Determination of the Isotopic Enrichment of 13C- and 2H-Labeled

---

Subscriber access provided by READING UNIV

Article 13

2

Determination of the Isotopic Enrichment of C- and HLabeled Tracers of Glucose Using High-Resolution Mass Spectrometry – Application to Dual- and Triple-Tracer Studies Martin Trötzmüller, Alexander Triebl, Amra Ajsic, Juergen Hartler, Harald C. Köfeler, and Werner Regittnig Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.7b03134 • Publication Date (Web): 31 Oct 2017 Downloaded from http://pubs.acs.org on November 4, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 14

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Determination of the Isotopic Enrichment of 13C- and 2H-Labeled Tracers of Glucose Using High-Resolution Mass Spectrometry – Application to Dual- and Triple-Tracer Studies Martin Trötzmüller,†,◊,‡ Alexander Triebl,†,‡ Amra Ajsic,§ Jürgen Hartler,∥,◊ Harald Köfeler,†,◊ and Werner Regittnig*,§ †

Center for Medical Research, Core Facility for Mass Spectrometry, and §Department of Internal Medicine, Division of Endocrinology and Diabetology, Medical University of Graz, Auenbruggerplatz 2, A-8036, Graz, Austria ◊ Omics Center Graz, BioTechMed-Graz, Stiftingtalstrasse 24, 8010 Graz, Austria ∥Institute for Computational Biotechnology, Bioinformatics Group, Graz University of Technology, Petersgasse 14, A-8010 Graz, Austria ABSTRACT: Multiple-tracer approaches for investigating glucose metabolism in humans usually involve the administration of stable and radioactive glucose tracers and the subsequent determination of tracer enrichments in sampled blood. When using conventional, low-resolution mass spectrometry (LRMS), the number of spectral interferences rises rapidly with the number of stable tracers employed. Thus, in LRMS, both computational effort and statistical uncertainties associated with the correction for spectral interferences limit the number of stable tracers that can be simultaneously employed (usually two). Here we show that these limitations can be overcome by applying high resolution mass spectrometry (HRMS). The HRMS method presented is based on the use of an Orbitrap mass spectrometer operated at a mass resolution of 100000 to allow electrospray-generated ions of the deprotonated glucose molecules to be monitored at their exact masses. The tracer enrichment determination in blood plasma is demonstrated for several triple combinations of 13C- and 2H-labeled glucose tracers (e.g., [1-2H1]-, [6,6-2H2]-, [1,6-13C2]glucose). For each combination it is shown that ions arising from 2H-labeled tracers are completely differentiated from those arising from 13C-labeled tracers, thereby allowing the enrichment of a tracer to be simply calculated from the observed ion intensities using a standard curve with curve parameters unaffected by the presence of other tracers. For each tracer, the HRMS method exhibits low limits of detection and good repeatability in the tested 0.1-15.0% enrichment range. Additionally, due to short sample preparation and analysis times, the method is well suited for high-throughput determination of multiple glucose tracer enrichments in plasma samples.

Whole-body glucose fluxes are usually determined by means of isotopic tracer techniques.1-4 These techniques typically involve an infusion of an isotopic tracer of glucose into the bloodstream, sampling of blood plasma after mixing between tracer and body glucose pool is complete, and measuring the isotopic enrichment in plasma glucose. From the enrichment measurements and tracer infusion rates, the rate by which glucose is produced by the body can then be calculated. During fasting, this endogenous glucose production (EGP) rate equals the overall rate of tissue glucose disposal (Rd). In response to ingesting a carbohydrate-containing meal, however, the situation is more complex, as EGP and Rd as well as the appearance of glucose arising from the meal (Ra meal) vary with time. To simultaneously quantitate these time-varying glucose fluxes, dual- and triple-tracer techniques have been commonly applyed.1,5-8 In many clinical studies applying these techniques, stable-labeled glucose isotopologues are used as tracers (e.g., [6,6-2H2]glucose, [1-13C1]glucose, [U-13C6]glucose) and measurement of their enrichment in sampled blood plasma is accomplished by quadrupole-mass spectrometers (Q-MS) and, less frequently, by isotope ratio mass spectrometers. Typically, these low resolution mass analyzers are coupled to gas chromatography (GC) or liquid chromatography (LC) systems and used to measure the relative intensities of selected ions

deriving from unlabeled glucose molecules (tracee) and labeled glucose tracers in the blood sample. From the observed relative ion intensities, the tracer enrichments, usually expressed as tracer-to-tracee ratios (TTR), are then estimated. However, because of the natural occurrence of heavier isotopes in the monitored ions (e.g., ~1.1% of carbons are 13Catoms), heavier isotopic variants of ions deriving from the tracee will be present in the mass analyzer and may make significant contributions to the signal in the mass channels designated for the glucose tracers. Therefore, prior to calculating TTR values from relative ion intensity data, it is necessary to correct the raw MS data for intensity contributions arising from naturally occurring isotopic variants of the monitored ions. Unfortunately, correction of ion intensity data for these interferences becomes considerably more complicated when two or more glucose tracers are used at once and when recycling of one or more of the adopted glucose tracers occurs during experiments. For instance, in a human study using [113 C1]-, [6,6-2H2]-, and [U-13C6]glucose simultaneously as tracers,8 tracer enrichments were estimated from relative ion intensity data obtained by GC-Q-MS analysis of the pentaacetate derivatives of glucose isotopologues using the positive chemical ionization mode and selective ion monitoring at mass-tocharge ratios (m/z) of 331 (M+0 ion), 332 (M+1 ion), 333

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(M+2 ion), and 337 (M+6 ion), corresponding to tracee, [113 C1]-, [6,6-2H2]-, and [U-13C6]glucose molecules, respectively. Due to the high number of carbon, hydrogen and oxygen atoms in the monitored ions (C14H19O9), signal intensity contributions from isotopic variants of ions deriving from the tracee molecules in the M+1 and M+2 channels are ~13.2% and ~2.5%, respectively. Thus, in this case, tracee isotopologues significantly overlap with both the [1-13C1]-, and [6,62 H2]glucose tracers. In addition, ~12.4% of ions deriving from the [1-13C1]glucose tracer appears in the M+2 channel and, hence, greatly interferes with the intensity measurement of the M+2 ions deriving from the [6,6-2H2]glucose tracer. Moreover, recycling of the infused [U-13C6]glucose tracer occurred via the gluconeogenesis pathway, thereby generating significant amounts of glucose isotopologues with masses of M+1, M+2, and M+3. Thus, it was necessary to also determine the signal intensity contributions in the M+1 and M+2 channels arising from the recycled glucose isotopologues. This was done by additionally monitoring M+3 ions while assuming that equal amounts of M+1, M+2 and M+3 glucose isotopologues were generated during tracer recycling. To correct the ion intensity data for all the mentioned interferences, the authors had to solve a rather complicated system of algebraic equations using matrix calculus.8 Besides the increased risk of introducing systematic errors,9 the use of complex procedures for correcting ion intensity data also gives rise to additional uncertainties (or random errors) that contribute to the spread in the final tracer enrichment results. In general, the amount of these additional uncertainties depends on the magnitude and number of corrections applied.10,11 To reduce the complexity of correcting relative ion intensity data, previous studies have used a radioactive tracer (e.g., [6-3H1]glucose) in place of one of the three stablelabeled tracers.1,6 Due to the high analytical sensitivity associated with the measurement of radioactivity, radioactive tracers can be infused at low rates. Thus, the amount of tracer present in the blood plasma is generally below the level of detection of the MS analyzers and, therefore, no interference with mass spectral data occurs. However, in this case, the reduction in complexity of correcting relative ion intensity data comes at the price of exposing study subjects to radiation. Thus, the potential hazard of radiation exposure precludes the combined use of radioactive and stable-labeled tracers in studies involving particular population groups, such as children or pregnant women. Another way of reducing the complexity may be the use of multiply-labeled instead of singly-labeled glucose tracers (e.g., [U-13C6; 1,2,3,4,5,6,6-2H7]glucose in place of [113 C1]glucose).7 Thereby, ions deriving from multiply-labeled glucose tracers will appear in higher mass channels (e.g., M+8) and, thus, may not interfere with intensity measurements of ions (e.g., M+2 ion and M+4 ion) deriving from tracers with lower masses (e.g., [6,6-2H2]- and [U-13C6]glucose tracers).7 However, the significantly higher costs of highly-substituted glucose tracers may limit their use in dual- and triple-tracer studies. Furthermore, in the case of recycling of highlysubstituted glucose tracers, it is very likely that glucose isotopologues of higher mass are formed.7 Ions arising from these isotopologues (e.g., M+5 and M+4 ions) may interfere with intensity signals in mass channels for the other tracers used in the experiment (e.g., [U-13C6]glucose). This may then necessitate additional corrections of raw MS intensity data.7 Because of the mass defect in the atomic nuclei, the increase in molecular mass arising from substitution of 12C by 13C

Page 2 of 14

(∆1.003355 Da) differs slightly from that arising from substitution of 1H by 2H (∆1.006277 Da).12 Since this small difference in the increases in molecular mass (2.922 mDa) can be resolved by high-resolution mass spectrometry (HRMS),13,14 we reasoned that by using HRMS in dual and triple tracer studies, 13C- and 2H-containing isotopologues of glucose having the same nominal mass, but slightly different exact masses, could be distinguished. This, in turn, would allow the enrichment of individual 13C- and 2H-labeled glucose tracers to be determined without mutual interference. For example, in the multiple-tracer case with [6,6-2H2]- and [1-13C1]glucose, determination of the [6,6-2H2]glucose tracer enrichment will no longer be influenced by the presence of the [1-13C1]glucose tracer, as under high resolution conditions, the ions deriving from the isotopic variants of the [1-13C1]glucose tracer can be completely differentiated from those deriving from the isotopic variants of the [6,6-2H2]glucose tracer (Table 1). Previously, HRMS has, for example, been used to analyze the stable isotope-labeling patterns in fatty acids,15,16 peptides,16,17 and proteins17 that arose from the administration of stable isotopelabeled precursors of these metabolites (e.g., 2H2O). To the best of our knowledge, HRMS has not been applied so far to the measurement of enrichments of metabolite tracers, like the enrichment of tracers of glucose. Here we report on the development and validation of an HRMS method for the enrichment determination of multiple 13C- and 2H-labeled glucose tracers in human blood plasma.

EXPERIMENTAL SECTION Chemicals. Acetonitrile, methanol (both Chromasolv grade), ammonium hydroxide solution (28% NH3 in H2O), D[1,6-13C2]glucose (99% 13C), D-glucose, D-fructose, Dsorbitol, and D-mannitol were purchased from Sigma-Aldrich (St. Louis, MO, USA). D-[6,6-2H2]glucose (99% 2H), D-[12 H1]glucose (98% 2H), D-[1-13C1]glucose (99% 13C), and D[U-13C6]glucose (99% 13C) were purchased from Cambridge Isotope Laboratories (Andover, MA, USA). D-mannose and D-galactose were purchased from Fluka Analytical (Buchs, Switzerland). Deionized water was obtained from a MilliQ Gradient A10 system (Millipore, Billerica, MA, USA). Sample preparation. Following administration of the glucose tracers to the study subjects, blood samples were collected in vials (2 ml FX Vacuette PET tube; Greiner Bio-One, Kremsmünster, Austria) containing an anticoagulant (potassium oxalate) and an anti-glycolytic agent (sodium fluoride). Collected blood samples were immediately put on ice, centrifuged at 4°C (2000 x g for 15 min), and supernatants (200 µL) transferred to glass vials (2 ml-screw cap vial, Agilent Technologies, Warsaw, Poland). To this, 600 µL of acetonitrile were added for protein precipitation. After vortexing, the samples were centrifuged (3000 x g) for 10 min at 20 °C. Afterwards, 400 µL of the supernatants were transferred to new glass vials and the solvents evaporated in a vacuum centrifuge (SC250EXP SpeedVac, Thermo Fisher Scientific Inc., Waltham, MA, USA). The dry samples were then reconstituted in 200 µL of acetonitrile/H2O (1:1; v/v), filtered through a 0.2 µm nylon syringe filter (Phenex-NY, Phenomenex, Torrance, CA, USA) into a 0.2 ml-HPLC vial (TopSert-TPX, VWR, Radnor, PA, USA) and stored at -80 °C prior to analysis. Instrumentation. Chromatographic separation was performed using an LC system (Dionex Ultimate 3000 RSLC; Thermo Fisher Scientific Inc.) equipped with an Acquity

ACS Paragon Plus Environment

Page 3 of 14

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

UPLC BEH Amide column (2.1 x 100 mm, 1.7 µm; Waters Corp., Milford, MA, USA). The column temperature was maintained at 35 °C. A sample volume of 4 µL was injected and isocratic elution was performed with acetonitrile/H2O (8:2; v/v) containing 0.1 vol% of the NH3-solution as additive at a flow rate of 150 µL/min for 15 min. The autosampler tray was kept at 6 °C. Before each sample injection, the LC system was flushed with 50 µL of acetonitrile/deionized water (1:1; v/v). The LC system was connected to an Orbitrap Velos Pro hybrid mass spectrometer (Thermo Scientific Inc.) through a heated-electrospray ionization interface (HESI II Probe), operating in negative ionization mode. The electrospray parameters were as follows: source voltage, 3.8 kV; source heater temperature, 250 °C; sheath, auxillary and sweep gas, 30, 12 and 1 arbitrary units, respectively; capillary temperature, 300 °C. The Orbitrap mass analyzer was operated in full scan mode at a resolution of 100 000 (defined at m/z 400) with a scan window of m/z 100 – 350. Automatic gain control (AGC) was set to allow 105 ions to enter the mass analyzer. Using the water loss of monoisotopic glucose (m/z 161.0455) as lock mass, deprotonated ions of glucose and sorbitol/mannitol were monitored at m/z-values given in Table 1. Integration of the observed ion peaks was performed using the LDA Software Tool.18 The chromatographic and mass spectrometric settings described above were determined in several sets of tuning experiments. Use of these optimized settings may provide proper chromatographic peak shapes and may avoid overfilling and under-filling of the Orbitrap. Overfilling of the Orbitrap may result in coalescence of closely spaced isotopologue peaks19 (see Supporting Information Figure S1), while under-filling may lead to inaccurate and imprecise relative ion intensity measurements.20 Table 1. Natural Abundance and Exact Mass of Stable Isotopologues of Glucose Abundance b (%)

Mass of [M-H](Da)

100.000

179.05611

C1 C5 H12 O6 C61H1217O116O5 12 C62H11H1116O6

6.48944 0.22856 0.13802

180.05947 180.06033 180.06239

2.922 2.060 0.000

12

1.23300 0.17547 0.01483 0.00022 0.00896 0.00032 0.00009 n.a.

181.06036 181.06282 181.06368 181.06455 181.06574 181.06661 181.06867 181.07176

8.308 5.844 4.982 4.119 2.922 2.060 0.000 3.097