From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
The Extracytoplasmic Domain of the Erythropoietin Receptor Forms a
Monomeric Complex With Erythropoietin
By Mei-Gang Yet and Simon S. Jones
W e have generated a truncated form of the erythropoietin
receptor (EPO-R), the extracytoplasmic ligand-binding domain, that is secreted from a transfected Chinese hamster
ovary (CHO) cell line. The truncated receptor is readily purified from CHO conditioned media as a 33-Kd glycosylated
protein, which is converted to a 25-Kd species upon treatment with protein N-glycan glycosidase F. Cross-linking of
radioiodinatedEPO to the secreted receptor yielded a complex of 72 Kd. Also, the growth of the EPO-dependent cell
line, FDCPE, was inhibited in a dose-responsive manner by
the truncated receptor. The complex of the secreted receptor and EPO was isolated by gel filtration and shown to be a
one-to-one complex of the receptor and growth factor by
quantitative amino terminal sequencing. Finally, analysis
of the interaction of the receptor and growth factor by gel
filtration indicated an apparent dissociation constant of
1.1 nmol/L for the truncated receptor.
0 1993 by The American Society of Hematology.
E
case, only a small fraction (1 %) of the isolated fusion protein
was soluble and able to bind EPO with an apparent dissociation constant of 0.3 to 1.5 nmol/L. Also, the extracytoplasmic domain of the murine EPO-R has been expressed in a
mammalian cell line.26The purified truncated murine receptor exhibited a dissociation constant of 17 nmol/L. We
now report the purification and biochemical and biological
characterization of a recombinant ligand-binding domain
of the human EPO-R.
RYTHROPOIETIN (EPO) appears to be the primary,
if not sole, regulator of proliferation and differentiation of immature erythroblasts in mammalia.’ The murine
and human cDNAs encoding the putative EPO receptor
(EPO-R) have been recently
Yet the structure of
the EPO/receptor complex, the mechanism of signal transduction, and whether there are other components in the
receptor complex remain to be defined.
Transfection of the EPO-R cDNA in SV40-transformed
COS-1 monkey kidney cells (COS) or Chinese hamster
ovary (CHO) cells yields both high- and low-affinity EPObinding site^.'.^.^ In addition, cross-linking of radioiodinated EPO to the receptor generates two complexes of 105
Kd and 140 Kd. These results are analogous to those observed with various cell lines and immature erythrobla~ts,~
in which the equivalent complexes are 125 Kd and 140 Kd.
Transfection of different murine interleukin-3 (IG3)-dependent cell lines with the EPO-R cDNA is sufficient to
confer EPO dependen~y.’“~Indeed, EPO-dependent cell
lines can be prepared expressing chimeric receptors consisting of the murine EPO-R extracytoplasmic domain coupled
to the cytoplasmic domain of AIC2A,I5 a subunit of the
murine IL-3 receptor.I6
The molecular size of the major protein expressed by the
EPO-R cDNA is 66 K d although this is consistent with the
105-Kd cross-linked complex (66 Kd for the receptor and
35 to 40 Kd for EPO), the structure of the 140-Kd complex
is more difficult to explain in terms of multimeric structures
of EPO and/or the receptor. There is evidence that the equivalent complexes of 125 Kd and 140 Kd in cell lines and
immature erythroblasts consist of EPO (40 Kd) cross-linked
to 85-Kd and 100-Kd components that are unrelated immunologically to the EPO-R.17 Although we and other^^.'^,'^
have found that the nondenatured cross-linked complexes
can be immunoprecipitated by anti-EPO-R antisera, this
result may be due to indirect precipitation of EPO crosslinked to the 100-Kd and, possibly, the 85-Kd (66 Kd in the
case of COS- 1 or CHO cells) components through their association with the EPO-R.5*’7,Ls,Zo
To investigate the structure of the EPO/receptor complex, we have expressed the extracytoplasmic EPO-binding
have found cDNAs
domain as a secreted protein.
encoding potentially secreted forms of the EPO-R, although
the presence of the protein in the serum remains to be determined. Recently, the extracytoplasmic domain of the human EPO-R has been expressed in Escherichia coli as an
amino terminal fusion to glutathione S-transferase.” In this
Hmd, Vol82,NoG(September 15). 1993: pp 1713-1719
MATERIALS AND METHODS
Cell lines and cell culture. FDC-P I cells were cultured in RPMI
media with 10%heat-inactivated fetal calf serum (HIFCS)and 10%
WEHI-conditioned medium as described.*’ COS-1 cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10%HIFCS. CHO cells were maintained in 01 medium
with 10%dialyzed HIFCS supplemented with adenosine, deoxyadenosine, and thymidine (each 10 pg/mL), as described.”
Expression of the truncated EPO-R. The full-length polymerase
chain reaction (PCR) clone, no. 18, of the human EPO-R3 was
subcloned into the EcoRI site of M 13mp19. The codon of the first
putative amino acid of the transmembrane domain, leucine 25 1,
was changed to a stop codon by M13 single-stranded site-specific
mutagenesis using the oligonucleotide primer, CTAGCGACCTGGACCCCWTCCTGACGCTCTCCCTC. The mammalian expression plasmid, psEPOR, was generated by the ligation of
the Xho I-Bgl I1 fragment containing the mutation with the Cla
I-Sal I fragment of the expression plasmid, PED:’ and pig-R3 restricted with Cla I and Xho I, where the Bgl I1 and Sal I ends of the
first two fragments had been treated with the Klenow fragment of
DNA polymerase I.
COS-I cells were transfected3with psEPOR and pulse-labeled 42
hours posttransfection with 0.2 mCi of [’SS]methioninein 2 mL of
methionine-free medium for 30 minutes and chased for various
periods of time in complete medium. Cell lysates (2 X lo6cpm) or
conditioned media (2 X lo5cpm) were immunoprecipitated as described3’with l to 2 pL of a rabbit polyclonal antisera raised against
From the Genetics Institute Inc, Cambridge, MA.
Submitted October 23, 1992; accepted M a y 3, 1993.
Address reprint requests to Simon S. Jones, PhD, 87 Cambridge
Park Dr, Cambridge, MA 02140.
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 I734 solely to
indicate this fact.
0 I993 by The American Society of Hematolofl.
0006-4971/93/8206-00Ol$3.00/0
1713
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
YET AND JONES
1714
M
I
2
3
4
5
7
6
8
9 IO
II
- 200.
- 97
*
- 69 - 45.
- 30,
- 14.3
a peptide based on the amino terminal sequence of the murine
EPO-R.9 The immunoprecipitates were deglycosylated with protein
N-glycan glycosidase F, endoglycosidase H. or neuraminidase, as
described.” Immunoprecipitates were analyzed on a sodium dodecyl sulfate (SDS) 12% polyacrylamide gel under reducing conditions, as described previously.M CHO cells were transfected with
psEPOR by the method of lipofection (GIBCO BRL, Grand Island,
NY). Clones expressing the secreted receptor were selected in a
medium without added nucleosides. Subsequent expression was
amplified by selection of clones in increasing concentrations of
methotrexate (20 nmol/L. 125 nmol/L, 3.13 pmol/L. and 15.6
pmol/L), as described.”
Cross-linking if (‘’’4EPO. Filtered (0.22 p ) COS-I -conditioned medium 42 hours after transfection with psEPOR was incubated with [‘251] EPO 2 nmol/L in the presence and absence of 100
nmol/L of unlabeled EPO at 4OC for I8 hours and cross-linked as
described previously.’ The cross-linked complexes were immunoprecipitated with the antimurine EPO-R antisera as described
above. The immunoprecipitates were analyzed on an SDS 10%polyacrylamide gel under reducing conditions.
Piir$cation of the rriincated EPO-R. Roller bottles (surface
area, 1,700 cm’; Corning, Corning, NY) were seeded with the truncated EPO-R CHO line, s8.15.6, at I x IO’ cells/bottle. The cells
were cultured for 2 to 3 days in RI media, I01 dialyzed HIFCS,
15.6 pmol/L methotrexate. and 20 mmol/L HEPES, pH 7.5 (400
mL). The cells were then washed twice with phosphate-buffered
saline (PBS) and cultured in OptiMEM (Gibco BRL), 15.6 pmol/L
methotrexate, and 20 mmol/L HEPES, pH 7.5 (200 mL). The medium was collected after 24 hours and fresh medium was added to
the roller bottle: this process was repeated three times. The conditioned medium was clarified by centrifugation, filtered through a
0.45-p filter, and concentrated IO- to 20-fold in an Amicon (Beverly, MA) stirred cell using a YM filter with a 10,000 molecular
weight cut-off. Ammonium sulfate was added to the concentrate to
300/0 saturation, the pellet was discarded, and ammonium sulfate
was added to the supernatant to 55% saturation. The 55% pellet was
Fig 1. SDS-polyacrylamide gel electrophoresis
(SDS-PAGE)of immunoprecipitatesof cell extracts
and medium from pulse-labeled ([3sS]methionine)
COS-1 cells transfected with the truncated EPO-R
cDNA. COS-1 cell extracts either mock-transfected (lane M) or transfected with psEPOR (lanes
1 through 6; 0 , 3 0 , and 60 minutes, and 3, 6, and
22 hours postpulse, respectively) and medium
(lanes 7 through 1 1 ; 30 and 60 minutes, and 3, 6,
and 22 hours postpulse, respectively) were treated
with the antimurine EPO-R antisera and analyzed
by SDS-PAGE, as described in the Materials and
Methods.
extracted twice with one-tenth the original concentrate volume of
35% saturated ammonium sulfate, 25 mmol/L HEPES, pH 7.5
(buffer A). and loaded onto a TosoHaas (Montgomeryville, PA)
phenyl toyopearl column ( I O mg protein/mL of support). The column was developed with a gradient of buffer A to 25 mmol/L
HEPES, pH 7.5. over 25 minutes: fractions containing the receptor
were concentrated fivefold to IO-fold using an Amicon YM- IO centricon and loaded ( I O mg total protein) onto a TosoHaas
TSK2000SW gel filtration column (7.5 mm X 60 cm) developed
with PBS. The purified receptor was fractionated on an SDS I 2 1
polyacrylamide gel. the protein was transferred to nitocellulose, and
the blot was probed with the antimurine EPO-R antisera at a 1:200
dilution. The immune complex was visualized using the ECL system (Amersham, Arlington Heights, IL) and protein A horseradish
peroxidase at a I : 1.000 dilution.
Inhibition qf EPO-depcndenr growth qf the cell line. FDCPE. expressing the human EPO-R. The EPOdependent cell line was
prepared from the murine IL-3-dependent cell line, FDC-PI, as
described previously.’ EPO-dependent cells were washed twice with
RPMl and seeded in a 96-well plate at 1 X 104/well.The cells were
incubated for 20 hours with varying concentrations of EPO in the
presence and absence ofthe truncated EPO-R in triplicate, and then
pulsed for 4 hours with 0.5 pCi/well of [’Hlthymidine. The cells
were harvested and counted.
The striicIiire of the EPO/triincated receptor complex. EPO (8 I
pmol) and the truncated receptor (0 to 300 pmol) were incubated in
PBS or Tris-buffered saline (TBS), pH 7.5, at 25°C or 4°C for up to
60 minutes. The products were analyzed on a TosoHaas TSKG3000SW gel filtration column (7.8 mm X 30 cm) equilibrated in
TBS at 25°C with a flow rate of0.7 mL/min.’’ The amounts of free
and bound receptor were calculated by intergrating the surface area
of the respective peaks using the Waters Expert 860 software
(Waters. Milford, MA). Quantitative N-terminal sequencing was
performed for 8 cycles using an Applied Biosystem (Foster City,
CA) gas-phase sequenator on the complex formed from EPO (400
pmol) and secreted EPO-R (500 pmol). The complex was isolated
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
1715
EPO-R FORMS MONOMERIC COMPLEX WITH EPO
B
A
U N
F
1 2
H
- 200
- 97
- 97
- 69
rw
- 69
- 45
- 45
z
- 30
- 30
+ Fig 2. (A) Deglycosylation of immunoprecipitates of the truncated EPO-R from COS-1 cell-conditioned medium. Immunoprecipitates of 3-hour postlabel conditioned medium containing the
truncated receptor (lane U) were treated with neuraminidse (lane
N), protein N-glycan glycosidase F (lane F), or endoglycosidase H
(lane H) and analyzed by SDS-PAGE, as described in the Materials
and Methods. (B) Cross-linking of radiolabeled EPO to the truncated EPO-R. COS-1 cell-conditioned medium was incubated with
2 nmol/L '251-EP0 in the presence (lane 1 + ) and absence (lane
2 ) of 100 nmol/L of unlabeled EPO. The mixture was incubated
with 0.2 mmol/L disuccinimidyl suberate (DSS) for 1 hour at 4°C
and an excess of 1 mol/LTris-HCI, pH 9, over DSS, was added. The
mixture was treated with the antimurine EPO-R antisera and analyzed by SDS-PAGE, as described in the Materials and Methods.
-
as discussed and desalted using an Amicon centricon-IO by extensive washing with water.
RESULTS
Expression of the truncated form of the human EPOR. The cDNA for the truncated EPO-R, whose coding sequence ends at the codon for amino acid no. 250, proline,
was expressed either transiently in COS-I cells or from
stable CHO cell lines. In pulse-chase studies of these metabolically labeled COS- l cells, two products of 25 Kd and 28
Kd were observed in cell extracts (Fig I, lane 1) that were
chased over a 3-hour period (Fig I , lane 4) into the medium
as a 33-Kd secreted product (Fig I , lane 9). This product was
specifically immunoprecipitated by an antimurine EPO-R
antisera, which has been shown to immunoprecipitate the
full-length EPO-R when expressed in COS-1 cells.' The se-
creted EPO-R was stable in conditioned medium for at least
22 hours (Fig I. lane I I ) . Immunoprecipitates of the secreted receptor were treated with protein N-glycan glycosidase F, endoglycosidase H, or neuraminidase, enzymes that
specifically remove complex, high-mannose-type glycans
and terminal sialic acid residues. The secreted EPO-R has a
theoretical molecular weight of 24 Kd and one potential site
of N-linked glycosylation. Protein N-glycan glycosidase F
converted the 33-Kd secreted product to a 25-Kd species
(Fig 2A. lane F), whereas endoglycosidase H had no effect
(Fig 2A, lane H) and neuraminidase produced a small reduction in size of I to 2 Kd (Fig 2A, lane N). Both protein
N-glycan glycosidase F and endoglycosidase H converted
the 28-Kd intermediate glycosylated product (Fig I , lane I ,
upper band) in the cell extract to a 25-Kd species, presumably the completely deglycosylated protein that is coincident with the lower band in lane 1 of Fig I (data not
shown). Similar results were obtained when the truncated
EPO-R was expressed in CHO cells (data not shown).
When COS- 1 -conditioned medium containing the secreted receptor was incubated with radioiodinated EPO, followed by treatment with disuccinimidyl suberate, a 72-Kd
species was generated (Fig 2B, lane 2-). This product was
immunoprecipitated by the antimurine EPO-R antisera,
and was not formed in the presence of a 50-fold excess of
unlabeled EPO (Fig 2B, lane I +). Cross-linking of radioiodinated EPO, in the absence ofthe secreted receptor, did not
generate any multimeric EPO species in the range of 72 Kd
(data not shown).
Purijication of the triincated receptor $-omCHO-conditioned medium. Transfection of CHO cells with the truncated EPO-R cDNA generated several CHO lines expressing the receptor at various levels by immunoblotting (data
not shown). The expression level was further amplified by
selection in increasing concentrations of methotrexate
yielding a line, s8.15.6, producing 1 to 3 pg/mL per I X IO6
cells/24 h of receptor or I % to 2% of total protein in the
medium. Conditioned medium was collected under serumfree conditions and the truncated EPO-R was enriched by
sequential concentration, ammonium sulfate fractionation.
The receptor was purified by hydrophobic interaction chromatography on a phenyl toyopearl column in which the
major peak (Fig 3A, fractions no. 23 through 29) contained
the receptor by immunoblot. These fractions were combined and size-fractionated on a gel filtration column (Fig
3B, major peak). By this route, 4 L of conditioned medium
containing 300 mg of protein generated 2 mg of truncated
EPO-R with a purity ofgreater than 90% (Fig 3C, fractions
no. 31 through 33). The purified truncated receptor was
recognized by the antimurine EPO-R antisera (Fig 3D).
Inhibition of EPO-dependent growth of FDCPE cells by
the truncated receptor. The EPO-dependent cell line,
FDCPE, was prepared by transfection of the full-length human EPO-R cDNA into FDC-PI cells followed by selection
in medium containing EPO ( I U/mL). Colonies of EPOdependent cells formed within 10 to 14 days. However, mocktransfected or untreated FDC-PI cells did not give rise to
EPO-dependent clones when cultured for extensive periods
of time in EPO-containing medium.
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
YET AND JONES
1716
1.4
Fraction Number
- B
Fraction Number
C
D
28 29 30 31 32 33 34 35 36
- 66
- 106
- 80
- 45
- 50
- 31
- 21
- 33
- 28
- 19
Fig 3. (A) Hydrophobicinteraction chromatography of the truncated receptor. The extracted ammonium sulfate pellet (see the Materials
and Methods) was applied t o a phenyl hydrophobic interaction column. The column was developed with a reverse gradient of saturated
ammonium sulfate from 35%t o 0%.Fractions no. 23 through 29 were pooled. (B) Chromatographic profile of the truncated EPO-R purification by gel filtration. Fractions containing the truncated EPO-R from the hydrophobic(phenyl) interaction chromatographic step were loaded
onto a TSK-2000 gel filtration column developed in PBS. (C) SDS-PAGE of fractions from the gel-filtration column. Samples of fractions
across the major peaks of protein eluted from the gel-filtration column were examined by SDS-PAGE on a 12%gel. Protein was visualizedby
silver staining (BioRad, Richmond, CA). (D) lmmunoblot of the purified truncated €PO-R. Fraction no. 32 (1 fig) from the gel-filtrationcolumn
was fractionated by SDS-PAGE on a 12%gel. The proteins were transferred to nitrocellulose and the blot was probed with the antimurine
EPO-R antisera, as described in the Materials and Methods.
The FDCPE cells proliferate in a dose-responsive manner
over a range of 0.5 to 10 U/mL of EPO (Fig 4, circles),
expressed as a percentage ofthe maximum signal. The proliferative response is inhibited by increasing concentrations of
the truncated receptor (Fig 4,0.02 to 0.32 pg/mL) in a dosedependent manner, ie, increasing concentrations of the secreted receptor require higher concentrations of EPO to generate an equivalent proliferative signal.
The structure and dissociation constant of the EPO/truncaled receptor complex. By gel-filtration chromatography,
the molecular weights of EPO and the truncated receptor
were estimated to be 51 Kd (Fig SA) and 32 Kd (Fig SB),
respectively; the small peak at 12.4 minutes (Fig 5B) a p
pears to be a dimer of the receptor by immunoblotting (data
not shown). When the two components were mixed, a new
species appeared at 89 Kd and an excess of EPO eliminated
all free receptor (Fig SC), whereas a one-to-one mixture of
both components generated one product (Fig 5D). Quantitative amino terminal sequencing of the 89-Kd complex in
Fig 5 showed two major sequences: APPRLI(C)D--- for
EPO and PPPNLPDP--(AWAPPPNL---, 20%alternative processing) for the truncated receptor in the mole
ratio of 1.1 :1, respectively.
A saturation binding curve (Fig 6) was generated by vary-
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
1717
EPO-R FORMS MONOMERIC COMPLEX WITH EPO
100
80
-
60
0
.-0
‘
U)
40
2c
0.01
0.io
too
10.0
100.0
1000.0
EPO (pM)
Fig 4. Inhibition of EPO-dependent growth of the cell line,
FDCPE, by the truncated EPO-R. Cells were incubated in various
concentrations of EPO in the absence (0)
or presence of the truncated EPO-R ([*I 0.02; [O]0.04; [VI 0.08; [A] 0.16; and [U] 0.32
pg/mL). The activity was plotted as the percentage of the maximum signal of each curve.
ing the quantity of truncated EPO-R added to a fixed quantity of EPO and measuring the amount of free and bound
receptor by intergrating the area under the respective peaks
separated by gel-filtration chromatography. The data were
replotted according to S~atchard,~’
giving a straight line (Fig
6, inset) indicating one population of receptors with a single
dissociation constant of 1.1 nmol/L.
DISCUSSION
We have expressed the ligand-binding domain of the human EPO-R as a secreted protein from a CHO cell line. The
truncated receptor (33 Kd) is immunoprecipitated by the
anti-EPO-R antisera that recognizes the murine and human full-length receptor. The 33-Kd protein is glycosylated
with complex glycan that, when removed, generates a 25Kd species that is close to the theoretical molecular weight
of the protein backbone. However, the apparent molecular
mass difference between the predominant protein (66 Kd)
expressed by the full-length EPO-R cDNA and the deglycosylated form (62 Kd) is only 4 Kd.6The most likely explanation for this difference is that the truncated EPO-R is similar
in terms of carbohydrate structure to the extracellular domain of the more glycosylated, but much less abundant,
forms of the EPO-R of 72 Kd6 or 78 Kd.33The truncated
receptor was readily purified from serum-free conditioned
medium by a series of simple chromatographic steps yielding substantial quantities of pure protein (>90%),which is
recognized by the antimurine EPO-R antisera.
The secreted extracytoplasmic domain retains the ability
to bind EPO and generates a 72-Kd complex upon chemical
cross-linking. In addition, the truncated EPO-R inhibits the
growth of the EPO-dependent cell line, FDCPE, in a dosedependent fashion.
Finally, analysis of the formation of the EPO/truncated
receptor complex by gel filtration, and quantitative amino
terminal sequencing of the complex, indicates that it consists of a one-to-one complex of receptor and growth factor,
with a dissociation constant of 1.1 nmol/L. This finding is
in contrast to human growth hormone, which can bind two
molecules of the cognate
We have not observed the EPO-induced formation of a dimeric truncated
EPO-R complex under our experimental conditions, which
are similar to those for growth hormone/receptor complex
formation.
These results are generally consistent with those observed
for the truncated murine EPO-R expressed in CHO cellsz6
and as a fusion protein produced in E coli.’’ The discrepancies are that, in the former, the isolated receptor could not
be cross-linked to EPO in the presence of DSS and had a kd
AU
0008 0004 -
Fig 5. Complex formation between EPO and the
truncated EPO-R. (A) EPO, 83 pmol; (B) truncated
receptor, 2 5 pmol; (C)EPO plus truncated receptor,
83 pmol and 2 5 pmol, respectively; and ID)EPO
plus truncated receptor, 83 pmol and 100 pmol, respectively, were incubated and chromatographed
on a TSK-G3000SW gel filtration column, as described in the Materials and Methods. The column
was calibrated with thyroglobulin (670 Kd), bovine
gamma globulin (15 8 Kd), chicken ovalbumin (44
Kd). and equine myoglobin (17 Kd), indicated by
arrows.
0.020
0.010
....
- D
0.04 0.12-
0.08 -
670
158
44
t
17.5
i6
min
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
YET AND JONES
1718
-
K,
m.
9.00 --
m
' 0.000
0.2
0.4
0.6
v (fraction bound)
1.00
0.00
0.00
5.00
10.00
15.00
20.00
25.00
0.8
1.0
30.00
[ receptor] free (nM)
of 17 nmol/L, and, in the latter, preliminary results suggest
the formation of dimeric receptor complexes upon crosslinking of the fusion protein with EP0.36Our approach to
studying the truncated EPO-R does not rely on expressing
the protein as a fusion protein in a nonmammalian system,
or using an EPO affinity column in the purification step,
which requires acidic conditions to elute the receptor. These
conditions were found26to significantly reduce the EPObinding capability of the receptor, but apparently not when
the receptor is expressed as a fusion protein.25In addition,
the use of gel-filtration chromatography allows a rapid and
direct study of the EPO/receptor complex formation rather
than by immunoprecipitation, by attachment of the receptor to beads, or through inhibition of EPO binding to cell
lines.
In conclusion, the EPO-R cDNA encodes the principal, if
not the only, EPO-binding protein, and implies that expression of the receptor is sufficient to generate EPO dependency. Outside the context of a cell membrane, the ligandbinding domain forms a low-affinity (kd, 1.1 nmol/L)
one-to-one complex with EPO. These results suggest that
the active EPO/receptor complex may differ from the
growth hormone example, even though the EPO-R ligandbinding domain is probably structurally related in that it
consists of two subdomains. Yet, the question remains of
how the high-affinity EPO-binding site (kd, 50 to 100 pmol/
L) is formed in the membrane. Also undetermined is the
structure of the two EPO cross-linked complexes and the
relationship of these to the biologically relevant receptor
complex. One possibility is that other membrane-bound
proteins, p85 and p100,5,17~'8*20
are required to generate an
active complex in conjunction with the EPO-R. Although
we did not observe the formation of dimeric truncated EPOR complexes with EPO, this does not preclude oligomerization of the EPO-R in the context of a cell membrane,6,37,38
such as the granulocyte colony-stimulating factor receptor,39as another possible route to the activation ofthe EPOR. We have observed EPO-like activity with anti-EPO-R
antibodies, but not with the Fab fragments,6which suggests
that multimerization of the receptor could play a role in
signal transduction by the EPO-R.
The availability of the purified form of the ligand-binding
35,00
Fig 6. Saturation binding of EPO by the truncated EPO-R. A constant amount of EPO (81 pmol)
was incubated with the truncated EPO-R (0to 300
pmol). The bound and free receptor for each point
were separated by gel filtration, the quantities of
each calculated as described in the Materials and
Methods, and expressed in terms of concentration
based on the elution volume of the components.
The saturation data were replotted (inset) as
v/[free receptor] against v, where v is the fraction
of receptor bound/total amount of EPO.
domain of the EPO-R will provide an invaluable tool for
dissecting the structure and biologic nature of the EPO/receptor complex.
ACKNOWLEDGMENT
We thank Alan DAndrea for the antimurine EPO-R antibody
and many helpful discussions; Micheal Brenner for amino acid sequencing; and Steve Clark, Dale Cumming, John Knopf, Hubert
Scoble, and Jas Seehra for their suggestions, support, and encouragement.
REFERENCES
1. Krantz SB: Erythropoietin. Blood 77:419, 1991
2. DAndrea AD, Lodish HF, Wong GG: Expression cloning of
the murine erythropoietin receptor. Cell 57:277, 1989
3. Jones SS, DAndrea DA, Haines LL, Wong GG:Humanerythropoietin receptor: Cloning, expression, and biologic characterization. Blood 76:31, 1990
4. Winkelmann JC, Penny LA, Deaven LL, Forget BG, Jenkins
RB: The gene for the human erythropoietin receptor: Analysis of
the coding sequence and assignment to chromosome 19p. Blood
76~24,1990
5. Dong YJ, Goldwasser E: Evidence for an accessory component that increases the affinity of the murine erythropoietin receptor. Blood 78:9a, 1991 (abstr, suppl 1)
6. Jones SS: Perspectives on the structure and mechanism of
signal transduction by the erythropoietin receptor, in Sasaki R, Kitagawa Y (eds): Advances in Biochemistry and Biology of Membranes, vol. Signal Transduction Through Growth Factor Receptors. Greenwich, CT, JAI (in press)
7. D'Andrea AD, Zon LI: Erythropoietin receptor. J Clin Invest
86:681, 1990
8. DAndrea AD, Jones SS: Activation of the erythropoietin receptor in stable lymphoid and myeloid transfectants. Semin Hemato1 28: 152, 199 l
9. Li JP, DAndrea AD, Lodish HF, Baltimore D: The Friend
spleen focus-forming virus gp55 glycoprotein binds to the erythropoietin receptor and activates cell growth. Nature 343:762, 1990
10. Carroll MP, Spivak JL, McMahon M, Weich N, Rapp UR,
May WS: Erythropoietin induces Raf- I activation and Raf- 1 is required for erythropoietin-mediated proliferation. J Biol Chem
266: 14964, 1991
I I. Miura 0, DAndrea A, Kabat D, Ihle JN: Induction of tyrosine phosphorylation by the erythropoietin receptor correlates with
mitogenesis. Mol Cell Biol 11:4895, 1991
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
EPO-R FORMS MONOMERIC COMPLEX WITH EPO
12. Quelle DE, Wojchowski DM: Localized cytosolic domains
of the erythropoietin receptor regulate growth signaling and downmodulate responsivenessto granulocyte-macrophagecolony-stimulating factor. Proc Natl Acad Sci USA 88:480 I , I99 1
13. Yoshimura A, DAndrea AD, Lodish H F Friend spleen
focus-formingvirus glycoproteingp55 interactswith the erythropoietin receptor in the endoplasmicreticulum and affects receptor metabolism. Proc Natl Acad Sci USA 87:4139, 1990
14. Yoshimura A, Lodish H F In vitro phosphorylation of the
erythropoietin receptor and an associated protein, pp130. Mol Cell
Biol 12:706, 1992
15. Zon LI, Moreau J-F, Koo J-W, Mathey-Prevot B, DAndrea
A D The erythropoietin receptor transmembrane region is necessary for activation by the Friend virus gp55 glycoprotein. Mol Cell
Biol 12:2949, 1992
16. Miyajima A, Kitamura T, Harada N, Yokota T, Arai K-I:
Cytokine receptors and signal transduction. Annu Rev Immunol
10:295, 1992
17. Mayeux P, Lacombe C, Casadevall N, Chretien S, Dusanter
I, Gisselbrecht S: Structure of the murine erythropoietin receptor
complex. J Biol Chem 266:23380, 1991
18. Sawyer ST: Evidence for a two subunit structure of the erythropoietin receptor. Blood 78:10a, 1991 (abstr, suppl 1)
19. Casadevall N, Lacombe C, Muller 0, Gisselbrecht S,
Mayeux P: Multimeric structure of the membrane erythropoietin
receptor of murine erythroleukemia cells (Friend cells). J Biol
Chem 266:16015, 1991
20. Winkelmann JC: The human erythropoietin receptor. Int J
Cell Cloning 10:254, 1992
21. Kuramochi S, Ihawa Y, Todokoro K: Characterization of
murine erythropoietin receptor genes. J Mol Biol 216567, 1990
22. Todokoro K, Kuramochi S, Nagasawa T, Abe T, Ikawa Y:
Isolation of a cDNA encoding a potential soluble receptor for human erythropoietin. Gene 12:283, 1991
23. Maouche L, Toumamille C, Hattab C, Boffa G, Cartron J-P,
Chretien S Cloning ofthe gene encoding the human erythropoietin
receptor. Blood 78:2557, 1991
24. Noguchi CT, Bae KS, Chin K, Wada Y, Schechter AN, Hankins WD: Cloning of the human erythropoietin receptor gene.
Blood 78:2548, 1991
25. Hams KW, Mitchell RA, Winkelmann JC: Ligand binding
properties of the human erythropoietin receptor extracellular domain expressed in Escherichia coli. J Biol Chem 267: 15205, 1992
26. Nagao M, Masuda S, Abe S, Ueda M, Sasaki R: Production
and ligand-binding characteristics of the soluble form of murine
erythropoietin receptor. Biochem Biophys Res Commun 188:888,
1992
1719
27. Hapel AJ, Warren HS, Hume DA: Different colony-stimulating factors are detected by the “interleukin-3”-dependent cell lines
FDC-PI and 32D cl-23. Blood 64:786, 1984
28. Kaufman RJ: Selection and coamplification of heterologous
genes in mammalian cells, in Goeddel D (ed): Methods in Enzymology; Gene Expression Technology, vol 185. New York, NY, Academic, 1990, p 537
29. Kaufman RJ, Davies MV, Wasley LC, Michnick D Improved vectors for stable expression of foreign genes in mammalian
cells by use of the untranslated leader sequence from EMC virus.
Nucleic Acids Res 19:4485, 199 I
30. Dorner AJ, Kaufman RJ: Analysis of the synthesis, processing and secretion of proteins expressed in mammalian cells, in
Goeddel D (ed): Methods in Enzymology; Gene Expression Technology, vol 185. New York, NY, Academic, 1990, p 577
3 1. Andreu JM: Measurement of protein-ligand interactions by
gel chromatography,in Hirs CHW, Timasheff SN (eds): Methods in
Enzymology Part J, vol 117. New York, NY, Academic, 1985, p
346
32. Scatchard G: The attraction of proteins for small molecules
and ions. Ann NY Acad Sci 5 I :660, 1949
33. Sawyer ST, Hankins WD: The functional form of the cloned
EPO receptor is a 78 Kd glycoprotein. Blood 80:147a, 1992 (abstr,
SUPPI 1)
34. Cunningham BC, Ultsch M, de Vos AM, Mulkemn MG,
Clauser KR, Wells JA: Dimerization of the extracellulardomain of
the human growth hormone receptor by a single hormone molecule. Science 254921, 1991
35. De Vos AM, Ultsch M, Kossiakoff AA Human growth hormone and extracellulardomain of its receptor: Crystal structure of
the complex. Science 255:306, 1992
36. Hanis KW, Winkelmann J C A high molecular weight complex is formed by the chemical crosslinking of erythropoietin to the
recombinant erythropoietin receptor extracellular domain. Blood
80:21a, 1992 (abstr, suppl 1)
37. Watowich SS, Yoshimura A, Longmore GD, Hilton DJ, Yoshimura Y, Lodish HF: Homodimerization and constitutive activation of the erythropoietin receptor. Proc Natl Acad Sci USA
89:2140, 1992
38. Yonekura S, Chem Y, Donahue KA, Feldman L, Vanasse
GJ, Sytkowski AJ: Erythropoietin receptors induced by dimethyl
sulfoxide exhibit positive cooperativity associated with an amplified biologic response. Proc Natl Acad Sci USA 88:2535, 1991
39. Fukunaga R, Ishizaka-Ikeda E, Nagata S: Purification and
characterization of the receptor for murine granulocyte colonystimulating factor. J Biol Chem 265: 14008, 1990
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
1993 82: 1713-1719
The extracytoplasmic domain of the erythropoietin receptor forms a
monomeric complex with erythropoietin
MG Yet and SS Jones
Updated information and services can be found at:
http://www.bloodjournal.org/content/82/6/1713.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.