- Email: [email protected]
Application of accelerator mass spectrometry in aluminum metabolism studies
536
Nuclear
Instruments
and Methods
in Physics
Research
B52 (1990) 536-539 North-Holland
Application of accelerator mass spectrometry metabolism studies
in aluminum
0. Meirav I) *, R.A.L. Sutton 2), D. Fink 3), R. Middleton 3), J. Klein 3), V.R. Walker 2), A. Halabe 2), D. Vetterli ‘) and R.R. Johnson ‘) ” Deparrment of Physics and ” Department of Medicine, University of Brirish Columbia, Vancouver, BC, Canada V6T I W5 ” Department of Physics, University of Penkylvania,
Philadelphia, Pennsylvania 19104. USA
The recent recognition that aluminum causes toxicity in uremic patients and may be associated with Alzheimer’s disease has stimulated many studies of its biochemical effects. However, such studies were hampered by the lack of a suitable tracer. In a novel experiment, we have applied the new technique of accelerator mass spectrometry to investigate aluminum kinetics in rats, using as a marker the long-lived isotope 26A1. We present the first aluminum kinetic model for a biological system. The results clearly demonstrate the advantage this technique holds for isotope tracer studies in animals as well as in humans.
1. Introduction Despite being the third most abundant element in the Earth’s crust, aluminum was until recently believed to have no significant biochemical role in health or disease, nor was it thought to participate in major metabolic processes [l]. It was considered to be very poorly absorbed from the gastrointestinal tract, and that fraction absorbed was assumed to be efficiently eliminated by the kidneys. Research over the past decade, however, presents a very contrary assessment of its effect on the human body - aluminum is now referred to as a toxic metal with sufficient evidence to implicate it in several disease states. The accumulation of ahnninum in the tissues of patients with chronic renal failure has been established [2-S] as the cause of certain toxic phenomena, including dialysis encephalopathy, osteomalacic dialysis osteodystrophy (a specific metabolic bone disease) and anemia. Although paranteral exposure to aluminum has been minimized by removing it from the dialysate fluid, toxicity continues to occur in uremic patients whether dialyzed or not. In addition, a number of reports [6,7 and references therein] have strongly suggested that aluminum plays an important role in the pathogenesis of neurofibrillary tangle formation. Since the lesion is one of the principal histologic features of Alzheimer’s disease, the concept has arisen that the presence of aluminum in certain populations of neurons serves as a pathogenetic, and possibly etiologic, factor in the disease. These sugges-
* Address for correspondence: Wesbrook
Mall, Vancouver,
0168-583X/90/$03.50
Oded Meirav, TRIUMF, BC, Canada V6T 2A3.
4004
Q 1990 - Elsevier Science Publishers
tions are, however, still controversial mainly because of the methodological difficulties that were encountered in measuring aluminum concentrations in brain samples. Although there is now overwhelming data in support of ahuninum toxicity, research into its mechanism has been severely limited by the absence of a suitable isotope with which to perform conventional tracer studies, as well as other methodological difficulties. Since aluminum has only a single stable isotope and no radioactive ones with life-times suitable for traditional radiation detection methods, kinetic studies using isotopic tracers have been precluded. Because of its very low concentration in biological systems (ppm in tissues and ppb in blood [8]) monitoring aluminum is a formidable task [9]. Consequently, the sole method of studying ahuninum kinetics prior to the resent work had been P, to administer large dosages of Al mto the body, followed by measurements of its level in blood, urine and several organs [lo-141. Since these loads of aluminum were very large compared to the total body burden, experiments in humans are impractical and the information obtained for animals cannot reflect the steady state kinetics. A new opportunity to investigate the biological behavior of aluminum has arisen with the development of accelerator mass spectrometry (AMS), whereby radionuclides themselves, rather than their decay products, are detected. This technique enables measurements to be made on milligram samples containing as few as a million atoms of a specific radionuclide [15]. Typically, the sensitivity expressed as a radioisotope to stable-isotope ratio is of the order to 10-‘4-10-‘6. It is particularly suited to long-lived radioisotopes, thus providing an analytical method by which the radioisotope ‘“Al
B.V. (North-Holland)
0. Meirav et al. / AMS in AI metabolism studies (Tl,,2 = 7.2 x 10’ yr) can be measured with a sensitivity
a million times greater than that of conventional decay counting. Hence, it is now possible to use 26Al as a marker, with amounts of material that are neither prohibitively costly nor hazardous. The idea of employing 26A1, in conjunction with AMS, as a tracer to study aluminum behavior in the human body was first proposed by Litherland 2161and discussed recently in more detail by Elmore 1171.In the present paper we describe an experiment aimed at assessing the feasibility of such a proposal and present the first results of this new method.
2. Mehod
A 385 g rat was injected intravenously with 0.35 ng of 26A1and 40 ng of 27A1dissolved in 0.5 ml of a slightly acidic (pH = 5) saline solution. A second rat, which was used as a control, received 39 ng of “A1 and no 26Xl. Serum and urine samples were collected over a three week period, with the first serum sample taken five hours after injection. Volumes of 0.05 to 0.5 ml of urine and 0.01 to 0.1 ml of serum were used for preparation of the AMS samples. Prior to chemical processing, 5 mg of “A was added as carrier and all serum samples were treated with trypsin in order to release the aluminum from the transferrin to which most of the upturn in the blood is bound [13,18]. Two techniques (fig. 1) were used to convert the aluminum in the solutions to
A
A
5 mg of Al I
I
2 ml of 5% oxine
iii1 r--z&~7 r dryness j / to
filter
I
Fig. 1. Plow diagram of extraction of aluminum from blood and urine samples.
53-I
aluminum oxide. In one, the entire solution was dried and the resulting residue ashed. In the other, oxine was used to selectively precipitate the burnt 1191 prior to the ashing. 26A1concentrations in the serum and urine samples were determined from 26A1/27A1ratios measured at the AMS facility at the University of Pennsylvania [20]. The ratios ranged from lo-‘* to lo-r4 allowing their deter~nation with statistical and systematic~ errors of l-10%. The conversion of these ratios to 26A1concentrations did not require the determination of the “AI concentrations because the carrier added to the samples completely dominated the aluminum native to the urine and serum. In addition, the carrier ensured that the measured 26Al,/27Al ratio would not be affected if any ‘?A1 was picked up during handling. The 26A1 background level was determined from samples collected before the injections and from the control rat (the natural background of the 26A1/27A1ratio is less than lo-r4 [20]) and was found to be negligible.
3. Rest&s Fig. 2 shows the fraction of the administered ‘!4l remaining in the serum and excreted in the urine as a function of time after injection. There is a rapid exponential decrease of the level of ‘“A1 over the first couple of days. Thereafter, from a concentration of about a tenth of a percent of the injected dosage, it decreases at a much slower rate. A three-compartment model (fig. 3) was constructed with one compartment representing the blood and the other two representing exchangeable pools in active equilibrium with the blood. The common approach of first order kinetics in which the clearance rate of an isotope from a compartment is proportional to the total amount of the isotope in that compartment was employed. In general, isotopic fr~tionation between “141 and 27A1will not occur if there is no kinetic isotope effect and if the two isotopes are in the same chemical form. Whilst the former probably holds, it is not known how fast the injected aluminum equilibrates with the native aluminum as aluminum can exist in various forms depending on the chemical evironment (m particular the pH). However, since most of the blood aluminum is attached to transferrin and since iron occupies only one-third of the metal binding sites [21], there are ample sites available for aluminum binding and hence we assumed that the administered aluminum was quickly incorporated into the transferrin molecules resulting in identical fractional clearance coefficients (&‘s). Under this assumption it was possible to use the same compartmental analysis for both isotopes. For 26A1there is no oral input and hence the model is IV(c). BIOMEDICAL
SCIENCES
538
0. Meirav et al. / AMS in AI metabolism studies
respectively, and the matrix A is given by:
IO0
-(R,+R,+R,) A(t)
10-l
w a,
R2
R,
RI
-R,
0
R4
0
-&
‘I.
(3)
For 27A1at steady state, dY(t)/dt = 0 and so:
4
-2
:
Pl(27Al)
'FlO .-
: ‘--d___
.
The aluminum
i
‘1% --__---__
1O-4
1o-5
I
0
5
-+ _
20
25
described mathematically by the following set of linear differential equations:
dY dt =AY,
(2) where B is the amount of the isotope in the blood, PI P2 the amounts in the first and second pools,
and
FOOD
INTESTINES _BLOOD
% R5
KIDNEYS
(6)
to solve eq. (1) analytically:
Fig. 2. The data shownis the fraction of the administered*“Al remainingin the blood and excreted in the urine as a function of time after injection as measuredin the present experiment. The horizontal bars indicate the duration of the urine collection. The curves are the resultsof a three-compartmentsimulation usingthe clearancevaluesgivenin the text.
R4
(5)
In general, the Ri’s are time dependent and may vary if steady state does not pertain. The amounts of aluminum injected in the present experiment were comparable with the total aluminum in blood but negligible compared with the total rat aluminum content. Thus it was assumed that the steady state was not disturbed and the Ri’s were taken to be constant. In this case it is possible
_yum ?-..
IO 15 days
(4)
lost daily by excretion is
U= R,B.
3
a:
R, = B(27Al)Rlr
P2(27A1)R5 = B(27A1)R4.
--,-_“‘!t$__
:lo-3 ?4
Ed
=
Y(t)
= i
c,E,, eat’,
(7)
i=l
where ai and Ea, are the eigenvalues and eigenvectors of the matrix A and the ci’s are normalization coefficients determined by the initial conditions for the administered 26A1: B(0) = 1; P,(O) = 0. A best fit to the data was obtained (fig. 2) for clearance coefficients of R, = 115, R2 = 22, R, = 27, R, = 3.5 and R, = 0.07 ((g/g)/day). The total aluminum blood pool was estimated to be 100 ng based on a total of - 13 ml of serum in a rat [22] and an aluminum concentration of 5-10 l.tg per litre of serum [14,23]. Thus, under the above mentioned assumptions, namely, no isotopic fractionation and constant Ri’s, our results correspond to an alumirmm excretion of 2.7 pg/day (eq. (6)) and to pool sizes of PI = 0.5 l.tg (eq. (4)) and Pz = 5.0 ltg (eq. (5)). These two pools represent about 0.3% and 3% of the total aluminum in a rat (based on estimated total aluminum in humans of 30-40 mg [7,8]). Approximately 75% of the 26A1was excreted in the urine during the first few days following the injection. The residence time of aluminum in the blood is l/( R, + R3 - 10 minutes, that in PI is l/R, - 1 hour and + R4) that in P2 is l/R, - 2 weeks.
Rl
1
6
,
R2
1 I
R3 URINE
Fig. 3. Box diagramof the three-compartmentaluminummodel that was used to simulate the experiment. The Ri’s are the fractionalclearancecoefficients.
4. summary The present work is the first application of AMS in the detection of a long-lived radioisotope tracer in a biological system and conclusively demonstrates its potential to enhance our understanding of aluminum metabolism and toxicity. Future studies will include intravenous and oral administration of the tracer, stud-
0. Meirav et al. / AMS ies of longer
duration in normal and uremic rats, and measurements of aluminum distribution in the brain as well as other organs and bones. Experiments similar to the present one are feasible in humans since the administration of - 30 pCi of 26Al, entailing a radiation dose of less than 0.1% of the natural background [24], would be sufficient. We believe that this approach represents a distinct breakthrough in the research of aluminum behavior in the human body.
Acknowledgements
We thank E. Nelson (Archaeology, Simon Fraser University) and J. Vogel (LLNL) for fruitful discussions, C. Orvig (Chemistry, University of British Columbia) and T. Ruth (TRIUMF) for advising on the chemistry, H. O’Brain (Los Alamos National Lab) and R. Kavanagh (Kellogg Radiation Lab, CALTECH) for providing the 26Al and D. Ridout (Medicine, University of Britisch Columbia) for assisting in sample collection and preparation.
References
111M.R. Wills and J. Savory, in: Aluminum
and Other Trace Elements in Renal Disease, ed. A. Taylor (Bailliere Tindall, London, 1986) pp. 24-37. PI A.C. Alfrey, G.R. LeGendre and W.D. Kaehny, N. Engl. J. Med. 294 (1976) 184. 131 Kidney Int. 29 suppl. 18 (1986) sl-~124. of Metals: Clinical and ex141 A.C. Alfrey, in: Toxicology perimental Research, eds. S.S. Brown and Y. Kodama (Ellis Horwood, Chichester, 1987). pp. 201-210.
539
in AI metabolism studies
[5] M.R. Wills and J. Savory, in: Metal Ions in Biological Systems, vol. 24, ed. H. Sigel (Marcel Dekker, New York, 1988) pp. 315-345. [6] D.R. Perl, ibid., pp. 259-283. [7] T.P. Kruck and D.R. McLachIan. ibid., pp. 285-314. [8] A.C. Alfrey, Kidney Int. 29 suppl. 18 (1986) s8-~11. [9] J. Savory, S. Brown, R.L. Bertholf, N. Mendoz and M.R. Wills, in: Methods in Enzymology, vol. 158, eds. J.F. Riordan and B.L. Vallee (Academic Press, San Diego, 1988) pp. 289-301. [lo] R.R. Rocker, A.J. Blotchy, J.A. Leffler and E.P. Rack, J. Lab. Clin. Med. 90 (1977) 810. [ll] W.D. Kaehny. A.P. Hegg and A.C. Alfrey, N. Engl. J. Med. 296 (1977) 1389. [12] J.E. Gorsky, A.A. Dietz, H. Spencer and D. Osis, Clin. Chem. 25 (1979) 1739. [13] S.K. Gupta, D.H. Waters and P.R. Gwilt, J. Pharma. Sci. 75 (1986) 586. [14] J.C.M. Ghan, M. Jacob, S. Brown, J. Savory and M.R. Wills, Nephron 48 (1988) 61. [15] D. Elmore and F.M. Phillips, Science 236 (1987) 543. 1161 A.E. Litherland, Ann. Rev. Nucl. Sci. 30 (1980) 437. [17] D. Elmore, Biol. Trace Elements Res. 12 (1987) 231. 1181 G.A. Trapp, Life Sci. 33 (1983) 311. [19] A. Vogel, A Text Book of Quantitative Inorganic Analysis (Longman, New York, 1978) pp. 436-437. [20] R. Middleton and J. Klein, Phil. Trans. R. Sot. London A323 (1987) 121. [21] A.C. Alfrey, in: (Marcel Dekker, [22] H.J. Baker, J.R. Laboratory Rat, pp. 411-412.
Aluminum and Health, ed. H.J. Gitelman New York and Basel, 1989) pp. 101-124. Lindsey and S.H. Weisbroth (eds.). The vol. 1. (Academic Press, New York, 1979)
[23] A. Taylor, B. Starkey and A.W. Walker, in: Aluminum and Other Trace Elements in Renal Disease, ed. A. Taylor (Bailhere Tindall, London, 1986) pp. 24-37. [24] E.E. Watson,M. Stabin and W.E. Bloch, MIRDOSE code 1988 version, Oak Ridge Associated Universities, Oak Ridge, Tennessee 37831, USA.
IV(c). BIOMEDICAL
SCIENCES
Nuclear
Instruments
and Methods
in Physics
Research
B52 (1990) 536-539 North-Holland
Application of accelerator mass spectrometry metabolism studies
in aluminum
0. Meirav I) *, R.A.L. Sutton 2), D. Fink 3), R. Middleton 3), J. Klein 3), V.R. Walker 2), A. Halabe 2), D. Vetterli ‘) and R.R. Johnson ‘) ” Deparrment of Physics and ” Department of Medicine, University of Brirish Columbia, Vancouver, BC, Canada V6T I W5 ” Department of Physics, University of Penkylvania,
Philadelphia, Pennsylvania 19104. USA
The recent recognition that aluminum causes toxicity in uremic patients and may be associated with Alzheimer’s disease has stimulated many studies of its biochemical effects. However, such studies were hampered by the lack of a suitable tracer. In a novel experiment, we have applied the new technique of accelerator mass spectrometry to investigate aluminum kinetics in rats, using as a marker the long-lived isotope 26A1. We present the first aluminum kinetic model for a biological system. The results clearly demonstrate the advantage this technique holds for isotope tracer studies in animals as well as in humans.
1. Introduction Despite being the third most abundant element in the Earth’s crust, aluminum was until recently believed to have no significant biochemical role in health or disease, nor was it thought to participate in major metabolic processes [l]. It was considered to be very poorly absorbed from the gastrointestinal tract, and that fraction absorbed was assumed to be efficiently eliminated by the kidneys. Research over the past decade, however, presents a very contrary assessment of its effect on the human body - aluminum is now referred to as a toxic metal with sufficient evidence to implicate it in several disease states. The accumulation of ahnninum in the tissues of patients with chronic renal failure has been established [2-S] as the cause of certain toxic phenomena, including dialysis encephalopathy, osteomalacic dialysis osteodystrophy (a specific metabolic bone disease) and anemia. Although paranteral exposure to aluminum has been minimized by removing it from the dialysate fluid, toxicity continues to occur in uremic patients whether dialyzed or not. In addition, a number of reports [6,7 and references therein] have strongly suggested that aluminum plays an important role in the pathogenesis of neurofibrillary tangle formation. Since the lesion is one of the principal histologic features of Alzheimer’s disease, the concept has arisen that the presence of aluminum in certain populations of neurons serves as a pathogenetic, and possibly etiologic, factor in the disease. These sugges-
* Address for correspondence: Wesbrook
Mall, Vancouver,
0168-583X/90/$03.50
Oded Meirav, TRIUMF, BC, Canada V6T 2A3.
4004
Q 1990 - Elsevier Science Publishers
tions are, however, still controversial mainly because of the methodological difficulties that were encountered in measuring aluminum concentrations in brain samples. Although there is now overwhelming data in support of ahuninum toxicity, research into its mechanism has been severely limited by the absence of a suitable isotope with which to perform conventional tracer studies, as well as other methodological difficulties. Since aluminum has only a single stable isotope and no radioactive ones with life-times suitable for traditional radiation detection methods, kinetic studies using isotopic tracers have been precluded. Because of its very low concentration in biological systems (ppm in tissues and ppb in blood [8]) monitoring aluminum is a formidable task [9]. Consequently, the sole method of studying ahuninum kinetics prior to the resent work had been P, to administer large dosages of Al mto the body, followed by measurements of its level in blood, urine and several organs [lo-141. Since these loads of aluminum were very large compared to the total body burden, experiments in humans are impractical and the information obtained for animals cannot reflect the steady state kinetics. A new opportunity to investigate the biological behavior of aluminum has arisen with the development of accelerator mass spectrometry (AMS), whereby radionuclides themselves, rather than their decay products, are detected. This technique enables measurements to be made on milligram samples containing as few as a million atoms of a specific radionuclide [15]. Typically, the sensitivity expressed as a radioisotope to stable-isotope ratio is of the order to 10-‘4-10-‘6. It is particularly suited to long-lived radioisotopes, thus providing an analytical method by which the radioisotope ‘“Al
B.V. (North-Holland)
0. Meirav et al. / AMS in AI metabolism studies (Tl,,2 = 7.2 x 10’ yr) can be measured with a sensitivity
a million times greater than that of conventional decay counting. Hence, it is now possible to use 26Al as a marker, with amounts of material that are neither prohibitively costly nor hazardous. The idea of employing 26A1, in conjunction with AMS, as a tracer to study aluminum behavior in the human body was first proposed by Litherland 2161and discussed recently in more detail by Elmore 1171.In the present paper we describe an experiment aimed at assessing the feasibility of such a proposal and present the first results of this new method.
2. Mehod
A 385 g rat was injected intravenously with 0.35 ng of 26A1and 40 ng of 27A1dissolved in 0.5 ml of a slightly acidic (pH = 5) saline solution. A second rat, which was used as a control, received 39 ng of “A1 and no 26Xl. Serum and urine samples were collected over a three week period, with the first serum sample taken five hours after injection. Volumes of 0.05 to 0.5 ml of urine and 0.01 to 0.1 ml of serum were used for preparation of the AMS samples. Prior to chemical processing, 5 mg of “A was added as carrier and all serum samples were treated with trypsin in order to release the aluminum from the transferrin to which most of the upturn in the blood is bound [13,18]. Two techniques (fig. 1) were used to convert the aluminum in the solutions to
A
A
5 mg of Al I
I
2 ml of 5% oxine
iii1 r--z&~7 r dryness j / to
filter
I
Fig. 1. Plow diagram of extraction of aluminum from blood and urine samples.
53-I
aluminum oxide. In one, the entire solution was dried and the resulting residue ashed. In the other, oxine was used to selectively precipitate the burnt 1191 prior to the ashing. 26A1concentrations in the serum and urine samples were determined from 26A1/27A1ratios measured at the AMS facility at the University of Pennsylvania [20]. The ratios ranged from lo-‘* to lo-r4 allowing their deter~nation with statistical and systematic~ errors of l-10%. The conversion of these ratios to 26A1concentrations did not require the determination of the “AI concentrations because the carrier added to the samples completely dominated the aluminum native to the urine and serum. In addition, the carrier ensured that the measured 26Al,/27Al ratio would not be affected if any ‘?A1 was picked up during handling. The 26A1 background level was determined from samples collected before the injections and from the control rat (the natural background of the 26A1/27A1ratio is less than lo-r4 [20]) and was found to be negligible.
3. Rest&s Fig. 2 shows the fraction of the administered ‘!4l remaining in the serum and excreted in the urine as a function of time after injection. There is a rapid exponential decrease of the level of ‘“A1 over the first couple of days. Thereafter, from a concentration of about a tenth of a percent of the injected dosage, it decreases at a much slower rate. A three-compartment model (fig. 3) was constructed with one compartment representing the blood and the other two representing exchangeable pools in active equilibrium with the blood. The common approach of first order kinetics in which the clearance rate of an isotope from a compartment is proportional to the total amount of the isotope in that compartment was employed. In general, isotopic fr~tionation between “141 and 27A1will not occur if there is no kinetic isotope effect and if the two isotopes are in the same chemical form. Whilst the former probably holds, it is not known how fast the injected aluminum equilibrates with the native aluminum as aluminum can exist in various forms depending on the chemical evironment (m particular the pH). However, since most of the blood aluminum is attached to transferrin and since iron occupies only one-third of the metal binding sites [21], there are ample sites available for aluminum binding and hence we assumed that the administered aluminum was quickly incorporated into the transferrin molecules resulting in identical fractional clearance coefficients (&‘s). Under this assumption it was possible to use the same compartmental analysis for both isotopes. For 26A1there is no oral input and hence the model is IV(c). BIOMEDICAL
SCIENCES
538
0. Meirav et al. / AMS in AI metabolism studies
respectively, and the matrix A is given by:
IO0
-(R,+R,+R,) A(t)
10-l
w a,
R2
R,
RI
-R,
0
R4
0
-&
‘I.
(3)
For 27A1at steady state, dY(t)/dt = 0 and so:
4
-2
:
Pl(27Al)
'FlO .-
: ‘--d___
.
The aluminum
i
‘1% --__---__
1O-4
1o-5
I
0
5
-+ _
20
25
described mathematically by the following set of linear differential equations:
dY dt =AY,
(2) where B is the amount of the isotope in the blood, PI P2 the amounts in the first and second pools,
and
FOOD
INTESTINES _BLOOD
% R5
KIDNEYS
(6)
to solve eq. (1) analytically:
Fig. 2. The data shownis the fraction of the administered*“Al remainingin the blood and excreted in the urine as a function of time after injection as measuredin the present experiment. The horizontal bars indicate the duration of the urine collection. The curves are the resultsof a three-compartmentsimulation usingthe clearancevaluesgivenin the text.
R4
(5)
In general, the Ri’s are time dependent and may vary if steady state does not pertain. The amounts of aluminum injected in the present experiment were comparable with the total aluminum in blood but negligible compared with the total rat aluminum content. Thus it was assumed that the steady state was not disturbed and the Ri’s were taken to be constant. In this case it is possible
_yum ?-..
IO 15 days
(4)
lost daily by excretion is
U= R,B.
3
a:
R, = B(27Al)Rlr
P2(27A1)R5 = B(27A1)R4.
--,-_“‘!t$__
:lo-3 ?4
Ed
=
Y(t)
= i
c,E,, eat’,
(7)
i=l
where ai and Ea, are the eigenvalues and eigenvectors of the matrix A and the ci’s are normalization coefficients determined by the initial conditions for the administered 26A1: B(0) = 1; P,(O) = 0. A best fit to the data was obtained (fig. 2) for clearance coefficients of R, = 115, R2 = 22, R, = 27, R, = 3.5 and R, = 0.07 ((g/g)/day). The total aluminum blood pool was estimated to be 100 ng based on a total of - 13 ml of serum in a rat [22] and an aluminum concentration of 5-10 l.tg per litre of serum [14,23]. Thus, under the above mentioned assumptions, namely, no isotopic fractionation and constant Ri’s, our results correspond to an alumirmm excretion of 2.7 pg/day (eq. (6)) and to pool sizes of PI = 0.5 l.tg (eq. (4)) and Pz = 5.0 ltg (eq. (5)). These two pools represent about 0.3% and 3% of the total aluminum in a rat (based on estimated total aluminum in humans of 30-40 mg [7,8]). Approximately 75% of the 26A1was excreted in the urine during the first few days following the injection. The residence time of aluminum in the blood is l/( R, + R3 - 10 minutes, that in PI is l/R, - 1 hour and + R4) that in P2 is l/R, - 2 weeks.
Rl
1
6
,
R2
1 I
R3 URINE
Fig. 3. Box diagramof the three-compartmentaluminummodel that was used to simulate the experiment. The Ri’s are the fractionalclearancecoefficients.
4. summary The present work is the first application of AMS in the detection of a long-lived radioisotope tracer in a biological system and conclusively demonstrates its potential to enhance our understanding of aluminum metabolism and toxicity. Future studies will include intravenous and oral administration of the tracer, stud-
0. Meirav et al. / AMS ies of longer
duration in normal and uremic rats, and measurements of aluminum distribution in the brain as well as other organs and bones. Experiments similar to the present one are feasible in humans since the administration of - 30 pCi of 26Al, entailing a radiation dose of less than 0.1% of the natural background [24], would be sufficient. We believe that this approach represents a distinct breakthrough in the research of aluminum behavior in the human body.
Acknowledgements
We thank E. Nelson (Archaeology, Simon Fraser University) and J. Vogel (LLNL) for fruitful discussions, C. Orvig (Chemistry, University of British Columbia) and T. Ruth (TRIUMF) for advising on the chemistry, H. O’Brain (Los Alamos National Lab) and R. Kavanagh (Kellogg Radiation Lab, CALTECH) for providing the 26Al and D. Ridout (Medicine, University of Britisch Columbia) for assisting in sample collection and preparation.
References
111M.R. Wills and J. Savory, in: Aluminum
and Other Trace Elements in Renal Disease, ed. A. Taylor (Bailliere Tindall, London, 1986) pp. 24-37. PI A.C. Alfrey, G.R. LeGendre and W.D. Kaehny, N. Engl. J. Med. 294 (1976) 184. 131 Kidney Int. 29 suppl. 18 (1986) sl-~124. of Metals: Clinical and ex141 A.C. Alfrey, in: Toxicology perimental Research, eds. S.S. Brown and Y. Kodama (Ellis Horwood, Chichester, 1987). pp. 201-210.
539
in AI metabolism studies
[5] M.R. Wills and J. Savory, in: Metal Ions in Biological Systems, vol. 24, ed. H. Sigel (Marcel Dekker, New York, 1988) pp. 315-345. [6] D.R. Perl, ibid., pp. 259-283. [7] T.P. Kruck and D.R. McLachIan. ibid., pp. 285-314. [8] A.C. Alfrey, Kidney Int. 29 suppl. 18 (1986) s8-~11. [9] J. Savory, S. Brown, R.L. Bertholf, N. Mendoz and M.R. Wills, in: Methods in Enzymology, vol. 158, eds. J.F. Riordan and B.L. Vallee (Academic Press, San Diego, 1988) pp. 289-301. [lo] R.R. Rocker, A.J. Blotchy, J.A. Leffler and E.P. Rack, J. Lab. Clin. Med. 90 (1977) 810. [ll] W.D. Kaehny. A.P. Hegg and A.C. Alfrey, N. Engl. J. Med. 296 (1977) 1389. [12] J.E. Gorsky, A.A. Dietz, H. Spencer and D. Osis, Clin. Chem. 25 (1979) 1739. [13] S.K. Gupta, D.H. Waters and P.R. Gwilt, J. Pharma. Sci. 75 (1986) 586. [14] J.C.M. Ghan, M. Jacob, S. Brown, J. Savory and M.R. Wills, Nephron 48 (1988) 61. [15] D. Elmore and F.M. Phillips, Science 236 (1987) 543. 1161 A.E. Litherland, Ann. Rev. Nucl. Sci. 30 (1980) 437. [17] D. Elmore, Biol. Trace Elements Res. 12 (1987) 231. 1181 G.A. Trapp, Life Sci. 33 (1983) 311. [19] A. Vogel, A Text Book of Quantitative Inorganic Analysis (Longman, New York, 1978) pp. 436-437. [20] R. Middleton and J. Klein, Phil. Trans. R. Sot. London A323 (1987) 121. [21] A.C. Alfrey, in: (Marcel Dekker, [22] H.J. Baker, J.R. Laboratory Rat, pp. 411-412.
Aluminum and Health, ed. H.J. Gitelman New York and Basel, 1989) pp. 101-124. Lindsey and S.H. Weisbroth (eds.). The vol. 1. (Academic Press, New York, 1979)
[23] A. Taylor, B. Starkey and A.W. Walker, in: Aluminum and Other Trace Elements in Renal Disease, ed. A. Taylor (Bailhere Tindall, London, 1986) pp. 24-37. [24] E.E. Watson,M. Stabin and W.E. Bloch, MIRDOSE code 1988 version, Oak Ridge Associated Universities, Oak Ridge, Tennessee 37831, USA.
IV(c). BIOMEDICAL
SCIENCES