Enhanced Resolution of Chiral Amino Acids with Capillary ... - Doi.org

Dec 5, 2016 - Journal of Separation Science 2017 18,. Thermal probe design for Europa sample acquisition. Mera Horne. Acta Astronautica 2017 , ...
0 downloads 0 Views 677KB Size
Subscriber access provided by UNIVERSITY OF WISCONSIN MILWAUKEE

Article

Enhanced resolution of chiral amino acids with capillary electrophoresis for biosignature detection in extraterrestrial samples Jessica S. Creamer, Maria F Mora, and Peter Athol Willis Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b04338 • Publication Date (Web): 05 Dec 2016 Downloaded from http://pubs.acs.org on December 5, 2016

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 10

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

Enhanced resolution of chiral amino acids with capillary electrophoresis for biosignature detection in extraterrestrial samples Jessica S. Creamer, Maria F. Mora, and Peter A. Willis* Jet Propulsion Laboratory, California Institute of Technology 4800 Oak Grove Dr. Pasadena, CA 91109 ABSTRACT: Amino acids are fundamental building blocks of terrestrial life as well as ubiquitous byproducts of abiotic reactions. In order to distinguish between amino acids formed by abiotic versus biotic processes it is possible to use chemical distributions to identify patterns unique to life. This article describes two capillary electrophoresis methods capable of resolving 17 amino acids found in high abundance in both biotic and abiotic samples (seven enantiomer pairs D/L-Ala, -Asp, -Glu, -His, -Leu, -Ser, -Val and the three achiral amino acids Gly, β-Ala, and GABA). To resolve the 13 neutral amino acids one method utilizes a background electrolyte containing γ-cyclodextrin and sodium taurocholate micelles. The acidic amino acid enantiomers were resolved with γ-cyclodextrin alone. These methods allow detection limits down to 5 nM for the neutral amino acids and 500 nM for acidic amino acids and were used to analyze samples collected from Mono Lake with minimal sample preparation.

Despite the many differing theories on how life originated on Earth, the one thing all the theories agree on is the fact that life would not have been possible without the presence of liquid water and an abundance of simple organic molecules. 1 As life evolved, the chemical inventory of organic molecules was refined to reflect the needs of biotic reactions. For example, of the hundreds of possible amino acid configurations seen abiotically, terrestrial biology uses an “alphabet” of only twenty. 2 Furthermore, while abiotic reactions generate a racemic mixture of amino acids, homochirality is necessary for proper protein folding and for this reason life on Earth uses exclusively lefthanded amino acid enantiomers. By surveying the abundances of amino acids present in a sample, these distributions (both type and chirality) could serve as biosignatures to differentiate between amino acids produced by biology from those formed by abiotic reactions. In the search for extraterrestrial life it is reasonable to expect that similar chemical patterns could emerge beyond Earth. However, the in situ analysis of amino acids in extraterrestrial environments remains a challenge. Because of the low expected abundances of organics in planetary samples (amino acid content in terrestrial soil and oceans can be part-per-billion or lower) in situ sampling techniques are preferable over optical ones because they provide increased sensitivity.3 Historically, the in situ analytical technique of choice has been GC coupled to MS. Having a mass analyzer has proven invaluable in being able to determine the distribution of volatile species on Mars,4,5 Titan,6 and Enceladus.7 However, when it comes to polar organic molecules, GC-MS methodology requires additional sample preparation to get these species into the gas phase. Most recently, the SAM instrument aboard the Curiosity rover included the first space-flight GC-MS with methodology developed specifically to detect amino acids and carboxylic acids and determine the presence of organic matter on the surface of Mars. SAM uses a ‘‘one-pot’’ extraction and derivatization protocol with N-methyl-N-(tert-

butyldimethylsilyl) trifluoroacetamide (MTBSTFA) and DMF. In the lab, the sensitivity of this technique was reported to be around 1 part-per-million for alanine.8 However, the hydrated minerals and oxides present in Martian samples react rapidly with MTBSTFA, making in situ derivatization and subsequent detection of amino acids and carboxylic acids by MS problematic.8 Additionally, most derivatization reagents react rapidly with water which would mean that future missions to the icy bodies Europa and Enceladus would require sample processing steps to remove water and salts prior to analysis by GC-MS, increasing the complexity of the instrument. Alternatively, capillary electrophoresis (CE) is an extremely promising analytical technique for analysis of polar organics in environments where water and salts9,10 are present. Since CE is a liquid-based technique, all sample preparation can be done without leaving the aqueous phase. CE can be coupled to a wide variety of detection methods including laser induced fluorescence detection (LIF) to achieve sensitive detection. Moreover, CE-LIF has been used for many decades for chiral amino acid analysis.11-14 To improve portability, CE can be miniaturized using a microfluidic platform known as microchip electrophoresis (ME). Because both CE and ME operate under the same separation principles it is possible to transfer methods developed on CE to ME.15 As a portable and sensitive technique for in situ analysis, MELIF has been of interest to the astrobiology community for many years.3,16-18 There have been several attempts towards the development of instrumentation to perform automated end-toend ME-LIF analysis.19,20 Most recently, our lab demonstrated for the first time an automated end-to-end sample processing, separation, and detection of amino acids with a single microfluidic device.21 Ongoing work in our group is focused on the development of the Chemical Laptop, and all-in-one battery-powered ME-LIF instrument for automated in situ analysis.3

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

Yet, for the analysis of chiral amino acids, the methods developed for these instruments have been limited in their capacity to resolve multiple pairs of astrobiologically relevant enantiomers simultaneously. To date, most of the work in this area has been focused towards the optimization of the Mars7 standard, a mixture of 12 amino acids (Gly, 2-aminoisobutyric acid (AIB) and D/L-Ser, -Val, -Ala, -Glu, and -Asp) chosen for their common occurrence in meteorites.22 The methods that exist for the separation of the Mars7 standard have yet to resolve all the 12 species simultaneously. Both traditional ME22,23 as well as counterflow gradient electrofocusing methods such as temperature gradient focusing24 and gradient elution isotachophoresis25 have been attempted. For each of these methods either one or both of the Ala and Ser enantiomers remain unresolved. To work around the resolution limitation, Hutt et al. were able to quantify Ala and Ser by analyzing the sample before and after a treatment with periodate to selectively remove Ser.23 Looking beyond the Mars7, Chiesl et al. improved the capacity of the analysis to resolve 25 astrobiologically relevant amino acids by employing two separation conditions in sequence, one with capillary zone electrophoresis (CZE) and the other with micellar electrokinetic chromatography (MEKC).26 The MEKC method was able to resolve Ala and Ser from each other, yet only partial chiral resolution was achieved for their enantiomer pairs (D/L resolution 0.73 for Ala and 0.58 for Ser). Additionally, Ala and Ser were the only two proteinogenic enantiomer pairs targeted by this method for chiral analysis (D/L-citrulline was also investigated). Here, we present two CE methods capable of resolving 17 amino acids labeled with 5-carboxyfluorescein succinimidyl ester (CFSE). These 17 amino acids (seven enantiomer pairs D/L -Ala, -Asp, -Glu, -His, -Leu, -Ser, -Val and the achiral Gly, βAla, and GABA) represent amino acids found in high abundance in biotic and abiotic samples and form what we are calling the Signature17 standard. Resolution of the neutral amino acids was achieved using cyclodextrin (CD) mediated MEKC, with dual chiral selectors sodium taurocholate (STC) and γ-CD. The acidic amino acids were resolved by CZE using γ-CD alone. These methods were then validated for their ability to label and detect amino acids in high salinity samples from Mono Lake, CA.

EXPERIMENTAL SECTION Chemicals and reagents. L- and D-amino acids, sodium tetraborate, STC, β-CD, γ-CD, acetonitrile (ACN), isopropyl alcohol (IPA), DMF, HPLC grade DI water, methanol (MeOH), hydrochloric acid (HCl), and sodium hydroxide (NaOH) were obtained as analytical grade reagents from Sigma-Aldrich (St. Louis, MO). CFSE was purchased from Thermo Fisher Scientific (Waltham, MA). Instrumentation. Separations were performed on a Beckman Coulter P/ACE MDQ (Brea, CA) with LIF detection (488 nm). The instrument was controlled using 32-Karat software. Fused silica capillaries (50 µm ID x 360 µm OD) were obtained from Polymicro Technologies (Phoenix, AZ) and cut to the desired length (30-60 cm). A small window was burned into the polyimide coating 10 cm from the capillary outlet for detection. Before the first separation the capillary was conditioned by rinsing for 5 min each with 0.5 N HCl, water, MeOH, water, 0.1 M NaOH, and water, followed by a 10 min rinse with the background electrolyte (BGE). Equilibration of the buffer within the

Page 2 of 10

capillary was expedited by applying a 0.5 kV/cm potential across the capillary following the conditioning rinses. To maintain the surface of the capillary, a rinse protocol of 3 min each of 0.5 N HCl, water, 0.1 M NaOH, water, and BGE was used prior to injection. Pressure injections of 0.5 psi for 4.0 s were used. Separation voltage ranged between 15 and 30 kV depending on the length of the capillary used. Data analysis was performed with OriginPro 2015 (Northampton, MA) and Peakfit v4.12 (Systat Software, San Jose, CA). Derivatization of amino acids Standard preparation for separation optimization. To prevent unreacted dye from generating fluorescent side products in the standards, reactions were done with excess amino acid to dye during separation optimization experiments. Each amino acid was labeled individually by adding 10 µL of 1 mM CFSE in DMF to 90 µL of 1 mM amino acid dissolved in 100 mM sodium tetraborate (final concentration of CFSE was 100 µM). After a 2 h reaction in the dark, the labeled standards were kept in the refrigerator for up to one week. For the separation optimization, the labeled standards were combined daily to create amino acid mixtures: 10 µL of each labeled amino acid was used and the final volume brought to 500 µL with 100 mM sodium tetraborate. Finally, before analysis, the mixtures were diluted once more 1:10 into BGE. Limit of detection. For limit of detection (LOD) measurements a stock containing the Signature17 amino acids at 1 mM each was prepared in 100 mM sodium tetraborate. Serial dilutions of this stock mixture were made in 100 mM sodium tetraborate in order to provide the amino acid concentrations used for the calibration curve (500, 250, 175, 100, 50, 25, 10, and 5 nM). Each concentration was labeled as is by combining 247.5 µL of the amino acid mixture with 2.5 µL 2 mM CFSE in DMF (resulting in a final concentration of CFSE of 20 µM) and left to react in the dark for 2 h. These samples were analyzed directly without further dilution. Mono Lake sample preparation. Water samples were collected at Mono Lake (37.977999, -119.128618) in 15 mL Falcon™ tubes (Sigma-Aldrich) in March 2012. They were transported to the laboratory in a cooler and then frozen and stored at -20 °C until analysis. Prior to analysis the samples were melted in the refrigerator for 2 h before aliquots were removed and the sample was refrozen. The melted lake water was diluted 1:1 with 100 mM sodium tetraborate pH 9.2. Labeling was done by combining 247.5 µL of the diluted sample with 2.5 µL of 2 mM CSFE in DMF and left to react in the dark for 2 h. These samples were analyzed without further dilution.

RESULTS AND DISCUSSION Fluorescent tag selection. A wide array of amine-reactive fluorophores exist for the derivatization of primary amines. In particular, fluorescein-based probes are a popular choice due to their large molar absorptivity and high fluorescence quantum yield. Yet, while picomolar LOD for CE-LIF analysis of fluorescein isothiocyanate (FITC) derivatized amino acids have been reported, they were obtained with diluted samples following derivatization at a high concentration.27 Due to the slow reaction kinetics of the isothiocyanate moiety with the amine group and the competition of this reaction with hydrolysis at high pH,28 micromolar concentrations of amino acids are needed for adequate derivatization with FITC when the amino acid is the limiting reagent.27,29-31 This means that the true LOD

ACS Paragon Plus Environment

Page 3 of 10

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

of an assay is actually determined by the lowest concentration of amino acid that can be derivatized by a given fluorophore. Although it is generally accepted that the search for biosignatures on worlds like Europa will require ultrasensitive instrumentation, it is difficult to assign a specific value for the LOD requirements. In the absence of an accurate planetary model, we can look at life in extreme environments on Earth to define the instrument requirements. For example, in the subglacial Lake Vostok in Antarctica, amino acids exist at low-nanomolar concentrations, corresponding to part-per-trillion (ppt) levels (mole/mole).32 Therefore, the protocols for in situ instrumentation seeking amino acids in this type of environment should at least be able to achieve this sensitivity. By changing the reactive moiety on fluorescein from isothiocyanate to succinimidyl ester, Banks et al. reported improved reactivity with amines and the formation of more stable conjugates.33 Furthermore, Lau et al. were able to achieve derivatization of 1 nM GABA with CFSE while forming fewer fluorescent side products than reactions with FITC.30 Due to the improvement in the derivatization efficiency (and therefore the LOD), CFSE was chosen as the fluorescent dye for this work. Amino acid target selection. While having a test that could distinguish between every possible amino acid and its enantiomer would be extremely powerful, it is neither practical nor necessary. It is not practical because the majority of amino acids have similar size and charge, making it infeasible to resolve hundreds of species simultaneously. Yet, even if it were possible, it would not be necessary because the majority of amino acids are found in such low abundance that it is unlikely they would be detected in situ. Instead, we chose to focus on the development of a separation for a select group of amino acids that occur in high abundance in both abiotic and biotic samples. While the terrestrial amino acid alphabet contains 20 species, just 8 amino acids (Ala, Asp, Glu, Gly, His, Leu, Ser, and Val) make up roughly 62% of the E. coli protein mass.34,35 Given that life likely utilized materials available in high abundance in the environment,1 it is not surprising that many of these amino acids are also present in abiotic samples, such as meteorites. Ala, Asp, Glu, Gly, Leu, Ser, and Val have been found repeatedly in CM2 and CR2 class carbonaceous chondrite meteorites. 36-40 Abiotic processes also produce a high abundance of AIB, isovaline (Iva), β-Ala, and GABA. While β-Ala and GABA are not incorporated into proteins, they can exist naturally in biotic samples in low concentrations (and in higher concentrations as degradation products of proteinogenic amino acids); AIB and Iva are only found abiotically. With so many overlapping amino acid species found in abiotic and biotic samples it becomes necessary to determine biosignatures (distribution patterns) that allow us to distinguish between amino acids produced by biology from those formed by abiotic reactions. Towards this end we have identified three chemical patterns: 1) which amino acids are present; 2) the relative abundance of the proteinogenic amino acids to glycine; 3) the presence of an enantiomeric excess. For the first pattern, it is possible to simply assess which amino acids are present in a sample. Figure 1a shows a Venn diagram which divides the 12 amino acid species found in the highest concentration in biotic (E. coli cultures 34,35,41) and abiotic (CR237,40 and CM238,39 meteorites) samples into three categories: terrestrial biotic, proteinogenic, and abiotic. Using these categories and given a population of amino acids in an unknown sample, one could begin to draw conclusions about its abiotic

versus biotic origin. For example, His is not produced abiotically. If one were to encounter a sample during a planetary mission that contained His, this would be consistent with the hypothesis that this material was either terrestrial biotic (as a form of contamination), or possibly extraterrestrial biotic. If one were to encounter a high abundance of AIB with low abundances of proteinogenic amino acids, this would be consistent with an abiotic source. However, the mere detection of amino acid such as Leu would, on its own, not be suggestive of either biotic or abiotic processes.

Figure 1. The 12 most common amino acids found in abiotic (meteorites) and biotic (E. coli) samples, presented in a Venn diagram that highlights the overlap of these amino acids between biotic, proteinogenic, and abiotic classifications. B) The relative abundance of the proteinogenic amino acids with respect to Gly. Data for the E. coli values comes from averaging the concentrations reported in Lobry et al.34 and Nishikawa et al.35 Data for the meteorites comes from averaging the concentrations of amino acids reported for several CM237,40 and CR238,39 meteorites.

To build a stronger case on the abiotic or biotic origin of a sample, it is useful to look at the second pattern: the relative abundance of the proteinogenic amino acids with respect to glycine. Abiotic samples have a much higher abundance of smaller, easier to synthesize amino acids, with Gly making up to 23% of the total amino acid mass of the Murchison meteorite.37 Due to the inherent Gibbs free energy of formation cost, 42 the larger molecular weight proteinogenic amino acids are present in lower abundances relative to Gly as seen in Figure 1b. Alternatively, proteins with higher functionality require amino acids with more complexity. Therefore, the higher molecular weight amino acids such as Ala, Asp, and Glu are more abundant than Gly in biotic samples (Figure 1b). These two drastically differ-

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

Page 4 of 10

Figure 2. Separation optimization of neutral amino acids through the addition of chiral selectors to a BGE of 80 mM sodium tetraborate pH 9.2. A) 20 mM β-CD and 30 mM STC, conditions previously described by Lu et al.; B) 20 mM β-CD and 30 mM STC with 5% v/v ACN; C) 20 mM γ-CD and 30 mM STC with 5% v/v ACN; D) 30 mM γ-CD and 30 mM STC with 5% v/v ACN. Peaks: 1. D-His; 2. D-Leu; 3. DVal; 4. L-His; 5. L-Leu; 6. D-Ser; 7. GABA; 8. L-Val; 9. D-Ala; 10. L-Ser; 11. β-Ala; 12. L-Ala; 13. Gly; 14. D-Glu; 15. D-Asp; 16. L-Glu; 17. L-Asp; *Dye side products

ent abundance patterns can be used to draw further conclusions different about the presence of higher-order biological processes. Finally, we can look at the third pattern: chiral distribution. The enantiomeric excess of amino acids has long been touted as the key piece of evidence needed to prove the existence of extraterrestrial life.43-46 Proper function and folding of terrestrial proteins requires enantiomeric purity. Because of this, in a biotic sample the enantiomeric ratio is very low (D/L 0.01-0.2 for Ala);47 whereas abiotic reactions generate a racemic mixture containing equal amounts of both D- and L-amino acids (D/L = 1.0 for all amino acids). To be able to recognize the three signatures addressed above, we chose to optimize the assay for a standard mix of the 20 amino acid found in the highest abundance in abiotic and biotic samples: four achiral amino acids (AIB, Gly, GABA, and βAla) and eight enantiomer pairs (D/L-Ala, -Asp, -Glu, -His, -Iva, -Leu, -Ser, and -Val). However, initial labeling experiments showed low reaction efficiency of CFSE with Iva and AIB. This is likely because the primary amine of these species is attached

to a tertiary carbon (i.e. a carbon atom with three carbon neighbors). In this case, the steric hindrance limits the accessibility of the primary amine to react with the succinimidyl ester as previously reported.22,26 Given the poor efficiency of the labeling reaction, even when the reaction was done with excess amino acid to CFSE, it’s unlikely that this method would be able to detect these species with the necessary sensitivity to find them in situ. The remaining 17 amino acids showed excellent reactivity with CFSE and were selected to create the Signature17 standard. Separation optimization Neutral amino acids. There are 13 neutral amino acids in the Signature17 standard: three achiral amino acids (Gly, GABA, and β-Ala) and five enantiomer pairs (D/L-Ala, -His, -Leu, -Ser, and -Val). Due to their identical charge and similar size (particularly after the addition of the fluorophore), the simultaneous separation of these species with CE is challenging. Furthermore, for this work, the initial separation optimization was done on a 20 cm effective length capillary. This was chosen to mimic the microchip design of the current prototype of the portable

ACS Paragon Plus Environment

Page 5 of 10

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

Figure 3. Electropherograms showing the effect of capillary length on the resolution of the 13 neutral amino acid species of the Signature17 standard. The field strength was maintained at 0.5 kV/cm for all capillary lengths (effective length shown in figure). BGE: 80 mM sodium tetraborate, 30 mM γ-CD, 30 mM STC, and 5% v/v ACN. Peaks: 1. D-His; 2. D-Leu; 3. D-Val; 4. L-His; 5. L-Leu; 6. D-Ser; 7. GABA; 8. LVal; 9. D-Ala; 10. L-Ser; 11. β-Ala; 12. L-Ala; 13. Gly; *Dye side products

ME instrument developed at JPL, the Chemical Laptop. This is less than half the length of capillaries typically used for CE separations of enantiomers and provides less transit time for analyte resolution to take place, which makes the simultaneous separation of all species even more challenging. Enantiomer separations can be achieved in CE through the addition of chiral selectors to the BGE. The most commonly used chiral selectors are CDs. For better versatility CDs can be paired with surfactant micelles to perform CD-assisted MEKC. When these micelles are formed from chiral surfactants such as bile salts, they become chiral selectors themselves. It has been shown that the combination of β-CD and bile salt surfactants have a cooperative effect for enantiomeric separations, providing better chiral resolution than having either of the chiral selectors alone.48-52 Lu et al. demonstrated the benefits of this dual chiral selector effect on FITC-labeled amino acids in an 80 mM borate BGE with 20 mM β-CD and 30 mM STC.49 While Lu et al. were able to resolve each enantiomer pair of all 19 chiral proteinogenic amino acids individually, they only reported a separation of 6 enantiomer pairs simultaneously (D/L-Ala, -Arg, -Asp, -Leu, -Gln, and -Glu).The BGE conditions described by Lu et al. were used as a starting point for the optimization of the Signature17 separation. However, due to the short channel, the

amino acids were largely unresolved from one another (Figure 2a). In order to improve resolution, the effect of adding organic solvents to the BGE was evaluated. Organic solvents can improve the resolution between analytes in two ways. First they modify the polarity of the BGE and effect the kinetics of the binding interactions between the analytes and the chiral selectors. Second, they change the viscosity and the dielectric constant of the BGE, which alters the electroosmotic flow (EOF). The reduction of the EOF allows for longer separation times, increasing the likelihood of interaction between the analytes and the chiral selectors. Several organic solvents (IPA, MeOH, and ACN) were added to the BGE at 5% v/v. Of those tested, ACN showed the best improvement to the resolution (Figure 2b). At concentrations above 5% v/v ACN, the longer migration times lead to broadening of the peaks with no additional improvement to the resolution. The addition of organic modifier was an improvement over the original BGE used by Lu et al., but it was not enough to fully resolve all of the amino acids in the Signature17 standard. To further improve the resolution of the neutral species in the Signature17 standard a variety of modified and unmodified CDs were tested. CDs are able to act as effective chiral selectors through the hydrophobic incorporation of the analyte into the

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

cavity formed by the ring of glucose molecules. Due to the size of the dye, the hydrophobic character of each analyte is dominated by the fluorescein moiety which is roughly 10 Å across 53 and more than twice as long as an amino acid.54 By increasing the size of the CD from a 7 (β-CD) to 8 (γ-CD) glucose monomer ring, the diameter of the inclusion cavity is increased from 7.8 to 9.5 Å. The larger cavity can better accommodate the fluorescein moiety, increasing the likelihood of the secondary chiral interactions between the chiral centers of the amino acid and the glucose, making it possible to differentiate between L- and D-amino acids. Keeping the concentration of STC constant, β-CD was exchanged for γ-CD (20mM). This improved the resolution of the neutral amino acids (Figure 2c). The concentration of γ-CD was further increased by 5 mM increments from 20 to 40 mM. It was determined that 30 mM was optimal as the most amino acids were either fully or partially resolved from one another (Figure 2d) at this concentration. On the 20 cm effective length capillary with the current method it was possible to partially resolve 12 of the 13 species (β-Ala and L-Ala remain comigrated). All further attempts at BGE optimization, including the use of modified CD and dual-CD BGEs, yielded no improvement upon this method. Finally, to provide increased resolution without additional BGE optimization, the length of the capillary was increased. Effective lengths of 30 to 50 cm were tested while keeping the field strength constant (0.5 kV/cm) (Figure 3). It was observed that with a 40 cm capillary it is possible to achieve baseline resolution of all 13 neutral amino acid species. Additionally, the enantiomeric resolution for the chiral amino acids ranged from Rs 7.8-16.5 (supporting information). This result will inform the design of the next portable ME system for in situ analysis of amino acids built at JPL so that longer channels and higher separation potentials are possible. Acidic amino acids. As shown in Figure 2, the chiral resolution of the neutral species requires a BGE containing several additives which increase the migration time of the analytes. However, this negatively affects the peaks for the late-migrating acidic species such as Asp and Glu. A slow EOF and a negative mobility lead to diffusion of the analyte zone and broadening of the peaks. By simply removing the STC and the 5% ACN from the BGE, the migration times are decreased and simultaneous resolution of Asp and Glu enantiomers (D/L Rs 15.9 and 14.9 respectively) is achieved on a 40 cm capillary (Figure 4). Both the neutral and acidic chiral amino acid methods can be run back to back in under an hour for the detection of all the species in the Signature17 standard. Limit of detection. As previously discussed, the LOD is not only a function of the LIF detection system, but is fundamentally dependent on the labeling of low concentration amines with a given fluorophore. Often the difficulties associated with the latter are not addressed when touting the benefits of LIF detection. In order to drive the reaction to generate high yield of the labeled amines, it is necessary to have a sufficient excess of the fluorescent label. However, many of these fluorophores are also susceptible to hydrolysis and other degradation pathways generating many fluorescent side products. At high concentrations of dye, the signal from these side products can exceed the signal generated by the low concentration analytes of interest.

Page 6 of 10

Figure 4. Electropherograms showing the resolution of the acidic amino acids enantiomers. BGE: 80 mM sodium tetraborate, 30 mM γ-CD. 40 cm effective length capillary (50 cm total), with 25 kV separation voltage. Peaks: 1. D-Glu; 2. D-Asp; 3. L-Glu; 4. L-Asp; *Dye side products

The lowest reported limits of detection for labeling low concentration amino acids was achieved by Chiesl et al. with Pacific Blue succinimidyl ester.26 Using this derivatization reagent with overnight labeling they were able to derivatize Val down to 75 pM while keeping the dye at 200 µM for all concentrations of Val. For CFSE, as shown by Lau et al., it was possible to derivatize GABA down to 1 nM.30 To achieve this, they performed the amine labeling with 1 mM CFSE (a million fold excess). Fortunately, with a borate BGE, their single analyte (GABA) was resolved from the dye side products that were formed. However, as the separation becomes more complex (e.g. simultaneous resolution of 13 labeled amino acids), it is necessary to use a BGE with additives capable of distinguishing between analytes with small structural changes. With our CZE and CDMEKC BGEs we see the separation of many fluorescent side products formed by CFSE, and unfortunately, several of these side products comigrate with both the neutral and acidic amino acids. To mitigate the interference of these side products, the final concentration of the dye was kept to 20 µM. This reduction of the amount of excess dye will not only reduce the dye side products, but it will also decrease the yield of the labeling reaction. However, this is a necessary compromise to reduce the interference between the analytes and the dye side products and achieve lower detection limits. Despite dye interference it is still possible to distinguish nanomolar levels of amino acids from the baseline with these methods (Table 1). Calibration curves for the neutral amino acids were made by derivatizing samples containing equimolar amounts of each amino acid from 5 nM to 500 nM with 20 µM CFSE. For the species that did not comigrate with the dye, the LOD was determined to be the lowest point on the linear curve (R2 > 0.998) where S/N > 3. For some of these amino acids it is possible to derivatize down to 5 nM. In pure water this concentration correlates to 90 ppt (mole/mole). This is a two order of magnitude improvement from what was possible with FITC (1 µM).30 For the neutral species that comigrate with the dye side products (D-Val, L-Ser, β-Ala, L-Ala, and Gly), the peak area of the blank was compared to that of the labeled samples. Here the

ACS Paragon Plus Environment

Page 7 of 10

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

LOD was the lowest derivatizable concentration where the peak area was statistically different from the blank with a ≥ 99.9% confidence (p value ≤ 0.001). Even with the dye side-product comigration all of the amino acids were detectable between 5100 nM. The only neutral species that we were unable to quantify was D-Val. This is because D-Val comigrates with a dye peak that changes in intensity over time as the dye degradation occurs. Table 1. Lowest derivatized concentration of amino acids Neutral amino acid

LOD (nM)

Acidic amino acid

LOD (nM)

D-Ala

25

D-Asp

500

L-Ala

25

L-Asp

750

β-Ala

100

D-Glu

NA

GABA

5

L-Glu

750

Gly

5

D-His

10

L-His

10

D-Leu

10

L-Leu

5

D-Ser

10

L-Ser

25

D-Val

NA

L-Val

10

NA – Not available due to comigration with dye side-products

For the acidic amino acids the LODs are higher because the Asp and Glu enantiomers migrate in the same window as several high intensity dye side product peaks (supplemental information). BGE optimization does not have a significant effect on the separation of Asp and Glu from these side products. For DAsp, D-Glu, and L-Glu a calibration curve using concentrations between 100-1500 nM was used, and the lowest derivatizable concentration of 500-750 nM was necessary to identify the peaks above the blank. Unfortunately, L-Asp comigrates with a large dye peak and the signal is lost so it cannot be quantified. Analysis of Mono Lake samples. Water plays a critical role for life on Earth, yet water is an excellent solvent for salts and minerals and can exist at a wide range of pHs. It has become clear that in harsh environments previously assumed to be inhabitable, there are diverse populations of extremophile microorganisms that thrive. These sites can be used as analogs for astrobiologically relevant targets.55 It is possible that Mars used to have an abundance of water but now all that remains is the potential for very briny water seeps that come and go at the surface.56 Similarly, both Europa57-60 and Enceladus61,62 are predicted to have high salinity oceans. Therefore, in order to detect biosignatures in complex matrixes encountered on spaceflight missions, it is necessary to have robust analytical techniques that can perform well in the presence of salts. To test the ability of our assay to function in the presence of high salt concentrations we analyzed samples from Mono Lake. Mono Lake is a shallow soda brine lake in Mono County, California. The high pH (9.8) and hypersalinity (81 g/L) supports the growth of a variety of halophilic and alkaliphilic extremophiles in all three domains (Bacteria, Archaea, and Eukayra) that can tolerate the two extremes.55,63,64

The analysis of the samples from Mono Lake showed nanomolar concentrations of several of the amino acids from the Signature17 standard (Figure 5a). It is important to note that the extreme salinity (3x saltier than the ocean) did not affect the ability of this assay to label and detect amino acids, despite minimal sample preparation. In the lake water diluted 1:1 with tetraborate buffer, the efficiency of the labeling reaction was corrected for by the addition of an internal standard. D-His was chosen as the internal standard for these experiments after determining it was not present in the original sample. The concentrations of the amino acids in Mono Lake were back-calculated from the calibration curves and then corrected for labeling efficiency as well as dilution factor. Using this data (Table 2) the three biosignatures introduced in this paper were evaluated. The first biosignature, which amino acids are present, was positive through the detection of L-His in the sample. L-His is only synthesized biotically giving the first indication of chemistry beyond abiotic reactions. Interestingly, a high concentration of GABA was also detected in these samples. This is unusual for typical biotic samples, however, this can be attributed to the extreme conditions of Mono Lake. Small non-protein amino acids, including GABA, can be used by cells as osmolytes to prevent lysis in extreme salt environments.65 GABA has been found in Mono Lake in a previous study at all depths at abundances between 0.1-1.1% of all amino acids in the sample.66 The second signature, relative abundance of the proteinogenic amino acids to Gly, showed that the pattern of abundance for Mono Lake amino acids favored the higher molecular weight amino acids to Gly, similar to that of the biotic “fingerprint” (Figure 5B). While Glu and Asp are expected to be present in this sample they were not detected due to dye interference. Table 2. Amino acids detected in Mono Lake Neutral amino acid

Concentration (nM)

Acidic amino acid

Concentration (nM)

D-Ala

50.52 ± 0.06

D-Asp

ND*

L-Ala

189.8 ± 0.1

L-Asp

ND*

β-Ala

ND

D-Glu

ND*

GABA

15.8 ± 0.1

L-Glu

ND*

Gly

166.80 ± 0.06

D-His

ND

L-His

46.5 ± 0.1

D-Leu

ND

L-Leu

101.20 ± 0.03

D-Ser

ND

L-Ser

62.6 ± 0.4

D-Val

ND*

L-Val

107.5 ± 0.1

ND - not detected above the LOD; ND* - not detected due to dye side-product interference. Error propagated from standard deviation of 3 consecutive analyses of a single sample.

Finally, our analysis also provides information pertaining to the third biosignature. Of the five neutral enantiomer pairs detected with this method, all showed an excess of L-amino acids.

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

Page 8 of 10

Figure 5: A) Electropherogram of the analysis for the Mono Lake sample compared to the blank. Separation conditions optimized for neutral amino acids, BGE: 80 mM sodium tetraborate pH 9.2, 30 mM γ-CD, 30 mM STC, and 5% v/v ACN. 40 cm effective length channel with a 25 kV separation voltage. Peaks: 4. L-His; 5. L-Leu; 7. GABA; 8. L-Val; 9. D-Ala; 10. L-Ser; 12. L-Ala; 13. Gly; *dye side products, peak numbers bolded and underlined denote amino acid species which comigrate with dye side product peaks, unmarked peaks are unidentified. Inset: Photo of the Mono Lake sample site (37.977999, -19.128618). B) The relative abundance of the amino acids in Mono Lake in comparison to that of meteorite and E. coli samples, for biosignature number two.

D-His, D-Leu, D-Ser,

and D-Val were not detected in concentrations above their LOD. The only enantiomer that was detected in both the L- and D-form was alanine. The D/L ratio was calculated to be 0.266 ± 0.001, slightly higher than what would be predicted from the contribution of D-Ala in the peptidoglycan in the bacterial cell wall (D/L 0.01-0.2).47 However, Mono Lake is a unique environment and may differ from other bodies of water when it comes to the production and cycling of peptidoglycan.66 In fact, Jorgenson et al. have reported an enantiomeric excess of up to 0.52 for Ala in Mono Lake depending on the year and season.

CONCLUSIONS This paper demonstrates two CE methods for the improved chiral resolution of the 17 amino acids found in high abundance in abiotic and biotic samples. The protocol presented here is the first report of the simultaneous separation of this many astrobiologically relevant amino acids enantiomers by CE-LIF. To date, this is the best method available for the detection of amino acid biosignatures on other worlds. Using the Signature17 standard, it is possible to identify the three distribution patterns that indicate the presence of chemistry that does not conform to what we observe abiotically. Additionally, the limit of detection reported here are up to 10,000 times better than what has been used on flight missions in the past (SAM GC-MS).8 By using the fluorescent label CFSE, the neutral species can be derivatized at concentrations as low as 5-25 nM. This is a two order of magnitude increase from what is possible with the previous fluorescent tag, FITC. The acidic amino acids still remain a challenge for the low concentration analysis with CFSE due to the interference of peaks formed from dye side products. Further work is underway to decrease the LOD for the acidic amino acids species.

In addition to the improvements on resolution and LOD, an important finding from this study is that water samples containing high concentrations of salts (from Mono Lake, CA) can be analyzed with minimal sample preparation. The hypersaline lake water samples were analyzed by simply mixing 1:1 with a tetraborate buffer and then adding CFSE for derivatization. No desalting or preconcentration was necessary. With this method, the presence of all three biosignatures were detected in the Mono Lake sample. These data clearly illustrate how the methods and biosignature patterns described here could be used to identify the presence of life on a planetary mission. These results highlight the applicability of the protocols described here for the detection of amino acid biosignatures in salty samples during possible missions to Europa or Enceladus, two prime destinations for the search of life beyond Earth.

ASSOCIATED CONTENT Supporting Information The Supporting Information is available free of charge on the ACS Publications website. Electropherograms showing the interference of dye side products with the acidic amino acid peaks (PDF). Table of the enantiomer resolution of the chiral amino acids in the Signature17 standard (PDF).

AUTHOR INFORMATION Corresponding Author *[email protected]

Author Contributions All authors have given approval to the final version of the manuscript.

ACS Paragon Plus Environment

Page 9 of 10

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

Copyright 2016 California Institute of Technology. US Government sponsor acknowledged.

ACKNOWLEDGMENTS The work done in this article was done at the Jet Propulsion Laboratory, California Institute of Technology, under a contract with the National Aeronautics and Space Administration. Financial support was provided by The Planetary Instrument Concepts for the Advancement of Solar System Observations (PICASSO) Program and the NASA Postdoctoral Program (NPP) at the Jet Propulsion Laboratory, administered by Universities Space Research Association through a contract with NASA.

REFERENCES (1) Bada, J. L. Earth Planet. Sci. Lett. 2004, 226, 1-15. (2) McKay, C. P. PLoS Biol 2004, 2, 1260-1263. (3) Willis, P. A.; Creamer, J. S.; Mora, M. F. Anal. Bioanal. Chem. 2015, 407, 6939-6963. (4) Biemann, K.; Oro, J.; Toulmin, P.; Orgel, L.; Nier, A.; Anderson, D.; Simmonds, P.; Flory, D.; Diaz, A.; Rushneck, D. Science 1976, 194, 72-76. (5) Biemann, K.; Oro, J.; Toulmin, P.; Orgel, L.; Nier, A.; Anderson, D.; Simmonds, P.; Flory, D.; Diaz, A.; Rushneck, D. J. Geophys. Res. 1977, 82, 4641-4658. (6) Brown, R.; Lebreton, J. P.; Waite, J. Titan from CassiniHuygens; Springer Science & Business Media, 2009. (7) Waite, J. H.; Combi, M. R.; Ip, W.-H.; Cravens, T. E.; McNutt, R. L.; Kasprzak, W.; Yelle, R.; Luhmann, J.; Niemann, H.; Gell, D. Science 2006, 311, 1419-1422. (8) Stalport, F.; Glavin, D. P.; Eigenbrode, J.; Bish, D.; Blake, D.; Coll, P.; Szopa, C.; Buch, A.; McAdam, A.; Dworkin, J. Planet. Space Sci 2012, 67, 1-13. (9) Han, Y.; Zuo, M.; Qi, L.; Liu, K.; Mao, L.; Chen, Y. Electrophoresis 2006, 27, 4240-4248. (10) Stockton, A. M.; Chiesl, T. N.; Lowenstein, T. K.; Amashukeli, X.; Grunthaner, F.; Mathies, R. A. Astrobiology 2009, 9, 823-831. (11) Fanali, S. J. Chromatogr. A 2000, 875, 89-122. (12) Giuffrida, A.; Maccarrone, G.; Cucinotta, V.; Orlandini, S.; Contino, A. J. Chromatogr. A 2014, 1363, 41-50. (13) Wan, H.; Blomberg, L. G. J. Chromatogr. A 2000, 875, 43-88. (14) Verleysen, K.; Sandra, P. Electrophoresis 1998, 19, 2798-2833. (15) Nagl, S.; Schulze, P.; Ohla, S.; Beyreiss, R.; Gitlin, L.; Belder, D. Anal. Chem. 2011, 83, 3232-3238. (16) Mora, M. F.; Stockton, A. M.; Willis, P. A. Electrophoresis 2012, 33, 2624-2638. (17) Willis, P. A.; Stockton, A. M. Microchip capillary electrophoresis for in situ planetary exploration, 2013, p 277-291. (18) Castro, E. R.; Manz, A. J. Chromatogr., A 2015, 1382, 66-85. (19) Chiesl, T. N.; Benhabib, M.; Stockton, A. M.; Mathies, R. A. In Astrobiology Science Conference, 2010, p 5021. (20) Skelley, A. M.; Scherer, J. R.; Aubrey, A. D.; Grover, W. H.; Ivester, R. H. C.; Ehrenfreund, P.; Grunthaner, F. J.; Bada, J. L.; Mathies, R. A. Proc. Nat. Acad. Sci. 2005, 102, 1041-1046. (21) Mora, M. F.; Greer, F.; Stockton, A. M.; Bryant, S.; Willis, P. A. Anal. Chem. 2011, 83, 8636-8641. (22) Skelley, A. M.; Mathies, R. A. J. Chromatogr. A 2003, 1021, 191-199. (23) Hutt, L. D.; Glavin, D. P.; Bada, J. L.; Mathies, R. A. Anal. Chem. 1999, 71, 4000-4006. (24) Danger, G.; Ross, D. Electrophoresis 2008, 29, 3107-3114. (25) Danger, G.; Ross, D. Electrophoresis 2008, 29, 4036-4044. (26) Chiesl, T. N.; Chu, W. K.; Stockton, A. M.; Amashukeli, X.; Grunthaner, F.; Mathies, R. A. Anal. Chem. 2009, 81, 2537-2544. (27) Cheng, Y.-F.; Dovichi, N. J. Science 1988, 242, 562-564. (28) Patai, S. Chemistry of cyanates and their thio derivatives; J. Wiley, 1977. (29) Albin, M.; Weinberger, R.; Sapp, E.; Moring, S. Anal. Chem. 1991, 63, 417-422.

(30) Lau, S. K.; Zaccardo, F.; Little, M.; Banks, P. J. Chromatogr. A 1998, 809, 203-210. (31) Simó, C.; Barbas, C.; Cifuentes, A. Electrophoresis 2005, 26, 1306-1318. (32) Christner, B. C.; Royston-Bishop, G.; Foreman, C. M.; Arnold, B. R.; Tranter, M.; Welch, K. A.; Lyons, W. B.; Tsapin, A. I.; Studinger, M.; Priscu, J. C. Limnol. and Oceanogr. 2006, 51, 24852501. (33) Banks, P. R.; Paquette, D. M. Bioconjugate Chem. 1995, 6, 447458. (34) Lobry, J. R.; Gautier, C. Nucleic Acids Res. 1994, 22, 31743180. (35) Nishikawa, K.; Ooi, T. J. Biochem. 1982, 91, 1821-1824. (36) Cobb, A. K.; Pudritz, R. E. Astrophys. J. 2014, 783, 140-152. (37) Cronin, J.; Pizzarello, S. Adv. Space Res. 1983, 3, 5-18. (38) Martins, Z.; Hofmann, B.; Gnos, E.; Greenwood, R.; Verchovsky, A.; Franchi, I.; Jull, A.; Botta, O.; Glavin, D.; Dworkin, J. Meteorit. Planet. Sci. 2007, 42, 1581-1595. (39) Pizzarello, S.; Holmes, W. Geochim. Cosmochim. Ac. 2009, 73, 2150-2162. (40) Shimoyama, A.; Harada, K.; Yanai, K. Chem. Lett. 1985, 14, 1183-1186. (41) Aubrey, A. D. Amino Acid Biosignatures-Implications for the Detection of Extinct or Extant Microbial Communities on Mars. ProQuest2008. (42) Higgs, P. G.; Pudritz, R. E. Astrobiology 2009, 9, 483-490. (43) Bada, J. L. Science 1997, 275, 942-943. (44) Bada, J. L.; McDonald, G. D. Anal. Chem. 1996, 68, 668A673A. (45) Halpern, B. Appl. Opt. 1969, 8, 1349-1353. (46) Kvenvolden, K. A. Space Life Sci. 1973, 4, 60-68. (47) Veuger, B.; Middelburg, J. J.; Boschker, H. T.; Houtekamer, M. Limnol. Oceanogr. Methods 2007, 5, 192-194. (48) Chen, F.; Zhang, S.; Qi, L.; Chen, Y. Electrophoresis 2006, 27, 2896-2904. (49) Lu, X.; Chen, Y. J. Chromatogr. A 2002, 955, 133-140. (50) Okafo, G. N.; Bintz, C.; Clarke, S. E.; Camilleri, P. J. Chem. Soc., Chem. Commun. 1992, 1189-1192. (51) Okafo, G. N.; Camilleri, P. J. Microcolumn Sep. 1993, 5, 149153. (52) Otsuka, K.; Terabe, S. J. Chromatogr. A 2000, 875, 163-178. (53) Jin, L. J.; Rodriguez, I.; Li, S. F. Electrophoresis 1999, 20, 1538-1545. (54) Ainavarapu, S. R. K.; Brujić, J.; Huang, H. H.; Wiita, A. P.; Lu, H.; Li, L.; Walther, K. A.; Carrion-Vazquez, M.; Li, H.; Fernandez, J. M. Biophys. J. 2007, 92, 225-233. (55) DasSarma, S. Microbe 2006, 1, 120-126. (56) Ojha, L.; Wilhelm, M. B.; Murchie, S. L.; McEwen, A. S.; Wray, J. J.; Hanley, J.; Masse, M.; Chojnacki, M. Nature Geosci 2015, 8, 829-832. (57) Hand, K.; Carlson, R. Geophys. Res. Lett. 2015, 42, 3174-3178. (58) Hand, K. P.; Chyba, C. F. Icarus 2007, 189, 424-438. (59) McCord, T.; Hansen, G.; Fanale, F.; Carlson, R.; Matson, D.; Johnson, T.; Smythe, W.; Crowley, J.; Martin, P.; Ocampo, A. Science 1998, 280, 1242-1245. (60) Muñoz-Iglesias, V.; Bonales, L. J.; Prieto-Ballesteros, O. Astrobiology 2013, 13, 693-702. (61) Brown, R. H.; Clark, R. N.; Buratti, B. J.; Cruikshank, D. P.; Barnes, J. W.; Mastrapa, R. M. E.; Bauer, J.; Newman, S.; Momary, T.; Baines, K. H.; Bellucci, G.; Capaccioni, F.; Cerroni, P.; Combes, M.; Coradini, A.; Drossart, P.; Formisano, V.; Jaumann, R.; Langevin, Y.; Matson, D. L.; McCord, T. B.; Nelson, R. M.; Nicholson, P. D.; Sicardy, B.; Sotin, C. Science 2006, 311, 1425-1428. (62) Zolotov, M. Y. Geophys. Res. Lett. 2007, 34. (63) Humayoun, S. B.; Bano, N.; Hollibaugh, J. T. Appl. Environ. Microbiol. 2003, 69, 1030-1042. (64) Winkler, D. W. An ecological study of Mono Lake, California; Institute of Ecology, University of California, 1977. (65) Roberts, M. F. Saline Syst. 2005, 1, 1-30. (66) J⊘rgensen, N. O. G.; Engel, P.; Jellison, R.; Hollibaugh, J. T. Geomicrobiol. J. 2008, 25, 38-55.

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

Table of Contents artwork:

Keywords: Astrobiology, chiral analysis, laser induced fluorescence detection, life detection, and limit of detection

ACS Paragon Plus Environment

Page 10 of 10