Geochemistry Geophysics Geosystems
3
Forum
G
Volume 2 July 2, 2001 Paper number 2001GC000142
AN ELECTRONIC JOURNAL OF THE EARTH SCIENCES Published by AGU and the Geochemical Society
ISSN: 1525-2027
Exploiting the multivariate isotopic nature of organic compounds Kai-Uwe Hinrichs Hanse Institute of Advanced Study, D-27753 Delmenhorst, Germany Permanently at Department of Geology and Geophysics, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543 (
[email protected])
Geoffrey Eglinton Hanse Institute of Advanced Study, D-27753 Delmenhorst, Germany Department of Marine Chemistry and Geochemistry, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543, USA Permanently at Biogeochemistry Centre, Earth Sciences Department, University of Bristol, Bristol BS81TH, England, UK (
[email protected])
Michael H. Engel Department of Geology and Geophysics, University of Oklahoma, Norman, Oklahoma 73019, USA (
[email protected])
Roger E. Summons Australian Geological Survey Organisation, GPO Box 378, Canberra, ACT 2601, Australia (
[email protected]) Keywords: Organic biomarkers; multi-isotopic molecular analysis; astrobiology; isotopic fractionation; extraterrestrial samples. Index terms: Isotopic composition/chemistry; organic geochemistry; techniques applicable in three or more fields. Received January 3, 2001; Revised March 26, 2001; Accepted March 27, 2001; Published July 2, 2001. Hinrichs, K.-U., G. Eglinton, M. H. Engel, and R. E. Summons, 2001. Exploiting the multivariate isotopic nature of organic compounds, Geochem. Geophys. Geosyst., vol. 2, Paper number 2001GC000142 [2961 words, 4 figures]. Published July 2, 2001.
NASA’s latest plans for Mars sample returns [Lawler, 2000] mean that pristine rocks from carefully selected sites on the Red Planet could be back in the Houston Space Center as early as 2014. With a mission price tag of over $2 billion, the scientists bidding for the right to analyze these samples will need to be able to
[1]
Copyright 2001 by the American Geophysical Union
offer the most powerful, sensitive, and precise techniques known, just as was done in 1969 with the lunar returned samples [e.g., Lunar Sample Analysis Planning Team, 1970]. Whether or not it turns out that Mars has, or had, living organisms on or beneath its surface, we will want to know everything we can about its
Geochemistry Geophysics Geosystems
3
G
hinrichs et al.: nature of organic compounds
molecular organic chemistry. However, until more basic information on the chemistry of the Martian surface is transmitted back from unmanned landers over the next few years, we will need to develop a wide variety of analytical approaches. These must include the distributions of isotopes, both intermolecular and intramolecular patterns, since we now know that these hold unique insights into the key planetary processes, both biotic and abiotic. Preparation for this challenge requires not only developing and testing novel analytical capabilities but also expanding our fundamental understanding of isotopic fractionation processes. Isotopes become unevenly distributed during virtually all chemical and biological transformations as well as by many processes that accompany planetary evolution. This leads to isotopic signatures that reveal the history and provenance of particular samples. All of the light elements, with the exception of phosphorus, that are most often present in organic compounds (i.e., the bioelements C, N, H, O, and S) exist as two or more stable isotopes. Even the simplest but globally important organic molecule methane, CH4, can be characterized by three-dimensional isotopic data: d13C, d14C, and dD. [2]
[3] Measurement of the elemental isotopic compositions of bulk samples of organic matter has been a routine technique for many years but generally requires significant amounts of material (mg to mg) and a specialized approach for each element. Emerging developments in analytical capability now mean that all of the above elements can be potentially measured on much smaller amounts of individual compounds (ng to mg) using a gas chromatograph coupled to an isotope ratio mass spectrometer via relatively straightforward, continuous flow combustion [Hayes et al., 1990] or pyrolysis [Burgoyne and Hayes, 1998] interfaces that are
2001GC000142
operated under different conditions to convert each component into a gas appropriate (i.e., CO2, N2, H2, CO, and SO2) for isotopic measurement. This simple, sensitive, and precise tool promises many new opportunities in Earth, environmental, and biological science. On November 9 and 10, 2000, a group of 16 researchers [van Bergen et al., 2000] convened at the Hanse-Wissenschaftskolleg (see http:// www.h-w-k.de) in Delmenhorst, Germany, to identify and discuss research topics that might be advanced by this technology. If science is to make the most of the opportunity provided by having access to returned Mars samples, we now need to intensify research into the fundamentals of isotope fractionation processes. This will have immediate impact in terms of enhanced understanding of biogeochemical cycling on Earth and ensure that scientists are prepared for possible unexpected combinations of geochemical and other data as was the case with the analyses of the Martian meteorite ALH84001 from Antarctica [e.g., Thomas-Keprta et al., 2000]. [4]
While relatively rare, some other types of meteorites, the carbonaceous chondrites, often contain a rich array of organic compounds. However, understanding their origins has generally been clouded by disagreement over whether or not the compounds present are truly extraterrestrial or are earthly contaminants. One potentially unambiguous means for testing a compound’s indigeneity is the characterization of the stable isotope compositions of its light elements. For example, the structural distribution of amino acids in carbonaceous chondrites alone is insufficient to reveal their origin; the additional stable isotope information of carbon and nitrogen helped to distinguish between terrestrial contaminants and genuine extraterrestrial compounds [Engel and Macko, 1997]. Analysis of extraterrestrial samples is one very obvious example, but there are many others.
[5]
Geochemistry Geophysics Geosystems
3
G
hinrichs et al.: nature of organic compounds
Multicompound, multi-isotope data hold much promise in tracer studies in microbial ecology that utilize distinctly labeled components carrying both radioactive and stable isotopes to investigate specific biochemical reactions and other complex processes in organisms and the natural environment [e.g., Boschker et al., 1998]. In addition, multi-isotope analyses of natural isotopic variation in microbial biomarker compounds could significantly aid studies of geologic organic matter transformation with applications in both sediment and soil biogeochemistry. For example, considerable progress has recently been made in elucidating the fate of methane in the ocean floor on the basis of the carbon isotopic composition of chemotaxonomic archaeal ether lipids. Surprisingly, the d13C isotopic data reveal that the methane evolved by the methanogens is consumed anaerobically by other archaea, which are very close relatives [e.g., Hinrichs et al., 1999] and thrive intimately with sulfate-reducing syntrophic bacterial partners [Boetius et al., 2000]. Similarly, for the exploration of the Deep Biosphere the isotopic signatures of other elements, particularly hydrogen and oxygen, will undoubtedly reveal heretofore hidden transfers of elements and compounds between the geosphere and biosphere. Isotopic studies of biopolymers such as nucleic acids and proteins in geologic systems will advance our understanding of the linkages between biological and geochemical processes. Industrially produced chemicals released into the environment carry distinct multi-isotope signatures introduced from the feedstocks during manufacturing. For example, compound-specific 13C and 37Cl values of polychlorinated biphenyls (PCBs), which accumulate in biota and sediments on a global scale, have recently been reported and may be a useful diagnostic tool for tracing their sources, transport, and ultimate fate [Jarman et al., 1998; Reddy et al., 2000]. Climate change and climate history research is perhaps the [6]
2001GC000142
main field where isotopic analysis of multiple elements in individual organic compounds will be able to complement and expand existing isotopic databases for atmospheric gases and biominerals. For example, the carbon isotopic compositions of sedimentary haptophytederived long-chain alkenones have already been tested for partial reconstruction of ancient levels of atmospheric CO2 [Eglinton et al., 2001]. However, these and other plant and algal biochemicals also encode hydrogen and oxygen isotopes that are linked through, and give information about, the hydrologic cycle. For example, the oxygen isotope content of sedimentary alkenones could possibly give an independent assessment of climate-dependent hydrologic conditions in the oceanic euphotic zone at the time of their deposition. In lakes the deuterium content of molecules from aquatic primary producers will serve as recorders of the water’s deuterium content, which is intimately linked to paleo rainfall and evaporation (Figure 1) [Sauer et al., 2001]. Similarly, studies of soil organic matter using pyrolysis techniques [e.g., Gleixner et al., 1999] combined with multielement isotope analysis will reveal new information about carbon turnover and how different agricultural practices affect the potential of soil to act as a source or sink for CO2. Petroleum and natural gas deposits are another case where carbon isotopes have proven very effective in delineating their sources and emplacement history. Again, hydrogen, oxygen, sulfur, and nitrogen isotopes will add further dimensions to characterization and significantly improve knowledge that directly aids exploration for and production of these resources. Related to this are the obvious applications to early Earth history and paleoenvironmental reconstruction through analysis of organic matter in Archean and Proterozoic sediments. Biomarker hydrocarbons carry carbon isotopic signatures that are informative about precursor
[7]
Geochemistry Geophysics Geosystems
3
G
hinrichs et al.: nature of organic compounds
2001GC000142
-180
-200
δ D (C 28 -stenol)
-220
-240
-260
ε =197
HO
-280
-300
-320
-340 -150
-100
-50
0
δ D (water) Figure 1. Relationship between D/H ratios (dD, in % Vienna standard mean ocean water (SMOW)) for the C28 stenol biosynthesized by phytoplankton grown in different lakes and marine environments and that of the ambient waters (modified after Sauer et al. [2001]). The line represents constant fractionation e. It is apparent that the sterol is recording the D/H ratio of the ambient water. Hence the dD value of sedimentary sterols should permit proxy estimation of the dD of past water bodies, a key feature of paleohydrology.
organisms and the carbon cycle, while complementary sulfur isotopic information, carried in certain ubiquitous and abundant organosulfur compounds such as dibenzothiophene, could shed new light on carbon and sulfur cycle relationships. Presently, there is some uncertainty about how well hydrogen isotopic signatures are insulated from diagenetic overprints. If these signatures prove to be robust, there is enormous potential for the hydrogen isotopic contents of hydrocarbons to discriminate marine and terrestrial environments and
to serve as proxies in paleoclimatology. As with meteorites, these studies are always subject to doubt or argument because rocks are open to invasion and/or loss of their indigenous organic components. Multielement isotopic data have already resolved some of these uncertainties. Forensic testing of flavors, drugs, and archaeological remains is another area where carbon isotopes have proven to be effective tracers of origin and history. In contrast to the
[8]
Geochemistry Geophysics Geosystems
3
G
hinrichs et al.: nature of organic compounds
Guaviare region Putumayo - Caqueta region Huallaga and Ucayali Valleys Apurimac Valley Chapare Valley
δ 15N + 0.1TRUX, ‰
30
Putumayo-Caqueta
20
10
0 Apurimac Guaviare -10
2001GC000142
exclusive use of a single isotope, the addition of degrees of variability through hydrogen, nitrogen, and oxygen isotopes is now clearly possible and heralds huge improvements in unraveling and understanding complex origins. This is illustrated by recent work of Ehleringer et al. [2000], who used a combination of compound-specific carbon and nitrogen isotope data, compound abundances, and statistical manipulation to predict the geographical origin of 96% of a set of 200 cocaine samples (Figure 2). Another example of tracing the origins of individual compounds by multi-isotope approaches comes from the food industry. The commercially important component of vanilla essence, vanillin, is now screened for its provenance using the stable isotope contents of oxygen and carbon (Figure 3) [Hener et al., 1998]. Ecosystem studies are illustrated at the dietary level by the isotopic analyses of human hair samples, predominantly the protein akeratin, shown in Figure 4 [Macko et al., 1998].
Huallaga-Ucayali Chapare -20 -150
-130
-110
-90
-70
-50
For the geoscience community these results from other fields are a strong indication that many widely used biomarkers contain (paleo)environmental information encrypted in their multielement stable isotope content; for example, the ultimate provenance of terrigenous materials transported by air and rivers could be deciphered in a manner similar to that of cocaine and vanillin. Such an implementation of additional isotope dimensions may lead to profound consequences for biomarker research in general, resulting in a shift of emphasis from chemotaxonomic to process specificity.
[9] -30
δ 13C - 10 TMC, ‰ Figure 2. Identification of geographic regions in South America where coca is commonly grown, based on dual isotope information of cocaine base as well as abundance of minor alkaloid components. Plotted on both axes are mixed expressions, each consisting of an isotope term and a concentration term; the y axis is [d15N cocaine (% versus air) + 0.1 relative concentration of truxilline (%)], and the x axis is [d13C cocaine (% versus PDB) 10 concentration of trimethoxycocaine]. Truxilline and trimethoxycocaine occur as two trace alkaloids in coca leaves. In addition to the obvious benefits for forensics, this illustration demonstrates the potential value of multi-isotope biomarker approaches for the geosciences to distinguish different geographical, climatological, and ecological regimes. Furthermore, it illustrates the importance of innovative data manipulation in biomarker research, especially when multiple isotope dimensions are employed. After Ehleringer et al. [2000].
[10] The recent development of continuous-flow techniques has made possible the transfer of chromatographic effluents via conversion reactor interfaces directly into isotope ratio mass spectrometers. Presently, for the analysis of isotope ratios of different elements in individual compounds, separate reactors and reaction con-
Geochemistry Geophysics Geosystems
3
G
hinrichs et al.: nature of organic compounds
-15
O
H C (ä vs PDB)
H
OH
-25
Mexico
Comores
?
OMe
synthetic
13
δ
Tahiti
H
H
-20
2001GC000142
-30
-35
biotechnological -40 -7
-6
-5
-4
δ 18O
-3
-2
-1
0
1
(ä vs CO-STD)
Figure 3. Dual isotopic (d13C and d18O, in % versus PDB and of standard CO, respectively) for the flavor compound vanillin. The three vanillin extracts from the naturally grown vanilla beans have similar d13C values, even though they come from geographically widely spaced sites: Mexico and the islands of the Comores and Tahiti. The Mexican sample, however, does differ markedly in d18O, no doubt owing to major differences in d18O in the ambient water supply. Not surprisingly, major differences in both d13C and d18O are apparent in the synthetic and biotechnological products [Hener et al., 1998]. On the basis of their dual isotopic values, the three samples of unknown origin can be assigned to Mexico.
ditions are required. Furthermore, multi-isotopic molecular analyses are restricted by the absolute amounts required for each element, e.g., 0.8 nmol (10 ng) of carbon, 30 nmol of hydrogen, 5 nmol of oxygen, and 1.5 nmol of nitrogen. Taking for example a C30 n-alkanol component of a higher plant leaf wax, this translates into a requirement for dD of almost 20 times the amount of compound required for d13C analyses (and d18O would require 200
times more than d13C!). This wide range of the absolute amount of compound to be analyzed poses a serious challenge to the gas chromatographic separation technique. Simultaneously obtaining high-quality isotopic data for H, C, and O is clearly impracticable at present, in view of the marked disparities in sample size and the requirements for multiple reactors and mass spectrometry (MS) systems. Ideally, we need a single, multipurpose reactor and a single
Geochemistry Geophysics Geosystems
3
G
hinrichs et al.: nature of organic compounds
11
δ15N
10
9
8
7 -20
-19
-18
-17
-16
δ
13C
Figure 4. Twin element stable isotope distributions (d13C and d15N, in % versus PDB and air, respectively) of hair samples taken from individual students at the University of Virginia (modified after Macko et al. [1998]). Hair is largely composed of the fibrous protein a-keratin. Its isotope composition is reflective of the recent diet of an individual, generally increasing slightly (1% in d13C and 3% in d15N) with trophic level. The marked variability accords with the great diversity of modern diets. For comparison, George Washington’s hair sample testifies to a rather balanced diet.
MS detection system. Could this be achieved? Time will tell.
References Boetius, A., K. Ravenschlag, C. J. Schubert, D. Rickert, F. Widdel, A. Gieseke, R. Amann, B. B. Jørgensen, U. Witte, and O. Pfannkuche, A marine microbial consortium apparently mediating anaerobic oxidation of methane, Nature, 407, 623 – 626, 2000. Boschker, H. T. S., S. C. Nold, P. Wellsbury, D. Bos, W. der Graaf, R. Pel, R. J. Parkes, and T. E. Cappenberg, Direct linking of microbial populations to specific biogeochemical processes by 13C-labelling of biomarkers, Nature, 392, 801 – 805, 1998.
2001GC000142
Burgoyne, T. W., and J. M. Hayes, Quantitative production of H2 by pyrolysis of gas chromatographic effluents, Anal. Chem., 70, 5136 – 5141, 1998. Eglinton, T. I., M. H. Conte, G. Eglinton, and J. M. Hayes, Proceedings of a workshop on alkenone-based paleoceanographic indicators, Geochem. Geophys. Geosyst., 2 (Article), 2000GC000122 [4984 words], 2001. Ehleringer, J. R., J. F. Casale, M. J. Lott, and V. L. Ford, Tracing the geographical origin of cocaine, Nature, 408, 311 – 312, 2000. Engel, M., and S. Macko, Isotopic evidence for extraterrestrial non-racemic amino acids in the Murchison meteorite, Nature, 389, 265 – 268, 1997. Gleixner, G., R. Bol, and J. Balesdent, Molecular insight into soil carbon turnover, Rapid Commun. Mass. Spectrom., 13, 1278 – 1283, 1999. Hayes, J. M., K. H. Freeman, B. N. Popp, and C. H. Hoham, Compound-specific isotopic analyses: A novel tool for reconstruction of ancient biogeochemical processes, Org. Geochem., 16, 1115 – 1128, 1990. Hener, U., W. A. Brand, A. W. Hilkert, D. Juchelka, A. Mosandl, and F. Z. Podebrad, Simultaneous on-line analysis of 18O/16O and 13C/12C ratios of organic compounds using GC-pyrolysis-IRMS, Lebensm. Unters. Forsch. A, 206, 230 – 232, 1998. Hinrichs, K.-U., J. M. Hayes, S. P. Sylva, P. G. Brewer, and E. F. DeLong, Methane-consuming archaebacteria in marine sediments, Nature, 398, 802 – 805, 1999. Jarman, W. M., A. Hilkert, C. E. Bacon, J. W. Collister, K. Ballschmiter, and R. W. Risebrough, Compound-specific carbon isotopic analysis of Aroclors, Clophens, Kaneclors, and Phenoclors, Environ. Sci. Technol., 32, 833 – 836, 1998. Lawler, A., A more cautious NASA sets plans for Mars, Science, 290, 915 – 916, 2000. Lunar Sample Analysis Planning Team, Summary of Apollo 11 science conference, Science, 167, 449 – 451, 1970. Macko, S. A., M. H. Engel, and K. Freeman, Variability of isotope compositions in modern and fossil organic matter, Chem. Geol., 152, 1, 1998. Reddy, C. M., L. J. Heraty, B. D. Holt, N. C. Sturchio, T. I. Eglinton, N. J. Drenzek, L. Xu, J. L. Lake, and K. A. Maruya, Stable chlorine compositions of Aroclors and Aroclor-contaminated sediments, Environ. Sci. Technol., 34, 2866 – 2869, 2000. Sauer, P. E., T. I. Eglinton, J. M. Hayes, A. Schimmelmann, and A. L. Sessions, Compound-specific D/H ratios of lipid biomarkers from sediments as a proxy for environmental and climatic conditions, Geochim. Cosmochim. Acta, 65, 213 – 222, 2001. Thomas-Keprta, K. L., D. A. Bazylinski, J. L. Kirschvink,
Geochemistry Geophysics Geosystems
3
G
hinrichs et al.: nature of organic compounds
S. J. Clemett, D. S. McKay, S. J. Wentworth, H. Vali, E. K. Gibson Jr., and C. S. Romanek, Elongated prismatic magnetite crystals in ALH84001 carbonate globules: Potential Martian magnetofossils, Geochim. Cosmochim. Acta, 64, 4049 – 4081, 2000. van Bergen, P., J. Bijma, H. T. S. Boschker, M. Bo¨ttcher, W. Brand, G. Eglinton, M. H. Engel, G. Gleixner,
2001GC000142
A. Hilkert, K.-U. Hinrichs, W. Michaelis, J. Rullko¨tter, S. Schulte, M. A. Sephton, R. E. Summons, and G. Voordouw, Multi-element stable isotope contents of molecules as information sources: New opportunities in the next decade,paper presented at 3rd Hanse Round Table Workshop, Hanse Inst. of Adv. Study, Delmenhorst, Germany, Nov. 9 – 10, 2000.
