From www.bloodjournal.org by guest on September 17, 2016. For personal use only.

Survival of Recombinant Erythropoietin in the Circulation:

The Role of Carbohydrates

By Michiko N. Fukuda, Hiroshi Sasaki, Lily Lopez, and Minoru Fukuda

Recombinant human erythropoietin produced in trans- enriched by tomato lectin affinity chromatography. The
fected Chinese hamster ovary cells is glycosylated much lectin-bound fraction was cleared to a larger extent than
the same way as the erythropoietin present in human was the unfractionated erythropoietin, while the compo-
urine. To determine the role of carbohydrates in the nent that did not bind the lectin was found to be stable in
stability of recombinant human erythropoietin in vivo, the circulation. Authentic N-acetyllactosamine repeats
(polylactosaminoglycans) prepared from erythrocytes
I ‘lJ-Iabeled recombinant erythropoietin was intrave-
nously infused into rats. The erythropoietin was slowly were similarly rapidly cleared from the circulation to the
cleared from the blood with a half-life of approximately two liver, and this clearance was inhibitable with asialo-a1-acid
hours. Asialoerythropoietin, which was produced by treat- glycoprotein. These results suggest that (a) the sialic acid
ment of recombinant human erythropoietin with sialidase. of the recombinant erythropoietin is necessary for this
was found to be cleared rapidly from circulation within ten glycoprotein hormone to circulate stably and (b) glycopro-
minutes. These data suggest that the galactose binding teins with more than three lactosaminyl repeat units may
protein of hepatic cells is involved in the clearance of be cleared by the galactose binding protein of hepato-
asialoerythropoietin. Erythropoietin also contains N-gly- cytes.
cans with a few N-acetyllactosamine repeats, which can be a 1989 by Grune & Stratton. Inc.

E rythropoietin is a glycoprotein hormone with a molecu- Human peptide hormones may be produced abundantly by
bar weight (mob wt) of 3 1 kibodaltons, with 40% of its using recombinant DNA expression vectors in cultured cells.
molecular size accounted for by carbohydrates. These carbo- Glycoprotein hormones must be expressed in animal cells
hydrates have been shown to consist of one 0-linked (carbo- rather than bacteria or yeast to acquire proper carbohydrate
hydrate attached through the hydroxyl group of serine or structures. In animal cells transfected with an expression
theonine residues) and three N-linked oligosaccharides (car- vector containing a cDNA copy of the erythropoietin gene,
bohydrates attached through the amino group of asparagine the polypeptide produced will be subsequently modified by
residues).”2 Erythropoietin is synthesized in the kidney and the glycosyltransferases present in the host cells.7’8 The
circulates in the blood to stimulate red cell proliferation and therapeutic use of recombinant glycoprotein hormones will
differentiation in the bone marrow. This hormonal activity is depend on their mimicking the actions of naturally produced
completely dependent on the presence of sialic acid3 when hormones, including the rate of clearance from the circula-
measured in vivo. But, when the assay is done in vitro, the tory system. We have previously described the carbohydrate
asialoerythropoietin has been found to have enhanced activi- structure of erythropoietin produced in Chinese hamster
ty.5 The apparent loss of activity of the asiaboerythropoietin ovary cells that were transfected with a human erythropoie-
in vivo can be explained by the rapid clearance from the tin cDNA.’ Recently, Takeuchi et al also reported the
circulation by hepatic cells.5 When the terminal siabic acid is carbohydrate structure of recombinant erythropoietin.9 Both
removed from the oligosaccharide, a gabactose residue studies showed that the major carbohydrate units of the
becomes the new terminal sugar. These gabactose-terminated erythropoietin are sialylated tetraantennary saccharides and
glycoproteins may then be recognized by receptors present on that a portion contain N-acetyllactosamine repeats.”9 The
the cell surface of hepatocytes and are internalized by data presented here describe the role of these carbohydrates,
endocytosis, followed by lysosomal digestion.6 Thus it is particularly the sialic acid and N-acetyllactosamine moie-
apparent that the erythropoietin must be fully glycosylated, ties, with respect to the survival of the recombinant erythro-
including terminal siabic acids, to accomplish its hormone poietin in the circulation.
action in vivo.
MATERIALS AND METHODS
Erythropoietin. Chinese hamster ovary cells were transfected
From the La Jolla Cancer Research Foundation. Cancer with an expression vector that contains the human erythropoietin
Research Center. La Jolla, CA. cDNA.’ Erythropoietin was purified from the spent medium of those
Submitted March 7. 1988; accepted August 23, /988. cells as described previously.7 The purification procedure was modi-
Supported by National Institute ofHealth Grant ROl DK 37016 fled from that of Miyake et al,’0 and fractionation on a reverse-phase
(M.N.F) and ROl CA 33000 (M.F.). high-performance liquid chromatograph column was included.7
Dr Sasakis current address is Chugai Pharmaceuticals. Ltd. Erythropoietin was also purified from the urine of aplastic anemia
Research Institute at Fuji-Gotemba, Koma-kado 705-1. Gotemba- patients according to Miyake et al,’#{176}
with a similar modification.
shi, Shizuoka. Japan. These erythropoietin samples were prepared by Chugai Pharmaceu-
Address reprint requests to Michiko N. Fukuda. PhD. La Jolla tical, Ltd (Tokyo).
Cancer Research Foundation. Cancer Research Center, 10901 N Polylactosaminoglycans. Branched polylactosaminoglycan
Torrey Pines Rd. La Jolla. CA 92037. peptides were isolated from human adult erythrocyte membranes as
The publication costs ofthis article were defrayed in part by page described previously.” Linear polylactosaminogbycan peptides were
charge payment. This article must therefore be hereby marked purified from human new born (umbilical cord blood) erythrocyte
“advertisement” in accordance with 18 U.S.C. §1734 solely to membranes.’2 Both preparations were labeled with [3H] by the
indicate this fact. galactose oxidase/NaB[3H]4 method as described previously.’2
© I 989 by Grune & Stratton, Inc. Sodium dodecyl sulfate-polyacrylamide gel electrophore-
0006-4971/89/7301-00I5$3.00/0 sis. Gel electrophoresis was carried out according to Laemmli.’3

84 Blood, Vol 73, No 1 (January), 1989: pp 84-89

 From www.bloodjournal.org by guest on September 17, 2016. For personal use only.

ERVTHROPOIETIN CARBOHYDRATES 85

Autoradiography was performed by exposing dried gels to x-ray film

(XAR-5, Kodak, Rochester, NY) for several hours at - 100#{176}C
with
an intensifying screen.
Clearance oferythropoietins. A male Sprague-Dawley rat (275 100
to 300 g) was anesthetized by exposure to an atmosphere of
metoxyflurane (Pitman-Moore, Washington Crossing, NJ). One to
3 pmol of [1251] erythropoietin (specific activity, -3 to 10 x 106
cpm/pmol) was prepared by the chboramine-T method’4 and
injected through the vein at the penis. After injection, the tip of the
tail was removed, and blood was collected into heparinized tubes at
to
various intervals. The radioactivity of the collected blood was
monitored by a gamma counter. At the end of the experiment (30
>,
minutes or three hours after the injection) the rat was killed by an
>
overdose of metoxyflurane. The organs were isolated, and the 50
C.)
distribution of radioactivity in the various organs was measured.
Clearance ofpolylactosaminoglycans. [3H]-labeled erythrocyte 0
polylactosaminoglycans were injected into the rat in the same 0
manner as described earlier. The radioactivity of blood was counted
in a scintillation counter.
Sialidase treatment oferythropoietin. [‘‘I] erythropoietin was
dissolved in 50 tL of 0.02 mol/L Na-phosphate buffer, pH 7.0,
containing 0.15 mol/L NaC1 and 1% bovine serum albumin. To this
solution sialidase (5 mU) from Clostridium perfringens (type VIII,
Sigma Chemical Co. St Louis) was added, and the enzymic reaction
was allowed to continue at 37#{176}C for 20 hours.
Fractionation of f125jj erythropoietin by tomato lectin and 0 10 20 30
Datura stramonium agglutinin. [‘nb) erythropoietin (1 x 10
Minutes
cpm/l0 pmol) was added to 40 L of a suspension (gel:buffer, 1:1,
vol/vol) of tomato lectin-Sepharose 4B’5 in 0.05 mob/L Tris-HCI Fig 1 . Clearance of intact and desialyzed erythropoietins from
buffer, pH 7.5. After incubation at 4#{176}C for four hours the mixture the circulation of rats. [‘‘1)-Iab.IOd recombinant erythropoietin
was centrifuged, and the supernatant was saved as the unbound was injected IV into rats before (0) and after () sialidase
fraction. After washing the lectin-Sepharose gel with the same treatment. The radioactivity in blood collected at the time inter-
buffer four times, I mL of 0.1 mob/L of chitobiose dissolved in the vals shown was measured by a gamma counter. Typical results are
same buffer was added to the gel. Bound [‘“1] erythropoietin that shown here, and similar results were obtained in four separate
experiments on ach intact and asialoerythropoietin sample.
was eluted with chitobiose was saved as the tomato lectin-bound
fraction. The ratio of the radioactivity in the tomato bectin-unbound
fraction and in the bound fraction was approximately 7 to I . The
To determine the role of terminal sialic acid in the
[1251] erythropoietin was also fractionated into bound and unbound
clearance of recombinant erythropoietin, similar experi-
fractions by using D stramonium agglutinin Sepharose’6 (E-Y
Laboratory, San Mateo, CA) in the same manner as fractionation by ments were performed. The terminal sialic acid residues were
tomato bectin. enzymatically removed from the recombinant erythropoie-
Preparation of asialo-a,-acid glycoprotein. a,-Acid glycopro- tin. In contrast to the intact erythropoietin, almost all
tein from human serum was purchased from Sigma. The a,-acid recombinant asialoerythropoietin rapidly disappeared from
glycoprotein was treated with 0.01 N HCI at 90#{176}C for one hour to plasma within 6 minutes. It was noticed that between 10 and
remove sialic acids and was dialyzed against water extensively. This 30 minutes the radioactivity in plasma increased gradually
acid-treated a,-acid glycoprotein sample is termed here as asiabo- (Fig 1). However, as described later, this apparent reappear-
a,-acid glycoprotein.
ance of radioactivity was probably due to the degradation of
the erythropoietin. Similar clearance curves were obtained
RESULTS
when authentic human urinary erythropoietin and its asiabo
Clearance ofrecombinant erythropoie:infrom the plasma form were examined (data not shown).
circulation of Recombinant
rats. erythropoietin produced Sodium dodecyb sulfate (SDS)-pobyacrylamide gel elec-
in Chinese hamster ovary cells was labeled with [1251] and trophoresis of plasma components collected during clearance
injected into rats intravenously (IV). A portion of the intact experiments indicates that the recombinant erythropoietin
erythropoietin was cleared rapidly, but this apparent clear- remained intact up to 30 minutes (Fig 2A). On the other
ance is most likely due to specific binding of erythropoietin to hand, the asiaboerythropoietin (Fig 2B) may be seen to
bone marrow cells since this apparent early-phase clearance disappear quickly within six minutes of injection. No
disappears when a larger quantity (more than 4 pmol) of increase in specific polypeptide bands is observed that corre-
erythropoietin is injected (data not shown). This data suggest sponds to the increase in total serum radioactivity found
that the bone marrow cells are undersaturated with erythro- between the ten- and 30-minute time points, which suggests
poietin and a part of the exogenous erythropoietin is quickly that the increase in radioactivity was due to a form undetect-
taken up by bone marrow cells. The majority of the recombi- able by gel electrophoresis.
nant erythropoietin remained in circulation relatively stable, To investigate this possibility, serum samples of labeled
up to 30 minutes (Fig 1). asialoerythropoietin collected one and 30 minutes after injec-
 From www.bloodjournal.org by guest on September 17, 2016. For personal use only.

86 FUKUDA ET AL

A distribution of radioactivity was as follows: liver, 85%; kid-

ney, 9%; lung, 4%; and spleen, 2%. The predominance of
kD asialoerythropoietin in the liver strongly suggests the involve-
-92.5 ment of receptor-mediated endocytosis specific to galactose
in clearance. The clearance of serum glycoproteins has been
-662
well documented,6 and it has been shown that asiaboglycopro-
-45.0 teins or galactose-terminated glycoproteins are readily
bound to galactose binding proteins of hepatocytes and are
-31.0
quickly cleared from circulation.
Intact recombinant erythropoietin is relatively stable in
circulation after an initial drop in concentration (Fig 1). To
-21.5
determine the time required for clearance of the recombinant
-14.4 hormone, serum from rats injected with [‘251]-babeled glyco-
protein was collected for longer times, and the radioactivity
present in circulation was analyzed by a gamma counter.
Figure 3 shows the clearance of the intact recombinant
erythropoietin in rats up to 72 hours. A two-phase clearance
2 4 6 8 10 15 30 curve was obtained, as with the previous studies)7_’9 An
B initial rapid a-phase with a half-life (t112) of ten minutes was
kD followed by a longer fl-phase with a t,12 of 108 minutes.
The distribution of radioactivity in organs was also deter-
-9a5
mined. At 30 minutes after injection the relative amounts
-662 were as follows: liver, 64%; kidney, 24%; lung, 9%; and
spleen, 3%. The distribution of the radioactivity at three
-45.0
hours was as follows: liver, 53%; kidney, 39%; lung 6%; and
-31.0 spleen, 2%. After three hours, the erythropoietin located in
the kidney was found to be significantly increased, and this
label could well be excreted (urinary) material. It is most
-21.5 likely that the intact erythropoietin taken up by the liver

-14.4
contains less sialic acid or more N-acetyllactosamine repeats
(see the next section) than does the erythropoietin that
remains in the circulation.
The effect of N-acetyllactosamine repeats on the clear-
ance of erythropoietin. The majority of recombinant
erythropoietin in carbohydrates are tetraantennary sacchar-
1 2 3 4 6 8 10 15 30
ides.’9 N-acetyllactosamine repeats are attached to the
Fig 2.SDS-polyacrylamide gel electrophoresis of intact and tetraantennary core saccharide. Oligosaccharides with one
desialyzed recombinant erythropoietin recovered from rat plasma. lactosamine repeat account for 32.1%, those with two lacto-
Plasma was applied to an SDS-polyacrylamide gel (14% acrylam- samine repeats account for 16.5%, and those with three
ide) and the proteins separated by electrophoresis. [‘lJ-labeled
proteins were detected by autoradiography. (A) Intact erythro-
poietin. A labeled component at a mol wt of 66 kilodaltons is a
dimer of erythropoietin. (B) Desialyzed erythropoietin. The num-
bers at the bottom of each lane show the time (minutes) when 80
plasma was taken from the rat.
60

100
tion were analyzed by Sephadex G-50 gel chromatography.
The majority of radioactivity from the sample collected one
at
minute after injection was eluted in the void volume (data
not shown), as would be expected for an undegraded glyco-
protein. The sample collected 30 minutes after injection was
eluted at the same column volume from the column, thus I
.
4 5 6 Pia
indicating that these radioactive species were of much lower 0

mob wt. It is probable that the radioactivity increase after ten cr

minutes that is seen in the asiabo form oferythropoietin is due
to the release of degraded products by cells and that these -I-
j I I I I I tv -i-
products are too small to be detected by SDS-polyacryl- 1 2 3 4 5 6’ 24’48’72
amide gel analysis. Hotrs
To determine the sites of localization of asialoerythropoie-
Fig 3. Clearance of intact recombinant erythropoieti from
tin uptake, the radioactivity present in the four major organs plasma circulation over a period of 72 hours. Results obtained for
of the rat were analyzed 30 minutes after injection. The six rats are presented.
 From www.bloodjournal.org by guest on September 17, 2016. For personal use only.

ERYTHROPOIETIN CARBOHYDRATES 87

lactosamine
investigate
repeats
the stability
are 4.7%
of erythropoietin
of the
that has N-acetyl-
total saccharides.’ To
A
lactosamine repeats, we fractionated [125I] erythropoietin by
using tomato lectin-Sepharose, which binds carbohydrates i oo
with more than three N-acetyllactosamine repeats.’5 Experi-
ments similar to those just described were performed by
using the tomato lectin-bound and -unbound fractions. Data
presented (Fig 4A) demonstrate that the tomato lectin-
bound (containing lactosamine repeats) erythropoietin was
cleared more rapidly and to a greater extent than was the
unbound fraction (Fig 4A). The distribution of the radioac-
tivity in the organs 30 minutes after injection of the bound
fraction was as follows: liver, 67%; kidney, 18%; lung 10%; ‘

and spleen, 5%. The liver may be seen again to be the major 50
organ for removing erythropoietin from the serum.
Erythropoietin was also fractionated by Datura lectin,
which has affinity to the side chain of R -‘ GbcNAc9l - c
6Man of tetraantennary carbohydrate structures.’6 There
was no difference detected in the clearance pattern between
Datura-bound and -unbound components of recombinant
erythropoietin (Fig 4B).
These results suggest that the length of the bactosamine
repeats determines the rate of clearance of sialic acid-
containing erythropoietin. To determine whether this is a 0 10 20 30
general recognition signal, polylactosaminogbycans isolated
Minutes
from erythrocytes, which have been identified as containing
a biantennary core structure and pobylactosaminyl side B
chains,”2 were therefore tested. As shown in Fig 5 these
glycosylated peptides were rapidly cleared from the plasma
circulation, and more than 90% of the radioactivity was
found to be taken up in the liver. This clearance pattern was
identical regardless of whether the erythrocyte polylac-
tosamines were intact or desialylated or whether they were
linear or branched (data not shown).
To test whether there was a specific interaction of polylac-
tosamine with a receptor, competitors were added to the
assay. The addition of lactose had no effect on the clearance,
but asiabo-a,-acid glycoprotein released the polylactosamines
in the early stage of uptake (Fig 5). This effect is probably
due to the replacement of cell surface-bound polylactosam- >
inc with asiabo-a1-acid glycoprotein. Thus the polylactosami- 50
noglycans and the asiabo-glycoproteins appear to be cleared
through the same gabactose binding receptor in the liver.
Small amounts ofone lactosamine repeat have been shown
to be present in a,-acid glycoproteins,2#{176} but intact a,-acid C

glycoprotein does not compete with erythrocyte polybac-

tosamines for clearance (data not shown). This result
suggests that the lactosamine repeat in intact a,-acid glyco-
protein is not enough to release erythrocyte polylactosamino-
glycans from hepatocytes but that the galactose terminal
residue exposed by desiabylation of a,-acid glycoprotein can 1 I I
compete with the polylactosamines. These observations sug- o io 20 30
gest that a glycoprotein with a number of lactosaminyl
repeats may also be recognized by hepatocytes through the Minutes
galactose binding protein, even when the glycoprotein is Fig 4. Clearance of erythropoietin separated by lectins. (A)
sialylated. Clearance of erythropoietin separated by tomato lectin. 0, Tomato
lectin-unbound fraction (erythropoietin without lactosamine
DISCUSSION repeats or with less than three lactosamine repasts); #{149}. tomato
lectin-bound fraction (erythropoietin with more than three lacto-
The present study demonstrates the necessity of sialylated samine repeats). (B) Clearance of erythropoietin separated by
N-glycan modification of erythropoietin for it to achieve its Datura agglutin. 0, Datura lectin-unbound fraction; , Datura
lectin-bound fraction.
 From www.bloodjournal.org by guest on September 17, 2016. For personal use only.

88 FUKUDA ET AL

Binding to hepatocytes

I
c)

>‘

‘p
o-c==- I
U
Co
0
0
CO

R +

0 10 20 30
Minutes
+
Fig 5. Clearance of polylactosaminoglycans. [H]-labeled
branched polylactosaminoglycans isolated from erythrocytes (see
the text) were injected into the rat. At the time shown by the
Fig 6. The structure of the carbohydrate moiety of recombi-
arrow, asialo-a1-acid glycoprotein (#{149})
or lactose (5 mg in 0.2 ml. 0)
nant erythropoietin and its binding to hepatic galactose binding
was injected IV. Linear polylactosaminoglycans gave similar
receptor. Structures of carbohydrates are adopted from the
results (not shown). previous studies.” From the top. slalylated tetraantennary, asialo
tetraantennary, sialylated tetraantennary with three lactosamine
units. and asialo tetraantennary with three lactosamine units. The
hormone activity in vivo. The data presented here showed
carbohydrate that binds to the hepatocyte ( + in this Figure) is
that the carbohydrates are required for the survival of cleared quickly from the circulation and vice versa. Sialic acid.
recombinant erythropoietin in plasma. Our results also show *galactose. C; N-acetylglucosamine. #{149};
mannose. 0; fructose, t.
that N-acetyblactosamine repeats, if they are longer than
three repeating units, are subject to rapid clearance by the (Gal3l -+ 4GlcNAcf3l - 3), we could detect up to three
galactose binding receptor of the liver even if terminal sialic N-acetybbactosamine repeats on a portion of the recombinant
acid residues are present. The carbohydrate structures of glycoprotein by fast atom bombardment analysis.’ The
erythropoietin in relation to their clearance are summarized erythropoietin molecules containing three N-acetyllactosam-
in Fig 6. inc repeats were selected by their preferential binding to
It is well-established that sialic acid residues attached to tomato lectin, and those molecules were rapidly removed
triantennary and tetraantennary complex-type obigosac- from the circulation by receptors in the liver. We have shown
charides are indispensable for the survival of glycoproteins in in this study that polylactosaminoglycans generally are rap-
the circulation. Thus, once the sialic acid is removed, newly idly cleared from the circulation by the galactose binding
exposed gabactose residues are recognized by the galactose receptor in the liver. This indicates that glycoproteins with
binding protein in the hepatocytes, and the asiaboglycopro- longer lactosaminyl repeats are not stable once they are
teins are rapidly removed from circulation and internalized released to plasma.
by hepatocytes.6 Structural analysis of recombinant erythro- The behavior of pobylactosamines in plasma circulation is
poietin has shown that it contains mostly triantennary and consistent with the fact that polylactosamines are abundant
tetraantennary complex-type oligosaccharides, and the pres- in plasma membranes but rare in serum. As exemplified by
ent study confirms that the sialic acid attached to the the band 3 glycoprotein in erythrocyte membranes,”2 the
erythropoietin pobypeptide plays a protective role conducive polylactosaminoglycans are major glycoconjugates in some
for the survival of recombinant erythropoietin in circula- cells in the human body. Little is known, however, about the
tion.5 catabolism of polylactosamines. The present results suggest
The carbohydrate analysis of the recombinant erythro- that polylactosamines could be rapidly internalized by hepa-
poietin showed that the erythropoietin produced by the tocytes if they were released to plasma. The presence of
Chinese hamster ovary cells contains N-acetyllactosamine keratan sulfate, a glycosaminoglycan consisting of sulfated
repeats.”9 Although most of these N-acetyblactosamines are polylactosamines, in human serum2’ suggests that the sulfa-
short, with one additional N-acetyllactosamine unit tion at C-6 on galactose residues protects the polylactosam-
 From www.bloodjournal.org by guest on September 17, 2016. For personal use only.

ERYTHROPOIETIN CARBOHYDRATES 89

ines from recognition by the galactose receptor and subse- of carbohydrates of the recombinant erythropoietin being in
quent clearance by hepatocytes. the correct form to escape detection and removal by galac-
Genetic engineering technology has enabled the produc- tose binding receptors of hepatocytes. The determination of
tion of a large quantity of cytokines such as interferons, the correctly glycosylated forms of potential recombinant
interleukin-2, and tumor necrosis factor. Many of these therapeutic agents and selection of proper expression vectors
recombinant cytokines, however, have been found to be and hosts to elicit those structures may be very important in
unstable in vivo or cause side effects upon administration to future studies.
animals and humans.22 Fortunately, erythropoietin produced
by transfection of animal cells has been greatly successful for
the treatment of patients.23a4 The success of clinical trials of ACKNOWLEDGMENT
erythropoietin is most likely due to the fact that the erythro- The authors thank Dr T. Kawaguchi of Chugai Pharmaceuticals
poietin targets narrowly specific cells such as erythroblasts. for his helpful discussions. We also thank Dr V. Piller-Michel for
However, another factor contributing to its biologic activity providing tomato lectin-Sepharose. We are greatly indebted to Dr
in vivo is the stablity of the erythropoietin molecule. We have M. Williams for the critical reading and grammatical improvement
shown that stability is largely dependent on a high proportion of the manuscript.

REFERENCES

I. Sasaki H, Bothner B, Dell A, Fukuda M: Carbohydrate I 3. Laemmbi UK: Cleavage of structural proteins during
structure of erythropoietin expressed in Chinese hamster ovary cells assembly of the head of bacteriophage T4. Nature 227:680, 1970
by a human erythropoietin cDNA. J Biol Chem 262:12059, 1987 14. Greenwood FC, Hunter WM, Glove JS: The preparation of
2. Lai PH, Everett R, Wang FF, Arakawa T, Goldwasser E: ‘31I-labelled human growth hormone of high specific radioactivity.
Structural characterization of human erythropoietin. J Biol Chem Biochem J 89:1 14, 1963
261:3116, 1986 15. Merkle RK, Cummings RD: Relationship of the terminal
3. Lowry PH, Keighley G, Borsook H: Inactivation of erythro- sequence to the length of poly-N-acetyllactosamine chains in as-
poietin by neurominidase and mild substitution reactions. Nature paragine-linked oligosaccharides from the mouse lymphoma cell line
185:102, 1960 BW5147: Immobilized tomato bectin interacts with high affinity
4. Goldwasser E, Kung C K-H: Part II. Chemistry and purifica- with glycopeptides containing long poly-N-acetyllactosamine chains.
tion of erythropoietin: Progress in the purification of erythropoietin. J Biob Chem 262:8179, 1987
Ann NY Acad Sci 149:49, 1968 16. Cummings RD. Kornfeld 5: The distribution of repeating
5. Gobdwasser E, Kung C K-H, Eliason J: On the mechanism of [Gal$,,4GlcNAcfl,,3] sequences in aspargine-linked oligosaccha-
erythropoietin-induced differentiation; XIII. The role of sialic acid rides of the mouse lymmphoma cell line BW5147 and PHAR2.l. J
in erythropoietin action. J Biol Chem 249:4202, 1974 Biol Chem 259:6253, 1984
6. Morell AG, Irvine KA, Sternlieb 1, Scheinberg IH, Ashwell G: 17. Weintraub AH, Gordon AS, Becker EL, Camiscoli iF,
Physical and chemical studies on ceruboplasmin: V. Metabolic Contrera iF: Plasma and renal clearance of exogenous erythropoie-
studies on sialic acid-free ceruboplasmin in vivo. J Biol Chem tin in the dog. Am i Physiol 207:523, 1964
243:155, 1968 18. Emmanuel DS, Goldwasser E, Katz Al: Metabolism of pure
7. Jacobs K, Shoemaker C, Rudersdorf R, Neill SD, Kaufman human erythropoietin in the rat. Am J Physiol 247:168, 1984 (suppl
RM, Mufson A, Seehra J, Jones 55, Hewick R, Ritch EF, Kawakita 16)
M, Shimizu T, Miyake T: Isolation and characterization of genomic 19. Steinberg SE, Garcia iF, Matzke GR, Mladenovik i:
and cDNA clones of human erythropoietin. Nature 313:806, 1985 Erythropoietin kinetics in rats: Generation and clearance. Blood
8. Powell JS, Berkner KL, Lebo RV, Adamson JW: Human 67:646, 1986
erythropoietin gene: High level expression in stably transfected 20. Yoshima H, Matsumoto A, Mizouchi T, Kawasaki T, Kobata
mammalian cells and chromosome localization. Proc Natb Acad Sci A: Comparative study of the carbohydrate moiety of rat and human
USA 83:6465, 1986 plasma a,-acid glycoprotein. i Biol Chem 256:8476, 1981
21. Thonar E i-M, Meyer RF, Dennis RF, Lenz ME, Maldon-
9. Takeuchi M, Takasaki 5, Miyazaki H, Kato T, Hoshi 5,
adao B, Hassell JR. Hewitt AT, Stark Wi Jr, Stock EL, Kuettner
Kochike N, Kobata A: Comparative study of the asparagine-linked
KE, Klintworth GK: Absence of normal keratansulfate in the blood
sugar chains of human erythropoietins purified from urine and the
of patients with macular corneal dystrophy. Am i Opthalmol
culture medium of recombinant Chinese hamster ovary cells. J Biob
102:561, 1986
Chem 203:3657, 1988
Moertel CG: On lymphokines,
22. cytokines and breakthroughs.
10. Miyake T, Kung C K-H, Goldwasser E: Purification of
JAMA 256:3141, 1986 (editorial)
human erythropoietin. J Biol Chem 252:5558, 1977
23. Eschbach iW, Eggrie iC, Downing MR. Browne JK, Adam-
11. Fukuda M, Dell A, Oates JE, Fukuda MN: Structure of son JW: Correction of the anemia of end-stage renal disease with
branched lactosaminoglycan, the carbohydrate moiety of band 3 recombinant human erythropoietin: Results of a phase I and II
isolated from adult human erythrocytes. J Biol Chem 259:8260, clinical trial. N EngI i Med
1987 316:73,
1984 24. Winearls CG, Oliver DO, Pippard Mi, Reid C, Downing
12. Fukuda M, Dell A, Fukuda MN: Structure of fetal lactosami- MR, Cotes PM: Effect of human erythropoietin derived from
noglycan: The carbohydrate moiety of band 3 isolated from human recombinant DNA on the anemia of patients maintained by chronic
umbilical cord erythrocytes. J Biob Chem 259:4782, 1984 haemodialysis. Lancet 2:1 175, 1986
 From www.bloodjournal.org by guest on September 17, 2016. For personal use only.

1989 73: 84-89

Survival of recombinant erythropoietin in the circulation: the role of

carbohydrates
MN Fukuda, H Sasaki, L Lopez and M Fukuda

Updated information and services can be found at:

http://www.bloodjournal.org/content/73/1/84.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.