From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
Role of Antennary Structure of N-Linked Sugar Chains in Renal Handling of
Recombinant Human Erythropoietin
By Tadashi Misaizu, Shigeru Matsuki, Thomas W. Strickland, Makoto Takeuchi, Akira Kobata, and Seiichi Takasaki
To elucidate the role of the branched structure of sugar
chains of human erythropoietin (EPO) in the expression of
in vivo activity, the pharmacokinetic profile of a l e s s active
recombinant human EPO sample (EPO-bi) enrichedwith biantennary sugar chainswas compared with that of a highly
activecontrol EPO sampleenriched with tetraantennary
sugar chains. After an intravenous injection
in rats, lz5I-EPObi disappearedfrom the plasma with 3.2 times greater total
body clearance (Cl,J than control '%EPO. Whole-body autoradiography after 20 minutes of administration indicated
that the overall distribution of radioactivity is similar, but
'=I-EPO-bi showed a higher level of radioactivity in the kidneys than control '%EPO. Quantitative determination of radioactivity in the tissues also indicatedthat radioactivity of
'%EPO-bi in the kidneys was two times higher than that
of control lPI-EPO. The difference in plasma disappearance
between '251-EPO-biand control lZ51-EPOwas not observed
in bilaterally nephrectomized rats. The distribution of
EPO-bi to bone marrow and spleen was similarly inhibited
by simultaneous injection of excess amounts of either the
nonlabeledEPO-bi or control EPO.These resultsindicate
that the low in vivo biologic activityof EPO-bi results from
rapid clearance from the systemic circulation by renal handling. Thus, the well-branchedstructureof the Nlinked
sugar chain of EPO is suggested to play an important role
in maintaining its higher plasma level, which guarantees an
effective transfer to target organs and stimulation of erythroid progenitor cells.
0 1995 by The American Society of Hematology.
E
nary structure of N-linked sugar chains included in this hormone, we examined the pharmacokinetics of EPO-bi using
the usual highly active EPO sample as a control in this study.
RYTHROPOIETIN (EPO) is a glycoprotein hormone
that stimulates proliferation and differentiation of erythroid progenitor cells.' EPO has three N-linked and one 0linked sugar chains, and the sugar content is approximately
40% of its molecular weight of 30.4 k D . ' , 3 The recombinant
DNA technique has enabled us to produce a sufficient
amount of EPO? To date, its sugar chain structures have
been well in~estigated.~.'
From the analysis of EPO modified
by enzymatic and chemical treatment9"' or produced by the
recombinant DNA t e ~ h n i q u e , ~it~has
- ' ~ been shown that the
N-linked sugar chains of EPO play important roles in biosynthesis, secretion, and expression of biologic activity. In many
cases, modification of the sugar portion of EPO resulted in
a decrease or a complete loss of in vivo biologic activity,
whereas it maintained or increased in vitro activity. A typical
example is that the removal of terminal sialic acid residues
from the sugar chains of EPO causes complete loss of in
vivo biologic activity but increases in vitro biologic activity."," The loss of in vivo biologic activity of asialo-EPO
could be explained by a rapid removal from the systemic
circulation,'9~'owhich resulted from hepatic uptake mediated
by a galactose-binding protein'' and degradation in the lysosome.
Recombinant human EPO produced by Chinese hamster
ovary cells is usually highly active, and contains N-linked
sugar chains mostly composed of tetraantennary structure^.^.^
However, in our previous studyi4 we found a unique EPO
sample (EPO-bi) that has less in vivo biologic activity and
contains higher amounts of biantennary N-linked sugar
chains than the usual EPO sample. Analysis of several EPO
samples with different in vivo activities showed that there
is a positive correlation between the ratio of tetraantennary
to biantennary sugar chains included in EPO and its biologic
activity in vivo. Interestingly, the degree of galactose exposure did not correlate with the in vivo activity among the
EPO samples used. Considering that EPO-bi shows higher
in vitro activity than the usual EPO sample, it is suggested
that the lower in vivo activity of EPO-bi results from its less
efficient delivery to target cells in the bone marrow. To
find a clue to understanding the mechanism by which EPO
expresses in vivo biologic activity depending on the antenBlood, Vol 86,No 1 1 (December l), 1995: pp 4097-4104
MATERIALS AND METHODS
Preparation and iodination of EPO samples. Two EPO samples
usedin this study, EPO-bi and control EPO, were produced by
recombinant Chinese hamster ovary cells and purified as previously
de~cribed.~.'~
The in vivo biologic activities of EPO-bi and control
EPO determined by the exhypoxic, polycythemic mice bioassay"
were 53 and 240 IU/pg protein, respectively. In vitro activity of
EPO-bi assayed using cultured rat bone marrow cells" was 1.8-fold
higher than that of control EPO. These EPO samples were iodinated
by the chloramine-T methodz3 using carrier-free Iz5I (Amersham,
Buckinghamshire, UK) and purified by gel filtration using a PD-10
column (Pharmacia, Uppsala, Sweden). Human serum albumin was
added to the labeled samples at a concentration of 0.25%. Their high
purity was confirmed by gel-filtration chromatography (>96%) and
by the immunoprecipitation method (>99%). These iodinated samples retained biologic activities in vivo and in vitro.
Animals and administration. Male Sprague-Dawley rats aged 5
to 6 weeks were purchased from SLC (Shizuoka, Japan). The animals
were used for study at 7 to 8 weeks of age and weighing 220 to
329 g. For investigation of renal handling of EPO samples, rats
were anesthetized with sodium pentobarbital 50 mgkg administered
intraperitoneally to permit implantation of catheters and to perform
nephrectomy. Unilateral and bilateral nephrectomies were performed
From the Pharmaceutical Development Laboratory, Kirin Brewery, Maebashi, Gunma; the Central Laboratories for Key Technology, Kirin Brewery, Yokohama-shi, Kanagawa; the Department of
Biochemistry, Institute of Medical Science, University of Tokyo,
Tokyo, Japan; and Amgen lnc, Thousand Oaks, CA.
Submitted December 22, 1994; accepted July 26, 1995.
Supported by a grantfrom the Japan Health Science Foundation.
Address reprint requests to Seiichi Takasaki, PhD, Department
of Biochemistry, Institute of Medical Science, University of Tokyo,
4-6-1 Shirokane-dai, Minato-ku, Tokyo 108, Japan.
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
"advertisement" in accordance with 18 U.S.C. section 1734 solely to
indicate rhis fact.
0 1995 by The American Society of Hematology.
0006-4971/95/861 I -0025$3.00/0
4097
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
4098
by renal arteriovenous ligation, and a polyethylene catheter (PE-50)
was inserted into the femoral artery. The catheter was filled with
heparin (50 U/mL) in saline throughout the experiment. The shamoperated group was treated in the same way without nephrectomy.
The labeled EPO samples were diluted with 20 mmol/L sohum
citrate buffer, pH 7.0, containing 100 mmol/L sodium chloride and
0.25% human serum albumin to prepare a dosing solution at a final
concentration of 0.5 pg (306 to 581 kBq)/mL. Rats were given an
intravenous bolus injection of dosing solution (1 mLkg) via tail
vein.
Plasma clearance. Each group of five rats received an intravenous injection of labeled EPO samples at a dose of 0.5 pgkg. At
selected time intervals, blood was withdrawn from the tail veins and
collected into heparinized test tubes. After centrifugation at 2,000
X g for 5 minutes, aliquots (20 to 50 p L ) of plasma were taken
for determination of total, trichloroacetic acid (TCA)-insoluble, and
immunoreactive radioactivities. For investigation of renal handling
of EPO samples, blood was withdrawn from the femoral artery via
the catheter, and immunoreactive radioactivity in the plasma was
measured. TCA-insoluble radioactivity in the plasma was measured
after addition of 1 mL ice-cold 10% TCA, centrifugation at 1,000
X g for 10 minutes, and washing of the precipitate with 10% TCA.
Immunoreactive radioactivity was also measured according to the
immunoprecipitation method. Briefly, 0.1 mL rabbit antiserum
against recombinant human EPO diluted 1:2,000 in phosphate-buffered saline containing 1%bovine serum albumin and 0.05% sodium
azide was added to the aliquots. After incubation overnight at 4"C,
0.5 mL of a solution of Amerlex-M (donkey antirabbit IgG; Amersham) was added to the mixtures and incubated for 1 hour at room
temperature. The precipitated fraction was obtained by centrifugation
at 1,OOO X g for 10 minutes, and its radioactivity was measured.
The contents of labeled materials were estimated by their specific
ra&oactivity, with correction for counting efficiency and attenuation,
and expressed as nanogram equivalents of protein.
Tissue distribution. Each group of three rats were given an intravenous injection of labeled EPO samples (0.5 pgkg). In the competition experiments, each group was also given an intravenous injection
of labeled EPO-bi (0.05 pgkg) with or without 5-, 25-, and 125fold larger amounts of nonlabeled EPO-bi or control EPO. At selected times, organs and tissues to be. assayed were removed from
rats killed by exsanguination from the abdominal aorta under ether
anesthesia, excised, and washed with saline. Blood samples were
collected into heparinized tubes and centrifuged at 1,000 X g for 15
minutes to obtain the plasma. The femur freed from muscle was cut
off to collect the bone marrow. The liver, spleen, and kidneys were
weighed and homogenized in 4 v01 saline using a Polytron homogenizer (Kinematica, Littau-Luzern, Switzerland). Aliquots of blood,
plasma, tissue homogenates, and other tissue samples were weighed,
and their total radioactivities were measured. TCA-insoluble radioactivities in plasma, liver, spleen, and kidneys were also measured.
For whole-body autoradiography, animals were killed by ether
inhalation at 20 minutes after an intravenous injection of labeled
EPO samples and then frozen in dry ice-hexane and embedded in
4% carboxymethyl cellulose. Frozen whole-body sections 35 pm
thick were prepared using a cryomicrotome (450MP PMV, Stockholm, Sweden). After freeze-drying in cryostats, the sections were
placed in contact with x-ray film (MARG 'H type; Konica, Tokyo,
Japan) for 3 weeks, and the film was developed to obtain the autoradiogram.
Excretion into urine. The animals were placed into glass metabolic cages (Metabolica, Sugiyamagen-Iriki, Tokyo, Japan) after intravenous injection of labeled EPO samples. Urine was collected
over 0 to 6, 6 to 12, and 12 to 24 hours, and thereafter at intervals
of 24 hours for 72 hours. The samples were collected in bottles
containing 1 mL 1 % sodium azide and placed in a cooling bath kept
MlSAlZU ET AL
at 4°C. Aliquots (200 pL) of urine were taken to measure total and
immunoreactive radioactivities.
Pharmacokinetic parameters
and
statistical
analysis.
The
plasma clearance data of immunoreactive radioactivity obtained for
each animal were fitted to the following equation for a two-compartment model by nonlinear regression analysis with the NONLIN84
program" using a VAX8350 computer (Digital Equipment Corp,
Massachusetts): C, = Ae-"' + Be-O', where C, represents the concentration in serum at time t, and A, a,B, and p are paired constants
of compartments 1 and 2, respectively. The constants A and B are
the y (time 0)-intercepts, and a and p are the slopes obtained from
the first (a)and second (p) phases of the plot of log plasma EPO
concentration versus time, respectively. According to standard techn i q u e ~the
, ~ ~computer-estimated constants a,p, A, and B were used
for calculation of the following kinetic parameters: the half-lives in
the firstand second phases (t,,a and tl,& which are equal to
-0.6931~and -0.693/p, respectively), the volume of distribution of
the central compartment (V&), and the steady-state volume of distribution (Vd,,), which is the total volume of central and peripheral
compartments. The area under the concentration versus time curve
(AUC) was calculated by the trapezoidal method and extrapolated
to infinity. Total body clearance (CItoJ was calculated as the dose
divided by AUC. Volume and clearance terms were normalized to
the body weightof each rat and expressed as milliliters per kilogram
and milliliters per hour per kilogram, respectively. Statistical significance was analyzed by the F test and Student's t-test for comparison of EPO-bi versus control EPO, and by one-way analysis of
variance followed by Tukey's multiple range test for group comparisons.
RESULTS
Plasmaclearance. To understand the expression of in
vivo biologic activity of EPO depending on the antennary
structure of N-linked sugar chains, the metabolic behavior
of less active EPO-bi, which has been shown to be enriched
with biantennary sugar chains,I4was examined as compared
with highly active control EPO enriched with tetraantennary
sugar chains. Carbohydrate structures of these two EPO samples are shown in Fig 1. These samples were radioiodinated
and given to rats by single intravenous injections, and their
concentrations in plasma were determined at various time
intervals (Fig 2). Both control L251-EP0and '"I-EPO-bi disappeared in a biphasic manner. Any measurement of total,
TCA-insoluble, and immunoreactive radioactivities indicated that lZ51-EPO-bidisappeared from the circulation more
rapidly than control '"I-EPO. Plasma concentrations of immunoreactive radioactivity of control EPO accounted for
95%, 89%, 87%, 76%, and 63% of the levels of total radioactivity after 20 minutes and 2, 4, 8, and 12 hours of injection,
respectively. The corresponding values for EPO-bi were
95%, 70%, 54%, 35%, and 24%. These results suggest that
control Iz5I-EPOis more resistant to degradation than EPObi. Pharmacokinetic parameters estimated from the plasma
concentration data of immunoreactive radioactivity are summarized in Table 1. The statistically significant differences
are that EPO-bi shows a slightly shorter half-life in the elimination phase (tl,$) and approximately a two-fold larger Vd,,
as compared with control EPO, and that C&,, of EPO-b1 was
3.2 times greater than that of control EPO. When nonlabeled
EPO samples were given and their plasma disappearance
was measured by radioimmunoassay,26similar differences
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
4099
RENALHANDLING OF RECOMBINANTERYTHROPOIETIN
Table 1. Pharmacokinetic Parameters for Plasma Clearance of
Immunoreactive RadioiodinatedEPO Samples Aftar a Single
Intravenous Administration
Sugar chains
(mollmol protein)
control €PO EPO-bi
biantennary
*+-O-.-O\
0
.
0.2
l
*+-o-.-o'
tl/za (h)
1.l
(h twlj )
Vd, (mUkg)
Vd. (mVkg)
AUC- (ng Eq. h/mL)
C L , (mUh/kg)
triantennary
k+-O-.-o
*+-0-.\
*+-O-.'
*A
0.9
0.4
0.498
3.73
41.6
62.1
39.9
12.5
2 0.145
0.393 t 0.035
2.57 -C 0.25*
48.2 2 2.4
101.5 t 7.5*
12.5 2 0.6'
40.1 i- 1.9*
2 0.05
2 3.5
-C 4.3
2 1.7
2 0.6
Male rats were given 0.5 pg/kg of control '251-EP0or lZ51-EPO-bi.
Each value represents the mean 5 SD of 5 rats.
Significantly different from control '251-EP0( P < .01).
\
0-.-h
'Z51-EPO-bi
Control '"I-EPO
Parameter
*A
0
'
bi was similar to that of control IZ5I-EPO,except that EPObi accumulated in the kidneys, especially in the renal cortex,
more efficiently than control EPO.
tetraantennary
Tissue distribution was also examined by measuring radio*+-(O-Dln\
activity in various tissues, and the results are summarized
in Table 2. At 20 minutes of injection of Iz5I-EPO,the plasma
level of radioactivity was 8.90 ng equivalents of EPO/mL,
1.o
2.4
and the same level of radioactivity was detected in bone
marrow (Table 2 ) . Levels of radioactivity in the spleen and
the kidneys were approximately 20% to 30% of those in the
plasma, and levels in the liver were less than 7%. Other
organs such as the submaxillary gland, stomach, pancreas,
Fig 1. NLinked sugarchainsincluded in control EPO andEPObi.'.l4 ( e ) Sialicacid; ( 0 )galactose; 1.
Nacstylglucosamina; (0) skeletal muscle, prostate, testis, epididymis, thymus, lymph
mannose; (A) fucose.
nodes, and small intestine contained lesser amounts of radioactivity, which were not greater than 4% of the radioactivity
in plasma (data not shown). Radioactivity in most tissues
decreased in parallel with the decrease in the plasma level
were also observed between EPO-bi and control EPO (data
over 4 hours after injection. On the other hand, distribution
not shown).
of Iz5I-EPO-bidiffered from that of control 12SI-EP0in two
Tissue distribution. Tissue distributions of radioiodinated EPO-bi and control EPO at 20 minutes after intravenous
respects. First, levels of"'I-EPO-bi
in bone marrow were
lower than those of control IZ5I-EPO, asin the case of the
injections were examined by whole-body autoradiography
(Fig 3). In rats injected with control "'I-EPO, the high level
plasma level. This difference became more evident at 4 hours
of radioactivity was detected in bone marrow, blood, kidafter injection than earlier. Second, radioactivity in the kidneys, spleen, and liver. Levels of radioactivity in other orneys after 20 minutes of injection of '251-EPO-bi wasapproxgans were low. The overall distribution profile of '251-EPOimately two times higher than that of control "'I-EPO. This
-
Fig 2. Plasma
disappearance of iodinated EPO
samples. After asingleintravenousadministration
of 0.5 pglkg of control '"I-EPO (A) or "l-EPO-bi (B)
to male rats, total (0).
TCA-precipitable (A), and immunoreactive ( 0 ) radioactivities in plasma were
measured. Each point represents the mean ? SD of
5 rats.
administrationafter
f
a
iil
.l
0
Time
2
4
6
8
1012
(hr)
0
2
4
6
8
1012
Time after administration
(hr)
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
MlSAlZU ET AL
4100
Splem
Lung
Kidney
Bonc m a m w
Blood
B a u mmw
Liwr
Bnin
spinal code
.I
f
Liver
LiVCl
agrees with the result obtained by whole-body autoradiography.
To determine the mode of tissue distribution, the effect
of simultaneous injection of excess amounts of nonlabeled
EPO samples on the distribution profile of labeled EPO-bi
was then examined. At 20 minutes after injection of a trace
amount of '"I-EPO-bi (0.05 & k g ) , the highest concentration of radioactivity was observed in bone marrow, and was
reduced by simultaneous injection of fivefold to 125-fold
larger amounts of nonlabeled EPO-bi (Fig 4A) and control
Fig 3. Whole-body autoradiogram at 20 minutes after a single
intravenous injection (0.5 pg/kgl
of EPO samples in male rats. (AI
and (B) Control '=I-EPO; (C) and
(Dl '251-EPO-bi.
EPO (Fig 4B) depending on their dose levels. There was
also a similar reduction in the spleen. Replacement ability
was almost the same between EPO-bi and control EPO. On
the other hand, there was little effect of simultaneous injection of nonlabeled EPO samples on levels of labeled EPObi in kidneys, liver, and plasma. Thus, distribution to bone
marrow and spleen seems to occur in a receptor-mediated
manner, but distribution to kidneys or liver does not.
Excretionintourine.
In both cases of'*'I-EPO-biand
control '2sI-EP0, most of the radioactivity injected was fi-
ING
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
OF RECOMBINANT ERYTHROPOIETIN
RENAL
4101
Table 2. Tissue Distribution of Radioactivity After a Single Intravenous Administrationof Radioiodinated ControlEPO
or EPO-bi in Male Rats
Concentration of Radioactivity (ng Eq of control '251-EP0 or "51-EPO-bi per g or per mL)
20 min
Tissue
Plasma
Bone marrow
Kidney
Spleen
Liver
Lung
Adrenal gland
Heart
EPO-bi
Control EPO
8.90
9.22
2.86
2.29
0.59
1.06
1.16
0.70
6.21
8.03
6.11
2.23
1.15
1.02
1.06
0.57
2 0.18
? 0.48
2 0.29
2 0.33
-C 0.01
2 0.11
2 0.08
2 0.13
4h
l h
2 0.79*
2 1.33
2 0.48t
2 0.50
2 0.15
2 0.30
t 0.04
2 0.04
Control EPO
7.56 ? 0.435
6.02 2 0.87
2.55 i- 0.18
1.32 t 0.15
0.47 2 0.02
1.11 i- 0.17
0.95 2 0.03
0.84 2 0.06
Control EPO
EPO-bi
3.84
5.33
4.12
1.15
0.49
0.83
0.66
0.52
3.23
0.14
2.82
1.13
0.76
0.27
0.62
0.36
0.40
5 0.10$
t 0.53
2 0.49*
? 0.17
? 0.02
2 0.10
5 0.02$
2 0.05t
2 0.33
2 0.12
2 0.11
2 0.02
5 0.03
2 0.05
2 0.02
EPO-bi
1.23
1.66
0.93
0.42
0.21
0.33
0.19
0.17
2 0.13t
2 0.04$
2 0.10
i- 0.07*
2 0.02'
+- 0.03t
2 0.02'
2 0.01$
Rats were given 0.5 pg/kg of control '251-EP0or '251-EPO-bi, and radioactivities in plasma and tissues were measured. Each value represents
the mean 2 SD of 3 rats.
* P < .05, t P < .01, * P < ,001: Significantly different from control EPO.
and 64% of levels in sham-operated rats by unilateral and
bilateral nephrectomy, respectively. Parameters in sham-operated rats were similar to those in unoperated rats (Table
1). On the other hand, CI,,, (13.9 ? 0.6 mLhkg) and Vd,,
(64.2 ? 9.5 mwkg) of control EPO in bilaterally nephrectomized rats were similar to those in unoperated rats (Table
1). Thus, the significant difference in CI,, and Vd,, between
"'I-EPO and Iz5I-EPO-biobserved before the operation (Table l) was not observed after bilateral nephrectomy.
nally recovered from the urine (Fig 5). However, excretion
of radioactivity from rats injected with '251-EPO-biwas faster
than from rats injected with control "'I-EPO. Radioactive
materials in the urine mostly occurred in nonimmunoreactive
forms in both cases; the immunoreactive form obtained from
control L251-EP0was 2.9% of the dose, and that from "'IEPO-bi was not greater than 0.3%. Low recovery of the
intact form was also observed by gel-permeation chromatography (data not shown).
Renal handling of EPO-bi. From the results described
earlier, it is likely that EPO-bi is cleared from the systemic
circulation by renal handling. To confirm this assumption,
the effect of nephrectomy on plasma clearance was examined
(Fig 6). As compared with sham-operated rats, unilaterally
and bilaterally nephrectomized rats showed a slightly and a
dramatically slower disappearance of Iz5I-EPO-bifrom the
circulation, respectively. Plasma disappearance curves of
EPO-bi and control EPO were similar in bilaterally nephrectomized rats, unlike those in unoperated rats. Statistical analysis (Table 3) indicated that CI,, of 1251-EPO-biwas decreased by 20% (34.5 ? 2.7 mL/h/kg) and 60% (17.4 ? 1.3
mL/h/kg) after unilateral and bilateral nephrectomy, respectively, as compared with that in sham-operated rats (42.5 ?
3.7 mL/h/kg). Vd,, of '251-EPO-biwas also reduced to 76%
DISCUSSION
In this study, the metabolic basis of our previous observationI4 that an increased branching of N-linked sugar chains
in EPO positively correlates with expression of its in vivo
activity was examined. Comparison of phamacokinetic profiles between EPO-bi and control EPO, which differ in that
EPO-bi is enriched with biantennary sugar chains and is five
times less active than control EPO enriched with tetraantennary sugar chains, provided some interesting evidence.
Plasma disappearance curves of radioiodinated EPO samples after single intravenous administrations showed that
EPO-bi is more rapidlycleared from the systemic circulation
than control EPO. In terms of CI,,,, EPO-bi was cleared 3.2
times faster than control EPO. Vd,, of EPO-bi was approxi-
1.o
.-- "
2
B
0.8
m-
gr
ek
0.6
c w
0'2
Fig 4. Alteration in tissuedistribution of iodinated EPO-bi by simultaneous injection
of nonlabeled
EPO samples.Rats were given 0.05 pglkg of "%EPObi with or without indicated doses of nonlabeled
EPO-bi (AI or control EPO (B). After 20 minutes of
injection, radioactivities in plasma
(01, bone marrow
(01,kidney (A), spleen (U), and liver (W) were measured.Each point represents the mean
SD of 3
rats.
*
a&
.E M 0.4
- c
e-
Y
By
0.2
S
0.00
1.25
6.25
Non-labeled EPO-bi(pg/kg)
0.0
0.00
1.25
6.25
Non-labeled EPO(pg/kg)
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
4102
MlSAlZU ET AL
Total
100
T
80
Immunoreactive
60
40
d
20
v
l
0
24
48
72
0
0
24
48
Time afteradministration
72
(hr)
Fig 5. Cumulative excretion of total and immunoreactive radioactivities after a single intravenous injection (0.5 pglkg) of control
orlZ51-EPO-bi ( 0 )in male rats. Each point represents the
EPO (0)
mean ? SD of 3 rats.
mately two times larger than that of the control, suggesting
that some tissues may contribute to the faster removal of
EPO-bi from the systemic circulation. Subsequent analysis
of tissue distribution by whole-body autoradiography at 20
minutes after injection indicated that EPO-bi is more rapidly
distributed to the kidneys. Quantitative determination of radioactivity in each tissue also confirmed the rapid distribution of EPO-bi to the kidneys. Thus, renal handling is suggested to contribute to the rapid clearance of EPO-bi from
the circulation. This assumption is clearly supported by the
experimental nephrectomy, in which plasma clearance of
EPO-bi showed a significant decrease while that of control
EPO was hardly affected. There was also the reduction of
Vd,, of EPO-bi to 76% and 64% after unilateral and bilateral
nephrectomy. In contrast, CI,,,and Vd,, of control EPO did
not significantly change after nephrectomy. Consequently.
the significant difference in CI,,, and Vd,, observed between
EPO-bi and control EPO before the operation vanished after
bilateral nephrectomy.
The nonrenal metabolism of control EPO accords with the
observations that nephrectomy does not affect clearance of
urinary human EPO in rats" and sheep,28but somehow differs from others in that clearance of urinary human EPO in
ratsz9or of recombinant human EPO in dogs'" depends on
the kidney to a small extent. The nonrenal metabolism of
EPO has also been shown in humans." Since different s a m ples and animals were usedby these groups, this minor
discrepancy cannot be explained. However, these reports
coincide in that EPO is largely metabolized through nonrenal
mechanisms such as receptor-mediated endocytosis followed
by lysosomal digestion by erythroid cells."z" Therefore, it
is notable that EPO-bi, and not control EPO, used in our
study is cleared largely depending on the kidneys.
How do the kidneys contribute to the rapid clearance of
EPO-bi from the circulation? As shown by competition experiments (Fig 4), distribution of labeled EPO-bi to the kidneys was not significantly altered by simultaneous administration of excess amounts of nonlabeled EPO samples,
whereas distribution to bone marrow or spleen was greatly
affected. Thus, it is not likely that some specific receptor(s)
mediatesthe transfer of EPO-bi tothe kidneys, but it is
considered thatEPO-bi maybe cleared via a nonspecific
mechanism such as glomerular filtration. Actually, EPO-bi
was excreted into the urine more rapidly than control EPO.
Less than 0.3% of the dose of EPO-bi and 3% of control
EPO were recovered from the urine in an immunoreactive
form, respectively. This slight difference may be due to the
different degradation, possibly mediated by proteinases in
renal tubular cells.
Although the concentrations of EPO-bi distributed to the
tissues, except for the kidneys, were similar to or less than
those of control EPO (Table 2 ) , the tissue/plasma concentration ratios of EPO-bi were higher than those of control EPO.
Therefore, itis possible to consider thatEPO-bi either
crosses capillaries more easily or is retained more avidly in
the tissues. If this possibility is true, EPO-bi would distribute
more efficiently to the tissues than control EPO on the condition that plasma concentrations of both EPO samples are the
100
-
I
.L
.l
0
2
4
6
8
Timeafteradministration(hr)
Fig 6. Effect of nephrectomy on plasma concentration of EPO
samples. lX1-EPO-bi (0.5 pg/kg) was injected in sham-oparated (0).
unilaterally nephrectomized (A),or bilaterally nephrectomized (0)
rats. Control '%EPO (0.5 pg/kg) WM injected in bilaterally nephrectomized rats
At the indicatedtime after injection, immunoreactive radioactivity of the samples was measured. Each point represents the mean k SD of 3 rats.
(m).
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
RENAL HANDLING OF RECOMBINANT ERYTHROPOIETIN
4103
Table 3. Effect of Neohrectomv on Plasma Clearance of RadioiodinatadEPO and EPO-bi
Control 'z61-EP0
'2s1-EPO-bi
Parameter
r,o!
(h)
f n B (h)
Vd, (mUkg)
Vd.. (mUkg)
AUC(ng Eq.h/mL)
Cl,, (mUh/kg)
Sham Operation
Unilateral
Nephrectomy
Bilateral
Nephrectomy
Bilateral
Nephrectomy
0.339 2 0.093
1.87
2.66 t 0.64
41.0 t 1.3
100.0 2 15.8*
11.8 t 1.0*
42.5 2 3.7*
0.256 t 0.029
2 0.09
40.4 t 1.9
76.1 t 0.7t
14.5 t 1.1'
34.5 -t 2.7*
0.279 2 0.093
2.80 2 0.05
37.4 2 1.9
63.8 2 6.1*
28.9 2 2.1**
17.4 t 1.3*
0.304 t 0.260
3.73 t 1.34
39.7 t 5.4
64.2 t 9.5
35.9 2 1.6
13.9 t 0.6
Male rats were given 0.5 pg/kg of lZ51-EPOor 'z51-EPO-bi.Pharmacokinetic parameters for clearance were estimated from the plasma concentration of immunoreactive radioactivity. Each value represents the mean t SD of 3 rats.
Significantly different from control '251-EP0 in bilaterally nephrectomized rats ( P i.01).
t P i.05, * P .01: Significantly different from '251-EPO-biin sham-operated rats.
same. However, the plasma concentration of EPO-bi is
greatly affected by renal metabolism, as discussed earlier,
and therefore results in its lower tissue distribution. It is also
noteworthy that the distribution of radioiodinated EPO-bi to
the liver was notaltered by a simultaneous injection of either
nonlabeled EPO-bi or control EPO, suggesting its nonspecific manner of distribution. Thus, it is evident that the clearance of EPO-bi differs from the rapid clearance of asialoEPOL9-20
and EPO with sugar chains with greater than three
N-acetyllactosamine repeating units,'' which are mediated
by hepatic galactose-binding protein.21This accords with our
previous findingI4that the different in vivo activity of various
EPO samples, including EPO-bi and control EPO, did not
correlate with the extent of galactose exposure of sugar
chains included in the samples.
According to the present knowledge on the difference in
antennary structures of sugar chains between EPO-bi and
control EPO, different metabolic behaviors of the two samples may be due to their physicochemical properties, such
as the molecular weight of the sugar moiety and the negative
charge depending on sialic acid content. EPO has a molecular weight of approximately 30 D,
and this size seems to
be critical for glomerular filtration of low-molecular-weight
proteins.34Thus, it is likely that EPO is filtrated by glomeruli
more rapidly when its molecular size is diminished due to
fewer branched antennary structures of sugar chains and a
smaller content of sialic acid, as in the case of EPO-bi. The
lower sialic acid content in EPO-bi also results in a reduced
negative charge, and may cause a conformational change of
the molecule that accelerates glomerular filtration. Another
example of a recombinant glycoprotein that shows a rapid
accumulation in the kidneys is a human granulocyte-macrophage colony-stimulating factor (GM-CSF).35 Approximately one third of GM-CSF, two potential sites of which
are fully N-glycosylated, accumulated in the kidneys after
15 minutes of intravenous injection. The impaired glycosylation of GM-CSF by the site-directed mutagenesis of potential
N-glycosylation sites resulted in a faster clearance from the
circulation, but did not cause an increased accumulation in
the kidneys. This observation complicates the explanation
for the rapid distribution of EPO-bi to the kidneys. Further
studies are needed to determine the precise mechanism.
However, it is obvious that more rapid clearance of EPO-bi
with fewer branched sugar chains is caused by renal handling. Evidence obtained in this study suggests that the wellbranched structure of the sugar chains of EPO may play an
effective role in maintaining the higher plasma concentration
of this hormone, which guarantees its delivery to target organs and higher expression of in vivo biologic activity.
ACKNOWLEDGMENT
We thank Drs T. Tokiwa, T. Kaneko, and T. Sudo for helpful
discussions. We also thank T. Hirosawa for excellent technical assistance.
REFERENCES
1. Graber SE, Krantz SB: Erythropoietin and the control of red
cell production. Annu Rev Med 29:51, 1978
2. Lai P-H, Everett R, Wang F-F, Arakawa T, Goldwasser E:
Structural characterization of human erythropoietin. J Biol Chem
261:3116, 1986
3. Davis JM, Arakawa T, Strickland TW, Yphantis DA: Characterization of recombinant human erythropoietin produced in Chinese
hamster ovary cells. Biochemistry 26:2633, 1987
4. Lin F-K, Suggs S, Lin C-H, Browne JK, Smalling R, Egrie
JC, Chen KK, Fox GM, Martin F, Stabinsky Z, Badrawi SM, Lai
P-H, Goldwasser E: Cloning and expression of the human erythropoietin gene. Proc Natl Acad Sci USA 82:7580, 1985
5. Takeuchi M, Takasaki S, Miyazaki H, Kat0 T, Hoshi S, Kochibe N, Kobata A: Comparative study of the asparagine-linked sugar
chains of human erythropoietins purified from urine and the culture
medium of recombinant Chinese hamster ovary cells. J Biol Chem
263:3657, 1988
6. Sasaki H, Bothner B, Dell A, Fukuda M: Carbohydrate structure of erythropoietin expressed in Chinese hamster ovary cells by
a human erythropoietin cDNA. J Biol Chem 262:12059, 1987
7. Tsuda E, Goto M, Murakami A, Akai M, Ueda M, Kawanishi
G , Takahashi N, Sasaki R, Chiba H, Ishihara H, Mori M, Tejima S,
Endo H, Arata Y: Comparative structural study of N-linked oligosaccharides of urinary and recombinant erythropoietins. Biochemistry
27:5646, 1988
8. Inoue N, Takeuchi M, Asano K, Shimizu R, Takasaki S , Kobata A: Structures of mucin-type sugar chains on human erythropoietin purified from urine and the culture medium of recombinant Chinese hamster ovary cells. Arch Biochem Biophys 301:375, 1993
9. Dordal MS, Wang FF, Goldwasser E: The role of carbohydrate
in erythropoietin action. Endocrinology 116:2293, 1985
10. Takeuchi M, Takasaki S, Shimada M, Kobata A: Role of
sugar chains in the in vitro biological activity of human erythropoie-
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
4104
tin produced in recombinant Chinese hamster ovary cells. J Biol
Chem 265:12127, 1990
11. Tsuda E, Kawanishi G, Ueda M, Masuda S, Sasaki R: The
role of carbohydrate in recombinant human erythropoietin. Eur J
Biochem 188:405, 1990
12. Higuchi M, Oh-eda M, Kuboniwa H, Tomonoh K, Shimonaka
Y, Ochi N: Role of sugar chains in the expression of the biological
activity of human erythro-poietin. J Biol Chem 267:7703, 1992
13. Dube S, Fisher JW, Powell JS: Glycosylation at specific sites
of erythropoietin is essential for biosynthesis, secretion, and biological function. J Biol Chem 263:17516, 1988
14. Takeuchi M, Inoue N. Strickland TW, Kubota M, Wada M,
Shimizu R, Hoshi S, Kozutsumi H, Takasaki S, Kobata A: Relationship between sugar chain structure and biological activity of recombinanthuman erythropoietin produced in Chinese hamster ovary
cells. Proc Natl Acad Sci USA 86:7819, 1989
15. Wasley LC, Timony G, Murtha P, Stoudemire J, Domer A J ,
Car0 J, Krieger M, Kaufman RJ: The importance of N-and U-linked
oligosaccharides for the biosynthesis and in vitro and invivo biologic
activities of erythropoietin. Blood 772624, 1991
16. Yamaguchi K, Akai K, Kawanishi G, Ueda M, Masuda S,
Sasaki R: Effects of site-directed removal of N-glycosylation sites
in human erythropoietin on its production and biological properties.
J Biol Chem 266:20434, 1991
17. Lukowsky WA, Painter RH: Studies on the role of sialic acid
in the physical and biological properties of erythropoietin. Can J
Biochem 50:909, 1972
18. Goldwasser E, Kung CK-H, Eliason J: On the mechanism of
erythropoietin-induced differentiation. XIII. The role of sialic acid
in erythropoietin action. J Biol Chem 249:4202, 1974
19. Fukuda MH, Sasaki H, Lopez L, Fukuda M: Survival of
recombinant erythropoietin in the circulation: The role of carbohydrates. Blood 7334, 1989
20. Spivak JL, Hogans BB: The in vivo metabolism of recombinant human erythropoietin in the rat. Blood 73:90, 1989
21. Ashwell G, Morel1 AG: The role of surface carbohydrates in
the hepatic recognition and transport of circulating glycoproteins.
Adv Enzymol 41:99, 1974
22. Goldwasser E, Eliason JF, Sikkema D: An assay for erythropoietin in vitro at the milliunit level. Endocrinology 97:315, 1975
MlSAlZU ET AL
23. Greenwood FC, Hunter WM, Glover JS: The preparation of
"'I-labeled human growth hormone of high specific radioactivity.
Biochem J 89: 114, 1963
24. Metzler CM, Weiner DL: NONLIN84 User's Guide. Lexington, KY, Statistical Consultants, Inc. 1984
25. Gibaldi M, Perrier D: Pharmacokinetics (ed 2). New York,
NY, Dekker, 1982
26. Egrie JC, Cotes PM, Lane J, Das REG, Tam RC: Development of radioimmunoassay for human erythropoietin using recombinant erythropoietin as tracer and immunogen. J Immunol Methods
99:235, 1987
27. Steinberg SE, Garcia JF, Matzke GR, Mladenovic J: Erythropoietin kinetics in rat: Generation and clearance. Blood 67:646,
1986
28. Mladenovic J, Eschbach JW, Koup JR, Garcia JF, Adamson
JF: Erythropoietin kinetics in normal and uremic sheep. J Lab Clin
Med 105:659, 1985
29. Emmanouel DS, Goldwasser E, Katz AI: Metabolism of pure
human erythropoietin in the rat. Am J Physiol 247:F168, 1984
30. Fu J-S, Lertora JJL, Brookins J, Rice JC, Fisher JW: Pharrnacokinetics of erythropoietin in intact and anephric dogs. J Lab Clin
Med 1 11:669, 1988
31. Kindler J, Eckardt K-U, Ehmer B, Jandeleit K,Kurtz A,
Schreiber A, Scigalla P, Sieberth H-G: Single-dose pharmacokinetics
of recombinant human erythropoietin in patients with various degrees of renal failure. Nephrol Dial Transplant 4:345, 1989
32. Sawyer ST, Krantz SB, Goldwasser E: Bindmg and receptormediated endocytosis of erythropoietin in Friend virus-infected erythroid cells. J Biol Chern 2625554, 1987
33. Sawada K, Krantz SB, Sawyer ST, Civin Cl: Quantitation of
specific binding of erythropoietin to human erythroid colony-forming cells. J Cell Physiol 137:337, 1988
34. Maack T: Renal handling of low molecular weight proteins.
Am J Med 5857, 1975
35. Donahue RE, Wang EA, Kaufman RG, Foutch L, Leary AC,
Witek-Giannetti JS, Metzger M, Hewick RM, Steinbrink DR, Shaw
G, Kamen R, Clark SC: Effect of N-linked carbohydrate on the in
vivo properties of human GM-CSF. Cold Spring Harbor Symp Quant
Biol 51:685. 1986
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1995 86: 4097-4104
Role of antennary structure of N-linked sugar chains in renal handling
of recombinant human erythropoietin
T Misaizu, S Matsuki, TW Strickland, M Takeuchi, A Kobata and S Takasaki
Updated information and services can be found at:
http://www.bloodjournal.org/content/86/11/4097.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.