Chapter 17. Accelerator Mass Spectrometry in the Study of Vitamins and Mineral Metabolism in Humans

17 Accelerator Mass Spectrometry in the Study of Vitamins andMineral Metabolism in Humans Fabiana Fonseca de Moura, Betty Jane Burri, and Andrew J.. 554 HISTORICAL BACKGROUND Accelerator

17 Accelerator Mass Spectrometry in the Study of Vitamins and

Mineral Metabolism in Humans

Fabiana Fonseca de Moura, Betty Jane Burri,

and Andrew J Clifford

CONTENTS

Historical Background 545

Description of Accelerator Mass Spectrometry 547

Accelerator Mass Spectrometry Method 548

Considerations for Human Subjects 548

Mathematical Modeling 549

Human Folate Metabolism 549

Human Vitamin A and b-Carotene Metabolism 551

Calcium 553

Summary 553

Acknowledgments 554

References 554

HISTORICAL BACKGROUND

Accelerator mass spectrometry (AMS) harnesses the power of advanced nuclear instruments

to solve important and heretofore unsolvable problems in human nutrition and metabolism AMS methods are based on standard nuclear physics concepts Isotopes of a given element differ from one another by the number of neutrons in their nucleus Generally, the isotope with the lowest number of neutrons in its nucleus is the natural isotope (e.g.,1H,12C) Adding one neutron typically creates a stable isotope (e.g.,2H,13C), which is similar in most properties

to the natural isotope, but differs in mass and can thus be separated and detected by mass spectrometry Isotopes with even greater numbers of neutrons (e.g.,3H,14C) become unstable

An unstable nucleus such as14C has excess energy, which is released in the form of particles of radiation These radioisotopes can also be detected by mass spectrometry, while more common and familiar instruments such as liquid scintillation and Geiger counters can detect their radioactive decay products

The antecedents of AMS date back to the beginning of the nuclear era In 1903, Marie Curie and her husband Pierre Curie established quantitative standards for measuring

the rate of radioactive emission, and it was Marie Curie who found that there was a decrease in the rate of radioactive emissions over time (radioactive decay), which could be calculated and predicted.1 In 1911, Ernest Rutherford bombarded atoms with a-rays and defined the structure of the atom.2By 1912, more than 30 radioactive species were known and current isotope terminology was introduced The a-particle is a nucleus of the element helium, b-particles are electrons whereas g-radiation is composed of electromagnetic rays (their names were intended to be temporary until better identification could be obtained) By

1921, several instruments had been constructed to determine the masses of isotopes and their relative proportions These instruments evolved into what we now call mass spectrometers

In the 1930s, J.D Cockroft and E.T.S Walton were the first to construct a true accelerator at the Cavendish Laboratory, at Cambridge, UK.3 The Cockroft– Walton accelerator accelerated protons by driving off electrons from atoms In this acceler-ator, hydrogen protons were generated by an electric discharge in hydrogen gas The proton ions traveled inside an evacuated tube containing electrodes Each time the ions oscillated from one electrode to the other, they accelerated; by the time the ions passed through the tube they were accelerated into a narrow bundle or beam of particles that could be separated and measured This first accelerator generated a little over a million volts Shortly thereafter, Robert J Van de Graaff developed the eponymous generator, which uses static electricity to generate very high voltages In this accelerator, a pulley-driven rubber belt moves at high speed to generate electricity As the pulley rotates, the inside of the belt becomes negatively charged and the outside positive The positive charges are then collected in an outer metal sphere The Van de Graaff generators produced as much as

10 million volts

In 1932, the most famous of all accelerators, the Ernest O Lawrence cyclotron, was built

at the Radiation Laboratory of the University of California at Berkeley.4In this accelerator the particle beams circled, allowing the particles to pass through the same electrodes many times Between 1934 and 1939, a large number of radionuclides were produced, iden-tified, and characterized by bombarding elements with every available particle in accelerating machines

Cyclotrons could also in principle be used as extremely sensitive mass spectrometers, but

it was not until 1977 that a cyclotron was used in this way for radiocarbon dating.5 The cyclotron increased the sensitivity of radiocarbon dating dramatically because it allowed direct measurement of the actual mass of radioactive 14C, instead of the typical methods, which only count radioactive decays Mass spectrometry methods had also been suggested for the measurement of 14C=12C ratios for carbon dating, but had difficulties distinguish-ing between the14N and14C To solve that problem, two research groups in 1977 proposed using a tandem Van de Graaff accelerator instead of a cyclotron for radiocarbon dating.6,7The Van de Graaff accelerator can discriminate between14N and 14C and it is also capable of accelerating and separating all three carbon isotopes (12C,13C, and 14C) simultaneously.8 Nowadays, the Van de Graaff accelerator is the most commonly used accelerator for 14C measurements

In the early 1960s, before the advancement of AMS, there were 14C measurements

of human blood and tissues from individuals who were exposed to elevated atmospheric

14C from nuclear weapons testing,9,10as well as 14C studies of the metabolism of nutrients

in hospitalized patients.11,12However, these studies had to use large amounts of14C capable of being detected by a liquid scintillation counter The possibility of using AMS in biomedical research has been reported since 197813and was reenforced in a review published in 1987.14 However, it was not until the early 1990s that AMS began to be used regularly for biomedical and clinical applications.15–19

DESCRIPTION OF ACCELERATOR MASS SPECTROMETRY

A photo of the accele rator mass spectromet er at Lawre nce Liv ermore National Labor atory (Liver more, CA, USA) used in our studies is shown in Fig ure 17.1 An accele rator mass spectr ometer is a form of an isot ope ratio spec trometer, ideal for measuring long-l ived radioi sotopes because it measur es the actual mass rather than the radioac tive decay AMS separat es an d measures the individ ual atoms of isot opic specie s AMS is an extre mely sensi tive techni que, ab le to detect isot ope co ncentra tions to parts pe r quadril lion and qua ntify label ed elements to attom ole level s in milli gram-si zed samples 20

Since AMS measur es individua l isot opomers, it is milli ons of times more sen sitive than the more famili ar methods of Geiger cou nting an d liquid scintillati on coun ting, which only measur e radioac tive decays How ever, da ta from liquid scint illatio n coun ting a nd AMS can

be linea rly extrapo lated an d compared 8,21 Radio isotope metho ds have inherent superi orities

to stabl e and natural isoto pe methods Specif ically, the total radioisotop e acti vity can be colle cted and measur ed , regardl ess of whet her the comp ounds measur ed have be en identi fied This allows for the colle ction and measur ement of all the meta bolites, before they are identifi ed Stable isot ope methods , in contras t, are difficul t to use to identify meta bolites, and in general can only be used to measur e meta bolites that have a lready been identi fied by other methods A seco nd advantag e is that AMS is more sensitiv e than almos t all stabl e isoto pe methods cu rrently available; by us ing such smal l dosages, it allows researchers to cond uct true-tra cer studies This is especi ally advantag eous in nutri ent meta bolism resear ch, wher e the observed behavior s in nutri ent meta bolism may de pend on the size of the adminis-tered label ed dose

The most co mmon use of AMS in nutri tion is to measur e carbon or hyd rogen isot opes, although calcium and aluminum have also been measured.22–24In this chapter, we illustrate the use of AMS for human metabolism research, using folic acid, vitamin A, b-carotene, and calcium as examples

FIGURE 17.1 1 MV accelerator mass spectrometer at the Lawrence Livermore National Laboratory

ACCELERATOR MASS SPECTROMETRY METHOD

AMS methods require careful sample preparation Before AMS measurement, the carbon

of biological samples must be converted to graphite The first method for rapid production

of graphite from biological samples was developed in 1992.25A description of a high-throughput method for measuring14C is given below.26In the first step, dried biological samples are placed

in combustion tubes containing cupric oxide and heated to 6508C for 2.5 h All of the carbon present in the sample is oxidized to carbon dioxide In the second step, carbon dioxide is reduced to graphite in the presence of titanium hydride and zinc powder at 5008C for 3 h then 5508C for 2 h, using cobalt as catalyst.26The graphitized samples are then loaded into the AMS instrument and 1 mg (or more) of carbon is added to each sample in the form of 50 mL 33.3 mg=mL of tributyrin in methanol It is important that the biological material to be analyzed does not get contaminated with14C during sample preparation To avoid sample cross con-tamination, disposable materials are used throughout the entire process of graphitization Most AMS instruments use cesium as an ion source.20,27 Samples are bombarded with cesium vapor, which causes the graphitized samples to form negative ions that are extracted

by a series of plates held thousands of volts more positive than the ion source The negative ion beam enters an injection magnet where the ions are separated and selected by their mass-to-charge ratio, so that12C,13C, and14C ions pass through separately as a series of pulses in sequence.28 The pulsed ion beams pass into a tandem electrostatic Van de Graaff particle accelerator where the negative ions flow toward a positive terminal held at 1 to 5 million volts

As the ions travel, they attain very high energies, and these high-energy ion beams are focused

to collide with argon gas molecules (on a 0.02 mmol thin carbon foil) in a collision cell This collision strips the outer valence electrons from the atoms, so that the charge on the atoms changes from negative to positive and all molecular species are converted to atoms These positive atomic ion beams are now repelled by the positive high terminal voltage used and exit the accelerator The beams then pass into a high-energy analyzing magnet where the

12C,13C, and14C atoms are separated by their mass moment charge state ratio.12C and13C are measured with Faraday cups whereas the less abundant 14C beam is focused by a quadropole and electrostatic cylindrical analyzer and counted in a gas ionization detector The rare isotope (14C) count is compared to the abundant (12C) isotope count to determine the relative abundance of the14C atoms in the original sample.28Measurements are normalized to improve precision, by comparing the14C=12C ratio in the sample with the same ratio obtained from a known standard, graphitized sucrose with an accepted14C=12C ratio of 1.5081 modern (Australian National University [ANU], Canberra, Australia).26,27 14C determinations are made at the Center for Accelerator Mass Spectrometry at Lawrence Livermore Laboratory (Livermore, CA, USA)

AMS can also be used to detect 3H tracers in milligram-sized samples.29,30 Sample preparation for analysis by tritium AMS is a multistep process in which the organic samples are converted to titanium hydride.29First, the organic sample is oxidized to carbon dioxide and water Then the water is reduced to hydrogen gas, which reacts with titanium to produce titanium hydride The ratio of3H=1H is measured by AMS This technique is currently under development, but once established it can be a very powerful tool because3H is the most widely and least expensive radioisotope used in biomedical research In addition, 3H AMS could

be used with 14C AMS for double-labeled experiments to study the interaction of two compounds or the metabolites of a single compound labeled in two separate locations.31,32 CONSIDERATIONS FOR HUMAN SUBJECTS

Several studies conducted in the 1960s used relatively large doses of radioisotopes to study the metabolism of vitamins in hospitalized subjects Classic studies of vitamins A, C, E, and other

nutrients were all conducted this way The information from these studies formed the basis of our current understanding of nutrient metabolism and requirement Those experi-mental protocols would not meet current Institutional Review Boards’ requirements since the amount of radiation used ranged from 10 to 194 mCi AMS, an extremely sensitive technique, allows the use of radiation dosages that are several 1000 fold lower, on average, a radiation exposure of 100 nCi,35–39which corresponds to 11 mSv or 1.1 mrem This amount of radiation exposure is equivalent to that received during a 3 h flight in an airplane or from 1 day of cosmic radiation at sea level The U.S Food and Drug Administration defines a safe radiation dose as <3 rem to the whole body, blood-forming organs, lens of the eye, and gonads or 5 rem for the remaining organs.40Additionally, tissues and fluids with a specific activity >2 nCi=g must be declared as radioactive material The blood, urine, and fecal specimens from the low doses of14C used in current AMS studies (200 nCi) are below the

2 nCi=g cutoff; therefore, the specimens are not considered radioactive material by the U.S Federal Regulation.40

MATHEMATICAL MODELING

Our understanding of nutrient metabolism is hindered because metabolism occurs over time, often in inaccessible tissues It is very difficult, even impossible, to collect experimental data for some critical steps in nutrient metabolism in vivo Kinetic modeling is a systems analysis approach that constructs a quantitative overview of the dynamic and kinetic beha-vior of metabolism of a nutrient as it might occur in vivo A mathematical model is built to realize as complete a description as possible of the metabolic system under investigation The advance of computer hardware and modeling software makes it possible to solve (and manipulate) differential equations meant to predict kinetic behavior efficiently and accur-ately Therefore, mathematical modeling has become an attractive tool for collecting and processing research data and information needed to understand the dynamics of nutrient metabolism in vivo Kinetic models are built to mimic the metabolism of a nutrient as it might occur in vivo and to estimate values for critical parameters, so that unobserved portions of the dynamic and kinetic behavior of the nutrient under investigation can be predicted Specific information obtained about the nutrient under investigation includes the number of storage sites (pools) for the nutrient and their sizes, how they are connected, and how their masses change over time

Modeling begins with a thorough review existing knowledge of the metabolism of the nutrient under investigation to formulate an initial structure for the model Then initial constants (for transfer of nutrient to recipient compartments from donor com-partments) are estimated and adjusted in physiologically relevant ways until the model structure and rate constants predict best fits for the experimental data Final parameter values are generated using iterative nonlinear least squares routines The following references describe a series of conferences on mathematical modeling in nutrition and health sciences.41–47

HUMAN FOLATE METABOLISM

Folate is necessary for purine and pyrimidine synthesis and for the metabolism of homo-cysteine to methionine There have been extensive studies of folate metabolism in humans using pharmacological dosages of radiolabeled folate measured with liquid scintillation counting.48–52 These studies yield useful information about folate absorption, metabolism, and excretion However, all but one were of short duration and thus gave no information about long-term storage and metabolism of folate.52

We investigated short- and long-range human folate metabolism with AMS following an oral dose of 14C-pteroylmonoglutamate in healthy adults.38 Thirteen free-living adults re-ceived 0.5 nmol14C-pteroylmonoglutamate (100 nCi) plus 79.5 nmol nonlabeled pterolylmo-noglutamate orally in water The subjects were typical American adults with no known disease and had a mean dietary folate intake of 1046 nmol=day.14C was followed in plasma, erythrocytes, urine, and feces for 40 days Kinetic models were used to analyze and interpret the data Model parameters were optimized using the SAAM II kinetic analysis software such that hypotheses that were inconsistent with the datasets observed for each of the 13 subjects could be rejected A diagram of the final model is shown in Figure 17.2 Our model consisted

of four pools of folate: gastrointestinal tract (lumen), plasma, erythrocyte, and viscera (all other tissues)

Apparent absorption of 14C-pteroylmonoglutamate was 79% Mean total body folate was 225 mmol Pteroylpolyglutamate synthesis, recycling, and catabolism were 1985, 1429, and 556 nmol=day, respectively Mean residence times were 0.525 day as visceral pteroylmo-noglutamate, 119 days as visceral pteroylpolyglutamate, 0.0086 day as plasma folate, and 0.1 day as gastrointestinal folate

The kinetic model predicted that only 0.25% of plasma folate was destined for bone marrow, even though folate metabolism is important for healthy bones It also predicted an important role for bile in folate metabolism Most folate was recycled in tissues through bile Visceral pteroylmonoglutamate, which is transported to the gastrointestinal tract via bile, provided a large pool of extracellular pteroylmonoglutamate (5351 nmol=day) that could blunt between-meal fluctuations in folate supply to the cells to sustain folate concentrations during periods of folate deprivation Therefore, the digestibility of the dietary folate plus the folate recovered in the bile (1046þ 5351 nmol=day, respectively) was 92% We accounted for the gastric transit time of 1 day to the absorption site The 6.15 days erythron transit time was a new observation that fit well with the week-long maturation of hematopoietic progenitor cells.53

Intact pteroylmonoglutamate that was eliminated in the urine represented ~6%

of ingested folate, a value that compared well with already published values.54–56 However, the novel and testable hypothesis represented by our model is that fully one-half of excreted folate was derived from visceral pteroylpolyglutamate and appeared in the urine as p-aminobenzoylglutamate (and its metabolic successors)

PteGlu1= Pteroylmonoglutamate PteGlun= Pteroylpolyglutamate

p-ABA-Glu =p-Aminobenzoylglutamate FABP = Folate-binding protein

GI Tract

Plasma

Plasma

p-ABA-Glu

Urine

Viscera

FABP

5897

2.6

5982

5351

556

Marrow

Feces

Diet

Enterohepatic 1046

FIGURE 17.2 Kinetic model of folate metabolism The numbers represent steady-state folate fluxes (nanomoles per day)

The model makes several important predictions First, the fractional absorption

of folate was high and independent of the gastrointestinal folate load Second, ~33% of visceral pteroylmonoglutamate was converted to the polyglutamate form Third, most of the body folate was visceral (>99%), and most of the visceral folate was pteroylpolyglutamate (>98%) Fourth, the model predicted that bile folate was 25 times greater than prior estimates and that steady-state folate distributions were approximately fivefold larger than prior estimates.57

In addition, the model predicted two distinct chemical forms of folate in plasma: pteroylmonoglutamate and p-aminobenzoylglutamate For visceral pteroylglutamate to be recycled by conversion to visceral pteroylmonoglutamate was no surprise, but for visceral pteroylpolyglutamate to also be converted directly to p-aminobenzoylglutamate is a new path-way that fits nicely with other recent discoveries in pteroylpolyglutamate catabolism.58

HUMAN VITAMIN A AND b-CAROTENE METABOLISM

Vitamin A (retinol and its metabolites) plays an important role in vision, growth, cell division, and differentiation.59 Retinol status has been difficult to assess using nonisotopic methods, because its serum concentrations are tightly regulated and 90% or more of its body stores are in inaccessible tissues such as liver and kidney.60–64Therefore, much of what is known about the human absorption and metabolism of retinoids is based on one small radioisotope study.65

Recently we fed deuterated retinyl acetate to adult men and women Our results show that

a single large peak appears in the blood at ~4 –8 h postdose, reaching its maximum at 12–24 h postdose.66 The vitamin A half-lives ranged from 75 to 241 days for men fed

a vitamin A-deficient diet65and 56 to 243 days for men and women fed a vitamin A-adequate diet.36,67,68

The reasons for the large variations in metabolic half-life are unknown, but the main factors that appear to influence vitamin A metabolism are the individual’s vitamin A nutri-tional status and dietary intake People with higher retinol status appear to absorb retinol more efficiently than people with lower retinol status.36,65,66 Very low intakes of retinol appear to reduce (rather than increase) retinol utilization, even when retinol stores are still adequate.65Other factors, such as, gender, race, and body composition, did not have a strong influence on vitamin A metabolism in our studies, but might well have an impact in more heterogeneous groups.66

Although retinoids are key essential nutrients, they are not widely dispersed among foods In developing countries, b-carotene, found in yellow-orange fruits and vegetables, is the major source of vitamin A.69 b-carotene has also been reported to have various biological effects; among them are enhancement of the immunological system, to stimulate gap junction communication between cells in vitro, and a possible antioxidant activity.70–72We conducted a long-term kinetic study of b-carotene30 using 14C-b-carotene derived by growing spinach in an atmospherically sealed chamber pulsed with 14CO2

A healthy 35 year old male received a single oral dose of 14C-b-carotene (306 mg; 200 nCi) and the tracer was followed for 209 days in plasma, 17 days in urine, and 10 days in feces Aliquots of plasma (30 mL), urine (100 mL), and stool (150 mL) samples were analyzed Plasma 14C-b-carotene, 14C-retinyl esters, 14C-retinol, and several 14C-retinoic acids were separated by reversed phase HPLC

The results showed that 57.4% of the administered dose was recovered in the stool within 48 h postdose; therefore, 42.6% of b-carotene was bioavailable Urine was not a major excrete route for intact b-carotene There was a 5.5 h delay between dosing and the

appearance of14C in plasma The losses of14C-b-carotene and its metabolites after an oral dose

of14C-b-carotene are shown in Figure 17.3.14C-b-carotene and14C-retinyl esters presented similar kinetic profiles for the first 24 h Both14C-b-carotene and14C-retinyl esters rose to

a plateau spanning between 14 and 21.3 h The concentration of14C-retinol rose linearly for

28 h postdose before declining Therefore, the substantial disappearance of retinyl esters from plasma between 21 and 25 h closely preceded the transition from increasing retinol concentrations This observation suggests that retinyl ester was handed off to retinol into circulation The area under the curve suggested a molar vitamin A value of 0.53 for b-carotene, with a minimum of 62% of the absorbed b-carotene cleaving to vitamin A The pattern of total14C,14C-b-carotene,14C-retinyl esters, and14C-retinol in plasma is shown in Figure 17.4

Days postdose, log scale

−0.005 0.000 0.005 0.010 0.015 0.020 0.025 0.030

In 14C-β-carotene

In 14C-retinol

In14C-retinyl esters

In14C-total

which is associated with the labeled retinyl esters, retinol, and b-carotene fractions, accounts for about one-half of the total radioactivity The remainder is associated with yet-unidentified carotenoid and retinoid metabolites, possibly epoxides, apo-carotenals, and retinoic acids

Days postdose

−0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6

Feces

−0.04 0.00 0.04 0.08 0.12 0.16 0.20 0.24

Urine

S.L., Dueker, S.R., Follett, J.R., Lin, Y., Carkeet, C., Buchholz, B.A., Vogel, J.S., and Clifford, A.J.,

J Lipid Res., 44 (8), 1591, 2003.)

In addition, the effect of vitamin A nutritional status on b-carotene metabolism was elucidated.36 Two healthy adult women received an oral dose of 14C-b-carotene (0.5–1.0 nmol with specific activity of 98.8 mCi=mmol) in a banana ‘‘milk shake.’’ Seven weeks after this first dose, both women began taking a vitamin A supplement, 3000 RE (10,000 IU) retinyl palmitate per day They consumed the supplement for 21 days and then received a second dose of14C-b-carotene The women continued taking the 3000 RE supple-ments for 14 days after the second dose was given, then the amount of vitamin A supplement was decreased to 1500 RE (5000 IU) per day for the remainder of the study Concentrations

of 14C-b-carotene,14C-retinyl esters, and 14C-retinol in plasma were measured for 46 days after the first dose and 56 days after the second dose

Using AUCs and irreversible losses of14C in feces and urine, a yield of 0.54 mol14C-vitamin

A from 1 mol of14C-b-carotene before supplementation and 0.74 mol14C-vitamin A after supplementation was calculated These data indicate that more vitamin A was formed from b-carotene when subjects were taking vitamin A supplementation This suggests that retinoid status can influence carotenoid status and vice versa

CALCIUM

Calcium is the most abundant divalent cation of the human body and is important for the maintenance of bone mineral density, blood clotting, nerve conduction, muscle contraction, enzyme regulation, and membrane permeability.73 Calcium has three radio-isotopes, 47Ca, 45Ca, and 41Ca:47Ca and 45Ca have relatively short half-lives (4.5 and 165 days, respectively) but41Ca is a very long-lived radioisotope (t1=2 116,000 years) Osteo-porosis, the decrease in bone mass and density due in part to loss of calcium, is a growing problem as people age; therefore, long-term studies on bone calcium turnover and bone resorption are extremely important.74,75 Short-term studies of calcium metabolism can be done by a variety of stable and radioisotope techniques, using47Ca and45Ca, but these cannot resolve long-term small but significant differences in bone resorption The advent of AMS has made possible the use of the long-lived radioisotope 41Ca, which potentially could be traced for decades.76 In 1990, Elmore et al assessed the potential for using 41Ca for bone resorption study by measuring41Ca by AMS in dogs.77The authors demonstrated that41Ca behaves identically to 45Ca in vivo Freeman et al developed an improved protocol,78then administered 5 nCi of41Ca dissolved in orange juice to 25 subjects and measured the tracer

in urine by AMS.24Fink et al described the protocols for measuring 41Ca=40Ca ratios to a sensitivity of 6 1016.79These studies have clearly demonstrated the feasibility of the AMS approach Freeman et al have also shown that the osteoporosis drug, alendronate, markedly suppressed bone resorption by effecting 41Ca loss in urine.24 The use of 41Ca and AMS to better understand long-term calcium metabolism in humans, and to trace the impact on osteoporosis of minute differences in calcium metabolism, occurring over many years, offers many exciting possibilities

SUMMARY

AMS is an isotope ratio method ideal for measuring the ratios of long-lived radioisotopes such as14C=12C for biological and chemical research It is capable of measuring nutrients and their metabolites in attomol (1018) concentrations in milligram-sized samples The detection sensitivity and small sample size requirements of AMS satisfies both the analytical and ethical requirements for tracer applications in human subjects

The literature on the AMS method has been published mostly by the researchers at the Center for Accelerator Mass Spectrometry at Lawrence National Laboratory (LLNL) in the United States, the Center for Biomedical Accelerator Mass Spectrometry in York, UK, and the Radiocarbon Laboratory of the GeoBiosphere Center in Lund, Sweden All of the work reported by our laboratory was performed in collaboration with the researchers at the Center for AMS at LLNL The best current description of the method for preparing AMS samples is given in: Getachew, G., Kim, S.H., Burri, B.J., Kelly, P.B., Haack, K.W., Ognibene, T.J., Buchholz, B.A., Vogel, J.S., Modrow, J., and Clifford, A.J How to convert biological carbon into graphite for AMS Radiocarbon, 48, 325–336, 2006

REFERENCES

1 Glasstone, S., Source Book on Atomic Energy, 3rd ed D Van Nostrand Company, Inc., New York,

1967, chap 2

2 Faires, R.A and Parks, B.H., Radioisotope Laboratory Techniques, 1st ed George Newnes, Ltd., London, 1958, chap 1

3 Wilson, R.R and Littauer, R., Accelerators: Machines of Nuclear Physics, 1st ed Anchor Books, New York, 1960, chap 3

4 Heilbron, J.L and Seidel, R.W., Lawrence and his Laboratory: A History of the Lawrence Berkley Laboratory, 1st ed University of California Press, Berkley and Los Angeles, 1989, chap 1

5 Muller, R.A., Radioisotope dating with a cyclotron, Science, 196, 489–494, 1977

6 Nelson, D.E., Korteling, R.G., and Stott, W.R., Carbon-14: direct detection at natural concentra-tions, Science, 198, 507–508, 1977

7 Bennett, C.L et al., Radiocarbon dating using electrostatic accelerators: negative ions provide the key, Science, 198, 508–510, 1977

8 Garner, R.C., Barker, J., Flavell, C., Garner, J.V., Whattam, M., Young, G.C., Cussans, N., Jezequel, S., and Leong, D., A validation study comparing accelerator MS and liquid scintillation

Anal., 24 (2), 197–209, 2000

9 Broecker, W.S., Schulert, A., and Olson, E.A., Bomb carbon-14 in human beings, Science, 130 (3371), 331–332, 1959

10 Libby, W.F., Berger, R., Mead, J.F., Alexander, G.V., and Ross, J.F., Replacement rates for human tissue from atmospheric radiocarbon, Science, 146, 1170–1172, 1964

11 Goodman, D.S., Blomstrand, R., Werner, B., Huang, H.S., and Shiratori, T., The intestinal absorption and metabolism of vitamin A and beta-carotene in man, J Clin Invest., 45 (10), 1615–1623, 1966

12 Blomstrand, R and Werner, B., Studies on the intestinal absorption of radioactive beta-carotene and vitamin A in man Conversion of beta-carotene into vitamin A, Scand J Clin Lab Invest., 19 (4), 339–345, 1967

13 Keilson, J and Waterhouse, C., First Conference on Radiocarbon Dating with Accelerators H.E Gove University of Rochester, Rochester, 1978

14 Elmore, D and Phillips, F.M., Accelerator mass spectrometry for measurement of long-lived radioisotopes, Science, 236, 543–550, 1987

15 Turteltaub, K.W., Felton, J.S., Gledhill, B.L., Vogel, J.S., Southon, J.R., Caffee, M.W., Finkel, R.C., Nelson, D.E., Proctor, I.D., and Davis, J.C., Accelerator mass spectrometry in biomedical dosimetry: relationship between low-level exposure and covalent binding of heterocyclic amine carcinogens to DNA, Proc Natl Acad Sci., USA 87 (14), 5288–5292, 1990

16 Shapiro, S.D., Endicott, S.K., Province, M.A., Pierce, J.A., and Campbell, E.J., Marked longevity

weapons-related radiocarbon, J Clin Invest., 87 (5), 1828–1834, 1991

17 Turteltaub, K.W., Frantz, C.E., Creek, M.R., Vogel, J.S., Shen, N., and Fultz, E., DNA adducts in model systems and humans, J Cell Biochem Suppl., 17F, 138–148, 1993