From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
Comparison of the Subacute Toxicity and Efficacy of 3-Hydroxypyridin-4-One
Iron Chelators in Overloaded and Nonoverloaded Mice
By J.B. Porter, K.P. Hoyes, R.D. Abeysinghe, P.N. Brooks, E.R. Huehns, and R.C. Hider
Five orally effective iron chelators of the 3-hydroxypyridin-4one series have been administered intraperitoneallyto ironoverloaded and nonoverloaded male mice at a dose of 200
m g l kgl24 h for a total of 60 days to investigatethe effect on
iron loading and toxicity. There was a significant reductionin
hepatic iron at the end of the study in the iron-overloaded
mice with all compounds studied using chemical iron quantitation (P < .001) and with Perls' stain (P < .01). Liver iron
removal with the hydroxypyridinones ranged from 37% with
CP20 to 63% with CP51, compared with 46% removal for
desferrioxamine (DFO). There was no significant reduction in
splenic or cardiac iron with any chelator. There were no
s
EVE=
hydroxypyridin-4-one iron chelators have
been shown to be effective at increasing the excretion
of 59Fewith short-term studies in mice, rats, and rabbits by
the oral route.'-' However the choice of which compound
should be developed for use in humans is not ~lear.5.~
Kontoghiorghes et aI8have chosen to develop 1,Zdimethyl3-hydroxypyridin-4-one (CP2O)(L1) for initial studies in
humans and increased urinary iron excretion has been
shown in a number of iron-overloaded patients.8s9However,
there is evidence to suggest that this compound is both less
active7and produces a number of toxic effects not characteristic of the series of compounds as a w h ~ l e . We
~ ~ have
~~'~
identified some compounds that appear to have a superior
therapeutic safety margin than CP2O(L1) in short-term
animal ~ t u d i e sThe
. ~ purpose of the present investigations
was, therefore, to clarify which compound was best suited
for further development.
Because there is evidence both with desferrioxamine
(DFO)",'* and with the hydro~ypyridinones~
that iron overload protects against some of the toxic effects of iron
chelation, we chose to compare the subacute toxicity in
normal and iron-overloaded mice. The dose selected, 200
mg/kg/d, was deliberately greater than that which would be
administered to humans to identify possible target organs
for toxicity. Because previous work had shown that the
intraperitoneal (IP) and oral efficacy of these compounds
were ~omparable,~
the IP route of administration was
chosen for this study so that the long-term efficacy and
toxicity could be compared with DFO. Macroscopic post
mortem and microscopic histologic examinations of selected organs were performed at the end of the study period
as well as biochemical and hematologic analysis of plasma
and blood. Iron content of selected tissues (liver, spleen,
and heart) was also examined to compare the long-term
relative efficacy of these compounds and to determine from
which tissues iron was removed.
MATERIALS AND METHODS
Chemicals. The hydroxypyridin-4-ones, CP20, CP21, CP51,
CP93, and CP94 (Fig l), supplied as a powder, were synthesized
according to published procedures." The purity of each compound
was tested by elemental analysis. Samples of each material were
dissolved in phosphate-buffered saline (PBS) pH 7.4 once each
week and samples were stored for up to 1 week at 4°C. Stability
when stored under these conditions was monitored by high
Blood, Vol78, No 10 (November 15). 1991: pp 2727-2734
deaths in iron-overloadedanimals receivingany ofthe hydroxypyridin-4-ones, but significantly more deaths in the nonoverloaded groups as a whole (P < .03). No weight loss was
observed with any chelator. Significant reductions in hemoglobin and white cell count were observed with CPZO(L1). No
histologic abnormalities of kidney, spleen, bone marrow, or
stifle joints were observed. Intracytoplasmic inclusion bodies were observed in the centrilobular hepatocytes of animals administered each of the hydroxypyridin-4-ones, while
the DFO-treatedand control groups showed no such changes.
0 1991 by The American Society of Hematology.
performance liquid chromatography (HPLC) using a PLRP-S
column (150 x 4.6 mm I.D.; Polymer Laboratories, Shropshire,
UK) with an isocratic mobile phase of 20 mmol/L phosphate, 2
mmol/L EDTA (pH 7.0) containing 15% methanol. Detection was
at 280 nm. All compounds showed no detectable change when
stored at 4°C for periods up to 1month as shown by quantitation of
the pyridine peak height and a deliberate search for breakdown
products. This procedure is in contrast to DFO, which is less stable
and was therefore dissolved twice weekly. All chemicals were of
analytical grade unless otherwise stated.
Animal specification and housing. Six- to 8-week-old male
BALB-c mice were purchased from Harlan OLAC (Oxford, UK)
and divided into those to receive iron loading and those not to be
iron overloaded. For iron overloading, mice received IP injections
of 2 mg of iron dextran (approximately equivalent to 100 mglkg) at
weekly intervals for 4 weeks. One month was allowed for equilibration of iron after overloading.
Groups of iron-overloaded or age-matched, nonoverloaded mice
were placed in polypropylene cages with stainless steel lids. Tap
water and diet were given ad libidum with rat and mouse S.Q.C.
Expanded Diet No. 1 (Special Diet Services Ltd, Essex, UK).
Dosing and observations. Mice were administered IP injections
of 0.2 mL of each chelator (200 mg/kg) or PBS alone for a total of
60 doses on a minimum of 5 days each week. Mice were observed
immediately before dosing each day and again for at least 15
minutes after. Animals observed to be suffering or clearly unwell
were killed by cervical dislocation before the end of the study. Body
weights were measured weekly throughout the study.
Postmortem and histopathologic aramination. After 60 days
treatment, the mice were anesthetised with Hypnorm (0.01 mLl30
g) (Jansen Pharmaceuticals, Oxford, UK) and Diazepam (5 mglkg)
From the Department of Clinical Haematology, University College
and Middlesex School of Medicine and the Department of Pharmacy,
Kings College, London, UK.
Submitted November 26, 1990; accepted July 9, 1991.
Supported by grants from the National Institutes of Health (No.
ROI-HL-42800-01)and the British Thalassaemia Society.
Address reprint requests to J.B. Porter, MB, MA, MRCP, MRCPath,
Dept. of Clinical Haematology, Faculty of Clinical Sciences, University
College and Middlesex School of Medicine, 98 Chenies Mews, London
WClE 6HX, UK.
The publication costs of this article were defrayed in palt by page
charge payment. This article must therefore be hereby marked
"advertisement" in accordance with 18 U.S.C.section I734 solely to
indicate this fact.
0 I991 by The American Society of Hematology.
0006-4971191 17810-0032$3.00/0
2727
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
2728
PORTER ET AL
HYDROXYPYRIDIN-4-ONE STRUCTURES.
some mice, insufficient blood was obtained for blood chemistry
analysis.
Quantitation and distribution of tissue iron. After each organ
0
had been weighed, two pieces of tissue were dissected from the
main lobe of the liver and one from the apex of the heart and the
spleen. These samples were lyophilized for 48 hours and the dry
weight determined. After extraction of the non-heme iron with 3
mol/L hydrochloric acid (Aristar, BDH Chemicals, Poole, UK)
and 10% trichloracetic acid (TCA; Aristar, BDH Chemicals)," the
iron content of each sample was determined by atomic absorption
spectroscopy at 248.3 nm using an air/acetylene flame for ironoverloaded tissue samples (Perkin Elmer 2280) and a graphite
furnace (Perkin Elmer 1100 B) (atomization temperature 2,400"C)
for samples with low iron content. Tissue iron was expressed as
R2
nanomoles per milligram of dry weight of tissue.
Perl's staining technique was performed on spleen, heart, and
liver sections, analyzed 'blind,' and graded for intensity of Perl's
staining. Liver sections were analyzed separately for granular and
CP20
diffuse hemosiderin, representing reticuloendothelial and parenchyCP21
mal deposits, respectively.
CP5 1
Statistical analysis. Continuous numerical data were examined
CP93
for
statistical significance by analysis of variance using Scheffe's
CP94
test. This test allowed the contrasting of means and groups of
_-_-_____-____---___----__--------------means both between and within overloaded and nonoverloaded
treatment groups of mice. Discontinuous histologic data were
Fig 1. The general structure for the hydroxypyridin-4-ones is
graded on a 1 through 5 increasing severity scale and the results
shown together with the substitutions, R, and R2,for the compounds
studied.
were compared by x2 analysis for differences in the incidence of
lesions and by Kruskal-Wallis one-way nonparametric analysis of
variance for the comparison of severity grades.
(Pheonix Pharmaceuticals, Gloucester, UK) by intramuscular injection and killed by exsanguination from the carotid artery. MacroRESULTS
scopic postmortem examination of the abdominal, thoracic, and
intracranial tissues was recorded and weights of liver, spleen,
Effects of chelators on weight gain and survival. Each
kidneys, heart, and brain measured. Tissues were preserved in 10%
animal was weighed weekly and the percent weight gain for
buffered formalin. Tissue sections were cut at 5 pm and stained by
each group after 60 days of treatment is tabulated in Table
hematoxylin and eosin (H&E). Coded samples were examined
1, with a breakdown into overloaded and nonoverloaded
histologically so that their identities were unknown at the time of
animals. While all groups of mice gained weight, this was
microscopy. Those with autolytic changes or poor fixation were
less so with all hydroxypyridinone compounds, both in
excluded from analysis.
iron-loaded and nonloaded animals. Weight gain was least
Tissues examined by light microscopy were bone marrow, eye,
with CP51, CP93, and CP21 and greatest with DFO. None
gall bladder, heart, kidney, liver, spleen, and stifle joint. Bone
of these differences in weight gain reached statistical
marrow was graded for the presence of hypoplasia and pigment
significance and no consistent difference in weight gain
deposits; the heart for pigment deposits; the spleen for pigment
deposits, extramedullary hemopoiesis, and atrophy; and the liver
between overloaded and nonoverloaded mice is seen.
for the presence of parenchymal pigment deposits, inflammatory
The number of animals alive at the end of 60 days of
cell infiltration, hepatocellular necrosis, and the presence of
treatment in each group is shown in Table 1 together with
eosinophilic material in centrilobular hepatocytes.
the days of dosing on which each mouse died. It can be seen
Hematologic investigations. Whole blood was taken under termithat there were no deaths among the overloaded mice who
nal anesthesia into labeled tubes and 130 KL aliquoted into
received iron chelators, the only death being in one unpolypropylene Eppendorf tubes containing 195 kg of dried dipotastreated control animal. By contrast, in the nonoverloaded
sium EDTA and mixed thoroughly. Samples were analyzed on a
group, seven animals died before the end of the study
Coulter STK-R (Luton, UK) for hemoglobin (Hb), white blood cell
(P < .03, Fisher two-sided exact test).
count (WCC), platelet count (Plts), mean cell volume (MCV), and
Among the nonoverloaded mice, the proportion survivmean cell Hb (MCH). Samples with clots present were excluded
from analysis. Blood films were also made on each sample and
ing to the end of the study period was less with some
stained by Romanowsky staining. The remainder of the blood
chelators than with others (Table 1). Whereas there were
without anticoagulant was left for 30 minutes to allow clot
no deaths with DFO, there were deaths with all the
retraction and then spun at 1,SOOgfor 5 minutes to obtain serum for
hydroxypyridin-4-ones studied except CP21. None of the
blood chemistry analysis.
deaths occurred immediately after administration of chelaBlood chemistry investigations. Serum was frozen and stored at
tors but, generally, some hours later. In most, death was not
-20°C before analysis. Samples were analyzed on a multichannel
preceded by modification of behavior. Animals observed to
analyzer for urea, creatinine, albumin, bilirubin, total protein,
be clearly unwell were killed before the end of the study. In
alanine transaminase (AST), and alkaline phosphatase (Alk Phos).
such cases, histologic examination was generally perDue to a small degree of hemolysis in the majority of the
formed, but on occasions when the mice died before the
specimens, serum potassium and sodium were not measured. In
I
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
2729
IRON CHELATORS: SUBACUTE TOXICITY AND EFFICACY
Table 1. Weight Gain and Survival
Compound
~
Control
CP20
CP21
CP51
CP93
CP94
DFO
~.
Nonoverloaded
Overloaded
Total
Nonoverloaded
Overloaded
Total
Nonoverloaded
Overloaded
Total
Nonoverloaded
Overloaded
Total
Nonoverloaded
Overloaded
Total
Nonoverloaded
Overloaded
Total
Nonoverloaded
Overloaded
Total
No. of
No.of
Mice
Before No. of
Dosing Deaths
Doses
Before
Death*
8
5
13
9
6
15
6
6
12
6
6
12
5
5
10
8
5
13
8
6
14
Weight
Gain
(mean%) SEM
-
0
1
1
3
0
3
0
0
0
2
0
2
1
0
1
1
0
1
0
0
0
17.3
18.5
17.7
12.3
12.7
12.6
11.8
8.0
9.9
3.3
10.6
7.7
6.0
7.0
6.6
13.8
13.0
13.4
16.6
14.3
15.6
36
46,49,58
-
47,59
31
57
-
0.9
1.5
0.7
3.7
2.1
1.9
1.9
2.5
1.5
5.3
1.8
2.6
1.4
4.6
2.4
2.6
4.4
2.3
5.5
7.4
4.3
The mean weight gain 2 SEM after 60 days of treatment of mice is
shown, where n is the number of animals in each group. For animals
dying before the end of the study, the number of doses after which this
occurred is shown.
*There were significantly more deaths in the nonoverloaded mice
than in the iron-nonoverloaded mice as a whole (P < .03).
end of the study, histologic examination was not always
possible due to autolysis.
Post mortem findings. Post mortem examination was
performed macroscopically on each mouse at the end of the
study and on selected organs microscopically.No consistent
macroscopic abnormality was observed with any group at
the time of post mortem. There was no significant change in
organ weights of kidney or heart in any group. With the
spleen, the mean weight in the CP51 nonoverloaded group
(0.08 r 0.03 g) and the DFO nonoverloaded group was
reduced from the control (0.12 ? 0.01 g), although this
failed to reach significance. Liver weight was increased
(1.59 ? 0.09 g) in CP21-treated overloaded mice, compared
with the other iron-overloaded hydroxypyridinone-treated
mice as a group (P = .04).
Histologic findings. Liver, spleen, kidney, lung, bone
marrow, heart, stifle joints, and eyes were examined microscopically on H&E-stained tissues in each animal. Microscopic examination of all these tissues except the liver
showed no significant abnormality in any group. In the liver,
however, treatment-related abnormalities were noted (Table 2). Hepatocytes immediately surrounding the central
veins contained intracytoplasmic eosinophilic inclusion bodies in many of the animals treated with the hydroxypyridinones, but not with DFO. The presence of this material was
not associated with evidence of necrosis or inflammatory
cell infiltrate and no other abnormality was noted. The
number of centrilobular hepatocytes affected and the quantity of intracytoplasmic eosinophilic material present varied
between different groups of mice. In Table 2 it can be seen
Table 2. Liver Histology
Compound
Control
CP20
CP21
CP51
CP93
CP94
DFO
Cytoplasmic
Inclusions
Nonoverloaded
Overloaded
Total
Nonoverloaded
Overloaded
Total
Nonoverloaded
Overloaded
Total
Nonoverloaded
Overloaded
Total
Nonoverloaded
Overloaded
Total
Nonoverloaded
Overloaded
Total
Nonoverloaded
Overloaded
Total
Minimal
Granular
Slight
Granular
Moderate
Globular
Marked
Globular
5
4
9
1
4
5
0
0
0
0
0
0
1
2
3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
1
0
0
0
0
0
0
0
0
Absent
0
0
0
0
3
3
6
Severe
Globular
0
0
0
0
0
0
6
6
12
0
0
0
0
0
0
0
0
0
0
3
3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
4
5
5
6
1
1
1
1
2
4
3
7
2
0
2
0
0
11
0
n
X2
P
5
4
9
3
6
9
6
6
12
3
6
9
4
4
8
4
5
9
5
6
11
2.9
1.5
2.8
10.0
9.0
20.0
5.0
9.0
14.5
8.0
7.0
16.0
5.0
0.8
4.8
NS
NS
0.0
0.0
0.0
<.05
<.01
<.Ol
1.001
<.05
1.01
<.001
<.01
c.01
c.001
<.05
NS
<.05
NS
NS
NS
The presence of intracytoplasmic inclusion bodies in centrilobular hepatocytes after 60 days of treatment is graded for each group of mice. Each
column shows the number of mice possessing a given grade of intracytoplasmic inclusion. The significance, P,of the difference from the respective
iron-overloaded or nonoverloaded controls is shown for each group of mice.
Abbreviation: NS, not significant (P > .05).
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
2730
PORTER ET AL
that the condition is most marked with CP51 and CP21 and
least marked with CP20 and CP94.
Because adrenal enlargement had been noted in rats
treated with some hydroxypyridinones” (authors’ unpublished observations),z adrenal glands were examined in a
supplementary study in groups of nonoverloaded mice
(n = 3) after 60 doses of CP94, CP20, or DFO at 200
mg/kg. No enlargement or histologic differences were
observed from control (PBS)-treated mice.
Serum chemisby. Table 3 shows the serum values 2 the
SEM for urea, creatinine, albumin, bilirubin (total), AST,
Alk Phos, and total protein on overloaded and nonoverloaded mice, with the number of samples analyzed in
parentheses. There is no statistically significant difference
between control and chelator-treated mice for any values
shown except for albumin, which is higher than control
values with all hydroxypyridinones, reaching statistical significance where shown (Table 3). There is a wide variation
for AST both in the control and treated groups. Of interest
is the finding of a higher Alk Phos in the iron-overloaded
than in the nonoverloaded animals as a group (P = .04).
This difference is most marked in the control group and
with CP20.
Hematologic findings. Table 4 shows the values of Hb,
MCV, MCH, red blood cell count (RCC), WCC, percentage neutrophils (% Neuts), percentage lymphocytes (%
Lymphs), and Plts in overloaded and nonoverloaded mice.
There is a statistically significant reduction in WCC with
CP20-treated mice compared with control mice (P = .02)
and compared with the other hydroxypyridinone-treated
mice together (P = .0002). Although the WCC is less than
in control animals with the other hydroxypyridinones and
with DFO, this is not as marked and fails to reach statistical
significance. Differential counting showed a leucopenia
rather than a selective neutropenia in CP2O-treated mice.
There is also a significant reduction in the Hb in CP20treated mice compared with the other hydroxypyridinones
as a group (P = .04) associated with a reduction in the RCC
(P = .0002). The MCV is significantly increased in nonoverloaded mice administered CP20 compared with control
(P = .001) but not in the corresponding overloaded group.
Additionally there is a small decrease in the Plts in
nonoverloaded mice treated with CP51, although this fails
to reach significance from other groups of mice.
Tissue iron content and distribution. In the iron-overloaded mice, tissue iron distribution was examined by Perls’
stain of liver, spleen, and heart and iron was also quantitated per milligram of dry weight of these tissues.
Liver. The degree of granular and diffuse Perl’s positive
material in the liver of the different groups of ironoverloaded mice, together with the statistical significance of
the differences, is tabulated in Table 5. All hydroxypyridin4-ones and DFO produced a highly significant reduction in
diffuse pigment deposits (P < .001), the most marked
reduction being with CP51 and the least with CP20 and
CP93. By contrast, the reduction in granular deposits is not
significant for any compound.
Liver iron content per milligram of dry weight of tissue
after 60 days of treatment with the chelators in ironoverloaded mice is shown in Fig 2A, all compounds producing a highly significant reduction in liver iron (P < .001). It
can be seen that the changes in liver iron measured by
Table 3. Blood Chemistry
Compound
Control
CP20
CP21
CP51
CP93
CP94
DFO
Nonoverloaded
Overloaded
Total
Nonoverloaded
Overloaded
Total
Nonoverloaded
Overloaded
Total
Nonoverloaded
Overloaded
Total
Nonoverloaded
Overloaded
Total
Nonoverloaded
Overloaded
Total
Nonoverloaded
Overloaded
Total
n
Urea
(mmol/L)
Creatinine
(&mol/L)
Albumin
b/L)
Bilirubin
(wmol/L)
AST
(IUIL)
Alk Phos
(IUIL)’
Protein
19lL)
5
4
9
3
6
9
6
6
12
4
6
10
4
5
9
5
5
10
5
6
11
9.2 f 0.8
9.1 f 0.8
9.1 f 0.5
7.5 2 0.7
8.9 2 0.8
8.4 f 0.6
8.9 f 0.6
8.5 2 0.4
8.7 f 0.3
9.6 f 0.8
9.5 f 0.8
9.5 f 0.7
9.9 2 1.1
10.3 ? 1.2
10.0 f 0.7
10.8 f 1.2
11.4 f 1.3
11.1 f 0.9
9.8 f 1.1
9.8 f 0.6
9.8 f 0.6
56.8 7.4
61.5 f 10.2
5 8 . 9 ~5.4
56.0 f 11.0
60.7 f 10.1
59.1 f 7.3
68.3 2 8.6
61.7 2 10.3
63.6 f 4.8
58.5 -c 7.5
60.2 f 6.5
59.4 f 4.7
57.5 2 10.4
61.3 f 1.6
61.2 f 4.2
57.2 f 5.2
67.8 f 8.2
62.5 f 5.3
55.2 f 6.2
63.5 f 3.6
60.0 f 3.4
24.6 -c 0.3
24.5 2 0.6
24.6 2 0.3
28.3 t 1.5
25.2 2 0.3
26.2 t 0.7
28.0 2 0.4
27.2 2 0.7
27.7 2 0.3t
27.8 t 2.1
28.0 2 0.3
27.9 f 0.8t
27.0 z 1.7
26.7 f 0.4
26.9 +- 0.6
27.6 +- 0.3
26.6 +- 0.8
27.1 f 0.4
22.8 f 0.7
24.2 f 0.4
23.5 f 0.5
6.2 f 0.5
5.5 f 1.5
5.9 f 0.6
5.0 f 1.9
4.5 f 1.0
4.7f 0.9
4.8 2 0.8
3.3 f 0.7
4.1 f 0.6
6.0 f 0.8
5.3 f 1.0
5.6 f 0.6
4.5 2 1.2
5.7 f 0.6
5.3 f 0.6
6.6 f 1.7
5.0 f 0.9
5.8 2 0.6
4.4 f 0.9
6 . 0 f 0.7
5.3 f 0.6
589 2 156
443 2 72
524 2 87
428 2 184
440 t 52
436 f 57
386 f 32
376 2 58
396 f 39
497 2 105
424 f 66
4 5 5 f 55
413 f 91
418 2 9
422 f 33
732 2 127
600 f 116
666 f 83
499 f 68
487 f 38
493 f 35
147 2 19
217 f 26
178 f 19
152 2 5
206 2 15
188 2 14
152 f 7
160 f 14
159 f 8
139 f 34
167 10
155 f 14
134 2 19
162f9
150 f 9
148 f 6
187 2 28
170 2 15
129 f 17
169 2 40
149 f 22
50 h 0.6
51 3 0.9
51 h 0.5
51 3 2.8
51 2 0.8
51 3 0.9
53 2 0.7
53 h 0.9
53 h 0.7
54 h 2.3
52 h 0.8
53 2 1.0
53 3 1.3
5 3 2 1.4
53 h 0.9
54 2 1.2
5 4 3 1.2
54 -c 0.8
48 2 1.3
52 2 0.5
50 2 0.8
*
Serum chemistry values for each group of mice after 60 days of treatment with different iron chelators are shown. Values are the mean of n
observations f SEM.
*A significant difference of Alk Phos in comparing all the iron-overloaded with all the nonoverloaded animals (P < ,051.
tSerum albumin significantly different from control values (P < .01).
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
2731
IRON CHELATORS: SUBACUTE TOXICITY AND EFFICACY
Table 4. Hematology
ComDound
Control
CP20
CP21
CP51
CP93
CP94
DFO
Nonoverloaded
Overloaded
Total
Nonoverloaded
Overloaded
Total
Nonoverloaded
Overloaded
Total
Nonoverloaded
Overloaded
Total
Nonoverloaded
Overloaded
Total
Nonoverloaded
Overloaded
Total
Nonoverloaded
Overloaded
Total
6
4
10
5
6
11
5
5
10
3
5
8
4
4
8
6
5
11
8
5
13
4224
2824
38’4
39 f 3
51 f 5
47 f 4
33 f 7
25 f 4
29f4
483-9
26k 3
34 f 6
39 f 5
30 f 9
34 f 4
40f5
31 f 5
3424
34 2 7
35 f 4
3524
7.72 1.4
7.2 2 1.6
7 . 5 k 1.0
2.8 2 0.5
3.1 f 0.5
2.9f0.3t
5220.8
6.3 f 2.3
5.7 f 0 . 6
4.3 20.2
5.1 f 1.0
4.8 f 0.6
5.7 f 0.9
7.4 1.2
6.6 3- 0.8
5.02 1.0
5.6 2 0.9
5 . 2 ~
0.7
5.7 k 0.8
7.9 f 1.7
6 . 6 f 0.9
55f4
71 f 4
6024
59 2 2
49 2 6
5324
642 5
72 f 4
6824
5 0 f 10
72 2 4
64 f 5
59 f 5
68 2 9
64 2 5
59f3
66 2 6
6324
66 2 7
61 2 5
64f4
1.1962 107
1,0952 230
1,156 2 105
919 2 123
1,050 2 57
998 2 60
1.3142 105
1,464 f 69
1.389264
816289
992 f 43
926 f 51
1,161 f 76
1,188 f 116
1,210 f 67
1,131 k 135
992 f 142
1,068r96
1,257 f 82
1,370 f 58
1,300f55
15.3 f 0.3
15.9 f 0.2
15.5 f 0.2
13.8 f 0.5
13.7 f 0.7
13.7 f 0.4$
15.8 2 0.5
14.8 f 0.5
15.3 f 0.4
17.5 2 1.6
14.7 3- 0.6
15.8 2 0.8
16.2 2 0.1
15.7 3- 0.5
16.0 3- 0.3
16.5 2 0.6
14.8 2 0.3
15.7 2 0.4
14.4 2 0.9
14.9 2 0.5
14.7 k 0.5
9.3 f 0.5
9.9 f 0.1
9.5 f 0.3
7.5 f 0.6
8.5 f 0.4
8.1 f 0.45
9.8 f 0.4
9.3 Z 0.2
9.6 f 0.2
10.4 f 0.8
9.6 f 0.2
9.9 3- 0.3
10.1 f 0.2
10.1 2 0.3
10.1 f 0.2
10.3 f 0.3
9.4 f 0.2
9.9 f 0.2
9.3 f 0.4
9.7 f 0.4
9.6 f 0.3
46.9
46.8
46.8
55.5
47.8
51.3
46.6
46.5
46.5
50.7
47.2
48.6
48.2
46.1
47.2
47.3
45.9
46.6
47.5
46.1
46.9
f 0.5
f 0.3
f 0.3
f 3.0*
f 0.8
f 1.8
f 0.5
0.2
0.2
f 1.2
f 0.2
f 0.8
f 0.6
f 0.1
f 0.5
3- 0.2
2 0.3
2 0.3
5 0.3
3- 0.3
f 0.3
2
3-
Hematologic values after 60 days of treatment with iron chelators are shown as the mean k SEM of n observations.
*The MCV is significantlygreater than in control nonoverloaded mice (P= ,001).
tsignificant difference in WCC from corresponding control values (P= .02).
SHb differs significantlyfrom other hydroxypyridinones as a group (P= ,041.
SRCC differs significantlyfrom other hydroxypyridinones as a group (P = ,002).
atomic absorption parallel the reduction in diffuse pigment
deposits shown with the Perl’s stain (Table 5). CP51 is the
most effective chelator by both techniques with 63% of liver
iron removed. CP51 and CP21 were both significantly more
effective than CP20 (P < .05) with 37% of liver iron
removed (Fig 2A). This result compares with iron removal
of 44% for DFO and 50% for CP94. In contrast to the
Table 5. Perl’s Stain on Liver of Iron-OverloadedMice
Perk
Stain
Minimal Slight Moderate Marked Total
Compound Intensity
(n)
(n)
(nl
(n)
(n)
Control
CP20
CP21
CP51
CP93
CP94
DFO
Granular
Diffuse
Granular
Diffuse
Granular
Diffuse
Granular
Diffuse
Granular
Diffuse
Granular
Diffuse
Granular
Diffuse
0
0
0
0
1
3
0
6
0
1
0
1
0
2
0
0
0
4
1
3
0
0
2
3
0
4
2
4
3
3
2
2
4
0
6
0
2
0
3
0
0
0
1
1
4
0
0
0
0
0
0
0
2
0
4
0
4
4
6
6
6
6
6
6
4
4
5
5
6
6
x2
P
1.5
4.5
2.5
7.1
1.5
8.6
2.8
5.9
0.2
6.8
0.2
7.2
NS
<.01
NS
<.001
NS
<.001
NS
<.01
NS
<.001
NS
<.001
The appearance of the Perl‘s stain on liver tissue of iron-loaded mice
after 60 days of treatment is tabulated, where n is the number of mice
with each grade of staining intensity. The significance of both diffuse
and granular stain reduction, P,from iron-overloaded control mice is
shown.
Abbreviation: NS, not significant (P> .05).
iron-overloaded mice, there was no significant reduction in
liver iron in nonoverloaded mice treated with any compound (not shown).
Spleen. Figure 2B shows the effect of 60 days of treatment on splenic nonheme iron. It can be seen that no
significant reduction in splenic iron was shown either by
Perl’s staining or iron quantitation.
Heart. Figure 2C shows the quantity of nonheme iron
per milligram of dry weight of heart tissue at the end of the
study. Although there appear to be small reductions of iron
content with CP21, CP51, CP93, and CP94, these were not
statistically significant. Using the Perl’s stain, the majority
of stainable hemosiderin was not in the myocytes but in
reticuloendothelial cells and significant differences in stainable iron could not be seen.
DISCUSSION
A number of compounds have shown promise as effective
oral iron chelators over the past two decades only to be
rejected later because of unacceptable toxi~ity.~,’~~’’
A thorough investigation of toxicity-efficacyrelationships of putative oral iron chelators is therefore important before
commencing formal toxicity testing and clinical trials. It is
also clearly important to distinguish which toxic effects are
protected against by iron loading and which are independent of it, so that chelators are not rejected as too toxic
simply because they have chelated iron effectively, or
because toxicity is found in nonoverloaded animals.
For these reasons we chose to study the relative efficacy
and toxicity of several of our most promising bidentate
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
2732
PORTER ET AL
Fig 2. The tissue iron concentration in iron-overloadedmice after 60 days of treatment with chelators at a dose of 200 mg/kg IP daily is shown
as the mean ? SEM for liver (A), spleen (E), and heart (C), expressed as nanomoles of iron per milligram of dry weight of tissue. ***signifies that
values are significantly less than control values (P < .OOl).
hydroxypyridin-4-ones over 2 months in overloaded and
nonoverloaded mice and to compare these with DFO, the
only drug of proven clinical value. All compounds were
administered IP, it having been shown previously that
similar iron excretion was caused by the hydroxypyridin-4ones selected’ whether administered orally o r IP.
The finding that survival was significantly better in the
iron-loaded hydroxypyridinone-treated mice than in the
nonloaded mice receiving the same dose suggests that iron
loading protects against the toxic effects of prolonged
dosing with these compounds as well as against acute
toxicity. This finding has important implications for the
design of more detailed toxicity testing in animals and for
clinical studies in humans. There is a reduced weight gain in
both iron-loaded and non-ironloaded mice with three of the
hydroxypyridinone treatment groups (CP21, CP51, and
CP93) together with a smaller reduction in weight gain with
CP20 and CP94. While this fails to reach significance, the
trend suggests that 200 mg/kg is not far below a dose
causing more serious toxicity, and the deaths observed in
some groups of mice would support this.
Due to the number of compounds compared in this study,
histology was limited to major organs and those at that time
considered to be potential targets for chelation toxicity.
Thus, the eyes were examined because DFO has been
shown to be toxic in high doses clinically.’x The absence of
treatment-related abnormalities in the organs examined,
with the exception of the liver, is encouraging. The i n c h sions in the liver only occur with the hydroxypyridinones
and are absent with DFO. They are therefore likely to
represent effects caused directly by this group of cornpounds. The mechanism and nature of these liver changes
are unclear, but the lack of a significant difference between
the overloaded and nonoverloaded mice suggests that the
changes are unlikely to be directly due to iron chelation.
Similar cytoplasmic inclusions have been noted t o occur
spontaneously in the hepatocytes of otherwise normal mice
as well as in chemically treated mice and in benign
hepatomas.” The compounds associated with the most
marked histologic changes in the liver, CP21 and CP51, are
also those with the largest increase of liver weight in
nonoverloaded mice. Enlargement of the adrenal glands
has been observed in rats administered CP20” and we have
observed similar changes in male and female rats administered CP94 for 28 days.*’ However, the lack of adrenal
changes in mice at similar doses in this study suggests that
there are interspecies differences in the metabolism o r
toxicity of these compounds.
The range of serum chemistry analyses was limited by the
volume of serum that was obtainable from the mice. The
absence of significant abnormalities of renal function suggest that the hydroxypyridin-4-ones are unlikely to have
major toxic effects on this organ at this dose and this is
confirmed by histology. The significance of the increase in
the serum albumin in hydroxypyridinone-treated mice is
unclear. There is no increase in AST despite the inclusions
in the liver. The higher Alk Phos in iron-overloaded
compared with nonoverloaded mice suggests that the overload is adversely affecting liver function. The finding that
this difference remains greatest with the compound that is
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
IRON CHELATORS: SUBACUTE TOXICITY AND EFFICACY
least effective at removing liver iron (CP20) would support
this hypothesis.
It is unclear why the WCC should be depressed more
with CP20 than with the other hydroxypyridinones studied.
It is known that iron chelators may inhibit ribonucleotide
reductase, arresting dividing cells in G1 or S phases of the
cell cycle, and this has been shown for the hydroxypyridin-4ones.'" However, it would be expected that CP20 would
have had less of an effect than the other hydroxypyridinones
studied because it removes less iron from cellsz' and from
animals in short-term7 as well as in the present longer-term
studies (Fig 2A). One explanation for this difference may
be that CP20 is metabolized differently from the hydroxypyridin-4-ones with larger substituents in either the 1 or 2
positions. Analysis of urine from rats treated with CP20 or
CP94 shows significant differences in their metabolic degradation." The fact that the WCC was reduced in both the
iron-overloaded and the nonoverloaded animals suggests
that iron overload does not protect against this potentially
toxic effect and that the mechanism may be independent of
iron chelation. However, the raised MCV is only seen in the
nonoverloaded mice (Table 4). Leucopenia has also been
noted in rats administered CP20 at 200 mg/kg orally.23How
these changes relate to the occurrence of neutropenia and
thrombocytopenia in two patientsz4 receiving CP20 is not
clear as the reduction in mice is of both lymphocytes and
neutrophils. Further studies on the mechanisms of these
effects are therefore required.
One theoretical objection to the short-term animal models used for screening many chelators is that the radio-iron
(usually "Fe) may overestimate the amount of iron mobilized. The demonstration that both DFO and the hydroxypyridin-4-ones can remove approximately half of the total
liver iron in overloaded mice over a 2-month period
suggests that 59Feexcretion using previous models'.' was
2733
representative of net iron excretion and not simply redistribution of 59Fe.In this context, it is interesting to note that
the compound that showed the greatest efficacy in shortterm mouse studies using "Fe (eg, CP51)7 is the most
effective in this subacute model. Conversely, the least active
hydroxypyridinone in this study measuring liver iron removal, CP20, was also the least effective of these compounds in the short-term model measuring "Fe e~cretion.~
After 60 days of treatment at this dose, iron has not been
removed significantly from the heart or the spleen (Fig 2B
and C). Analysis of Perl's stain of the heart showed that the
excess iron in the loaded mice was mainly epicardial and
endocardial rather than myocardial in distribution (not
shown). However, because most of the cardiac iron was not
in the myocardium, this model may not reflect what
happens in clinical iron overload. The results suggest that
iron is chelated from hepatocytes before reticuloendothelial or cardiac iron is removed in significant quantities.
In conclusion, this study confirms findings in hepatocyte
cultures" and in acute mouse models' that several of the
hydroxypyridin-4-ones are at least as effective as DFO at
mobilizing cellular iron, and that the liver is likely to be the
major organ for iron chelation initially in vivo. Several of
these or related hydroxypyridinones require further detailed toxicologic and metabolic studies to aid the selection
of a compound for clinical use. On the basis of this and
previous investigations, CP94 has the best balance between
toxicity and efficacy of the hydroxypyridin-Cones studied by
us to date. We have therefore chosen this compound for our
initial evaluations of the efficacy, metabolism, and pharmacokinetics of the bidentate hydroxypyridin-4-ones in ironoverloaded human volunteers. Formal toxicity studies in
two other animal species are in progress with this compound before clinical trials.
REFERENCES
1. Gyparaki M, Porter JB, Burke LC, Huehns ER, Hider RC:
8. Kontoghiorghes GJ, Aldouri MA, Sheppard L, Hofirand
Evaluation of in vivo hydroxypyridinone iron chelators intended for
AV: 1,2-dimethyl-3-hydroxypyrid-4-one,
an orally active chelator
treatment of iron overload by the oral route. Biochem SOCTrans
for treatment of iron overload. Lancet 1:1294,1987
14:1181,1986
9. Olivieri NF, Koren G, Hermann C, Chung D, McClelland R,
2. Gyparaki M, Porter JB, Burke LC, Huehns ER, Hider RC: In
Freedman M, St Louis P, Templeton D: Effective iron chelation
vivo evaluation of hydroxypyridinone iron chelators in a mouse
with L1 in patients with thalassaemia major. Iron balance and dose
model. Acta Haematol (Basel) 78:217,1987
response studies. Blood 74:51a, 1989 (abstr, suppl 1)
3. Kontoghiorghes GJ, Sheppard L, Hoffbrand AV, Charalam10. Porter JB, Hoyes KP, Abeysinghe R, Huehns ER, Hider RC:
bous J, Tiperkae J, Pippard MJ: Iron chelation studies using
Animal toxicology of iron chelator L1. Lancet 2:156,1989 (letter)
desferrioxamine and the potential oral chelator, 1,2,-dimethyl-311. Blake DR, Winyard P, Lunec J, Williams A, Good PA,
hydroxypyrid-4-one, in normal and iron overloaded rats. J Clin
Crewes ST, Gutteridge JM, Rowley D, Halliwell B, Cornish A,
Pathol40:404, 1987
Hider RC: Cerebral and occular toxicity induced by desferrioxam4. Kontoghiorghes GJ, Hofirand AV: Orally active alphaine. Q J Med 56:345,1985
ketohydroxy pyridine iron chelators intended for clinical use: In
12. Porter JB, Jaswon MS, Huehns ER, East CA, Hazel1 JWP:
vivo studies in rabbits. Br J Haematol62:607,1986
Desferrioxamine ototoxicity: Evaluation of risk factors in thalas5. Huehns ER, Porter JB, Hider RC: Selection of hydroxypyridinsaemic patients and guidelines for safe dosage. Br J Haematol
4-ones for the treatment of iron overload using in v i m and in vivo
73:403, 1990
models. Haemoglobin 12593, 1988
13. Hider RC, Kontoghiorghes G, Silver J: Pharmaceutical
6. Porter JB: Oral iron chelators, prospects for future developcompositions. UKpatent GB 2118176A,1982, London, UK
ment. Eur J Haematol43:271, 1989
7. Porter JB, Morgan J, Hoyes KP, Burke LC, Huehns ER,
14. Torrance JD, Bothwell TH: A simple technique for measurHider RC: Relative oral efficacy and acute toxicity of hydroxypyriing storage iron concentration in formalinised liver samples. S Afr J
din-4-one iron chelators in mice. Blood 76:2389, 1990
Med Sci 33:9, 1968
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
2734
15. Toth K, Sugar J: Intranuclear and intracytoplasmic inclusions in normal and neoplastic mouse hepatocytes, in Jones TC,
Mohr U, Hunt RD (eds): Digestive System. New York, NY,
Springer-Verlag, 1985, p 92
16. Cerami A, Grady RW, Peterson CM, Bhargava K The
status of new iron chelators. Ann NY Acad Sci 344:425,1980
17. Peter HH: Industrial aspects of iron chelators. Pharmaceutical applications, in Spik G, Montreuil J, Crichton RR, Mazurier J
(eds): Proteins of Iron Storage and Transport. New York, NY,
Elsevier, 1935,p 283
18. Davies SC, Marcus RE, Hungerford JL, Miller MH, Arden
GB, Huehns ER: Ocular toxicity of high dose intravenous desferrioxamine. Lancet 2:181,1983
19. Grady RW, Srinivasan R, Dunn JB, Hilgartner M W Toxicity associated with DMHP (Ll), the new oral iron chelator. Pediatr
Res 142A, 1990
20. Hoyes KP, Burke LC, Thomas NSB, Porter JB: Cell cycle
synchronisation in leukaemia cell lines using hydroxypyridinone
chelators. Br J Haematol74:27,1990 (abstr, suppl 1)
PORTER ET AL
21. Porter JB, Gyparaki M, Burke LC, Huehns ER, Sarpong P,
Saez V, Hider R C Iron mobilisation from hepatocyte monolayer
cultures by chelators: The importance of membrane permeability
and the iron binding constant. Blood 72:1479,1988
22. Hider RC, Singh S, Porter JB, Huehns ER: Development of
hydroxypyrinin-Cones as orally active iron chelators. Ann NY
Acad Sci 612327,1991
23. KontioghiorghesGJ, Nasseri-Sina P, Goddard JG, Barr JM,
Nottey f,Sheppard LN:Safety of oral chelator t l . Lancet 2:457,
1989 (letter)
24. Bartlett AN, Hoarand AV, KontoghiorghesGJ: Long term
trial with the oral iron chelator 1,2-dimethyl-3-hydroxypyrid-4-one
(Ll). Br J Haematol76:301,1990
25. Porter JB, Abeysinghe RD, Hoyes KP, Huehns ER, Hider
EC: Oral efficacy and formal toxicity of 1,2diethyl-3-hydroxypyridin4-one (CP94). Preclinical evaluation. Br J Haematol 77:1991
(abstr)
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
1991 78: 2727-2734
Comparison of the subacute toxicity and efficacy of
3-hydroxypyridin-4- one iron chelators in overloaded and
nonoverloaded mice
JB Porter, KP Hoyes, RD Abeysinghe, PN Brooks, ER Huehns and RC Hider
Updated information and services can be found at:
http://www.bloodjournal.org/content/78/10/2727.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.