Anti-Erythropoietin (EPO) Receptor Monoclonal Antibodies

Anti-Erythropoietin (EPO) Receptor Monoclonal Antibodies Distinguish
EPO-Dependent and EPO-Independent Erythroid Progenitors in
Polycythemia Vera
By Michael J. Fisher, Jaroslav F . Prchal, Josef T. Prchal, and Alan D. D’Andrea
Erythroid progenitorcells isolated from patients
with polycythemia vera (PV) proliferate and differentiate in methylcellulose in the absence of exogenous erythropoietin (EPO). To
investigate the potential role of the erythropoietin
receptor
(EPO-R) in the pathogenesis of PV, we cultured bone marrow-derived or peripheral blood-derived erythroid progenitors in the presence of neutralizing monoclonal antibodies
(MoAbs) specific for EPO or EPO-R. Mononuclear cells were
obtained from 9 healthy adults and 9 PV patients byFicollHypaque gradients and cultured with or without EPO in
methylcellulose for 12 days under standard or serum-free
conditions. Neutralizing anti-EPO and anti-EPO-R MoAbs,
added t o cultures at day 0, caused dose-dependent growth
inhibition of all normal burst-forming units-erythroid(BFUE) derived from healthy adult controls. The MoAbs had no
effect on the growth
of nonerythroid progenitorcells under
the same culture conditions. In contrast, neutralizing anti-
bodies distinguished two classes of BFU-E derived from PV
patients. Class I BFU-E from PV patients were EPO-dependent. These progenitors, like those derived from healthy
adults, had normalEPO dose-dependent growth characteristics and showed
a normal period of
EPO requirement in vitro
that extended 6 days after the initiation of culture. These
results indicatethat EPO exerts its critical effect early during
erythroid differentiation; the addiiion of neutralizing antibodies t o normal progenitors after 6 days had no effect on
the subsequent size or maturation of the colonies. Class II
BFU-E from PV patients wereEPO-independent. They proliierated and differentiated even in the presence of high concentrations of neutralizing anti-EPO or anti-EPO-R MoAbs.
We conclude that the class II BFU-E from PV patients are
independent of free EPO.
0 1994 by The Americen Society of Hematology.
P
ture erythroblasts. The human EPO-R is a 508 amino acid
membrane protein and is a member of the cytokine receptor
~uperfamily.’~,’~
The number of EPO-R per cell surface increases as a BFU-E matures to a colony-forming unit-erythroid (CFU-E) but then decreases serially as a CFU-E matures to a reticu10cyte.l~~’~
Mature erythrocytes have no cell
surface EPO-R (D’Andrea, unpublished observation). Ectopic expression of the human EPO-R polypeptide in a murine cell line, Ba/F3, confers EPO binding, EPO-dependent
growth, and EPO-dependent erythroid differentiation.”
Several lines of evidence suggest that the EPO-Rmay
play a role in PV. First, the EPO-R plays a central role in
the pathogenesis of one form of murine polycythemia?’ In
this disease, the glycoprotein gp55 of the Friend spleen focus-forming virus binds to the murine EPO-R and constitutively activates the receptor, giving rise to EPO-independent
polycythemia and eventually to erythroleukemia. Secondly,
some studies suggest that the EPO-R is altered or defective
in the erythroblasts from human subjects with PV. Whereas
normal CFU-E appear to have both high- and low-affinity
EPO-binding sites:’ CFU-E from patients with PVhave only
a single class of low-affinity EPO-R (kd = 720 pmoVL).”
Loss of the high-affinity binding site onPV erythroblasts
could account for the loss of EPO-dependence. Third, mutations in the EPO-R polypeptide alter its inherent sensitivity
to EP0.23Truncation of the carboxy terminal 70 amino acids
of the human EPO-R is one cause of familial erythro~ y t o s i s . ’ ~It. ~is
~ therefore plausible that mutations in the
human EPO-R could account for the altered sensitivity of
PV cells to EPO for at least a subpopulation of patients.
Other mutations in the EPO-R have been found that result
in the constitutive activation of this receptor. A point mutation (R129C) results in the constitutive homodimerization
of the murine EPO-R and signal transduction in the absence
of EP0.27Such a mutation could account for the EPO-independent growth of erythroblasts from patients withPV.
Moreover, the effect of an activated EPO-Rneednot
be
restricted to erythroid cells. Recent studies have shown that
OLYCYTHEMIA VERA (PV) is a clonal myeloproliferative disorder, characterized by erythrocytosis and,
in some cases, by elevations of granulocytes and platelets.
The increased redblood cell production in PV is not the
result of increased erythropoietin (EPO) production. For PV
patients, circulating serum levels of EPO are normal’ or
lower than normal.’ Instead, erythrocytosis results from proliferation of an abnormal clone of erythroid progenitor^.^.^
Some studies suggest that the abnormal clone is independent
of EPO.”’ Other studies suggest that the abnormal clone is
hypersensitive, but still dependent, on EP0.9”2 More recently, the growth characteristics of PV burst-forming unitserythroid (BFU-E) in response to other hematopoietic growth
factors has been evaluated. BFU-E from PV subjects were
found to be hypersensitive to insulin-like growth factor-]
(IGF-l)” or to interleukin-3 (IL-3).I4
EPO initiates its cellular response by binding to the EPO
receptor (EPO-R) that is expressed on the surface of imrnaFrom the Division of Pediatric Oncology and Division of Cellular
and Molecular Biology, Dana-Farber Cancer Institute, Harvard
Medical School, Boston, MA; the Departments of Medicine and
Oncology of McGill University, Montreal, Quebec, Canada; and the
Department of Medicine, Division of Hematology/Oncology, University of Alabama at Birmingham, AL.
Submitted October 5, 1993; accepted May 13, 1994.
Supported by a grant from National Institutes of Health (Award
No. R 0 1 DK 43889-01) (A.D.D.),by the Sandoz Corp (A.D.D.),and
by the HHMI (M.J.F.). A.D.D.is a Lucille P. Markey Scholar. Also
supported in part by a grant from the Lucille P. Markey Charitable
Trust.
Address reprint requests to Alan D.D’Andrea, MD, Pediatric
Oncology, Dana-FarberCancer Institute, 44 Binney St, Boston, MA
02115.
The publication costsof this article weredefrayed in part by page
charge payment. This article must therefore be hereby marked
“advertisement” in accordance with 18 V.S.C. section 1734 solely to
indicate this fact.
0 1994 by The American Society of Hematology.
0006-497I/94/8406-OOI 7$3.00/0
1982
Blood, Vol 84, No 6 (September 15). 1994 pp 1982-1991
1983
ANTI-€PO RECEPTOR MONOCLONAL ANTIBODIES
the R129C EPO-R may be functionally expressed in megakaryocytes, perhaps accounting for the thrombocytosis observed in PV.28.29
To study the potential role of the EPO-R in the pathogenesis of PV, we have used neutralizing monoclonal antibodies
(MoAbs) that bind to either EP03’ or EPO-R.31Our results
confirm that both EPO-dependent erythroid progenitors and
EPO-independent erythroid progenitors are present in the
bone marrow and peripheral blood of patients with PV. The
EPO-dependent progenitor cells from PV patients are indistinguishable from the progenitors of normal adult controls
with regard to their EPO sensitivity and their developmental
stage of EPO requirement. In contrast, the EPO-independent
progenitor cells from PV are not inhibited by neutralizing
concentrations of anti-EPO or anti-EPO-R at any time during their 12-day development into burst colonies.
MATERIALS AND METHODS
Subjects. PV patients were selected according to the following
guidelines. All patients had increased red blood cell volume (males,
2 3 6 mL/kg; females, 2 3 2 mwkg) and normal arterial oxygen saturation (292%) and either splenomegaly or two of the following:
thrombocytosis (>400,000/pL), leukocytosis (> 12,000 cells/pL),
increased leukocyte alkaline phosphatase, or increased serum vitamin BIZ(>900 pg/mL). No subjects had been treated with hydroxyurea in the 3 months before study. Normal subjects were adult volunteers. All peripheral blood or bone marrow samples were obtained
via the guidelines of our institutional review boards for human subject experimentation.
Materials. Recombinant human EPO was a generous gift from
Kirin Brewery (Tokyo, Japan) (specific activity, 300,000 U/mg).
Stem cell factor (SCF) and recombinant human L - 3 were generously
provided by Genetics Institute (Cambridge, MA). MoAbs were previously described. MoAb 5.5.1 1 is a neutralizing anti-EPO MoAb.”
MoAb 16.5.1 is a neutralizing anti-EPO-R MoAb?‘ MoAb stocks
were prepared as 1 mg/mLin phosphate-buffered saline (PBS;
GIBCO, Grand Island, NY). All growth factors and MoAbs were
stored at -80°C.
Cell cultures. Peripheral blood was obtained by venupuncture,
and bone marrow was obtained by aspiration from the posterior iliac
crest. Mononuclear cells were isolated by separation over FicollPaque (Pharmacia, Uppsala, Sweden). Cells were washed three times
in Iscove’s modified Dulbecco’s medium (IMDM; GIBCO) and
plated at various concentrations (50,000 to 200,000 cells/mL) in
standard l-mL methylcellulose (MC) cultures containing 0.9% MC
(Teny Fox Laboratories, Vancouver, BC, Canada), 30% heat-inactivated fetal calf serum (FCS; HyClone Laboratories, Logan, UT), 2mercaptoethanol
m o m ; Sigma, St Louis, MO), and either
10% leukocyte-conditioned medium or 0.9% bovine serum albumin
(BSA; Fraction V; Sigma) that was deionized with an analytical
grade mixed bed resin (BioRad, Richmond, CA), 0.05% NaHC03,
2 mmoyL glutamine, penicillin, streptomycin, 50 ng/mL SCF, and
20 ng/mL IL-3.” Alternatively, for serum-free cultures, FCS was
omitted and transferrin (300 pg/mL; Boehringer Mannheim, Indianapolis, IN), bovine pancreas insulin (10 pg/mL; Sigma), and BSAmom) were added.
adsorbed cholesterol (2 X
Cultures were plated in duplicate in the presence or absence of 1
U/mL EPO. Before plating, one of the following was added to the
plates: (1) no Ab, (2) anti-EPO-R MoAb, (3) anti-EPO MoAb, or
(4) control MoAb (either a nonneutralizing anti-EPO-R MoAb, a
nonneutralizing anti-EPO MoAb, or MOP-C21 [an IgGlk MoAb
provided by Genetics Institute]). Alternatively, in the indicated experiments, cells were plated in MC cultures first and MoAbs were
added dropwise to the center of the plate at various times after the
initiation of culture. Final MoAb concentration was 60 nmol/L in
l-ml MC plates. Cultures were incubated at 37°C and 5% CO2 until
assessment of colony formation at day 7 (CFU-E) or day 12 (BFUE and nonerythroid colonies) using standard criteria for colony clas~ification.~~
BFU-E colony size ranged from 50 to 1,000 hemoglobinized cells. Results are expressed as the average number of
colonies from duplicate cultures per 1 6 cells.
BaF3 cells were maintained in RPMI plus 10% FCS plus 10%
WEHI-conditioned media, as previously described. The human
erythroleukemia cell lines K562,34OCIMI?’ and TF-l” were grown
in RPMI plus 10% FCS.
Generation of BdF3 cells expressing a constitutively activated
human EPO-R. PXM-hEPO-R, encoding the full-length human
EPO-R cDNA, has previously been described.” The R129C mutation of the human EPO-R (hEPO-R-C) was generated by the M13
mutagenesis method (BioRad). PXM-hEPO-R or PXM-hEPO-R-C
were electroporated into parental, IL-3-dependent BaF3 cells, as
previously de~cribed.’~
Individual BaF3 subclones were isolated by
limiting dilution. Expression of either wild-type hEPO-R or hEPOR-C polypeptides was confirmed by immunoprecipitation, as previously de~cribed.’~
Flow cytornetric analysis. The indicated cell lines (2 X 10’ cells)
were incubated withan anti-hEPO-R MoAb (MoAb 16.5.1; 13
nmom) and were stained with fluorescein isothiocyanate (FITC)conjugated antimouse second antibody. Cells were analyzed by
FACScan (Becton Dickinson, San Jose, CA), as previously de~cribed.~’
EPO-dependentgrowthassay.
The indicated cell lines were
grown in plain media (RPMI + 10% FCS with no supplemental
growth factor) or EPO-supplemented media, as previously des~ribed.’~
To assay factor-independent growth, cells were washed
twice in plain media and resuspended in plain media. Cells (2 X
104/100pL) were added to multiple wells of a 96-well plate. Supplemental EPO or anti-EPO-R MoAb was added as indicated and
growth was measured after 48 hours by the M’IT reduction assay.”
RESULTS
Anti-EPO-R MoAbs inhibit the growth of normal erythroid progenitors. To establish that the neutralizing
MoAbs specifically inhibit erythroid progenitor growth and
to determine appropriate neutralizing concentrations, progenitor cells from a normal subject were plated in cultures
containing 1 U/mL of EPO in the presence of various concentrations of anti-EPO-R MoAb, anti-EPO MoAb, or a
control Ab. Cultures were incubated until assessment of colony formation on day 12.Anti-EPO-R MoAb and anti-EPO
MoAb demonstrated dose-dependent inhibition of EPO-mediated growth of erythroid colonies (Fig 1A). The concentration of MoAb that yielded one-half maximal inhibition was
1 nmoVL for both the anti-EPO-R MoAb and the anti-EPO
MoAb. This concentration was similar to that required for
inhibition of EPO-dependent cell line^.^',^' In contrast, neither MoAb inhibited the formation of nonerythroid colonies
in methylcellulose culture (Fig 1B).
PV patients have EPO-dependent and EPO-independent
erythroid progenitors. To examine the importance of the
interaction between EPO and the EPO-R in PV, progenitor
cells were plated in standard methylcellulose cultures in the
presence or absence of 1 U/mL EPO. Before plating, neutralizing concentrations of one of the following was added
to the plates: no Ab, anti-EPO-R MoAb, anti-EPO MoAb,
FISHER ET AL
i
A
B
c
‘5:
10-5 104 109 10-2 IO-1 loo lo1
102
103
1 I
MoAb Concentratlon (nM)
0
IO=
lo4
loo 10’ lo2
MoAb Concentration (nM)
loJ lo4 10‘’
10’
10‘
Fig 1. Anti-EPO-R MoAbs inhibit the growthof normal erythroid progenitors. Progenitor cells (1.5 x lo5)from a normal subject were plated
in standard methylcellulose cultures containing l UlmL of EPO in the presence of various concentrations of anti-EPO-R MoAb (01, anti-EPO
MoAb (m), or a control MoAb (0).
Cultures were assessed on day 12 for (A) BFU-E and (B) nonerythroid colony growth.
or a control Ab. Cultures were then incubated at 37°C until
assessment of colony formation at day 12.
A representative example of 1 of 9 PV patients examined
as well as a normal control are shown in Fig 2. Normal
erythroid progenitors exhibited no growth without the addition of EPO. When EPO was added, there was growth of
approximately SO colonies. This normal erythroid colony
growth was blocked by anti-EPO-R and anti-EPO MoAbs,
but not by the control Ab (Fig 2A).
The PV subject had EPO-independent BFU-E that persisted even in the presence of neutralizing concentrations of
anti-EPO-R MoAb and anti-EPO MoAb. In addition, the
PV subject had normal erythroid progenitors that were dependent on EPO and were suppressed by the MoAbs. The
inhibition was to the level of the EPO-independent number
of colonies (Fig 2B). These findings suggest that, in addition
tonormal EPO-dependent erythroid progenitors, an EPOindependent population of erythroid progenitors exists in PV.
This population is not sensitive to anti-EPO or anti-EPOR MoAbs. Depicted in Fig 2C and D is the nonerythroid
colony formation for the same two subjects. There was no
effect of any of the MoAbs on nonerythroid colony formation. The inhibitory effects of the MoAbs were therefore
specific for erythroid colony formation and not caused by a
nonspecific toxic effect.
To rule out the possibility thatthe PV progenitors are
responding to a minute amount of EPO thought to be present
in the fetal bovine serum usedin the culture media, we
repeated this experiment in serum-free cultures (Fig 3). As
above, a PV subject had a subset of erythroid progenitors
that was EPO-independent in serum-free cultures (approximately 50 colonies) and a subset of normal erythroid progenitorsthat responded to EPO (approximately 100 colonies;
Fig 3B). As in Fig 2, the EPO-dependent colonies, but not
the EPO-independent colonies, were inhibited by anti-EPO
and anti-EPO-R MoAbs. This inhibitory effect was specific;
none of the MoAbs effected nonerythroid colony growth
(Fig 3D).Normal control progenitor cells wereplatedin
parallel (Fig 3A and C).
The results of progenitor studies from 18 subjects (9 normal controls and 9 PV patients) are presented in Table 1.
All 9 PV subjects had a mixture of EPO-independent and
EPO-dependent erythroid progenitors. No differences were
observed in the size of the EPO-independent and EPO-dependent colonies. All 9 normal controls had only EPO-dependent progenitors. These results were obtained irrespective
of the source of mononuclear cells (bone marrow or peripheral blood) and irrespective of the culture conditions (serum
versus serum-free). The anti-EPO and anti-EPO-R MoAbs
did not inhibit the growth of nonerythroid colonies in any
of the cultures (data not shown). For PV subjects, on average,
71.2% of the colonies were EPO-dependent (range, 43.5%
to 92.4%) and 28.8% of the colonies were EPO-independent
(range, 7.6% to 56.6%). There was no correlation between
the percentage of EPO independent colonies and the hematocrits of the individual patients (data not shown).
EPO-dependent growth of erythroid progenitors fromPV
patients and from normal controls. To further evaluate the
possibility that PV erythroid progenitors are hypersensitive
to the growth effects of EPO, progenitor cells from 3 PV
and 3 normal subjects wereplatedin parallel in standard
methylcellulose cultures in the presence of various concentrations of EPO (Fig 4). Cultures were scored at day 12 for
colony formation. For the 3 PV subjects, 16% to 24% of the
bursts arose from EPO-independent progenitor cells (Fig 4,
Y-intercept). When the growth of these EPO-independent
cells was subtracted and the PV EPO-dependent progenitors
were evaluated separately, theyhad EPO responsiveness
similar to progenitors from the normal controls (Fig 4, inset,
P > .lo).
Anti-EPO and anti-EPO-R MoAbs define a critical period
of EPO requirement early after the initiation of culture.
Normal erythroid progenitors mature through defined stages,
from undifferentiated BFU-E to C m - E to mature erythro-
ANTI-EPO RECEPTOR MONOCLONAL ANTIBODIES
1985
PV Subject
Normal Subject
80
-v)
3
1
''1T
60
v)
807
A
0
h
40
Y
i
-*
E
m
0
20
- EPO
+ EPO
- EPO
+ EPO
- EPO
+ EPO
T- D
- EPO
+ EPO
Fig 2. PV patients have EPO-dependent and EPO-independent erythroid progenitors
in serum-containing cultures. Progenitor cells were
plated in standard methylcellulose culturesin t h e presence or absence of 1 U/mL EPO. Before plating, a neutralizing concentration (60 nmol/
L) of one of the following was added to the plates: no MoAb
(W), anti-EPO-R MoAb (B),anti-EPO MoAb LO), or a control MoAb (7).
Cultures
were assessed o n day 12 for BFU-E and nonerythroid colony formation. ( A ) Bone marrow BFU-E colonies from a normal subject. (B1 Bone
(D) Bone marrow nonerythroid
marrow BFU-E colonies from a PV subject. (C) Bone marrow nonerythroid colonies from a normal subject.
colonies from a PV subject.
cyte.I7 'x.'x To determine the period of timc during which thc
EPOEPO-R interaction is critical. ncutraliLing concentrations of anti-EPO-R MoAb or anti-EPO MoAb were added
to normalerythroidProgenitors i n culture atvarioustimes
after cell plating. By day 2 after plating, both anti-EPO and
anti-EPO-R MoAbs no longer inhibit CFU-E formation (Fig
SA). In contrast, the MoAbs were still inhibitory to RFU-E
(a more immature erythroid precursor) as late as 6 days
afterinitiation of cell cultures (Fig SB). These results are
consistent with previous studiey that show that approximately 4 days are required fur a BFU-E to differentiate into
a CFU-E." This inhibitory effect was again specific to the
erythroid colonies; addition of MoAb did not inhibit nonerythroid colony formation at any time after initiation of culture
(Fig SC).
We next compared the critical period of EPO-dependence
for progenitors from a PV sub,ject versus a normal control
(Fig 6). Mononuclear cells were isolatedandcultured
in
slandard rnethylccllulose conditions. A ncutralihg concentration of the anti-EPO MoAb was added a t different times
after the initiation of cell culture. This PV sub.ject had both
EPO-dependent colonies (78%) and EPO-indcpcndent colonies (22%; Y-axis). The EPO-depcndcnt progenitors from
the PV subject developed similarly to progenitors from the
normal subject; they displayed similar neutralization kinetics. The cells became independent of I'l-ee EPO aftcr 6 days
of culture.
A c,orl.siirLltil,el~,
c l c t i w l z ~ ~ t wEPO-R
~ n
(hEPO-K-C)c w n f i J l : s
EPO-irldt,lletltlcnt g , - o ~ t ' l h011 H d F 3 c~clls. Because of thc
limited acccss to PV cells, we attempted io crexte an in vitro
system to model the subset of PV progenitors that are EPOindependent. We generateda novel cDNA encodinga constitutively activated form of the human EPO-R (hEPO-R-C).
ThehEPO-R-C polypeptide contains a misscnsc mutation
(R129C) previously shown to constitrutivcly activatc the m u rine EPO-R p~lypeptide.~'
BdF3 cells arc an early erythroid
FISHER ET AL
1986
PV Subject
Normal Subject
I
n
5.
z
m
F
50-
c
0
25-
at
0-
- EPO
+ EPO
- EPO
+ EPO
- EPO
+ EPO
50
C
u)
8
40
40
.
.-
v)
v)
0
0
c
F
.-K
v)
30
Q)
C
0
0
S
20
Q20
10
z
YC
W
t
B
30
0
10
r
'c
0
0
SL
at
n
- EPO
+ EPO
n
-I
Fig 3. PV patients have EPO-dependent and EPO-independent erythroid progenitors in serum-free cultures. Progenitor cells were plated in
serum-free MC cultures in the presence or absence of 1 U/mL EPO. Before plating, a neutralizing concentration (60 nmol/L) of one of the
following was added to the plates: no MoAb (W), anti-EPO-R MoAb (B), anti-EPO MoAb (U), or a control MoAb (0).
Cultures were assessed
on day 12 for BFU-E and nonerythroid colony formation. (A) Blood BFU-E colonies from a normal subject. (B) Blood BFU-E colonies from a PV
subject. (Cl Blood nonerythroid colonies from a normal subject. (D) Blood nonerythroid colonies froma PV subject.
cell line"' that lack EPO-R and are dependent on murine 1L3 for growth. After transfection with the cDNA encoding the
wild-type human EPO-R, these cells show EPO-dependent
growth.Half-maximalgmwthwas
at 0.1 to 0.2U/mL of
EPO (Fig 7). Transfection of BdF3 cells with the cDNA
encoding
hEPO-R-C
resulted in EPO-independent cell
growth (Fig 7).
Anti-EPO-R MoAhs,firil to inhibit the growth of et-ythroImkernic c.el1.s. We next analyzed theBalF3 subclones, as
well as three human erythroleukemia cell lines, for growth
characteristics in the presence of neutralizing anti-EPO-R
MoAbs (Table 2). These erythroleukemia cell lines, likc PV
erythroid progenitors, grow and differentiate in the absence
of EPO. Anti-EPO-R MoAbinhibited thc growth of the
EPO-dependent BdF3-hEPO-R cells. In contrast, the antiEPO-R MoAb did not inhibit the growth of Ba/F.I-hEPOR-C. OCTMI, TF-1, or K562 cells. These cells continued to
grow when EPO was removed from the medium. By FACS
analysis, all cell lines, except parental Ba/F3 and KS62, expressed ccll surface EPO-R (Fig 8). Like the EPO-independent erythroblasts from PV subjects, these cell lines did not
require the interaction of EPO and EPO-R for their transformed EPO-independent phenotype.
DISCUSSION
Previous studics have suggested that abnormal erythroid
progenitors from patients with PV are either hypersensitive
toEPOor
independent of EPO.However, the true EPO
independence of the PV crythroidprogenitorcells has remained controversial. Low levels of EPO activity in the
methylcellulose cultures used could not bc ruled out."." To
exclude any effect of serum-derived EPO, we have used
neutralizing anti-EPO and anti-EPO-R MoAbs a s well as
serum-free cultures. The neutralizing MoAbs, when used in
appropriately high concentrations,completelyblockthe
EPO-dependentgrowth of normalerythroblastprogenitor
ANTI-EPO RECEPTOR MONOCLONAL ANTIBODIES
1987
Table 1. BFU-E Colonies/106 Cells From Hematopoietic Culturesof All Subjects
+ EPO
- EPO
Subject
Condition
Source
S
S
S
S
N1
N2
N3
N4
N5
N6
N7
N8
N9
PV l
PV 2
PV 3
PV 4
PV 5
PV 6
PV 7
PV 8
PV 9
SF
SF
SF
SF
SF
S
S
S
S
S
SF
SF
SF
SF
15
49
62
61
Abbreviations: N. normal;
BM
BM
BM
PB
PB
PB
PB
PB
PB
BM
PB
PB
PB
PB
PB
PB
PB
PB
No Ab
a-EPO-R
a-EPO
Control Ab
0
0
0
0
0
0
0
0
ND
0
0
0
0
0
ND
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
13
51
69
31
9
44
5
6
60
15
48
44
13
13
37
7
ND
60
S, serum-containing
ND
ND
59
33
14
0
0
1
0
0
17
45
48
24
4
46
9
10
70
No Ab
52
74
242
54
49
61
67
11
21
63
85
150
92
25
155
76
132
140
a-EPO-R
0
0
14
0
0
0
1
0
a-EPO
Control Ab
0
ND
0
0
0
0
ND
1
51
74
200
53
54
67
57
9
25
60
69
132
114
27
165
72
127
143
0
0
12
36
69
14
6
56
32
18
54
17
46
52
26
6
53
11
24
59
culture; SF, serum-free culture; BM, bone marrow; PB, peripheral blood; ND, not done.
cells. Some PV erythroblasts, in contrast, grow in the presence of the MoAbs. Our results show that the abnormal
erythroblasts from PV patients are truly independent of EPO.
Several investigators have suggested that PV patients have
two populations of erythroid progenitors: one population of
normal progenitors and another derived from an abnormal
cl0ne.4".~~
We find that patients with PV have both normal
EPO-dependent progenitors (class I) and EPO-independent
progenitors (class 11). Class I progenitors are indistinguishable from erythroid progenitors of normal adult controls by
three criteria. First, these cells show EPO-dependent growth
and are inhibited by the addition of either anti-EPO or anti-
120
100
E
Y
80
3
U
m
60
la
E
4z
40
r
0
$
20
0
0.0
0.2
0.4
0.6
0.8
1 .o
1.2
EPO Concentration
(unitslml)
Fig 4. EPO-dependent growth of erythroid progenitors from PV patients and normal controls. Progenitor cells from PV patients
10) and
normal subjects (m1 were plated in standard MC cultures in the presence of various concentrations of EPO. Cultures were scored at day 12.
for each PV subject. For statistical analysis,we
Inset showsa plot of the same data after the EPO-independent colonies had been subtracted
examined the between subjects effects of PV versus normal adult controls. We used a two-way repeated measure analysisof variance. We
found no significant difference between PV and normal groups (f = 7.27, P .lo).
FISHER ET AL
A
-.
0
2
4
s
s
.
I
10
T h e of MoAb Additlon (days)
Fig 6. Kinetics of anti-EPO MoAb neutralization ofPV erythroid
progenitors. Neutralizing concentrations (60 nmol/L) ofanti-EPO
MoAb were added to PV ( 0 )and normal (m) bone manowderived
mononuclear cells in culture at various times after plating. Cultures
were assessed a t day 12 for BFU-E colony formation. Results are
expressed asthe percentage of inhibition of maximal colony formetion.
0
2
4
0
2
4
6
8
6
8
0
I
1
Previous studies4' suggested that EPO maintained progenitor viability and prevented programmed cell death. Our results suggest that the critical activity of EPO is relatively
early during erythropoiesis. The period of EPO-dependence
is coincident with the period of peak EPO-R expression on
the cell ~ u r f a c e . " ~Even
' ~ ~ ~when
~
the activity of EPO or
EPO-R is inhibited after 6 days, the colonies continue to
mature into normal bursts in methylcellulose cultures. The
size and morphology of these bursts is normal when scored
at 12 days (data not shown), suggesting that free EPO is
no longer required for the growth or differentiation of the
established colonies after 6 days.
The peripheral blood and bone marrow ofPV patients
Time of MoAb Addition (days)
Fig 5. A critical period ofEPO requirement during normal erythroid differentiation. Neutralizing concentrations (60 nmol/L) of
anti-EPO-R MoAb ( 0 )or anti-EPO MoAb (m) or a control MoAb (0)
were added to normel bone marrow-derived erythroid progenitors
in culture at various times after plating. Cultures were assessed at
day 7 for CFU-E growth (A) or at day 12 for BFU-E (B) and nonerythroid (C) colony formation. Results are expressed asthe percentage
of inhibkion of maximal colony formation.
EPO-R MoAbs (Figs 2 and 3). Second, they have similar
EPO-responsiveness relative to normal control progenitors
(Fig 4). These results are consistent with previous studies
that show normal EPO sensitivity ofPV BFU-E at EPO
concentrations known to affect normal erythroid progenit o r ~ , " .as~ ~
well as recent studies convincingly showing the
normal EPO-sensitivity of PV EPO-dependent erythroid progenitors in serum-free cultures.4oThird, as in normal control
subjects, the class I erythroid progenitors lose their dependence on free EPO after 6 days in methylcellulose culture
(Fig 6).
0.0
0.0
0.5
1 .o
1 .S
EPO Concentration (unltr/ml)
Fig 7. A constitutively activated human EPO-R polypeptida conf e n EPO-indopendentgrowth in transfooted BelF3d b . BalF3 wlb
expreaslng either wild-type hLPO-R (m), e condtutively activated
hEPO-R IR129C) W, or no hatwologous potypaptide (0)ware grown
in varying concentrations of EPO. After 48 hours, cell viabiiity was
measured by the MlT reduction assay?'
ANTI-EPO1989
RECEPTOR MONOCLONAL ANTIBODIES
Table 2. Effect of Anti-EPO-R MoAbs on the Growthof Erythroleukemia Cell Lines
Growth in EPO
in Growth
Plain Media
Cell Surface
Expression of EPO-R
-Anti-EPO-R
+Anti-EPO-R
-Anti-EPO-R
+Anti-EPO-R
Baff 3
Baff3-hEPO-R
Ba/F3-hEPO-R-C
K562
TF-l
OClMl
also contain EPO-independent erythroid progenitors (class
11). These progenitors comprise a relatively small fraction
(29%)of the total progenitors scored in our assays. Our data
suggest that class I1 PV erythroblasts are not hypersensitive
to EPO. This is consistent with a recent report4 that provides
evidence for hypersensitivity of PV erythroblasts to IGF-1,
but not to EPO. The failure of anti-EPO-R MoAbs to inhibit
the growth of class I1 progenitors from PV patients has several possible explanations. First, EPO-independent erythroblasts may lack cell surface EPO-R. Second, these PV erythroblasts may have a mutant cell surface EPO-R that fails to
bind MoAb or that is not inhibited. Third, the EPO-independent erythroblasts may express normal cell surface EPO-R
but may not require the EPO-R for cell growth. A growth
D
Fig 8. Cell surface expression
of EPO-R on erythroleukemia cell
lines. The indicated cell lines (2
x lo6cells) were incubatedwith
or without
an
anti-hEPO-R
MoAb and thenstained
with
FITC-conjugated antimouse second antibody. Fluorescence was
analyzed by FACScan as described in Materials and Methods. (AI Ba/F3, (B1 BdF3-hEPOR, (C) BdF3-hEPO-R-C, ID) K562,
(E) TF-1, and (F1 OCIM1.
99
' +
F
Fluorescence
1990
FISHER ET AL
signal, distal to the EPO-R in the signal transduction pathway, may be constitutively activated. Alternatively, another
growth factor, such as insulin or IGF-l, may be implicated
in the proliferation ofPV erythroid progenitors. We have
not yet distinguished among these possibilities.
Interestingly, the anti-EPO-R MoAb failed to inhibit the
growth of erythroleukemic cell lines that have cell surface
EPO-R (Table 2). Ba/F3-hEPO-R-C cells express a novel,
constitutively activated EPO-R polypeptide on the cell surface, yet are not inhibited by the MoAb. By analogy, the
growth of class I1 PV erythroblasts could be accounted for
by a constitutively activated form of the EPO-R that arose
by somatic mutation. Previous studies have shown that neutralizing MoAbs can inhibit cellular growth mediated by
constitutively activated receptor polypeptides. For instance,
fibroblasts transformed by the constitutively activated m u
oncogene product are inhibited by MoAbs to its cell surface
produ~t.~’
The failure of our antibodies to inhibit the constitutively activated hEPO-R suggests several possible models
of EPO-R activation. First, the hEPO-R-C may be generating
a growth stimulus from an intracellular compartment, and
would therefore be inaccessible to cell surface antibody inhibition. However, based on our FACS analysis, at least some
of the hEPO-R-C is expressed on the cell surface (Fig 8).
Secondly, the anti-EPO-R MoAbs may block EPO-induced
dimerization of the EPO-R but fail to block the constitutive
homodimerization induced by the R129C mutation.
Recent studies have shown that mutations in the EPO-R
account for certain disease states. A truncation of 70 amino
acids from the carboxy terminus of the human EPO-R results
in an EPO-R polypeptide that is hypersensitive to EP0.”.25
This mutation accounts for at least some cases of autosomal
dominant familial erythrocytosis, although it does not account for all known cases.43 A missense mutation in the
murine EPO-R (R129C) results in a constitutively activated
EPO-R.” Mice infected with a virus encoding this mutant
mEPO-R develop erythrocytosis, mild thrombocytosis, splenomegaly, and EPO-independent erythroid colony formation
in ~ i t r o . ’ This
~ . ~ ~same mutation, in the context of the human
EPO-R polypeptide, also confers constitutive growth (Table
2). The results of this study suggest that somatic mutations
of the EPO-R could, theoretically, account for the EPOindependence of erythroid progenitors fromat least some
patients with PV. On this basis, we are currently screening
PV patients for the presence of mutant EPO-R polypeptides.
ACKNOWLEDGMENT
We thank Colin Sieff, Andrew Laurie, Bernard Mathey-Prevot,
Martin Carroll, Jon Ellen, and David Nathan for helpful discussions.
REFERENCES
1. Koeffler HP, Goldwasser E: Erythropoietin radioimmunoassay
in evaluating patients with polycythemia. Ann Intern Med 94:44,
1981
2. Cotes PM, Dore CJ, Yin JAL, Lewis SM, Messinezy M, Pearson TC, Reid C: Determination of serum immunoreactive erythropoietin in the investigation of erythrocytosis. N Engl J Med 315:283,
1986
3. Adamson JW, Fialkow PJ,Murphy S, Prchal JF, Steinmann
L: Polycythemia vera: Stem cell and probable clonal origin of the
disease. N Engl J Med 295:913, 1976
4. Prchal JF, Adamson JW, Murphy S, Steinmann L, Fialkow PJ:
Polycythemia vera. The in vitro response of normal and abnormal
cell lines to erythropoietin. J Clin Invest 61:1044, 1978
5. Prchal JF, Axelrad AA: Bone-marrow responses in polycythemia vera. N Engl J Med 290: 1382, 1974
6. Eridani S, Dudley JM, Sawyer BM, Pearson TC: Erythropoietic
colonies in a serum-free system: Results in primary proliferative
polycythaemia and thrombocythaemia. Br J Haematol 67:387, 1987
7. de Wolf JTM, Beentjes JAM, Esselink MT, Smit JW, Halie
RM, Clark SC, Vellenga E: In polycythemia vera human interleukin
3 and granulocyte-macrophage colony-stimulating factor enhance
erythroid colony growth in theabsence of erythropoietin. Exp Hemato1 17:981, 1989
8. Najman A, Baillou C, Drouet X, Krystal G, Isnard F, Gonn
NC: Late erythroid progenitors in polycythemia vera are independent
of erythropoietin. Blood 70:264a, 1987 (abstr, suppl 1)
9. Golde DW, Bersch N, Cline MJ: Polycythemia vera: Hormonal
modulation of erythropoiesis in vitro. Blood 49:399, 1977
10. Casadevall N, Vainchenker W, Lacombe C, Vinci G, Chapman J , Breton-Conus J, Varet B: Erythroid progenitors in polycythemia vera: Demonstration of their hypersensitivity to erythropoietin
using serum-free cultures. Blood 59:447, 1982
11. Zanjani ED, Lutton JD, Hoffman R, Wasserman LR: Erythroid colony formation by polycythemia vera marrow in vitro. Dependence on erythropoietin. J Clin Invest 59:841, 1977
12. Eaves CJ, Eaves AC: Erythropoietin (Ep) dose-response
curves for three classes of erythroid progenitors in normal human
marrowandin
patients with polycythemia vera. Blood52:1196,
1978
13. Correa P, Axelrad AA: Circulating erythroid progenitors in
polycythemia vera are hypersensitive to IGF-I. Blood 78:381a, 1991
(abstr, suppl 1)
14. Dai CH, Krantz SB, Means RT, Horn ST, Gilbert HS: Polycythemia vera blood burst-forming units-erythroid are hypersensitive
to interleukin-3. J Clin Invest 87:391, 1991
15. Jones SS, D’Andrea AD, Haines LI, Wong GC: Human erythropoietin receptor: Cloning, expression and biological characterization. Blood 76:31, 1990
16. Winkelman JC, Penny LA, Deaven LL, Forget BC, Jenkins
RB: The gene for the human erythropoietin receptor: Analysis of
the coding sequence and assignment to chromosome 19p. Blood
76:24, 1990
17. Sawada K, Krantz SB, Dai CH, Koury ST, Horn ST, Glick
AD, Civin CI: Purification ofhumanblood
burst-forming unitserythroid and demonstration of the evolution of erythropoietin receptors. J Cell Physiol 142:219, 1990
18. Wickrema A, Krantz SB, Winkelmann JC, Bondurant MC:
Differentiation and erythropoietin receptor gene expression in human
erythroid progenitor cells. Blood 80:1940, 1992
19. Liboi E, Carroll M, D’Andrea AD, Mathey-Prevot B: The
erythropoietin receptor signals both proliferation and erythroid-specific differentiation. Proc Natl Acad Sci USA 90:1135 1, 1993
20. Ben-David Y, Bernstein A: Friend virus-induced erythroleukemia and the multistage nature of cancer. Cell 662331, 1991
21, Sawada K, Krantz SB, Sawyer ST, Civin CI: Quantitation of
specific bindmg of erythropoietin to human erythroid colony-forming cells. J Cell Physiol 137:337, 1988
22. Means RT, Krantz SB, Sawyer S, Gilbert HS: Erythropoietin
receptors in polycythemia vera. J Clin Invest 84:1340, 1989
23. D’Andrea AD, Yoshimura A, Youssoufian H, Zon LI, Koo
J-W, Lodish HF: Thecytoplasmic region of the erythropoietin receptor contains nonoverlapping positive and negative growth-regulatory
domains. Mol Cell Biol 1 1 : 1980, 1991
ANTI-EPO RECEPTOR MONOCLONALANTIBODIES
24. de la Chapelle A, Sistonen A, Lehvaslaiho P, Ikkala H, Juvonen E: Familial erythrocytosis genetically linked to erythropoietin
receptor gene. Lancet 34132, 1993
25. de la Chapelle A, Traskelin A, Juvonen E Truncated erythropoietin receptor causes dominantly inherited benign human erythrocytosis. Proc Natl Acad Sci USA 90:4495, 1993
26. Juvonen E, Ikkala E, Fyhrquist F, Ruutu T: Autosomal dominant erythrocytosis caused by increased sensitivity to erythropoietin.
Blood 78:3066, 1991
27. Yoshimura A, Longmore G, Lodish HF: Point mutation in
the exoplasmic domain of the erythropoietin receptor resulting in
hormone-independent activation and tumorigenicity. Nature 348:
647, 1990
28. Pharr PN, Hankins D, Hofbauer A, Lodish HF, Longmore G:
Expression of a constitutively active erythropoietin receptor in primary hematopoietic progenitors abrogates erythropoietin dependence and enhances erythroid colony-forming unit, erythroid burstforming unit, and granulocyte/macrophage progenitor growth. Proc
Natl Acad Sci USA 90:938, 1993
29. Longmore CD, Pharr P, Neumann D, Lodish HF: Both megakaryocytopoiesis and erythropoiesis are induced in mice infected
with a retrovirus expressing an oncogenic erythropoietin receptor.
Blood 82:2386, 1993
30. D’Andrea AD, Szklut PJ, Lodish HF, Alderman EM: Inhibition of receptor binding and neutralization of bioactivity by antierythropoietin monoclonal antibodies. Blood 75:874, 1990
31. D’Andrea AD, Rup BJ, Fisher MJ, Jones S: Anti-erythropoietin receptor (EPO-R) monoclonal antibodies inhibit erythropoietin
binding and neutralize bioactivity. Blood 82:46, 1993
32. Sieff CA, Emerson SG, Mufson A, Gesner TG, Nathan DC:
Dependence of highly enriched human bone marrow progenitors
on hematopoietic growth factors and their response to recombinant
erythropoietin. J Clin Invest 77:74, 1986
33. Gregory CJ, Eaves AC: Human marrow cells capable of erythropoietic differentiation in vitro: Definition of three erythroid colony
responses. Blood 49:855, 1977
34. Anderson LC, Jokinen M, Gahmberg CC: Induction of ery-
1991
throid differentiation in the human leukaemia cell line K562. Nature
278:364, 1979
35. Papayannapoulou T, Raines E, Collins S, Nakamoto B,
Tweeddale M, Ross R:Constitutive and inducible secretion of platelet-derived growth factor analogs by human leukemic cell lines coexpressing erythroid and megakaryocytic markers. J Clin Invest
79:859, 1987
36. Kitamura T, Tange T, Terasawa T, Chiba S, Kuwake T, Miyagawa K, Pia0 Y, Miyazono K, Urabe A, Takaku F Establishment
and characterization of a uniquehuman cell line that proliferates
dependently on GM-CSF, IL-3, and erythropoietin. J Cell Physiol
140:323, 1989
37. Mosmann T: Rapid colorimetric assay for cellular growth
and survival: Application to proliferation and cytotoxicity assays. J
Immunol Methods 65:55, 1983
38. Broudy VC, Lin N, Brice M, Nakamoto B, Papayannopoulou
T: Erythropoietin receptor characteristics on primary human erythroid cells. Blood 77:2583, 1991
39. Lacombe C, Casadevall N,Varet B: Polycythemia vera: In
vitro studies of circulating erythroid progenitors. Br J Haematol
44:189, 1980
40. Correa PN, Eskinazi D, Axelrad AA: Circulating erythroid
progenitors in polycythemia vera are hypersensitive to insulin-like
growth factor-l in vitro: Studies in an improved serum-free medium.
Blood 83:99, 1994
41. Koury MJ, Bondurant MC: Maintenance by erythropoietin of
viability and maturation of murine erythroid precursor cells. J Cell
Physiol 137:65, 1988
42. Drebin JA, Link VC, Stem DF, Weinberg R A , Greene MI:
Down-modulation of an oncogene protein product and reversion of
the transformed phenotype by monoclonal antibodies. Cell 41:695,
1985
43. Sokol L, Prchal JF, Prchal JT: Primary familial and congenital
polycythemia. Lancet 342:115, 1993
44. Longmore CD, Lodish HF: An activating mutation in the
murine erythropoietin receptor induces erythroleukemia in mice: A
cytokine receptor superfamily oncogene. Cell 67:1089, 1991