NMR-Based Identification of Metabolites in Polar and Non-Polar Extracts of Avian Liver

H OH OH metabolites Communication NMR-Based Identification of Metabolites in Polar and Non-Polar Extracts of Avian Liver Fariba Fathi 1 , Antonio Brun 2 ID , Katherine H. Rott 2 , Paulo Falco Cobra 1 , Marco Tonelli 1 , Hamid R. Eghbalnia 1 , Enrique Caviedes-Vidal 3 , William H. Karasov 2 and John L. Markley 1, * 1 2 3 * Biochemistry Department, University of Wisconsin-Madison, 433 Babcock Drive, Madison, WI 53706, USA; (F.F.); (P.F.C.); (M.T.); (H.R.E.) Department of Forest and Wildlife Ecology, University of Wisconsin-Madison, 1630 Linden Drive, Madison, WI 53706, USA; (A.B.); (K.H.R.); (W.H.K.) Instituto Multidisciplinario de Investigaciones Biológicas de San Luis, Consejo Nacional de Investigaciones Científicas y Técnicas, and Departamento de Bioquímica y Ciencias Biológicas, Universidad Nacional de San Luis, San Luis 5700, Argentina; Correspondence: ; Tel.: +1-608-263-9349 Received: 3 October 2017; Accepted: 8 November 2017; Published: 16 November 2017 Abstract: Metabolites present in liver provide important clues regarding the physiological state of an organism. The aim of this work was to evaluate a protocol for high-throughput NMR-based analysis of polar and non-polar metabolites from a small quantity of liver tissue. We extracted the tissue with a methanol/chloroform/water mixture and isolated the polar metabolites from the methanol/water layer and the non-polar metabolites from the chloroform layer. Following drying, we re-solubilized the fractions for analysis with a 600 MHz NMR spectrometer equipped with a 1.7 mm cryogenic probe. In order to evaluate the feasibility of this protocol for metabolomics studies, we analyzed the metabolic profile of livers from house sparrow (Passer domesticus) nestlings raised on two different diets: livers from 10 nestlings raised on a high protein diet (HP) for 4 d and livers from 12 nestlings raised on the HP diet for 3 d and then switched to a high carbohydrate diet (HC) for 1 d. The protocol enabled the detection of 52 polar and nine non-polar metabolites in 1 H NMR spectra of the extracts. We analyzed the lipophilic metabolites by one-way ANOVA to assess statistically significant concentration differences between the two groups. The results of our studies demonstrate that the protocol described here can be exploited for high-throughput screening of small quantities of liver tissue (approx. 100 mg wet mass) obtainable from small animals. Keywords: NMR spectroscopy; liver tissue; extraction protocol; metabolite identification by NMR; diet, effect of on liver metabolites 1. Introduction The liver, the most metabolically active organ, has multiple vital functions related to digestion, metabolism, and immunity. It processes the nutrients absorbed by the small intestine and produces bile that is released into the intestine for digesting fat. Therefore, liver has been called the body’s “chemical factory”. Owing to its important role in the digestive system, researchers have developed several methods to assay different liver fractions, including NMR-based metabolomics [1,2]. 1 H NMR spectroscopy has been utilized since the 1970s to determine metabolic profiles in biological fluids and extracts [3]. NMR has several advantages over mass spectrometry for quantifying metabolite levels, including limited handling and preprocessing of biological samples, high reproducibility, non-destructive analysis, and high-throughput fingerprinting [4,5]. Metabolites 2017, 7, 61; doi:10.3390/metabo7040061 www.mdpi.com/journal/metabolites Metabolites 2017, 7, 61 2 of 9 A crucial step for metabolomic studies of fluids is sample preparation. Several solvent systems for liver extraction have been investigated [6,7], and, from the various possibilities, we chose the methanol/chloroform/water solvent combination [8] for the current study. This solvent system yields NMR spectra of the polar fraction with improved baseline and linewidth, thanks to the exclusion of lipids and proteins from the aqueous phase. The chloroform phase contains non-polar metabolites. The methanol/chloroform/water extraction method described here offers a rapid approach to simultaneous characterization of both the hydrophilic and hydrophobic metabolome. Because defects in lipid metabolism are known to be associated with diseases [9], finding a method to assay non-polar metabolites in liver can be very useful for functional and pathological studies of this tissue. The Karasov laboratory is investigating the effects of diet on the metabolism of small (approx. 7–15 g) nestling birds. Our primary goal was to develop an NMR-based approach for identifying and quantifying the polar and non-polar metabolites present in liver tissue from these birds. Our secondary goal was to develop a database of metabolites from liver tissue that can be detected by NMR spectroscopy for use in future studies. To maximize the sensitivity of these sample-limited experiments, we made use of a 600 MHz NMR spectrometer equipped with a cryogenic 1.7 mm NMR probe, which enabled analysis of 70 µL extracts from 100 mg tissue. As reported here, our protocol enabled the detection of 52 polar and nine non-polar metabolites and revealed statistically different levels of non-polar metabolites in nestling birds fed on diets with either high carbohydrate or high protein but with identical lipid contents. 2. Results NMR data were collected on a 600 MHz NMR spectrometer equipped with a 1.7 mm cryogenic probe, which afforded the sensitivity needed for these small samples. Representative one-dimensional (1D) 1 H NMR spectra of the polar and non-polar fractions from the liver tissue investigated in this study are presented in Figures 1 and 2, respectively. We used peak intensities from assigned peaks in the 1 H NMR spectra of individual samples to determine the metabolite concentrations with assistance of Chenomx NMR suite 8.2 software (Chenomx Inc., Edmonton, AB, Canada). The 1 H NMR peak assignments to individual metabolites were verified and refined with reference to a 2D 1 H,13 C HSQC spectrum. To acquire the 2D 1 H,13 C HSQC spectrum (shown in Supplementary Materials Figure S1) in a reasonable amount of time, we combined the polar fractions from three birds. We assigned the peaks by comparing the chemical shifts of the 1D 1 H and 2D 1 H,13 C HSQC NMR spectra with those from reference spectra deposited in the Biological Magnetic Resonance data Bank (BMRB) [10] and the Human Metabolome Database (HMDB) [11]. We were able to identify 52 metabolites present in the polar fraction and nine in the non-polar fraction. Tables 1 and 2 list the metabolites identified in the polar and nonpolar phases, respectively. In the polar fraction, we identified essential amino acids (valine, isoleucine, leucine, lysine, methionine, phenylalanine, threonine, and tryptophan), conditional amino acids (arginine, glutamine, tyrosine, glycine, proline, and serine), and nonessential amino acids (alanine, asparagine, and aspart (...truncated)