabolism in chronic free living situations common to everyday life. Moreover, the intensive and invasive nature may contraindicate their use within certain more vulnerable populations such as the very frail or adolescent.

Deuterium oxide: The renaissance of a non-substrate specific universally acting SIT

Recent developments involving the use of the novel stable isotope tracer: Deuterium Oxide, (D₂O/²H₂O) or ‘heavy water’, have provided the opportunity to overcome some of these limitations. The use of D₂O features in some of the earliest SIT studies where it was apparent that deuterium is incorporated into many metabolic substrates. Soon after it was shown that by maintaining a constant level of D₂O in the body, the kinetics of various substrates could be obtained by measuring the amount of deuterium incorporated.

While the use of D₂O was sporadic until the end of the last century, subsequent advances in analytical instrumentation pioneered by Stephen Previs and Marc Hellerstein was key in re-establishing the application of D₂O in the measurement of protein, nucleic acid and lipid metabolism. D₂O is primarily administered by oral consumption with body water enrichment being monitored simply through collection of saliva samples. This avoids the need for preparation of sterile IV infusions, IV cannulation, and the need for constant subject monitoring in controlled environments. This greatly reduces the invasiveness of studies. D₂O rapidly equilibrates throughout the body water pool and is incorporated into any metabolic pathway that exchanges hydrogen with body water (ergo, most if not all!). Indeed, deuterium atoms incorporate into many substrates including amino acids, glucose, fatty acids and nucleotides, effectively enabling the body to create its own SIT labeled compounds (Figure 3). This engenders potential for determining simultaneous rates of turnover and flux in multiple substrate pools simultaneously. Crucially, relatively slow turnover of the body water pool permits measurement over longer periods i.e. days-weeks-months. This permits experiments to be extended beyond standard limits for traditional SIT studies of ~24h, thereby providing a vital mechanism for monitoring chronic, cumulative metabolic rates of greater translational and ‘real life’ relevance.

This key feature engenders it well to the study of muscle protein turnover in particular, which unlike many other metabolic pools turns over at a relatively slow rate (~1-1.5%/d), therefore through the provision of D₂O, MPS can be monitored over periods of several hours to several months in order to gain insight into muscle protein regulation on a long-term basis. Indeed, this technique has already shown its unique capabilities highlighting that increased MPS over the first three weeks of RET matched early, plateauing muscle hypertrophy, with these chronic D₂O MPS derived measures correlating well with long term muscle hypertrophy due to RET. This suggests that D₂O can provide a predictive real life representation of exercise adaptation, something which is not observed with traditional acute SIT.

Further to this, the ease of application and lack of invasiveness will have great utility in vulnerable and clinical populations, with measurements already being made in ageing, and renal and cancer patients.

Isotopomics – The next generation

Tissues are complex biological structures made up of thousands of individual components. Assessing changes in the dynamic flux/turnover of these tissues are generally (although not exclusively) made through their primary constituents i.e. mixed muscle or myofibrillar proteins for muscle. However, these are crude fractions and represent the collective actions of many proteins and do not provide knowledge on how single proteins turnover and how they contribute to muscle homeostasis. To investigate the role of individual proteins stable isotopes have recently been combined with ‘OMICS’ methodologies. For instance, using D₂O ingestion in humans, Hellerstein et al., devised a method to isolate peptides using nanoLC-MS/MS to quantify enrichment of deuterium within individual proteins to calculate rates of turnover (Price et al., 2012).

Not unsurprisingly, these methods are rapidly being utilized to provide unique information regarding tissue and biofluid proteome dynamics in both health and disease. Rather intriguing, recently the same research group has reported development of the so called ‘virtual biopsy’. Using D₂O derived dynamic proteomics, plasma proteins, such as the muscle specific creatine kinase M-type (CK-M) and carbonic anhydrase 3 (CA-3) have been shown to accurately represent rates of turnover of the same proteins (or crude fractions) sampled in muscle. Such novel techniques could prove useful in situations where muscle biopsies may be contraindicated, such as in young children or frail elderly, or where they are not easily obtainable such as during exercise or in ICU patients.

Alongside this novel proteomics application, the emerging application of SIT alongside metabolomics now enables ‘fluxomics’ (Figure 3), whereby flux through multiple metabolic pathways can be dynamically monitored. For example, using a [U-¹³C¹⁵N]-valine SIT, and monitoring the enrichment of ¹³C labeling in downstream metabolites of valine and the ¹⁵N labeling of transamination metabolites, it was shown that high aerobic capacity in a outbred rat model selected for high or low intrinsic or inborn aerobic capacity, was associated with increased flux through BCAA catabolic pathways combined with more efficient fatty acid utilization. Where these fluxomics techniques could become more powerful and informative is through the inclusion of D₂O rather than a substrate specific SIT, as this has the potential to provide flux data from a multitude of different pathways simultaneously, providing a more holistic picture of dynamic whole body metabolism and its regulation/dysregulation in health and disease.

Accessibility and costs

The issue of accessibility and costs for stable isotope tracers will always be somewhat of a limiting factor in setting up and running such experiments. Stable isotope tracers by their nature are expensive to purchase, this is due to the specialized way in which they are manufactured, and the limited number of companies that provide them worldwide. Moreover, there are additional costs incurred through the need to prepare sterile infusate, in addition to clinical consumables for placing of IV lines and collection of multiple biological samples, at least for traditional substrate specific tracers. As an example the average cost for an acute substrate specific SIT study to be performed within our laboratories is ~£450/study (inclusive of all consumables and pharmacy costs for IV tracer prep). This cost still doesn’t take into account costs for the analysis of samples once collected, which usually means requiring access to the appropriate analytical platforms such as gas chromatography mass spectrometry, which dependent on lab and the type and number of compounds being analysed can range from £10 to >£100 per sample. Taking all this into account, the cost of performing a stable isotope tracer study can seem rather daunting, and some may consider that the initial cost outlay may outweigh the benefits provided by the outcomes of the study, we would argue this would not be the case in a well designed experiment. However, one should not be discouraged, there are ways of designing your studies to minimise some of these cost implications. D₂O, unlike substrate specific tracers is considerably cheaper (~£200/study) to produce and the cost of this tracer has been steadily decreasing over recent years, as its popularity as a technique has gradually increased.

Moreover, the need for sterile preparations, I.V. lines and other clinical consumables is greatly reduced due to the oral route of administration. In addition, as highlighted earlier this tracer is highly versatile, being applicable to measure changes in metabolism over periods as short as a few hours, acting as an alternative to many substrate specific tracers, whilst also being capable of measuring over more chronic free living periods of days, weeks or even months. Whilst this  greatly saves on experimental cost, analytical costs remain. Isotope ratio mass spectrometry (the traditional workhorse for stable isotope tracers) can cost anything upwards of £150,000 and tends to only be available in specialist labs. However, improvements in design and sensitivity of (GC/LC based) tandem mass spectrometers (a more affordable and widely available piece of analytical equipment) increases accessibility to these analytical techniques whilst also reducing overall costs for analysis. Many university labs have some form of mass spectrometric equipment available to them, through core or central facilities, and there are even commercial providers available at certain prices. Moreover, we find scientific colleagues are always keen to collaborate on important and interesting projects and ideas. If you want to use tracers within your work, the scope and availability of these techniques is ever increasing.

Conclusion

SITs are unique and useful tools for use within the physiological sciences and have an application to a far wider range of disciplines beyond that of purely muscle physiology. If you can sample the tissue and model the metabolic pathways involved in its physiology, then SIT may provide benefit to your experiments. Moreover, its application can extend beyond human work and they are also routinely used within pre-clinical animal models, as well as in vitro cell culture. The availability of these techniques is increasing rapidly, and we highly recommend the follow reviews and books to the readers to provide them with more detailed insight into this fascinating field of research (Wolfe & Chinkes, 2005; Gasier et al., Wilkinson, 2016; Kim et al., 2016; Wilkinson et al., 2016).

References

Gasier HG, Fluckey JD & Previs SF (2010). The application of 2H2O to measure skeletal muscle protein synthesis. Nutr Metab (Lond) 7, 31

Holmes WE, Angel TE, Li KW & Hellerstein MK (2015). Dynamic Proteomics: In Vivo Proteome-Wide Measurement of Protein Kinetics Using Metabolic Labeling. Methods Enzymol 561, 219–276.

Kim I-Y, Suh S-H, Lee I-K & Wolfe RR (2016). Applications of stable, nonradioactive isotope tracers in in vivo human metabolic research. Exp Mol Med 48, e203.

Price JC, Holmes WE, Li KW, Floreani N a, Neese R a, Turner SM & Hellerstein MK (2012). Measurement of human plasma proteome dynamics with (2)H(2)O and liquid chromatography tandem mass spectrometry. Anal Biochem 420, 73–83.

Ussing H (1941). The rate of protein renewal in mice and rats studied by means of heavy hydrogen. Acta Physiol Scand 2, 209–221.

Wilkinson DJ (2016). Historical and contemporary stable isotope tracer approaches to studying mammalian protein metabolism. Mass Spectrom Rev 47, 987–992.

Wilkinson DJ, Brook MS, Smith K & Atherton PJ (2016). Stable isotope tracers and exercise physiology: Past, present and future.J Physiol; DOI: 10.1113/JP272277..

Wolfe RR & Chinkes DL (2005). Isotope Tracers in Metabolic Research: Principles and Practice of Kinetic Analysis, 2nd edn. John Wiley & Sons, Inc., Hoboken, NJ.29