CLIN. CHEM, 30/5, 745-747 (1984)
Electrothermal Atomic Absorption Determination of Aluminum in Tissues
Dissolved in Tetramethyl Ammonium Hydroxide
Brian J. Stevens
in this rapid, precise method for the accurate determination
of aluminum in biological tissue, the only preparative step
required is the dissolution of the sample in hot aqueous
tetramethyl ammonium hydroxide, followed by dilution with
ethanol. Aluminum is measured by electrothermal (graphite
furnace) atomic absorption spectroscopy, with direct reference to aqueous standards. The CV5 for the method within-
‘day and day-to-dayare 3.0% and 6.8%, respectively.Analytical recoveryof added aluminum is86 to 108%, and matrix
effects are minimal. Measurement of aluminum in tissue from
normal laboratory rats and rats injected with aluminum gave
values close to those found by an acid digestion method, but
higher than those obtained by extraction with a saturated
solution of ethylenediaminetetraacetate.
AddItIonal Keyphrases: trace elements
rats
ments
dialysis encephalopathy
bone disease
Accurate determination
of aluminum
trace ele-
in blood, serum,
and biological tissue has become of importance in relation to
studies on the etiology of dialysis-associated
encephalopathy
(1) and renal-associated bone disease (2). Although several
methods of analysis for aluminum in serum have been
described (3-5), there are few procedures specifically
directed to itsmeasurement in tissue.
Atomic absorption spectroscopy is the most commonly
used instrumental technique for measuring trace metals in
biological material. Aluminum
has been determined
in
tissue by flame mode (6) or electrothermal atomization (7).
In either case, preliminary solubilization of the tissue in
mineral acids (6) or extraction of the aluminum with a
chelating agent (7) is generally required.
Tetramethyl ammonium hydroxide (miMi) has been used
as a tissue solubilizer in methods for analysis for several
metals (8, 9), but only one publication (10) deals with its
application to aluminum, and in this case unfavorable
results were reported.
In the present method, a dilute
aqueous solution of the quaternary base is used to dissolve
small samples of tissue before analysis by electrothermal
atomization. The method obviates wet or dry ashing, allows
direct standardization without additions, and avoids possible problems of incomplete extraction.
Materials and Methods
Apparatus
I used a Model 460 atomic absorption spectrophotometer
fitted with a Model 500 graphite furnace and Model AS4O
automatic sampler (all from Perkin-Elmer Corp., Norwalk,
CT 06852). The aluminum hollow-cathode lamp was supplied by S. and J. Juniper and Co., Harlow, Essex, England.
Department of Applied Biology, Royal Melbourne Institute of
Technology, Melbourne, Victoria, 3001, Australia.
ReceivedJuly 13, 1983; accepted February 7, 1984.
Results were recordedwith a chart recorder. Background
correction was used throughout.
I used screw-capped 12-mL polypropylene tubes (cat.no.
PGCT; Johns Professional
Products, Clayton, Victoria,
3167, Australia) in the tissue dissolution.
All tubes, volumetric ware, disposable pipette tips, etc.,
were pro-cleaned by soaking in a 20 g/L solution of disodium
ethylenediaminetetraacetate
(EDTA;
B.D.H.
Chemicals
(Australia) Pty. Ltd.,Port Fairy, Victoria, 3284, Australia),
rinsed with copious quantities of aluminum-free water, and
allowed to dry in a clean atmosphere. Polypropylene cups for
the sampler were stored in 100 g/L nitric acid and rinsed
immediately before use.
Reagents
All water used to wash laboratory ware and prepare
solutions and standards was purified by reverse osmosisand
mixed-bed de-ionization (Permutit Co. of Australia Pty.
Ltd., Brookvale, N.S.W. 2100, Australia). The water and
other reagents used were free of detectable aluminum.
Stock-standard aluminum solution was prepared by procleaning analytical-grade aluminum wire (B.D.H. Chemicals), then dissolving 1.000 g of it in the minimum necessary
amount of reagent-grade sulfuric acid. This solution was
dilutedto 1 L with water.An intermediate10 mg/L standard was prepared by further diluting 10.0 mL of stock
standard to 1 L with water immediately before use.
A 50 g/L aqueous solution of mi.i-z (cat. no. ‘11505; Sigma
Chemical Co., St. Louis, MO 63178) was prepared freshly for
use, not stored.
Ethanol (“Analar” grade, B.D.H. Chemicals) was used for
final dilution of the aqueous sample.
Materials for this work was obtained during the course of
a separate study. Nephrectomized and non-nephrectomized
male albino rats were injected subcutaneously
with either
isotonic saline (control group) or aluminum chloride solution once a day during 30 days. The animals were then
decapitated and various soft tissues taken for analysis.
Procedures
Cut small pieces of soft tissue (about 30 to 90 mg wet
weight), using stainless-steel blades and forceps, and rinse
the sample briefly in de-ionized water. Place the samples in
pre-cleaned plastic tubes and dry them overnight in a hotair oven at 120 #{176}C.
Cool the tubes and contents in a
desiccator, over silica gel, and weigh the samples accurately
after transferring them to pre-cleaned polypropylene tubes
fitted with screw-caps.Dry-sample weights should range
between 10 and 30 mg.
Add 2.0 mL of the aqueous miu solution, stopper the
tubes, and place them in a hot-air oven at 90 #{176}C
for 1 to 2 h.
Shake the tubes occasionally on a vibrating mixer during
this period. The tissue samples will be completely dissolved.
Cool the solutions to room temperature, dilute to 10 mL
with ethanol, and mix well.
Now prepare a blank and working standards by adding 0,
50, 100, and 150 pL ofthe intermediate aluminum standard
CLINICALCHEMISTRY, Vol.30, No. 5, 1984 745
to 2.0 mL of aqueous m.ii,
and dilute to 10 mL with
0.4
ethanol.
1.0 and
Table
Analyze
The final aluminum concentrations are then 0,0.5,
1.5 zg per tube.
1 gives the instrument settings for the analysis.
the solutions in duplicate for aluminum within 2 to
3 h of their preparation. Results are considered satisfactory
when peak heights agree to within 0.03 A over the working
range. If sample absorbances higher than that for the 1.5 pg
standard are obtained, repeat the measurement, using a
smaller sample volume.
-
0.2
-
0
z
4
0
Results
U)
Figure 1 illustrates a typical recorder trace of the standard curve. In the method as described, absorbance and
concentration are linearly related up to approximately 1 pg
per tube, although satisfactory sensitivity is seen over a
wide working range.
Precision studies. Replicate samples of dried powderedliver tissue, analyzed on a within-run basis, showed CVs of
2.0% and 3.0% at concentrations of 2.0 pg/g and 27.5 pg/g
dry weight, respectively (n = 20). The same samples when
analyzed on a day-to-day basis gave CV of 4.3% and 6.8%,
respectively (n = 20).
I compared results with the present method with those
obtained by using preliminary acid digestion of tissue, and
also by extraction of the aluminum with saturated EDTA
solution (7). Fifteen liver samples from normal and aluminum-treated rats were dried, powdered in an agate mortar,
and subdivided intothree portions.Aluminum
was measured in the following ways. (a) The present method was
applied as described here. (b) In the second method, the
weighed tissue samples were digested by heating with a
mixture of nitric, perchloric, and sulfuric acids (10:5:4 by
vol) and continuing the treatment until all the tissue was
oxidized and only the sulfuric acid remained. The samples
were then diluted to 10 mL with de-ionized water and
analyzed in the graphite furnace by comparison with blank
and standards similarly treated. (c) In the third method,the
weighed samples were extracted for 2 h with a saturated
aqueous solution of disodium EDTA (shown to be Al free), on
4
a rotary
0.3
0.1
0.0
A
B
Y
.971k+
Sy.U3i8
R1.998
Sintercept 1.17
Ssbp
.024
0
a
z
-J
4
Ya.758X-.378
Sy.a3.O2
graphite furnace with standards prepared in a similar
Samplevolume,20 .tL
Aluminum hollow-cathode lamp, 10 mA
Slit width, 0.7-nm bandpass
Wavelength, 309.3 nm
Deuterium background corrector, On
Graphitefurnacetube, pyrolyticcoated
Purge gas, argon, 300 mLimin
Dry cycle, 130#{176}C,
10-s ramp, 5-s hold
Char cycle, 1500#{176}C,
18-s ramp, 6-s hold
Atomizecycle, 2700 #{176}C,
0-s ramp, 6-s hold
Gas flow:interruptat atomization
746
CLINICAL CHEMISTRY, Vol. 30, No. 5, 1984
R.994
= 1.11
S810p. p.023
medium.
Table 1. instrumental Conditions for
Determination of Al in Tissue
D
.164
shaker. These extracts were then compared in the
The results are summarized in Figure 2. Although correlations are very high for both comparisons (p <.001),
statistical analysis by use of dependent t-tests on logtransformed differences between respective values at each
point show that the TMAHIacid digestion methods are equivalent (t = 0.943, not sigrnf.) whereas the EDTA extraction
method gives significantly lower values (t = 6.069,p <.001).
Deuterium
background correction was found to be necessary in this method, because non-atomic absorption enhancedthe total signal of tissue solutions by about 10%, but
contributed only negligibly to the standards.
C
Fig.1. Recorder tracing for duplicate standards of aluminuminrw,i#{247}ethanol
ank; 8, 50 g/L; C, 100 g/L; D, 150
0
20
ALUMINUM
(iJg)
Fig. 2. TissueAl as measuredby the presentmethodand by EDTA
extractioncompared with mineral acid digestion
O-O,
acid digestion (x.axis)vs EDTAextraction(y.axis);#{149} #{149},
acid
digestion(x-axts)vs TMAH dissolution(y.axis)
The predominant
inorganic ions in liver, kidney, and
brain are Nat, K, C1, PO,
and SO (12). To study
possible matrix interference by these ions, I added them as
Al-free salts to solutions of aluminum in miMi-ethanol,in
concentrations ranging up to three times physiological (24
mg/L for PO, 12 mg/L for the other ions)without measurable effects on the background-corrected aluminum signal.
Such studies were done at two aluminum concentrations, 50
and 100 p.g/L.
It is not possible to check the accuracy of aluminum
analyses in tissue by reference to independent standard
material
such as the U.S. National Bureau of Standards
Bovine Liver, Standard Reference Material no. 1577, because no certified figure is available. However, I used the
method to measure
aluminum
in three
NBS Standard
Reference Materials without specified values. Results were:
SRM
1577 Bovine Liver, <0.5 pglg; SRM
1566 oyster tissue,
a final diluting agent for aqueous Th1.AH
provides
a low-
13.3 (SD 1.2) pglg; SRM 50 AlbacoreTuna 3.6 (SD 0.6) pg/g.
These results were obtained for five separatemeasurements
viscosity medium that gives excellent precision without
automatic sample injection, and allows addition of aqueous
on each tissue. Analytical recoveries of aluminum added to
solutions of various tissues in miMi-ethanol in concentrations ranging from 13 to 107 pg/g were within the limits of
86 to 108% (Table 2).
standard
because
Discussion
In trace-metal
analysis, the risk of errors due to contami-
nation is high. Aluminum
is a particularly
difficult element
in this regard because of its ubiquitousdistribution.
It is
therefore desirable to minimize sample handling and the
number of reagents used. The direct technique used in the
present method is rapid, and it obviates external ashing
procedures. The same tube may be used for weighing,
dissolution,
and dilution of the sample, thus avoiding the
need for homogenizing or grinding procedures, which produce trace-metal contamination (11).
The method-comparison studies showed that acid digestion of the tissue and solution of the tissue in miMi-ethanol
both give higher results than does the EDTA method. This
suggests the possibility of incomplete extraction by EDTA
because of (e.g.) protein binding.
As little as 0.5 pg of aluminum per gram dry weight of
tissue can be determined at the limit of detection of the
method, assuming that 20 mg of tissue were to be analyzed
in a final volume of 10 mL. The sensitivity of the method is
such that matrix effects can be diluted out, allowing direct
standardization and avoiding the need for standard additions.
TMAH dissolved in toluene has been used by some workers
as a tissue-solubilizing agent, and lack of precision due to
viscosity has been reported (10). Also, if a toluene solution is
used it is not possible to add aqueous standard aliquots,
because these solvents are immiscible. The use of ethanol as
Table 2. AnalytIcal Recovery of Added Al from
TMAH Digests of Various Rat Tissues
Al present
Al added
Al found
pg Al/tube
Recovery, %
Muscle
0.21
1.40
0.5
0.5
0.55
1.32
0.5
0.5
0.20
0.81
0.68
1,94
108
0.98
1.86
86
108
0.5
0.68
0.5
1.27
96
92
94
Liver
solutions.
There is no build-up of residue in the graphite tube during
the course of analysis. The alkaline nature of the solution
appears to favor increased usable lifeof the graphite tube,
200-300
firings
can usually be recorded without
any deterioration in precision or sensitivity. This is longer
than the lifetime observed with acid solutions.
I thank Timothy Humphery, Ph.D., and Gregory Willis, Ph.D., of
Prince Henry’s Hospital, Melbourne, for providing the rat tissues;
Nola Biddle, B.A., B.Sc.,for statistical assistance; and Ruth Lennie
for excellent laboratory assistance. The work was entirely supportedbya grant from the Research Committeeof the Royal Melbourne
Institute
of
Technology.
References
1. Alfrey AC, LeGendre GR, Kaehny WD. The dialysis encephalopathy syndrome.Possiblealuminum intoxication.NEng1J Med 294,
184-188 (1976).
2. Ward MK, Feest TG, Ellis HA, et al. Osteomalacicdialysis
osteodystrophy;evidence for a water borne aetiological agent,
probablyaluminium. Lancet I, 841-845 (1978).
3. Alderman FR, Gitelman HJ. Improved electrothermal determinationof aluminum in serum by atomic absorptionspectroscopy.
Clin Chem 26, 258-260 (1980).
4. King SW,Wills MR, Savory J. Electrothermalatomicabsorption
determinationof aluminum in blood serum. Anal Chim Acta 128,
221-224 (1981).
5. Leung FY, Henderson AR. Improved determination of aluminum in serum and urine with use of a stabilized temperature
platform furnace. Clin Chem 28, 2139-2143 (1982).
6. Julshamn K, Andersen K-J, Willassen Y, Braekkan OR. A
routine method for the determination of aluminium in human
tissue samples using standard additionand graphite furnace atomic
absorption spectrophotometry. Anal Biochem 88, 552-559 (1978).
7. LeGendre GR, AlfreyAC. Measuring picogram amounts of
aluminum in biological tissue by flameless atomic absorption
analysis of a chelate. Clin Chem 22, 53-56 (1976).
8. Gross SB, Parkinson ES. Analysesof metals in human tissues
using base (mt..n) digests and graphite furnace atomic absorption
spectrophotometry.
At Absorpt Newsl 13, 107-108 (1974).
9. Murthy L, Menden EE, Eller PM, Petering HG. Atomic absorption determination of zinc, copper,cadmium, and lead in tissues
solubilized by aqueous tetramethylammonium hydroxide. Anal
Biochem 53, 365-372 (1973).
10. Julshamn K, Andersen K-J. A studyonthe digestionof human
muscle biopsies for trace metal analysis using an organic tissue
solubilizer.Anal Biochem 98, 315-318 (1979).
11. Bunker VW, Delves HT, Fautley RF. A system to minimise
trace metal contamination of biological material during homogenisation. Ann Clin Biochem 19, 444 445 (1982).
Brain
12. Documenta Geigy ‘Scientific Tables’, 6thed.,Geigy Pharmaceuticals Division, of Geigy (Australasia) Pty Ltd., St. Leonards,
N.S.W., Australia, 1962, p 516.
CLINICALCHEMISTRY,Vol.30, No. 5, 1984 747