(Circulation. 1997;96:3561-3569.)
© 1997 American Heart Association, Inc.
Articles |
In Utero Cardiac Gene Transfer via Intraplacental Delivery of Recombinant Adenovirus
From the Departments of Surgery (Y.J.W., T.J.G.), Medicine (G.P.R., J.L.S., M.E.R.), and Neuroscience (R.J.B.-G.), University of Pennsylvania School of Medicine (Philadelphia). Dr Swain is now at the Department of Medicine, Stanford University Medical Center, S-102, Stanford, CA 94305-5109.
Correspondence to Rita J. Balice-Gordon, PhD, Department of Neuroscience, University of Pennsylvania School of Medicine, 215 Stemmler Hall, Philadelphia, PA 19104-6074. E-mail rbaliceg{at}mail.med.upenn.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background The relationship among the maternal, placental, and uniquely shunted embryonic circulation was explored to provide access to the embryonic cardiovascular system in utero. Manipulation of gene expression in the developing heart would be particularly useful for studying the effects of altered gene expression on cardiac development and in the etiology of congenital cardiac anomalies.
Methods and Results Dye studies demonstrated that intraplacental
injection allows direct access to the embryonic cardiac and systemic
circulation. To evaluate the efficacy of cardiac gene transfer using
this approach, replication-deficient recombinant adenoviral vectors
encoding luciferase or ß-galactosidase as reporter genes were
injected intraplacentally into embryonic day (E)12.5 murine embryos, an
age at which the mass of the heart was observed to be large compared
with other organs. Embryos were assayed for transgene expression at
E15.5 and at birth. Survival rates at these times were similar among
vector-injected and control groups. At E15.5 and at birth, luciferase
activity within the heart was 9- and 23-fold higher, respectively, than
in the remainder of the embryo, although levels of expression were
generally lower at birth than during embryonic life. ß-Galactosidase
expression was observed within all regions of the embryonic heart and
was localized to 
15% of atrial and ventricular
cells.
Conclusions Intraplacental delivery of adenovirus at embryonic day 12.5 results in somatic gene transfer to the murine embryonic heart, which persists at least until birth. The combination of intraplacental injection to directly access the fetal coronary circulation and injection at E12.5 when the mass of the heart is large compared with other organs results in transgene expression in cardiac cells. Intraplacental injections early in embryonic life may thus be useful to study the effects of temporal manipulation of gene expression on cardiac development and disease.
Key Words: genes • viruses • heart defects, congenital • cardiovascular diseases
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
A variety of congenital cardiac diseases have begun to be characterized by spatial and temporal alterations in gene expression or function during early development that lead to profound changes in subsequent cardiac function.1 Current approaches for studying the effects of altered gene expression on cardiac development rely primarily on costly and time-consuming transgenic technology. The ability to alter gene expression at any point during embryonic development and then study the effects on subsequent development provides a new and powerful tool for studying cardiogenesis that may provide new insights into the treatment of congenital heart disease.
Because of their high efficiency of gene transfer, replication-deficient recombinant adenoviruses have been studied extensively in mammalian systems. Adenoviral-mediated gene transfer into cardiomyocytes has been demonstrated in vitro,2,3 and limited cardiac gene transfer in vivo has been achieved in neonatal and adult animals via several approaches.3–9 The persistence of transgene expression in these and other adult animal models has been shown to be limited by the immune system of the animal,10 which may be mitigated by the ability to induce immunotolerance to alloantigens.11 Adenovirally mediated somatic gene transfer has been used to target the pulmonary epithelium to develop an in utero therapy for cystic fibrosis. Attempts to deliver adenoviral vectors via intra-amniotic injection in ovine and rodent models produced primarily cutaneous and proximal respiratory and alimentary tract transgene expression.12–14 Ovine fetal intratracheal vector delivery yielded proximal airway transgene expression.14,15 Cardiac transgene expression was not detected with either of these delivery methods.
In this study, the relationship between the placental circulation and the uniquely shunted embryonic circulation was used to directly access the embryonic cardiovascular system in the hope of transiently circumventing the systemic circulation to achieve greater viral delivery to the myocardium. We focused on delivery at E12 to E13 because at this stage, the mass of the heart is large compared with other organs; thus, viral transduction in the heart might be relatively greater than that in the liver and spleen at this age. Replication-deficient recombinant adenoviruses encoding the RSV promoter/enhancer driving the luciferase or ß-galactosidase reporter transgenes were delivered intraplacentally to gestational age E12.5 mouse embryos. Luciferase was used as the reporter in most experiments because of the absence of endogenous activity in mammalian tissue, consistency of assay conditions between experiments, and ability to specifically quantify transgene expression. Transgene expression was then evaluated at either E15.5 or postnatal day (p) 1 after spontaneous term delivery. Intraplacental delivery of recombinant adenovirus resulted in transgene expression primarily in the embryonic heart that was observed to persist postnatally. Transgene expression was observed in all cardiac regions and was localized to 14% of cardiomyocytes in the atria and 19% of cardiomyocytes in the ventricles. Thus, the combination of intraplacental injection to access the fetal coronary circulation directly and injection at E12.5 when the mass of the heart is large compared with other organs results in transgene expression in cardiac cells. The intraplacental approach described here to effect adenovirally mediated in utero gene transfer at early embryonic stages may be useful for studying the role of selected proteins in cardiac development and may provide a basis for the development of new therapies for congenital heart disease.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Construction and Propagation of Recombinant Adenoviral Expression Vectors
An adenoviral expression vector was constructed encoding firefly luciferase16 driven by the RSV promoter/enhancer. The recombinant vector AdRSVLuc was generated via homologous recombination with an E1A-deleted, replication-deficient human serotype 5 adenovirus17 (courtesy of Dr Brent French, Houston, Tex). AdRSVLacZ, encoding the ß-galactosidase reporter under the control of the RSV promoter, was similarly generated (courtesy of Dr James Wilson, Philadelphia, Pa). AdRSVLuc and AdRSVLacZ were then propagated in transformed human embryonic kidney 293 stem cells, which are permissive for viral replication.18
The vectors were subsequently purified by ultracentrifugation on two discontinuous cesium chloride density gradients, desalted by column chromatography, and then resuspended in and dialyzed against 3% sucrose in PBS (pH 7.4).19 Viral particle density was measured spectrophotometrically at OD260, and pfu density was determined by plaque assay on agar overlays of 293 cell monolayers. Luciferase and ß-galactosidase reporter transgene expression were then assessed in vitro by infecting subconfluent HeLa cells with varying titers of AdRSVLuc or AdRSVLacZ and, after 3 days, measuring luciferase bioluminescence activity or evaluating ß-galactosidase expression by X-gal staining, respectively.
Because of the potentially detrimental effects that replication-competent adenovirus might exert on embryonic development and subsequent viability, the absence of replication-competent wild-type adenoviral contaminant in the AdRSVLuc and AdRSVLacZ preparations was confirmed by PCR using primers for the deleted E1A region.20 Primers for the E2A region were used as positive controls, and the analyses were simultaneously performed on replication-competent wild-type dl7001 adenoviral DNA. PCR products were analyzed by agarose gel electrophoresis. No wild-type adenoviral DNA was detected in the AdRSVLuc or AdRSVLacZ preparations with PCR. The absence of replication-competent wild-type adenoviral contaminant in the vector preparations was further confirmed in A549 human lung carcinoma cells incubated with 5x107 pfu of virus/150-mm and then serially passaged three times. These cells were readily lysed when infected with wild-type adenovirus and were unaffected by the replication-deficient vectors AdRSVLuc and AdRSVLacZ.
Intraplacental Delivery of Adenoviral Vector
By convention, the day of vaginal plug discovery was designated
E0.5.21 Gravid C57/BL6J mice at E12.5 (term=E19.5 to E20.5)
were anesthetized with intraperitoneal
ketamine (1.5 mg/kg) and xylazine (15 mg/kg) in
0.9% NaCl. Terbutaline (0.5 mg/kg) in 0.9% NaCl was
administered subcutaneously to diminish uterine
contractility. With the use of a stereo microscope
(Leica MZ8) and fiber optic illumination, a limited low midline
laparotomy was performed, and both uterine horns were externalized.
Heat-pulled glass micropipettes (Sutter Instrument Co) with tip diameters of <50 µm were connected to a pneumatic microinfusion pump (WPI) and used to deliver 3 µL containing 5x107 pfu of AdRSVLuc or AdRSVLacZ or vehicle (3% sucrose in PBS) into each placenta at 5 psi of pressure. An injection volume of 3 µL was selected on the basis of two criteria. First, this volume could be easily injected without causing tissue distortion or swelling; second, it was a convenient volume to measure in a small glass micropipette. The uterus was then returned to the abdomen, which was then closed, and the mother was allowed to recover. In an initial series of pilot experiments, the technical aspects of the procedure were refined. All experiments were performed in accordance with institutional guidelines.
A limited pilot study established a working dose of AdRSVLuc based on
embryonic survival and quantitative transgene expression. Three
doses—5x10,6 5x107, and 5x108
pfu—were evaluated. Luciferase activity (see below) in embryos
receiving 5x106 pfu was 2 orders of magnitude less than
those receiving 5x107 pfu. Survival rates were 61% (17 of
28) at 5x106 pfu, 94% (16 of 17) at 5x107
pfu, and 8% (1 of 12) at 5x108 pfu. The low survival rate
of embryos injected with 5x108 pfu was observed in a
single pregnant female mouse and was not explored further, given that
embryos injected with 5x107 pfu showed high levels of
survival and transgene expression. The mechanism of viral toxicity at
5x108 pfu was not evaluated. The dose of
5x107 pfu was used in all subsequent experiments.
Measurement of Heart and Liver Mass
To evaluate the mass ratio of the heart compared with other
organs, the heart and liver of E12.5 (n=9 embryos), E15.5 (n=12), and
E18.5 (n=7) embryos, P1 neonates (n=9) and adults (n=9) were dissected,
blotted to remove excess fluid, and weighed.
Analysis of Transgene Expression
A total of 261 murine embryos received an injection of either
AdRSVLuc or vehicle at E12.5. Three days after injection, a repeat
laparotomy was performed on the mothers of 199 of the embryos. The
viability of the embryos was confirmed by the presence of a visibly
beating heart and normal gross morphology. The 3-day time point was
chosen because in vivo experiments in adult animals suggest that this
time period provides maximal transgene expression.4 Viable
embryos and their placentas, as well as a portion of the maternal
liver, were removed to assay for luciferase activity.16
Specimens were blotted to remove excess fluid, weighed,
homogenized in a buffer consisting of 100
mmol/L KPO4 and 1 mmol/L dithiothreitol,
freeze-thawed three times, and centrifuged at 15{ths}000
rpm for 10 minutes. Then, 50 µL of supernatant was assayed for
luciferase activity in a reaction mixture containing 250
µmol/L D-luciferin (Boehringer-Mannheim),
10 mmol/L MgCl2, and 5 mmol/L ATP
in 350 µL of buffer at 20°C. Light output was measured with a
luminometer (Analytical Luminescence Laboratory), and purified firefly
luciferase (Boehringer-Mannheim) was used as a positive control
and to construct a standard curve for quantification of luciferase
activity in tissue. Results were expressed as picograms of luciferase
per gram of tissue to permit comparison of the degree of transgene
expression among different tissues.
To evaluate normal progression of development in manipulated embryos and the persistence of transgene expression in the early postnatal period, 62 of the embryos receiving AdRSVLuc or vehicle were allowed to proceed to term. Normal spontaneous vaginal delivery ensued, and live-born neonates were observed for viability and assayed on postnatal day P1 for transgene expression.
Localization of Transgene Expression
To determine the site of transgene expression within embryos, in
60 of the AdRSVLuc-injected embryos alive at E15.5, the heart, lungs,
and liver were individually dissected; the head was separated from the
body; and each of these components was assayed for transgene expression
by luciferase bioluminescence assay. Twelve of the live-born neonates
that had received AdRSVLuc were similarly dissected and assayed.
To determine the cellular localization of luciferase, immunohistochemical localization using a rabbit polyclonal anti-luciferase antibody (Cortex Biochem) was attempted in a variety of embryonic tissues. Fluorescent and biotin conjugated IgG followed by fluorescent strepavidin was used to visualize sites of primary antibody binding. A number of different fixation, permeabilization, blocking, and incubation conditions were explored. Cultures of HeLa cells infected with AdRSVLuc were used as positive controls. Despite this effort, it was not possible to detect a reproducible signal in the heart and liver of injected embryos due to lack of specificity of the primary antibody. Thus, virally mediated expression of ß-galactosidase followed by X-gal histochemistry was used to evaluated transgene expression at the cellular level.
Twenty embryos received an intraplacental injection of 5x107 pfu of AdRSVLacZ at E12.5. Three days later, embryonic organs were dissected as described above, and all components were fixed in 2% paraformaldehyde and 0.1% glutaraldehyde in PBS (pH 7.4) for 30 minutes at 4°C, washed twice with PBS, and stained overnight at 37°C in X-gal solution consisting of 2 mmol/L 5-bromo-4-chloro-3-indolyl-ß-D-galactopyranoside, 4 mmol/L potassium ferricyanide, 4 mmol/L potassium ferrocyanide, and 1 mmol/L MgCl2 in PBS (pH 7.4).22 Tissues were photographed, further fixed in 4% paraformaldehyde and 0.2% glutaraldehyde in PBS, cryoprotected in 20% sucrose in PBS, and frozen in an embedding compound (OCT; Miles Laboratory) on a dry ice–ethanol slurry. Then, 12-µm sections were obtained and counterstained with H&E.
Statistical Analysis
Statistical significance was evaluated with the unpaired
two-tailed Student's t test for heart to liver mass ratios
and survival rates. The paired two-tailed Student's t test
was used to evaluate statistical significance in the quantitative
transgene localization studies.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Intraplacental Injection Provides Access to the Embryonic Cardiovascular System
Previously reported approaches to accomplish embryonic gene transfer used, among other methods, intra-amniotic delivery, which provided no access to the cardiovascular system.12–14 To determine whether intraplacental injection would provide effective access to the embryonic cardiovascular system, the placentas of E12.5 embryos (n=15) were injected with a 1% solution of the opaque dye Fast Green (Sigma Chemical) (Fig 1A
and 1B).
Hysterotomy was performed immediately after dye injection (Fig 1C), and
the embryo and placenta were dissected and examined for the
distribution of dye staining. The placenta, umbilical vein, and
embryonic heart were visibly stained. Although the remainder of the
embryo displayed diffuse staining of lesser intensity, several blood
vessels exhibited staining, most notably in the head, indicating that
dye had entered the systemic circulation (Fig 1D). The amniotic fluid
was typically clear in appearance, suggesting that little or no dye was
injected directly into the amniotic sac. However, some variability in
the degree of embryonic staining was noted in early experiments. A deep
injection into the placenta tended to result in dye infusion into the
amniotic cavity, whereas a shallow injection resulted in only placental
staining without dye entry into the embryonic circulation. These dye
studies thus confirmed that intraplacental injection provides a route
of vascular access for vector delivery to the embryonic heart.
|
Decline in the Mass Ratio of Heart Compared With Liver From E12.5
to E15.5
The dye studies described above were performed at E12 to E13, ages
at which it was apparent that the size of the heart was large compared
with other organs, such as the liver. The mass of the heart was
compared to the mass of the liver at embryonic, neonatal, and adult
ages (Fig 2). The mean mass of the heart
at E12.5 was 1.8±0.09 mg compared with 2.7±0.18 mg for the liver
(P<.001), a ratio of 0.69±0.05. This ratio was observed to
decline significantly by E15.5, when the mass of the heart was 3.2±0.4
mg compared with 22.9±3.7 mg for the liver, a ratio of 0.18±0.03
(P<.001). The heart-to-liver ratio was observed to decrease
during the remainder of embryonic life, declining to a ratio of
0.11±0.01 at 3 months of age. This analysis shows that at
E12.5, the mass of the heart is significantly greater than that of the
liver. The largest decrease in the heart-to-liver ratio was observed
between E12.5 and E15.5, reflecting the rapid growth of the liver
during this period. Thus, E12.5 was selected as an age at that to
evaluate the efficacy of intraplacental viral delivery to target the
heart.
|
Embryonic and Term Survival After Intraplacental Injection
To determine whether intraplacental delivery of adenovirus
affected subsequent embryonic development and viability, 261 embryos
were injected at E12.5 with 3 µL of 5x107 pfu of
recombinant adenovirus using the RSV promoter to drive expression of
the reporter gene luciferase (AdRSVLuc; n=183) or vehicle (n=78).
Embryos were evaluated for viability either 3 days later at E15.5
(n=199) or at term (n=62). At E15.5, 77% (111 of 144) of
AdRSVLuc-injected embryos and 82% (45 of 55) of vehicle-injected
embryos were alive (P=NS). Viability was determined by the
presence of a visibly beating heart and embryonic size and gross
morphology comparable to those of age-matched, nonmanipulated embryos.
Adenoviral infection did not affect embryonic viability at 3 days after
injection, and the observed rates of embryonic demise are within the
reported range of normal gestational loss during murine
development.23,24
Among 23 embryos injected with vehicle, 17 neonates (74%) were born; among 39 embryos injected with AdRSVLuc, 26 neonates (67%) were born. These delivery rates were similar (P=NS) and comparable to the E15.5 survival rates (P=NS), suggesting no decrease in the ability of adenovirus-infected embryos to proceed to normal term delivery. During the first postnatal day, neonates were observed to be active, healthy in appearance, and nursed by their mothers.
Embryonic, Placental, and Maternal Transgene Expression
To evaluate the extent of gene transfer and transgene expression
at E15.5, 95 embryos receiving 3 µL of 5x107 pfu
AdRSVLuc and 29 embryos receiving vehicle were harvested, and transgene
expression was evaluated using a luciferase bioluminescence assay. The
embryos injected with vehicle exhibited background luciferase activity
(
1 pg luciferase/g of tissue). Twenty-one percent (20 of 95) of
embryos exhibited transgene expression in the range of 1 to 10 pg
luciferase/g of tissue, 42% (40 of 95) expressed 10 to 100
pg/g, and an additional 21% (20 of 95) expressed 100 to 1000
pg/g (Fig 3A). Four percent (4 of
95) of embryos expressed luciferase in excess of 10 ng/g. Thus,
the intraplacental route of adenovirus delivery resulted in substantial
gene transfer to the embryo. Intraplacental adenovirus injection
resulted in a wide range of transgene expression across embryos at both
of these ages. Given that some variability in staining was observed in
the dye injection experiments depending on the localization of the
injection, differences related to the mechanical aspects of injection
probably account for the range of transgene expression observed.
|
Placentas were similarly evaluated for transgene expression. Fifty-six
percent (53 of 95) of the placentas exhibited luciferase activity in
the range of 100 to 1000 pg luciferase/g of tissue (Fig 3B
). The
luciferase activity in the 29 control placentas was at background
levels (1 pg/g). Given that the placenta was the site of viral
injection, it is not surprising that placental luciferase activity was
higher than that observed in the embryo. However, no correlation
between embryonic and placental expression was observed (n=95 embryos
and placentas; r=.00).
To determine what portion of the adenovirus injected into the placenta
entered the maternal circulation, maternal livers were evaluated for
luciferase activity. Eighty-three percent (15 of 18) of maternal livers
were observed to express the transgene in the range of 1 to 10 pg
luciferase/g of tissue, normalized to the number of embryos in each
mother receiving AdRSVLuc (Fig 3C). Maternal hepatic luciferase
activity was 2 orders of magnitude lower than that observed in the
placenta, suggesting minimal vector leakage from injected placentas
into the maternal circulation. There was no correlation between
placental and maternal liver expression levels (n=95 placentas and 18
maternal livers; r=.03). The correlation between embryonic
expression and maternal liver expression was higher (r=.51),
perhaps reflecting a higher likelihood of virus entering the maternal
circulation when the embryonic circulation was successfully
accessed.
Persistence of Transgene Expression in Neonates
To evaluate the persistence of transgene expression, term neonates
were assayed for luciferase activity. Twenty-six neonates injected with
AdRSVLuc at E12.5 and 17 neonates injected with vehicle at E12.5 were
carried to term and assayed for transgene expression on P1. Total
neonatal transgene expression of animals receiving AdRSVLuc varied over
a wide range (Fig 4) similar to that seen
with embryos assayed at E15.5 (Fig 3A). Control neonates injected with
vehicle exhibited background luciferase activity of 1 pg luciferase/g
of tissue.
|
To assess stability of transgene expression, luciferase activity in P1
neonates was compared with that of E15.5 embryos by eliminating as a
factor the significant increase in mass during late gestation. Thus,
the total luciferase activity expressed as picograms per embryo or
neonate was compared (Fig 5). Among the
E15.5 embryos, 32% (30 of 95) expressed in the range of 1 to 10 pg,
and 38% (36 of 95) expressed 10 to 100 pg. Among the P1 neonates, 25%
(6 of 24) expressed 1 to 10 pg, and 42% (10 of 24) expressed 10 to 100
pg. With the exception of the few exhibiting high transgene expression
in the embryonic group, these results suggest stable expression of the
luciferase transgene, at least during the 8- to 9-day period we
evaluated.
|
Localization of Transgene Expression to the Developing
Heart
The distribution of transgene expression within the embryo was
determined by individually assaying the embryonic organs, head, and
body at E15.5 in 60 viable embryos that receiving AdRSVLuc at E12.5.
Transgene expression was observed in all parts of the embryo (Fig 6A), with 3.8±1.1 pg of luciferase in
the heart and 10.1±2.1 pg in the liver. Absolute levels of transgene
expression were highest in the body and liver. Comparison among
embryonic components on a per–gram of tissue basis (Fig 6B) showed
that the highest level of transgene expression per gram of tissue was
observed in the heart (1650±550 pg luciferase/g of tissue) and liver
(610±190 pg luciferase/g of tissue). Transgene expression in the
embryonic heart was 2.7-fold higher than in the liver
(P<.05), 13.7-fold higher than in the lungs
(P<.005), 14.5-fold higher than in the head
(P<.005), and 12.2-fold higher than expression in the
remainder of the body (P<.01). Overall, there was 9.1-fold
higher transgene expression in the heart compared with the total embryo
(P<.01). These results demonstrate that the intraplacental
route of vascular delivery of adenovirus to the E12.5 murine embryo
results in significantly higher cardiac transgene expression compared
with other embryonic tissues.
|
To evaluate postnatal cardiac expression, the distribution of transgene
expression within 12 neonates was assayed in the head, heart, lungs,
liver, and remainder of the body. Luciferase activity in the neonatal
heart was 6.6±2.7 pg compared with 17.0±6.9 pg in the liver (Fig 7A). Absolute levels of luciferase
activity were observed to be higher than in E15.5 embryos. When
normalized for individual tissue mass, luciferase activity was found to
be 490±210 pg/g in the heart compared with 21±8 pg/g in
the remainder of the neonate (P<.05) (Fig 7B). Cardiac and
hepatic transgene expressions were observed to be markedly higher than
in other neonatal tissues (Fig 7B). Transgene expression in the
neonatal heart was 1.6-fold higher than in the liver with near
statistical significance (P<.07), 23.4-fold higher than in
the lungs (P<.05), 118-fold higher than in the head
(P<.05), and 349-fold higher than in the body
(P<.05). The neonatal cardiac expression
represented a 23.6-fold enhancement of transgene expression
compared with the total neonate (P<.05). Thus,
substantially higher cardiac and hepatic relative transgene expression
was observed at term compared with the rest of the neonate.
|
Localization of Transgene Expression Within the Developing
Heart
The data presented above demonstrate that luciferase
activity is localized primarily to the embryonic heart after
intraplacental injection of AdRSVLuc. Due to the lack of a highly
specific antibody against luciferase (see "Methods"), cellular
localization of transgene expression was conducted using an adenovirus
encoding the reporter ß-galactosidase AdRSVLacZ. Intraplacental
delivery of 3 µL of 5x107 pfu AdRSVLacZ was performed at
E12.5 (n=20 embryos). Twelve viable embryos (60%) were harvested at
E15.5. In 5 of the 12 embryos, X-gal staining of the intact heart
revealed a diffuse pattern of transgene expression throughout all
regions of the heart (Fig 8A). X-gal
staining of other intact organs showed diffuse expression throughout
the liver, expression in the placenta localized around the site of
viral injection, and minimal detectable expression in the embryonic
lungs, head, and body (data not shown). These qualitative results are
consistent with the quantitative localization of luciferase
activity in E15.5 embryos.
|
ß-Galactosidase expression in three of the five positive-staining
hearts was further evaluated with frozen sections. Transgene expression
was observed in the atria and atrial appendages (Fig 8B) and in the
ventricular septum and biventricular
endocardium and subendocardium (Fig 8C). The number of lacZ-positive
cells was quantified in the atria and ventricles, with a mean of 14%
of the cells in the atria (n=3, range, 3% to 27%) and a mean of 19%
in the ventricles (n=3, range, 6% to 37%) expressing the transgene,
In embryonic liver, a mean of 14% of hepatocytes expressed
the trangene (n=3, range, 9% to 17%). Examination at higher
magnification after counterstaining with H&E demonstrated transgene
expression within cardiomyocytes (Fig 8D). Thus, the
qualitative localization of embryonic transgene expression using
AdRSVLacZ revealed an expression pattern consistent with the
quantitative results observed using AdRSVLuc. Furthermore, the
ß-galactosidase transgene was expressed throughout the embryonic
heart and was observed within individual
cardiomyocytes.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This study demonstrates that the intraplacental injection of replication-deficient recombinant adenovirus at E12.5 results in robust gene transfer to the developing mouse embryo. Gene transfer was found predominantly in the embryonic heart, with transgene expression localized to cardiomyocytes. The technique of in utero intraplacental somatic gene transfer at the early embryonic age reported here permits spontaneous delivery of normal full-term neonates with preservation of the transgene expression pattern observed during embryonic life.
Through use of the relationship between the placenta and the embryonic
circulation, a straightforward means of access to the embryonic
cardiovascular system was established. During embryonic
development, nutrient-rich oxygenated blood returning from
the placenta via the umbilical vein bypasses the liver via the ductus
venosus and bypasses the lungs via the foramen ovale and ductus
arteriosus. The newly formed coronary circulation25
is the first capillary bed encountered; thus, blood is effectively
shunted from the placenta directly to the myocardium and
subsequently enters the systemic embryonic circulation. Although
adenovirus traveling via this circulatory pathway will initially bypass
the liver, a significant portion will still obtain access to the liver
once in the systemic circulation, and this likely accounts for the
substantial hepatic expression observed. Intraplacental injection of an
opaque dye resulted in intense staining of the placenta, umbilical
vein, embryonic heart, and blood vessels in the head, confirming access
to the embryonic circulation (Fig 1D). However, the distribution of dye
may not accurately predict the distribution of adenovirally mediated
transgene expression, given obvious differences in mass,
hydrophobicity, and other factors (eg, viral receptor number) governing
viral access and transgene expression.
The extent of cardiac transgene expression observed in this study exceeds that seen in previous studies of intravenous administration of adenovirus in neonatal and adult animals.6,9 A significant contributing factor to the cardiac expression we observe is likely due to the size of the heart compared with the liver at E12.5. The heart-to-liver mass ratio is 0.69 at E12.5 and is decreased dramatically by E15.5 to 0.18, decreasing further by adulthood to a ratio of 0.11. If the relative size of organs were a contributing factor to transgene expression, the change in relative cardiac size suggests that delivery later in embryonic life would not have resulted in efficient cardiac gene transfer. Delivery of AdCMVlacZ to the fetal circulation in E12 to E13 mouse embryos resulted in predominant expression of ß-galactosidase in the heart 24 to 72 hours later, whereas delivery at E15 resulted in marked attenuation of cardiac expression compared with the endothelium of other organs.26 In neonatal mice, intravenous injection of AdRSVLacZ has been shown to yield transgene expression in only 0.2% of myocardial cells, with far greater expression in hepatic tissue.6 In adult animals, even with the addition of hepatic vascular exclusion to limit hepatic uptake of systemic adenovirus, <5% of total body transgene activity has been observed in the heart after intravenous injection (Y.J.W. and J.L.S., unpublished observations). Thus, it is unlikely that there is a specific tropism of adenovirus for the heart, an unknown activity of the RSV promoter in cardiac tissue, or some other mechanism to account for the efficient gene transfer to embryonic heart observed in the present study. Taken together, the route of gene transfer via the placental and uniquely shunted embryonic circulation in E12.5 embryos (when the heart-to-liver mass ratio is high) may account for the markedly more efficient cardiac gene transfer than that observed previously.
One aspect of embryonic gene transfer not well examined in previous reports is the extent of maternal transgene delivery and expression. With any in utero method of vector delivery, maternal exposure to vector, subsequent transgene expression, and potential immune sensitization are important issues. Because the liver is known to be the site of most avid uptake of systemic adenovirus,6,9 the degree of maternal hepatic transgene expression was used to measure the extent of vector leakage after placental injection. The results demonstrated that the intraplacental injection of AdRSVLuc yielded transgene expression in the maternal liver of <1% of that seen in the placenta, indicating minimal maternal exposure to the vector.
There are a number of potential applications of an intraplacental route of in utero viral delivery to accomplish cardiac gene transfer. This approach will provide a powerful tool with which to study mammalian cardiac development in vivo, particularly in evaluating the temporal effects of manipulating gene expression. For many developmental and disease processes, the approach described here may prove to be simpler and more versatile than producing transgenic mice. In conjunction with transgenic knockout animals, this method may be useful in attempting phenotype rescue. The use of tissue- and development-specific regulatory elements to direct transgene expression may permit more targeted expression with respect to both the organ distribution and developmental stage of transgene expression. Another application of intraplacental embryonic gene transfer is in the correction of genetic diseases in utero before phenotypic manifestations of abnormalities. This method of in utero gene transfer may have important human applications in the prenatal treatment of inherited genetic diseases and may not be limited to the cardiovascular system. Furthermore, fetal gene transfer may pose a unique opportunity to establish immune tolerance with the hope of achieving indefinite transgene persistence. The intravenous delivery of agents to the human fetus has already been accomplished via ultrasonographically guided umbilical vein injection.27 The rapid advances that are occurring in the in utero manipulation of human fetuses are likely to further enhance the ability to effect fetal gene transfer. Targeted prenatal genetic manipulation may ultimately provide an important therapeutic option in the treatment of congenital diseases of the heart and other organs.
|
Selected Abbreviations and Acronyms |
|---|
|
|
Acknowledgments |
|---|
This work was supported by National Institutes of Health grants R-29-NS-34373 (R.B-G.) and R-37-HL-26831–16 (J.L.S.), an unrestricted research award from Bristol-Myers Squibb (J.L.S.), and a McKnight Neuroscience Scholar award (R.B.-G.). A portion of this work has been presented in abstract form. We thank Francis P. Ruggiero and Daniel B.J. Smith for excellent technical assistance and Dr Mark Rich for helpful discussions and comments on the manuscript.
Received May 20, 1997; revision received July 21, 1997; accepted August 1, 1997.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. Payne RM, Johnson MC, Grant JAW, Strauss AW. Toward a molecular understanding of congenital heart disease. Circulation. 1995;91:494–504.
2. Kirshenbaum LA, McClellan WR, Mazur W, French BA, Schneider MD. Highly efficient gene transfer into adult ventricular myocytes by recombinant adenovirus. J Clin Invest. 1993;92:381–387.
3.
Kass-Eisler A, Falck-Pedersen E, Alvira M, Rivera J,
Buttrick PM, Wittenberg BA, Cipriani L, Leinwand LA. Quantitative
determination of adenovirus-mediated gene delivery to rat cardiac
myocytes in vitro and in vivo. Proc Natl Acad Sci
U S A.. 1993;90:11498–11502.
4.
Guzman RJ, Lemarchand P, Crystal RG, Epstein SE,
Finkel T. Efficient gene transfer into myocardium by direct
injection of adenovirus vectors. Circ Res. 1993;73:1202–1207.
5.
French BA, Mazur W, Geske RS, Bolli R. Direct in
vivo gene transfer into porcine myocardium using
replication-deficient adenoviral vectors. Circulation. 1994;90:2414–2424.
6. Stratford-Perricaudet LD, Makeh I, Perricaudet M, Briand P. Widespread long-term gene transfer to mouse skeletal muscles and heart. J Clin Invest. 1992;90:626–630.
7. Barr E, Carroll J, Kalynych AM, Tripathy S, Kozarsky K, Wilson JM, Leiden JM. Efficient catheter-mediated gene transfer into the heart using replication-defective adenovirus. Gene Ther. 1994;1:51–58.[Medline] [Order article via Infotrieve]
8. Li JJ, Ueno H, Pan Y, Tomita H, Yamamoto H, Kanegae Y, Saito I, Takeshita A. Percutaneous transluminal gene transfer into canine myocardium in vivo by replication-defective adenovirus. Cardiovasc Res. 1995;30:97–105.[Medline] [Order article via Infotrieve]
9. Huard J, Lochmuller H, Ascadi G, Jani A, Massie B, Karpati G. The route of administration is a major determinant of the transduction efficiency of rat tissues by adenoviral recombinants. Gene Ther. 1995;2:107–115.[Medline] [Order article via Infotrieve]
10.
Yang Y, Nunes FA, Berencsi K, Furth EE, Gonczol E,
Wilson JM. Cellular immunity to viral antigens limits E1-deleted
adenoviruses for gene therapy. Proc Natl Acad Sci U S A.. 1994;91:4407–4411.
11. Billingham RE, Brent L, Medawar PB. Actively acquired tolerance of foreign cells. Nature.. 1953;172:603–606.[Medline] [Order article via Infotrieve]
12. Sekhon HS, Larson JE. In utero gene transfer into the pulmonary epithelium. Nat Med.. 1995;1:1201–1203.[Medline] [Order article via Infotrieve]
13. Holzinger A, Trapnell BC, Weaver TE, Whitsett JA, Iwamoto HS. Intra-amniotic administration of an adenoviral vector for gene transfer to fetal sheep and mouse tissues. Pediatr Res. 1995;38:844–850.[Medline] [Order article via Infotrieve]
14. McCray PB Jr, Armstrong K, Zabner J, Miller DW, Koretzky GA, Couture L, Robillard JE, Smith AE, Welsh MJ. Adenoviral-mediated gene transfer to embryonic pulmonary epithelia in vitro and in vivo. J Clin Invest. 1995;95:2620–2632.
15. Vincent MC, Trapnell BC, Baughman RP, Wert SE, Whitsett JA, Iwamoto HS. Adenovirus-mediated gene transfer to the respiratory tract of embryonic sheep in utero. Hum Gene Ther. 1995;6:1019–1028.
16.
deWet JR, Wood KV, DeLuca M, Helsinki DR, Subramani S.
Firefly luciferase gene: structure and expression in mammalian cells.
Mol Cell Biol. 1987;7:725–737.
17.
French BA, Mazur W, Ali NM, Geske RS, Finnigan JP,
Rodgers GP, Roberts R, Raizner AE. Percutaneous
transluminal in vivo gene transfer by recombinant adenovirus
in normal porcine coronary arteries, atherosclerotic arteries,
and two models of coronary restenosis.
Circulation. 1994;90:2402–2413.
18. Kozarsky KF, Wilson JM. Gene therapy: adenovirus vectors. Curr Opin Genet Dev. 1993;3:499–503.[Medline] [Order article via Infotrieve]
19. Graham FL, Prevec L. Methods for the construction of adenovirus vectors. Mol Biotech. 1995;3:207–220.[Medline] [Order article via Infotrieve]
20. Zhang WW, Koch PE, Roth JA. Detection of wild-type contamination in a recombinant adenoviral preparation by PCR. BioTechniques. 1995;18:444–447.[Medline] [Order article via Infotrieve]
21. Kaufman MH. The Atlas of Mouse Development. London, UK: Academic Press; 1992.
22. Sanes JR, Rubenstein JLR, Nicolas JF. Use of a recombinant retrovirus to study post-implantation cell lineage in mouse embryos. EMBO J. 1986;5:3133–3142.[Medline] [Order article via Infotrieve]
23. Muneoka K, Wanek N, Bryant SV. Mouse embryos develop normally exo utero. J Exp Zool. 1986;239:289–293.
24. Rugh R. The Mouse: Its Reproduction and Development. London, UK: Oxford Science Publications; 1968.
25. Viragh S, Challice CE. The origin of the epicardium and the embryonic myocardial circulation. in the mouse. Anat Record. 1981;201:157–168.[Medline] [Order article via Infotrieve]
26. Schachtner SK, Crivello K, Buck CA, Baldwin HS. Gestational age affects tissue expression of in utero gene transfer in the mouse. Circulation. 1996;94(suppl I):I-11. Abstract.
27. Touraine JL, Raudrant D, Royo C, Rebaud F, Barbier F, Roncarolo MG, Touraine F, Laplace S, Gebuhrer L, Betuel H, Frappaz D, Freycon F, Vullo C. In utero transplantation of hemopoietic stem cells in humans. Transplant Proc. 1991;23:1706–1708.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  G. Christensen, S. Minamisawa, P. J. Gruber, Y. Wang, and K. R. Chien High-Efficiency, Long-Term Cardiac Expression of Foreign Genes in Living Mouse Embryos and Neonates Circulation, January 18, 2000; 101(2): 178 - 184. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. S. Lipshutz, R. Sarkar, L. Flebbe-Rehwaldt, H. Kazazian, and K. M. L. Gaensler Short-term correction of factor VIII deficiency in a murine model of hemophilia A after delivery of adenovirus murine factor VIII in utero PNAS, November 9, 1999; 96(23): 13324 - 13329. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Carmeliet, L. Moons, and D. Collen Mouse models of angiogenesis, arterial stenosis, atherosclerosis and hemostasis Cardiovasc Res, July 1, 1998; 39(1): 8 - 33. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||