Deuterium isotopomers as a tool in environmental research

(1)

UMEÅ UNIVERSITY MEDICAL DISSERTATION

New Series No. 1070 ISSN: 0346-6612 ISBN: 91-7264-218-1 Edited by the Dean of the Faculty of Medicine

Deuterium isotopomers as a tool

in environmental research

by

Tatiana Betson

Department of Medical Biochemistry and Biophysics, Umeå University

(2)

Department of Medical Biochemistry and Biophysics Umeå University

SE-901 87 Umeå, Sweden

New Series No. 1070

ISBN 91-7264-218-1 Printed in Sweden by Print and Media

(3)

(4)

(5)

Abbreviations

D (or 2H) Deuterium

p,p’-DDT p,p’-dichlorodiphenyltrichloroethane o.p’-DDD o.p’-dichlorodiphenyldichloroethane

DID Deuterium Isotopomer Distribution FID Free Induction Decay

FTIR Fourier Transform Infra Red spectroscopy 1_H _{Protium (or proton)}

IRMS Isotope Ratio Mass Spectrometry KIE Kinetic Isotope Effect

NMR Nuclear Magnetic Resonance spectroscopy PHC Polyhalogenated Compounds

POP Persistent Organic Pollutants TMS Tetramethylsilane

VPDB Vienna PeeDee Belemnite

(7)

Abstract

Deuterium isotopomers as a tool in environmental research

This thesis describes the development and the use of quantitative deuterium Nuclear Magnetic Resonance spectroscopy (NMR) as a tool in two areas of environmental research: the study of long term climate-plant interactions and the source tracking of persistent organic pollutants.

Long-term interactions between plants and climate will influence climate change during this century and beyond, but cannot be studied in manipulative experiments. We propose that long tree-ring series can serve as records for tracking such interactions during past centuries.

The abundance of the stable hydrogen isotope deuterium (D) is influenced by physical and biochemical isotope fractionations. Because the overlapping effects of these fractionations are not understood, studies of the D abundance of tree rings led to conflicting results. We hypothesized that both types of fractionations can be separated if the D abundance of individual C-H groups of metabolites can be measured, that is if individual D isotopomers are quantified.

The first paper describes a technique for quantification of D isotopomers in tree-ring cellulose by NMR. The technique showed that the D isotopomers distribution (DID) was non-random. Therefore, the abundance of each isotopomer potentially contains individual information which suggests an explanation for the conflicting results obtained by measuring the overall D abundance (δD).

In the second paper, this technique was used to study hydrogen isotope exchange during cellulose synthesis in tree rings. This revealed that some C-H positions exchange strongly with xylem water, while others do not. This means that the exchanging C-H positions should acquire the D abundance of source water, which is determined by physical fractionations, while non-exchanging C-H positions of tree-ring cellulose should retain biochemical fractionations from the leaf level. Therefore, the abundance of the corresponding D isotopomers should contain information about climate and physiology. When analysing tree-ring series, the DIDs should reflect information about temperature, transpiration and regulation of photosynthesis. In the third paper, we showed that CO2 concentration during photosynthesis

determines a specific abundance ratio of D isotopomers. This dependence was found in metabolites of annual plants, and in tree-ring cellulose. This result shows that D isotopomers of tree-ring series may be used to detect long-term CO2 fertilisation

effects. This information is essential to forecast adaptations of plants to increasing CO2 concentrations on time scales of centuries.

In the fourth paper, the source of persistent organic pollutants in the environment was tracked using DID measurements. The δD values of two compounds of related structures were not enough to show indisputably that they did not originate from the same source. However, the DIDs of the common part between the two compounds proved that they did not originate from the same source. These results underline the superior discriminatory power of DIDs, compared to δD measurements.

The versatility of DID measurements makes them a precious tool in addressing questions that cannot be answered by δD measurements.

Key words: Deuterium, NMR, isotopomers, CO2 response, climate reconstruction,

(8)

Publication List

This thesis is based on the following publications, which will be referred to in the text by their roman numerals:

I - Quantification of deuterium isotopomers in tree-ring cellulose

Tatiana R. Betson, Angela Augusti and Jürgen Schleucher, Analytical

Chemistry, 2006, in press.

II - Hydrogen exchange during cellulose synthesis distinguishes climatic and biochemical isotope fractionations in tree ring,

Angela Augusti, Tatiana R. Betson and Jürgen Schleucher, New

Phytologist, 2006, vol 172, issue 3, p 490-499.

III – Deuterium Isotopomers record a CO2 response of plants in

leaves and tree-rings

Angela Augusti, Tatiana R. Betson and Jürgen Schleucher,

Manuscript.

IV - Baseline isotopic data of polyhalogenated compounds

Walter Vetter, Wolfgang Armbruster, Tatiana R. Betson, Jürgen Schleucher, Thomas Kapp and Katja Lehnert, Analytica Chimica Acta, 2006, vol 577, issue 2, p 250-256.

(9)

Introduction

Most chemical elements of biochemical interest such as hydrogen, carbon, nitrogen and oxygen are composed of different stable isotopes. The abundances of the heavy stable isotopes (2H, also known as Deuterium (D), 13_C,15_N,17_{O and}18_{O) are much smaller} compared to the light stable isotopes (1H, 12C, 14N and 16O) and are modified by physical and biochemical processes. One common use of heavy stable isotopes is to label compounds in a harmless way. This provides a non-invasive tool in medicine to diagnose metabolic disorders (Goetze et al., 2005) and to detect potential disease (Levine

et al., 2004) and in medical studies to better understand human

metabolism (Jin et al., 2004; Ribeiro et al., 2005). Stable isotopes are also widely used to track biogeochemical fluxes through the carbon and nitrogen cycles (Hobbie et al., 2002; Nasholm et al., 1998). In chemistry and biochemistry, labelling is also used in NMR structure determination (Wider & Wuthrich, 1999), to study reaction mechanisms (Ankianiec et al., 1994; Barta et al., 1994; Cook & Cleland, 1981) and to determine kinetic isotope effects (Cleland, 1980).

Because their natural abundances are modified by physical and chemical fractionations, stable isotopes are also a precious tool at natural abundance. They are used to study enzyme mechanisms (Cleland, 2003), metabolic fluxes (Smith & Ziegler, 1990; Zhang et

al., 1995; Zhang et al., 1994) and to determine kinetic isotope effects

(Singleton & Thomas, 1995; Lee et al., 2004). In biogeosciences, stable isotopes at natural abundance are used to integrate biogeochemical fluxes from local to global scale. Prominent examples are the study of the carbon and nitrogen cycles within ecosystems (Ekblad & Hogberg, 2001; Hogberg, 1997) and quantification of sources and sinks of the greenhouse gases CO2, N2O and CH4 (Francey et al., 1995; Dore et al., 1998; Lowe et al., 1994; Quay et al., 1999). They are also an invaluable tool to reconstruct past climate from archives such as ice cores and tree rings (Petit et al., 1999;

(10)

Jouzel et al., 2006; Buhay & Edwards, 1995; Masson-Delmotte et al., 2005). Most of the measurements of stable isotope abundance are made using isotope ratio mass spectrometry (IRMS). This technique measures the ratio of heavy to light isotope of whole molecules. However, it has been shown for all heavy isotopes that their abundance in non-equivalent intramolecular positions is variable (Martin & Martin, 1981; Schmidt, 2003; Robins et al., 2003). Because a molecule carrying a heavy isotope in a particular position is called an isotopomer, this variation can be described by isotopomer distributions. These distributions contain information about the origin of compounds (Zhang et al., 2002). Isotopomer distributions have only been measured by IRMS in a few cases. To measure isotopomer distributions, nuclear magnetic resonance (NMR) is probably the most effective tool. Isotopomer analysis by NMR is used in food authentication (Martin et al., 1988; Remaud et al., 1997), biomechanisms (Zhang et al., 1995) and kinetic isotope effects studies (Lee et al., 2004). But isotopomer measurements are still uncommon compared to measurements of isotope ratios. However, limitations of isotope ratio measurements are becoming more and more apparent. For example, attempts to reconstruct climate data from the D composition of tree-ring cellulose have proved difficult (Waterhouse

et al., 2002; McCarrol & Loader, 2004) despite the strong correlation

between δD of precipitation and climate (Dansgaard, 1964). We propose that the isotopomer distribution of tree-ring cellulose has the potential to provide clearer information. It is therefore necessary to develop new techniques to exploit the information present in isotopomer distributions. In this thesis, a technique was developed to measure the D isotopomers distribution (DID) of tree-ring glucose using NMR. The technique was applied to tree rings, and it was transferred to persistent organic pollutants, demonstrating the versatility of DID measurements in environmental sciences.

(11)

Stable isotopes

General considerations

An element (such as hydrogen, carbon or oxygen) can have several stable isotopes which have the same number of protons and electrons but a different number of neutrons. For example, H is composed of two stable isotopes: protium (1H) and deuterium (2H or D), deuterium having an additional neutron compared to protium (see Table 1 for other examples of stable isotopes). Stable isotopes are non- radioactive.

Element Isotope Neutrons Abundance (%) NMR

1_{H 0 99.985}_Active Hydrogen ₂ H or D 1 0.015 Active 12_{C 6 98.89}_Inactive Carbon ₁₃ C 7 1.11 Active 14_{N 7 99.63}_Active Nitrogen ₁₅ N 8 0.37 Active 16_{O 8 99.759}_Inactive 17_{O 9 0.037}_Active Oxygen 18_{O 10 0.204}_Inactive 32_{S 16 95.00}_Inactive 33_{S 17 0.76}_Active 34_{S 18 4.22}_Inactive Sulfur 36_{S 20 0.014}_Inactive

Table 1. Average terrestrial abundances of the stable isotopes of

major elements of interest in ecological studies alongside the number of neutrons and if the atoms give a signal in NMR.

The stable isotopes of an element are usually referred to as the light and heavy isotope(s) (the light isotope containing the least

(12)

neutrons) as the differing numbers of neutrons give them a different mass. Stable isotopes are naturally present in every compound in different abundances (see Table 1). For the elements listed in Table 1, the heavy isotopes are also rare isotopes, and can be considered as natural tracers in biological systems.

Definitions

The isotope abundance (A) describes the frequency of the heavy isotope of an element and is calculated as follows:

A = L H

H

+ equation 1

with H the number of heavy atoms and L the number of light atoms. The isotope ratio (R) is also a way to describe the frequency of the heavy isotope of an element and is calculated as follows:

R = L H

equation 2

However, it is hard to measure precisely these two values with the common techniques and the instruments available do not guarantee long-term stability. Therefore, the isotope ratio of a sample is usually expressed relative to the isotope ratio of an international reference (such as the Vienna Standard Mean Ocean Water (VSMOW) for D and 18_{O and the Vienna PeeDee Belemnite (VPDB) for}13_{C and} 18_{O) using the following equation:}

δ ( ‰) = 1 1000 R R ref s _× ⎟⎟ ⎠ ⎞ ⎜⎜ ⎝ ⎛ − equation 3

(13)

with δ the isotopic composition of the sample compared to that of the reference and Rs and Rref the isotope ratio of the sample and of the reference, respectively. Because it is generally the difference in isotope ratio between compounds which is used to draw conclusions, the measurements against a reference give very precise data. For example, in the northern hemisphere, the δ13_{C increases between} spring and autumn by about 2‰ due to the discrimination against 13_CO

2 of plants during photosynthesis. In this case, the difference is of interest because it shows the discrimination occurring during photosynthesis.

The isotopic composition of a compound is defined by two types of mechanisms which occur during the synthesis of that compound: physical and chemical fractionations.

Mechanisms of isotope fractionation

The light and the heavy isotope(s) of an element have slightly different physical and chemical properties. The bond between a heavy isotope and another atom is stronger and shorter than between the light isotope and that atom. This explains why a molecule containing a heavy isotope will not behave in exactly the same way as a molecule containing only light isotopes.

1) Physical fractionation

Physical properties such as the melting point and vapour pressure as well as the diffusion rate are altered by the presence of a heavy isotope in a molecule. For example, H2O evaporates faster than HDO because HDO has a slightly lower vapour pressure. This also implies that HDO condenses faster than H2O. This is true for all the heavy isotopes present in H2O (D, 17O and 18O) and explains isotope fractionations in the global hydrological cycle. When water evaporates over the oceans, the atmospheric moisture is depleted in the heavy isotopes. When precipitation forms from this moisture, the remaining

(14)

atmospheric water vapour becomes depleted in the heavy isotopes. When precipitation is formed in isotopic equilibrium with the ambient atmospheric water, the δ of the remaining atmospheric water can be calculated as follows using the Rayleigh law:

δf = δ0 - ε logf equation 4

with δ the stable isotope composition in ‰, ε the equilibrium fractionation factor and f the fraction of water remaining in the system (1> f >0). When freshly evaporated oceanic moisture (f = 1) moves inland or up in latitude, it rains in successive precipitation events (decreasing f) occurring at consecutively lower temperatures as the air becomes dryer. In parallel with this, the heavy isotopes rain out preferentially, and the remaining moisture becomes depleted in heavy isotopes. Therefore, precipitation formed at lower temperatures (higher altitude or in more continental locations) is depleted in heavy isotopes. This explains the influence of altitude, latitude and distance-from-coast observed on the isotopic composition of precipitation (Dansgaard, 1964). At a particular location, the isotopic composition of the precipitation depends mainly on the ambient temperature at the time of the precipitation event (Dansgaard, 1964; Dawson, 1993) and that is why the isotopic composition of ice cores in D has been successfully used as a thermometer to reconstruct past climate, on time scales ranging from years to ice age cycles (Petit et al., 1999; EPICA community members, 2004). The isotopic variation expected from temperature only is around 200 ‰ for δD and 20 ‰ for δ18_O. 2) Chemical fractionation

During a chemical or biochemical reaction, the light isotope of each element involved usually reacts faster, leading to an enrichment of the remaining substrate in the heavy isotope(s). This is what is called a kinetic isotope effect (KIE) and is calculated as follows:

(15)

KIE = L H k k equation 5

where kH and kL are the reaction rate constants when there is a heavy isotope and when there is a light isotope, respectively. If a KIE>1, it is called a “regular (or normal) KIE” and this means that the molecule containing the light isotope reacts faster. Thus, the reaction depletes the product and enriches the remaining reactant in the heavy isotope during the reaction. If a KIE<1, it is called an “inverse KIE” and it means that the molecule containing the heavy isotope reacts faster, resulting in an enrichment of the product (depletion of the reactant) in the heavy isotope during the reaction. There are two particular types of KIE: primary KIEs occur when the reaction involves the scission or the formation of a bond in which the heavy isotope is involved; secondary KIEs take place when a heavy isotope is present in the molecule and influences the reaction even though it is not directly involved in the reaction. The magnitude of both KIEs depends on the activation energy of the reaction and on the structure of the transition state. That is why both primary and secondary KIEs have been extensively used to elucidate chemical and biochemical reaction mechanisms (Cook & Cleland, 1981; Barta et al., 1994; Ankianiec et

al., 1994; Lee et al., 2004). Primary KIEs are the largest. D isotope effects can theoretically be as large as 18, 13C isotope effects 1.25 and 18_{O isotope effects 1.19 (Bigeleisen & Wolfsberg, 1958). Secondary} KIE are much smaller and can theoretically range for example from 0.46 to 1.74 for D and 0.983 to 1.012 for 13C (Bigeleisen & Wolfsberg, 1958). The physical isotope fractionations described above are usually smaller than primary KIEs, and of the same magnitude as secondary KIEs.

In a non-symmetric molecule like glucose, D atoms (and analogously the other heavy isotopes) are introduced into each C-H group by specific reactions with specific D isotope effects. Because of that, the distribution of the light and the heavy isotope(s) of an element in a given compound is non-random and is determined by the synthesis processes of that particular compound (Martin & Martin,

(16)

1999; Schmidt, 2003; Robins et al., 2003). This means that the relative abundance of a heavy isotope at each position in a compound is determined by the (bio)chemical and physical history of that compound.

Isotopomers

A molecule containing a stable isotope is called an isotopomer. In this work, we focused our interest on the stable isotopes of hydrogen. Molecules containing a D atom are called D isotopomers. As an example, a glucose molecule is shown in Figure 1. There are seven C-H groups and therefore glucose can have seven different D isotopomers. The hydrogen atoms within the hydroxyl groups exchange fast with H2O (or other labile H of the environment) and cannot be considered as true isotopomers but only as transient.

Figure 1. Glucose molecule carrying a D atom at carbon 2: D2 isotopomer. Each C-H position can potentially contain a D. Because of the natural rarity of D atoms, a molecule is unlikely to carry two D. Therefore, glucose has seven different mono-deuterated isotopomers.

At natural abundance, the probability of having two D in one molecule is very small because of the low abundance of D (Table 1) and therefore only mono-deuterated isotopomers are considered when dealing with natural compounds of low molecular weight (<1000 Da) (Robins et al., 2003). Each D isotopomer is named depending on which carbon the D atom is bound to. For the carbon 6 of glucose (see Figure 1), the D atom can be at two positions which leads to different configuration (S or R) as the carbon becomes chiral. The isotopomers are then called D6R or D6S isotopomer depending on the configuration they give to the carbon. The abundance of each D isotopomer in a

(17)

compound can be described by the Deuterium Isotopomer Distribution (DID). The DID of glucose (or any other molecule) is the result of all the physical and biochemical fractionations that happened during its synthesis. Measuring the DID (or other heavy isotope isotopomer distribution) therefore contains precious information about the compound synthesis and origin (e.g. Martin et al., 1988; Caer et al., 1991; Remaud et al., 1997; Zhang et al., 2002; Schmidt, 2003; Robins

et al., 2003). If the overall isotope abundance is measured by traditional techniques, this information is, at least partially, lost

Measurements of isotope and isotopomer abundance

Three major techniques are used to measure the natural stable isotope abundance of compounds.

1) Isotope Ratio Mass Spectrometry (IRMS)

IRMS is the standard method to measure δ values. A specially designed mass spectrometer is used to measure accurately the isotope ratio and is able to separate gases with a different isotopic composition such as H2 and HD (or 12CO2 and 13CO2, 16O2 and 18_O16_{O). Complex molecules have to be converted into a gas prior to} measurement. Thus, to measure the isotope ratio in D of a compound, it has to be reacted to yield H2 and therefore only the average D/H ratio of the whole molecule can be measured by established techniques.

To measure isotopomer abundances by IRMS, two methods have been used. First, the chemical breakdown of a compound yields smaller fragments containing only a limited number of atom positions. Each fragment has to be separated and analysed. The isotopomer abundance is obtained at each position by direct measurement (if the fragment contains only one atom position) or by calculation (if the fragment contains several atom positions). To date, several compounds have been successfully measured in this way using IRMS.

(18)

For example, in 1982, Monson and Hayes (Monson & Hayes, 1982) measured the 13C isotopomers of bacterial fatty acids after chemical degradation while Rossmann and Schmidt in 1991 (Rossmann et al., 1991) measured 13C isotopomers of glucose after chemical and biochemical degradation. However, this approach is very laborious and entails a high risk of alteration of the original isotopic composition due to KIEs during the degradation/derivatization steps. Therefore, high yields and strict control of each reaction step are required. Moreover, for each compound, a new derivatization process has to be set up and checked for the absence of isotope effects. The second method used to measure isotopomer abundances was proposed by Corso and Brenna in 1997 (Corso & Brenna, 1997) for 13C isotopomers. In this technique, the fragmentation of the molecule is done by pyrolysis before introduction to the IRMS. The compound is pyrolised into fragments which are separated using gas chromatography and analysed by IRMS. Because the pyrolysis, the gas chromatography separation, the combustion and the IRMS measurement are done in one, coupled apparatus, the method is called “on-line”. It avoids some of the drawbacks of chemical breakdown, but the technique can generate secondary reactions (Corso & Brenna, 1999) which lead to isotope scrambling. Therefore, it does not work for every compound. Another approach was proposed by Brenninkmejer and Rockmann in 1999 (Brenninkmeijer & Rockmann, 1999) to measure 15N isotopomers of N2O. This technique is known as fragment ion IRMS. After ionization in the source of the isotope ratio mass spectrometer, part of the N2O fragments into NO. The mass of N2O and NO are measured and the abundance of the two 15N isotopomers (15N14NO and 14N15NO) calculated. However, this method is special for small molecules like N2O where one fragmentation step can give access to all isotopomers.

Although chemical degradation and pyrolysis can be used to measure 13C isotopomers, these techniques cannot be adapted to quantify D isotopomers as the risk of isotope effects during degradation and isotope scrambling during pyrolysis is much greater

(19)

than with 13C. The fragment ion IRMS uses a particular property of N2O and is restricted to that particular compound. Moreover, the high risk of isotope scrambling would not make this technique suitable for D isotopomer quantification.

2) Fourier transform infrared spectroscopy

Fourier Transform Infrared Spectroscopy (FTIR) is a powerful tool for identifying types of chemical bonds in a molecule by producing an infrared absorption spectrum that is like a molecular "fingerprint". The technique basically measures the vibration frequencies of all bonds in the molecule. The presence of a heavy isotope in a bond will change slightly the characteristics of that bond and the higher mass of the heavy isotope changes its vibration frequency significantly. It is therefore possible to identify the presence of isotopomers in a compound and to quantify them. For example, ozone isotopomers have been successfully measured using FTIR (Christensen et al., 1996). However, all applications to date required gaseous samples and the technique may be fundamentally limited to small molecules, because the spectra of biochemical metabolites like glucose may be too crowded to resolve the signal of each isotopomer. 3) Nuclear Magnetic Resonance Spectroscopy (NMR)

NMR is able to measure individual isotopomers in parallel in a single compound. It is based on the property of certain nuclei to have a magnetic moment. In a magnetic field, these nuclei acquire a magnetization parallel to the magnetic field. A quick and intense radio frequency pulse (µs range and about 100W power) turns the magnetization of one selected nucleus perpendicular to the magnetic field direction. In the perpendicular plane, the magnetization then rotates at a frequency characteristic for the nucleus and depending on the magnetic field strength. In non-symmetric molecules, the frequency is modified to a small degree by the chemical environment

(20)

of each intramolecular position and is therefore called a chemical shift. The rotation frequencies of all atoms in a sample are recorded and the signal (so-called Free Induction Decay (FID)) Fourier-transformed. Because only frequency differences between the atoms in the molecule are of chemical interest, they are expressed relative to a reference compound (usually tetramethylsilane (TMS) for hydrogen, D and 13C). Because the deviation from the reference is small, the chemical shifts are expressed in ppm (parts per million) (see Figure 2). The chemical shift of each atom gives information about its chemical environment which depends on its position in the molecule (as an example see Figure 2a). This means that it is possible to obtain one signal per isotopomer in a spectrum (Figure 2b). NMR can also distinguish both H positions of prochiral CH2 groups. Using appropriate measurement parameters, the intensity of each NMR signal is proportional to the number of atoms sharing the same environment, which means the number of atoms at a particular position. The signal integrals can then be obtained by integration, which is best done by a lineshape fit (“deconvolution”). In Figure 2a, there is a 1H spectrum of 2,4-dibromoanisole. In this spectrum, each integral reflects the number of 1H in each position. Because the abundance of 1H is almost 100%, the integrals are not modified by the presence of isotopomers. But when looking at the 2H (or D) spectrum of the same compound (Figure 2b), the four peaks corresponding to the four different H positions do not have an integral which only depends on the number of equivalent H at that position. The abundance of 2H at each position actually differs slightly due to D fractionations which occurred during the synthesis of the 2,4-dibromoanisole. The integral of the peaks is directly proportional to the abundance of the isotopomers. Thus, NMR quantifies the abundance ratios of all isotopomers of a particular compound in a single measurement: it measures the isotopomer distribution of a compound. Unlike for IRMS, it is not possible to measure the sample and a standard in an alternating fashion, to measure the isotope abundance difference between the two.

(21)

Figure 2. a) 1H spectrum and b) 2H spectrum of 2,4-dibromoanisole (formula displayed). Both spectra show a peak for each of the four different H positions (A to D) of the compound and their relative integral is displayed. In the 1H spectrum, the integrals reflect the number of equivalent H at each position and the peaks are split due to the presence of neighbouring 1H. However, in the 2H spectrum, the integral reflects the relative abundance of each monodeuterated (no splitting) isotopomer. The chemical shift is expressed in ppm relative to TMS (see text) and in MHz.

(22)

This means that NMR cannot measure the isotopic ratio at each position unless a standard of known isotope ratio can be introduced in a known molecular ratio. Further, the sample should be pure to avoid peak overlap between the compound of interest and possible contaminants.

NMR can only work with NMR-active stable isotopes (see Table 1) which includes D and 13C but not 18O, for example. NMR analysis requires bigger samples (50 mg for NMR against 5 µg for IRMS for 13_{C measurements) and is about 2 to 10 times less precise than IRMS} analysis. Still, it is particularly suitable to quantifying D isotopomers since the large D KIEs during the synthesis of a compound leads to large isotopomer abundance variations, which are invisible in δD measurements by IRMS. DID measurements by NMR have been successfully used for many years in food authentication (Martin et al., 1988; Remaud et al., 1997), studies of biosynthesis in plants (Zhang et

al., 2002; Schleucher et al., 1999) and humans (Jin et al., 2004; Ribeiro et al., 2005) and in the determination of KIEs to elucidate reaction mechanisms (Grant et al., 1982). The DID of various natural compounds (e. g. carbohydrates, aromatic compounds, fatty acids, terpenes and alcohols) showed a large variation (typically ± 20%) which is expected from the size of KIEs. However, this variation is larger than the δD variation expected from climate (± 10%). This means that measuring δD on organic archives to reconstruct past climate can be seriously hindered by the KIEs that occurred during the archives synthesis. Instead, DID has a higher content of information which grants a higher chance of accessing data about climate and biochemical processes in parallel. In principle, 13C isotopomers quantification by NMR can be used in much the same way as that of D to determine KIEs during chemical and enzymatic reactions (Singleton & Thomas, 1995; Lee et al., 2004) and in food authentication (Caer et al., 1991; Tenailleau et al., 2004). However, quantification of 13C isotopomers by NMR is much more demanding compared to D, because 13C KIEs are smaller, while error sources for quantification of 13C NMR spectra are much larger.

(23)

Climate change

Open questions

The atmospheric CO2 concentration ([CO2]) has varied over the history of the Earth (IPCC, 2001) and, during the successive warm and cold periods, [CO2] and temperature were strongly coupled. During the last 200 years, the atmospheric [CO2] has increased at an abnormal speed mainly due to the large emissions of CO2 from anthropogenic activities. At present, the atmospheric [CO2] is 380 ppm, a level unprecedented for the past 23 millions years (Pearson & Palmer, 2000). Due to the greenhouse effect of CO2, the atmospheric [CO2] is regarded as a likely forcing mechanism on global climate because of its large and predictable effect on temperature (Crowley, 2000). In the last 200 years, the excess of CO2 release due to anthropogenic activities has contributed to a global rise in temperature that is forecasted to last into the coming centuries. The rise in annual global temperature leads to the rise of sea level because of ocean expansion and ice melting at the poles and at mountain glaciers. It also leads to the loss of numerous plant and animal species (Thomas et al., 2004; Pounds & Puschendorf, 2004) and is suspected to generate more frequent climate extremes, such as heavy precipitation events (floods), storms and extreme temperature (heat waves) (IPCC, 2001). To forecast the possible influence of the rising atmospheric [CO2] in the coming century, the amplitude of the rise has to be estimated. Unfortunately, such estimates are hampered by several uncertainties concerning possible CO2 sources and sinks. The different scenarios forecast an atmospheric [CO2] of 550 to 1000 ppm for the end of the 21st century, with predicted concomitant temperature rises between 1.8 to 6ºC. To refine climate models and predictions, it is important to address the remaining uncertainties. One such uncertainty is if vegetation will be a stronger sink of CO2 than at present. This would occur if increasing atmospheric [CO2] augments global plant productivity (“CO2 fertilisation effect”). To answer this question,

(24)

manipulative experiments have been conducted by exposing plants to higher CO2 levels and recording the plants’ response. But these experiments covered only short time scales (up to 10 years) (Long et

al., 2004) and may observe transient responses that do not hold over longer time scales. It is therefore difficult to extrapolate conclusions to century time scales. To cover such a time span, it is relevant to study the response of vegetation over centuries to the past variation in atmospheric [CO2], especially during the last 200 years.

Another uncertainty concerns the high frequency climate variability in the past. Available instrumental records span the last 200 years at best and cover only a limited geographical area. To access past climate data on longer time scale covering most of the surface of the world, proxy records have to be used. However, many proxy records have low time resolution and therefore only show long-term variability (i.e. averaged over 5-10 years or up to 50 years) (Feng & Epstein, 1994; Feng & Epstein, 1996; Libby et al., 1976; Petit et al., 1999). To refine climate models, climatic data with a high resolution over a long time scale are needed.

To address both uncertainties, tree rings are designated archives. They may contain a record of plants’ response to increasing atmospheric [CO2] but also of climate. They have a wide geographical distribution and tree rings span centuries with a very high time-resolution. That is why we aimed to develop a tool to extract such precious information.

Tree rings as a proxy for climate (dendroclimatology)

Trees are interesting archives of environmental changes. They synthesize a new ring of wood in their trunk each year, often with a distinct difference between wood formed in spring and summer. They are long-lived plants, have the same photosynthetic pathway (C3 species) as most plants and are wide-spread geographically. Also, their physiology is influenced by environmental factors such as temperature and atmospheric [CO2]. Therefore, they can contain

(25)

chronological data about climate covering century-time scales with a yearly resolution or even seasonal on a global scale. The study of the relationship between tree-ring characteristics and climate is called dendroclimatology. It exploits either the physical or chemical characteristics of the tree rings. Because many trees have experienced the unusually fast increase in atmospheric [CO2] in the past two centuries, it is interesting to study their physiological response to such a change. This should help to understand the long-term interaction between the biosphere and atmosphere.

1) Physical characteristics of tree rings

Wood density and tree-ring width depend on environmental conditions to which the tree is exposed during growth. Since the factors influencing the growth of the tree are numerous, trees for dendroclimatological studies have to be carefully selected so that their growth is mainly influenced by one factor (usually temperature or precipitation). The careful selection of trees avoids background noise from other environmental factors (Fritts, 1976). For example, the trees of the northern hemisphere at the tree line are sensitive to temperature and have therefore been used to reconstruct past temperature (Jacoby & Darrigo, 1989). This technique has also been applied in the southern hemisphere to trees in New Zealand (Palmer & Xiong, 2004). A recent work by Briffa et al. (Briffa et al., 2002) used a large scale sampling around the globe to reconstruct the temperature over the last 600 years, mainly using late wood density (i.e. formed during summer). The results of such studies have been used as key evidence to demonstrate the anthropogenic effect on global warming during the last millennium. However, it can be difficult to separate the effect of each influence on the tree’s growth and therefore to get clear climate data (Mann et al., 1999).

The growth of the tree has also been used in some studies (DeLucia et al., 1999; Oren et al., 2001) to monitor the effect of atmospheric [CO2] on tree physiology. However, these studies were

(26)

manipulative experiments covering only a few years and, under natural growth conditions, it is difficult to isolate the trees’ response to the increase of [CO2] over century time scales.

2) Chemical characteristics of tree rings

Wood is mainly composed of cellulose formed from primary carbohydrates synthesised from CO2 and H2O during photosynthesis at leaf level. Therefore, the isotopic composition of that CO2 and H2O influences the isotopic composition of the leaf carbohydrates. Part of these carbohydrates is transferred to the trunk where they enter tree- ring synthesis. Consequently, hydrogen, carbon and oxygen stable isotopes of tree rings have been used to reconstruct environmental factors such as temperature, precipitation, relative humidity and to study the water use efficiency as a response to atmospheric [CO2] increase (for a complete review see McCarrol & Loader, 2004). δD should be a very good indicator of ambient temperature (Dansgaard, 1964) and has been used to reconstruct past climate over millennia from tree rings (Feng & Epstein, 1994). However, to improve the correlation with temperature, the δD is measured or pooled to average several years (5 to 50 years) and thus only reflects low frequency climate variability. Moreover, most work attempting to link tree-ring isotope ratios with climate has used 13C or 18O isotopes because several studies have found δD measurements on tree-ring cellulose hard to interpret (Lipp et al., 1991; Waterhouse et al., 2002). A likely explanation for the difficulties encountered in δD-based climate reconstructions lies in D fractionations happening within the tree. These fractionations alter the isotopic composition of the metabolites compared to the environmental source, but could not be separated in past studies and were assumed to be constant. To describe the D composition of tree-ring cellulose, we use a model that includes four distinct mechanisms, illustrated in Figure 3. We assume that, when separating climate and physiological influences in this way, the

(27)

isotopic composition of tree rings can yield information about both environmental factors and tree physiology.

The first mechanism is the D fractionation in the hydrological cycle which confers an air temperature signal to the δD of precipitation water (Dansgaard, 1964) and influences the δD of photosynthates.

Second, leaf transpiration discriminates against water molecules containing heavy isotopes, which enriches the remaining leaf water in D. Leaf transpiration and leaf water D enrichment increase with decreasing air humidity. Therefore, the leaf water contains a humidity signal which is passed on to photosynthates during photosynthesis.

Third, enzyme isotope effects reduce the abundances of specific D isotopomers of photosynthates. If the D isotopomer fractionations due to enzyme isotope effects can be identified in tree rings, the size of the fractionations may carry physiological signals.

Fourth, enzyme-catalysed isotope exchange occurs between C-H and xylem sap water during cellulose synthesis (C-Hill et al., 1995; Roden & Ehleringer, 2000). This exchange can overwrite the effects of all leaf-level processes in the trunk. It also re-introduces the temperature signal into tree-ring cellulose because xylem sap water has the same δD as precipitation water (first mechanism).

If the δD of tree-ring cellulose is measured, only the average D isotopomer abundance is accessed. Therefore, the D fractionations due to the four mechanisms cannot be separated, and possible corresponding signals cannot be isolated. This thesis describes an approach to separate the mechanisms by D isotopomer measurements. If each signal can be separated, it provides the unique opportunity to access parallel information about past climate and tree physiology.

(28)

Figure 3. Scheme of the four different, successive mechanisms that influence the hydrogen isotope composition of tree-ring cellulose.

(29)

Persistent organic pollutants in the environment

Definition

Persistent organic pollutants (POPs) are organic compounds that are resistant to environmental degradation through chemical, biological, and photolytic processes. Because of this, they have been observed to persist in the environment, to be capable of long-range transport and bioaccumulation in human and animal tissue, and to have potentially significant impacts on human health and the environment (Ritter et al., 1995; Bernes & Naylor, 1998). Some POPs are polyhalogenated compounds (PHC) often composed of aromatic rings substituted with halogens such as chlorine and bromine. They include chloropesticides (e.g. DDT, lindane, toxaphene, and chlordane), polychlorinated biphenyls (PCBs) and brominated flame retardants (e.g. polybrominated diphenyl ester). Chloropesticides have been used as potent pesticides which presented the advantage of being cheap to produce and effective on a long-term as their solubility in water is limited. Brominated flame retardants are used in many household items, such as carpets, computer plastic covers, cables, etc… to prevent ignition and hinder the spread of fire, saving the life of thousands of people every year. Both groups of compounds and PCBs contain halogens to increase their chemical stability and to lower their water solubility, which makes their degradation very slow in the environment and facilitates bioaccumulation in top predators. Because of the persistence of POPs in the environment, these compounds accumulate over long time periods and eventually reach levels that have adverse effects on the fauna such as toxic effects (anemia), neurotoxic effects (behavorial alterations), and interferences with hormones (impaired reproduction).

(30)

Identification of their origin and fate

Following the fate of POPs in the environment can help to understand how and at what speed they can be eliminated. It is also important to understand how these compounds move in the environment. Moreover, some closely-related compounds are also produced by natural sources which can lead to the identification of structural features that make a compound degradable or not. Furthermore, it is interesting to attribute the source/origin of POPs to identify the polluters because in some countries the polluters have to clean contaminated sites. It can also reveal the misuse and/or illegal sale of remaining stocks. To follow and to determine the origin of these compounds, isotope ratios are frequently used, because for an isolated compound, they are the only variables that can identify a source. Previously, only the isotope ratios δ13_{C and δ}37_{Cl have been} used (Vetter et al., 2005; Reddy et al., 2000) because the δD of polyhalogenated compounds could not be measured online by IRMS. Recently, however, a new technique was developed to measure δD as well (Armbruster et al., 2006). Assuming that isotope ratios are not changed by transport processes, they can distinguish the same compound from different sources. But when a compound is converted, the isotope ratios of the starting material and the structurally different product cannot be compared any longer. Thus, closely related compounds can have strongly different δ values. In this case, measuring the DID of these compounds using D NMR allows us to compare isotopomer abundances of their common parts and complement IRMS in the identification of their origin.

(31)

Results and conclusions

The research described in this thesis provides an important basis for the use of D isotopomer measurements by NMR as a tool in environmental research. In this section, the main results and conclusions of papers I to IV are reviewed.

Paper I

The D abundance of tree-ring cellulose has been hard to interpret, presumably because of variation in D isotopomer abundance. We expected this variation to originate from signals reflecting tree physiology and climate. We assumed that these signals may be separated if the DID is measured. However, wood samples cannot be measured by NMR without prior solubilisation, derivatization and purification. This paper describes a technique to prepare a glucose derivative from wood and measure the DID of the glucose units. We showed that the preparation technique did not alter the isotopic content of the original material and that the measurement conditions led to quantitative measurements. Using this technique, we found a common DID pattern in leaf glucose of annual plants and in tree-ring glucose, which demonstrates that the DID of tree rings retains signals stemming from enzyme D fractionations on the leaf level. The development of this technique was the first step in separating and extracting information from D in tree rings.

Paper II

The aim of this paper was to understand if individual isotopomers of tree-ring glucose contain specific signals related to climate or physiology. At the outset of this research, it had been established that there is hydrogen isotope exchange between C-H groups and trunk water during cellulose synthesis. We reasoned that this exchange should not be equal for each C-H group and that the

(32)

level of exchange would reveal which isotopomer would carry which signal. To quantify the exchange for each C-H group, we grew two tree species under controlled conditions and used D-labelled water to track the exchange during cellulose synthesis. We compared the DID of leaf glucose to the DID of tree-ring glucose to identify the exchanging positions. We showed that different C-H groups exchange to strongly variable degrees, with the H at carbon 2 exchanging most and the two H at carbon 6 least. The large H exchange at carbon 2 should imprint the temperature signal present in precipitation on the abundance of the D2 isotopomer, irrespective of all leaf-level processes. In contrast, the abundances of the two D6 isotopomers should be transmitted from leaf-level metabolites to tree rings, and can store signals on leaf-level processes in tree rings. The results of this paper showed that D isotopomer measurements open a new way of using D of tree rings to reconstruct climate and to study long-term physiological adaptations of plants to environmental changes.

Paper III

In this work, we aimed to identify a potential response of plants to an increase in atmospheric [CO2] recorded in the DID. We analysed glucose from three classes of carbohydrates (soluble sugars, starch and cellulose) extracted from annual plants of different species, as well as from tree rings. The plants were subjected to different [CO2], ranging from the pre-industrial to forecasted atmospheric levels. The results showed that the ratio of the two D6 isotopomers’ abundance was correlated with the [CO2] to which the plant was exposed. We proposed that the dependence of this isotopomer abundance ratio on [CO2] can be used to estimate the proportion of photorespiration to photocarboxylation. Because we had demonstrated in paper II that the D6 isotopomers exchanged the least with xylem sap water during cellulose synthesis in the trunk (paper II), we expected the CO2 dependence to be transmitted (slightly dampened) to tree-rings. The results obtained from tree-ring samples confirmed this hypothesis.

(33)

Applied to tree-ring series, we expect DID measurements to provide a unique tool to access information about the physiological response of trees (and vegetation) to atmospheric [CO2] increase during the last two centuries, or during glacial-interglacial cycles. Such information should give insights about how the global vegetation will respond to increasing [CO2] in the coming century, which will improve climate predictions.

Paper IV

In this paper, NMR measurements are used as an accompanying technique for IRMS measurements. Polyhalogenated compounds (PHCs) are persistent organic pollutants in the environment. Isotope measurements are used to identify the origin and degradation rate of the PHCs, but whole-molecule isotope ratios (δ values) are of limited usefulness if related but structurally different compounds must be compared. Therefore, we measured the DIDs of two compounds (2,4-dibromophenol and 2,4-dibromoanisole) that were related structurally, but showed differing δ13_{C and δD. Because of the structural difference} between the compounds, the differences in δ values cannot prove that the two compounds originate from different sources. In contrast, the DID measurements showed that the common part of the two compounds had different DIDs, proving that they could not originate from a common synthesis pathway. This result illustrates the potential of DID measurements in tracking the origin and affiliations among environmental pollutants. This also demonstrates that DID measurements can be easily adapted to different compound classes and different research areas.

Unpublished results related to paper IV

DDT is a well-known insecticide, and was one of the first compounds where accumulation in food chains and environmental toxicity have been observed. However, there is currently renewed

(34)

interest in its use, because it is one of the few effective agents to fight malaria cheaply. For these reasons, monitoring of its use and knowledge of its turnover in the environment are essential for its safe application. We therefore compared DIDs of DDT (dichloro-diphenyltrichloroethane) and the structurally related by-product DDD (dichlorodiphenyl-dichloroethane). While the two compounds were assumed to originate from a common synthesis process, IRMS measurements showed a difference in δD values of 80 ‰ (on the VSMOW scale). This difference was considered incompatible with a common source, because DDD contains only one additional hydrogen compared to nine hydrogens which are common for both compounds (see Figure 4).

Figure 4. Chemical structure of p,p’-DDT and o,p’-DDD. In DDD, one hydrogen is present in the aliphatic part of the molecule.

To explain the δD difference, the additional hydrogen would need to have a δD of approximately +800 ‰, corresponding to nearly twice the natural abundance of D, while typical δD variation is ±100 ‰. Using general rules for the recording of high-quality D NMR spectra presented in paper I, we measured the DIDs of both compounds. Combined with the δD values of the whole molecules, we calculated δD values of each C-H group of DDT and DDD. We found a value of +800‰ for the additional H in DDD, while the rest of the C-H groups showed similar δD values for DDT and DDD. This result proved that the two compounds could originate from a common reaction although their δD values were very different. This illustrates the potential of combining IRMS and NMR measurements in determining the origin of persistent pollutants in the environment.

(35)

General Conclusion

This thesis work focused on developing a technique to measure the DID of tree-ring glucose in order to access information regarding tree physiology and past climate that is inaccessible with conventional techniques. The first results, applying the technique to tree rings from a CO2 enrichment site, support our hypothesis that DID measurements on tree rings can retrieve information about physiological response to environmental changes on long time scales. This information is of prime interest to improve current climate models.

DID measurements are not limited to tree ring studies and we easily adapted the technique to determine the origin of persistent pollutants in the environment. This demonstrates that DID measurements can be a very powerful tool in environmental research. In the future, we expect the use of DID measurements to become more common, addressing questions that could not be answered with previously available techniques.

(36)

Acknowledgements

I would like to thank everybody at the department of Medical Biochemistry and Biophysics for the pleasant working environment. Some special thanks are also in order to:

Jürgen Schleucher, my supervisor, for coming up with the project ideas, having always new interesting ideas to propose and for all the animated work discussions. And also for guiding me trough my Ph D work when I sometimes felt lost or letting me try things after warning to learn by myself. I also want to thank you for all the cakes, waffles and other food experiences and for simply being a very nice person and friend. I have learned a lot with you, both from a professional and personal point of view.

Angela Augusti, my dear friend and co-worker. Life in Umeå and in the lab would just not have been the same without you. You are a truly wonderful person and I am glad to have shared all these moments in and out of work with you. And how could we have survived lab accidents without each other?? I wish you the best for your own thesis defence.

The current NMR group: Aster, Göran, Katja, Linus, Peter, Tobias T. (and Gustav); and the previous members: Bin, Elke, Frédéric, Nora, Sara, Shintaka, Sybren, and Tobias S. for such a friendly working environment.

Our collaborators, but especially Peter Högberg for the support in this project; Marc Filot and Markus Leuenberger for a wonderful time in Bern, both from a personal and professional point of view; and Walter Vetter for a fruitful and friendly collaboration.

(37)

Ingrid Råberg and Urban Backman for their incredibly precious help in filling administrative papers and in organising the teaching, respectively.

Gerhard and Burkhard for their friendship and all the nice outdoor activities shared.

The Umeå French Mafia (Catherine, Charleen, Christine, Johan, Julien, Laurent, Marco, Nathalie, Olivier et Pauline) pour être toujours au rendez vous et pour les soirées passées à s’interroger sur les moeurs suédoises qui nous laissent parfois (souvent) perplexes. Grosses bises à tous.

Merci aussi à toute ma famille pour leur support moral tout au long de ces années passées loin d’eux. Merci en particulier à ma maman qui ne comprend toujours pas ce que je fais mais qui m’a toujours fait confiance et supportée moralement dans mes décisions. Je vous aime tous très fort.

My friends out of University in Umeå (Heather, François, Maria, Karin and Simeon) and in France (Carole, Emmanuelle, Johannes, Sylvie et Alexandre, Véronique et Philippe, Bruce et Jean-Marie) for their support and friendship.

Last but not least, Nicholas Betson, my beloved husband for his every day support, his indestructible good mood but also for his useful comments and our long discussions about work. Without you I would feel that something is missing in my life. Je t’aime mon petit Loulou.

Tatiana

This thesis work was financially supported by the Centre for environmental research in Umeå (CMF) but also grants from the Kempe and the Wallenberg foundations, the Swedish Royal Academy of Science (KVA) and the European program SIBAE.

(38)

References

Ankianiec BC, Christou V, Hardy DT, Thomson SK, Young GB.

1994. Mechanisms of thermolytic rearrangement of cis-bis(silylmethyl)platinum(ii) complexes - beta-carbon transfer predominates over hydrogen transfer. Journal of the American

Chemical Society 116: 9963-9978.

Armbruster W, Lehnert K, Vetter W. 2006. Establishing a chromium-reactor design for measuring δ2H values of solid polyhalogenated compounds using direct elemental analysis and stable isotope ratio mass spectrometry. Analytical and Bioanalytical

Chemistry 384: 237-243.

Barta NS, Kirk BA, Stille JR. 1994. Alpha-deuterium and beta-deuterium isotope effects in the Mgx(2) and methylaluminoxane promoted intra-molecular olefin insertion of Cp(2)ticlr complexes - insight into co-catalyst dependence and chain end control in ziegler-natta polymerization. Journal of the American Chemical Society 116: 8912-8919.

Bernes C, Naylor M. 1998. Persistent organic pollutants: a swedish

view of an international problem. Stockholm: Swedish Environmental Protection Agency.

Bigeleisen J, Wolfsberg M. 1958. Theoretical and experimental aspects of isotope effects in chemical kinetics. Advances in Chemical

Physics 1: 15-76.

Brenninkmeijer CAM, Rockmann T. 1999. Mass spectrometry of the intramolecular nitrogen isotope distribution of environmental nitrous oxide using fragment-ion analysis. Rapid Communications in

Mass Spectrometry 13: 2028-2033.

Briffa KR, Osborn TJ, Schweingruber FH, Jones PD, Shiyatov SG, Vaganov EA. 2002. Tree-ring width and density data around the Northern Hemisphere: Part 1, local and regional climate signals.

Holocene 12: 737-757.

Buhay WM, Edwards TWD. 1995. Climate in south-western Ontario, Canada, between AD 1610 and 1885 inferred from oxygen and

(39)

hydrogen isotopic measurements of wood cellulose from trees in different hydrologic settings. Quaternary Research 438-446.

Caer V, Trierweiler M, Martin GJ, Martin ML. 1991. Determination of site-specific carbon isotope ratios at natural abundance by 13C nuclear magnetic resonance spectroscopy.

Analytical Chemistry 63: 2306-2313.

Christensen LK, Larsen NW, Nicolaisen FM, Pedersen T, Sorensen GO, Egsgaard H. 1996. Far-IR spectroscopy of ozone as a means of quantification of ozone isotopomers. Journal of Molecular

Spectroscopy 175: 220-233.

Cleland WW. 1980. Measurements of isotope effects by the equilibrium perturbation technique. In: Purich, DL, ed. Methods in

Enzymology. vol. 64. New York: Academic Press, 104-125.

Cleland WW. 2003. The use of isotope effects to determine enzyme mechanisms. Journal of Biological Chemistry 278: 51975-51984.

Cook PF, Cleland WW. 1981. Mechanistic deductions from isotope effects in multi-reactant enzyme mechanisms. Biochemistry 20: 1790-1796.

Corso TN, Brenna JT. 1997. High-precision position-specific isotope analysis. Proceedings of the National Academy of Sciences of the

United States of America 94: 1049-1053.

Corso TN, Brenna JT. 1999. On-line pyrolysis of hydrocarbons coupled to high-precision carbon isotope ratio analysis. Analytica

Chimica Acta 397: 217-224.

Crowley TJ. 2000. Carbon dioxide and Phanerozoic climate. In: Huber BT, MacLeod KG, Wing SC, eds. Warm Climates in Earth

History. Cambridge: Cambridge University Press, 425-444.

Dansgaard W. 1964. Stable isotopes in precipitation. Tellus 16: 436-468.

Dawson TE. 1993. Water sources of plants as determined from xylem-water isotopic composition: perspectives on plant competition, distribution, and water relations. In: Ehleringer JR, Hall AE, Farquhar

(40)

GD, eds. Stable isotopes and plant carbon-water relations. New York: Academic Press, 465-496.

DeLucia EH, Hamilton JG, Naidu SL, Thomas RB, Andrews JA, Finzi A, Lavine M, Matamala R, Mohan JE, Hendrey GR, Schlesinger WH. 1999. Net primary production of a forest ecosystem with experimental CO2 enrichment. Science 284: 1177-1179.

Dore JE, Popp BN, Karl DM, Sansone FJ. 1998. A large source of atmospheric nitrous oxide from subtropical North Pacific surface waters. Nature 396: 63-66.

Ekblad A, Högberg P. 2001. Natural abundance of 13C in CO2 respired from forest soils reveals speed of link between tree photosynthesis and root respiration. Oecologia 127: 305-308.

EPICA community members. 2004. Eight glacial cycles from an Antarctic ice core. Nature 429: 623-628.

Feng XH, Epstein S. 1994. Climatic implications of an 8000 year hydrogen isotope time series from bristlecone pine trees. Science 265: 1079-1081.

Feng XH, Epstein S. 1996. Climatic trends from isotopic records of tree rings: The past 100-200 years. Climatic Change 33: 551-562.

Francey RJ, Tans PP, Allison CE, Enting IG, White JWC, Trolier M. 1995. Changes in oceanic and terrestrial carbon uptake since 1982.

Nature 373: 326-330.

Fritts HC. 1976. Tree rings and climate. London: Academic Press.

Goetze O, Wieczorek J, Mueller T, Przuntek H, Schmidt WE, Woitalla D. 2005. Impaired gastric emptying of a solid test meal in patients with Parkinson's disease using 13_{C-sodium octanoate breath} test. Neuroscience Letters 375: 170-173.

Grant DM, Curtis J, Croasmun WR, Dalling DK, Wehrli FW, Wehrli S. 1982. NMR determination of site-specific deuterium isotope effects. Journal of the American Chemical Society 104: 4492-4494.

(41)

Hill SA, Waterhouse JS, Field EM, Switsur VR, Rees TAP. 1995. Rapid recycling of triose phospates in oak stem tissue. Plant, Cell and

Environment 18: 931-936.

Hobbie EA, Tingey DT, Rygiewicz PT, Johnson MG, Olszyk DM.

2002. Contributions of current year photosynthate to fine roots estimated using a 13C-depleted CO2 source. Plant and Soil 247: 233-242.

Hogberg P. 1997. Tansley review No 95: 15N natural abundance in soil-plant systems. New Phytologist 137: 179-203.

IPCC. 2001. Climate change 2001: The scientific basis. Contribution

of Working Group I to the Third Assessment Report of the Inter-goverrnmental Panel on Climate Change. Cambridge: Cambridge University Press.

Jacoby GC, Darrigo R. 1989. Reconstructed northern hemisphere annual temperature since 1671 based on high-latitude tree-ring data from North-America. Climatic Change 14: 39-59.

Jin ES, Jones JG, Merritt M, Burgess SC, Malloy CR, Sherry AD.

2004. Glucose production, gluconeogenesis, and hepatic tricarboxylic acid cycle fluxes measured by nuclear magnetic resonance analysis of a single glucose derivative. Analytical Biochemistry 327: 149-155.

Jouzel J, Lorius C, Raynaud D. 2006. Quaternary climate and atmospheric composition: new ice cores. Comptes Rendus Palevol 5: 45-55.

Lee JK, Bain AD, Berti PJ. 2004. Probing the transition states of four glucoside hydrolyses with 13C kinetic isotope effects measured at natural abundance by NMR spectroscopy. Journal of the American

Chemical Society 126: 3769-3776.

Levine A, Shevah O, Miloh T, Wine E, Niv Y, Bujanover Y, Avni Y, Shirin H. 2004. Validation of a novel real time 13C urea breath test for rapid evaluation of Helicobacter pylori in children and adolescents.

The Journal of Pediatrics 145: 112-114.

Libby LM, Pandolfi LJ, Payton PH, Marshall J, Becker B, Giertz-Sienbenlist V. 1976. Isotopic tree thermometers. Nature 261: 284-288.

(42)

Lipp J, Trimborn P, Fritz P, Moser H, Becker B, Frenzel B. 1991. Stable isotopes in tree ring cellulose and climatic change. Tellus 43B: 322-330.

Long SP, Ainsworth EA, Rogers A, Ort DR. 2004. Rising atmospheric carbon dioxide: plants FACE the future. Annual Review

of Plant Biology 55: 591-628.

Lowe DC, Brenninkmeijer CAM, Brailsford GW, Lassey KR, Gomez AJ, Nisbet EG. 1994. Concentration and 13C records of atmospheric methane in New Zealand and Antarctica - Evidence for changes in methane sources. Journal of Geophysical

Research-Atmospheres 99: 16913-16925.

Mann ME, Bradley RS, Hughes MK. 1999. Northern hemisphere temperatures during the past millennium: Inferences, uncertainties, and limitations. Geophysical Research Letters 26: 759-762.

Martin GJ, Guillou C, Martin ML, Cabanis MT, Tep Y, Aerny J.

1988. Natural factors of isotope fractionation and the characterization of wines. Journal of Agricultural and Food Chemistry 36: 316-322.

Martin GJ, Martin ML. 1981. Deuterium labelling at natural abundance level as studied by high field quantitative 2H NMR.

Tetrahedron Letters 22: 3525-3528.

Martin ML, Martin GJ. 1999. Site-specific isotope effects and origin inference. Analusis 27: 209-213.

Masson-Delmotte V, Raffalli-Delerce G, Danis PA, Yiou P, Stievenard M, Guibal F, Mestre O, Bernard V, Goosse H, Hoffmann G, Jouzel J. 2005. Changes in European precipitation seasonality and in drought frequencies revealed by a four-century-long tree-ring isotopic record from Brittany, western France. Climate

Dynamics 24: 57-69.

McCarrol D, Loader NJ. 2004. Stable isotopes in tree rings.

Quaternary Science Reviews 23: 771-801.

Monson KD, Hayes JM. 1982. Carbon isotopic fractionation in the biosynthesis of bacterial fatty acids. Ozonolysis of unsaturated fatty

(43)

acids as a means of determining the intramolecular distribution of carbon isotopes. Geochimica et Cosmochimica Acta 46: 139-149.

Näsholm T, Ekblad A, Nordin A, Giesler R, Högberg M, Högberg P. 1998. Boreal forest plants take up organic nitrogen. Nature 392: 914-916.

Oren R, Ellsworth DS, Johnsen KH, Phillips N, Ewers BE, Maier C, Schafer KVR, McCarthy H, Hendrey G, McNulty SG, Katul GG. 2001. Soil fertility limits carbon sequestration by forest ecosystems in a CO2-enriched atmosphere. Nature 411: 469-472.

Palmer JG, Xiong LM. 2004. New Zealand climate over the last 500 years reconstructed from Libocedrus bidwillii Hook. f. tree-ring chronologies. Holocene 14: 282-289.

Pearson PN, Palmer MR. 2000. Atmospheric carbon dioxide concentrations over the past 60 million years. Nature 406: 695-699.

Petit JR, Jouzel J, Raynaud D, Barkov NI, Barnola JM, Basile I, Bender M, Chappellaz J, Davis M, Delaygue G, Delmotte M, Kotlyakov VM, Legrand M, Lipenkov VY, Lorius C, Pepin L, Ritz C, Saltzman E, Stievenard M. 1999. Climate and atmospheric history of the past 420,000 years from the Vostok ice core, Antarctica.

Nature 399: 429-436.

Pounds JA, Puschendorf R. 2004. Ecology - Clouded futures.

Nature 427: 107-109.

Quay P, Stutsman J, Wilbur D, Snover A, Dlugokencky E, Brown T. 1999. The isotopic composition of atmospheric methane. Global

Biogeochemical Cycles 13: 445-461.

Reddy CM, Heraty LJ, Holt BD, Sturchio NC, Eglinton TI, Drenzek NJ, Xu L, Lake JL, Maruya KA. 2000. Stable chlorine isotopic compositions of aroclors and aroclor-contaminated sediments.

Environmental Science & Technology 34: 2866-2870.

Remaud GS, Martin YL, Martin GG, Martin GJ. 1997. Detection of sophisticated adulterations of natural vanilla flavors and extracts: Application of the SNIF-NMR method to vanillin and

(44)

p-hydroxybenzaldehyde. Journal of Agricultural and Food Chemistry

45: 859-866.

Ribeiro A, Caldeira MM, Carvalheiro M, Bastos M, Baptista C, Fagulha A, Barros L, Barosa C, Jones JG. 2005. Simple measurement of gluconeogenesis by direct 2H NMR analysis of menthol glucuronide enrichment from (H2O)2H. Magnetic Resonance in Medicine 54: 429-434.

Ritter L, Solomon KR, Forget J, Stemeroff M, O'Leary C. 1995. A review of selected persistent organic pollutants. PCS/95.39. Report prepared for The International Programme on Chemical Safety.

Robins RJ, Billault I, Duan JR, Guiet SÃ, Pionnier SÃ, Zhang BL.

2003. Measurements of 2_{H distribution in natural products by} quantitative 2H NMR: An approach to understanding the metabolism and enzyme mechanisms? Phytochemistry Reviews 2: 87-102.

Roden JS, Ehleringer JR. 2000. Hydrogen and oxygen isotope ratios of tree ring cellulose for field-grown riparian trees. Oecologia 123: 481-489.

Rossmann A, Butzenlechner M, Schmidt HL. 1991. Evidence for non-statistical carbon isotope distribution in natural glucose. Plant

Physiology 96: 609-614.

Schleucher J, Vanderveer P, Markley JL, Sharkey TD. 1999. Intra-molecular deuterium distributions reveal disequilibrium of chloroplast phosphoglucose isomerase. Plant, Cell and Environment

22: 525-533.

Schmidt HL. 2003. Fundamentals and systematics of the non-statistical distributions of isotopes in natural compounds.

Naturwissenschaften 90: 537-552.

Singleton DA, Thomas AA. 1995. High-precision simultaneous determination of multiple small kinetic isotope effects at natural abundance. Journal of the American Chemical Society 117: 9357-9358.

Smith BN, Ziegler H. 1990. Isotopic fractionation of hydrogen in plants. Botanica Acta 103: 335-342.

Deuterium isotopomers as a tool in environmental research