Content uploaded by Anita C Rudy
Author content
All content in this area was uploaded by Anita C Rudy on Sep 25, 2019
Content may be subject to copyright.
ELSEVIER
Vaccine, Vol. 15, No. 12113. pp. 1314-1318, 1997
@ 1997 Elsevier Science Ltd. All rights reserved
Printed in Great Britain
PII: SO264-410X(97)00041-8 0264-410X/97 $17+0.00
In viva absorption of aluminium-
containing vaccine adjuvants using
26
Al
Richard E. Flarend”, Stanley L. Hem-l-II, Joe L. White$, David Elmore”,
Mark A. Suckow§, Anita C. RudyY and Euphemie A. Dandashlit
Aluminium hydroxide (AH) and aluminium phosphate (AP) udjuvants, labelled with
‘6Al, were injected intramuscularly (i.m.) in New Zealand White rabbits. Blood and
urine samples were collected for 28 days and analysed for -“Al using accelerutor mass
s
?F
ectrometly to determine the absorption and elimination of AH and AP adjuvants.
_ Al was present in the first blood sample (I h) ,for both adjutants. The area under the
blood level curve for 28 days indicates that three times more nluminium was absorbed
porn AP adjuvant than AH adjuvant. The distribution profile of aluminium to tissues
was the same for both adjuvants (kidney > spleen > liver > heart > lymph node >
brain). This study has demonstrated that in vivo mechanisms are available to eliminnte
aluminium-containing ndjuvants after i.m. administration. In addition, the pharmaco-
kinetic profiles of AH and AP adjuvants are different. 0 1997 Elsevier Science Ltd.
Keywords: adjuvant absorption. antigen desorption. “‘Al
Vaccines usually contain an antigen and an adjuvant,
which potentiates the immune response to the antigen.
The adjuvant effect of aluminium-containing
compounds was first observed in 1926’. Since that time
aluminium hydroxide adjuvant and aluminium
phosphate adjuvant have been widely used in both
human and animal vaccines. These are the only
adjuvants that are currently approved for use in human
vaccines by the United States Food and Drug Admini-
stration (FDA).
A recent study’ has shown that aluminium hydroxide
(AH) adjuvant is crystalline aluminium oxyhydroxide,
AIOOH. It has a fibrous morphology and dissolves very
slowly in simulated interstitial fluid’. Aluminium
phosphate (AP) adjuvant is amorphous aluminium
hydroxyphosphate. It has a platy morphology and
dissolves more rapidly in simulated interstitial fluid
than AH adjuvant. Interstitial fluid contains three
organic acids which have an cc-hydroxy carboxylic acid
group (citric, lactic and malic acids), and are therefore
capable of chelating aluminium3-‘. A recent in vitro
study’ showed that citrate anion was able to dissolve
*Department of Physics, Purdue University, West Lafayette,
IN 47907, USA. tDepartment of Industrial and Physical
Pharmacy, Purdue University, West Lafayette, IN 47907,
USA. *Department of Agronomy, Purdue University, West
Lafayette, IN 47907, USA. §Laboratory Animal Program,
Purdue University, West Lafayette, IN 47907, USA.
llDivision of Clinical Pharmacology, Department of
Medicine, School of Medicine, Indiana University, Indian-
apolis, IN 46202, USA. I/Author to whom all correspond-
ence should be addressed. (Received 2 August 1996;
revised version received 8 January 1997; accepted 9
January 1997)
both AH and AP adjuvants, although AP adjuvant
dissolved more rapidly.
Vaccines containing AH or AP adjuvants arc usually
administered intramuscularly. The FDA limits the
quantity of the adjuvant to no > 0.85 mg aluminium
per dose. The disposition of aluminium-containing
adjuvants after intramuscular (i.m.) administration is
not understood. This is largely because the low dose of
aluminium does not cause detectable changes in the
concentration of aluminium normally present in blood,
urine or tissues. Measurement of -“Al by accelerator
mass spectrometry (AMS)‘.” offers the first opportunity
to directly determine if aluminium-containing
adjuvants are removed from the site of injection by
dissolution in interstitial fluid. In addition, AMS allows
the absorption, distribution and elimination profiles of
aluminium-containing adjuvants to be studied and
optimized.
MATERIALS AND METHODS
Adjuvants
‘“Al-containi;F AH adjuvant was prepared by adding
0.596 g, of an
‘“Al g AlC13 solution in 0.1 N HCI (170 Bq
or 0.24 pg ‘“Al g ‘) to 45 ml of 0.2 M AlCl+
Forty-five milliliters of a 0.6 N NaOH and 4 M NaCl
solution was added dropwise over 30 min to the AlCl,/
“‘AICI, solution with vigorous agitation. The precipitate
was repeatedly washed with 50 ml portions of double
distilled water (ddH:O) after centrifugation until the
supernatant was free of chloride as determined by the
absence of a precipitate when 0.1 M AgN03 was
1314 Vaccine 1997 Volume 15 Number 12/l 3
In vivo absorption of Al-containing vaccine adjuvants: R.E. Flared et al.
added. The washed precipitate was resuspended in
50 ml of ddH?O, filled into a sealed container and
placed in an 8O’C oven for 24 h. After heating, the
volume was adjusted to 57.1 ml with ddH?O. The
adjuvant suspension was autoclaved at 121°C for
20 min. A dose of 0.20 ml contains 0.85 mg Al. The
preceding procedure without the ‘“AIC13 was followed
to product an AH adjuvant for testing. The tests
showed that the AH adjuvant prepared by this
procedure exhibited the X-ray diffraction pattern and
infrared spectrum which are typical of AH adjuvant’.
“‘Al-containing AP adjuvant was prepared by
dissolving 3.7 g of alum [KAI(S0&12 H20] in enough
ddH,O to make 68 ml and adding 0.519 g of the ‘“AlCl,
solution in 0.1 N HCl (170 Bq ‘“Al g-’ or 0.24 /lg
“‘Al g ‘). A phosphate solution was prepared (0.3403 g
NaHJPO,*H,O, 0.3501 g NazHPOI and 5.5796 g NaCl)
in enough ddH:O to make 800 ml. The alum solution
was slowly added to the phosphate solution and
agitated until the solution was clear. The solution was
titrated with 1N NaOH with agitation until the pH was
7.1-7.2 to precipitate aluminium hydroxyphosphate.
The suspension was agitated for 2 h and the pH
readjusted to 7.1-7.2 with 1 N NaOH. The precipitate
was washed three times with 0.9% NaCl by centrifuga-
tion. After the third wash, the sediment was dispersed
in enough 0.9% NaCl to make 50 ml. The adjuvant
suspension was autoclaved at 121°C for 20 min. A dose
of 0.20 ml contains 0.85 mg Al. The preceding
procedure without the ‘“AlCl, was followed to produce
an AP adjuvant for testing. The tests showed that the
AP adjuvant prepared by this procedure was
amorphous by X-ray diffraction and the infrared
spectrum was typical of AP adjuvant’.
“‘Al-containing aluminium citrate was prepared by
dissolving 0.7606 g AIC13.6 Hz0 in enough ddH,O to
make 10 ml. Twenty-one microliters of the “‘AICI1
solution in 0.1 N HCl ( 170 Bq ‘“Al g ’ or 0.24 /lg
‘“Al g -‘) was added with mixing. A citric acid solution
was prepared by dissolving 0.6620 g of citric acid in
enough ddHIO to make 10 ml. The citric acid solution
was added to the AICl$“AlCl, solution and mixed. The
pH was adjusted to 7.4 with 0.1 N NaOH.
The specific activity of the ‘“Al-labelled adjuvants
was 15.9 Bq ml ’ for the AH adjuvant and
15.5 Bq ml ’ for the AP adjuvant. The specific activity
of the ‘“Al-labelled aluminium citrate solution was
1.07 Bq ml-~‘. Thus, the doses contained 3.2 Bq for the
AH adjuvant (i.m.), 3.1 Bq for the AP adjuvant (i.m.)
and 0.32 Bq for the aluminium citrate solution (intra-
venous; i.v.). Calibration errors were 3-5%.
Rabbits
Six female New Zealand White rabbits were used to
determine the in L?\JO absorption of the ‘“Al-labelled
adjuvants. They were conditioned for 21 days before
the study and their weights were 2.5-2.8 kg at the
beginning of the study and 3.2-3.7 kg at the end of the
study.
Two rabbits received an i.m. injection (0.2 ml of
‘“Al-labelled adjuvant followed by 0.1 ml of sterile 0.9%
NaCl to wash the syringe) of ‘“Al-labelled AH
adjuvant, two rabbits received a similar i.m. injection of
‘“Al-labelled AP adjuvant, one rabbit received an
equivalent iv. injection (0.3 ml of ‘“Al-labellcd
aluminium citrate followed by 0.1 ml of sterile 0.9%
NaCl to wash the syringe) of ‘“Al-labelled aluminium
citrate, and one rabbit received an equivalent i.m. dose
of AP adjuvant containing no ‘“Al as a cross-contami-
nation monitor. All rabbits received a total of 0.85 mg
aluminium.
The rabbits were killed 28 days after the injections
by sodium pentobarbital overdose. This study was
approved by the Purdue University Animal Care and
USC Committee and performed in accordance with all
federal regulations.
Sample collection
One milliliter of whole blood was collected at 0, 1, 2,
4, 6, 10. and 12 h and at 1, 2, 4, 6, 8, 12, 16 and 21 days.
Three milliliters of blood were collected at 28 days.
The samples were collected in 3 ml vials with premea-
sured ethylenediaminetetra-acetic acid and refrigerated
immediately.
Urine was collected for 24 h before dosing and for
the following intervals: O-5. 5-9 and 9-24 h, l-2, 2-4,
4-6, 6-8, 11-12, 15-16. 20-21 and 27-28 days. Urine
was collected in screened pans placed under the cages.
The pans were tilled with 2 I of water at the beginning
of each collection period. At the end of the collecting
period, the pans were agitated and 40 ml aliquots were
placed in 50 ml polypropylene centrifuge tubes and
immediately refrigerated. The total volume of liquid in
the pans when the aliquot was collected was recorded.
Tissue samples were collected after the rabbits were
killed on day 28. Whole brain, heart, left kidney, liver,
mesenteric lymph node and spleen tissues were
collected and frozen in comercial plastic freezer bags.
Bone (femur) samples were also collected, but these
samples were lost during chemical preparation. The
brain sample for one of the AP-dosed rabbits was also
lost during chemical preparation.
Sample preparation
Blood and urine samples were prepared for AMS
analysis by the addition of l-100 mg - Al carrier from
AliCl (ICP 10000 p.p.m. “AI standard). The samples
were then repeatedly digested in nitric acid (70%) at
80°C in a porcelain crucible and allowed to evaporate
to dryness. After two digestions in nitric acid, the
samples were ashed at 800°C to yield A120i powder.
This AlzOJ powder was then mixed with silver powder
in a 1:3 ratio by mass and analysed by AMS.
Tissues were prepared by first dissolving the tissue
in 20-200 ml (depending on tissue size) of nitric acid
(70%) in polyethylene bottles. Aliquots of the dissolved
tissue were then prepared as described above except
that hydrogen peroxide (30%) was used as well as
nitric acid in the wet digestion.
Data analysis
Since AMS measures relative amounts of lhAl and
“Al in samples. the actual recovery percentage of
aluminium during sample preparation is irrelevant
provided that the carrier “Al is homogenized with the
-“AI native to the sample. In order to test the repro-
ducibility of the carrier addition, sample digestion, and
AMS analyses, ten samples were separately prepared in
triplicate. The results for each of these samples agreed
Vaccine 1997 Volume 15 Number 12/13 1315
In vivo absorption of Al-containing vaccine adjuvants: FE Flarend et al
within 10% (standard error of the mean) or within the
AMS precision.
Cross-contamination of ‘“Al between the animals
was monitored by the measurement of samples from
the rabbit receiving no ‘“Al dose. Data was rejected if
the ‘“Al concentration in a given sample was not at
least five times higher than the equivalent sample from
the cross-contamination monitor. Also, the “‘Al
concentration in blood, urine and tissue samples from
the cross-contamination monitor rabbit was subtracted
from the ‘“Al concentration in equivalent samples of
the other rabbits.
Cross-contamination of ‘“Al between samples during
chemical preparation was monitored with the prepara-
tion of chemistry blanks. In no case did these blanks
indicate more than a 1% cross-contamination during
chemical preparation. Chemistry blanks are samples
that are prepared alongside experimental samples.
These blanks undergo the same preparation procedure
in order to monitor any possible cross-contaminatin of
‘“Al between samples during the chemical preparation
of experimental samples.
All AMS analyses were conducted at the Purdue
Rare Isotope Measurement Laboratory, PRIME Lab”.
Although all samples were analysed for ‘hAl content,
data is reported in terms of aluminium arising from the
‘“Al-labelled adjuvants or ‘hAl-labclled aluminium
citrate. The result for the 4 h blood sample for rabbit 1
was rejected and not included in any analysis due to an
error in the recording of data for that sample.
RESULTS
Figure 1 shows the time profile for the aluminium
blood concentration of the four rabbits receiving the
‘“Al-labelled adjuvants. The blood level curve of both
adjuvants exhibit an absorption phase and an elimina-
tion phase, as is typical of i.m. administration. It is
noteworthy that ‘“Al was found in the blood at the first
sampling point (1 h) for both adjuvants. Thus dissolu-
tion of the adjuvants in interstitial fluid begins upon
0 200 400 600 800
elapsed time (hr)
Figure 1 Blood concentration profile after i.m. administration of
ZGAl-labelled aluminium hydroxide adjuvant: n , rabbit 1; l , rabbit
2; A, mean; or aluminium phosphate adjuvant: J, rabbit 3; ,
rabbit 4; A, mean
1316 Vaccine 1997 Volume 15 Number 12/13
administration. The aluminium concentration produced
by AH adjuvant at 1 h was similar to the concentra-
tions found from 2 to 28 days.
The mean area under the blood concentration
versus time curve (AUC) from days 0 to 28, deter-
mined using the trapezoid rule, was 1.6 x IO ’ mg h g ’
for the i.v. dose of ‘“Al-labellcd aluminium citrate
(n = I): X.1 x 10 a mg h g ’ for the “‘Al-labelled AP
adjuvant (n = 2); and 2.7 x 10 ’ mg h g ’ for the
“‘Al-labelled AH adjuvant (17 = 2). Thus. three times as
much aluminium was absorbed from the AP adjuvant
as from the AH adjuvant within 28 days. However,
during the first 48 h (Figure I insert), the AUC of the
AH adjuvant was 1.4 times the AUC of the AP
adjuvant. These data also indicate that 17% of the AH
adjuvant and 51%. of the AP adjuvant were absorbed
within 28 days based on the AUC of the i.v. dose of
“‘Al-labelled aluminium citrate. The blood concentra-
tion of aluminium for each of the rabbits receiving an
adjuvant had not reached a terminal elimination phase
by day 28.
Cumulative urinary excretion of aluminium (Figure
2) indicates that the body is able to eliminate the
aluminium absorbed from the adjuvants. The cumula-
tive amount of aluminium eliminated in the urine
during the 28 days of the study was 6% of the AH
adjuvant dose and 22% of the AP adjuvant dose.
Aluminium from both adjuvants was still being
excreted at a steady rate at day 28.
The pharmacokinetic parameters determined from
the blood and urine data are presented in Tut& 1.
Distribution of aluminium in tissues 28 days after
administration of AH and AP adjuvants is shown in
Figure 3. For each tissue. the concentration of
aluminium was greater in the rabbits which received
AP adjuvant. The average aluminium tissue concentra-
tion was 2.9 times greater for AP adjuvant than for AH
adjuvant.
DISCUSSION
It is noteworthy that the aluminium concentration
produced by AH adjuvant at the first sampling point
3.OE-1
2.5E-1
2.OE-1
1.5E-1
1 .OE-1
5.OE-2
O.OE+O
0 200 400 600 800
elapsed time (hr)
Figure 2 Cumulative urinary excretion of aluminium after i.m.
administration of ‘“Al-labelled aluminium hydroxide adjuvant: .,
rabbit 1; l , rabbit 2; 4, mean; or aluminium phosphate adjuvant:
D, rabbit 3; -, rabbit 4; ,L, mean. Error bars of ~5% are not
shown
In vivo absorption of Al-containing vaccine adjuvants: R.E. Flarend et al.
(1 h) was similar to the 2-28 day concentrations. This
indicates that dissolution of aluminium-containing
adjuvants in interstitial fluid begins quickly after i.m.
administration. It is surprising that the aluminium
concentrations were greater during the first 24 h for
crystalline AH adjuvant than for the amorphous AP
adjuvant. This suggests that the initial rate of dissolu-
tion from the edges of the fibrous AH adjuvant
particles is greater than from the platy AP adjuvant
particles.
The rapid appearance of aluminium in the blood
may have implications for theories regarding the
mechanism of adjuvant action of aluminium-containing
adjuvants. The most widely accepted theory is the
repository effect”‘, whereby the antigen adsorbed by
the aluminium-containing adjuvant is slowly released
after i.m. administration. The rapid appearance of
aluminium as seen in the insert of Figure I challenges
the repository mechanism as it is likely that the
adsorbed antigen would be quickly desorbed as a result
of the fast initial dissolution of the substrate.
After 2 days, the absorption rate for AP adjuvant
was considerably more than the AH adjuvant which
confirms the difference in irz vitro dissolution rates in
simulated interstitial fluid3. The blood concentration of
aluminium was fairly steady from days 2 to 28
Table 1 Pharmacokinetic parameters after i.m. injection of
“Al-containing aluminium hydroxide and aluminium phosphate
adjuvants
Cumulative
AUC for aluminium in
O-28 days % Absorbed urine after
Adjuvant (mg h g-‘) in 28 days 28 days (%)
Aluminium hydroxide
Rabbit 1 2.0 x 1om4 13 5.0
Rabbit 2 3.5x10-” 22 6.2
Average 2.7~10~~ 17 5.6
Aluminium phosphate
Rabbit 3 2.7 x 10m4 47 10
Rabbit 4 8.7 x 10m4 55 33
Average 8.1 x 10m4 51 22
1 E-4
1 E-8
do
.
A
n
Kidney Spleen
MA*
indicating a relatively constant absorption rate for each
adjuvant even 28 days after i.m. administration. No
terminal phase had been reached for the blood concen-
tration of aluminium so it is difficult to determine the
mean residence time of each adjuvant, It is clear,
however, that AP adjuvant will be eliminated before
AH adjuvant because the long term absorption rate of
the AP adjuvant is greater.
The measured increase in the plasma concentration
of aluminium from the i.v. dose was ca 600 ng ml ‘,
which is considerably more than the increase of
2 ng ml ’ from the i.m. dose. Since it has been shown
that the pharmacokinetics of aluminium depend on the
concentration in the blood”, the pharmacokinetics of
the i.v. bolus dose were probably somewhat different
from those of the i.m. dose. Thus the AUC from the
i.v. dose may not provide a completely accurate
baseline for determining the fraction of the aluminium
absorbed from the i.m. administration of the AH and
AP adjuvants. However. this does not affect the
relative comparison of the AH and AP adjuvants.
The two rabbits which received AH adjuvant
exhibited very similar pharmacokinetic characteristics.
The blood level data for the two rabbits receiving AP
adjuvant were also very similar. However, the cumula-
tive urinary excretion of aluminium differed by a factor
of three between the two rabbits which received AP
adjuvant. This difference is probably due to intersub-
ject variability in the elimination of aluminium”. In
spite of this intersubject variation, the cumulative
urinary excretion of aluminium after 28 days in each
rabbit receiving AP adjuvant was greater than the
cumulative urinary excretion of aluminium in the
rabbits receiving AH adjuvant.
The normal pla:ma aluminium concentration in
rabbits is 30 ng ml . The maximum increase in the
plasma aluminium concentration from the 0.85 mg
aluminium doses of either adjuvant was ca 2 ng ml
This small increase would have been masked by the
aluminium background if ‘“Al-labelled adjuvants were
not used. If the same dose of these adjuvants was
administered i.m. to adult humans, an increase in the
plasma aluminium concentration of CN 0.04 ng ml-’
mAO
do
.
A
q @
3
A
I DA
f .*
Liver Heart L.N. Brain
Figure 3 Aluminium tissue concentration 28 days after administration of 26Al-labelled aluminium hydroxide adjuvant: ., rabbit 1; l , rabbit
2; A. mean; or aluminium phosphate adjuvant: cj, rabbit 3; ‘_, rabbit 4; a, mean. L.N., lymph node. Error bars of ~5% are not shown
Vaccine 1997 Volume 15 Number 12/13 1317
In vivo absorption of Al-containing vaccine adjuvanrs: RI. Flarend et al.
could he expected based on the larger blood volume of
humans and assuming the same rate of dissolution in
interstitial fluid. This represents a 0.8% increase in
plasma aluminium concentration based on a normal
value of 5 ng ml I’. This small change explains the
safety of aluminium-containing adjuvants and empha-
sizes the utility of AMS for studying aluminium
concentration in live.
The relative tissue distribution was the same for
both adjuvants (kidney > spleen > liver > heart >
lymph node > brain). This distribution pattern is
typical of results obtained when ‘“Al was given by other
routes of administration15. Since the concentration of
aluminium was 2.9 times greater on average in each
tissue (F&WY 3) for the rabbits which received AP
adjuvant, the tissue data is consistent with the ratio of
3.0 which was observed for the AUC of AP adjuvant
compared to AH adjuvant. Thus, the relative ‘“Al
tissue concentrations can be inferred from the ‘“Al
blood concentrations.
Since the adjuvants are being dissolved by interstitial
fluid which flows directly into the lymphatic system,
one may expect the aluminium concentration to be
quite high in the lymph tissue that was collected.
However, the i.m. doses were given in the hind quarter
where the ncarcst lymph node is difficult to isolate. For
this reason, the mesenteric lymph node. located in the
abdominal cavity, was removed. Thus the aluminium
from the dissolved adjuvants does not flow directly to
the lymph tissue that was collected and measured.
Dissolution, absorption, distribution and elimination
of aluminium-containing adjuvants after i.m. admini-
stration has been demonstrated by the use of
“‘Al-1abclled adjuvants. The two adjuvants studied
exhibited significantly different dissolution rates in
interstitial fluid which were rcflccted in different blood.
urinary excretion and tissue profiles. Human studies
using “‘Al-labelled adjuvants can be performed since
the radiation exposure to “‘Al is negligible. There was
I.6 Bq “‘Al used in each rabbit. In humans, CII 74 Bq
“‘Al would need to bc used resulting in a maximum
whole body exposure to radiation of CI~ 15 !tSv year ’
compared to the natural background exposure of
3000 I&V year “.
The application of AMS to the in \il>o performance
of vaccines should lead to a fuller understanding of the
mechanism of adjuvant action of aluminium-containing
adjuvants. The ability to label an aluminium-containing
compound with ‘“Al, as demonstrated in this study.
may prove useful in studying the in L~\YI absorption,
distribution, metabolism and elimination protilcs of
other aluminium-containing compounds.
ACKNOWLEDGEMENTS
This research was supported in part by the Showalter
Trust. PRIME Lab is supported by the National
Science Foundation.
REFERENCES
1
2
3
4
5
6
7
8
9
IO
11
12
13
14
15
Glenny, A.T., Pope, C.G.. Waddington, H. and Wallace, U. J.
The antigenic value of toxoid precipitated by potassium alum.
J. fathol. Bacterial. 1926, 29, 31-40.
Shirodkar, S.. Hutchinson, R.L., Perry, D.L., White, J.L. and
Hem, S.L. Aluminum compounds used as adjuvants in
vaccines. Pharm. Res. 1990. 7, 1282-1288.
Seeber, S.J., White, J.L. and Hem, S.L. J. Solubilization of
alummum-containing adjuvants by constituents of interstitial
fluid. Parenter. Sci. Technol. 1991, 45, 156-l 59.
Frisell. W. Human Biochemistry. McMillan, New York, 1982, p.
552.
Bell, G.H.. Emslie-Smith. D. and Patterson, C.K. Textbook of
Physiology and Biochemistry, 9th edn. Churchill Livingston,
Edinburgh, 1976, p. 416.
Selkurt, E.E. Physiology, 4th edn. Little, Brown, Boston, 1976,
p. 537.
Flarend, R.E. and Elmore, D. Aluminium in Infant’s Health and
Nutrition. eds P. Zatta, and A.C. Alfrey. World Scientific,
London, in press.
Elmore, D. and Phillips, F.M. Accelerator mass spectrometry
for measurement of long-lived radioisotopes. Science 1987,
236,543-550.
Elmore, D., Dep, L. and Flack, Ft. et al. The Purdue rare
isotope measurement laboratory. Nucl. Instrum. Methods
Phys. Res. 1994, 892, 65568.
World Health Organization. Immunological Adjuvants. World
Health Organization Technical Report Series No. 595, World
Health Organization, Geneva, 1976, pp. 6-8.
Wilhelm, M., Zhang, X.-J., Hafner, D. and Ohnesorge, F.K.
Single-dose toxicokinetics of aluminum in the rat. Arch.
Toxicol. 1992, 66, 700-705.
Talbot. R.J.. Newton, D., Priest, N.D.. Austin, J.G. and Day,
J.P. Intersubject variability in the metabolism of aluminum
following intravenous injection as citrate. Hum. Exp. Toxicol.
1995, 14, 595-599.
Ahn, H.-W., Fulton, B., Moxon, D. and Jeffrey, E.H. Interactive
effects of fluoride and aluminum: uptake and accumulation in
bones of rabbits administered both agents in their drinking
water. J. Toxicol. Environ. Health 1995, 44, 337-350.
Alfrey, A.C. Aluminium and Health: A Critical Review, ed. H.J.
Gitelman. Dekker, New York, 1989, pp. 101-124
Meirav. 0.. Sutton, R.A. and Fink, D. Accelerator mass
spectrometry: application to study of aluminum kinetics in the
rat. Am. J. Physiol. 1991, 260, F466-F469.
1318 Vaccine 1997 Volume 15 Number 12/l 3
