Praena-Fernández, Maria J. Dominguez-Simeon, Jose Villar, Rafael Moreno-Luna and Juan M.
Rocio Muñoz-Hernandez, Maria L. Miranda, Pablo Stiefel, Ruei-Zeng Lin, Juan M.
Forming Cells in
Decreased Level of Cord Blood Circulating Endothelial Colony
Print ISSN: 0194-911X. Online ISSN: 1524-4563
Copyright © 2014 American Heart Association, Inc. All rights reserved.
is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
published online April 21, 2014;
World Wide Web at:
The online version of this article, along with updated information and services, is located on the
Data Supplement (unedited) at:
is online at:
Information about subscribing to
Information about reprints can be found online at:
document. Permissions and Rights Question and Answer this process is available in the
click Request Permissions in the middle column of the Web page under Services. Further information about
Office. Once the online version of the published article for which permission is being requested is located,
can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial
Requests for permissions to reproduce figures, tables, or portions of articles originally published
at Harvard University on April 22, 2014 at Harvard University on April 22, 2014 at Harvard University on April 22, 2014 at Harvard University on April 22, 2014 at Harvard University on April 22, 2014 at Harvard University on April 22, 2014 at Harvard University on April 22, 2014http://hyper.ahajournals.org/ http://hyper.ahajournals.org/ http://hyper.ahajournals.org/http://hyper.ahajournals.org/http://hyper.ahajournals.org/http://hyper.ahajournals.org/ http://hyper.ahajournals.org/Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from
and fetal morbidity and mortality.1,2 Offspring of preeclamp-
tic pregnancies have an increased risk of developing postnatal
cardiovascular events, including hypertension and stroke.3,4
Epidemiological studies have shown that several cardiovas-
cular diseases have origins during development.5 However,
the effects of preeclampsia on the fetal cardiovascular system
remain poorly understood.
Endothelial colony–forming cells (ECFCs) are circulat-
ing progenitor cells that give rise to highly vasculogenic
endothelial cells.6,7 ECFC levels in fetal blood are elevated
during the third trimester of pregnancy,8–10 and these cells
are postulated to contribute to the rapid formation of fetal
vasculature and to the maintenance of vascular integrity.11,12
Recent studies have shown that cord blood ECFC level and
function are impaired in several pregnancy-related disorders
reeclampsia is a multisystem syndrome affecting 2% to
8% of pregnancies, and it is a major cause of maternal
associated with long-term cardiovascular risks, including
gestational diabetes mellitus, fetal bronchopulmonary dys-
plasia, and intrauterine growth restriction.13–16 However, it
remains unclear whether cord blood levels of ECFCs are also
altered during preeclampsia.
Here, we conducted a prospective cohort study to determine
the umbilical cord blood levels of ECFCs in preeclampsia and
analyzed the results in light of potential confounding obstetric
factors. We also compared the functional properties of ECFCs
derived from preeclamptic and normal pregnancies.
Fifteen (preeclampsia) and 35 (control) white mother–offspring pairs
were included in this study. Preeclampsia was defined as high blood
pressure (>140/90 mm Hg) and excess protein in the urine (>0.3
g in 24 hours) after 20 weeks of pregnancy. Pre-existing chronic
Abstract—Preeclampsia is a pregnancy-related disorder associated with increased cardiovascular risk for the offspring.
Endothelial colony–forming cells (ECFCs) are a subset of circulating endothelial progenitor cells that participate
in the formation of vasculature during development. However, the effect of preeclampsia on fetal levels of ECFCs
is largely unknown. In this study, we sought to determine whether cord blood ECFC abundance and function are
altered in preeclampsia. We conducted a prospective cohort study that included women with normal (n=35) and
preeclamptic (n=15) pregnancies. We measured ECFC levels in the umbilical cord blood of neonates and characterized
ECFC phenotype, cloning-forming ability, proliferation, and migration toward vascular endothelial growth factor-A
and fibroblast growth factor-2, in vitro formation of capillary-like structures, and in vivo vasculogenic ability in
immunodeficient mice. We found that the level of cord blood ECFCs was statistically lower in preeclampsia than in
control pregnancies (P=0.04), a reduction that was independent of other obstetric factors. In addition, cord blood ECFCs
from preeclamptic pregnancies required more time to emerge in culture than control ECFCs. However, once derived in
culture, ECFC function was deemed normal and highly similar between preeclampsia and control, including the ability
to form vascular networks in vivo. This study demonstrates that preeclampsia affects ECFC abundance in neonates.
A reduced level of ECFCs during preeclamptic pregnancies may contribute to an increased risk of developing future
cardiovascular events. (Hypertension. 2014;64:00-00.) • Online Data Supplement
Key Words: fetal blood ■ preeclampsia ■ pregnancy
Received December 20, 2013; first decision January 13, 2014; revision accepted February 26, 2014.
From the CSIC/Universidad de Sevilla, Unidad Clínico-Experimental de Riesgo Vascular (UCAMI-UCERV) (R.M.-H., M.L.M., P.S., M.J.D.-S., J.V.,
R.M.-L.) and Unidad de Asesoría Estadística, Metodología y Evaluación de Investigación, Fundación Pública Andaluza para la Gestión de la Investigación
en Salud de Sevilla (FISEVI) (J.M.P.-F.), Instituto de Biomedicina de Sevilla (IBiS) and Hospital Universitario Virgen del Rocío, Seville, Spain; Department
of Cardiac Surgery, Boston Children’s Hospital, MA (R.M.-H., R.-Z.L., R.M.-L., J.M.M.-M.); Department of Surgery, Harvard Medical School, Boston,
MA (R.-Z.L., R.M.-L., J.M.M.-M.); and Harvard Stem Cell Institute, Cambridge, MA (J.M.M.-M.).
The online-only Data Supplement is available with this article at http://hyper.ahajournals.org/lookup/suppl/doi:10.1161/HYPERTENSIONAHA.
Correspondence to Juan M. Melero-Martin, Department of Cardiac Surgery, Boston Children’s Hospital, 300 Longwood Ave, Enders 349, Boston, MA
02115. E-mail email@example.com or Rafael Moreno-Luna, Vascular Pathophysiology Laboratory, National Hospital Paraplegic,
SESCAM, Finca la Peraleda s/n, 45071 Toledo, Spain. E-mail firstname.lastname@example.org
© 2014 American Heart Association, Inc.
Hypertension is available at http://hyper.ahajournals.org
Decreased Level of Cord Blood Circulating Endothelial
Colony–Forming Cells in Preeclampsia
Rocio Muñoz-Hernandez, Maria L. Miranda, Pablo Stiefel, Ruei-Zeng Lin,
Juan M. Praena-Fernández, Maria J. Dominguez-Simeon, Jose Villar, Rafael Moreno-Luna,
Juan M. Melero-Martin
at Harvard University on April 22, 2014http://hyper.ahajournals.org/Downloaded from
2 Hypertension July 2014
hypertension was not an exclusion criterion for preeclampsia. All
preeclamptic mothers were treated with α-methyldopa; in addition,
5 patients also received labetalol. Intrauterine growth restriction was
defined as a fetus with an individualized weight percentile <10%
and with asymmetry in several ultrasound measurements, including
a significant decrease in abdominal perimeter compared with long
bone length and biparietal diameter. Exclusion criteria included mul-
tiple gestation, maternal infections, respiratory disease, and women
who carried fetuses with chromosomal abnormalities or congenital
malformations. In the control group, women with hypertensive dis-
orders were excluded. The local ethics committee at the Hospital
Universitario Virgen del Rocío approved this research, and all the
parents gave written informed consent for extraction of data from
their obstetric records and for the use of umbilical cord blood in
accordance with the Declaration of Helsinki. Methods on obstetric
factors are described in Materials and Methods in the online-only
Enumeration and Characterization of ECFCs
Umbilical cord blood samples (20–50 mL) were collected ex utero
using heparinized tubes and processed within 2 hours. Enumeration
and characterization of ECFCs were performed following previously
described methods17–19; details can be found in expanded Materials
and Methods in the online-only Data Supplement.
Data from preeclampsia and control subjects were compared and
analyzed with IBM SPSS v. 19.0 software (IBM Corp, Armonk,
NY). Categorical variables were expressed by absolute frequencies
and percentages (n, %). Noncategorical variables were expressed by
mean±SD or median and 25th to 75th interquartile range. Categorical
variables were analyzed with Fisher exact tests except for tobacco
use and offspring sex, which were analyzed with Pearson χ2 tests.
Noncategorical variables were analyzed with 2-tailed unpaired
Student t tests, with the exception of maternal age and gestational
age, which were not normally distributed and therefore analyzed with
Mann–Whitney U tests. Shapiro–Wilk tests were used to determine
normality. Univariate correlations were performed with the use of
Spearman correlation coefficient. Data from experiments performed
in vitro and in mice were analyzed using GraphPad Prism v. 5 soft-
ware (GraphPad Software, La Jolla, CA). These data were expressed
as mean±SE and mean values were compared using unpaired Student
t tests. For all analyses, P<0.05 was considered significant.
We studied 15 (preeclampsia) and 35 (control) mother–off-
spring pairs (Table). Based on the severity of the pathology,
the preeclampsia group included subjects with mild (n=6),
severe (n=7), and hemolysis, elevated liver enzymes, and low
platelets (HELLP) syndrome (n=2). In addition, 2 subjects in
the preeclampsia group had pre-existing chronic hyperten-
sion. The prevalence of cesarean deliveries in the preeclamp-
sia group was statistically higher than in control (P=0.001).
Offspring born from mothers with preeclampsia had lower
gestational age, birth weight, and birth weight percentile than
those in the control group (P=0.001, P=0.004, and P=0.006,
respectively). Preeclamptic mothers had higher pregestational
diastolic blood pressure (P=0.003) than control. There were
no statistical differences in the remainder of the obstetric char-
acteristics analyzed (P>0.05).
Cord Blood Levels of ECFCs in Preeclampsia
We quantified the number of ECFCs in the umbilical cord
blood of neonates at the time of delivery. ECFCs were
identified in culture as outgrown colonies containing ≥50
endothelial cells. The endothelial nature of the colonies was
corroborated by the cobblestone-like morphology of the cells
(Figure 1A) and by binding of fluorescently labeled Ulex
europaeus agglutinin type 1 lectin (Figure 1B). Colonies in
the control group emerged in culture as early as 1 week (7%
of the colonies), and most of the colonies emerged between
2 weeks (60%) and 3 weeks (31%; Figure 1C), which is con-
sistent with previous reports.18 In contrast, the time needed
for colony appearance in the preeclampsia group was higher,
and a substantial proportion of colonies (44%) emerged in
the fourth week of culture (Figure 1C). Total ECFC level
Table. Obstetric Characteristics of Preeclampsia and Control
Primipara, n (%)
In vitro fertilization, n (%)
Cesarean delivery, n (%)
Tobacco use, n (%)
Gestational diabetes mellitus, n (%)
Pregestational BMI, kg/m2
<25, n (%)
25–30, n (%)
>30, n (%)
Gestational weight gain, kg
Pregestational blood pressure,† mm Hg
Blood pressure at onset of PE, mm Hg
Male, n (%)
Gestation age, wk
restriction, n (%)
Preterm birth, n (%)
Birth weight, kg
Birth weight percentile, %
Cord blood MNC level,‡ millions
per 10 mL blood
Categorical variables are represented by absolute frequencies and
percentages (n, %). Noncategorical variables are represented by mean±SD.
Categorical variables were analyzed with Fisher exact tests except for tobacco
use and offspring sex that were analyzed with Pearson χ2 tests. P values are
from comparison of control and preeclampsia groups.
BMI indicates body mass index; MNC, mononuclear cell; and PE, preeclampsia.
*Noncategorical variables that were normally distributed were analyzed with
Student t tests. Noncategorical variables that were not normally distributed were
analyzed with Mann–Whitney U tests.
†Values for pregestational blood pressure are from n=27 control subjects.
‡Values for MNC level are from n=29 control subjects.
at Harvard University on April 22, 2014http://hyper.ahajournals.org/ Downloaded from
Muñoz-Hernandez et al Neonatal Endothelial Progenitors in Preeclampsia 3
in each group was determined after 4 weeks in culture. The
median ECFC level in control was 5 colonies per 10 mL of
cord blood with a broad 25th to 75th interquartile range of 0.5
to 13 colonies. Meanwhile, ECFC level in preeclampsia was
statistically lower than in control (P=0.04), with a median
abundance of 1 colony per 10 mL of cord blood and a 25th to
75th interquartile of 0 to 4 colonies (Figure 1D). Moreover, a
significant portion of the preeclamptic group in the study had
no measurable ECFCs. Statistical analyses performed in both
preeclampsia and control groups demonstrated that the level
of ECFCs was independent (P>0.05) of most obstetric fac-
tors (Tables S1 and S2 in the online-only Data Supplement),
including maternal age (P=0.06 and P=0.77 in preeclampsia
and control, respectively; Table S2), mode of delivery (P=0.53
and P=0.83; Table S1), offspring sex (P=0.40 and P=0.64;
Table S1), offspring birth weight (P=0.27 and P=0.87; Table
S2), offspring birth weight percentile (P=0.26 and P=0.71;
Table S2), gestational weight gain (P=0.31 and P=0.08; Table
S2), and cord blood mononuclear cell level (P=0.95 and
P=0.97; Table S2). Moreover, the level of cord blood ECFCs
in preeclampsia was independent of both the severity of the
pathology (mild/severe/hemolysis, elevated liver enzymes,
and low platelets (HELLP) syndrome; P=0.06), the time
of onset of preeclampsia (early/late; P=0.42), diastolic and
systolic pregestational blood pressure (P=0.51 and P=0.94;
Table S2), and diastolic and systolic blood pressure at the
time of onset of preeclampsia (P=0.52 and P=0.27; Table S2).
Variation of Cord Blood ECFC Levels With
Maternal Body Mass Index and Gestational Age
Previously, we demonstrated that maternal body mass index
(BMI) is a potential confounding factor for cord blood levels
of ECFCs.18 To address whether the difference in ECFC abun-
dance between preeclampsia and control was confounded by
maternal weight, we categorized the study into prepregnancy
maternal BMI <25 kg/m2 (normal weight; n=15/n=5 control/
preeclampsia), 25 to 30 kg/m2 (overweight; n=12/n=6), and
>30 kg/m2 (obese; n=8/n=4; Figure 2A). ECFC levels in control
subjects increased from normal prepregnancy maternal weight
(mean of 4 colonies) to overweight (11 colonies) and obese
(7 colonies) subjects, with statistically significant differences
between these subgroups (Figure 2A). In contrast, the level of
ECFCs in preeclampsia was consistently low, irrespective of
the value of maternal BMI, with mean ECFC abundances of 2,
3, and 3 colonies in cord blood samples from normal weight,
overweight, and obese mothers, respectively (Figure 2A). In
addition, the difference in ECFC levels between control and
preeclampsia for maternal BMI 25 to 30 kg/m2 was statistically
significant (P<0.05). Taken together, these results confirmed
that maternal BMI is a confounding factor for ECFC level and
demonstrated that the reduction in ECFC abundance observed
in preeclampsia was more prominent among subjects in the
overweight (BMI=25–30 kg/m2) group.
To address whether the difference in ECFC abundance
between preeclampsia and control was influenced by ges-
tational age, we categorized the study into premature (<37
gestational weeks; n=5/n=5 control/preeclampsia) or term
(≥37 weeks; n=30/n=10) deliveries (Figure 2B). Our study
did not include extremely premature infants, and the lowest
gestational age for both groups was 31 weeks. We observed
that ECFC abundance in the control group was increased in
prematurity (Figure 2B), which is consistent with previous
reports.8,18 However, the level of ECFCs in preeclampsia did
not change with gestational age (P>0.05), and it remained sig-
nificantly lower than the control for both preterm (4±1 colo-
nies in preeclampsia and 14±2 colonies in control; P=0.02)
Figure 1. Cord blood levels of endothelial colony–forming cells
(ECFCs) in preeclampsia. A, Phase contrast micrograph of a
representative ECFC colony from a preeclamptic (PE) pregnancy.
Arrowheads delimitate the border of the colony (scale bar,
200 μm). B, Binding of fluorescently labeled Ulex europaeus
agglutinin type 1 lectin (UEA-1) to a colony of ECFCs (scale bar,
200 μm). C, Weekly appearance of ECFC colonies in culture.
Bars represent mean±SE levels of ECFCs in 10 mL of cord blood.
Percentages represent the proportion of total ECFCs appeared
each week. D, Total number of ECFC colonies in 10 mL of cord
blood from normal (n=35) and PE (n=15) pregnancies. Lines
represent mean ECFC abundance. n values are denoted on top
of each group. *P<0.05. CTR indicates control.
Figure 2. Variation of cord blood endothelial
colony–forming cell (ECFC) levels with maternal
body mass index (BMI) and gestational age. A,
ECFC abundance in cord blood from subjects
categorized by prepregnancy maternal BMI.
B, Cord blood level of ECFCs from deliveries
categorized by gestational age as preterm (<37
weeks) and term (≥37 weeks). Lines represent
mean ECFC abundance in 10 mL of cord blood.
n values are denoted on top of each group.
*P<0.05 between control (CTR) and preeclamptic
(PE) groups for maternal BMI 25 to 30 kg/m2 and
gestational age <37 weeks.
at Harvard University on April 22, 2014http://hyper.ahajournals.org/ Downloaded from
4 Hypertension July 2014
and term deliveries (2±1 colonies in preeclampsia and 6±2
colonies in control; P=0.06; Figure 2B). In addition, the dif-
ference in ECFC levels between control and preeclampsia
for gestational age <37 weeks was statistically significant
(P<0.05). These results confirmed that gestational age is a
confounding factor for ECFC level (increased in prematurity)
and demonstrated that the overall reduction in ECFC abun-
dance observed in preeclampsia was more prominent among
Phenotypic and Functional Characteristics of Cord
Blood ECFCs in Preeclampsia
We then examined whether there were functional differences
among ECFCs from the control and preeclampsia groups. To
this aim, ECFCs were first expanded in culture and purified by
virtue of CD31 expression (Figure 3A). The endothelial phe-
notype of CD31-selected cells was verified via expression of
CD31 and vascular endothelial-cadherin at the cell–cell bor-
ders, and the expression of von Willebrand factor in a punctu-
ate pattern in the cytoplasm (Figure 3B). Quantitative reverse
transcription polymerase chain reaction analyses demon-
strated similar levels of expression of endothelial cell markers
(CD31, von Willebrand factor, vascular endothelial-cadherin,
and endothelial nitric oxide synthase) and absence of mesen-
chymal cell markers (CD90 and platelet-derived growth fac-
tor receptor-β) in ECFCs from both preeclampsia and control
(P>0.05; Figure 3C). We also observed that ECFCs from both
groups expressed high levels of growth factor receptors vas-
cular endothelial growth factor (VEGF) receptor-1, VEGF
receptor-2, and fibroblast growth factor (FGF) receptor-1 and
low levels of VEGF receptor-3, FGF receptor-2, and FGF
receptor-3 (Figure 3D), which is consistent with a vascular
To assess ECFC function, we randomly selected 6 ECFC
cultures from each group and performed several in vitro func-
tional assays (Figure 4). In the preeclampsia group, one of the
ECFC cultures selected corresponded to a subject with intra-
uterine growth restriction. We observed no statistical difference
between preeclampsia and control in ECFC cloning-forming
ability (Figure 4A; P=0.48 and Figure 4B; P=0.60) and in
the capacity of ECFCs to assemble into capillary-like struc-
tures on Matrigel (Figure 4C and 4D; P=0.95 and Figure 4E;
P=0.81). We observed a moderate decrease in the mitogenic
and migratory response to VEGF-A (Figure 4F; P=0.52
and Figure 4H; P=0.20) and FGF-2 (Figure 4F; P=0.15 and
Figure 4H; P=0.29) in cord blood ECFCs from preeclamp-
sia, although these differences were not statistically sig-
nificant for n=6. We also examined the in vivo vasculogenic
ability of ECFCs after transplantation into immunodeficient
mice (Figure 5). In both preeclampsia and control groups,
transplanted ECFCs formed extensive networks of perfused
microvessels by day 7, as revealed by hematoxylin and eosin–
stained sections of the explants (Figure 5A) and confirmed
by immunohistochemical staining of human-specific CD31
(Figure 5B). Microvessels also stained positively for Ulex
europaeus agglutinin type 1 lectin, a lectin that specifically
binds to human (but not murine) endothelial cells (Figure 5C).
In addition, ECFC-lined microvessels had extensive perivas-
cular coverage at day 7, as revealed by positive α-smooth
muscle actin expression (Figure 5C), which indicated vascular
stability. Importantly, quantitative histological evaluation of
human-specific microvessel density demonstrated no statisti-
cal difference between ECFCs from preeclampsia and control
(Figure 5D; P=0.71).
The mechanisms that govern the abundance of ECFCs in
health and disease are insufficiently known. The maternal vas-
cular pathophysiologic features of preeclampsia are well char-
acterized and involve widespread endothelial dysfunction.21
However, the effects of preeclampsia on fetal levels of circu-
lating progenitor cells have not been examined systematically.
A previous study by Hwang et al22 demonstrated a decrease of
cord blood AC133+/KDR+/CD34+ endothelial progenitor cells
and their progeny in pregnancies complicated by preeclamp-
sia. However, there is increasing consensus on the distinction
between cells that originate from AC133+/KDR+/CD34+ endo-
thelial progenitor cells and those that are defined as ECFCs.23–25
Indeed, Yoder et al23 demonstrated that early endothelial pro-
genitor cells that generate endothelial cell colony-forming
Figure 3. Phenotype of cord blood endothelial
colony–forming cells (ECFCs) in preeclampsia.
A, Phase contrast micrograph of CD31-selected
culture-expanded ECFCs from preeclamptic
(PE) cord blood (scale bar, 200 μm). B, ECFC
expression of CD31, vascular endothelial (VE)-
cadherin, and von Willebrand factor (vWF)
demonstrated by indirect immunofluorescence.
Cell nuclei were counterstained with 4',6-diamidino-
2-phenylindole (DAPI; scale bar, 50 μm).
C, Quantitative reverse transcription polymerase
chain reaction analyses of ECFCs for endothelial
(CD31, vWF, VE-cadherin, and endothelial nitric
oxide synthase [eNOS]) and mesenchymal (platelet-
derived growth factor receptor [PDGFR]-β, CD90)
cell markers and for (D) vascular endothelial
growth factor receptors (VEGFRs; VEGFR-1,
VEGFR-2, and VEGFR-3) and fibroblast growth
factor receptors (FGFRs; FGFR-1, FGFR-2, and
FGFR-3). Bars represent mean±SE (n=6) number of
mRNA transcripts normalized to 106 copies of 18S
ribosomal RNA. CTR indicates control.
at Harvard University on April 22, 2014http://hyper.ahajournals.org/ Downloaded from
Muñoz-Hernandez et al Neonatal Endothelial Progenitors in Preeclampsia 5
units are hematopoietic in origin, fail to form perfused ves-
sels in vivo, and are clonally distinct from ECFCs. Thus, in
addition to variations in the number of AC133+/KDR+/CD34+
endothelial progenitor cells, it remains unclear whether pre-
eclampsia alters baseline levels of cord blood ECFCs. Here,
we unambiguously identified ECFCs based on well-known
endothelial cell markers and functional properties and dem-
onstrated that the level of cord blood circulating ECFCs is
decreased in preeclampsia. This reduction was statistically
significant, independent of common obstetric factors, and was
not associated with changes in cell phenotype or function.
Recent studies have emphasized the importance of several
confounding factors on circulating levels of ECFCs, includ-
ing maternal BMI and gestational age.8,18 Previously, we dem-
onstrated a positive correlation between maternal BMI and
ECFC abundance in umbilical cord blood of neonates born
from nonobese healthy mothers with nonpathological preg-
nancies.18 This association suggested a potential physiological
adaptation that occurs in the rapidly growing fetus in response
to intrauterine conditions imposed by maternal weight. In this
study, we examined the influence of maternal prepregnancy
weight and found that ECFCs levels were consistently lower
in preeclampsia than in control pregnancies, irrespective of
maternal BMI. Gestational age has also been recognized as
a source of variation for ECFC levels. Previous studies have
shown that levels of circulating ECFCs are more elevated
in premature deliveries (gestational age, 28–35 weeks) than
at term,8 although extremely premature infants (<28 weeks)
have been associated with fewer ECFCs.9,14 We examined the
influence of gestational age in an equal number of premature
infants (<37 gestational weeks) and observed that indepen-
dent of gestational age, ECFCs levels were consistently low
in the pathological group. This implicated that the difference
in ECFC abundance between preeclampsia and control was
more significant in premature deliveries than at term. Taken
together, our data suggest an impaired mobilization of ECFCs
in preeclampsia that is more evident in preterm deliveries and
is independent of common obstetric factors.
Emerging evidence indicates that besides inflicting varia-
tions in abundance, deleterious conditions during fetal life
can also impair ECFCs function.13,16,26 For instance, ECFCs
from newborns of diabetic mothers display premature senes-
cence and reduced proliferative and vasculogenic properties,
Figure 4. In vitro functional properties of cord
blood endothelial colony–forming cells (ECFCs) in
preeclampsia. A, Clonogenic properties of ECFCs
expressed as (A) percentage of cells with cloning-
forming ability and (B) mean number of cells per
colony after 10 days in culture. Cell nuclei were
identified by 4',6-diamidino-2-phenylindole (DAPI)
staining (inset). C, Representative phase contrast
micrographs of capillary-like networks formed by
ECFCs on Matrigel (scale bar, 300 μm). The ability
to form capillary-like networks was quantified and
expressed as (D) total number of capillaries per
field and (E) total capillary length per field. F, Cell
proliferation in response to vascular endothelial
growth factor (VEGF)-A (10 ng/mL) and fibroblast
growth factor (FGF)-2 (1 ng/mL) expressed as fold
increase in cell number. G, Representative phase
contrast micrographs depicting the closure of a gap
created in an ECFC monolayer (scale bar, 200 μm).
Gap closure was monitored in response to VEGF-A
(10 ng/mL) and FGF-2 (1 ng/mL). H, Migratory
capacity of ECFC in response to VEGF-A and FGF-
2 expressed as percentage of gap closure after
15 hours. Bars represent mean±SE (n=6). CTR
indicates control; and PE, preeclamptic.
Figure 5. In vivo vasculogenic properties of cord blood endothelial
colony–forming cells (ECFCs) in preeclampsia. ECFCs were
combined with mesenchymal stem cells in Matrigel and the mixture
subcutaneously injected into nude mice for 7 days. A, Hematoxylin
and eosin (H&E)–stained section of a representative explant
revealing numerous perfused blood vessels at day 7. Macroscopic
view of the explant is depicted in the inset (scale in mm).
B, Immunohistochemical staining with an antibody against human-
specific CD31 (hCD31) revealing numerous human blood vessel
lumens (yellow arrowheads). Cell nuclei were counterstained with
hematoxylin. C, Perivascular coverage was assessed by double
immunofluorescence staining using Ulex europaeus agglutinin
type 1 lectin (UEA-1; red) and an antibody against α-smooth
muscle actin (α-SMA; green; white arrowheads indicate double
positive lumens). Nuclei were counterstained with 4',6-diamidino-2-
phenylindole (DAPI). All images (A–C) are representative of implants
that were seeded with ECFCs from preeclamptic (PE) pregnancies
(scale bar, 100 μm). D, Microvessel density determined as the
number of ECFC-lined blood vessels per unit of area in implants
that were seeded with ECFCs from either PE or control (CTR)
pregnancies. Bars represent mean±SE (n=6).
at Harvard University on April 22, 2014http://hyper.ahajournals.org/Downloaded from
6 Hypertension July 2014
including a decrease in the ability to form chimeric vessels
after transplantation into immunodeficient mice.13 Similarly,
ECFCs derived from pregnancies complicated by intrauter-
ine growth restriction exhibit altered vasculogenic poten-
tial.16 In this study, we observed a considerable delay in the
average time of colony appearance in preeclampsia, with
a significant proportion of ECFC colonies emerging during
the fourth week of culture. However, with the exception of
the delayed endothelial colony formation, ECFCs from pre-
eclamptic pregnancies were otherwise deemed functionally
normal. The ability to grow at clonal density and the capacity
to form capillary-like networks were highly similar between
ECFCs from the preeclamptic group and their nonpathologi-
cal counterparts. The proliferative and migratory responses
to angiogenic factors VEGF-A and FGF-2 were reduced in
ECFCs from the preeclamptic group, although the differences
with control ECFCs were not statistically significant. More
importantly, ECFCs from the preeclamptic group displayed
full vasculogenic capacity after transplantation into immuno-
deficient mice, forming extensive networks of perfused blood
vessels with complete perivascular coverage. Taken together,
ECFC function was deemed similar between preeclampsia
and control. Nevertheless, whether a larger sample size may
reveal small functional differences not appreciated in our
study remains a possibility.
In this study, we demonstrated a decreased level of umbilical
cord blood circulating ECFCs in preeclampsia. Cord blood
ECFCs from preeclamptic pregnancies required more time
to emerge in culture as endothelial colonies than control
ECFCs, but they displayed otherwise normal vascular activ-
ity in vitro and in vivo. Epidemiological studies have indi-
cated that several cardiovascular diseases originate during
development, and thus there is increasing interest in under-
standing the relation between the activity of fetal progeni-
tor cells and the appearance of cardiovascular pathologies
in the offspring. To date, the pathophysiological implica-
tions of having reduced levels of circulating ECFCs during
pregnancy are not well understood. Further studies should
examine whether the reduced level of cord blood ECFCs
observed in preeclampsia correlates with elevated risk of
developing subsequent cardiovascular events, such as stroke
We thank Dan Li and Dr Shou-Ching Jaminet (Center for Vascular
Biology, Department of Pathology, Beth Israel Deaconess Medical
Center) for quantitative reverse transcription polymerase chain reac-
tion analyses and the study midwives (Unidad Clínica de Obstetricia
y Urgencias del Hospital de la Mujer, Hospital Universitario Virgen
del Rocio) for assistance in the collection of umbilical cord blood
Sources of Funding
This work was supported by the National Institutes of Health
(R00EB009096 to Dr Melero-Martin); Sistema Andaluz de Salud,
Consejería de Salud (SAS111241 to Dr Moreno-Luna) and Consejería
de Economía, Innovación y Ciencia (P08-CVI-4352 to Dr Villar) de
la Junta de Andalucía, and Instituto de Salud Carlos III (PI10/02473
to Dr Stiefel).
1. Lain KY, Roberts JM. Contemporary concepts of the pathogenesis and
management of preeclampsia. JAMA. 2002;287:3183–3186.
2. Redman CW, Sargent IL. Latest advances in understanding preeclampsia.
3. Øglaend B, Forman MR, Romundstad PR, Nilsen ST, Vatten LJ. Blood
pressure in early adolescence in the offspring of preeclamptic and normo-
tensive pregnancies. J Hypertens. 2009;27:2051–2054.
4. Kajantie E, Eriksson JG, Osmond C, Thornburg K, Barker DJ.
Pre-eclampsia is associated with increased risk of stroke in the adult off-
spring: the Helsinki birth cohort study. Stroke. 2009;40:1176–1180.
5. Barker DJ, Winter PD, Osmond C, Margetts B, Simmonds SJ. Weight in
infancy and death from ischaemic heart disease. Lancet. 1989;2:577–580.
6. Melero-Martin JM, Khan ZA, Picard A, Wu X, Paruchuri S, Bischoff J. In
vivo vasculogenic potential of human blood-derived endothelial progeni-
tor cells. Blood. 2007;109:4761–4768.
7. Melero-Martin JM, De Obaldia ME, Kang SY, Khan ZA, Yuan L, Oettgen
P, Bischoff J. Engineering robust and functional vascular networks in
vivo with human adult and cord blood-derived progenitor cells. Circ Res.
8. Baker CD, Ryan SL, Ingram DA, Seedorf GJ, Abman SH,
Balasubramaniam V. Endothelial colony-forming cells from preterm
infants are increased and more susceptible to hyperoxia. Am J Respir Crit
Care Med. 2009;180:454–461.
9. Javed MJ, Mead LE, Prater D, Bessler WK, Foster D, Case J, Goebel WS,
Yoder MC, Haneline LS, Ingram DA. Endothelial colony forming cells
and mesenchymal stem cells are enriched at different gestational ages in
human umbilical cord blood. Pediatr Res. 2008;64:68–73.
10. Ligi I, Simoncini S, Tellier E, et al. A switch toward angiostatic gene
expression impairs the angiogenic properties of endothelial progenitor
cells in low birth weight preterm infants. Blood. 2011;118:1699–1709.
11. Hirschi KK, Ingram DA, Yoder MC. Assessing identity, phenotype, and fate of
endothelial progenitor cells. Arterioscler Thromb Vasc Biol. 2008;28:1584–1595.
12. Yoder MC. Human endothelial progenitor cells. Cold Spring Harb
Perspect Med. 2012;2:a006692.
13. Ingram DA, Lien IZ, Mead LE, Estes M, Prater DN, Derr-Yellin E,
DiMeglio LA, Haneline LS. In vitro hyperglycemia or a diabetic intra-
uterine environment reduces neonatal endothelial colony-forming cell
numbers and function. Diabetes. 2008;57:724–731.
14. Borghesi A, Massa M, Campanelli R, Bollani L, Tzialla C, Figar TA,
Ferrari G, Bonetti E, Chiesa G, de Silvestri A, Spinillo A, Rosti V, Stronati
M. Circulating endothelial progenitor cells in preterm infants with bron-
chopulmonary dysplasia. Am J Respir Crit Care Med. 2009;180:540–546.
15. Baker CD, Balasubramaniam V, Mourani PM, Sontag MK, Black CP, Ryan
SL, Abman SH. Cord blood angiogenic progenitor cells are decreased in
bronchopulmonary dysplasia. Eur Respir J. 2012;40:1516–1522.
16. Sipos PI, Bourque SL, Hubel CA, Baker PN, Sibley CP, Davidge ST,
Crocker IP. Endothelial colony-forming cells derived from pregnancies
complicated by intrauterine growth restriction are fewer and have reduced
vasculogenic capacity. J Clin Endocrinol Metab. 2013;98:4953–4960.
17. Lin RZ, Dreyzin A, Aamodt K, Dudley AC, Melero-Martin JM. Functional
endothelial progenitor cells from cryopreserved umbilical cord blood.
Cell Transplant. 2011;20:515–522.
18. Moreno-Luna R, Muñoz-Hernandez R, Lin RZ, Miranda ML, Vallejo-Vaz
AJ, Stiefel P, Praena-Fernández JM, Bernal-Bermejo J, Jimenez-Jimenez
LM, Villar J, Melero-Martin JM. Maternal body-mass index and cord blood
circulating endothelial colony-forming cells. J Pediatr. 2014;164:566–571.
19. Lin RZ, Moreno-Luna R, Zhou B, Pu WT, Melero-Martin JM. Equal
modulation of endothelial cell function by four distinct tissue-specific
mesenchymal stem cells. Angiogenesis. 2012;15:443–455.
20. Klagsbrun M, D’Amore PA. Vascular endothelial growth factor and its
receptors. Cytokine Growth Factor Rev. 1996;7:259–270.
21. Davidge ST, Signorella AP, Lykins DL, Gilmour CH, Roberts JM. Evidence
of endothelial activation and endothelial activators in cord blood of infants
of preeclamptic women. Am J Obstet Gynecol. 1996;175:1301–1306.
22. Hwang HS, Maeng YS, Park YW, Koos BJ, Kwon YG, Kim YH. Increased
senescence and reduced functional ability of fetal endothelial progenitor
cells in pregnancies complicated by preeclampsia without intrauterine
growth restriction. Am J Obstet Gynecol. 2008;199:259.e1–259.e7.
23. Yoder MC, Mead LE, Prater D, Krier TR, Mroueh KN, Li F, Krasich R,
Temm CJ, Prchal JT, Ingram DA. Redefining endothelial progenitor cells
at Harvard University on April 22, 2014http://hyper.ahajournals.org/ Downloaded from
Muñoz-Hernandez et al Neonatal Endothelial Progenitors in Preeclampsia 7
via clonal analysis and hematopoietic stem/progenitor cell principals. Blood.
24. Fadini GP, Losordo D, Dimmeler S. Critical reevaluation of endothelial
progenitor cell phenotypes for therapeutic and diagnostic use. Circ Res.
25. Yoder MC. Endothelial progenitor cell: a blood cell by many other names
may serve similar functions. J Mol Med (Berl). 2013;91:285–295.
26. Acosta JC, Haas DM, Saha CK, Dimeglio LA, Ingram DA, Haneline LS.
Gestational diabetes mellitus alters maternal and neonatal circulating endothe-
lial progenitor cell subsets. Am J Obstet Gynecol. 2011;204:254.e8–254.e15.
What Is New?
•?To our knowledge, this is the first prospective cohort study that examines
cord blood endothelial colony–forming cell (ECFC) level and function in
What Is Relevant?
•?Preeclampsia is a pregnancy-related disorder associated with increased
cardiovascular risk for the offspring. ECFCs participate in the formation
of new vasculature and the maintenance of vascular integrity; thus, an
impaired ECFC level during pregnancy may contribute to an increased
risk of developing postnatal cardiovascular events.
Cord blood ECFC function is normal and highly similar between
preeclampsia and control. However, ECFC level is significantly
decreased in preeclampsia. This reduction in ECFC abundance is
independent of other obstetric characteristics, including gesta-
tional age and maternal body mass index. Further studies should
examine whether a reduced level of cord blood ECFCs correlates
with elevated risk of developing subsequent cardiovascular events,
such as stroke and hypertension.
Novelty and Significance
at Harvard University on April 22, 2014http://hyper.ahajournals.org/Downloaded from
Decreased Level of Cord Blood Circulating Endothelial Colony-Forming Cells in Preeclampsia
Rocio Muñoz-Hernandez 1,2, Maria L. Miranda 1, Pablo Stiefel 1, Ruei-Zeng Lin 2,3, Juan M. Praena-
Fernández 4, Maria J. Dominguez-Simeon 1, Jose Villar 1, Rafael Moreno-Luna 1,2,3 and Juan M. Melero-
1 Instituto de Biomedicina de Sevilla (IBiS), Hospital Universitario Virgen del Rocío/CSIC/Universidad
de Sevilla, Unidad Clínico-Experimental de Riesgo Vascular (UCAMI-UCERV), Seville, Spain
2 Department of Cardiac Surgery, Boston Children's Hospital, Boston, MA
3 Department of Surgery, Harvard Medical School, Boston, MA
4 Unidad de Asesoría Estadística, Metodología y Evaluación de Investigación. Fundación Pública
Andaluza para la Gestión de la Investigación en Salud de Sevilla (FISEVI). Hospital Universitario
Virgen del Rocío, Instituto de Biomedicina de Sevilla (IBiS), Seville, Spain
5 Harvard Stem Cell Institute, Harvard University, Cambridge, Massachusetts, USA
Juan M. Melero-Martin, Ph.D.
Department of Cardiac Surgery
Boston Children’s Hospital
300 Longwood Ave., Enders 349
Boston, MA 02115
Tel.: (617) 919-3072
Fax: (617) 730-0235
Rafael Moreno-Luna, Ph.D.
Vascular Pathophysiology Laboratory
National Hospital Paraplegic, SESCAM
Finca la Peraleda s/n
45071 Toledo, Spain
Running title: Neonatal endothelial progenitors in preeclampsia
EXPANDED MATERIAL AND METHODS
The following maternal and neonatal data were obtained from the obstetric records: severity of
preeclampsia (mild/severe/HELLP syndrome); time of onset of preeclampsia (early/late); maternal age;
mode of delivery (cesarean/vaginal delivery); mode of conception (natural/in vitro fertilization); parity
(primipara/multipara); offspring sex; offspring birth weight; maternal height; pre-gestational (6-8 weeks
gestation) maternal weight; end-of-pregnancy (right before delivery) maternal weight; pre-gestational
blood pressure; and blood pressure at the onset of preeclampsia. Brachial systolic and diastolic blood
pressures were measured on the right arm with a handheld analog device operated by trained nursing
staff. Patients were lying in supine position with a bed slope of 45° (semi-sitting) for 10 min. Three
blood pressure readings were taken at 5 min intervals and the mean was used for data analysis.
Gestational age was recorded according to the obstetricians’ best estimate of gestation. Maternal BMI
was calculated as the weight in kilograms divided by the square of the height in meters (kg/m2).
Gestational weight gain was calculated as the difference between the weight at the end of pregnancy and
the weight at first consultation. Newborn birth weight percentiles were calculated according to World
Health Organization (WHO) standards. Maternal pre-pregnancy tobacco use was dichotomized into
never- and ever-users.
Enumeration of endothelial colony-forming cells
Umbilical cord blood samples (20–50 mL) were collected ex utero using heparinized tubes and
processed within 2 hours. Mononuclear cells (MNC) were obtained and cryopreserved as previously
described 1,2. Cryopreserved mononuclear cells were thawed, thoroughly washed, and cultured on
fibronectin-coated 6-well tissue culture plates (BD Bioscience, San Jose, CA, USA) using endothelial
cell-medium (EGM-2 without hydrocortisone, Lonza; 20% FBS; 1X glutamine-penicillin-streptomycin)
3. Unbound cells were removed at 48 hours and the bound fraction maintained in endothelial cell-
medium, with media being replenished every 2-3 days. Endothelial colonies were identified as well-
circumscribed monolayers of ≥ 50 cells with cobblestone morphology. Colonies were enumerated on
days 7, 14, 21, and 28 by visual inspection using an inverted microscope.
Characterization of endothelial colony-forming cells
Colonies were incubated for 20 min with fluorescently labeled Ulex Europaeus Agglutinin type 1 lectin
(UEA-1; 1:200) to corroborate their endothelial nature. ECFCs were expanded in culture and purified by
expression of CD31, as we have previously shown 3. Endothelial cell phenotype was characterized in
vitro by testing: (1) expression of endothelial cell markers, (2) cloning-forming ability, (3) proliferation
and migration towards VEGF and FGF-2, and (4) formation of capillary -like structures, using methods
previously described by our laboratory 2,4. The vasculogenic ability of ECFCs was evaluated in vivo
using a previously developed xenograft model of human endothelial cell transplantation into
immunodeficient mice 3,5. Animal experiments were conducted under a protocol approved by the
Institutional Animal Care and Use Committee at Boston Children’s Hospital.
Lin R-Z, Dreyzin A, Aamodt K, Dudley AC, Melero-Martin J. Functional endothelial progenitor cells from
cryopreserved umbilical cord blood. Cell Transplant. 2011;20:515–522.
Moreno-Luna R, Munoz-Hernandez R, Lin R-Z, Miranda ML, Vallejo-Vaz AJ, Stiefel P, Praena-Fernandez
JM, Bernal-Bermejo J, Jimenez Jimenez LM, Villar J, Melero-Martin JM. Maternal Body-Mass Index and
Cord Blood Circulating Endothelial Colony-Forming Cells. J Pediatr. 2014;164:566–571.
Melero-Martin J, Khan ZA, Picard A, Wu X, Paruchuri S, Bischoff J. In vivo vasculogenic potential of
human blood-derived endothelial progenitor cells. Blood. 2007;109:4761-4768.
Lin R-Z, Moreno-Luna R, Zhou B, Pu WT, Melero-Martin J. Equal modulation of endothelial cell function
by four distinct tissue-specific mesenchymal stem cells. Angiogenesis. 2012;15:443-455.
Melero-Martin JM, De Obaldia ME, Kang SY, Khan ZA, Yuan L, Oettgen P, Bischoff J. Engineering robust
and functional vascular networks in vivo with human adult and cord blood-derived progenitor cells. Circ
Table S1. Cord blood ECFC levels in control and preeclampsia groups: analysis of categorical obstetric variables
Control (n=35) Preeclampsia (n=15)
P Value YES NO
In vitro fertilization
Intrauterine growth restriction
ECFC levels are represented by mean ± SD. P values are from comparison of YES and NO subgroups within control and preeclampsia groups
(Mann–Whitney U). ECFC = endothelial colony-forming cell.
Table S2. Cord blood ECFC levels in control and preeclampsia groups: analysis of quantitative obstetric variables
Obstetric variable n
P Value n
Age 35 -.05 .77 15 .50 .06
35 .16 .35 15 .07 .81
Gestational weight gain 35 .30 .08 15 -.28 .31
Pre-gestational blood pressure
Diastolic 27 -.10 .61 15 .18 .51
Systolic 27 -.40 .04 15 .02 .94
Blood pressure at onset of PE
Diastolic - - - 15 -.18 .52
Systolic - - - 15 -.31 .27
Gestation age 35 -.21 .24 15 -.37 .17
Birth weight 35 -.03 .87 15 -.31 .27
Birth weight percentile 35 -.07 .71 15 -.31 .26
Cord blood MNC level 29 .01 .97 15 -.02 .95
Univariate correlations between each obstetric variable and ECFC levels were examined with Spearman rho test. n = number of subjects
analyzed for each obstetric variable. r = Spearman correlation co-efficient. P = statistical significance for each correlation. ECFC =
endothelial colony-forming cell; BMI = body-mass index; PE = preeclampsia; MNC = mononuclear cell.