Vol. 284, Issue 4, R1126-R1137, April 2003
The embryo makes red blood cell progenitors in every tissue
simultaneously with blood vessel morphogenesis
Maria Luisa S.
Sequeira Lopez1,
Daniel R.
Cherñavvsky1,
Takayo
Nomasa1,
Lee
Wall2,
Masashi
Yanagisawa3, and
R. Ariel
Gomez1
1 Department of Pediatrics, University of Virginia,
Charlottesville, Virginia 22908; 2 Department of Medicine,
Université de Montréal, Quebec H2L 4M1, Canada; and
3 Howard Hughes Medical Institute, Department of Molecular
Genetics, University of Texas Southwestern Medical Center, Dallas,
Texas 75390-9050
 |
ABSTRACT |
During embryonic life,
hematopoiesis occurs first in the yolk sac, followed by the
aorto-gonado-mesonephric region, the fetal liver, and the bone marrow.
The possibility of hematopoiesis in other embryonic sites has been
suspected for a long time. With the use of different methodologies
(transgenic mice, electron microscopy, laser capture microdissection,
organ culture, and cross-transplant experiments), we show that multiple
regions within the embryo are capable of forming blood before and
during organogenesis. This widespread phenomenon occurs by
hemo-vasculogenesis, the formation of blood vessels accompanied by the
simultaneous generation of red blood cells. Erythroblasts develop
within aggregates of endothelial cell precursors. When the lumen forms,
the erythroblasts "bud" from endothelial cells into the forming
vessel. The extensive hematopoietic capacity found in the embryo helps
explain why, under pathological circumstances such as severe anemia,
extramedullary hematopoiesis can occur in any adult tissue.
Understanding the intrinsic ability of tissues to manufacture their own
blood cells and vessels has the potential to advance the fields of
organogenesis, regeneration, and tissue engineering.
development; hematopoiesis; homeostasis; kidney
| |
INTRODUCTION |
IN THE VERTEBRATE
embryo, hematopoietic and vascular endothelial cells are the first
cells to differentiate in response to induction of the mesoderm
(2). In mice, endothelial cells and hematopoietic cells
are first observed (when primitive hematopoiesis begins) in the yolk
sac at 7.5 days post coitum (9). Primitive hematopoiesis
is restricted to the formation of nucleated erythrocytes that express
embryonic hemoglobin (43) and macrophages
(7).
Definitive hematopoiesis is the process whereby all types of blood
cells are formed followed by their differentiation, including the
enucleation of erythrocytes. The initial site of definitive hematopoiesis is controversial. It has been proposed that the first
site is in the splanchnopleural/aorto-gonado-mesonephric (AGM) area (10, 14, 24) and that early in
development endothelial cells from the AGM region may themselves give
rise to hematopoietic cells (6, 20, 37). As embryonic
development continues, the site for hematopoiesis changes to the fetal
liver and as the animal matures to the bone marrow, where hematopoiesis
persists in the adult. The possibility of hematopoiesis occurring in
other embryonic sites has not been explored. During development of the yolk sac, hematopoiesis is intimately linked to the development of
blood vessels. In fact, the ontogenic relationship between hematopoietic cells and the endothelial cells of the blood islands of
the yolk sac has been observed almost 100 years ago (34). A similar observation has also been made in the aorta and post umbilical and vitelline arteries where hematopoietic clusters seem to
originate from hemogenic endothelium (6, 18, 20, 37). The
existence of a common precursor for endothelial cells and hematopoietic
cells, the hemangioblast, has been suspected for a long time (25,
30, 31, 34). Taken altogether, these observations suggest that
vasculogenesis and hematopoiesis are part of the same process, in which
formation of a blood vessel is accompanied by the simultaneous in situ
production of blood cells within that vessel (hemo-vasculogenesis).
Our experiments show that 1) hematopoiesis is widespread in
the embryo: in addition to the AGM area, fetal liver, and bone marrow,
the embryo makes erythroblasts in every tissue examined before and
during organogenesis and 2) hematopoiesis is tied to and
integrated with the differentiation and morphogenesis of blood vessels:
formation of a vessel is accompanied by the generation of red cell
precursors, suggesting that hematopoietic and vascular cells share a
common precursor.
| |
MATERIALS AND METHODS |
Animals.
Several mouse strains were studied: expressing Lac-Z driven by the
Tie2 promoter (marks endothelial cells) (35) or
under the control of the 
-globin LCR-promoter combination
(marks erythroid cells) (16), expressing cre recombinase
driven by the Tie2 promoter (22), ubiquitously
expressing Lac-Z (ROSA 26), the ROSA26 Lox P reporter mice
(38), and wild-type C57B6. Embryos at 11.5 days of
gestation (E11.5) were the source of fetal kidneys used for in vitro culture and for grafting under the kidney capsule. Time-dated pregnant mice were mated overnight and the females were checked for
vaginal plugs the following morning. The day of detection of a vaginal
plug was regarded as day 0.5 of gestation. Grafting of
embryonic kidneys under the kidney capsule was performed as previously
described (36). All procedures were performed in accordance with the Guiding Principles for Research Involving Animals and Human Beings by the American Physiological Society (1) and were approved by the University of Virginia Animal Care Committee.
Immunohistochemistry and X-gal reaction.
Embryonic tissues from mice were subjected to the X-gal reaction,
paraffin embedded, sectioned (5 µm), and immunostained as previously
described (36). The antibodies used were anti-
-smooth muscle actin (-SMA) (Sigma) and anti-hemoglobin (DAKO).
Electron microscopy studies.
E9.5 and E11.5 mouse embryos were dissected and
fixed with 4% PFA and 2.5% glutaraldehyde at 4°C overnight, then
postfixed in OsO4, embedded in epoxy resin by conventional
methods, cut into 70- to 80-nm sections, and examined using a JEOL 100 CX transmission electron microscope.
Metanephric kidney culture.
E11.5 kidneys from
-globin/LacZ + embryos (n = 24) were cultured for 72 h at 37°C as previously described
(29). The defined serum-free medium (DMEM-F12) was
supplemented with 10 mM HEPES, 1.1 mg/ml NaHCO3, 5 µg/ml
insulin and transferrin, 2.8 nM selenite (Sigma), 25 ng/ml
PGE1 (Sigma), 32 pg/ml triiodothyronine (Sigma), 50 U/ml
penicillin G, and 50 U/ml mycostatin. The medium with additives was
changed daily.
Metanephric kidney culture on top of embryonic stem cells.
Embryonic stem (ES) cells from 129 SvEv mice were grown on top of
irradiated feeder layers of primary murine embryonic fibroblasts placed
on top of cell culture inserts with 3-µm pore size (Falcon) on a
six-well dish. Forty-eight hours later, E12 kidneys
(n = 17) from ROSA26 mice were dissected as previously
described (36) and placed on top of the ES cell colonies
for 5 days. Medium containing DMEM-H's (GIBCO) supplemented with 15%
FBS (GIBCO), 1% sodium pyruvate (GIBCO), 0.0008% -mercaptoethanol
(Sigma), 5 µg/ml insulin and transferrin, 2.8 nM selenite (Sigma), 25 ng/ml PGE1 (Sigma), 32 pg/ml triiodothyronine (Sigma), and
15 mU/ml erythropoietin (Sigma) was changed daily. Embryonic kidneys
were fixed and subjected to the X-Gal reaction as previously described
(36).
Peripheral blood extraction.
Mouse embryos at E12.5 and E14.5 were dissected
from the uterus, membranes were opened, and the umbilical cord was cut
to let blood flow out from the cord. Blood samples were then collected using a P10 pipette. From anesthetized adult animals, blood was collected from the abdominal aorta using a 1-ml syringe and a 23-gauge needle.
Laser capture microdissection.
Mouse embryos (E11.5) were harvested and fixed in 70%
ethanol overnight. After dehydration with different alcohol gradients and xylenes, tissues were embedded in paraffin and cut into 5-µm consecutive sections. Sections were deparaffinized and stained with
hematoxylin and eosin as follows: 70% ethanol for 30 s,
hematoxylin for 30 s, deionized H2O for 30 s,
bluing reagent (Richard-Allan Scientific, Kalamazoo, MI) for 30 s,
70% ethanol for 1 min, 95% ethanol for 1 min, eosin for 30 s,
95% ethanol for 1 min two times, 100% ethanol for 1 min, xylenes for
5 min two times, and then air-drying for 20 min. Subsequently, the
tissues were microdissected using a Pix Cell II laser capture
microscope with an infrared diode laser (Arcturus engineering, Santa
Clara, CA) as described (3, 11). Briefly, the dehydrated
section was overlaid with a cap that contains a thermoplastic membrane,
and then focal melting of the membrane through laser activation
captured the cells. After visual control of the completeness of
dissection, the cap was removed and the captured cells were immersed in
50 µl of extraction buffer (PicoPure RNA isolation Kit, Arcturus,
Mountain View, CA).
RNA extraction, DNase treatment, and reverse transcription.
With the use of laser capture microdissection (LCM), three different
kinds of samples were extracted: 1) immature blood cells (vessel contents), 2) endothelial cells (vessel walls), and
3) "budding" cells (cells that were clearly budding from
the vessels into the lumen). Samples of whole blood (E12.5,
E14.5, and adult mouse) were also collected. RNA from these
cells was obtained (PicoPure RNA isolation kit, Arcturus) following the
manufacturer's protocol (5). The RNA pellets were
dissolved in water and then treated with RNase-free DNase (Ambion,
Austin, TX). RNA reverse transcription was done using 5 µl of RNA, 1 µg oligo(dt) primer, 25 µmol/l dNTPs, and 200 U M-MLV reverse
transcriptase (Promega, Madison, WI) in 14 µl total volume. A mock
reaction without addition of reverse transcriptase was also performed
for each sample. Erythroid, endothelial, and smooth muscle markers were
detected by PCR. -SMA mRNA was tested in peripheral blood samples
from E12.5, E14.5, and adult mice and in LCM
samples of the three groups of cells (immature blood cells, endothelial
cells, and "budding" cells). -Globin mRNA was tested in LCM
samples of the three groups of cells mentioned above. Tie2 mRNA was
tested in peripheral blood samples from E12.5,
E14.5, and in LCM samples of blood cells. Primers and PCR
conditions for mouse -SMA were 5'-TAT GTA GCT CTG GAC TTT GAA-3' and
5'-CAG AGC AGG GGG GAC TTA GAA-3'. Annealing temperature was 55°C,
and the number of cycles was 40 (Eppendorf Mastercycler, Westbury, NY).
Expected product of the cDNA amplification was 492 bp.
-Globin primers were 5'-CAC AAC CCC AGA AAC AGA CA-3' and 5'-CTG ACA
GAT GCT CTC TTG GG-3'. Conditions were annealing temperature 55°C,
and the number of cycles was 40. The expected product size was 525 bp.
For Tie2, we performed nested RT-PCR. Outer primers were 5'-TGT CAA TCA
GGC CTG GAA ATA C-3' and 5'-GAG GAG GGA GAA TGT CAC TAA GG-3'.
Conditions were annealing temperature 56.6°C, and the number of
cycles was 40. The expected product size was 464 bp. Inner primers were
5'-TAC TTG GAG CCG CGG ACT GAC-3' and 5'-CGC CTT GGT GTT GAC TCT
AGC-3'. Conditions were annealing temperature 56.8°C, and the number
of cycles was 40. The expected product size was 351 bp. The
amplification was performed using Taq DNA polymerase 0.75 U (Promega),
2 µl for the RT, and 25 µl total volume reaction. Negative controls
included reaction without reverse transcriptase and without RNA. The
PCR product was separated on 1% agarose gel stained with ethidium bromide.
| |
RESULTS AND DISCUSSION |
Erythropoiesis throughout the embryo.
To study the lineage relationship between endothelial and erythroid
cells and to define whether the latter cells and blood vessels develop
simultaneously from a common precursor as a widespread phenomenon
occurring throughout the embryo, we examined mouse embryos expressing
LacZ under the control of the Tie2 promoter (marks endothelial cells) (35) or under the control of the
-globin LCR-promoter combination (marks erythroid cells)
(16). In addition, to define whether endothelial
cell precursors give origin to blood cells, we examined embryos derived
from the cross of Tie2-Cre (22) with the ROSA26
Lox P reporter mice (38). In mice derived from this cross,
endothelial cells and their progeny express LacZ because
once Cre-mediated recombination occurs, it activates
LacZ expression and the recombined LacZ transgene
is inherited through the cellular lineage. Thus, any cell that
expresses the Tie2-Cre transgene and its descendants,
whether they are still expressing the transgene or not, will be blue
with the X-gal reaction.
At E8.5, there is already widespread expression of
-globin in erythroblasts and endothelial cells: it is not
limited to the aorta and major vessels. In fact, -globin
expression is encountered in multiple tissues throughout the embryo
(Fig. 1, A-E)
including, but not limited to, the areas surrounding the optic pit of
the developing head (Fig. 1B), the neural tube (Fig. 1,
C and E), and in between the developing somites
(Fig. 1D). Figure 1E shows -globin/LacZ-positive erythroid cells
"budding" from endothelial cells, a phenomenon that still persists
at E11.5 while the vessels are still forming (Fig. 1,
F-H). Endothelial cells of "budding" areas share -globin expression with erythroblasts.
Confirming those findings, similar results are obtained in mice
harboring a different promoter-reporter transgene. In
Tie2/LacZ mouse embryos, as shown at
E10.5 (Fig. 2,
A-F and H-J),
expression of LacZ is observed both in endothelial cells and
hematopoietic cells and in their mesenchymal precursors throughout the
embryo. Figure 2, B, C, and F, shows
that mesenchymal derivatives surrounding the developing brain contain
LacZ-expressing cells in endothelium and blood precursors.
This tight relationship between endothelial cells and blood precursors
is also evident during organogenesis in multiple organs of the
developing embryo. For example, Fig. 2G shows this
relationship in the developing kidney at E16.5.
View larger version (132K):
[in this window]
[in a new window]
|
Fig. 1.
Distribution of -globin expression as assessed by the
X-Gal reaction in -globin/LacZ embryos at
embryonic day 8.5 (E8.5; A to
E) and at E11.5 (F to H).
LacZ expression in the whole embryo (A),
surrounding the optic pit (op; B), surrounding the neural
tube (nt; C), in between developing somites (s) (arrows)
(D), and lining the dorsal aorta (arrowheads). E:
LacZ expression is evident in both erythroid cells ("budding" from)
and in endothelial cells (arrowhead), adjacent to the neuroepithelium
(n). F: LacZ expression in "budding" cells (arrows) and
endothelial cells (arrowheads) in the cephalic mesenchyme, in the
developing plexus coroideus (G), and in a marginal vein in
the left hindlimb bud (H). Scale bars: 200 µm
(A); 100 µm (C); 50 µm (B,
D, and G); 25 µm (E, F,
and H).
|
|
View larger version (113K):
[in this window]
[in a new window]
|
Fig. 2.
Tie2 distribution, as assessed by the X-Gal reaction in
Tie2/LacZ embryos at E10.5
(A-F and H-J) and
in the developing kidney at E16.5 (G).
H-J: smooth muscle cell identification by
specific immunostaining with antibody against -smooth muscle actin
(SMA) (36). LacZ staining in the whole embryo
(A), higher magnification of (A) showing abundant
Tie2 expression in cephalic mesenchyme (B),
cephalic mesenchyme adjacent to the neuroepithelium (n) (C),
dorsal mesenteric vessel showing cells budding from the endothelium
(arrows) (D), concomitant expression of Tie2 by
endothelial and circulating blood cells in a primary head vein
(E), Tie2-expressing cells budding from
endothelial cells in the cephalic mesenchyme (arrows) (F).
G: developing capillaries adjacent to developing glomeruli
show blood cells "budding" from endothelial cells (arrows).
Insets, top, middle, and
bottom: higher-magnification areas depicted by the arrows.
H: coexistence of -SMA expression and Tie2
(LacZ) in circulating blood cells of a peripheral blood
vessel. I: circulating blood cells expressing -SMA in the
lumen of the aorta (a) and in a developing intersegmental artery
adjacent to the dermomyotome (d) component of somites. J:
higher magnification of I at the level of the intersegmental
artery; the arrow indicates a blood cell expressing -SMA. Scale
bars: 200 µm (A); 100 µm (B and
I); 50 µm (C-H and
J).
|
|
To define the lineage relationship between endothelial and erythroid
cells, we took advantage of the Cre-Lox system. Tie2-Cre transgenic mice crossed to ROSA26 LoxP reporter mice generated embryos expressing LacZ, after Cre-mediated recombination,
in endothelial cells and their descendants. As shown in Fig.
3, the expression pattern was similar to
that seen in the Tie2/LacZ and -globin/LacZ transgenic mice, supporting the
concept that erythroid cells share a common precursor(s) with
endothelial cells. Because occasionally some mammalian cells exhibit
endogenous -galactosidase activity during development or
apoptosis, we performed the X-gal reaction in all littermates
(Tie2-Cre alone, ROSA26 LoxP alone, and wild type) resulting
from the aforementioned cross (Tie2-Cre × ROSA26 LoxP)
at different embryonic and postnatal ages (E8.5, E9.5, E10.5, E11.5, E13.5,
E16.5, newborns 2 and 5 days old) and none of them showed
-galactosidase staining, demonstrating that the staining was
specific and due to Cre-mediated recombination in the
appropriate cell types.
View larger version (148K):
[in this window]
[in a new window]
|
Fig. 3.
LacZ distribution in Tie2-Cre/R26R
double transgenic embryos at E9.5
(A-C) and E10.5 (D).
A: LacZ expression is observed in the endothelial
cells of the forming vessels, in the cells that are budding (arrows)
from them, and in the blood cells within the vessels.
B-D: cells at different stages of budding
from the endothelium. Scale bars: 100 µm (A); 50 µm
(B-D).
|
|
In separate experiments, to corroborate whether LacZ expression
corresponded to mRNA expression, we obtained endothelial cells, budding
cells, and blood cells within vessels by the LCM technique (3,
11) and determined the expression of erythroid and endothelial markers by PCR. Endothelial cells from LCM samples were positive for
-globin, and blood cells from both LCM and E12.5 and
E14.5 peripheral blood samples were positive for Tie2 (Fig.
4). These results are in agreement with
those obtained using the transgenic animals described above.
View larger version (42K):
[in this window]
[in a new window]
|
Fig. 4.
RT-PCR gene expression analysis from laser capture microdissection
(LCM) (E11.5) and embryonic peripheral blood samples
(E12.5 and E14.5). LCM samples were obtained from
developing vessels, from which 3 different cells were picked: blood
cells present in the lumen (BC), endothelial cells (EC), and budding
cells (Bd). -SMA, -globin, and Tie2 transcripts were detected in
LCM and blood samples. Peripheral blood was obtained from blood freely
flowing from the umbilical cord. Reactions were performed in the
presence (+) or absence ( ) of reverse transcriptase.
|
|
The overlap of erythroid and endothelial marker expression suggests a
common origin between endothelial and erythroid cells. Altogether, the
experiments suggest that vasculogenesis and hematopoiesis are
intimately linked and occur concomitantly during vessel formation: as a
new vessel is formed, erythroblasts also form. Kisanuki et al.
(22) also observed circulating LacZ-positive
cells in embryos derived from crossing Tie2-Cre mice with
reporter mice as described above. Although it was hypothesized that
these cells could be circulating endothelial progenitors, the present
results suggest that they are more likely to be red cell precursors. To
clarify this issue, we immunostained Tie2/LacZ
embryos for hemoglobin and observed that the blue staining cells within
the forming vessels were also positive for hemoglobin (Fig. 5). We,
therefore, conclude that those circulating cells (derived from a common
precursor together with endothelial cells) belong to the erythroid
lineage. It also seemed possible that all cell types from a forming
vessel, including smooth muscle cells of arterioles, as well as
endothelial and erythroid cells, derive from the same precursor(s). To
address this issue, we performed immunostaining for -SMA, a marker
for smooth muscle cells of the vessel wall (4, 40). Figure
2, H-J, shows expression of -SMA protein
in blood cells within developing vessels throughout the
Tie2/LacZ embryos at E10.5. Confirming these results, -SMA mRNA expression was detected by PCR in blood from E12.5, E14.5 embryos, and adult mice (Fig.
4). -SMA mRNA expression was also present in samples of blood,
endothelium, and "budding" cells obtained by LCM (Fig. 4).
Recently, Yamashita et al. (45) demonstrated in vitro that
cells expressing -SMA, as well as hematopoietic and endothelial
cells, could arise from a population of cells originally derived from
ES cells that express the vascular endothelial growth factor receptor
2, Flk1. Our results suggest that this phenomenon also occurs in vivo
in the embryo, and together with the results of Yamashita et al.
(45) the findings indicate that all three cell types,
vascular smooth muscle, endothelial and erythroid cells, may arise from
a common precursor. These results support the notion for a common
ancestor for these cell types. Whether this progenitor is the suspected
hemangioblast (25, 30, 31, 34) remains to be determined.
View larger version (164K):
[in this window]
[in a new window]
|
Fig. 5.
Hemoglobin expression in Tie2/LacZ embryo at
E10.5 (A and B) and in
Tie2-Cre/R26R double transgenic embryo at E16.5
(C). A and B: arrows show that blood
cells and budding cell immunostained for hemoglobin (dark brown) also
express Tie2/LacZ (blue). C: concomitant
expression of LacZ and hemoglobin is observed in blood and
endothelial cells in embryonic tissue of skeletal muscle. Bars: 25 µm.
|
|
A summary of evidence supporting the relationship between endothelial
cells and erythroid cells is depicted in Table
1.
View this table:
[in this window]
[in a new window]
|
Table 1.
Summary of findings: evidence for relationship between endothelial
cells and erythroblasts during embryonic development
|
|
Hematopoiesis by budding.
Clusters of hematopoietic cells have previously been observed within
the aorta, vitelline, and umbilical arteries (13, 20, 27, 32,
44). Budding cells from endothelial cells in the floor of the
aorta have been described (13, 20, 32) and the concept of
a hemogenic endothelium has been proposed based on in vivo and in vitro
studies that demonstrate that hematopoietic cells can be generated from
endothelial cells (6, 18, 20, 26, 37).
Our results show that budding of hematopoietic cells from endothelial
cells is not restricted to a few areas and is observed throughout the
mouse embryo (Figs. 3 and 6). Also, the
results support the hypothesis of a common origin for both cell types. Hematopoietic cells are formed within and concomitantly with the development of new blood vessels. As a new blood vessel is formed, erythroid cells are seen "budding" from primitive endothelial cells
lining the developing endothelial tubes. Mice expressing LacZ under the control of -globin regulatory sequences
showed that endothelial cells and erythroid cells shared expression of -globin. Moreover, expression of -globin is limited to the areas where erythroid cells are actively developing from endothelial cells
and is not seen in adult or developed vessels. Figure 3, B
and C, describes different stages of the budding process
found throughout the embryo, and Fig. 6,
A-C, shows, at high magnification, the tight
relationship between an endothelial cell and an erythroblast during the
budding process, indicating that it is not nonspecific sticking of
erythroid cells or circulating endothelial cells. Clearly, the presence
of "budding" hematopoietic cells is a common occurrence throughout
the embryo and is encountered in a variety of tissue types, including
kidney, brain, skin, and others (see Figs. 1-3 and 6). The
"budding" process seen here in small developing vessels mimics the
budding of blood cells from endothelial cells that occurs in the aorta
as seen in Fig. 6, H and I. Furthermore, electron
microscopy studies of developing vessels within the head of mice
embryos (E9.5 and E11.5) allowed us to show the
intimate relationship between erythroblasts and endothelial cells
during vessel formation and to reconstruct the main stages of
hemo-vasculogenesis (Fig. 7).
View larger version (87K):
[in this window]
[in a new window]
|
Fig. 6.
Budding of hematopoietic cells from endothelial cells is observed
in multiple sites throughout the embryo in mice expressing
LacZ under -globin or Tie2 control.
A-C: consecutive sections of E11.5
-globin/LacZ embryo showing the close relationship
between endothelial and erythroid cells adjacent to the facio-acoustic
preganglion complex. A blue erythroid cell (arrow) is budding from an
endothelial cell (arrowhead). The budding process in cells expressing
-globin is seen in the cephalic mesenchyme adjacent to
the neuroepithelium (D) and also adjacent to the
neuroepithelium at E8.5 (E). Mice expressing
LacZ under Tie2 control show the budding process
in dorsal posterior root ganglion (F) in the lower part of
the embryo (G) adjacent to the neuroepithelium and in the
aorta (H and I).
|
|
View larger version (134K):
[in this window]
[in a new window]
|
Fig. 7.
Stages of hemo-vasculogenesis. Electron microscopy of cephalic
mesenchyme adjacent to the neuroepithelium (ne) of E9.5
mouse embryos. A: erythroblast (e) surrounded by mesenchymal
cells (endothelial precursors). B: lumen starts to form but
the e is still attached to the endothelial precursor. C: EC
are lining the forming capillary; * indicates the area where the e and
EC show communication. D: higher magnification of the area
denoted by * in C showing "pore-like" communications
(arrowheads). E: e detaching from EC; ** indicates the
detaching region. F: higher magnification of the area
denoted by ** in E showing the detaching region in more
detail. Original magnification: A, ×27,000; B,
×17,820; C and E, ×8,910; D and
F, ×54,000.
|
|
Hemo-vasculogenesis starts as an aggregate of cells (Fig.
7A). Eventually, the cells
undergo further differentiation into erythroblasts and vascular
cells (Fig. 7B). Then, a cavity(s) is formed followed by
formation of a lumen and the separation of erythroblasts from
endothelial cells (Fig. 7, B-F). The fact that the process starts as a cell aggregate containing both cell types
(erythroblasts and endothelial cells) before lumen formation has occurred suggests that both cell types develop in situ, a process
that we described as hemo-vasculogenesis. The mechanisms governing
hemo-vasculogenesis are unclear. The Cbfa2 gene (also known as
AML1, Runx1, and PEBPA2A) is required for the formation of intra-aortic
hematopoietic clusters and cystic detachment of cells into the lumen
during embryonic life (27, 28). Those findings resemble
the results presented here for hemo-vasculogenesis in the developing
head. It remains to be determined whether this gene or related ones are
involved in the process of hemo-vasculogenesis in other tissues besides
the aorta.
Kidney as a model for hematopoiesis during organogenesis.
To address the question whether hematopoiesis occurs during
organogenesis, we chose the kidney as a model to perform further experiments. We demonstrate herein that the embryonic kidney is capable
of producing its own blood cells, particularly cells of the erythroid
lineage. Although it has been suggested that the AGM area is a site for
blood formation in the early mouse embryo (10, 14, 20,
24), the formation of blood by the definitive (metanephric)
kidney during the period of active organogenesis was previously
unknown. The studies presented below clearly show that blood
precursors, presumably erythroblasts, are present in the
undifferentiated metanephric mesenchyma before nephrogenesis has ensued
(Fig. 8, A and C).
When these embryonic kidneys are grown in vitro, nephrogenesis takes
place. However, under the usual culture conditions, glomerular
development is avascular. To study the capacity of the embryonic kidney
to generate its own blood and taking advantage of the mouse model that
expresses LacZ under the control of the -globin promoter
(16), E11.5 kidneys from
-globin/LacZ + embryos were cultured for
72 h. As shown in Fig. 8, A and C, at
E11.5, when the embryonic kidney is still avascular, the
metanephric blastema possesses only few and scattered individual
LacZ + cells, mostly surrounding the developing
ureteric bud. After 72 h in culture, there was an increase in the
number of erythroid cells (Fig. 8, B and D) and
the distribution of the LacZ + cells resembled the
normal pattern encountered in the E14.5 kidneys from animals
that develop in utero (Fig. 8E). These results indicate that
erythropoietic precursors within the metanephric mesenchyma continue to
proliferate under culture conditions.
View larger version (91K):
[in this window]
[in a new window]
|
Fig. 8.
Hemo-vasculogenesis in embryonic kidneys grown in vitro
(A-D), on top of embryonic stem (ES) cells
(F-H), and under the kidney capsule
(I and J). In vitro culture of E11.5
kidneys from -globin/LacZ embryos. A:
E11.5 whole kidney with scattered cells expressing
-globin. B: after 72 h in culture, there
is an increase in the number of -globin expressing cells.
C: section of A with scattered blue cells
surrounding the developing ureter (arrows); D: section from
B shows an increase in the number of
-globin-positive cells. For comparison, E
shows the normal in vivo increase in -globin expressing
cells at E14.5. Hemo-vasculogenesis in ROSA26 embryonic
kidneys either grown (F-H) on top of ES
cells or transplanted under the kidney capsule of an adult mouse
(I and J). E11.5 ROSA26 embryonic
kidneys were cultured for 5 days on top of wild-type ES cells.
F: embryonic kidneys were invaded by ES cells (pink) that
underwent branching morphogenesis and developed tubular structures.
G and H: however, the metanephric mesenchyme
derived from the embryonic kidney underwent hemo-vasculogenesis
(arrows). H: enlarged view of a focus of hemo-vasculogenesis
in the embryonic kidney. The budding process also observed here is
indistinguishable from the one observed in vivo. X-gal reaction and
counterstaining with nuclear fast red (36) show that the
cells derived from the ROSA26 kidneys have blue cytoplasm and pink
nuclei, whereas the ones derived from the wild-type ES cells have pink
nuclei and cytoplasm. I and J: ROSA26 embryonic
kidneys transplanted under the kidney capsule of a wild-type mouse
differentiated, forming blue nephrons, blood vessels, and blood cells
(arrows) demonstrating that these structures originated from the
embryonic kidney. J: blood cell within the developing
arteriole expresses -SMA (arrow), further suggesting a common origin
of vascular and blood cells. Smooth muscle cells are identified by
specific immunostaining (brown color) with antibody against -SMA
(4). Bars: 100 µm (F); 50 µm (D,
E, G, I, and
J).
|
|
To strengthen the hypothesis that the kidney produces its own blood, we
cultured embryonic kidneys from ROSA 26 mice (in which all cells are
blue with the X-gal reaction) on top of wild-type ES cells for 5 days.
This in vitro system showed that whereas ureteric branches developed
from ES cells (Fig. 8F), -gal + blood islands
originated from the ROSA 26 embryonic kidney (Fig. 8, G and
H) demonstrating the embryonic kidney's intrinsic capacity to form its own blood. Furthermore, in these culture conditions, blood
cells also originate by "budding" (Fig. 8H) as it occurs in the intact embryo, suggesting that this is a built-in process programmed to occur even under in vitro conditions. Similarly, in vivo
experiments, in which E11.5 Rosa 26 embryonic kidneys (all
cells are blue) were transplanted under the kidney capsule of adult
wild-type mice kidneys for 7 days, revealed that all vascular elements,
including blood cells, stained blue after the X-gal reaction (Fig. 8,
I and J), suggesting the intrinsic metanephric origin of blood and vascular cells. Although we cannot completely exclude the possibility that errand cells coming from the AGM region
may have seeded the embryonic kidney, the possibility seems remote
considering that hemo-vasculogenesis occurs initially as an aggregate
of cells that later differentiate in situ into erythroblast and
endothelial cells. If the origin of the blood cells were extrarenal, it
would not invalidate the fact that they undergo further differentiation plus cell division in situ. Our data suggest, however, that the process
occurs in situ, not from migrating cells but very likely from
mesenchymal precursors residing in the embryonic organ. The lineage
relationship between "hemo-vasculogenic" cells and other cell types
in the kidney remains to be studied. The ability of the kidney to make
its own blood has also been observed in early nonmammalian species,
such as adult frogs, suggesting that this mechanism is phylogenetically
conserved. The presumptive mesonephric anlagen in nonmammalian
vertebrates generates hematopoietic precursors and the promesonephros
serves as a hematopoietic site (42). Interestingly, as we
showed for the whole embryo, these cross-transplantation experiments
also revealed the expression of -SMA protein by blood cells within
the transplanted embryonic kidney (Fig. 8J) reflecting the
fact that blood and blood vessel development have a common origin and
occur concomitantly.
It has been well recognized that during embryonic life, hematopoiesis
occurs first in the yolk sac followed by the AGM region, the fetal
liver, and then the bone marrow. In the present study, we used
different approaches and techniques and showed that multiple embryonic
tissues have the capacity to produce blood cells. The present findings
provide strong evidence for hemo-vasculogenesis, the simultaneous,
local generation of blood and vessels likely derived from related
precursor cell(s). The results do not invalidate the possibility that
cells originating from distant sites (i.e., AGM region) could coexist
in situ or in the circulation with blood cells that developed locally
simultaneously with vessel morphogenesis. However, it should be noted
that in many organs, we observed erythroblasts before vascularization
to that organ had occurred, implying that distant cells could not
access the tissue via the circulation. It is still possible that
seeding from distant sites may occur through other mechanisms such as
cell migration. Future work will be needed to ascertain this possibility.
The fact that hemo-vasculogenesis is widespread throughout the
whole embryo during development may explain why extramedullary hematopoiesis occurs in adult life in almost any tissue and organ [i.e., skull (21), brain (12), liver,
spleen, kidneys, adrenal glands, breast, paravertebral and presacral
areas (41), skin, testicles (33), heart
(17), lung (46), gastrointestinal tract
(39), pancreas (8), prostate
(19), and others] when blood availability is affected. It
would appear that hematopoiesis can be reactivated as a
compensatory mechanism in organs or regions where it previously
occurred during embryonic and fetal life.
The above observations suggest that many adult tissues have the
capacity to reactivate gene expression after a period of latency under
circumstances of abnormal physiology or disease. This process can occur
in vivo under pathophysiological circumstances as a way of compensating
the lack or malfunction of a specific tissue. As discussed above, the
ability of adult tissues to generate blood could be explained either by
the reactivation of a gene program in differentiated cells that have
retained the capacity to reacquire an embryonic phenotype, or likely,
the differentiation of resident precursor cells with hemo-vasculogenic
potential when the physiological conditions demand it to maintain
homeostasis. Examples for both possibilities seem to occur in nature
for other systems (15, 23, 31). Further work will be
necessary to define which one of these possibilities occurs during
extramedullary hematopoiesis.
A fundamental problem in experimental organogenesis is to obtain
appropriate tissue circulation. A substantial amount of experimental work will be needed to understand the mechanisms underlying
hemo-vasculogenesis. Nevertheless, because hemo-vasculogenesis can be
reproduced in vitro and in transplanted embryonic tissues, it offers
new opportunities in tissue regeneration and organogenesis.
| |
ACKNOWLEDGEMENTS |
We are grateful to O. Smithies, A. N. Goldfarb, B. Gumbiner,
and D. De Simone for discussions and suggestions. We thank E. S. Pentz for advice in primer design and ES cell culture.
| |
FOOTNOTES |
This work was supported by National Institutes of Health Grants (Center
of Excellence in Pediatric Nephrology, DK-52612; RO1, HL-66242; CHRC,
HDO-1421-01). Dr. Sequeira Lopez is a Howard Hughes Medical
Institute Physician Postdoctoral Fellow.
Address for reprint requests and other correspondence: R. Ariel
Gomez, Robert J. Roberts Professor of Pediatrics and Biology, Vice
President for Research and Public Service, MR4 Bldg., Rm. 2001, Univ.
of Virginia, 300 Lane Road, Charlottesville, VA 22908 (E-mail:
rg{at}virginia.edu).
The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement"
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
First published December 19, 2002;10.1152/ajpregu.00543.2002
Received 6 September 2002; accepted in final form 17 December 2002.
| |
REFERENCES |
1.
American Physiological Society.
Guiding principles for research involving animals and human beings.
Am J Physiol Regul Integr Comp Physiol
283:
R281-R283,
2002[Free Full Text].
2.
Baron, M.
Induction of embryonic hematopoietic and endothelial stem/progenitor cells by hedgehog-mediated signals.
Differentiation
68:
175-185,
2001[Web of Science][Medline].
3.
Bonner, RF,
Emmert-Buck M,
Cole K,
Pohida T,
Chuaqui R,
Goldstein S,
and
Liotta LA.
Laser capture microdissection: molecular analysis of tissue.
Science
278:
1481-1483,
1997[Free Full Text].
4.
Carey, AV,
Carey RM,
and
Gomez RA.
Expression of -smooth muscle actin in the developing kidney vasculature.
Hypertension
19:
II168-II175,
1992[Medline].
5.
Chomczynski, P,
and
Sacchi N.
Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction.
Anal Biochem
162:
156-159,
1987[Web of Science][Medline].
6.
Ciau-Uitz, A,
Walmsley M,
and
Patient R.
Distinct origins of adult and embryonic blood in Xenopus.
Cell
102:
787-796,
2000[Web of Science][Medline].
7.
Cline, MJ,
and
Moore MA.
Embryonic origin of the mouse macrophage.
Blood
39:
842-849,
1972[Abstract/Free Full Text].
8.
Crider, S,
Kroszer-Hamati A,
and
Krishnan K.
Isolated pancreatic extramedullary hematopoiesis.
Acta Haematol
99:
38-40,
1998[Web of Science][Medline].
9.
Cumano, A,
and
Godin I.
Pluripotent hematopoietic stem cell development during embryogenesis.
Curr Opin Immunol
13:
166-171,
2001[Web of Science][Medline].
10.
De Bruijn, MF,
Speck NA,
Peeters MC,
and
Dzierzak E.
Definitive hematopoietic stem cells first develop within the major arterial regions of the mouse embryo.
EMBO J
19:
2465-2474,
2000[Web of Science][Medline].
11.
Emmert-Buck, MR,
Bonner RF,
Smith PD,
Chuaqui RF,
Zhuang Z,
Goldstein SR,
Weiss RA,
and
Liotta LA.
Laser capture microdissection.
Science
274:
998-1001,
1996[Abstract/Free Full Text].
12.
Fucharoen, S,
Suthipongchai S,
Poungvarin N,
Ladpli S,
Sonakul D,
and
Wasi P.
Intracranial extramedullary hematopoiesis inducing epilepsy in a patient with -thalassemia-hemoglobin E.
Arch Intern Med
145:
739-742,
1985[Abstract/Free Full Text].
13.
Garcia-Porrero, JA,
Godin IE,
and
Dieterlen-Lievre F.
Potential intraembryonic hemogenic sites at pre-liver stages in the mouse.
Anat Embryol (Berl)
192:
425-435,
1995[Medline].
14.
Godin, IE,
Garcia-Porrero JA,
Coutinho A,
Dieterlen-Lievre F,
and
Marcos MA.
Para-aortic splanchnopleura from early mouse embryos contains B1a cell progenitors.
Nature
364:
67-70,
1993[Medline].
15.
Gomez, RA,
Chevalier RL,
Everett AD,
Elwood JP,
Peach MJ,
Lynch KR,
and
Carey RM.
Recruitment of renin gene-expressing cells in adult rat kidneys.
Am J Physiol Renal Fluid Electrolyte Physiol
259:
F660-F665,
1990[Abstract/Free Full Text].
16.
Guy, LG,
Kothary R,
DeRepentigny Y,
Delvoye N,
Ellis J,
and
Wall L.
The -globin locus control region enhances transcription of but does not confer position-independent expression onto the lacZ gene in transgenic mice.
EMBO J
15:
3713-3721,
1996[Web of Science][Medline].
17.
Hill, DA,
and
Swanson PE.
Myocardial extramedullary hematopoiesis: a clinicopathologic study.
Mod Pathol
13:
779-787,
2000[Web of Science][Medline].
18.
Hirai, H,
Ogawa M,
Suzuki N,
Yamamoto M,
Breier G,
Mazda O,
Imanishi J,
and
Nishikawa SI.
Hemogenic and non-hemogenic endothelium can be distinguished by the activity of fetal liver kinase (Flk)-1 promoter/enhancer during mouse embryogenesis.
Blood
101:
886-893,
2002[Medline].
19.
Humphrey, PA,
and
Vollmer RT.
Extramedullary hematopoiesis in the prostate.
Am J Surg Pathol
15:
486-490,
1991[Web of Science][Medline].
20.
Jaffredo, T,
Gautier R,
Eichmann A,
and
Dieterlen-Lievre F.
Intraaortic hemopoietic cells are derived from endothelial cells during ontogeny.
Development
125:
4575-4583,
1998[Abstract].
21.
Joseph, M,
Rajshekhar V,
and
Chandy MJ.
Haematopoietic tissue presenting as a sphenoid sinus mass: case report.
Neuroradiology
42:
153-154,
2000[Web of Science][Medline].
22.
Kisanuki, YY,
Hammer RE,
Miyazaki J,
Williams SC,
Richardson JA,
and
Yanagisawa M.
Tie2-Cre transgenic mice: a new model for endothelial cell-lineage analysis in vivo.
Dev Biol
230:
230-242,
2001[Web of Science][Medline].
23.
Lemischka, I.
Stem cell dogmas in the genomics era.
Rev Clin Exp Hematol
5:
15-25,
2001[Medline].
24.
Medvinsky, A,
and
Dzierzak E.
Definitive hematopoiesis is autonomously initiated by the AGM region.
Cell
86:
897-906,
1996[Web of Science][Medline].
25.
Murray, PDF
The development in vitro of the blood of the early chick embryo.
Proc R Soc Lond B Biol Sci
11:
497-521,
1932.
26.
Nishikawa, SI,
Nishikawa S,
Hirashima M,
Matsuyoshi N,
and
Kodama H.
Progressive lineage analysis by cell sorting and culture identifies FLK1+VE-cadherin+ cells at a diverging point of endothelial and hemopoietic lineages.
Development
125:
1747-1757,
1998[Abstract].
27.
North, T,
Gu TL,
Stacy T,
Wang Q,
Howard L,
Binder M,
Marin-Padilla M,
and
Speck NA.
Cbfa2 is required for the formation of intra-aortic hematopoietic clusters.
Development
126:
2563-2575,
1999[Abstract].
28.
North, TE,
de Bruijn MF,
Stacy T,
Talebian L,
Lind E,
Robin C,
Binder M,
Dzierzak E,
and
Speck NA.
Runx1 expression marks long-term repopulating hematopoietic stem cells in the midgestation mouse embryo.
Immunity
16:
661-672,
2002[Web of Science][Medline].
29.
Norwood, VF,
Garmey M,
Wolford J,
Carey RM,
and
Gomez RA.
Novel expression and regulation of the renin-angiotensin system in metanephric organ culture.
Am J Physiol Regul Integr Comp Physiol
279:
R522-R530,
2000[Abstract/Free Full Text].
30.
Ogawa, M,
Fraser S,
Fujimoto T,
Endoh M,
Nishikawa S,
and
Nishikawa SI.
Origin of hematopoietic progenitors during embryogenesis.
Int Rev Immunol
20:
21-44,
2001[Medline].
31.
Orkin, SH,
and
Zon LI.
Hematopoiesis and stem cells: plasticity versus developmental heterogeneity.
Nat Immun
3:
323-328,
2002.
32.
Pardanaud, L,
and
Dieterlen-Lievre F.
Manipulation of the angiopoietic/hemangiopoietic commitment in the avian embryo.
Development
126:
617-627,
1999[Abstract].
33.
Ruberto, E,
Espinola R,
Brusco J,
Bacchiocchi M,
Zoppi J,
and
Paz R.
Idiopathic myelofibrosis with extramedullary hematopoiesis foci in the skin and testicles. Report of a case.
Sangre (Barc)
40:
157-160,
1995[Medline].
34.
Sabin, FR.
Origin and development of the primitive vessels of the chick and the pig.
Contrib Embryol
6:
61-124,
1917.
35.
Schlaeger, TM,
Bartunkova S,
Lawitts JA,
Teichmann G,
Risau W,
Deutsch U,
and
Sato TN.
Uniform vascular-endothelial-cell-specific gene expression in both embryonic and adult transgenic mice.
Proc Natl Acad Sci USA
94:
3058-3063,
1997[Abstract/Free Full Text].
36.
Sequeira Lopez, ML,
Pentz ES,
Robert B,
Abrahamson DR,
and
Gomez RA.
Embryonic origin and lineage of juxtaglomerular cells.
Am J Physiol Renal Physiol
281:
F345-F356,
2001[Abstract/Free Full Text].
37.
Smith, RA,
and
Glomski CA.
"Hemogenic endothelium" of the embryonic aorta: does it exist?
Dev Comp Immunol
6:
359-368,
1982[Web of Science][Medline].
38.
Soriano, P.
Generalized lacZ expression with the ROSA26 Cre reporter strain.
Nat Genet
21:
70-71,
1999[Web of Science][Medline].
39.
Sunderland, K,
Barratt J,
and
Pidcock M.
Extramedullary hemopoiesis arising in the gut mimicking carcinoma of the cecum.
Pathology
26:
62-64,
1994[Web of Science][Medline].
40.
Takahashi, Y,
Imanaka T,
and
Takano T.
Spatial and temporal pattern of smooth muscle cell differentiation during development of the vascular system in the mouse embryo.
Anat Embryol (Berl)
194:
515-526,
1996[Medline].
41.
Tunaci, M,
Tunaci A,
Engin G,
Ozkorkmaz B,
Dincol G,
Acunas G,
and
Acunas B.
Imaging features of thalassemia.
Eur Radiol
9:
1804-1809,
1999[Web of Science][Medline].
42.
Turpen, JB,
Knudson CM,
and
Hoefen PS.
The early ontogeny of hematopoietic cells studied by grafting cytogenetically labeled tissue anlagen: localization of a prospective stem cell compartment.
Dev Biol
85:
99-112,
1981[Web of Science][Medline].
43.
Wong, PM,
Chung SW,
Chui DH,
and
Eaves CJ.
Properties of the earliest clonogenic hemopoietic precursors to appear in the developing murine yolk sac.
Proc Natl Acad Sci USA
83:
3851-3854,
1986[Abstract/Free Full Text].
44.
Wood, HB,
May G,
Healy L,
Enver T,
and
Morriss-Kay GM.
CD34 expression patterns during early mouse development are related to modes of blood vessel formation and reveal additional sites of hematopoiesis.
Blood
90:
2300-2311,
1997[Abstract/Free Full Text].
45.
Yamashita, J,
Itoh H,
Hirashima M,
Ogawa M,
Nishikawa S,
Yurugi T,
Naito M,
Nakao K,
and
Nishikawa S.
Flk1-positive cells derived from embryonic stem cells serve as vascular progenitors.
Nature
408:
92-96,
2000[Medline].
46.
Yusen, RD,
and
Kollef MH.
Acute respiratory failure due to extramedullary hematopoiesis.
Chest
108:
1170-1172,
1995[Medline].
Am J Physiol Regul Integr Comp Physiol 284(4):R1126-R1137
0363-6119/03 $5.00
Copyright © 2003 the American Physiological Society
TOP
ABSTRACT
INTRODUCTION
