Rapid Quantification and Isotopic Analysis of Dissolved Sulfur Species

Research Article

Received: 19 September 2016 Revised: 6 January 2017 Accepted: 23 February 2017 Published online in Wiley Online Library

Rapid Commun. Mass Spectrom. 2017, 31, 791–803 (wileyonlinelibrary.com) DOI: 10.1002/rcm.7846 Rapid quantification and isotopic analysis of dissolved sulfur species

Derek A. Smith1*,† , Alex L. Sessions1, Katherine S. Dawson1, Nathan Dalleska2 and Victoria J. Orphan1 1Division of Geological and Planetary Sciences, California Institute of Technology, 1200 E California Blvd, Pasadena, CA 91125, USA 2Environmental Science and Engineering, California Institute of Technology, 1200 E California Blvd, Pasadena, CA 91125, USA

RATIONALE: Dissolved sulfur species are of significant interest, both as important substrates for microbial activities and as key intermediaries in biogeochemical cycles. Species of intermediate oxidation state such as sulfite, thiosulfate, and thiols are of particular interest but are notoriously difficult to analyze, because of low concentrations and rapid oxidation during storage and analysis. METHODS: Dissolved sulfur species are reacted with monobromobimane which yields a fluorescent bimane derivative that is stable to oxidation. Separation by Ultra-Performance Liquid Chromatography (UPLC) on a C18 column yields baseline resolution of analytes in under 5 min. Fluorescence detection (380 nm excitation, 480 nm emission) provides highly selective and sensitive quantitation, and Time-of-Flight Mass Spectrometry (TOF-MS) is used to quantify isotopic abundance, providing the ability to detect stable isotope tracers (either 33Sor34S). RESULTS: Sulfite, thiosulfate, methanethiol, and bisulfide were quantified with on-column detection limits of picomoles (μM concentrations). Other sulfur species with unshared electrons are also amenable to analysis. TOF-MS detection of 34S enrichment was accurate and precise to within 0.6% (relative) when sample and standard had similar isotope ratios, and was able to detect enrichments as small as 0.01 atom%. Accuracy was validated by comparison to isotope-ratio mass spectrometry. Four example applications are provided to demonstrate the utility of this method. CONCLUSIONS: Derivatization of aqueous sulfur species with bromobimane is easily accomplished in the field, and protects analytes from oxidation during storage. UPLC separation with fluorescence detection provides low-μM detection limits. Using high-resolution TOF-MS, accurate detection of as little as 0.01% 34S label incorporation into multiple species is feasible. This provides a useful new analytical window into microbial sulfur cycling. Copyright © 2017 John Wiley & Sons, Ltd.

Sulfur exists in the natural environment in myriad forms, oxygen minimum zones of the world’s oceans,[5,6] the low 2– fi ’ ’ [7,8] ranging in oxidation state from sulfate (SO4 ) to sul de sulfate methanogenic zone of marine sediments, salt – (HS ) and including many species of intermediate oxidation marshes,[9] freshwater sediments,[10,11] and within symbiotic fi – 0 [9,12] state, such as polysul de (HSx ), elemental sulfur (S ), microbial associations. In all cases, microbial sulfur thiosulfate, trithionate, sulfite, and others. Species such as cycling is thought to play key roles in the remineralization of thiosulfate, trithionate, and sulfite in particular are thought organic matter. to be important intermediaries in microbial reactions that Sulfur speciation also dramatically affects the solubility and – transform (either oxidize, reduce, or disproportionate) sulfur bioavailability of trace metals,[1 3] abiotic oxidation and and are frequently coupled to the biogeochemical cycles of reduction of manganese and iron,[1,4] and sulfurization of – carbon, oxygen, and iron.[1 4] ’Cryptic’ sulfur cycles driven organic matter.[13,14] Studies of all these processes, and others, by microbial metabolism have been hypothesized to occur in would benefit greatly from improved methods for quantifying dissolved sulfur species, particularly the most transient, intermediate compounds. Moreover, an ability to distinguish * Correspondence to: D. A. Smith, Department of Earth and stable isotope labels in such compounds would enable a Planetary Sciences, Washington University in Saint Louis, wide variety of stable isotope probing experiments that Campus Box 1169, Rudolph Hall Rm 110, 1 Brookings could help illuminate the cryptic microbial sulfur cycle. A Drive, St. Louis, MO 63130, USA. variety of techniques to measure dissolved sulfur species E-mail: [email protected] have previously been described. One of the most widely † Current address: Department of Earth and Planetary utilized techniques, the Cline assay, employs absorption Sciences, Washington University in Saint Louis, Campus spectrophotometry of a N,N-dimethyl-p-phenylene diamine, Box 1169, Rudolph Hall Rm 110, 1 Brookings Drive, St. chloride, and sulfide complex (methylene blue) to quantify 791 Louis, MO 63130, USA. hydrogen sulfide.[15]

Ion chromatography (IC) has been employed to separate have been previously used to quantify sulfur species from multiple anions of sulfur (i.e., [16,17]). However, IC eluent a range of diverse samples including hydrothermal vent fl 2– 2– chemistry can alter sulfur speciation; some sulfur species, uids, sulfur-oxidizing bacterial cultures (SO3 ,S2O3 ), – – e.g. polysulfides, are too reactive for accurate quantification HS in human blood, and cysteine and HS in – with IC; further, IC provides no isotopic information, methanogens.[27 31] although it has been combined with multi-collector To improve on the separation of derivatized sulfur species inductively coupled plasma mass spectrometry (MC-ICP- by HPLC, which takes ~60 min per sample,[24,25] we employ MS) to achieve this.[18,19] Derivatization of sulfur species with Ultra-Performance Liquid Chromatography (UPLC) for rapid methyl trifluoromethanesulfonic acid (methyl triflate) is (<5 min) separation of derivatized sulfur species. For capable of stabilizing polysulfides with quantification detection and quantitation, we evaluated both a conventional down to approximately 25 μM by UV absorption fluorescence detector and a state-of-the-art Time-of-Flight spectrophotometry coupled to high-performance liquid (TOF) mass spectrometer (TOF-MS). Fluorescence detection chromatography (HPLC); however, with extraction and is less expensive and more sensitive than MS, and evaporation post-extraction, lower limits are reported.[20] nondestructive so it can be coupled with preparative fraction Drawbacks of the methyl triflate method include the toxicity collection of individual sulfur species, e.g. for subsequent of the reagent, multistep extraction procedure, addition of high-precision analysis of natural abundance isotope ratios. sulfur to the derivatized product, and applicability only to However, fluorescence itself provides no isotopic information, reduced sulfur species. Cyclic voltammetry and in situ probes which is the main impetus for using MS. The high-resolution have also been used to measure and quantify many of the TOF-MS employed here allows us to distinguish isobaric reactive sulfur species in environmental samples at relatively interferences in the molecular ions, and accurately quantify high spatial resolutions (mms).[9,21,22] The limitations of cyclic 34S enrichment down to levels of ~0.6%. The same voltammetry are that it cannot distinguish between the methodological approach could presumably be used for 33S isotopes of sulfur. enrichments. We note that while TOF-MS was used in this Here, we adapt a previously described method for study, quadrupole or other low-resolution MS could be an preserving aqueous nucleophilic sulfur species as bimane alternative, with the caveat that isotopic discrimination might derivatives, and update their separation and detection to suffer. modern instrumentation. Key features of the derivatization To refine and test this updated methodology, we focused on 2– 2– – are that (1) reaction with bromobimane (mBBr) is four sulfur analytes: SO3 ,S2O3 ,CH3SH, and HS . Other quantitative thus causing no isotopic fractionation; (2) the sulfur species are also potentially accessible with this resulting bimane derivatives are very stable towards approach. These analytes were measured in standard subsequent oxidation, providing a convenient means to solutions (both a dilution series of decreasing concentration, preserve samples in the field; and (3) bimane is strongly and a series of varying isotopic composition), then fluorescent, enabling non-destructive fluorescence characterized in four separate proof-of-principle applications: – detection.[23 26] The quantitative mBBr reaction proceeds as (1) measurement of sulfite and thiosulfate concentrations in a nucleophilic substitution of sulfur compounds containing marine sediment pore-waters, where they are present at low an unshared pair of electrons (Fig. 1), which includes sulfite μM concentrations; (2) measuring 34S appearance in bisulfide 2– 2– fi (SO3 ), thiosulfate (S2O3 ), organic thiols (R-SH), bisul de in a pure culture of sulfate-reducing bacteria amended with – fi 2– 0 34 2– 34 fi (HS ), and polysul des (Sn ) but not elemental sulfur (S ) SO4 tracer; (3) measuring S in bisul de in methane seep 2– 34 2– or sulfate (SO4 ). mBBr is a neutral molecule capable of sediments incubated with SO4 tracer; and (4) identifying crossing the cellular membrane, enabling reactions with both previously unreported intermediate sulfur species in a culture extra- and intracellular sulfur pools. Bimane derivatives of sulfur-oxidizing bacteria.

Figure 1. The nucleophilic substitution reaction occurs where a nucleophilic sulfur molecule replaces bromide from bromobimane. The reaction is buffered by HEPES at a pH of 8 and EDTA is used as a chelating agent. In the case of bisulfide, two moles of bromobimane are reacted per one mole of bisulfide. However, most other nucleophilic 792 sulfur molecules form a terminal group at the end of a single bimane.

EXPERIMENTAL Mass spectrometry The TOF mass spectrometer employed electrospray ionization Derivatization protocol (ESI) in both negative and positive ion mode, optimized as Monobromobimane (hereafter just bromobimane or mBBr) follows: source temperature 120°C, desolvation temperature powder (≥97%; Sigma Aldrich, St. Louis, MO, USA) was 500°C, desolvation gas flow 800 L/h, capillary voltage dissolved in acetonitrile (ACN) to produce a working stock 2.5 kV, cone voltage 50 V, cone gas flow 40 L/h, and flight tube solution of 50 mM mBBr. This was diluted prior to use to a 9.0 kV. Collision energy was 6.0 eV. Acquisition mass range concentration of ~2-fold stoichiometric excess relative to was 50–800 Da and the calibration mass range was 91.183– sulfur analytes. Some sulfur species have multiple reaction 1993.753 Da. A lock mass of 556.2771 Da was used to verify – sites, e.g. HS binds two bimanes, necessitating the excess mass accuracy of ±3 mDa. The scan time was 0.2 s with an reagent. On the other hand, too much mBBr creates an HPLC interscan time of 0.014 s from 0.0 to 4.5 min. Data was collected peak that can coelute with bimane thiols and produce large in centroid format. Sulfur species were quantified using the amounts of background fluorescence; thus we avoided molecular ion. adding a large excess. A typical mBBr reaction mixture was prepared by combining the 50 mM working stock solution Quantification of sulfur species of mBBr 1:1 with 50 mM HEPES and 5 mM EDTA in MilliQ fi water (at pH 8). An aliquot of 300 μL from the reaction Calibration standards for quanti cation were created by fi mixture was added to 100 μL of sample (e.g. a ratio of serial dilution of sodium sul te, sodium thiosulfate, sodium fi 3:1 v/v). Samples were reacted with mBBr for 1 h in the dark sul de and sodium thiomethoxide (all from Sigma-Aldrich) at room temperature. The reaction was then quenched with in N2-bubbled MilliQ water. Concentrations ranged from 600 μL 65 mM methanesulfonic acid in MilliQ water to 400 μL 3.335 nM to 6.667 mM. The integrated mass spectral peak fl of sample and bromobimane reaction mixture (e.g. ratio of area and integrated uorescence peak area were recorded 3:2 v/v). Samples were stored in amber glass vials at –20°C for each standard and sample using MassLynx® and until analysis. Both the mBBr reaction mixture and QuanLynx® software, which produced linear regressions of integrated peak area vs. concentration with average R2 methanesulfonic acid were bubbled with N2 gas prior to values ≥0.99. use to remove O2, and the entire derivatization procedure was carried out in an anaerobic chamber to minimize oxidation of analytes. Similar precautions to minimize Quantification of stable sulfur isotope enrichment oxidation during field sampling and derivatization are Several approaches for extracting isotopic enrichment data strongly recommended. from LC/MS spectra are possible. Here we adopt the approach of measuring the molecular ion and ~7–10 isotope peaks 34 Liquid chromatography (depending on level of S enrichment) for each analyte species, in both the sample and one or more standards. For Separation of the bimane derivatives was conducted on an each analyte, the average molecular weight of the bimane ACQUITY I-Class UPLC® system (Waters, Milford, MA, derivative was calculated as the weighted mean of the USA) equipped with a flow-through needle autosampler. molecular ion and isotopic peaks, using peak height as the Analytes were separated on an ACQUITY UPLC® BEH C18 weighting factor for abundance because peak area is not column (2.1 mm × 50 mm, 1.7 μm, 130 Å) maintained at 45°C available for data collected in centroid mode: using an injection volume of 1 or 2 μL. The elution profile contained a mobile phase comprised of 0.1% formic acid 10 M ¼ ∑ MnHn=∑Hn (LCMS grade), 1% ACN (LCMS grade), and 98.9% MilliQ water n¼1 (solvent A) and 100% ACN (LCMS grade) (solvent B) (Table 1). Effluent from the UPLC system was directed sequentially where Mn and Hn are the accurate mass and height, through an ACQUITY fluorescence (FLR) detector and a Xevo respectively, of a given mass spectral peak. M can then be G2-S TOF mass spectrometer (both Waters). The FLR detector compared between sample and standard to calculate the was operated at an excitation wavelength of 380 nm and an shift in mean molecular weight of the sample: emission wavelength of 480 nm. Δ ¼ M Msamp Mstd

Table 1. Optimized elution program for separation of Although this analysis confounds contributions from bimane derivatives of inorganic sulfur species isotopes of S, C, H, and N, the latter three are entirely derived from atoms in the bimane derivative, which is the same for Time (min) Flow rate %A %B both sample and standard and so cancel in the calculation of ΔM. Isotopes of oxygen are not necessarily identical, e.g. in 1 Initial 0.40 90.00 10.00 S2O3 or SO3, but with a difference of 26 mDa between 2 0.05 0.40 90.00 10.00 16O18O and 34S, the TOF mass spectrometer employed here 3 3.25 0.40 75.00 25.00 easily resolves these two isobars. ΔM can thus be interpreted 4 3.75 0.40 20.00 80.00 directly as the change in average mass due to sulfur isotope 5 3.95 0.40 20.00 80.00 substitution, and is linearly proportional to the fractional 6 4.00 0.40 90.00 10.00 incorporation of isotope label. The slope of the relationship 7 4.50 0.40 90.00 10.00 793 (between ΔM and atom% label) is ~1 if 33S is the label and

~2 if 34S is the label. For most applications then, a Values of δ34S were calibrated relative to an internal standard ‰ straightforward approach is simply to generate a linear of AgS2 (18.03 ); precision and accuracy for these calibration curve for ΔM using multiple standards of known measurements are estimated as <0.2‰. isotopic abundance. To achieve cross-calibration of the two methods, two To be able to compare data collected by LC/MS to isotope aliquots from seven sulfide samples with increasing atomic ratios measured by isotope-ratio mass spectrometry (IRMS), percentages of 34S, ranging from 4.27 to 17.4%, were we first converted ΔM into the absolute (fractional) abundance generated. In an anaerobic chamber, 9.8 mL of a 23 mM of 34S using: 4.27% 34S sulfide solution was combined with 0.2 mL of a 34 fi 34 185 mM 100% S sul de solution to produce a 17.4% S stock Δ þ 34 M MVCDT solution. That stock was serially mixed with aliquots of the F ¼ 16:08997 34 fi 1:98759 4.27% S sul de solution to produce solutions of intermediate 34S abundance. Parallel aliquots of these standards were

where MVCDT is the mean atomic weight of sulfur in the VCDT derivatized with mBBr and measured by UPLC/TOF-MS, (Vienna Canyon Diablo Troilite) international reference and precipitated as zinc sulfide for measurement by material, and the constants 1.98759 and 16.08997 derive from EA/IRMS. the masses and abundances of sulfur isotopes in that reference material. The isotope ratio was then calculated Experiments with sulfate-reducing and sulfur-oxidizing from the following equation: bacterial cultures Stocks of Desulfococcus multivorans DSM 2059 and Sulfurovum 34S 34F 34R def ¼ lithotrophicum DSM 23290[34] were obtained from the German ½32S 134F Collection of Microorganisms (DSMZ) and initially propagated in the lab using the recommended media. and an enrichment factor (EnrF, i.e., fractionation factor) was Triplicate anaerobic cultures of D. multivorans were grown in calculated as: 30 mL butyl rubber stoppered serum bottles at 28°C to fi [34] 34 exponential phase in a modi ed MJ medium (g/L): NaCl RA α = ¼ • • A B 34 30.0, K2HPO4 0.434, CaCl2 2H2O 0.28, MgSO4 7H2O 3.40, RB • • MgCl2 6H2O 4.18, KCl 0.33, NH4Cl 0.35, NiCl2 6H2O • 0.50 mg, Na2SeO3 5H2O 0.50 mg, FeCl2 0.01, NaHCO3 1.5, analogous to the calculation typically employed by natural- • abundance IRMS measurements. Na2S2O3 5H2O 1.5, NaNO3 2.0, with the addition of 10 mL Sequentially diluted sulfur standards with concentrations DSM 141 Trace Element solution and 10 mL of DSM 141 and isotopic compositions that bracketed those of the samples Vitamin solution. The medium was supplemented with pyruvate to a final concentration of 10 mM, and sulfate were used to quantify S isotope enrichment. To determine the δ34 ‰ solution with a SVCDT value of +2000 was added to a effect that non-bracketed standards have on the determined fi EnrF of a sample, we performed the isotopic analysis nal concentration of 0.28 mM. The medium was kept described above with standards of bisulfide that were more anaerobic by saturation with 80:20 N2:CO2 gas, followed by adjustment of the pH to 7.0. During exponential phase than three orders of magnitude less concentrated than sample μ concentrations. This exercise produced an additional 6.2% (47.5 h after inoculation), 100 L from each culture was collected into a 1 mL syringe and transferred after filtration decrease in the recorded EnrF of the sample; however, as μ fi μ standard concentrations approach and overlap sample through a 0.2 m acrodisk lter into 300 L of the mBBr concentrations, this additional error is greatly decreased. Peak reaction mixture. Continuous culturing of S. lithotrophicum was carried out in a heights and exact masses were recorded either with fi MassLynx® or using a partially automated analysis with BioFlo115 fermentor/bioreactor (New Brunswick Scienti c). Oxidant availability, in this case the concentration of nitrate MAVEN (Metabolomics Analysis and Visualization – Engine).[32,33] As a test of precision, a zero-enrichment (NO3 ), was the growth-limiting factor. A working volume of experiment (i.e. comparison of two aliquots of the same 1.75 L was maintained with an agitation rate of 150 rpm. A constant pressure of 0.5 bar from a 80:20 N2:CO2 gas mixture reference standard) yielded enrichment factors of fi 1.000 ± 0.006 for both sulfide and thiosulfate. continuously saturated the growth vessel. Modi ed MJ medium as described above was used[34] (and DSMZ Medium • Isotope-ratio mass spectrometry (IRMS) 1011) with the addition of 1.5 mL 400 mM Na2S 9H2O and no addition of isotopically labeled substrates. An excess of the Sulfur isotope ratios and weight percentage for precipitated autotrophic carbon substrate, CO2, was continually supplied ZnS weighed into tin capsules (5 × 7 mm) were determined viathe80:20N2:CO2 gasmixture.Temperaturewasmaintained ™ on a Delta XLPlus isotope-ratio mass spectrometer (Thermo at 28°C. Influent anoxic sterile media was supplied from a 2 L Quest) attached to a NC2500 elemental analyzer (EA; CE glass bottle that was connected to a 3 L mylar gas bag over fl Instruments). All sulfur isotopic compositions are reported pressurized with the 80:20 N2:CO2 gas mixture and ef uent as per mil variations relative to VCDT following the waste was collected into a sterile anoxic 2 L glass bottle with a conventional delta notation: non-pressurized 3 L mylar gas bag. A low dilution rate of –1 ! 0.036 day was maintained throughout the experiment to ðÞ= promote slow growth. That dilution rate was approximately 34S 32S Sample – δ34S ¼ 1 half of the optimal growth rate of 0.069 day 1 determined in 794 ðÞ34S=32S VCDT batch culture. Once growth conditions of the bioreactor

wileyonlinelibrary.com/journal/rcm Copyright © 2017 John Wiley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2017, 31, 791–803 Quantification and isotopic analysis of dissolved sulfur species reached steady state, 25 mL aliquots were removed from the Table 2. List of representative sulfur compounds and their vessel; from these a 100 μL aliquot was collected into a 1 mL predicted reaction with bromobimane syringe and ﬁltered through a 0.2 μm acrodisk ﬁlter into 300 μL of the bromobimane reaction mixture. Molecular Reaction with Name formula mBBr

Sediment incubations – Bisulfide HS + 2– Sediment samples were collected on May 9, 2013, from the Sulfite SO3 + 2– Santa Monica Basin, offshore California, as part of an Thiosulfate S2O3 + – expedition led by the Monterey Bay Aquarium Research Polysulfides HSSn + Institute on the R/V Western Flyer. Using the ROV Doc Ricketts Cysteine C3H7NO2S+ (Dive DR463), sediment push cores PC 61 and PC 55 were Glutathione C10H17N3O6S+ collected in an extensive white microbial mat overlying an Coenzyme A C21H36N7O16P3S+ – active methane seep at a water depth of 860 m and in situ Coenzyme M C2H5O3S2 + temperature of 4°C (lat. 35.473431 N, long. 118.4007 W). Upon Organic Thiols R-SH + 2– collection shipboard, PC 55 was processed shipboard in 1 cm Polythionates SnO6 0 [35] intervals according to Orphan et al. Pore-water aliquots Elemental Sulfur S0 0 ’ from each depth horizon were collected using an argon- Adenosine 5 - C10H14N5O10PS 0 pressurized Reeburgh-type squeezer, preserved for phosphosulfate (APS) – geochemical analyses, and stored at –20°C. PC 61 was Sulfonates (i.e., HEPES) R-SO3 0 2– sectioned into two 6 cm intervals and heat sealed under anoxic Sulfate SO4 0 conditions in a mylar bag flushed with argon and stored at 4°C + indicates positive reaction, 0 indicates reaction will not occur until processing for microcosm incubation experiments R indicates alkyl or organic substituent (established 40 days after collection). Data sourced from Fahey and Newton[27] and Zopfi et al.[39] Stable isotope probing (SIP) experiments were conducted at 10°C in 5 mL butyl stoppered serum bottles with 1 mL of sediment slurry (0–6 cm PC 61) mixed with anoxic – artificial seawater containing 5 mM unlabeled sulfide and collection and analyzed 29 months later, yielded HS 34 2– 28 mM 100% SO4 for 40 days. Residual unlabeled SO4 concentrations that were identical within error to those 2– in the sediment slurry diluted the tracer to a level of ~80% obtained shipboard by the Cline assay, and SO3 and 34 2– S, based on initial sampling. The incubation was allowed S2O3 concentrations that were similar to previous reports to settle before time points of the supernatant were collected. (data not shown). Underivatized sulfide and sulfite do not The supernatant was centrifuged at 20,000 g for 5 min. A last nearly this long in the presence of O2, even when stored 50 μL aliquot of the overlying liquid was added to the mBBr at –20°C. reaction mixture and allowed to react as described above. The optimized UPLC elution gradient described above and Before analysis this solution was thawed and diluted 1:10 in Table 1 was able to fully resolve the bimane derivatives of 2– 2– – in 20:80 ACN:65 mM methanesulfonic acid. SO3 ,S2O3 ,CH3SH, and HS within 4.5 min (Fig. 2(A)), making this a very rapid separation method. However, the composition of the eluent was found to influence the separation and required modification depending on the RESULTS AND DISCUSSION 2– species of interest. For example, LC peaks representing SO3 – bimane and S O 2 bimane were observed to co-elute under Quantitation of dissolved sulfur analytes 2 3 conditions containing >90% of the polar eluent (0.1% formic The use of bimane derivatives for quantitation of dissolved acid, 1% ACN, and 98.9% MilliQ water). Similarly, >70% fi – sulfur species has many bene ts. Experimental testing of this organic eluent (100% ACN) was required to separate CH3S – assay showed reactivity with a broad range of aqueous sulfur bimane from S2 dibimane by 3.75 min. Additionally, as the species, and provided rapid, quantitative measurement with peak of unreacted mBBr elutes close to sulfide dibimane, the addition of a ~2-fold molar excess of the bromobimane excessively large reagent concentrations can lead to a broad (mBBr) reagent, and resulted in no isotopic fractionation. peak area that can influence the sulfide measurement. Limiting Bromobimane is reactive towards most sulfur species with the mBBr reagent to a 2-fold excess largely avoided this unpaired electrons, and so has the ability to target a wide problem, while still enabling quantitative preservation of variety of trace, labile species in addition to those studied sulfur intermediates. here (Table 2). Sulfate and elemental sulfur are notable Performance characteristics for quantitation of analytes by exceptions that do not react with mBBr. The reagent, while either FLR or TOF-MS are given in Table 3. Comparison of moderately expensive, is safe and easily deployed in the field the two methods revealed a higher degree of sensitivity using to preserve samples immediately after collection.[24,25,27] fluorescence detection over TOF-MS. The TOF-MS method Although we did not quantitatively study the stability of was found to be sensitive down to the low micromolar range bimane derivatives, qualitative evidence suggests they are (sample concentration before dilution in the reaction mixture); highly stable when stored at –20°C. Mixtures of derivatized based on an injection volume of 1 μL, equivalent to <1 pmol standards stored in this way were measured with no loss injected on-column. Both the FLR and TOF-MS detectors in signal for at least 3 months (longer times were not tested). became saturated at ≥3 mmol sulfur on-column, yielding a 795 Marine pore-water samples, derivatized at the time of dynamic range of ~9 orders of magnitude. Both detectors

Figure 2. Chromatography of sulfur species using bromobimane preservation and UPLC/TOF-MS analysis showing a 2– 2– – μ separation of a mixture of sulfur standards consisting of SO3 ,S2O3 ,CH3SH, and HS each at 250 M. (A) Extracted ion chromatogram of 1: sulfite bimane 0.55 min 271.0389 Da, 2: thiosulfate bimane 0.96 min 303.0609 Da, 3: methanethiol bimane 2.97 min 239.0854 Da, and 4: sulfide dibimane 3.25 min 415.1440 Da. (B–E) Mass spectra with the chemical structure, chemical formula, and isotopic distribution for the negative ion of sulfite bimane (B), negative ion of thiosulfate bimane (C), positive ion of methanethiol bimane (D), and positive ion of sulfide dibimane (E). [Color figure can be viewed at wileyonlinelibrary.com]

Table 3. Performance characteristics for FLR and TOF-MS detection of bimane derivatives

Sulfite bimane Thiosulfate bimane Sulfide dibimane

FLR Sensitivitya (S/N) 916:1 1333:1 1833:1 TOF-MS Sensitivitya (S/N) 165:1 369:1 371:1 FLR Precision (relative %) 2.35 6.10 3.95 TOF-MS Precision (relative %) 4.71 3.81 4.00 FLR Accuracya (± μM) 0.63 0.86 2.32 TOF-MS Accuracya (± μM) 3.31 3.64 4.04 a μ

796 Quantified at 33.35 M.

wileyonlinelibrary.com/journal/rcm Copyright © 2017 John Wiley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2017, 31, 791–803 Quantification and isotopic analysis of dissolved sulfur species yielded linear calibration curves over this range. Precision and Further validation of method accuracy is complicated by accuracy were similar for the two detectors, typically a few the fact that the standard analytical technique, IRMS, is not percent relative to sample concentration. In summary, both designed to measure the relatively large isotopic enrichments detectors were shown to be capable of quantitation of being targeted by our LC/MS method. Certified international dissolved sulfur species. Quantitation of sulfur species by reference materials for calibration span a range of only a few TOF-MS is more expensive and complex than the FLR tens of permil,[36] whereas an artificial enrichment in 34Sof detection but has the advantage of enabling positive only 1% (absolute abundance) constitutes a shift in δ34Sof identification of unknown sulfur species, while the nearly 250‰. These issues notwithstanding, we measured nondestructive FLR detector is ideally suited for preparative the same series of sulfide isotopic standards (see fraction collection. Experimental section) using both UPLC/TOF-MS and EA/IRMS (Fig. 3). Seven samples ranging from 4.27 to 17.4 atom percent 34S Measurement of sulfur isotopic enrichment were measured. Data from both instruments were highly Several different strategies were tested for quantifying reproducible and linear over the entire range of isotopic isotope abundance based on the relative intensity of isotopic compositions. However, both yielded apparent 34S peaks of the molecular ion (Figs. 2(B)–2(E)). These included abundances significantly lower than those calculated for calculation of M + 2/M ion abundance ratios, Σ(M + 2/ the standards based on mixing volumes. The measured M+…M + 7/M) ion abundance ratios, and fitting the values averaged 98.8% and 90.4% of the theoretical values complete envelope of isotopic peaks with a forward model. for the TOF-MS and IRMS data, respectively (Fig. 3(A)). In the end, the calculation of a weighted-mean mass shift Comparison of the two datasets with each other is also (as described in the Experimental section) was both the most highly linear (Fig. 3(B)). accurate and simplest approach. A minor drawback is that Both datasets (i.e., from TOF-MS and IRMS) were collected this analysis cannot distinguish between simultaneous by comparing unknown, 34S-enriched samples with a single enrichments of 33S and 34S (i.e., from separately labeled isotopic standard with a natural abundance of 34S. substrates), for which deconvolution of the full isotopologue Presumably, the deviations of measured data from correct envelope would be required. A spectrometer with sufficient isotope ratios therefore reflect instrumental fractionations that mass resolution to distinguish 18O from 34S isotopologues, do not scale uniformly with isotopic composition. Such effects such as the TOF analyzer employed here, is essential to this are well known in measurements of natural-abundance application. isotopes where they are often referred to as "scale An estimate of precision in isotopic measurements is compression/expansion". For this reason, two-point provided by the zero-enrichment experiment. The same calibration of isotopic compositions is almost universally – (natural isotopic abundance) standard solution used for recommended.[36 38] However, in our case no 34S-enriched calibration of the TOF-MS data was measured repeatedly as reference materials exist, so we are unable to explicitly test a sample, yielding enrichment factors for sulfide and accuracy across the range of isotopic compositions to a level thiosulfate of 1.000 ± 0.006, equivalent to 6‰ uncertainty in that approaches measurement precision. Nevertheless, the the δ34S nomenclature with no systematic error. Relative linearity and reproducibility of the responses of both precision is better at higher levels of isotopic enrichment. instruments in Fig. 3 strongly suggest that – with calibration

– Figure 3. Cross-calibration of the same samples of HS with increasing atomic percentage of 34S derivatized with bromobimane and precipitated with zinc acetate run using UPLC/TOF-MS and EA/IRMS, respectively: (A) Atomic percentage of 34S – in HS plotted against enrichment factor (red circles) (EnrF = –0.1077 + 0.2613*%34S, 2 δ34 δ34 R = 0.999), as determined by the UPLC/TOF-MS methodology, and SVCDT ( S= 1027.3 + 244.0*%34S, R2 = 0.999), as measured by EA/IRMS (blue squares). (B) A systematic and reproducible linear cross-calibration (gold circles) was established δ34 linking our enrichment factors (EnrF) with conventional IRMS SVCDT measurements of sulfur (EnrF =0.9935 + 0.001071*δ34S, R2 = 0.999). The solid black line represents the theoretical relationship between atom percentage of 34S, enrichment factors, and δ34 797 SVCDT. [Color figure can be viewed at wileyonlinelibrary.com]

34 – 2– standards having suitable S-enriched compositions depth (Fig. 4(B)). We did not observe an increase in SO3 2– – 2– accuracy for isotopic abundance should approach or S2O3 concentrations at the HS and SO4 transition as 2– 2– measurement precision, i.e., 0.6% relative. might be expected. The concentrations of SO3 and S2O3 Specific reasons for the deviation of TOF-MS data at high 34S were reproducible and fall within the range of previously enrichments are not currently known. Fractionations during reported environmental marine sediments (see Zopfi ionization, transmission, and detection are all possible. Ion et al.[39] for review). shading, in which the earlier-arriving 32S ions induce a slight electric field at the detector that repels the later-arriving 34S Application 2: Tracing 34S-labeled sulfate in a ions, is a possibility specific to the TOF that would clearly scale sulfate-reducing bacterial culture nonlinearly with isotope ratio. The second experimental demonstration of this method was to 34 34 – – trackthe conversionof S-labeledsulfateinto S-sulfide,using Application 1: Quantitation of pore-water SO 2 and S O 2 3 2 3 batch cultures of the sulfate-reducing bacterium (SRB) in methane seep sediments Desulfococcus multivorans. Cultures were amended with ’ μ 34 As an initial test of the method s ability to quantify trace sulfur 280 M labeled sulfate ( FSO4 = 12.4%) and unlabeled 34 species in complex natural samples, UPLC separation with thiosulfate (6.44 ± 0.15 mM; FS2O3 = 4.2%). Here we FLR and TOF-MS detection were used to compare demonstrate the utility of the UPLC/TOF-MS analytical 2– environmental pore-water concentrations of SO3 and methodology for tracking the production of isotopically 2– fi S2O3 in a depth pro le of sediments from the Santa Monica enriched sulfur species (Fig. 5(A)), and through the addition Basin. The sediments were collected within an active of 34S-enriched sulfate, we examine the potential for sulfur methane seep, where microbial sulfur metabolism associated disproportionation and preferred sulfur substrate usage by D. with anaerobic oxidation of methane and chemosynthetic multivorans. Previous reports suggested the potential for sulfide oxidation is enhanced. Both the FLR and TOF thiosulfate reduction in addition to the canonical sulfate 2– 2– fi detectors were used to quantitate SO3 and S2O3 ;we reduction by D. multivorans, but the speci c preference of sulfur 2– 2– [40] present the FLR data for SO3 and the TOF S2O3 substrate by this organism has not been directly tested. concentrations highlighting the abilities of both instruments As the D. multivorans cultures reached the exponential – for accurate quantitation (Fig. 4(A)). The FLR and TOF phase of growth, HS accumulated to concentrations of 2– 2– detectors produced S2O3 and SO3 values that were 1.66 ± 0.49 mM. Abiotic control bottles for this experiment 2– 2– fi – 2– inconsistent with previous reports. SO3 and S2O3 reached had no quanti able HS or SO3 and an average thiosulfate maximum concentrations of 12.5 and 42.5 μM at 1.5 and concentration of 6.44 ± 0.15 mM, indistinguishable from the 2– fi 3.5 cm depth, respectively. SO3 concentrations were below amount added. The triplicate experimental bottles had nal 2– fi detection limits by 4.5 cm depth. S2O3 concentrations thiosulfate concentrations of 5.42 ± 1.35 mM. Signi cantly – μ 2– varied across much of the 18.5 cm core. Conventional more HS was produced than the 280 MSO4 provided, – 2– 2– 2– analysis of HS by the Cline assay and SO4 by ion which implies a contribution of both SO4 and S2O3 – – chromatography show typical profiles of HS building with reduction to the HS produced. The mass balance of sulfur 2– depth to a maximum of 12.5 mM by 10.5 cm and SO4 was conserved, within error; however, we cannot rule out – decreasing from 25 mM at the surface down to 2 mM at the possibility of a minor amount of non-enriched HS

Figure 4. Pore-water profile from methane seep sediments underlying sulfide- oxidizing microbial mat (push core PC55) recovered from the Santa Monica Basin. PC55 was sectioned at 1 cm intervals for the upper 6 cm, then at 3 cm intervals down to 18.5 cm: (A) Concentrations, with depth, of bromobimane- preserved intermediate sulfur species (sulfite and thiosulfate) measured by FLR and TOF detectors, respectively. (B) Colorimetric analysis (Cline assay) of sulfide concentrations and IC measurements of sulfate and dissolved inorganic carbon (DIC) measured from aliquots of the same pore-water samples. [Color 798 figure can be viewed at wileyonlinelibrary.com]

Figure 5. (A) Representative negative ionization chromatogram from triplicate cultures of Desulfococcus multivorans, a sulfate-reducing bacterium, incubated with 34S-labeled sulfate. Peaks shown in the chromatogram are labeled with the compound, elution time, and most abundant ion, and include 1: HEPES 0.31 min 237.095 Da, 2: sulfite bimane 0.52 min 271.042 Da, 3: thiosulfate bimane 0.86 min 303.015 Da, 6: bromine cluster 2.89 min 375.607 Da, 7: sulfide dibimane 3.11 min 413.130 Da. (B) Positive 34 2– ionization chromatogram from a methane seep sediment incubation with SO4 ;as this is a complex natural sample, more compounds were detected relative to pure cultures or standards. Peaks are 1: HEPES 0.31 min 239.111 Da, 3: thiosulfate bimane 0.89 min 305.030 Da, 4: bimane 1.51 min 193.102 Da, 5: 34S sulfide bimane 2.67 min 227.063 Da, 6: bromobimane 2.89 min 271.012 Da, 7: 34S sulfide dibimane 3.11 min 415.146 Da, 8: disulfide dibimane 3.46 min 469.098 Da. (C) Mass spectrum of the positive ion of sulfide dibimane from methane seep sediment incubations (peak 7 in (B)). The peak heights for 32S (415.1455 Da) and 34S (417.1416 Da) were approximately 34 2– 34 – equal due to the extensive conversion of SO4 into sulfide ( SH ) presumably associated with sulfate-coupled anaerobic oxidation of methane. [Color figure can be viewed at wileyonlinelibrary.com]

2– 2– transferred with the initial inoculum. Assuming that all of the that SO4 is being incompletely reduced to S2O3 during 34 2– – – SO4 was converted into HS , more than 1.3 mM of HS sulfate reduction, or possibly that thiosulfate is being slowly 2– μ – had to come from unlabeled S2O3 . There were also M formed from HS . 2– levels of SO3 detected in the experimental bottles, which might have been produced via a number of different Application 3: Tracing 34S-labeled sulfate in methane seep pathways. Notably, there was no detectable enrichment of sediment incubations 34 2– SinSO3 indicating it did not derive from the reduction 34 2– 34 of the SO4 substrate. At exponential phase, FHS was To test the utility of the sulfur SIP method in environmental 34 5.4 ± 0.4% and FS2O3 was 4.5 ± 0.5%. Approximately one- samples, we conducted 40-day incubations with labeled 34 – 34 third of the S-enriched HS was created by D. multivorans sulfate (28 mM, FSO4 = 1, i.e., 100% S label) added to methane reducing labeled sulfate, but the mechanism resulting in seep sediments. UPLC chromatograms of the mBBr- 2– enrichment of the S2O3 pool remains unclear. The minor derivatized supernatant contained many more peaks relative 799 incorporation of the 34S label into thiosulfate could suggest to the experiments with pure cultures of D. multivorans,

2– 2– including several peaks that were clearly not SO3 ,S2O3 , oxidation of methane (AOM). In addition to large quantities – – – CH3S ,orHS. Each peak in the environmental samples was of HS , the sediment incubations also contained 1.65 mM fi 2– 34 subsequently identi ed through comparisons with standards S2O3 with no detectable S enrichment (FS2O3 = 0.042). fi fi – or de novo determination from the mass spectral data (Fig. 5(B)). Disul de dibimane, corresponding to the polysul de HS2 , – – Two separate bimane derivatives of HS were observed: was also detected in the incubations. The zero valent (HS2 ) sulfide monobimane and sulfide dibimane. We infer that the has been hypothesized as the chemical intermediate in the amount of mBBr added to these samples (at the time of AOM microbial syntrophy between a sulfate-reducing collection) was insufficient to fully react with bisulfide due methanotrophic ANME archaea and a disulfide- to other reactive sulfur species. For natural samples precise disproportionating deltaproteobacteria.[12] Milucka et al.[12] concentrations of total reactive sulfur species are likely to be used the methyl triflate technique to measure polysulfides unknown; however, if an estimation can be provided, then a from 3-x S chain length; however, this method was incapable two-fold stoichiometric excess of mBBr can still be supplied. of directly measuring the primary proposed intermediate, – This caused considerable problems for quantitation because HS2 . Here we show that the TOF-MS method for measuring fi – generation of a sul de monobimane calibration curve is not bromobimane derivatives of sulfur is capable of HS2 straightforward. Nevertheless, based on sulfide dibimane we detection; however, to develop this into a quantitative assay, fi – observed the build-up of 4.35 mM sul de with FSO4 = 0.498 further work and reliable standards for HS2 are needed as (i.e., 50% 34S label) (Fig. 5(C)). Sulfate reduction was clearly we are currently unable to reliably create and derivatize prevalent in these sediments, likely coupled to the anaerobic individual chain-length polysulfides with bimane.

Figure 6. (A) Negative ionization chromatogram from slow growth and oxidant limited continuous culture of Sulfurovum lithotrophicum, a sulfur-oxidizing epsilonproteobacterium. The three resolvable peaks are 1: thiosulfate bimane 1.38 min – 303.0114 Da, 2: disulfanemonosulfonate ( S-S-SO H) bimane 2.06 min 334.9837 Da, 3: – 3 trisulfanemonosulfonate ( S-S-SO H) bimane 2.94 min 366.9554 Da. (B) Isotope model 3 – of the negative ion of trisulfanemonosulfonate ( S-S-SO3H) bimane with theoretical mass and isotopic distribution. (C) Mass spectrum of the negative ion of – trisulfanemonosulfonate ( S-S-SO H) bimane from the oxidant-limited chemostat

800 3 experiment (peak 3 in (A)). [Color figure can be viewed at wileyonlinelibrary.com]

Application 4: Identification of novel sulfur species in a CONCLUSIONS continuous culture of the sulfur-oxidizing bacterium Sulfurovum lithotrophicum We present a method for preservation, separation, and quantitative analysis of dissolved sulfur-containing analytes. Coupling bimane derivatization to mass spectral analysis It is based on derivatization with monobromobimane (mBBr), should enable the identification of potentially novel and which is rapid, quantitative, and easily adapted to field important intermediate redox species of sulfur. In this conditions. Qualitative evidence suggests that mBBr- application, we studied the sulfur-oxidizing bacterium derivatized samples are stable relative to unreacted samples, Sulfurovum lithotrophicum in oxidant-limited continuous with minimal oxidation over the course of months to years. culture, with thiosulfate as the supplied sulfur source. We Baseline separation of four key analytes, sulfite, thiosulfate, hypothesized that partial oxidation of thiosulfate could result methanethiol, and bisulfide, was achieved by UPLC in just in the accumulation of unusual and potentially important 5 min, making this a high-throughput analytical approach. sulfur species of intermediate redox state. Chromatograms Detection by either fluorescence or time-of-flight mass of mBBr-derivatized samples from these cultures revealed spectrometry provided highly sensitive, precise, and accurate two unidentified peaks (Fig. 6(A)) in addition to thiosulfate. quantitation. Using TOF-MS, we demonstrate the ability to Using de novo analysis of the mass spectral data, these two quantify 34S isotopic enrichments with ~0.6% relative unknown bimane-derivatized sulfur compounds were precision, over a dynamic range from natural abundance fi – tentatively identi ed as disulfanemonosulfonate ( S-S-SO3H) (4.2%) to at least 50%. In the proof-of-principle applications bimane (mass 334.9837 Da) and trisulfanemonosulfonate described here, we (1) demonstrated thiosulfate utilization – ( S-S-S-SO3H) bimane (mass 366.9553 Da) (Figs. 6(A), 6(B) during sulfate reduction by pure cultures of the sulfate and 6(C)). These compounds appeared to be produced reducing bacterium D. multivorans; (2) detected zero valent 2– fi – fi during S2O3 oxidation by S. lithotrophicum and were not disul de (HS2 ) in methane seep sediments, a polysul de observed in the standard solutions or in any of the other species that has been historically challenging to measure with samples we analyzed. Quantitation and identification, via other methods; and (3) resolved the identity of two unknown liquid chromatography and FLR detection, of these sulfur metabolites as disulfanemonosulfonate and compounds was not possible as authentic standards are trisulfonemonosulfonate produced by S. lithotrophicum during not available to construct calibration curves and retention thiosulfate oxidation under oxidant limited conditions. times. The methodology demonstrated here has numerous The chemistry of sulfanemonosulfonic acids has been potential applications. The ability to quantify reactive, low- previously described and the formation of these acids is abundance species in complex natural samples will be proposed to be a result of the nucleophilic degradation of extremely useful in studies of biogeochemical sulfur cycling, sulfur chains.[41] The proposed reaction mechanism especially in environments supporting rapid sulfur cycling 2– 2– [5,6] involves either SO3 or S2O3 interacting with elemental such as oxygen minimum zones, below the sulfate sulfur or polysulfide chains. Through microscopic reduction zone in marine sediments,[7,8] sulfur-metabolizing examination of S. lithotrophicum cells, structures consistent microbial symbioses,[9,42] and freshwater sediments.[10,11] with intracellular sulfur globules were visible (data not Moreover, coupling UPLC with nondestructive FLR shown).[42,43] Sulfanemonosulfonic acids were previously detection to preparative fraction collection provides an detected in the cell pellet of Allochromatium vinosum,a analytical route to high-precision IRMS-based isotopic purple photosynthetic sulfur oxidizer, but these sulfur analysis of these compounds at natural isotopic abundance. compounds were not detected in the supernatant.[28] Due The ability to detect a large dynamic range of isotopic to the analytical limitations of XANES spectroscopy, that enrichments is also extremely useful for stable-isotope study was unable to determine the chain length of these probing (SIP)-type experiments, either in environmental molecules. The experiments with S. lithotrophicum here sample incubations or in culture. This is likely to have demonstrate the added advantage of using TOF-MS, with applications in microbial ecology, biochemistry, and even its high mass resolving capabilities, for untargeted biomedicine.[45] identification and resolution of unknown sulfur compounds. The detection of sulfanemonosulfonic acids in two distantly Acknowledgements related sulfur-oxidizing bacteria points to the potential importance of these compounds. One possibility is that We thank chief scientist Peter Brewer (MBARI) and the intracellular sulfanemonosulfonic acids may represent a shipboard, scientific crew and pilots of the R/V Western Flyer pathway by which microorganisms first begin to utilize their and the ROV Doc Ricketts for allowing us to participate in the stored reserves of sulfur globules (elemental expedition. We also thank S. Scheller and H. Yu for sulfur/polysulfides). Indeed, previous experiments with A. providing and maintaining the methane seep sediment vinosum revealed a correlation between intracellular incubations, F. Wu and S. Connon for technical assistance, sulfur and the relative concentration/presence of and C. Neubaur, M. R. Raven and M. S. Sim for helpful sulfanemonosulfonic acids.[28] The reaction mechanism discussions. The work was funded by the Gordon and Betty originally proposed by Meyer et al.[41] would be consistent Moore Foundation through Grant No. GBMF3306 to VJO with the ability of dissimilatory sulfite reductase (Dsr) and ALS, and the NASA Astrobiology Institute Life 2– subunits to produce SO3 which could then interact with Underground (Grant Award # NNA13AA92A to VJO). This the intracellular elemental sulfur/polysulfides creating is NAI-Life Underground Publication Number 111. DAS was 801 potentially important intracellular sulfanesulfonic acids. supported as an Agouron Institute Geobiology Fellow.

REFERENCES [19] Y. Zakon, L. Halicz, F. Gelman. Isotope analysis of sulfur, bromine, and chlorine in individual anionic species by ion chromatography/multicollector-ICPMS. Anal. Chem. 2014, [1] J. Zopfi, M. E. Bottcher, B. B. Jorgensen. Biogeochemistry of 86, 6495. sulfur and iron in Thioploca-colonized surface sediments in [20] A. Kamyshny Jr, A. Goifman, J. Gun, D. Rizkov, O. Lev. the upwelling area off central Chile. Geochim. Cosmochim. Equilibrium distribution of polysulfide ions in aqueous Acta 2008, 72, 827. solutions at 25°C: A new approach for the study of [2] M. A. Huerta-Diaz, A. Tessier, R. Carignan. Geochemistry of polysulfides’ equilibria. Environ. Sci. Technol. 2004, 38, 6633. trace metals associated with reduced sulfur in freshwater [21] G. W. Luther III, B. T. Glazer, L. Hohmann, J. I. Popp, sediments. Appl. Geochem. 1998, 13, 213. M. Taillefert, T. F. Rozan, P. J. Brendel, S. M. Theberge, [3] J. W. Morse, G. W. Luther III. Chemical influences on trace D. B. Nuzzio. Sulfur speciation monitored in situ with solid metal-sulfide interactions in anoxic sediments. Geochim. state gold amalgam voltammetric microelectrodes: Cosmochim. Acta 1999, 63, 3373. polysulfides as a special case in sediments, microbial mats [4] A. Schippers, in Sulfur Biogeochemistry – Past and Present, and hydrothermal vent waters. J. Environ. Monit. 2001, 3, 61. (Eds: J. P. Amend, K. J. Edwards, T. W. Lyons). Geological [22] E. S. Boyd, G. K. Druschel. Involvement of intermediate Society of America, Colorado, 2004, pp. 49–62. sulfur species in biological reduction of elemental sulfur [5] D. E. Canfield, F. J. Stewart, B. Thamdrup, L. De Bradandere, under acidic, hydrothermal conditions. Appl. Environ. T. Dalsgaard, E. F. Delong, N. P. Revsbech, O. Ulloa. A Microbiol. 2013, 79, 2061. cryptic sulfur cycle in oxygen-minimum-zone waters off [23] G. L. Newton, R. Dorian, R. C. Fahey. Analysis of biological the Chilean coast. Science 2010, 330, 1375. thiols: Derivatization with monobromobimane and [6] D. T. Johnston, B. C. Gill, A. Masterson, E. Beirne, separation by reverse-phase high-performance liquid K. L. Casciotti, A. N. Knapp, W. Berelson. Placing an upper chromatography. Anal. Biochem. 1981, 114, 383. limit on cryptic marine sulphur cycling. Nature 2014, [24] R. D. Vetter, P. A. Matrai, B. Javor, J. O’Brien, in Biogenic 513, 530. Sulfur in the Environment, (Eds: E. S. Saltzman, W. J. Cooper). [7] L. Holmkvist, T. G. Ferdelman, B. B. Jorgensen. A cryptic American Chemical Society, Washington, DC, 1989, sulfur cycle driven by iron in the methane zone of marine pp. 243–261. sediment (Aarhus Bay, Denmark). Geochim. Cosmochim. Acta [25] J. Rethmeier, A. Rabenstein, M. Langer, U. Fischer. Detection 2011, 75, 3581. of traces of oxidized and reduced sulfur compounds in small [8] M. Aoki, R. Kakiuchi, T. Yamaguchi, K. Takai, F. Inagaka, samples by combination of different high-performance H. Imachi. Phylogenetic diversity of aprA genes in liquid chromatography methods. J. Chromatogr. A 1997, subseafloor sediments on the Northwestern Pacific Margin 760, 295. off Japan. Microbes Environ. 2015, 30, 276. [26] C. Arndt, F. Gaill, H. Felbeck. Anaerobic sulfur metabolism [9] E. G. Wilbanks, U. Jaekel, V. Salman, P. T. Humphrey, in thiotrophic symbioses. J. Exp. Biol. 2001, 204, 741. J. A. Eisen, M. T. Facciotti, D. H. Buckley, S. H. Zinder, [27] R. C. Fahey, G. L. Newton. Determination of low-molecular- G. K. Druschel, D. A. Fike, V. J. Orphan. Microscale sulfur weight thiols using monobromobimane fluorescent labeling cycling in the phototrophic pink berry consortia of the and high-performance liquid chromatography. Methods Sippewissett Salt Marsh. Environ. Microbiol. 2014, 16, 3398. Enzymol. 1987, 143, 85. [10] C. M. Hansel, C. J. Lentini, Y. Tang, D. T. Johnston, [28] B. Franz, T. Gehrke, H. Lichtenberg, J. Hormes, C. Dahl, S. D. Wankel, P. M. Jardine. Dominance of sulfur-fueled iron A. Prange. Unexpected extracellular and intracellular sulfur oxide reduction in low-sulfate freshwater sediments. The species during growth of Allochromatium vinosum with ISME J. 2015, 1. reduced sulfur compounds. Microbiology 2009, 155, 2766. [11] K. á. Norđi, B. Thamdrup, C. J. Schubert. Anaerobic [29] E. A. Wintner, T. L. Deckwerth, W. Langston, A. Bengtsson, oxidation of methane in an iron-rich Danish freshwater lake D. Leviten, P. Hill, M. A. Insko, R. Dumpit, E. VandenEkart, sediment. Limnol. Oceanogr. 2013, 58, 546. C. F. Toombs, C. Czabo. A monobromobimane-based assay [12] J. Milucka, T. G. Ferdelman, L. Polerecky, D. Franzke, to measure pharmacokinetic profile of reactive sulphide G. Wegener, M. Schmid, I. Lieberwirth, M. Wagner, species in blood. Br. J. Pharmacol. 2010, 160, 941. F. Widdel, M. M. M. Kuypers. Zero-valent sulphur is a key [30] Y. Liu, M. Sieprawska-Lupa, W. B. Whitman, R. H. White. intermediate in marine methane oxidation. Nature 2012, Cysteine is not the sulfur source for iron-sulfur cluster and 491, 541. methionine biosynthesis in the methanogenic Archaeon [13] M. R. Raven, A. L. Sessions, W. W. Fischer, J. F. Adkins. Methanococcus maripaludis. J. Biol. Chem. 2010, 285, 31923. Sedimentary pyrite δ34S differs from porewater sulfide in [31] J. Rodriguez, J. Hiras, T. E. Hanson. Sulfite oxidation in Santa Barbara Basin: Proposed role of organic sulfur. Chlorobaculum tepidum. Frontiers Microbiol. 2011, 2,1. Geochim. Cosmochim. Acta 2016, 186, 120. [32] E. Melamud, L. Vastag, J. D. Rabinowitz. Metabolomic [14] M. R. Raven, A. L. Sessions, J. F. Adkins, R. C. Thunell. Rapid analysis and visualization engine for LC-MS data. Anal. organic matter sulfurization in sinking particles from the Chem. 2010, 82, 9818. Cariaco Basin water column. Geochim. Cosmochim. Acta [33] M. F. Clasquin, E. Melamud, J. D. Rabinowitz. LC-MS data 2016, 190, 175. processing with MAVEN: A metabolomics analysis and [15] J. D. Cline. Spectrophotometric determination of hydrogen visualization engine. Curr. Protocols Bioinformatics 2012, sulfide in natural waters. Limnol. Oceanogr. 1969, 14, 454. https://doi.org/10.1002/0471250953.bi1411s37. [16] B. B. Jørgensen. A thiosulfate shunt in the sulfur cycle of [34] F. Inagaki, K. Takai, K. H. Nealson, K. Horikoshi. Sulfurovum marine sediments. Science 1990, 249, 152. lithotrophicum gen. nov., sp. nov., a novel sulfur-oxidizing [17] M. Yasuyuki, K. Fukasawa, T. Koh. Determination of sulfur chemolithoautotroph within the ε-Proteobacteria isolated anions at the ppb level by ion chromatography utilizing their from Okinawa Trough hydrothermal sediments. Int. J. catalytic effects on the postcolumn reaction of iodine with Systemat. Evolut. Microbiol. 2004, 54, 1477. azide. J. Chromatogr. A 1998, 804, 143. [35] V. J. Orphan, C. H. House, K. Hinrichs, K. D. McKeegan, [18] G. Paris, A. L. Sessions, A. V. Subhas, J. F. Adkins. MC-ICP- E. F. DeLong. Methane-consuming Archaea revealed by MS measurement of δ34S and Δ33S in small amounts of directly coupled isotopic and phylogenetic analysis. Science 802 dissolved sulfate. Chem. Geol. 2013, 45, 50. 2001, 293, 484.

[36] W. A. Brand, T. B. Coplen, J. Vogl, M. Rosner, T. Prohaska. [41] B. Meyer, L. Peter, K. Spitzer. Trends in the charge Assessment of international reference materials for isotope- distribution of sulfanes, sulfanesulfonic acids, ratio analysis (IUPAC Technical Report). Pure Appl. Chem. sulfanedisulfonic acids, and sulfurous acid. Inorg. Chem. 2014, 86, 425. 1977, 16, 27. [37] R. A. Werner, W. A. Brand. Referencing strategies and [42] A. S. Engel, M. L. Porter, L. A. Stern, S. Quinlan, P. C. Bennett. techniques in stable isotope ratio analysis. Rapid Commun. Bacterial diversity and ecosystem function of filamentous Mass Spectrom. 2001, 15, 501. microbial mats from aphotic (cave) sulfidic [38] T. B. Coplen, W. A. Brand, M. Gehre, M. Gröning, springs dominated by chemolithoautotrophic H. A. J. Meijer, B. Toman, R. M. Verkouteren. New Guidelines "EpsilonproteobacteriaI". FEMS Microbiol. Ecol. 2004, 51, 31. for δ13C Measurements. Anal. Chem. 2006, 78, 2439. [43] T. L. Hamilton, D. S. Jones, I. Schaperdoth, J. L. Macalady. [39] J. Zopfi, T. G. Ferdelman, H. Fossing, in Sulfur Metagenomic insights into S(0) precipitation in a terrestrial Biogeochemistry – Past and Present, (Eds: J. P. Amend, K. J. subsurface lithoautotrophic ecosystem. Frontiers Microbiol. Edwards, T. W. Lyons). Geological Society of America, 2015, 5,1. Colorado, 2004, pp. 97–116. [44] K. S. Dawson, S. Scheller, J. G. Dillon, V. J. Orphan. Distinct [40] K. S. Habicht, D. E. Canfied, J. Rethmeier. Sulfur isotope microbial ecophysiologies and sulfur-cycling in the fractionation during bacterial reduction and environment. Frontiers Microbiol. 2016, 7,1. disproportionation of thiosulfate and sulfite. Geochim. [45] J. W. Taraska. Mapping membrane protein structure with Cosmochim. Acta 1998, 62, 2585. fluorescence. Curr. Opin. Struct. Biol. 2012, 22, 507. 803