Clinical Chemistry 43: 2413-2417, 1997;
(Clinical Chemistry. 1997;43:2413-2417.)
© 1997 American Association for Clinical Chemistry, Inc.
Fully automated measurement of total iron-binding capacity in serum
Hachiro Yamanishia,
Shigeki Kimura,
Shigeru Iyama,
Yoshihisa Yamaguchi and
Takehiko Yanagihara
a Author for correspondence. Fax 81-6-879-6635.
 |
Abstract
|
|---|
We established a method for fully automated measurement of total
iron-binding capacity (TIBC) in serum without separation of the unbound
excess iron after saturating serum transferrin. After saturation of
serum transferrin with an excess amount of iron (first step), the
unbound iron was eliminated by formation of a complex with ferrozine,
which was used as a chromogenic reagent (second step). For the TIBC
assay, iron dissociated from transferrin by shifting the pH to acidic
was reacted with ferrozine, and the increase in the absorbance at 570
nm was measured (third step). Because the iron used as a calibrator,
which was added to saturate transferrin, reacted completely with
ferrozine in the second step (elimination of unbound iron), the change
in the absorbance to generate a calibration factor could not be
monitored in the third step. To solve this problem, we used
N-(2-hydroxyethyl)ethylenediamine-N,N',N'-triacetic
acid (HEDTA) to complex with the iron added to saturate transferrin in
the second step. This made it possible to form an iron–ferrozine
complex at acidic pH because iron was dissociated from HEDTA at acidic
pH. The within-run CVs of this method were 0.66–2.43% at 17.7–77.0
µmol/L, and the day-to-day CVs were 1.06–1.57% at 29.9–60.4
µmol/L (n = 10). The correlation between the values obtained
with this method (y) and those from the direct TIBC assay,
which involved removal of unbound iron by ion-exchange resin
(x), was: y = 0.963x + 0.29
µmol/L (r = 0.973, Sy|x =
2.83, n = 59), and with the TIBC values calculated from the serum
iron concentrations and the unbound iron-binding capacities measured by
a direct colorimetric method (x) was: y =
1.01x - 1.06 µmol/L (r = 0.994,
Sy|x = 1.66, n = 51).
|
Introduction
|
|---|
Total iron-binding capacity
(TIBC)1
indicates the total amount of iron to saturate plasma or serum
transferrin, which is a binding protein for iron. The measurement of
serum iron, TIBC, and the percent saturation of transferrin with iron
[(serum iron/TIBC)100] has been used to assess the state of iron
deficiency and for other purposes in clinical medicine
(1).
Among the several methods for measurement of serum iron, the direct
colorimetric methods, which do not involve sample deproteinization, are
widely used and have been modified as a fully automated assay for
routine analysis (2)(3). On the other hand,
the measurement of TIBC consists of three steps: saturation of
transferrin by addition of an excess amount of iron, removal of unbound
iron by absorption with magnesium carbonate (4)(5)(6) or
ion-exchange resin (7), and finally determination of iron
that is dissociated from transferrin at acidic pH.
However, the step for removal of unbound iron, requiring
centrifugation, is an impediment to adaptation of the TIBC assay to a
fully automated analysis. Moreover, it is difficult to generate a
calibration factor necessary for the assay. When iron solution without
binding protein is used as a calibrator, the unbound iron is removed in
the elimination step, and therefore the reaction to obtain a
calibration factor cannot be monitored in the next step. The ideal
calibrator needs to bind a known amount of added iron at alkaline pH
and dissociate completely at acidic pH, as serum transferrin does. To
overcome this problem, we used as a calibrator a chelating agent that
has a large stability constant with iron at an alkaline condition and
succeeded in establishing a fully automated method for measurement of
TIBC.
|
Materials and Methods
|
|---|
reagent
Analytical-grade sodium dihydrogen phosphate, manganese sulfate,
copper sulfate, thiourea, ferric chloride, and Triton X-100 were
purchased from Wako Pure Chemical Industries; sodium hydrogen
carbonate, citric acid, and L-ascorbic acid were
purchased from Katayama Chemical. Ferrozine and
N-(2-hydroxyethyl)ethylenediamine-N,N',N'-triacetic
acid (HEDTA) were purchased from Dojindo Laboratory. Tris was purchased
from Sigma Chemical Co., and lipemic turbidity and bilirubin conjugate
were purchased from International Reagent. Hemolysates were prepared
from human whole-blood samples with added EDTA.
preparation of reagents
The first reagent (R1) was made up with 300 mmol/L Tris, 150
mmol/L sodium hydrogen carbonate, and 4.2 g/L Triton X-100 (pH 8.4).
The second reagent (R2) was made up with iron calibrator solution (180
µmol/L Fe3+ in 10 mmol/L HCl), which was diluted
10-fold with R1. The third reagent (R3) was made up with 10 mmol/L
ferrozine and 40 mmol/L L-ascorbic acid in Tris buffer (50
mmol/L, pH 5.0). The fourth reagent (R4) was made up with 600 mmol/L
citric acid and 25.6 mmol/L thiourea, adjusted to pH 2.0 with 300
mmol/L sodium dihydrogen phosphate. Physiological saline (NaCl 150
mmol/L) and 20 mmol/L HEDTA in 0.03 mmol/L NaOH were used as zero and
high calibrators, respectively.
apparatus
The data obtained by the fully automated TIBC assay were compared
with those obtained with conventional methods by use of the Model 7070
automated analyzer from Hitachi (also known as the Hitachi 911).
assay principle
Our fully automated assay consisted of three steps of reactions.
Table 1
shows the scheme of each step in our method. We then used HEDTA
as a calibrator to generate a calibration factor. This chelating agent
has a large stability constant with Fe3+ metal ligand
(logKML = 19.8) at pH 7–12 (8).
In the first step, serum transferrin was saturated by the iron
calibrator solution (180 µmol/L). Then the unbound iron was reduced
to Fe2+ by ascorbic acid and eliminated by formation
of a complex with ferrozine (9) used as a chromogenic
reagent (second step).
Because iron in the calibrator solution was bound to HEDTA used as
a calibrator, ferrozine did not form a complex with iron in this
elimination step.
The next step, that is, the final process of the TIBC assay, was
dissociation of iron from transferrin and formation of a
ferrozine–iron complex that was proportional to the TIBC under acidic
pH . The increase in absorbance at this step was monitored at 570 nm.
The iron calibrator that was bound to HEDTA was dissociated from
HEDTA under the acidic condition and formed a colored complex with
ferrozine.
The calibration factor was obtained from the absorbance change at
this step.
assay procedure
The calibrator solution or a serum sample (50 µL) was mixed with
200 µL of R1, then incubated with 50 µL of R2, and kept at 37 °C
for 3.3 min. At 4.9 min after addition of 50 µL of R3, 100 µL of R4
was added. The absorbance change was recorded for 5.9 min at 570 nm
with reference at 660 nm.
method for comparison
We used TIBC CUPTM (ESA) for saturation of serum
transferrin (7), and the transferrin-bound iron was
determined by UnimateTM (Hoffmann-La Roche)
(9), a direct colorimetric method that has ferrozine as
the color reagent. TIBC CUP contains an ion-exchange resin that
includes bound iron. When serum is added to the TIBC CUP, transferrin
in the serum is saturated with the resin-bound iron, and all excess
iron remains bound to the resin beads. We also calculated the TIBC
value from the serum iron concentration and the unsaturated
iron-binding capacity (UIBC) value obtained by Unimate and Unimate UIBC
(Hoffmann-La Roche).
|
Results
|
|---|
optimization studies
We examined the effect of ferrozine and ascorbic acid
concentrations on the overall reaction. The maximal reaction rate was
achieved when the ferrozine concentration was >0.56 mmol/L and the
ascorbic acid concentration was >2.22 mmol/L. We thus used 1.1 mmol/L
ferrozine and 4.4 mmol/L ascorbic acid in the fully automated method.
We also examined the effect of the HEDTA concentration used as a
calibrator, which behaved in the same manner as transferrin. When the
HEDTA concentration was <5 mmol/L, a slight increase in absorbance was
observed after addition of R3 (elimination step). When the HEDTA
concentration was 20 mmol/L, there was no increase of absorbance,
indicating that dissociated iron was not present. We thus used this
concentration as the calibrator.
The pH of the mixture of the sample and calibrator after addition of R3
was 8.39 ± 0.03, and the pH of the mixture was 3.90 ± 0.05
at the final step. The typical time course of the calibrator and human
serum is shown in Fig. 1
. The reaction between unbound iron and ferrozine ended within 1
min, and dissociation of iron from transferrin and formation of the
ferrozine–iron complex ended within 2 min after addition of R4.
View larger version (23K):
[in this window]
[in a new window]
|
Figure 1. Typical time course of overall reaction.
 , calibrator solution (20 mmol/L HEDTA);  , zero calibrator
(physiological saline); •, human serum.
|
|
To examine dissociation of iron from transferrin after addition of R4,
we determined the serum iron concentration by our method but without
saturation of transferrin and compared the results with those from the
Unimate method. The correlation between values obtained with our method
(y) and the Unimate method (x) was:
y = 0.965x - 0.99 µmol/L
(r = 0.995, Sy|x = 0.93, n =
40).
assay evaluation
To study the linearity of the calibration curve, we assayed the
calibrator in duplicate with six concentrations of R2 (0–144
µmol/L). The calibration curve was straight for iron concentrations
up to 180 µmol/L. We also assayed several human sera that were
diluted with physiological saline. The dilution curve was linear for
TIBC values up to 90 µmol/L (Fig. 2
).
View larger version (24K):
[in this window]
[in a new window]
|
Figure 2. Dilution curve of human serum.
Several human sera were serially diluted with physiological saline up
to fivefold.
|
|
We also examined the detection limit of our method by assaying the zero
calibrator (physiological saline) 10 times; the result (mean ±
SD) was 0.63 ± 0.40 µmol/L. The detection limit, defined as the
mean TIBC value of the zero calibrator + 3 SD, was 1.83 µmol/L.
To examine the precision of our method, we assayed pooled human sera.
As shown in Table 2
, the within-assay CVs ranged from 0.66% to 2.43% and the
day-to-day CVs were from 1.06% to 1.57%. Our method was compared with
the TIBC CUP method as well as the Unimate method.
As shown in Fig. 3
, the correlation between values obtained with our method
(y) and the TIBC CUP method (x) was:
y = 0.963x + 0.29 µmol/L
(r = 0.973, Sy|x = 2.83, n =
59) and with the Unimate method (x) was: y =
1.01x - 1.06 µmol/L (r = 0.994,
Sy|x = 1.66, n = 51).
View larger version (20K):
[in this window]
[in a new window]
|
Figure 3. Comparison of serum TIBC results.
A, the present method vs the TIBC CUP method; B,
the present method vs the Unimate method.
|
|
We found no interference with TIBC ranging from 27.2 to 58.7 µmol/L
by addition of 150 mg/L bilirubin, 100 µmol/L Mn2+,
100 µmol/L Cu2+, and 2460 formazine turbidity units of
the lipemic material to pooled human sera. Table 3
shows the negative effects of hemolysis on this method. We then
tested the absorption spectrum of the overall reaction in a hemolytic
serum sample (100 mg/L hemoglobin). The reaction with unbound iron and
ferrozine showed two peaks at 542 and 576 nm. After addition of R4, the
increase of absorbance occurred in the range of 600 and 700 nm with a
peak at 630 nm. The cause of this peak is unknown. When the absorbance
at 660 nm (reference) was subtracted from the absorbance at 570 nm with
the hemolytic sample, the absorbance change was about -25%, which was
a negative error as compared with the nonhemolytic sample. This unknown
peak was not observed when physiological saline was used as the sample
and hemolysate was added.
We also examined the interference by ferritin iron. Serum samples
containing ferritin concentrations of 2000–12 000 µg/L were
measured by our method, and the results were compared with those
obtained by the TIBC CUP method and the Unimate method. The values
obtained by our method showed no positive error as compared with two
other methods.
|
Discussion
|
|---|
Although several methods exist for measurement of TIBC
(4)(5)(6)(7)(10)(11), these methods need
to absorb unbound iron after saturation of serum transferrin by
magnesium carbonate or ion-exchange resin and removal by
centrifugation. Therefore, it is difficult to adapt those methods to
fully automated assay of TIBC. The TIBC value can also be calculated
from the serum iron concentration and the UIBC value (TIBC = serum
iron + UIBC), but we believe that the total amount of iron that is
really bound to transferrin should be measured as TIBC.
A calibration factor could be generated by the use of HEDTA as a
calibrator. The final absorbance of the calibrator after addition of R4
was the same as that of the zero calibrator (Fig. 1
). This indicated
that iron formed a complex with ferrozine completely in the presence of
HEDTA.
The values obtained by our method without saturation of serum
transferrin indicated that the serum iron concentrations correlated
well with the values obtained with the Unimate method. There also was a
good correlation between our method and the TIBC CUP method or other
methods. These data suggested that the binding sites of serum
transferrin were well saturated with added iron and that iron was
dissociated from transferrin after addition of R4.
Recently, we reported in patients with hyperferritinemia that the serum
iron concentration measured by the method proposed by the International
Committee for Standardization in Haematology
(12)(13) and a constant-potential coulometric
method had positive errors caused by liberation of iron from
circulating ferritin by deproteinization or the acidic solvent
(14)(15). However, no interference from
ferritin was observed with the present method. This result means that
transferrin-bound iron was released specifically from transferrin in
the acidic condition, even at high concentrations of ferritin.
In summary, our fully automated method does not require
any absorbents for removal of unbound iron or centrifugation. Moreover,
the total amount of iron bound to serum transferrin is measured as TIBC
without interference from other cations, bilirubin, lipemic material,
or ferritin. We thus believe that our fully automated method for serum
TIBC may be suitable for routine clinical use in the laboratory.
|
Footnotes
|
|---|
The Central Laboratory for Clinical Investigation, Osaka University Hospital, 2-15, Yamadaoka, Suita, Osaka, 565, Japan.
1 Nonstandard abbreviations: TIBC, total iron-binding capacity; UIBC, unsaturated iron-binding capacity; HEDTA, N-(2-hydroxyethyl)ethylenediamine-N,N',N'-triacetic acid. 
|
References
|
|---|
-
Brittenham GM, Danish EH, Harris JW. Assessment of bone marrow and body iron stores: old techniques and new technologies. Semin Hematol 1981;18:194-221.
[Web of Science][Medline]
[Order article via Infotrieve]
-
Tietz NW, Rinker AD, Morrison SR. When is a serum iron really a serum iron? The status of serum iron measurements. Clin Chem 1994;40:546-551.
[Abstract/Free Full Text]
-
Fujita T, Hamasaki H, Furukata C, Nonobe M. New enzymatic assay of iron in serum. Clin Chem 1994;40:763-767.
[Abstract/Free Full Text]
-
Ramsay WNM. The determination of the total iron-binding capacity of serum. Clin Chim Acta 1957;2:221-226.
-
. International Committee of Standardization in Haematology. The measurement of total and unsaturated iron-binding capacity in serum. Br J Haematol 1978;38:281-287.
[Web of Science][Medline]
[Order article via Infotrieve]
-
Cerriotti F, Cerriotti G. Improved direct specific determination of serum iron and total iron-binding capacity. Clin Chem 1980;26:327-331.
[Abstract/Free Full Text]
-
Jacobs JC, Alexander NM. Colorimetry and constant-potential coulometry determinations of transferrin-bound iron, total iron-binding capacity, and total iron in serum containing iron-dextran, with use of sodium dithionite and alumina columns. Clin Chem 1990;36:1803-1807.
[Abstract/Free Full Text]
-
Sillen LG, ed. Stability constants of metal–ion
complexes, 2nd ed. London: The Chemical Society, Burlington House,
1964:642pp..
-
Stookey LL. Ferrozine—a new spectrophotometric reagent for iron. Anal Chem 1970;42:779-781.
-
Huebers HA, Eng MJ, Josephson BM, Ekpoom N, Rettmer RL, Labbé RF, et al. Plasma iron and transferrin iron-binding capacity evaluated by colorimetric and immunoprecipitation method. Clin Chem 1987;33:273-277.
[Abstract/Free Full Text]
-
Starr RT. Use of an alumina column in estimating total iron-binding capacity. Clin Chem 1980;26:156-158.
[Abstract/Free Full Text]
-
. International Committee for Standardization in Haematology. Recommendations for measurement of serum iron in human blood. Br J Haematol 1978;38:291-294.
[Web of Science][Medline]
[Order article via Infotrieve]
-
. Iron Panel of the International Comittee for Standardization in Haematology. Revised recommendation for measurements of serum iron in human blood. Br J Haematol 1990;75:615-616.
[Web of Science][Medline]
[Order article via Infotrieve]
-
Yamanishi H, Iyama S, Fushimi R, Amino N. Interference of ferritin in measurement of serum iron concentration: comparison by five methods [Tech Brief]. Clin Chem 1996;42:331-332.
[Free Full Text]
-
Yamanishi H, Iyama S, Amino N. Ferritin iron interferes with recommended method for measuring serum iron [Tech Brief]. Clin Chem 1996;42:2042-2043.
[Free Full Text]
The following articles in journals at HighWire Press have cited this article:
|
|
|
|
 
T. Shan, Y. Wang, Y. Wang, J. Liu, and Z. Xu
Effect of dietary lactoferrin on the immune functions and serum iron level of weanling piglets
J Anim Sci,
September 1, 2007;
85(9):
2140 - 2146.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
H. Yamanishi, S. Iyama, Y. Yamaguchi, Y. Kanakura, and Y. Iwatani
Total Iron-binding Capacity Calculated from Serum Transferrin Concentration or Serum Iron Concentration and Unsaturated Iron-binding Capacity
Clin. Chem.,
January 1, 2003;
49(1):
175 - 178.
[Full Text]
[PDF]
|
|
|
|
|
|
|
H. Yamanishi, S. Iyama, Y. Yamaguchi, Y. Kanakura, and Y. Iwatani
Modification of Fully Automated Total Iron-binding Capacity (TIBC) Assay in Serum and Comparison with Dimension TIBC Method
Clin. Chem.,
September 1, 2002;
48(9):
1565 - 1570.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
G. Siek, J. Lawlor, D. Pelczar, M. Sane, and J. Musto
Direct Serum Total Iron-binding Capacity Assay Suitable for Automated Analyzers
Clin. Chem.,
January 1, 2002;
48(1):
161 - 166.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
