Developmental changes in iron metabolism and erythropoiesis in mice with human gain‐of‐function erythropoietin receptor

Abstract

Iron availability for erythropoiesis is controlled by the iron‐regulatory hormone hepcidin. Increased erythropoiesis negatively regulates hepcidin synthesis by erythroferrone (ERFE), a hormone produced by erythroid precursors in response to erythropoietin (EPO). The mechanisms coordinating erythropoietic activity with iron homeostasis in erythrocytosis with low EPO are not well defined as exemplified by dominantly inherited (heterozygous) gain‐of‐function mutation of human EPO receptor (mtHEPOR) with low EPO characterized by postnatal erythrocytosis. We previously created a mouse model of this mtHEPOR that develops fetal erythrocytosis with a transient perinatal amelioration of erythrocytosis and its reappearance at 3–6 weeks of age. Prenatally and perinatally, mtHEPOR heterozygous and homozygous mice (differing in erythrocytosis severity) had increased Erfe transcripts, reduced hepcidin, and iron deficiency. Epo was transiently normal in the prenatal life; then decreased at postnatal day 7, and remained reduced in adulthood. Postnatally, hepcidin increased in mtHEPOR heterozygotes and homozygotes, accompanied by low Erfe induction and iron accumulation. With aging, the old, especially mtHEPOR homozygotes had a decline of erythropoiesis, myeloid expansion, and local bone marrow inflammatory stress. In addition, mtHEPOR erythrocytes had a reduced lifespan. This, together with reduced iron demand for erythropoiesis, due to its age‐related attenuation, likely contributes to increased iron deposition in the aged mtHEPOR mice. In conclusion, the erythroid drive‐mediated inhibition of hepcidin production in mtHEPOR mice in the prenatal/perinatal period is postnatally abrogated by increasing iron stores promoting hepcidin synthesis. The differences observed in studied characteristics between mtHEPOR heterozygotes and homozygotes suggest dose‐dependent alterations of downstream EPOR stimulation.

Full text

RESEARCH ARTICLE
Developmental changes in iron metabolism and erythropoiesis
in mice with human gain-of-function erythropoietin receptor
Barbora Kralova
1
| Lucie Sochorcova
1
| Jihyun Song
2
| Ondrej Jahoda
1
|
Katarina Hlusickova Kapralova
1
| Josef T. Prchal
2
| Vladimir Divoky
1
|
Monika Horvathova
1
1
Department of Biology, Faculty of Medicine
and Dentistry, Palacky University,
Olomouc, Czech Republic
2
Division of Hematology & Hematologic
Malignancies, The University of Utah School of
Medicine, Salt Lake City, Utah, USA
Correspondence
Monika Horvathova, Department of Biology,
Faculty of Medicine and Dentistry, Palacky
University, Hnevotinska 3, Olomouc
775 15, Czech Republic.
Email: monika.horvathova@upol.cz
Funding information
Ministry of Health of the Czech Republic,
Grant/Award Number: NV19-07-00412;
European Union - Next Generation EU,
Program EXCELES, Grant/Award Number:
LX22NPO5102; Internal grant of Palacky
University, Grant/Award Number:
IGA_LF_2022_003; Ministry of Education,
Youth and Sports (to animal facility BIOCEV),
Grant/Award Numbers: LM2018126,
CZ.02.1.01/0.0/0.0/18_046/0015861
[Correction added on January 06, 2023, after
first online publication: The copyright line was
changed.]
Abstract
Iron availability for erythropoiesis is controlled by the iron-regulatory hormone
hepcidin. Increased erythropoiesis negatively regulates hepcidin synthesis by eryth-
roferrone (ERFE), a hormone produced by erythroid precursors in response to
erythropoietin (EPO). The mechanisms coordinating erythropoietic activity with iron
homeostasis in erythrocytosis with low EPO are not well defined as exemplified by
dominantly inherited (heterozygous) gain-of-function mutation of human EPO receptor
(mtHEPOR) with low EPO characterized by postnatal erythrocytosis. We previously cre-
ated a mouse model of this mtHEPOR that develops fetal erythrocytosis with a transient
perinatal amelioration of erythrocytosis and its reappearance at 3–6 weeks of age. Pre-
natally and perinatally, mtHEPOR heterozygous and homozygous mice (differing in ery-
throcytosis severity) had increased Erfe transcripts, reduced hepcidin, and iron
deficiency. Epo was transiently normal in the prenatal life; then decreased at postnatal
day 7, and remained reduced in adulthood. Postnatally, hepcidin increased in mtHEPOR
heterozygotes and homozygotes, accompanied by low Erfe induction and iron accumu-
lation. With aging, the old, especially mtHEPOR homozygotes had a decline of erythro-
poiesis, myeloid expansion, and local bone marrow inflammatory stress. In addition,
mtHEPOR erythrocytes had a reduced lifespan. This, together with reduced iron demand
for erythropoiesis, due to its age-related attenuation, likely contributes to increased iron
deposition in the aged mtHEPOR mice. In conclusion, the erythroid drive-mediated inhi-
bition of hepcidin production in mtHEPOR mice in the prenatal/perinatal period is post-
natally abrogated by increasing iron stores promoting hepcidin synthesis. The
differences observed in studied characteristics between mtHEPOR heterozygotes and
homozygotes suggest dose-dependent alterations of downstream EPOR stimulation.
1|INTRODUCTION
Iron is essential for erythropoiesis and its levels are controlled by the iron-
regulatory hormone hepcidin.
1
Hepcidin (encoded by HAMP gene)
expression is regulated by a complex interplay of signals, predominantly by
inflammation, iron status, hypoxia, and erythropoietic drive.
1–4
Increased
erythropoietic activity negatively regulates hepcidin synthesis by erythro-
ferrone (ERFE), a hormone produced by erythroid precursors in response
Received: 28 June 2022 Accepted: 5 July 2022
DOI: 10.1002/ajh.26658
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium,
provided the original work is properly cited.
© 2022 The Authors. American Journal of Hematology published by Wiley Periodicals LLC.
1286 Am J Hematol. 2022;97:1286–1299.
wileyonlinelibrary.com/journal/ajh

to erythropoietin (EPO).
4
A tight control of iron balance is essential to sus-
tain iron demand and avoid iron deficiency or iron overload. Augmented
erythropoiesis whether productive or ineffective, via EPO-induced ERFE
expression, attenuates hepcidin production to increase iron availability for
hemoglobin synthesis.
4
In disease states characterized by ineffective
erythropoiesis such as thalassemia, dyserythropoietic congenital anemias,
and myelodysplastic syndrome, the erythropoiesis is hyperproliferative
and the affected anemic subjects develop secondary iron overload
caused by inappropriately suppressed hepcidin, resulting from excessive
accumulation of immature erythroblasts and exaggerated ERFE produc-
tion.
5,6
The transgenic mice with graded erythroid overexpression of Erfe
develop dose-dependent iron overload and relative hepcidindeficiency.
7
The co-regulation of erythropoiesis and iron metabolism as exem-
plified by ERFE is less clear in congenital diseases characterized by
chronically elevated effective erythropoiesis.
8
Decreased circulating
hepcidin concentrations and iron deficiency were reported among indi-
viduals with Chuvash polycythemia,
9
the congenital erythrocytosis char-
acterized by augmented hypoxia-induced transcription factors (HIFs)
due to a hypomorphic mutation of negative HIFs regulator von Hippel
Lindau gene (VHL
R200W
).
10
Similarly, erythrocytic mice overexpressing
human EPO (named Tg6)
11
displayed iron deficiency and hepcidin sup-
pression.
12
Patients with polycythemia vera (PV), an acquired erythrocy-
tosis with constitutively active JAK2
13
and subnormal EPO levels,
14,15
have slightly elevated ERFE and normal/insufficiently suppressed hepci-
din given the degree of expanded erythropoiesis and iron deficiency.
16
However, little is known about iron metabolism in congenital ery-
throcytosis characterized by low EPO levels; originally named primary
familial and congenital polycythemia (PFCP)
8,14
which is due to het-
erozygous gain-of-function EPO receptor (EPOR) mutations. These
EPOR mutations have typically truncated the cytoplasmic part of
EPOR with loss of its negative regulatory domain leading to aug-
mented EPOR-JAK2-STAT5 signaling, resulting in excessive prolifera-
tion, survival, and differentiation of erythroid progenitors.
8,17–20
We previously created a mouse model of PFCP bearing the gain-of-
function erythrocytosis-causing human EPOR gene mutation (mtHEPOR).
We generated not only the heterozygous mouse model analogous to
human disease but also the homozygous mice.
21,22
We have shown that
mtHEPOR embryos develop erythrocytosis around embryonic day (ED)
17.5, followed by transient amelioration of erythrocytosis in perinatal life
and its reappearance at 3–6weeksof age.
22
Similarly, erythrocytosis is
absent in patients with PFCP in the perinatal period and develops within
a few weeks of neonatal life.
8
The mtHEPOR homozygous mice have
even greater erythrocytosis than their heterozygous counterparts.
21,22
Here, we investigated how the developmental and age-related
changes of erythron in mtHEPOR heterozygotes and homozygotes are
interconnected with changes in iron homeostasis.
2|MATERIALS AND METHODS
2.1 |Animals
The gain-of-function human EPOR knock-in mice (substitution
c.1278C > G; designation of the mutation according to LOVD
database, www.lovd.nl) were produced on a C57BL/6 back-
ground.
21
Detailed information about mouse breeding and
assessment of hematological and iron status parameters,
23
and
detection of megakaryocytes
24
can be found in supplemental
Methods. All analyses were performed according to the regula-
tions of Palacky University (PU) Institutional Animal Care and
Use Committee.
2.2 |Real-time PCR analysis
Quantitative reverse transcriptase-polymerase chain reaction (q-PCR)
was performed as described in supplemental Methods. Relative gene
expression was normalized against β-Actin.
2.3 |Enzyme-linked immunosorbent assay (ELISA)
Commercially available ELISA kits, specified in supplemental
Methods, were used for the measurements of serum levels of
hepcidin, ferritin, erythroferrone, Epo, and selected inflammatory
cytokines.
2.4 |Flow cytometry
Bone marrow (BM) and spleen cells were isolated and co-stained with
FITC-conjugated CD71 and phycoerythrin-conjugated Ter119 anti-
bodies (BD Biosciences).
23,25
The surface expression of CD47 on red
blood cells (RBCs) was determined using FITC-conjugated anti-mouse
CD47 antibody (BD Biosciences).
26
The intensity of fluorescence was
measured by FC500 (Beckman-Coulter). Multi-parameter flow cyto-
metry analysis of hematopoietic stem and progenitor cell (HSPC)
populations was performed as a custom service, as specified in sup-
plemental Methods.
2.5 |Hematopoietic colony assay
Freshly isolated BM cells were plated in methylcellulose media
(StemCell Technologies) according to the manufacturer's instructions
as we previously described
21,27
and further specified in supplemental
Methods. Colonies were scored by standard morphologic criteria
under an inverted microscope (Olympus).
2.6 |RBC lifespan
Red blood cells were isolated from total blood and stained
in vitro with carboxyfluorescein diacetate succinimidyl ester
(CFSE, ThermoFisher Scientific).
28,29
CFSE labeled RBCs were
injected into the tail veins of control mice expressing mouse EpoR
(mEpoR) and their fluorescence was tracked for 35 days using
FC500.
KRALOVA ET AL.1287

2.7 |Human subjects
The Ethics Committee of PU Hospital approved the collection and
analysis of human subjects' samples according to an informed consent
obtained as per the Declaration of Helsinki. Hematocrit (Hct) was
measured using a Sysmex XE-500 analyzer (Sysmex Corp). Iron
parameters were determined with standard biochemical methods.
Hepcidin and ERFE serum concentrations were measured with com-
mercial ELISA kits according to the manufacturers' instructions (DRG
Instruments GmbH and Intrinsic LifeSciences, respectively); serum
EPO was measured by radioimmunoassay.
30
2.8 |Statistics
All data are presented as mean ± standard error of the mean (SEM).
An unpaired t-test with Welch's correction or linear regression analy-
sis was used to determine the statistical significance of the results and
calculated by GraphPad Prism 8 Software (GraphPad Software Inc.).
Correlation analyses were performed using the same software. p<.05
were considered statistically significant.
3|RESULTS
3.1 |Iron-restricted erythropoiesis in mtHEPOR
mice shortly after birth
We first evaluated the evolution of erythrocytosis in mtHEPOR mice
at different postnatal ages: at postnatal day (PD)7, 2.5 months
(referred to as young mice), 6.5 months (mature adult), and
16 months (old). Despite partial amelioration of erythrocytosis in
mtHEPOR mice, we previously reported at PD7,
22
Hct was signifi-
cantly increased in both mtHEPOR heterozygotes and homozygotes
compared to mEpoR controls and remained elevated at all subsequent
analyzed stages (Figure 1A). The highest Hct levels, detected in mtHE-
POR homozygotes, were accompanied by mild splenomegaly
(Figure 1B). A progressive reduction in Hct levels was observed with
aging in animals of all genotypes (Figure 1A). RBC count and hemoglo-
bin were also consistently higher in mtHEPOR heterozygotes and
homozygotes than in mEpoR littermates; both parameters decline with
postnatal aging (Figure S1A).
RBC indices, mean corpuscular hemoglobin (MCH), and mean cor-
puscular hemoglobin concentration (MCHC), indicated limited iron
supply for erythropoiesis in both mtHEPOR heterozygotes and homo-
zygotes in the perinatal period (Figure 1C) but not in mature adult and
old mice (Figure 1D). The absence of microcytosis at PD7 (Figure 1C),
typical for iron deficiency states, was likely masked in mtHEPOR het-
erozygous and homozygous neonates by an increased proportion of
macrocytic RBCs generated in the prenatal period,
31
as a result of
delayed switch from primitive to definitive erythropoiesis we previ-
ously reported in mtHEPOR mice.
22
Reduced RBC indices, anisocyto-
sis, and reticulocytosis observed in four-weeks-old mtHEPOR
heterozygotes and homozygotes (Figure S1B-D) partially or completely
normalized in their older counterparts (Figure 1D,E), including normali-
zation of red cell distribution width (RDW, Figure 1F). This indicated
the existence of a transient iron deficiency not sufficient for productive
erythropoiesis in an early life.
Consistently with the finding that low iron potentiates the com-
mitment of megakaryocytic (Mk)-erythroid progenitors (MEP) toward
the Mk lineage,
32
platelet (PLT) counts were elevated in both mtHE-
POR heterozygotes and homozygotes compared to mEpoR controls at
PD7 (Figure 1G). On the other hand, low PLT observed in mtHEPOR
homozygotes later in life (Figure 1G) resembled the human phenotype
wherein affected patients had slightly decreased PLT counts likely
due to a marked increase in RBC and total blood volumes, suggesting
that their normal PLT mass was diluted by increased blood volume.
8,14
3.2 |Developmental transition from iron
deficiency to increased tissue iron deposition in
mtHEPOR mice
To correlate the temporal changes in RBC indices with iron status,
selected iron parameters were assessed. At the time of prenatal devel-
opment of erythrocytosis at ED17.5,
22
strong downregulation of
Hamp mRNA expression was observed in both heterozygous and
homozygous mtHEPOR fetal livers (FL, Figure 2A). At PD7, signifi-
cantly decreased Hamp mRNA expression was also found in the liver
(Figure 2B) along with decreased serum hepcidin levels (Figure 2C).
This was consistent with lower ferritin levels, reduced non-heme liver
iron concentration (LIC), and diminished expression of the iron-store
regulator of hepcidin, Bmp6
2
in these mice (Figure 2D-F). The mea-
surements of serum iron concentration and transferrin saturation
(TSAT) were precluded at ED17.5 and PD7 due to a limited volume of
samples. The activation of Bmp/Smad pathway, responsible for iron-
induced hepcidin production, was partly blunted in both heterozygous
and homozygous mtHEPOR newborns, as documented by diminished
expression of two Bmp/Smad target genes,
7
Smad7 and Id1
(Figure S2A). Thus, available iron status data, together with reduced
RBC indices (MCH and MCHC), as a whole support the conclusion
about transient iron deficiency. Spleen iron content (SIC) was compa-
rable between mice of all genotypes at PD7 (Figure 2G), but this
may be influenced by reported accelerated destruction of RBCs in
mtHEPOR neonates compared to mEpoR littermates.
22
Normal to increased TSAT, ferritin levels, LIC, and SIC were
detected in mtHEPOR homozygotes at later postnatal age (Figure 2H-
K, Figure S2B,C). This was accompanied by the induction of hepcidin
(Figure 2L,M) and upregulation of hepatic Bmp6 transcripts, primarily
in the mature adult and old mtHEPOR homozygotes (Figure 2N).
Young mtHEPOR heterozygotes had slightly diminished iron stores
(Figure 2I-N), but all analyzed iron status parameters, including hepci-
din, became comparable to mEpoR controls with aging. The increase in
hepatic Smad7 and Id1 expression reflected postnatal restoration of
Bmp/Smad pathway activation (Figure S2D) in both mtHEPOR hetero-
zygotes and homozygotes.
1288 KRALOVA ET AL.

FIGURE 1 Evaluation of hematological parameters and spleen size in mice of all studied genotypes. (A) Hematocrit (Hct) levels in mice of
different postnatal ages: postnatal day 7 (PD7), 2.5 months old (young), 6.5 months old (mature adult), and 16 months old (old) mice. The
decline in Hct levels between young and old animals: 14% in mtHEPOR mice (p< .001), 12% in mEpoR/mtHEPOR mice (p< .001), and 9% in
mEpoR controls (p< .01). (B) Spleen weight index (the ratio of spleen weight to gross bodyweight) in young, mature adult, and old mice. (C and D)
RBC indices; mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), and mean corpuscular volume (MCV) at
PD7 (C) and in young, mature adult, and old mice (D). (E) Percentage of reticulocytes (Retics) in young, mature adult, and old mice. (F) Red cell
distribution width (RDW) in young, mature adult, and old mice. (G) Platelet (PLT) counts at PD7 and in young, mature adult, and old mice. All
parameters were measured in at least 5 mice per each group and genotype at all indicated time points. Values indicate mean ± SEM; *p< .05,
**p< .01, ***p< .001 versus mEpoR;
#
p< .05 and
##
p< .01, mtHEPOR versus mEpoR/mtHEPOR calculated by unpaired t-test with Welch's
correction
KRALOVA ET AL.1289

FIGURE 2 Legend on next page.
1290 KRALOVA ET AL.

Importantly, the ratio between Hamp mRNA and Bmp6 mRNA
(Figure 2O) as well as between Hamp mRNA and LIC (Figure 2P)was
lower in mtHEPOR heterozygotes and homozygotes than in mEpoR con-
trols only at PD7 and indicated inappropriate hepcidin suppression at this
stage. At later stages of postnatal life, normal ratios of Hamp mRNA to
Bmp6 mRNA (Figure 2O)andHamp mRNA to LIC (Figure 2P) excluded
the contribution of relatively decreased hepcidin to iron loading.
3.3 |Inverse relationship between ERFE and
hepcidin in mtHEPOR mice at embryonic and perinatal
periods but not in postnatal life
In agreement with the known inhibition of hepcidin by ERFE,
4
relative
hepatic and splenic Erfe mRNA expression was stimulated in both
mtHEPOR heterozygotes and homozygotes at ED17.5 and PD7
(Figure 3A,B). This corresponded to a significantly higher proportion
of immature erythroid progenitors in fetal hepatic and perinatal periph-
eral blood circulation, which we previously reported in mtHEPOR
homozygotes.
22
Erfe transcripts were only modestly increased in the BM and
spleen of both mtHEPOR heterozygotes and homozygotes at later
stages of postnatal life (Figures 3C,D). Moreover, a negative correla-
tion between Hamp and Erfe expression was only partly retained in
mtHEPOR heterozygotes, but not in mtHEPOR homozygotes, wherein
a positive correlation was detected (Figure S3A). Despite transcrip-
tional Erfe stimulation, serum Erfe levels were below the ELISA detec-
tion threshold in all but three analyzed mtHEPOR mice (one
heterozygote and two homozygotes) (Figure S3B), as were in all
mEpoR control samples. This showed that the degree of Erfe induction
in mtHEPOR mice is much lower compared to previously reported Erfe
up-regulation in a mouse model of β-thalassemia intermedia (Th3/
+
mice)
27
or in control mice after EPO administration
23
(Figure S3B,C).
3.4 |Modest expression of ERFE and absence of
excessive accumulation of immature erythroblasts in
the postnatal mtHEPOR BM and spleen
The assessment of erythroid differentiation and maturation
25
revealed
an increased percentage of BM erythroid cells expressing Ter119 in
young and mature adult mtHEPOR heterozygotes and homozygotes
compared to mEpoR controls (Figure 3E) and their progressive
decline with age (old vs. young mice: mEpoR by 5.82%; mtHEPOR
heterozygotes by 6.37%, and mtHEPOR homozygotes by 7.34%).
No major differences in the distribution of BM erythroid cells into
individual maturation stages were detected in young and mature adult
mice. However, the percentage of immature Ter119
high
CD71
high
BM
erythroblasts (region II,
25
Figure 3E) was lower in old mtHEPOR
homozygotes in comparison to their young counterparts and also
to the age-matched mEpoR controls and mtHEPOR heterozygotes.
Analyses of splenic erythropoiesis showed higher percentage of total
Ter119-positive cells in young and mature adult mtHEPOR heterozy-
gotes and homozygotes than in controls, normal maturation pattern,
and no decline with age (Figure S4). These data showed an absence of
excessive accumulation of immature erythroblasts (known to drive
substantial overproduction of ERFE)
5,6
in mtHEPOR mice during later
postnatal life and also indicated aging-related relative suppression of
BM erythropoiesis in mtHEPOR homozygotes.
We questioned whether a possible attenuation of the EPOR
signaling cascade during aging contributes to a relative decrease of
BM erythropoiesis in mtHEPOR mice. We, therefore, tested
whether the hypersensitive EPO dose–response phenotype of ery-
throid progenitors,
33
we previously published in young mtHEPOR
homozygotes,
21
is maintained in aged mtHEPOR animals. Analysis
of erythroid progenitors' responses to various EPO doses in serum-
containing cultures from old mice revealed EPO hypersensitivity of
mtHEPOR colony forming unit-erythroid (CFU-E) and burst forming
unit-erythroid (BFU-E) progenitors (Figure 3F), identical to that
published for three-months-old animals.
21
Therefore, the EPO
hypersensitivity phenotype, caused by EPO-induced EPOR signal-
ing overactivation,
34
is preserved in BM erythroid progenitors dur-
ing the aging of mtHEPOR animals. In addition, normalization of Erfe
expression (a direct Stat5 target) to an erythroid marker glyco-
phorin A (Gypa), revealed that Erfe mRNA expression is increased in
individual erythroid precursors in mtHEPOR heterozygotes and
homozygotes compared to control mice and that the degree of
induction is comparable at all analyzed stages, that is, ED17.5, PD7,
young, mature adult, and old mice (Figure S5). Additional analyses
comparing Stat5 activation in young versus aged erythroid progeni-
tors, such as immunohistochemical staining of Stat5 phosphoryla-
tion and analysis of other downstream targets were not conclusive.
FIGURE 2 Assessment of hepcidin, iron status parameters, and tissue iron content during prenatal, perinatal, and postnatal period. (A-G)
Parameters analyzed during prenatal and perinatal period: Hepatic hepcidin (Hamp) mRNA expression at embryonic day (ED)17.5 (A) and
postnatal day 7 (PD7) (B); serum hepcidin (C) and ferritin (D) levels at PD7; liver iron content (LIC) (E), hepatic Bmp6 mRNA expression (F), and
spleen iron content (SIC) (G) at PD7. (H-N) Parameters analyzed during postnatal life (in young, mature adult, and old mice): Transferrin saturation
(TSAT) (H), serum ferritin levels (I), LIC (J), SIC (K), hepatic Hamp mRNA expression (L), serum hepcidin (M), and hepatic Bmp6 mRNA expression
(N). (O and P) Hamp mRNA expression relative to Bmp6 mRNA expression (O) and the ratio of Hamp mRNA to LIC (P) at different stages of
ontogenesis. In Panels A, B, and F, the target gene mRNA expression determined by q-PCR was normalized to b-Actin and to the mRNA
expression levels of the mEpoR controls. Serum levels of hepcidin (C and M) and ferritin (D and I) were measured by ELISA. LIC (E and J) and SIC
(G and K) were quantified by colorimetric assay. In Panels L and N, the target gene expression analyzed by q-PCR and normalized to b-Actin,is
presented as fold change relative to young mEpoR control group. Values indicate mean ± SEM. *p< .05, **p< .01 versus mEpoR by unpaired t-test
with Welch's correction; n≥3 mice per analysis and genotype
KRALOVA ET AL.1291

FIGURE 3 Analysis of erythroferrone (Erfe) expression, bone marrow (BM) erythropoietic activity, sensitivity of erythroid progenitors to
erythropoietin (EPO), and Epo production. (A-D) Erfe mRNA expression (normalized to b-Actin) measured in the liver at embryonic day (ED)17.5
(A), in the spleen at postnatal day 7 (PD7, B), and in the BM (C) and spleen (D) of young, mature adult, and old mice by q-PCR. In panels C and D,
the expression is presented as fold change relative to young mEpoR control samples. (E) Flow cytometry analysis of CD71 and Ter119
expression on BM cells in young, mature adult, and old mice of mEpoR,mEpoR/mtHEPOR, and mtHEPOR genotype. The relative number of
cells in regions I to IV,
25
is expressed as a percentage of all viable erythroid cells. (F) EPO dose–response curves of colony forming-unit erythroid
(CFU-E) and burst forming-unit erythroid (BFU-E) progenitors; plots of the number of erythroid colonies as a percent maximum versus the
concentration of recombinant human (rh) EPO. Each point indicates mean ± SEM of three independent experiments (performed each in
duplicates), n =3 mice per each genotype. (G-I) Assessment of Epo production. Hepatic Epo mRNA expression (normalized to b-Actin) at ED17.5
(G) and Epo levels (pg/mL) in the serum at PD7 (H) and in young mice (I) were determined by q-PCR and ELISA, respectively. Values indicate
mean ± SEM; *p< .05, **p< .01, versus age-matched mEpoR controls;
#
p< .05 mtHEPOR versus mEpoR/mtHEPOR calculated by unpaired t-test
with Welch's correction; n≥3 mice per group and genotype at all indicated time points [Color figure can be viewed at wileyonlinelibrary.com]
1292 KRALOVA ET AL.

Nevertheless, the presented data suggest there is no dramatic mod-
ification of EPOR/Stat5 activation in mtHEPOR animals during
aging.
Since EPO is the primary driver of erythropoiesis and ERFE
production,
4,5
we measured its expression during ontogenesis in our
model. Epo expression was prenatally normal in both mtHEPOR
FIGURE 4 Evaluation of inflammatory cytokines, hematopoietic colony assay, assessment of hematopoietic stem and progenitor cell (HSPC)
populations, and Gata1 and PU.1 expression in the bone marrow (BM) during postnatal life. (A) mRNA expression of genes encoding
proinflammatory cytokines. IL-6,Inf-γ,Tgf-β1,andTnf-αexpression were assessed by q-PCR in the BM of mEpoR/mtHEPOR and mtHEPOR mice
compared to mEpoR controls at indicated stages of ontogenesis. Expression of all target genes was normalized to b-Actin and is presented as fold
change relative to young mEpoR control group. (B) Hematopoietic colony assay of BM cells isolated from young (left columns) and old (right columns)
mEpoR/mtHEPOR and mtHEPOR mice compared to mEpoR controls. The colonies were evaluated and counted on day 2: Colony forming unit (CFU)–
Erythroid (CFU-E) or on day 12: CFU-granulocyte, erythroid, macrophage, megakaryocyte (CFU-GEMM), burst forming unit-erythroid (BFU-E), and
non-erythroid colonies, designated as CFU-GM/M/G and comprising CFU-granulocyte, macrophage (CFU-GM), CFU–Macrophage (CFU-M), and
CFU–Granulocyte (CFU-G) under an inverted microscope (CKX41, Olympus). (C) Flow cytometry analysis of selected BM HSPC subpopulations in
young and old mEpoR and mtHEPOR mice. % of total viable Lin

cells is shown per each population; LSK: Lin

Sca-1
+
c-Kit
+
cells; LK: Lin

c-Kit
+
cell;
LT-LSK: Long term-LSK cells (CD150
+
CD48

CD34

LSK cells); GMP: Granulocyte and macrophage progenitors (CD34
+
FcRg
+
LK cells), MEP:
Megakaryocyte and erythroid progenitors (CD34

FcRg

LK cells). (D) Gata1 and PU.1 BM mRNA expression (normalized to b-Actin) assessed in
young and old mice of mEpoR/mtHEPOR and mtHEPOR genotype and age-matched mEpoR controls. The expression of PU.1 and Gata1 mRNA is
presented as fold change relative to young mEpoR controls. Values in all panels indicate mean ± SEM. *p<.05,**p<.01,***p<.001determinedby
unpaired t-test with Welch's correction; n≥3 each group and genotype [Color figure can be viewed at wileyonlinelibrary.com]
KRALOVA ET AL.1293

heterozygotes and homozygotes (Figure 3G), decreased at PD7
(Figure 3H), and remained low in early and late adulthood (Figure 3I
and Figure S6A), in agreement with our previously published data
22
and in accordance with the aforementioned hypersensitive EPO
dose–response phenotype of BM progenitors demonstrated in
young
21
as well as old mtHEPOR animals. This indicates no clear-cut
positive correlation of Epo with Erfe or stimulated erythropoiesis in
mtHEPOR mice. Changes in Epo expression in mtHEPOR mice during
ontogenesis, not seen in control mice, did not seem to be significantly
modulated by differences in hypoxia signaling because of unaltered
expression of several other target genes of HIFs (Figure S6B-D).
In agreement, we previously reported unaltered expression of HIF
target genes in EPOR-mutated patients.
35
The only exception, signifi-
cantly increased Slc2a1 transcripts (encoding glucose transporter
1-Glut1) in ED17.5 FL of mtHEPOR heterozygotes and homozygotes
(Figure S6B), likely reflects increased glucose uptake and meta-
bolisms
36
during the period of the highest erythroid expansion.
3.5 |Progressive augmentation of inflammatory
cytokines and age-related myeloid dominance in the
BM of mtHEPOR homozygotes
In order to evaluate the possible contribution of inflammation to the
regulation of hepcidin production
1,3
in mtHEPOR mice, serum levels of
selected inflammatory cytokines: IL-6, tumor necrosis factor-α(Tnf-α),
interferon-γ(Ifn-γ), and transforming growth factor-β(Tgf-β1) were
measured. No indications for systemic inflammation or local IL-6
mRNA induction in the liver were observed in mtHEPOR heterozy-
gotes and homozygotes (Figure S7A,B). However, q-PCR analysis of
corresponding inflammatory genes revealed their local stimulation
mainly in the BM of mtHEPOR homozygotes (Figure 4A), together
with moderate, but detectable trend toward increased ROS produc-
tion (Figure S7C), suggesting progressive aging-related alterations in
their BM microenvironment. These data indicate activation of inflam-
matory signaling mechanisms in aged mtHEPOR mice, known to pro-
mote BM remodeling, premature aging, and myelopoiesis in a
paracrine or an autocrine fashion.
37–40
We, therefore, evaluated the numbers of individual hematopoietic
progenitors in the BM of young and old animals by clonogenic assays.
While the numbers of colonies derived from a multilineage colony
forming unit granulocyte, erythroid, macrophage, megakaryocyte
(CFU-GEMM) progenitor did not differ, significantly increased num-
bers of BFU-E-derived and/or CFU-E-derived colonies were found in
both young heterozygous and homozygous mtHEPOR mice, compared
to mEpoR controls (Figure 4B). In old mtHEPOR heterozygotes and
homozygotes, however, we observed significant increase in the total
number of non-erythroid colonies derived from granulocyte and/or
macrophage progenitor cells compared to mEpoR controls (Figure 4B).
This was predominantly due to the elevation of CFU-macrophage
(CFU-M) and less so of CFU-granulocyte, macrophage (CFU-GM)
colonies (Figure S8A), and suggested myeloid lineage bias of
hematopoiesis.
38,41
Subsequent flow cytometry of subpopulations of hematopoietic
stem cells (HSCs) and primitive progenitors from mtHEPOR homozy-
gotes and mEpoR controls revealed a significant increase in the num-
bers of lineage (Lin)

Sca-1
+
c-Kit
+
(LSK) stem cells in old mtHEPOR
homozygotes compared to young counterparts, while this increase
was less prominent in mEpoR controls (Figure 4C). From the major
subsets within Lin

c-Kit
+
(LK) cells in mtHEPOR homozygotes,
granulocyte-monocyte progenitors (GMPs) were significantly
increased in old animals compared to young animals (Figure 4C). Since
megakaryocyte–erythroid progenitors (MEPs) dominated the GMPs in
young mtHEPOR mice (Figure 4C), the myeloid expansion in old mtHE-
POR homozygotes reflects the restoration of myelopoiesis, attenuated
in young mtHEPOR counterparts, when compared to controls, indicat-
ing a progressive reduction of erythropoietic drive and augmentation
of myelopoiesis during aging in mtHEPOR mice; to a far greater extent
than in control mice. In agreement, relatively increased transcripts of
the myeloid master regulator PU.1 and decreased transcripts of ery-
throid regulator Gata1
42
can be seen in the BM of old mtHEPOR het-
erozygotes and homozygotes, but not in mEpoR controls (Figure 4D).
It is known that pro-inflammatory cytokines and elevated megakaryo-
cytes in BM microenvironment are drivers of excessive myelopoiesis
in aged hematopoiesis.
38
Indeed, in addition to the aforementioned
local inflammation, a significant induction in the number of BM mega-
karyocytes (1.4-times) between young and old mice was observed in
mtHEPOR homozygotes (Figure S8B,C).
3.6 |Impaired survival of mtHEPOR RBCs
The recycling of iron through erythrophagocytosis by macrophages is
a major contributor to systemic iron homeostasis.
43
We previously
showed that the dramatic decrease of Hct in mtHEPOR homozygotes
at PD7 was paralleled by induced exposure of phosphatidylserine
(PS) on the membrane of mtHEPOR homozygous RBCs,
22
which likely
contributed to their enhanced recognition and destruction by reticulo-
endothelial macrophages.
44
Because PS exposure was comparable in
mtHEPOR homozygotes and mEpoR controls during adult life,
22
we
analyzed the surface expression of CD47, also known as “don't eat
me”signal, another marker priming RBCs for clearance.
45
The CD47
surface expression was reduced on young mtHEPOR heterozygous
and more significantly homozygous RBCs compared to mEpoR RBCs
(Figure S9A). In addition, the in vivo lifespan of mtHEPOR homozygous
RBCs was reduced to almost half compared to mEpoR controls
(Figure S9B) and less so in mtHEPOR heterozygotes. These data indi-
cate that induced RBC recycling in mtHEPOR mice may contribute to
increased iron deposition.
3.7 |Undetectable ERFE in PFCP patients
To investigate whether the results from the mouse model reflect char-
acteristics of human disease, we measured hematological and iron sta-
tus parameters of four PFCP, EPOR-mutant patients (Table S1). Two
1294 KRALOVA ET AL.

of them were available for ERFE and hepcidin levels assessment
(Table S1 and Figure S10). In parallel, patients with congenital erythro-
cytosis caused by augmented hypoxia signaling due to a gain-of-
function mutation of HIF 2-alpha (encoded by EPAS1 gene)
35
were
evaluated. Hereditary hemolytic anemias with ineffective erythropoie-
sis (pyruvate kinase deficiency or hereditary spherocytosis)
46,47
served as positive controls for elevated ERFE and low/inappropriately
normal hepcidin (Figure S10). All pediatric patients with both types of
erythrocytoses had elevated Hct and normal RBC indices (MCV,
MCH, and MCHC), serum iron and ferritin levels, and reduced TSAT
(Table S1). While slightly elevated ERFE and undetectable hepcidin
levels were observed in EPAS1-mutant patients, EPOR-mutant
patients had normal hepcidin and ERFE levels below the detection
threshold (Table S1 and Figure S10). Thus, EPOR-mutant patients with
moderate iron deficiency and normal hepcidin levels resemble young
mtHEPOR murine heterozygotes.
4|DISCUSSION
In this study, we evaluated developmentally-related changes in eryth-
ropoietic activity and iron metabolism in a mouse model of congenital
erythrocytosis with human gain-of-function EPOR.
21,22
We observed
that in the prenatal/perinatal period mtHEPOR heterozygotes and
homozygotes presented with significantly down-regulated hepcidin,
up-regulated Erfe, and iron deficient erythropoiesis (Figures 1C,2A-F,
and 3A,B). This corresponded to significantly higher proportions of
immature erythroid progenitors in the fetal hepatic circulation and
neonatal peripheral blood
22
and reflected an inverse correlation
between Erfe and hepcidin
4,6
and high iron demand by EPOR
mutation-directed augmented erythropoiesis
5
at these stages. In con-
trast, normal to elevated iron parameters and hepcidin levels were
seen in young, mature adult, and old mtHEPOR homozygous mice
(Figure 2H-N). Though young mtHEPOR heterozygotes had slightly
reduced iron stores, adult and old mice with this genotype were iron-
replete with physiological hepcidin levels (Figure 2H-N). Mild Erfe
induction (Figure S3B) was consistent with the absence of excessive
accumulation of immature erythroblasts
5,6
in the BM or spleen of
mtHEPOR heterozygotes and homozygotes during adult life (Figure 3E
and Figure S4). These changes in iron metabolism in the prenatal/
perinatal stage and postnatal life likely reflect the dynamics and
strength of opposed signals, that is, erythropoietic drive and iron
stores, regulating hepcidin production.
48
A dominant factor affecting
hepcidin expression in mtHEPOR fetuses/newborns, the erythropoi-
etic drive, is later overridden by hepatic iron accumulation
(Figure 2O,P).
Low Epo levels in mtHEPOR mice (Figure 3I and Figure S6A), as a
physiological response to erythroid progenitors' Epo hypersensitiv-
ity
21
(Figure 3F), were only observed after birth. During the prenatal
development, Epo production in mtHEPOR mice was normal
(Figure 3G). We suggest that this could be related to the proposed
role of EPO in non-erythroid tissue/organ development.
49
Since het-
erodimeric EPOR complexes, found in non-erythroid cells,
50
require
higher concentrations of EPO for signaling than homodimeric
EPOR,
51,52
the attenuation of EPO synthesis might be blunted during
the prenatal life of mtHEPOR mice. Also, fetal hypoxia and augmented
glucose metabolism, indicated by significantly increased Glut1 tran-
scripts in mtHEPOR FL (Figure S6B), likely linked to increased energy
expenditure associated with the erythroid lineage expansion at murine
FL stage,
36
was reported to require high EPO.
53
We next interrogated the changes in erythron and progenitor cell
populations during the aging of mtHEPOR mice. We observed a pro-
gressive decline in the percentage of Ter119-positive BM cells and
immature Ter119
high
CD71
high
erythroblasts in mtHEPOR heterozy-
gotes and homozygotes with age (Figure 3E). Concomitantly, suppres-
sion of early and late erythroid progenitors' numbers and induction of
non-erythroid progenitors were detected (Figure 4B) and paralleled by
the upregulation of PU.1 expression (Figure 4D). Assessment of
HSPCs revealed a higher increase of LSK cells in aged mtHEPOR mice
than in mEpoR mice (Figure 4C), hematopoietic progenitor cell type
known to increase during aging.
54
Significant increase in GMPs
(Figure 4C) in aged mtHEPOR animals was consistent with progressive
myeloid cell expansion during aging. This, together with significant
induction in the number of BM megakaryocytes in old mtHEPOR
homozygotes (Figure S8B,C), represents signs of premature hemato-
poietic aging.
38
These effects are likely due to a combination of cell-
autonomous as well as microenvironmental regulations. Our earlier
transplantation studies using mtHEPOR mice suggested BM progeni-
tor cell-intrinsic, EPOR-driven regulation of megakaryopoiesis.
55
The
megakaryocytes may then contribute to hematopoietic aging through
secretion of pro-inflammatory cytokines, as do JAK2
V617F
-positive PV
megakaryocytes.
56
In agreement, age-related induction of pro-
inflammatory cytokine mRNA expression was observed in the BM of
mainly mtHEPOR homozygotes (Figure 4A), suggesting alterations in
the BM microenvironment.
38,41,54,56
Increased levels of IL-6, TGF-β,
TNF-α, and/or INF-γare known to suppress erythropoiesis.
57–60
Overall, these data suggest progressive attenuation of BM erythroid
drive in mtHEPOR homozygotes and, to a lesser extent, also in mtHE-
POR heterozygotes when compared to mEpoR controls, further sup-
ported by a more profound decline in Hct levels between young and
old animals of both mtHEPOR heterozygotes and homozygotes com-
pared to mEpoR controls (Figure 1A).
The above-mentioned observations have raised the question of a
possible attenuation of the EPOR signaling cascade in our model dur-
ing aging. Nevertheless, our data do not favor such mechanism, since
prolonged EPO-induced Jak2 and Stat5 activation appears to be main-
tained in mtHEPOR BM erythroid progenitors during aging, as demon-
strated by the preserved EPO-hypersensitivity dose–response
phenotype (Figure 3F) and comparable extent of Erfe induction
(a Stat5 target) in individual erythroid precursors during all studied
stages of postnatal development (Figure S5). Other subtle in vivo
modifications of signaling downstream mutant EPOR during aging,
e.g., in MAPK and Akt kinase pathways, cannot be excluded but have
not been analyzed. Another intriguing possibility is that the erythroid
expression of transferrin receptor 2 (TfR2) could be altered in our
mtHEPOR model over time which could eventually affect the
KRALOVA ET AL.1295

presentation of EPOR/TfR2 complexes
61
at the cell membrane.
Indeed, dynamic changes in TfR2 mRNA expression in the erythroid
cells were observed in mtHEPOR mice during ontogenesis (data not
shown), suggesting that TfR2 may modulate EPOR surface presenta-
tion in our model. However, with respect to the complex and different
regulation of the presentation of truncated EPOR on the cell mem-
brane compared to murine EpoR,
62
together with the fact that also
Epo-dependent EpoR internalization is impaired for truncated
EPOR,
63
only a comprehensive genetic study involving breeding of
mtHEPOR mice with the BM-specific TfR2 knock-out mice would help
to address this question.
Instead, our data suggest that mtHEPOR HSCs in the late adult
BM may have an increased proliferative history, contributing to the
relative exhaustion of progenitors with erythroid lineage-priming pro-
gram in the aged marrow. In this regard, we have previously documen-
ted
64
a higher proportion of LSKs (Flt3

Rh123
low
subset of
cKit
+
Sca1
+
cells) to total BM cells in mtHEPOR mice than in knock-in
mice expressing wild-type human EPOR gene (wtHEPOR) that has
hypoactive EPOR signaling.
21,65
In addition, a reduced reconstitution
capacity of mtHEPOR HSCs compared to wtHEPOR HSCs was
observed.
64,66
In the current study, we revealed a significant increase
in the numbers of LSKs (i.e., Lin

Sca-1
+
c-Kit
+
) in old mtHEPOR homo-
zygotes compared to young counterparts, while this increase was less
prominent in aged mEpoR controls. These data collectively suggest
that life-lasting prolonged activation of EPOR-Jak2-Stat5 signaling
promotes the proliferation of LSKs. Increased HSCs activation then
presumably results in their increased proliferative history and aging
that potentially leads to increased myeloid bias.
67
This could be due
to a non-cell-autonomous effect of mild local BM inflammation
(Figure 4A) and moderate, but detectable increase in ROS production
(Figure S7C). This likely leads to paracrine signaling between HSCs
and other cell types in the mtHEPOR BM, facilitating HSC entry into
the cell cycle.
68
In addition, cell-autonomous proliferative stimuli
could also play a role, as suggested by recent studies demonstrating
physiological EpoR expression in a subset of HSCs and their immedi-
ate progeny.
69
In addition to age-related decline of BM erythropoiesis, we also
observed a reduced lifespan of mtHEPOR RBCs (Figure S9B) accompa-
nied by diminished surface expression of CD47 (Figure S9A), which
inhibits phagocytosis of erythrocytes by macrophages of the reticulo-
endothelial system.
45
Diminished surface expression of CD47, related
to enhanced erythrophagocytosis, was previously reported in other
erythrocytic mouse models including Tg6 transgenic mice overexpres-
sing Epo
26
and in JAK2
V617F
PV mouse model.
70
We, therefore, pro-
pose that the enhanced clearance of RBCs together with reduced iron
demand for erythropoiesis due to its age-related decline in the BM
causes elevation in liver iron stores and consequent increase in Bmp6
expression and hepcidin synthesis (Figure 2I,J,L-N).
1,2
We show here that in contrast to so far analyzed other mouse
models of erythrocytoses,
12,71,72
mtHEPOR mice neither present with
postnatal iron deficiency nor with hepcidin suppression (Figure 2H-N).
The absence of expansion of immature erythroblasts in the BM and
spleen of mtHEPOR mice (Figure 3E and Figure S4) is another
distinguishing characteristic. The discrepant phenotypic manifesta-
tions between erythrocytoses with high EPO levels and our model
likely result from differences in the cell-autonomous and non-cell-
autonomous consequences of signaling present in erythrocytosis dis-
orders having high EPO compared to low EPO, which is characteristic
of truncated gain-of-function EPOR of affected PFCP individuals as
well as observed in this mouse model. Studies using BM chimeras gen-
erated from Tg6 EPO transgenics
73
or mtHEPOR mice
55
and C57BL/6
mice as donors to C57BL/6 recipients revealed dissimilarities between
erythropoiesis driven by BM HSPCs after exposure to excessive EPO
versus those bearing gain-of-function EPOR. Besides the cell-intrinsic
mechanisms responsible for transient (Tg6) versus sustained (mtHE-
POR) differentiation potential of BM HSPCs toward erythroid lineage,
chronic exposure to EPO
73
versus prolonged EPOR signaling due to
EPOR gain-of-function,
34
may have different impacts on BM niche. In
this regard, EPOR expression in non-erythroid tissues, including eryth-
roblastic island macrophages, adipocytes, or osteoblasts
74,75
should
be taken into consideration. Moreover, a comparison of two mouse
models of PV, with either V617F- or exon 12-mutant JAK2, showed
that changes in iron metabolism (including Erfe and hepcidin) were
more pronounced in exon 12-mutant mice relative to V617F-mutant
mice.
71
This fits with more prominent erythrocytosis observed in
JAK2 exon 12 compared with JAK2
V617F
-mutant mice, hypothetically
related to exon 12-mutant JAK2 favoring interaction with EPOR.
76
This further indicates that qualitative differences in disordered EPOR/
JAK2 signaling variously modulate iron metabolism and/or erythropoi-
esis itself. Moreover, the differences in the severity of phenotype
between mtHEPOR heterozygotes and homozygotes observed in our
study suggest dose-dependent alterations related to different degrees
of downstream EPOR stimulation. Additional factors, absent in mtHE-
POR mice, including systemic inflammation
10,77
(Figure S7A) and/or
HIF-mediated mechanisms
35,78
(Figure S6B-D), may further impact
iron homeostasis and erythroid drive in PV and erythrocytoses with
augmented hypoxia signaling.
The absence of augmented hypoxia signaling and systemic inflam-
mation may also prevent clonal evolution and malignant transforma-
tion of hematopoiesis with gain-of-function EPOR mutation. PFCP
patients are known to carry polyclonal hematopoiesis
34
and do not
seem to accumulate additional somatic mutations known in PV,
79,80
although a predisposing potential of one of the activating EPOR germ-
line variants (EPOR
P488S
) for myeloproliferative neoplasms has
recently been documented.
81
The above-mentioned changes associ-
ated with mtHEPOR BM microenvironment are mild and do not neces-
sarily represent permanent challenges that would contribute to an
increased incidence of hematological malignancies in PFCP. Some of
the observed changes we describe in mtHEPOR homozygous mice
may not apply to individuals affected by PFCP as these patients are
always heterozygotes.
8,14
Nevertheless, a comprehensive understand-
ing of the predisposing potential of activating EPOR germline variants
for malignant transformation will require long-term monitoring of the
patients and further studies.
We conclude that the erythrocytic phenotype, as well as hepcidin,
Erfe, and Epo expression undergoes dynamic changes during
1296 KRALOVA ET AL.

ontogenesis in a mouse model of congenital erythrocytosis bearing
human gain-of-function EPOR. We propose that even in the absence
of systemic inflammation, albeit with possible paracrine inflammatory
signals, known to affect BM remodeling and hematopoietic aging, life-
lasting prolonged activation of EPOR-Jak2-Stat5 signaling promoted a
progressive decrease of committed erythroid progenitors and resulted
in an age-related decline of accelerated erythropoiesis in mtHEPOR
mice. Overall, our data demonstrate that in mtHEPOR mice with aug-
mented effective erythropoiesis, hepcidin production is to a greater
extent regulated by liver iron content which in turn reflects iron con-
sumption by erythroid cells. Consistently with the mouse model,
patients' analyses supported that ERFE is not the main regulator of
hepcidin in pediatric/young adult EPOR-mutated erythrocytosis with
low EPO levels (Table S1 and Figure S10).
AUTHOR CONTRIBUTIONS
Barbora Kralova: Performed most experiments, analyzed results, and
contributed to manuscript writing; Lucie Sochorcova,Jihyun Song,
Ondrej Jahoda, and Katarina Hlusickova Kapralova: Performed some
experiments and analyzed results. Vladimir Divoky: Contributed to
study design, results interpretation, and wrote the manuscript; Josef
T. Prchal: Contributed to study design, results interpretation, and
revised the manuscript. Monika Horvathova: Designed the research,
interpreted results, and wrote the manuscript. All authors have read
and agreed to the published version of the manuscript.
ACKNOWLEDGEMENTS
This work was supported by the Ministry of Health of the Czech Republic
(NV19-07-00412), the project National Institute for Cancer Research
(Programme EXCELES, ID Project No. LX22NPO5102) - Funded by the
European Union - Next Generation EU, and the Internal grant of Palacky
University (IGA_LF_2022_003). Further support (to animal facility BIO-
CEV) was provided from these projects: LM 2018126 Czech Center for
Phenogenomics by the Ministry of Education, Youth and Sports (MEYS)
OP RDE and CZ.02.1.01/0.0/0.0/18_046/0015861 CCP Infrastructure
Upgrade II by MEYS and ESIF. We thank prof. D. Pospisilova, Faculty
Hospital Olomouc and Palacky University Olomouc, Czech Republic, and
Dr.D.Prochazkova,J.E.PurkyneUniversity,UstinadLabem,
Czech Republic for providing patients' clinical data and patients' sampling.
CONFLICT OF INTEREST
The authors declare no conflict of interest.
DATA AVAILABILITY STATEMENT
The data are available from the corresponding author upon reasonable
request.
ORCID
Ondrej Jahoda https://orcid.org/0000-0001-7246-6449
Katarina Hlusickova Kapralova https://orcid.org/0000-0001-6235-
2915
Vladimir Divoky https://orcid.org/0000-0003-0202-245X
Monika Horvathova https://orcid.org/0000-0003-3857-8986
REFERENCES
1. Ganz T, Nemeth E. Hepcidin and iron homeostasis. Biochim Biophys
Acta. 2012;1823(9):1434-1443.
2. Andriopoulos B Jr, Corradini E, Xia Y, et al. BMP6 is a key endogenous
regulator of hepcidin expression and iron metabolism. Nat Genet.
2009;41(4):482-487.
3. Ganz T, Nemeth E. Iron homeostasis in host defense and inflamma-
tion. Nat Rev Immunol. 2015;15(8):500-510.
4. Kautz L, Jung G, Valore EV, Rivella S, Nemeth E, Ganz T. Identification
of erythroferrone as an erythroid regulator of iron metabolism. Nat
Genet. 2014;46(7):678-684.
5. Ganz T. Erythropoietic regulators of iron metabolism. Free Rad Biol
Med. 2019;133:69-74.
6. Kautz L, Jung G, Du X, et al. Erythroferrone contributes to hepcidin
suppression and iron overload in a mouse model of β-thalassemia.
Blood. 2015;126(17):2031-2037.
7. Coffey R, Jung G, Olivera JD, et al. Erythroid overproduction of ery-
throferrone causes iron overload and developmental abnormalities in
mice. Blood. 2022;139(3):439-451.
8. Prchal JT. Primary and secondary Erythrocytoses/polycythemias.
In Kaushansky K, Lichtman MA, Prchal JT, Levi M, Burns LJ,
Linch DC, eds. Williams Hematology, 10th ed. McGraw Hill; 2021:
941–960.
9. Ang SO, Chen H, Hirota K, et al. Disruption of oxygen homeostasis
underlies congenital Chuvash polycythemia. Nat Genet. 2002;32:
614-621.
10. Gordeuk VR, Miasnikova GY, Sergueeva AI, et al. Chuvash polycythe-
mia VHLR200W mutation is associated with down-regulation of hep-
cidin expression. Blood. 2011;118(19):5278-5282.
11. Ruschitzka FT, Wenger RH, Stallmach T, et al. Nitric oxide prevents
cardiovascular disease and determines survival in polyglobulic mice
overexpressing erythropoietin. Proc Natl Acad Sci USA. 2000;97(21):
11609-11613.
12. Díaz V, Gammella E, Recalcati S, et al. Liver iron modulates hepcidin
expression during chronically elevated erythropoiesis in mice. Hepa-
tology. 2013;58(6):2122-2132.
13. James C, Ugo V, Le Couédic JP, et al. A unique clonal JAK2 mutation
leading to constitutive signalling causes polycythaemia vera. Nature.
2005;434(7037):1144-1148.
14. Prchal JT, Crist WM, Goldwasser E, Perrine G, Prchal JF. Autosomal
dominant polycythemia. Blood. 1985;66(5):1208-1214.
15. Arber DA, Orazi A, Hasserjian R, et al. The 2016 revision to the World
Health Organization classification of myeloid neoplasms and acute
leukemia. Blood. 2016;127(20):2391-2405.
16. Ginzburg YZ, Feola M, Zimran E, Varkonyi J, Ganz T, Hoffman R. Dys-
regulated iron metabolism in polycythemia vera: etiology and conse-
quences. Leukemia. 2018;32(10):2105-2116.
17. de la Chapelle A, Träskelin AL, Juvonen E. Truncated erythropoietin
receptor causes dominantly inherited benign human erythrocytosis.
Proc Natl Acad Sci USA. 1993;90(10):4495-4499.
18. Sokol L, Luhovy M, Guan Y, Prchal JF, Semenza GL, Prchal JT. Primary
familial polycythemia: a frameshift mutation in the erythropoietin
receptor gene and increased sensitivity of erythroid progenitors to
erythropoietin. Blood. 1995;86(1):15-22.
19. Arcasoy MO, Harris KW, Forget BG. A human erythropoietin receptor
gene mutant causing familial erythrocytosis is associated with deregu-
lation of the rates of Jak2 and Stat5 inactivation. Exp Hematol. 1999;
27(1):63-74.
20. Huang LJ, Shen YM, Bulut GB. Advances in understanding the patho-
genesis of primary familial and congenital polycythaemia. Br J Haema-
tol. 2010;148(6):844-852.
21. Divoky V, Liu Z, Ryan TM, Prchal JF, Townes TM, Prchal JT. Mouse
model of congenital polycythemia: homologous replacement of
murine gene by mutant human erythropoietin receptor gene. Proc
Natl Acad Sci USA. 2001;98(3):986-991.
KRALOVA ET AL.1297

22. Divoky V, Song J, Horvathova M, et al. Delayed hemoglobin switching
and perinatal neocytolysis in mice with gain-of-function erythropoie-
tin receptor. J Mol Med (Berl). 2016;94(5):597-608.
23. Horvathova M, Kapralova K, Zidova Z, Dolezal D, Pospisilova D,
Divoky V. Erythropoietin-driven signaling ameliorates the survival
defect of DMT1-mutant erythroid progenitors and erythroblasts.
Haematologica. 2012;97(10):1480-1488.
24. Shammas FV, Engeset A. Glycogen content and PAS staining pattern
of human megakaryocytes. Scand J Haematol. 1986;37(3):237-242.
25. Socolovsky M, Nam H, Fleming MD, Haase VH, Brugnara C,
Lodish HF. Ineffective erythropoiesis in Stat5a/5b/mice due
to decreased survival of early erythroblasts. Blood. 2001;98(12):
3261-3273.
26. Bogdanova A, Mihov D, Lutz H, Saam B, Gassmann M, Vogel J.
Enhanced erythro-phagocytosis in polycythemic mice overexpressing
erythropoietin. Blood. 2007;110(2):762-769.
27. Garcia-Santos D, Hamdi A, Saxova Z, et al. Inhibition of heme oxyge-
nase ameliorates anemia and reduces iron overload in a β-thalassemia
mouse model. Blood. 2018;131(2):236-246.
28. Akel A, Wagner CA, Kovacikova J, et al. Enhanced suicidal death of
erythrocytes from gene-targeted mice lacking the cl-/HCO(3)()
exchanger AE1. M J Physiol Cell Physiol. 2007;292(5):C1759-C1767.
29. Zidova Z, Kapralova K, Koralkova P, et al. DMT1-mutant erythrocytes
have shortened life span, accelerated glycolysis and increased oxida-
tive stress. Cell Physiol Biochem. 2014;34(6):2221-2231.
30. Zadrazil J, Horak P, Horcicka V, Zahalkova J, Strebl P, Hruby M.
Endogenous erythropoietin levels and anemia in long-term renal
transplant recipients. Kidney Blood Press Res. 2007;30(2):108-116.
31. Lucarelli G, Howard D, Stohlman F Jr. Regulation of erythropoiesis.
XV. Neonatal erythropoiesis and the effect of nephrectomy. J Clin
Invest. 1964;43(11):2195-2203.
32. Xavier-Ferrucio J, Scanlon V, Li X, et al. Low iron promotes megakar-
yocytic commitment of megakaryocytic-erythroid progenitors in
humans and mice. Blood. 2019;134(18):1547-1557.
33. Watowich SS, Xie X, Klingmuller U, et al. Erythropoietin receptor
mutations associated with familial erythrocytosis cause hypersensitiv-
ity to erythropoietin in the heterozygous state. Blood. 1999;94(7):
2530-2532.
34. Pasquier F, Marty C, Balligand T, et al. New pathogenic mechanisms
induced by germline erythropoietin receptor mutations in primary
erythrocytosis. Haematologica. 2018;103(4):575-586.
35. Kapralova K, Lanikova L, Lorenzo F, et al. RUNX1 and NF-E2 upregu-
lation is not specific for MPNs, but is seen in polycythemic disorders
with augmented HIF signaling. Blood. 2014;123(3):391-394.
36. Montel-Hagen A, Blanc L, Boyer-Clavel M, et al. The Glut1 and Glut4
glucose transporters are differentially expressed during perinatal and
postnatal erythropoiesis. Blood. 2008;112(12):4729-4738.
37. Pietras EM. Inflammation: a key regulator of hematopoietic stem cell
fate in health and disease. Blood. 2017;130(15):1693-1698.
38. Ho YH, Del Toro R, Rivera-Torres J, et al. Remodeling of bone mar-
row hematopoietic stem cell niches promotes myeloid cell expansion
during premature or physiological aging. Cell Stem Cell. 2019;25(3):
407-418.e6.
39. Roy A, Wang G, Iskander D, et al. Transitions in lineage specification
and gene regulatory networks in hematopoietic stem/progenitor cells
over human development. Cell Rep. 2021;36(11):109698.
40. Zhao JL, Ma C, O'Connell RM, et al. Conversion of danger signals into
cytokine signals by hematopoietic stem and progenitor cells for regu-
lation of stress-induced hematopoiesis. Cell Stem Cell. 2014;14(4):
445-459.
41. Cho RH, Sieburg HB, Muller-Sieburg CE. A new mechanism for the
aging of hematopoietic stem cells: aging changes the clonal composi-
tion of the stem cell compartment but not individual stem cells. Blood.
2008;111(12):5553-5561.
42. Rekhtman N, Choe KS, Matushansky I, Murray S, Stopka T,
Skoultchi AI. PU.1 and pRB interact and cooperate to repress
GATA-1 and block erythroid differentiation. Mol Cell Biol. 2003;
23(21):7460-7474.
43. Ganz T. Macrophages and systemic iron homeostasis. J Innate Immun.
2012;4(5–6):446-453.
44. Lee SJ, Park SY, Jung MY, Bae SM, Kim IS. Mechanism for
phosphatidylserine-dependent erythrophagocytosis in mouse liver.
Blood. 2011;117(19):5215-5223.
45. Oldenborg PA, Zheleznyak A, Fang YF, Lagenaur CF, Gresham HD,
Lindberg FP. Role of CD47 as a marker of self on red blood cells. Sci-
ence. 2000;288(5473):2051-2054.
46. Mojzikova R, Koralkova P, Holub D, et al. Iron status in patients with
pyruvate kinase deficiency: neonatal hyperferritinaemia associated
with a novel frameshift deletion in the PKLR gene (p.Arg518fs), and
low hepcidin to ferritin ratios. Br J Haematol. 2014;165(4):556-563.
47. Sulovska L, Holub D, Zidova Z, et al. Characterization of iron metabo-
lism and erythropoiesis in erythrocyte membrane defects and thalas-
semia traits. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub.
2016;160(2):231-237.
48. Huang H, Constante M, Layoun A, Santos MM. Contribution of
STAT3 and SMAD4 pathways to the regulation of hepcidin by oppos-
ing stimuli. Blood. 2009;113(15):3593-3599.
49. Suresh S, Rajvanshi PK, Noguchi CT. The many facets of erythropoie-
tin physiologic and metabolic response. Front Physiol. 2020;10:1534.
50. Brines M, Grasso G, Fiordaliso F, et al. Erythropoietin mediates tissue
protection through an erythropoietin and common beta-subunit het-
eroreceptor. Proc Natl Acad Sci USA. 2004;101(41):14907-14912.
51. Broxmeyer HE. Erythropoietin: multiple targets, actions, and modify-
ing influences for biological and clinical consideration. J Exp Med.
2013;210(2):205-208.
52. Zhang YL, Radhakrishnan ML, Lu X, Gross AW, Tidor B, Lodish HF.
Symmetric signaling by an asymmetric 1 erythropoietin: 2 erythropoi-
etin receptor complex. Mol Cell. 2009;33(2):266-274.
53. Teng R, Gavrilova O, Suzuki N, et al. Disrupted erythropoietin signal-
ling promotes obesity and alters hypothalamus proopiomelanocortin
production. Nat Commun. 2011;2:520.
54. Valletta S, Thomas A, Meng Y, et al. Micro-environmental sensing by
bone marrow stroma identifies IL-6 and TGFβ1 as regulators of hema-
topoietic ageing. Nat Commun. 2020;11(1):4075.
55. Huang X, Pierce LJ, Chen GL, Chang KT, Spangrude GJ, Prchal JT.
Erythropoietin receptor signaling regulates both erythropoiesis and
megakaryopoiesis in vivo. Blood Cells Mol Dis. 2010;44(1):1-6.
56. Lee S, Wong H, Castiglione M, Murphy M, Kaushansky K, Zhan H.
JAK2V617F mutant megakaryocytes contribute to hematopoietic
aging in a murine model of myeloproliferative neoplasm. Stem Cells.
2022;40(4):359-370.
57. McCranor BJ, Kim MJ, Cruz NM, et al. Interleukin-6 directly impairs
the erythroid development of human TF-1 erythroleukemic cells.
Blood Cells Mol Dis. 2014;52(2–3):126-133.
58. Zermati Y, Fichelson S, Valensi F, et al. Transforming growth factor
inhibits erythropoiesis by blocking proliferation and accelerating dif-
ferentiation of erythroid progenitors. Exp Hematol. 2000;28(8):
885-894.
59. Johnson RA, Waddelow TA, Caro J, Oliff A, Roodman GD. Chronic
exposure to tumor necrosis factor in vivo preferentially inhibits eryth-
ropoiesis in nude mice. Blood. 1989;74(1):130-138.
60. Wang CQ, Udupa KB, Lipschitz DA. Interferon-gamma exerts its neg-
ative regulatory effect primarily on the earliest stages of murine ery-
throid progenitor cell development. J Cell Physiol. 1995;162(1):
134-138.
61. Forejtnikovà H, Vieillevoye M, Zermati Y, et al. Transferrin receptor
2 is a component of the erythropoietin receptor complex and is
required for efficient erythropoiesis. Blood. 2010;116(24):5357-5367.
1298 KRALOVA ET AL.

62. Watowich SS. The erythropoietin receptor: molecular structure and
hematopoietic signaling pathways. JInvestMed. 2011;59(7):1067-1072.
63. Sulahian R, Cleaver O, Huang LJ. Ligand-induced EpoR internalization
is mediated by JAK2 and p85 and is impaired by mutations responsi-
ble for primary familial and congenital polycythemia. Blood. 2009;
113(21):5287-5297.
64. Chen GL, Chang K, Huang X, Spangrude GJ, Prchal JT. Erythropoietin
signaling inhibits long term marrow reconstitution. Blood. 2006;
108(11):912a Abstract 3194.
65. Divoky V, Prchal JT. Mouse surviving solely on human erythropoietin
receptor (EpoR): model of human EpoR-linked disease. Blood. 2002;
99(10):3873-3874.
66. Chang K, Pastore YD, Nussenzveig RH, Prchal JT. Erythropoietin sig-
naling inhibits the short-term marrow reconstitution. Blood. 2005;
106(11):848a Abstract 3028.
67. Orford KW, Scadden DT. Deconstructing stem cell self-renewal:
genetic insights into cell-cycle regulation. Nat Rev Genet. 2008;9:
115-128.
68. Haas S, Trumpp A, Milsom MD. Causes and consequences of hemato-
poietic stem cell heterogeneity. Cell Stem Cell. 2018;22(5):627-638.
69. Zhang H, Wang S, Liu D, et al. EpoR-tdTomato-Cre mice enable iden-
tification of EpoR expression in subsets of tissue macrophages and
hematopoietic cells. Blood. 2021;138(20):1986-1997.
70. Wang W, Liu W, Fidler T, et al. Macrophage inflammation, Erythro-
phagocytosis, and accelerated atherosclerosis in Jak2
V617F
mice. Circ
Res. 2018;123(11):e35-e47.
71. Grisouard J, Li S, Kubovcakova L, et al. JAK2 exon 12 mutant mice
display isolated erythrocytosis and changes in iron metabolism favor-
ing increased erythropoiesis. Blood. 2016;128(6):839-851.
72. Wilkinson N, Pantopoulos K. IRP1 regulates erythropoiesis and sys-
temic iron homeostasis by controlling HIF2αmRNA translation. Blood.
2013;122(9):1658-1668.
73. Singh RP, Grinenko T, Ramasz B, et al. Hematopoietic stem cells but
not multipotent progenitors drive erythropoiesis during chronic ery-
throid stress in EPO transgenic mice. Stem Cell Reports. 2018;10(6):
1908-1919.
74. Noguchi CT, Wang L, Rogers HM, Teng R, Jia Y. Survival and prolifer-
ative roles of erythropoietin beyond the erythroid lineage. Expert Rev
Mol Med. 2008;10:e36.
75. Li W, Wang Y, Zhao H, et al. Identification and transcriptome analysis
of erythroblastic Island macrophages. Blood. 2019;134(5):480-491.
76. Kota J, Caceres N, Constantinescu SN. Aberrant signal transduction
pathways in myeloproliferative neoplasms. Leukemia. 2008;22(10):
1828-1840.
77. Spivak JL. Polycythemia Vera. Curr Treat Options Oncol. 2018;
19(2):12.
78. Gangaraju R, Song J, Kim SJ, et al. Thrombotic, inflammatory, and
HIF-regulated genes and thrombosis risk in polycythemia vera and
essential thrombocythemia. Blood Adv. 2020;4(6):1115-1130.
79. Rives S, Pahl HL, Florensa L. Molecular genetic analyses in familial
and sporadic congenital primary erythrocytosis. Haematologica. 2007;
92(5):674-677.
80. Al-Sheikh M, Mazurier E, Gardie B. A study of 36 unrelated cases
with pure erythrocytosis revealed three new mutations in the eryth-
ropoietin receptor gene. Haematologica. 2008;93(7):1072-1075.
81. Rabadan Moraes G, Pasquier F, Marzac C, et al. An inherited gain-of-
function risk allele in EPOR predisposes to familial JAK2V617F mye-
loproliferative neoplasms. Br J Haematol. 2022;198(1):131-136.
SUPPORTING INFORMATION
Additional supporting information can be found online in the Support-
ing Information section at the end of this article.
How to cite this article: Kralova B, Sochorcova L, Song J, et al.
Developmental changes in iron metabolism and erythropoiesis
in mice with human gain-of-function erythropoietin receptor.
Am J Hematol. 2022;97(10):1286‐1299. doi:10.1002/ajh.
26658
KRALOVA ET AL.1299