The 100 isotopologue challenge: spectrometry as a means of high‐dimension clumped and position‐specific isotope analysis John Eiler1, Brooke Dallas1, Elle Chimiak1, Johannes Schwieters2, Dieter Juchelka2, Alexander Makarov2, Jens Griep‐Raming2 1Division of Geological and Planetary Sciences, California Institute of Technology, Pasadena CA, USA Overview 2 Example of Use The distribution of rare isotopes (D, 13C, 15N, etc.) at molecular scales — site‐specific Thermo Fisher Scientific, Bremen, Germany Representative analyses of the site-specific 13C effects and/or multiple substitutions, or ‘clumping’ — offers many potential tools in structures of three amino acids from different environmental chemistry, geochemistry, biochemistry and other disciplines. Existing sources. C1, C2 and C3 are carboxyl, amine and technologies for observing molecular‐scale isotopic structures (NMR, sector mass Methods I: Results I: Resolution and Sensitivity methyl carbons, respectively. All reported 13C spectrometry, infrared spectroscopy) enable only a few such measurements, with values are relative to an arbitrary standard (VWR Here we explore Fourier transform mass spectrometry (FTMS) as a solution to this problem, using the A select portion of the ethyl benzene mass limits on sample types and sizes. We present the results of an experiment to develop Resolution (FWHM) of 877,000 stock L alanine). All three materials span a Q Exactive GC system— a recent innovation on Orbitrap‐based FTMS, where analytes are delivered by Mass resolution of the Orbitrap mass analyzer depends on TM TM TM 13 12 12 13 the Thermo Scientific QExactive GC Orbitrap™ instrument— an Orbitrap Fourier C C4H5 C5H6 narrow range in bulk  C but they differ 13C12C H acquisition time and analyte mass, but generally greatly exceeds PDB gas flow, are generated by impact, mass‐range selection is made with the ‘Advanced 4 5 12 transform mass — as a generalized tool for precise, accurate, rapid and C5H4D strongly in site specific ‘finger print’. Quadrupole Technology’ (AQT), and anything possible by the sector instruments used for isotope sensitive measurements of molecular isotopic structures. Results to‐date indicate this 3,4, 5 12 ratio analysis, and is clearly sufficient to separate common 5: The selected ions (or their collisional fragments) C5H6 mass analysis is made using the Intensity 12C H D technology can achieve uncertainties in the range of 0.1‐1‰(partperthousand, are stored in the C trap (repeat as desired) 5 4 Fields of Application Orbitrap detection. Peaks of interest isobars (upper left). Separation of isotopologues of organics and relative, in abundance ratio) for many isotopic species, analytes and sample sizes, and may also be subjected to an MS‐MS Mass (amu) their fragment ions is robust up to ~500 amu (lower left). • Forensics: Isotopic ‘fingerprinting’ that is rapid enough to measure ~10‐100 isotopic constraints on one compound. 4: (optional): Selected masses may be passed to experiment using the HCD collision Sensitivity varies with analyte and tuning, but is generally on the includes ~10‐100 independent measures of site‐ the HCD cell and subjected to fragmentation by order of 105 molecules per , with baselines of a few ions per cell. This system routinely achieves energetic collisions specific and multiple substitutions will greatly Introduction 1000000 second, implying sample sizes <10‐6 mols for many isotopologues advance the specificity of molecular sources in Most molecules of interest to forensics, medicine or other areas of applied chemistry mass resolutions of 105,andenables 3: The AQT is used to select a mass range of interest (≥0.4 amu range, with resolution of ~1000) Orbitrap 800000 6" (please see below). forensic studies of compounds relevant to exist in many isotopic forms — isotopologues that differ in their numbers and sites of rapid, automated analysis of large Resolution (FWHM) 2: All ions are extracted, 1012 environmental pollution, crime, performance isotopic substitution. For example, consider testosterone: numbers of peaks across a large mass 600000 transferred through the bent Abundance of isotopologue, as Cocaine Non‐ 2 M/∆M enhancing drugs, and national security. range. 4 5 flatapole, and collisionally 1010 fraction of all all ions • ~2x106 isotopologues adjacent 3 400000 • Biogeochemistry: Proportions of multiply 13C cooled • ~10,000, restricting the list to multiple 8 and 1 200000 10 substituted and site‐specific isotopologues are substitution those that are clearly abundant D collected

13 and 6 known to constrain temperatures and enough for analysis C 6 0 10 Adjacent multiple 0 200 400 600 800 1000 Minimum for 1 ‰ relative error • Almost all are both multiply 13 Ions mechanisms of molecular synthesis, identities of Site‐specific C or 4 substituted (‘clumped’) and ‘site substitution Mass (amu) 10 differences source substrates, and mechanisms and specific’ 6: After steps 1‐5 are performed 1 or more 10‐12 10‐10 10‐8 10‐6 10‐4 Mols analyte amounts of molecular destruction or loss. These times, all ions accumulated in the C trap are capabilities will open new windows on the Proportions of these isotopologues may vary as functions of many natural and injected into the Orbitrap and observed 1: Analyte is introduced and ionized by electron Results II: Precision and Accuracy origins of ancient biomarkers, paleoclimate engineered processes, and therefore an analysis that observes some large number of bombardment 10000 records based on organic compounds, among them could provide exceptionally detailed constraints on molecular origins. other uses. (s) Mass spectrometry is the only technology that currently seems capable of quantifying 1000 • Personalized medicine and metabolomics: Fatty acids and Hopanes ‰

Methods II: Sample Introduction 10 min 1

±1SE

large numbers of site‐specific and clumped isotope species in small samples (sub TM High‐dimensionality, sensitive measurements of (3) Two purged reservoirs — room temperature liquids and solids (2) Conflo open split — room temperature gases 900 ppm 100 micromol) of diverse compounds. This is because peaks in a molecule’s mass spectrum 126 molecular isotopic structure will enable

Helium delivery 1 Helium delivery 2 To Q Exactive GC system Xe reach 1 min 128 often come from some known (or knowable) combination of molecular sites, and such Xe to constraints on metabolite and drug budgets 1.9 % 10 He delivery Observed peaks exist in numerous isotopic forms. Sample reservoir 1 Sample reservoir 2 Gas 2 (‘out’ position) normally obtained from isotopic labeling Time C H + fragment experiments, but using the naturally occurring 2 5 Gas 1 (‘in’ position) 1 12C H 3 8 Shot noise ±1SE 0.001 0.01 0.1 1 isotopes. 12 13 C2 CH8 Relative abundance of rare isotope 12 + C3H7D 1 ‰ in 3 minutes; 0.4 ‰ in 12.5 minutes Methionine CH fragment 12 13 13 4 port valve To Ion Source 3 C C2H8 C3H8 (as fraction of largest peak concurrently collected) 12 13 12 13 (Reservoir 2 C2 CH7D C C2H7D Even the mass spectrum of something as intensity 12 A key requirement of accurate and Conclusions Full molecule 10 C H D 12 13 selected) 4 port valve 3 6 2 C CH6D2 Time (minutes) 12 simple as propane contains families of peaks Log (Sample to mass spec) External errors of Orbitrap analyses of C3H5D3 Vent precise isotope ratio analysis is 10 20 30 40 50 60 Orbitrap‐based FT isotope ratio mass spectrometry (FT‐IRMS) has several strengths: that sub‐sample different portions of the (vacuum) isotope ratios are generally limited by shot‐ 100 44 45 46 47 Vent Helium delivery 3 highly controlled sample/standard Mass resolution of substitutions involving isotopes of H, C, N, O and S at molecular

molecule, and each peak exists in several max) (vacuum) (column flow from GC) noise errors alone, down to limits of ~0.1‐ 80

singly‐ and multiply‐substituted forms, (% comparison. We achieve this using 60 or fragment masses ≤500 amu; the speed and mass range to measure many such potentially offering dozens of constraints on 1‰ (upper left). This fact implies that 40 12 13 any of three systems we have adapted to the Q Exactive species per sample in minutes to hours; the sensitivity for quantitative analysis of C C2H8 proportions of site‐specific and ‘clumped’ limits of precision are mostly controlled by 20

Intensity sub‐micro‐molar samples, or doubly and triply substituted species at their trace isotopologues. Injector GC instrument: (1) A GC column for analysis of eluted 0.164 (1) GC injection — mobile solutes relative peak intensities and time (upper solute peaks, adapted to send sample peaks through a 12 17 16 natural abundances; the ability to receive diverse analytes in several forms and 12C 13CH D C O O 0.162 Intensity 2 7 To Ion Source 12C H D right); the fastest measurements generally 12 18 16 3 6 2 reservoir that broadens them, increasing the time a C O O 0.160 delivery systems; and precision and accuracy comparable to existing technologies. Peak look at only a portion of the mass 46.070 46.075 broadener peak is observed; (2) an open split for analysis of room‐ 0.158 These capabilities will enable a large number of new applications in biological, Helium delivery spectrum at one time. Sample/standard 1.0201234567891011 GC Column temperature gases; and (3) a pair of reservoirs that can chemical and natural sciences. bracketing (lower right) permits Sample However, established mass spectrometric approaches lack the mass resolution to 4 port valve For research use only. Not for use in diagnostic procedures. stably deliver sample or standard to the ion source for Standard 1.018 study any but the smallest molecules (~50‐70 amu or less), and are far too slow to (Column flow to mass spec) standardization with accuracy similar to 18.6±0.2 ‰ © 2016 Thermo Fisher Scientific Inc. All trademarks are the property of Thermo Fisher Scientific and its subsidiaries unless minutes to hours. otherwise specified. This information is not intended to encourage use of these products in any manner that might infringe the Vent Column flow precision (tenths of ‰). 1.016 constrain more than a handful of isotopologues per molecule. intellectual property rights of others. (vacuum) 12345678910