| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Drug Monitoring and Toxicology |
1
Bone and Mineral Research Unit, Hospital Central de Asturias, Instituto Reina Sofía de Investigación Nefrológica, Julián Clavería s/n, Oviedo 33006, Spain.
2
Department of Physical and Analytical Chemistry, Faculty
of Chemistry, University of Oviedo, Oviedo 33006, Spain.
a Author for correspondence. Fax 34-8-5106142; e-mail cannata{at}hca.es.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
It is now well established that serum aluminum is largely bound (>80%) to high molecular weight proteins (4)(5)(6)(7)(8)(9). Therefore, its removal by dialysis is limited unless a chelating agent such as desferrioxamine is used.
Desferrioxamine (DFO)1 (Mr 560) is a naturally occurring trihydroxamic acid obtained by fermentation of the bacterium, Streptomyces pilosus. DFO effectively chelates trivalent ions such as iron and aluminum. The above chelated compounds are called ferrioxamine (FeDFO, Ks = 10) (10) and aluminoxamine (AlDFO, Ks = 10) (11), respectively. DFO has been widely used since 1963 to remove iron via the kidneys in patients with iron overload (12). In contrast, the chelator has been used in the diagnosis and therapy of aluminum accumulation/toxicity in dialysis patients only since 1979 (13).
Given the increasing number of reports dealing with side-effects resulting from DFO therapy and the variation in interindividual sensitivity to the drug, assessing the optimal therapeutic schedule of DFO treatment is imperative. Many protocols have varied the time of infusion, i.e., before (14), in the first (15) or last (16) hours, or after dialysis (17). Postdialysis administration has been advocated to avoid intradialysis loss of free DFO, whereas predialysis infusion has been proposed to exploit the intradialysis removal and reduce the risk of toxicity that is possibly linked to high peak values of aluminum. Nevertheless, thus far no definitive evidence has been provided to support a unique choice.
In view of the above information, this study was undertaken to investigate the effect of two periods of time (1 and 44 h) between DFO infusion and dialysis on the formation of AlDFO and FeDFO in serum from patients receiving a single dose of 15 mg/kg of DFO.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
apparatus
The chromatographic determination of DFO, AlDFO, and FeDFO in the
serum ultrafiltration fraction was carried out with an partially inert
HPLC system equipped with the following LKB components: two high
pressure pumps (Model 2150) with titanium pump heads, an injection
valve (Model 2154–002) with a 200-µL loop of titanium, a
variable-wavelength spectrophotometric detector (Model 2151) with a
10-µL flow cell, and a recording integrator (Model 2221). All
connections used titanium tubing (LKB, Model 2135–602). The
chromatographic separation was performed on a Tracer Spherisorb ODS-2
column [10-µm beads, 250 x 4.0 (i.d.) mm, Teknokroma]
preceded by a guard column (ODS-TRC-160; Teknokroma).
Determination of aluminum and iron concentrations in the serum samples and the ultrafiltration fractions was performed by Electrothermal Atomic Absorption Spectrometry (ETAAS), using the following Perkin–Elmer equipment: a graphite furnace (Model HGA-600) coupled to an atomic absorption spectrophotometer (Model Z-3030) possessing Zeeman background correction, an automatic sampler (Model AS-60), and a PR-100 printer. All the instrumentation was housed in a cleanroom equipped with a filtered laminar air supply to avoid sample contamination from aluminum in dust.
Ultramicrofiltration experiments were carried out using an Amicon Micropartition System (MPS-1) fitted with Amicon YM5 membranes (nominal cutoff, Mr 5000 ).
Statistical comparisons were performed using Student's t-test for paired data incorporated into the Systat, Ver. 5.2, statistical package for the Macintosh.
patient protocol
Five patients on regular hemodialysis (HD) were studied during a
period of 2 weeks, following the protocol in Fig. 1
. Our procedures complied with the Helsinki Declaration of 1975
for human subjects, as revised in 1983. DFO was administered at a dose
of 15 mg/kg body weight. The DFO was diluted in physiological solution
and then infused for 30 min by endovenous route either at the end of
the dialysis (Fig. 1
, dialysis A) or 1 h before the dialysis (Fig. 1
, dialysis C). The dialyses were performed using cuprophane membranes
for ~4 hours.
|
Blood samples (3–4 mL) were collected from the arterial line at the
beginning (Pre-HD), middle (Half-HD), and at the end (Post-HD) of the
three dialysis sessions under study (Fig. 1
, A-C). In the case of
dialysis C, an additional sample was taken at the beginning of the next
dialysis (Fig. 1
, dialysis D).
sample storage and preparation
Blood was allowed to clot and was centrifuged for 20 min at
800g. Serum was transferred into polystyrene tubes and
stored at -20 °C until time of analysis.
Before HPLC analysis, serum samples (700 µL) were placed in the ultramicrofiltration cell and centrifuged at room temperature at 2600g for 40 min. This protein-free ultrafiltrate was injected (200 µL) into the HPLC system. The recovery of DFO, AlDFO, and FeDFO from serum samples with this ultramicrofiltration procedure was ~90% (18).
hplc procedure
AlDFO, FeDFO, and DFO concentrations in serum ultrafiltrates were
determined by reversed-phase HPLC, using our previously reported method
(18). Briefly, the three solutes investigated were separated
with a mobile phase of 130 mL/L acetonitrile in phosphate buffer (5
mmol/L, pH 3.5) at a flow rate of 1.5 mL/min on a C18
column and detected by ultraviolet absorption at 210 nm (DFO) and 220
nm (AlDFO and FeDFO).
The calibration plots were linear over the range tested (1–18 mg/L for
DFO, 0.5–3.3 mg/L for AlDFO, and 0.3–1.6 mg/L for FeDFO). The
relative SD observed oscillated between 3.1% to 4.1% with a detection
limit for supplemented ultrafiltrate serum (3 
B
criterion) of 0.14 mg/L for DFO, 0.10 mg/L for AlDFO, and 0.08 mg/L for
FeDFO (18).
etaas procedure
Total aluminum and iron concentrations in serum samples and the
ultrafiltrates were determined by ETAAS, using operating conditions
similar to those described previously (19)(20)(21). Serum
samples were diluted with ultrapure water 1:1 and 1:30 (by volume) for
total aluminum and iron determinations, respectively. No additional
treatment was necessary for the ultrafiltrates. Calibration graphs were
constructed in both cases, using aqueous calibrators, and no matrix
effects were observed. Detection limits in serum ultrafiltrates were
0.5 µg/L and 1.7 µg/L for aluminum and iron, respectively.
precautions to avoid contamination
Precautions to avoid aluminum contamination during the ETAAS
determination were taken carefully, as described previously
(19). In addition, all the plastic material of the MPS-1
micropartition system was treated as indicated previously
(20). The YM5 membranes were washed by ultrafiltration twice
with 1 mL of 0.1 mol/L sodium hydroxide, then three times with 1 mL of
ultrapure water, and finally with 1 mL of DFO (100 mg/L). Subsequently,
the membranes were washed with copious amounts of ultrapure water. This
rinsing process was repeated until membranes were essentially aluminum-
and iron-free; the washing process was checked by measuring the
aluminum and iron concentration in the ultrafiltrate after
ultrafiltering 1 mL of a DFO solution (100 mg/L).
|
Results and Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
). The
concentration of DFO determined in the ultrafilterable fraction of
these samples ranged from 0.6 to 11.8 mg/L (mean 3.35 ± 4.79
mg/L).
The total and ultrafilterable aluminum concentrations found in the
samples analyzed are given in Table 1
. As can be seen, 44 h after DFO infusion (dialysis B) the
total aluminum in the serum increased about threefold, whereas the
ultrafiltered aluminum increased by more than 10-fold because of the
mobilization of this element from deposits in the body by chelation
with DFO. As a result, the percentage of ultrafilterable aluminum
increased from 17.1 ± 1.6% before DFO infusion to 78.6
± 19.3% 44 h after DFO infusion (Table 1
). Results obtained for the
samples collected 1 h after DFO infusion (dialysis C) showed that,
in this case, the total serum aluminum content did not increase, but
again, ~72.9 ± 7.4% of aluminum was ultrafilterable (Table 1
);
there were no substantial differences in the ultrafilterable aluminum
percentage between the two strategies of chelator administration. It
appears that after 1 h, DFO is able to displace aluminum from
serum proteins (transferrin) but not from other body deposits (e.g.,
bone). However, before dialysis D (48 h later), the concentrations of
total and ultrafilterable serum aluminum increased (Table 1
), again
indicating the release of aluminum from deposits within the body. The
results reported above confirm our group's previous results
(22).
|
The concentration of aluminum bound to DFO in the serum samples was
also determined. The percentage of serum aluminum bound to DFO was
significantly higher when DFO was infused 1 h before dialysis
(38.8 ± 7.7%) compared with the percentage bound when the drug
was administered 44 h before dialysis (15.8 ± 8.0%)
(P <0.02; Table 1
). Results showed that only a small amount
of ultrafilterable aluminum was bound to DFO. Considering the low
concentrations of free DFO in these samples, probably because DFO is
metabolized rapidly (23) and because the binding region in
nearly all the metabolites of DFO remain unaffected and thus retain
similar metal-chelating abilities (24)(25), the
most reasonable explanation these results is that the aluminum present
in the ultrafiltrate is bound to both DFO and DFO metabolites. Thus, as
shown in Table 1
, 44 h after DFO infusion, the concentration of
aluminum bound to DFO is similar to that obtained 1 h after drug
administration. However, the total ultrafilterable aluminum
concentration is higher. This suggests that DFO and DFO metabolites are
able to remove aluminum from stores in the body.
Fig. 2
shows the changes in total, ultrafilterable, and DFO-bound
serum aluminum concentrations vs time after DFO infusion. As expected,
in both strategies of DFO administration, the concentration of aluminum
bound to DFO decreased continuously during the intradialytic period
(HD-B and -C) because of the removal of the low molecular weight
(Mr 584) AlDFO complex by dialysis. The amount
of aluminum bound to DFO also decreased during the interdialytic period
(Fig. 2
, HD-C and -D), in this case, probably because DFO was
metabolized.
|
In contrast, the total aluminum concentration in serum and the
ultrafilterable fraction decreased intradialytically (elimination by
dialysis of aluminum bound to DFO and DFO metabolites) but not
interdialytically (Fig. 2
) because DFO and DFO metabolites removed
aluminum from stores in the body and from serum transferrin during the
interdialytic period (between dialyses C and D). As a result, the
percentage of ultrafilterable aluminum observed after DFO
administration remained virtually unchanged during all periods of time
investigated, in agreement with our previous results (22),
whereas the percentage of aluminum bound to DFO decreased with time
during (dialyses B and C) and between (dialyses C and D; Table 1
)
dialysis sessions.
The data in Table 1
also show that, after DFO infusion, ~76% of
total serum aluminum is present in the ultrafilterable fraction and
that this percentage does not seem to be influenced by the time (1 and
44 h) between DFO infusion and dialysis (22). However,
the percentage of aluminum bound to DFO decreases from 38.8% 1 h
after DFO infusion to 15.8% 44 h after DFO infusion.
To confirm the ideas discussed above, the aluminum distribution in
serum ultrafiltrate was investigated. A serum sample from a patient
collected at the beginning of the dialysis (dialysis B, 44 h after
DFO infusion) with a total aluminum content of 372 µg/L was analyzed
for AlDFO according to the HPLC procedure with UV detection described
in Materials and Methods. In addition, column fractions were
collected manually (750 µL) in aluminum-free automatic sampler cups
and immediately analyzed off-line for aluminum by ETAAS, using the
operating conditions described in Materials and Methods. The
results are shown in Fig. 3
. The peak (detected at 220 nm) that eluted at 9.15 min
corresponds to the quantitative elution of the AlDFO complex that we
reported previously (18). The aluminum concentration
corresponding to this peak (aluminum bound to DFO) was determined by
HPLC with UV detection to be 89.6 µg/L, only 25% of the total
aluminum (355 µg/L) present in the sample injected. The UV elution
profile in Fig. 3
also shows the presence of important peaks at
retention times up to 5 min, which correspond to different unidentified
UV-absorbing species present in ultrafilterable serum (18).
The ETAAS elution profile of total aluminum (bar graph in Fig. 3
)
indicates that aluminum is present in fractions (9–10 min)
corresponding to elution of the AlDFO complex detected at 220 nm (Fig. 3
). These fractions contain ~23% of the total aluminum injected, in
agreement with the results reported above for AlDFO, using UV
detection. In addition, it can be observed in Fig. 3
that aluminum is
also present in the fractions at 1–5 min; the amount of aluminum found
in these fractions represents 65% of the total aluminum injected
(total aluminum recovery ~90%).
|
According with Sing et al. (24) and Lehmann and Heinrich
(25), the main metabolites of DFO are produced by
bioconversion of the terminal amino group of the DFO molecule into a
carboxyl group, leading to a shortening of the hydrocarbon chain length
of DFO, in which methylene groups are removed but the hydroxamic
chelation sites remain intact. Therefore, the complexes of aluminum
with these DFO metabolites are less hydrophobic (fewer
-CH2- groups) than the AlDFO complex. Consequently, they
are expected to elute faster than AlDFO in reversed-phase liquid
chromatography, where solution retention occurs mainly by hydrophobic
interaction with the hydrocarbon stationary phase. In fact, this
chromatographic behavior has been already described for iron bound to
DFO and DFO metabolites (23)(24)(25). Therefore, the aluminum
peaks detected by ETAAS (bar graph in Fig. 3
) at retention times <5
min could correspond to aluminum bound to DFO metabolites.
Unfortunately, these peaks cannot be detected by UV absorption at 220
nm because in our chromatographic system they are overlapped by the
peaks produced by other UV-absorbing serum components (Fig. 3
).
Other authors, however, have proposed the formation of unknown species (aluminum-binding proteins or a ternary complex of aluminum-DFO-protein) after the infusion of DFO. The molecular weight of those species has been reported to be 8000 (26)(27) or in the range between 20 000 and 60 000 (28). It has been suggested that the presence of this unknown species is induced by high serum aluminum concentrations, even without the administration of DFO (26). The molecular weight of the unknown species of aluminum found in our study must be <5000 because this is the molecular weight cutoff used to deproteinate the serum samples before the chromatographic process. Thus, the aluminum-induced Mr 8000 protein described before (26) cannot explain our findings.
Our findings also suggest that a great part of the aluminum removed
with the use of DFO is not in the form of aluminum-DFO complex (Fig. 3
). Additional speciation studies after DFO administration are
necessary to know the serum carriers of aluminum and the mechanism of
action of DFO for the treatment of aluminum overload.
Table 2
shows the results obtained for the determination of total and
ultrafilterable iron concentrations in the serum samples included in
this study. As expected, a small amount of the total serum iron
concentration is found in the ultrafilterable fraction after DFO
infusion; therefore, only 6.8 ± 6.6% and 11.3 ± 5.4% of
total iron was ultrafiltered 44 h and 1 h after DFO infusion,
respectively (Table 2
). These results agree with previous reports that
DFO cannot displace in vivo iron from the serum protein carrier
transferrin (29), because replacement occurs slowly in vitro
(30).
|
Again, as in the case of aluminum, not all of the iron present in the ultrafiltrate was bound to DFO. In fact, the concentration of iron as FeDFO found in the samples investigated was always below the detection limit (7 µg/L) of the HPLC procedure used (18). These results confirm our idea that DFO and DFO metabolites are implicated in the mobilization of iron and aluminum in the body. In fact, Sing et al. (24) have described the presence of three iron chelates, ferrioxamine, and two iron-binding DFO metabolites in the serum ultrafiltrate and urine of two patients treated with DFO.
Finally, regarding the patient benefit of using our new protocol of DFO
administration (i.e., predialysis infusion), we want to stress that the
most important goal we want to achieve with DFO is the removal of the
greatest amount of aluminum with the least risk of toxicity for the
patient. Previous studies have shown great efficacy in the removal of
aluminum, with lower interdialytic peaks of serum aluminum, when DFO is
administered 1 h before HD (14). The present study
confirms previous findings (14), demonstrating that using
DFO before dialysis allows us to obtain lower serum peaks of total and
ultrafilterable aluminum at the beginning and at the end of dialysis,
with a lower chance of aluminum redistribution in tissues during the
interdialytic period (Table 1
). This fact may decrease the possibility
of reentrance of aluminum into some organs, e.g., the brain, limiting
the toxic effect of the use of DFO (14). The delayed effect
of a single infusion of DFO observed in the second and third dialyses
(Fig. 2
), also suggests that a lower dose of DFO could be effective, as
has been recently described in studies using 5 mg/kg (31)
and very low doses of DFO, such as 0.5 mg/kg (32). The
combination of all of these new approaches–new schedules of DFO
administration and lower doses–may contribute to the reduction of the
toxicity of DFO without a loss of efficacy.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
