Natural abundance 13 C NMR spectra of biological extracts are recorded in a single scan provided that the samples are hyperpolarized by dissolution dynamic nuclear polarization combined with cross polarization. Heteronuclear 2D correlation spectra of hyperpolarized breast cancer cell extracts can also be obtained in a single scan. Hyperpolarized NMR of extracts opens many perspectives for metabolomics.Lying at the interface of chemistry and biology, metabolomics is one of the youngest sprouts of the sprawling tree of "omic" sciences. The last decade has witnessed a growing interest in mapping metabolic pathways, in identifying new biomarkers, and in modeling metabolic fluxes. 1,2 Nuclear Magnetic Resonance (NMR) spectroscopy is a major analytical technique in this field, offering unambiguous information about the identity of metabolites and their time-dependent concentrations in complex samples such as biofluids, extracts and tissues. However, proton NMR spectra have a limited range of chemical shifts and often suffer from overlapping signals. Carbon-13, which has a larger chemical shift dispersion, can benefit from sensitive detectors 3 or from isotopic enrichment, 4 albeit at the cost of an enhanced complexity of the spectra due to 13 C-13 C couplings. Heteronuclear ( 1 H→ 13 C) correlation spectroscopy can in principle offer a satisfactory dispersion. 5,6 So far, 13 C NMR is not routinely used in metabolomic studies, but hyperpolarization by dissolution dynamic nuclear polarization (D-DNP) combined with cross polarization (CP) could boost its popularity in this field.Hyperpolarization techniques are capable of generating spin polarization levels that are about 50 000 times above Boltzmann equilibrium at room temperature. 7-9 Once prepared and transferred to a solution-state NMR spectrometer, the resulting magnetization can be exploited within a limited time-span that depends on the longitudinal relaxation time T 1 of the hyperpolarized nuclei. Among various methods for hyperpolarization, dynamic nuclear polarization (DNP) is least affected by the chemical substrate and sample preparation. 10 For solution-state NMR, particularly promising results can be obtained by dissolution DNP, where the sample to be analyzed is mixed with free radicals in a glass-forming solution, frozen to liquid helium temperatures, and hyperpolarized by irradiating in the vicinity of the electron spin resonance (ESR) of the unpaired electrons of the radicals. 11 Spins of 13 C nuclei may be polarized either directly or via cross-polarization from protons to 13 C. 12 Rapid melting and shuttling of the sample from the polarizer to a routine NMR spectrometer enables conventional solution-state NMR spectroscopy with a sensitivity that can be boosted by a factor of about 50 000. Most applications of D-DNP have been restricted to the injection of pure hyperpolarized 13 C-labeled molecules such as pyruvate into living organisms. 13 D-DNP has hardly been applied to complex mixtures of metabolites, 14 but only to a few wellchosen small molecules...

These data indicate that metabolites associated with energy and protein metabolism were involved in the response to a high-fat, high-fibre diet. Relevant plasma indicators of metabolic flexibility related to changes in body adiposity were then proposed.

Excessive deposition of body fat is detrimental to production efficiency. The aim of this study was to provide plasma indicators of chickens' ability to store fat. From 3 to 9 wk of age, chickens from 2 experimental lines exhibiting a 2.5-fold difference in abdominal fat content and fed experimental diets with contrasted feed energy sources were compared. The diets contained 80 vs. 20 g of lipids and 379 vs. 514 g of starch per kg of feed, respectively, but had the same ME and total protein contents. Cellulose was used to dilute energy in the high-fat diet. At 9 wk of age, the body composition was analyzed and blood samples were collected. A metabolome-wide approach based on proton nuclear magnetic resonance spectroscopy was associated with conventional measurements of plasma parameters. A metabolomics approach showed that betaine, glutamine, and histidine were the most discriminating metabolites between groups. Betaine, uric acid, triglycerides, and phospholipids were positively correlated (r > 0.3; P < 0.05) and glutamine, histidine, triiodothyronine, homocysteine, and β-hydroxybutyrate were negatively correlated (r < -0.3; P < 0.05) with relative weight of abdominal fat and/or fat situated at the top of external face of the thigh. The combination of plasma free fatty acids, total cholesterol, phospholipid, β-hydroxybutyrate, glutamine, and methionine levels accounted for 74% of the variability of the relative weight of abdominal fat. On the other hand, the combination of plasma triglyceride and homocysteine levels accounted for 37% of the variability of fat situated at the top of external face of the thigh. The variations in plasma levels of betaine, homocysteine, uric acid, glutamine, and histidine suggest the implication of methyl donors in the control of hepatic lipid synthesis and illustrate the interplay between AA, glucose, and lipid metabolisms in growing chickens.