| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Molecular/Cancer Biology Laboratory, Haartman Institute and Helsinki University Central Hospital and Ludwig Institute for Cancer Research, Biomedicum Helsinki, 00014 Helsinki, Finland; and
* Department of Physiology, Institute of Biomedicine, University of Turku, 20520 Turku, Finland
1Correspondence: Molecular/Cancer Biology Laboratory and Ludwig Institute for Cancer Research, Biomedicum Helsinki, P.O. Box 63 (Haartmaninkatu 8), 00014 University of Helsinki, Helsinki, Finland. E-mail: Kari.Alitalo{at}Helsinki.Fi
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: receptor tyrosine kinase • promoter • angiogenesis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Angiogenesis, the formation of new blood vessels from preexisting ones and from circulating endothelial progenitors, is needed for developmental expansion of the vascular system as well as for tumor growth and metastasis (2)
. Endothelial cells lining the blood and lymphatic vessels are dependent on growth factors and receptor tyrosine kinase-mediated signaling that regulate growth and differentiation responses of the target cells. Vascular endothelial cells express nonlineage-restricted and endothelial cell-restricted receptor tyrosine kinases. Two major families of receptor tyrosine kinases specifically expressed in the vascular endothelium consist of the vascular endothelial growth factor receptors (VEGFRs) and Tie (tyrosine kinase with immunoglobulin and epidermal growth factor homology domains) receptors (3
, 4)
. Two members of the Tie receptor family, Tie1 and Tie2, have been identified, as well as four Tie2 ligands, called angiopoietins 1–4 (5
6
7
; for a review, see ref 3
).
Tie1 is first expressed on embryonic day (E) 8–8.5 in angioblasts of the head mesenchyme, in endothelial cells of the dorsal aorta, and in the blood islands of the yolk sac; expression persists in endothelial cells throughout embryogenesis (8
, 9)
. Tie1 is required cell autonomously for endothelial cell survival and extension of the vascular network during late embryogenesis and adulthood (10
, 11)
. However, Tie1 mRNA expression is down-regulated in many adult tissues after development (8
, 9)
. In adult mice, increased Tie1 mRNA levels are detected in the endothelium at sites of neovascularization associated with corpus luteum formation during the ovulatory cycle and with wound healing (12)
. Increased levels of Tie1 mRNA and protein have been found in some pathological conditions such as in the vasculature of certain human tumors (13
14
15
16
17
18)
, suggesting that Tie1 is needed for tumor angiogenesis.
The promoters of human and mouse Tie1 genes have been isolated and partially characterized. We previously showed that the 0.8 kb AflII-ApaI mouse Tie1 promoter fragment contains all the elements needed to target gene expression specifically to endothelial cells in vivo (19)
. Transgenic mice expressing the LacZ gene under the control of the 0.8 kb Tie1 promoter fragment showed ß-galactosidase activity in endothelial cells during development. However, ß-galactosidase staining could not reveal gene expression in viable endothelium and it was difficult to compare Tie1 promoter activity between different tissues and developmental stages based on the intensity of staining.
Due to its autofluorescence capacity, the jellyfish green fluorescent protein (GFP) has been used recently as a marker in various applications in molecular biology. We report here generation and characterization of transgenic reporter mice expressing enhanced GFP (EGFP) or a recombinant protein consisting of the Zeosin resistance marker and EGFP (ZEGFP) under the control of the mouse Tie1 promoter. We describe purification of endothelial cells from these reporter mice by fluorescence-activated cell sorting; using a transplantable tumor model, we analyzed whether the regulatory elements needed for Tie1 expression in tumor neovasculature are present within the Tie1 promoter.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A.
|
Testing of the Tie1-Z/EGFP constructs in transfected cells
Mouse lung endothelial cells (LE-II) (20)
were grown in minimal essential medium (MEM) and 293T cells were grown in Dulbecco’s modified Eagles’ medium (DMEM) containing 10% fetal calf serum (FCS), glutamine, and antibiotics. The 293T cells were transfected using the FuGENE 6 reagent (Roche Molecular Biochemicals, Nutley, NJ) according to the manufacturer’s instructions. Cells were lysed in PLCLB (50 mM HEPES pH 7.0–7.5, 150 mM NaCl, 10% glycerol, 1% Triton-X, 1.5 mM MgCl2) lysis buffer supplemented with aprotinin, leupeptin, phenylmethylsulfonyl fluoride, and sodium vanadate. Equal amounts of total cell lysates were separated in 15% SDS-PAGE, transferred to nitrocellulose, and detected using a combination of three affinity-purified peptide antibodies against GFP (Living Colors Aequorea victoria peptide antibody, Clontech). The bound antibody was detected using HRP-conjugated swine-anti-rabbit antibodies (Dako A/S, Glostrup, Denmark) and the ECL method (Amersham Life Science, Amersham, UK).
For testing Zeosin resistance of the Tie1-ZEGFP-expressing LE-II cells, nontransfected cells and cells transfected with Tie1-EGFP construct were used as negative controls and pcDNA3.1/Zeor (Invitrogen) -transfected cells were used as a positive control. Two days after transfection, cells were subcultured and the selection medium containing 0.4 mg/mL Zeosin was added. The selection medium was changed every second day and colonies of cells surviving the selection were isolated after 2 or 3 wk by use of cloning rings.
Production and analysis of Tie1-Z/EGFP transgenic mice
The transgenes were isolated by AflII digestion and the transgenic mice were produced by microinjection of zygotes from superovulated FVB/n mice. After microinjections, the zygotes were transferred at one- or two-cell stage into oviducts of pseudopregnant NMRI foster mothers. PCR of genomic DNA was used to identify the transgenic offspring by using primers for EGFP (forward primer: 5'-GCCGACCACTACCAGCAGAACACCCCCATC-3' and reverse primer: 5'-ATTTTATGTTTCAGGTTCAGGGGGAGGTGT-3'). The expression of the transgene was studied at E10.5 by fluorescent microscopy. Age of embryos was estimated from the formation of the copulation plug on day 0.5.
Fluorescence microscopy
To detect Z/EGFP fluorescence from the embryos and adult tissues, Leica MZFLIII microscope and a GFP Plus fluorescence filter set, with excitation filter 480/40 nm and barrier filter 510 nm, were used. Cultured fluorescent cells were studied in the Zeiss Axioplan2 microscope equipped with a Sensicam digital camera (PCO CCD Imaging, Germany).
Generation of primary endothelial cell cultures, immunofluorescence staining, and FACS analysis
Fluorescent E9.5 embryos were dissected mechanically into small pieces and a droplet of trypsin-EDTA was added, followed by 15 min of incubation. The cell suspension was agitated and separated by low-speed centrifugation (1400 rpm, 1 min), after which the cells were resuspended in complete DMEM (containing 10% FCS, glutamine, and antibiotics), MV low serum endothelial cell growth medium (Promo Cell) supplemented with the endothelial cell supplement mix and FCS (4:4:2). Cell clumps were pelleted (800 rpm, 1 min), and subjected to another trypsinization, after which single-cell suspensions were combined. The cells were washed twice in phosphate-buffered saline (PBS) and resuspended into PBS.
Fluorescence-activated cell sorting was done using FACStar cell sorter (Becton Dickinson, San Jose, CA) and Consort 30 program. The number of EGFP-positive cells in the samples was analyzed with LYSYS II program and FACScan sorter (Becton Dickinson). After cell sorting, cells from the positive pool were resuspended to the culture medium and plated on gelatin (0.1%). After overnight culture, nonadherent cells were removed. Media was changed every second day; after the third passage, the gelatin coating was omitted.
For immunofluorescence stainings, cells isolated from E9.5 embryos or cultured after FACS sorting were grown on glass coverslips. After fixation in 3% paraformaldehyde for 5 min and blocking in 5% goat serum, cells were stained with primary monoclonal antibodies recognizing mouse CD31 (MEC13.3, PharMingen, San Diego, CA), VE-cadherin (11D4.1, PharMingen), VEGFR-2 (AVAS12
1, PharMingen), or VEGFR-3 (21)
for 30 min at room temperature. After washings, cells were incubated with tetramethylrhodamine isothiocyanate-conjugated goat-anti-rat IgG secondary antibodies (Jackson Immunoresearch, West Grove, PA). For flow cytometry analysis of CD31 expression, E11.5 embryos were minced into small 1–2 mm pieces and incubated in collagenase A (5 mg/mL; Roche Molecular Biochemicals) for 30 min at 37°C. During incubation, cells were dissociated by pipetting every 10 min to obtain single-cell suspension. Collagenase was removed by washing cells once with DMEM containing 10% FCS, followed by washes with PBS. Approximately 3 x 106 cells were used per sample; after blocking with 1% bovine serum albumin (BSA) for 30 min, cells were incubated in allophycocyanin-conjugated CD31 (1.5 µg/3x106 cells) for 1 h on ice, followed by washes with 1% BSA in PBS.
To immortalize the Tie1-ZEGFP-positive cells from embryos at E12, the cells were infected with a retrovirus encoding the polyoma middle T antigen and neomycin resistance marker (22)
. The virus was obtained from the GgP+E cells (23)
cultured in bicarbonate-buffered DMEM supplemented with high glucose (Life Technologies, Grand Island, NY) and 15% FCS. 1.2 x 106 GgP+E cells on a 10 cm plate were incubated for 24 h in 2 mL of HEPES-buffered DMEM high glucose (Life Technologies) supplemented with 15% FCS. The medium was then collected and centrifuged at 2800 rpm for 15 min. The supernatant was passed through a 0.45 µm filter and polybrene (hexadimethrine bromide; Sigma, St. Louis, MO) was added to a final concentration of 8 µg/mL. Embryonic cells were infected by incubation with this medium for 5 h. The infection was repeated on the following day. The selection with neomycin (800 µg/mL) alone and with various concentrations of Zeosin ranging from 100 µg/mL to 1.5 mg/mL was started 72 h after the infection. Noninfected embryonic cells were used as controls.
RNA isolation and analysis
Total RNA was isolated from mouse tissues and primary cell cultures by using the RNeasy kit (Qiagen, Chatsworth, CA) according to manufacturer’s instructions. For Northern blot analysis, RNA was fractionated in 1% agarose gels containing formaldehyde and blotted onto Hybond-N (Amersham, Arlington Heights, IL). The filters were hybridized in Ultrahyb Hybridization buffer (Ambion, Austin, TX) and washed in stringent conditions. The probes used were a 0.7 kb NotI/BamHI fragment containing the full-length EGFP cDNA, 1 kb ClaI-SalI fragment containing the 5' part from of the LacZ cDNA, 0.4 kb EcoRI fragment of mouse Tie1 cDNA clone ICID (12)
, and 2.4 kb HindIII-ApaI fragment containing the carboxyl-terminal part of human TIE2 cDNA.
For reverse transcription, genomic DNA was removed from RNA using a RNase-Free DNase Set Protocol (Qiagen). Reverse transcription used 300–500 ng of total RNA, Omniscript RT Kit (Qiagen) and random hexanucleotide primers. The EGFP cDNA was then PCR amplified using the same primers as in transgenic mouse screenings; Tie1 cDNA was amplified by using 5'-CTCACTGCCCTCCTGACTGG-3' and 5'-GCATGCAGATTTTCCTCT-3' as primers and the LacZ cDNA by using 5'-CTGGATCAAATCTGTCGATCCTT-3' and 5'-GCTGGATGCGGCGTGCGGT-3' as primers and Dynazyme (Finnzymes, Espoo, Finland) polymerase in a total volume of 25 µL with the following conditions: 94°C for 30 s, 60°C for 30 s, and 72°C for 30 s. All PCR experiments included reverse transcriptase-negative controls and a blank with no template.
ß-Galactosidase staining of tissues
Embryos, whole-mount adult tissues, and tumor samples were fixed in 0.2% glutaraldehyde, 2 mM MgCl2, 5 mM EGTA in 100 mM phosphate buffer pH 7.3 at room temperature for 15–30 min. Tissues were then washed at room temperature in wash buffer (2 mM MgCl2, 0.01% deoxycholate, 0.02% Nonidet P40 in 100 mM phosphate) and incubated in 1 mg/mL X-gal reaction mixture (1 mg/mL 4-chloro-5bromo-3-indolyl-ß-galactosidase, 5 mmol/L K3Fe(CN)6, 5 mmol/L K4Fe(CN)6 in wash buffer) at 37°C overnight. After the staining, samples were fixed overnight with 4% paraformaldehyde in PBS. Tissues were dehydrated through graded ethanols and embedded in paraffin. Sections were cut at 7 µm, mounted onto glass slides, dewaxed, and stained with nuclear fast red.
Lectin staining of blood vessels
Lectin staining was used to visualize blood vessels in the skin of adult mice as described previously (24)
. Biotinylated Lycopersicon esculentum lectin (Sigma; 100 µL of 1 mg/mL solution) was injected intravenously (i.v.) via the femoral vein into anesthetized mice and allowed to circulate for 2 min. The mice were killed and tissues were fixed by intracardial perfusion with 1% paraformaldehyde-0.5% glutaraldehyde in PBS. The ears were dissected, washed in PBS, and the cartilage was removed. Bound lectin was visualized by biotin-peroxidase staining, and the ears were mounted on slides and examined by light microscopy.
Tumor cell culture and transplantation
Lewis lung carcinoma (LLC) cells were grown in DMEM supplemented with 10% FCS, 1% L-glutamine, and antibiotics. For generation of tumors, 
1 x 106 tumor cells in 200 µL were injected subcutaneously (s.c.) on the back of four Tie1-Z/EGFP mice from three different founder lines as well as to three Tie1-LacZ transgenic mice (19)
, three Tie1+/LacZ knock-in mice (25)
, and two wild-type mice. The mice were killed 11–18 days after the injections, when the tumors were 1–1.5 cm x 0.5–1 cm in size. Tumors grown in Tie1-Z/EGFP mice were analyzed directly by fluorescence microscopy whereas tumors grown in Tie1-LacZ transgenic and Tie1+/LacZ heterozygous mice were stained for ß-galactosidase activity, followed by sectioning.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). 293T cells transfected with Tie1-EGFP or Tie1-ZEGFP constructs showed specific fluorescence for GFP. In Western blot, the protein products showed sizes expected on the basis of the nucleotide sequences: 30 kDa for EGFP and 43 kDa for ZEGFP (Fig. 1B
).
The Zeosin resistance encoded by the Tie1-ZEGFP construct was tested by transfecting cells of the mouse lung endothelial cell line LE-II. Only cells transfected with the Tie1-ZEGFP or pcDNA3.1/Zeo survived the Zeosin selection, whereas the nontransfected and Tie1-EGFP transfected cells died. Several surviving clones were picked and studied by fluorescence microscopy and flow cytometry. The ZEGFP protein was located in the cytoplasm and the nucleus (see Fig. 1C
). This was the distribution of the Tie1-EGFP protein in the 293T cells (data not shown). Between 96 and 98% of cells in Tie1-ZEGFP clones were fluorescent in FACS analysis, whereas none of the pcDNA3.1/Zeo clones were fluorescent by the same criteria (Fig. 1D
).
Tie1 promoter-driven expression of GFP during embryogenesis
Thirteen Tie1-EGFP and 14 Tie1-ZEGFP founders were obtained from the microinjections. Three founder lines showed fluorescence for EGFP and two for ZEGFP in the developing blood vessels of E10.5 embryos, the fluorescence cosegregating with the DNA positivity as determined by PCR. Although the fluorescence intensity varied between the founder lines, no differences were seen in the transgene expression patterns. We then compared reporter expression in the Tie1-Z/EGFP mice with the Tie1-LacZ mice, which have the same promoter fragment driving the transgene expression (19)
.
At E8, prominent fluorescence was observed in the forming endocardial heart tube and in the angioblasts of the head of Tie1-Z/EGFP embryos (Fig. 2
A, B). Few scattered fluorescent cells, presumably angioblasts, were detected in the yolk sac (Fig. 2A
). The paired dorsal aorta, located bilaterally in the trunk ventral to the somitic stalks, was also positive (Fig. 2C, D
). Whole-mount analysis at E9.5 revealed Tie1 promoter activity in virtually all developing vessels (Fig. 2E, F
). Blood vessels of the somites and the paired dorsal aorta were strongly fluorescent and endothelial cells of the developing heart could be visualized as well (Figs. 2G, H
). Tie1 promoter-driven Z/EGFP expression was high in the aortic sac, the branchial arch arteries, the atrial chamber of the heart, and the trabeculated wall of the ventricular chamber. Fluorescence was detected in the umbilical stalk connecting the embryo to the placenta (Fig. 2I, J
), the vitelline vessels, and the yolk sac vasculature (Fig. 2K, L
).
|
At E10.5, the distal part of the outflow tract of the heart opens into the region of the aortic sac, from which blood flows into the paired dorsal aortas via the aortic arch arteries. Analysis of the EGFP expression from the different founder lines revealed that Tie1 promoter activity is most prominent in the outflow tract of the heart and in the endocardium of the heart tube (Fig. 3
A). The truncus arteriosus/aortic sac forming the roots and proximal portion of the aorta, the pulmonary artery, and the sinus venosus, conus cordis forming the outflow tracts of ventricles, and bulbus cordis forming the trabeculated part of the right ventricle were all fluorescent (Fig. 3C
). The fluorescence of the head microvasculature appeared weaker than the fluorescence seen in the developing heart region (data not shown). Comparison with the previously generated Tie1-LacZ transgenic mice indicated that Z/EGFP was expressed in the same locations as the LacZ reporter (Fig. 3B, D
). However, differences in the promoter activity in different tissues could not be evaluated by the ß-galactosidase staining intensity, indicating that Z/EGFP is more suitable reporter for estimation of cell- and tissue-specific promoter activity in vivo. The Tie1 promoter continued to be highly active at E11.5 (Fig. 3E, F
). At high magnifications, fluorescence could be detected at single-cell resolution in virtually all blood vessels—for example, in the head and the somites (Fig. 3G, H
).
|
At E13.5, not all blood vessels were fluorescent. At this stage, arteries on the ventral side of the abdomen and the tail were positive whereas the adjacent large veins were largely negative (Fig. 4
A–C). No fluorescence was detected in the veins in the head region (data not shown). The most prominent fluorescence was seen in the outflow tract of the heart (Fig. 4D
). Some signal was detected in the descending aorta and in a subset of the lung endothelium (Fig. 4E
). A similar kind of EGFP expression pattern was seen at E14.5 (data not shown).
|
At E17.5, although the skin had become thicker, transgenic embryos could be identified by the fluorescent blood vessels of the developing limbs (Fig. 5
A). Again, the Tie1 promoter was highly active in the heart region and in the atriums and outflow tract of the heart (Fig. 5B, C
). Some cells in the ductus arteriosus, which allows most of the blood to bypass the pulmonary circulation on its way to the descending aorta, were fluorescent (Fig. 5C
). The descending aorta and the intercostal arteries were positive (Fig. 5D
). Again, analysis of the blood vessels in skin of the tail and thorax showed Tie1 promoter-driven EGFP expression in the arteries but not in the adjacent veins (Fig. 5E, F
).
|
Tie1 promoter activity in adult mice
No EGFP signal was detected in the lungs of adult Tie1-Z/EGFP transgenic mice, though Tie1 is known to be expressed in the adult lung endothelium (data not shown). No fluorescence was seen either in the kidney, skin, brain, liver, spleen, stomach, gut, heat, skeletal muscle, penis, ovary, uterus, or bulbourethral glands. Male mice from founder lines expressing the Tie1-ZEGFP transgene showed some fluorescence in the testis, whereas no fluorescence could be detected in the testes of Tie1-EGFP mice (data not shown). In contrast, several tissues from the adult Tie1-LacZ transgenic and Tie1+/LacZ heterozygotic mice showed ß-galactosidase staining. Some adult tissues such as the esophagus and the heart gave similar ß-galactosidase signals in these two mouse lines. However, in most tissues analyzed, as in the lung, brain, and mesenterium, staining was weaker for the Tie1-LacZ transgene than for the Tie1+/LacZ knock-in (data not shown).
Results from the Northern blot analysis confirmed that the testis was the only organ where the Tie1-ZEGFP was expressed although the endogenous Tie1 mRNA was detected in all of the adult mouse tissues analyzed (data not shown). In a very long exposure, a faint ZEGFP signal was obtained from lung RNA. Northern blot results were further confirmed by RT-PCR analysis, which demonstrated weak ZEGFP expression in the lung and strong expression in the testis (data not shown). The RT-PCR results showed low levels of EGFP mRNA in the testes of Tie1-EGFP mice (data not shown). LacZ reporter gene expression was detected in the testes of Tie1-LacZ transgenic and Tie1+/LacZ heterozygotic mice, though endogenous Tie1 transcription was low in the testis (data not shown). This suggests that expression of reporter genes in the testes of Tie1-Z/EGFP, Tie1-LacZ transgenic and Tie1+/LacZ heterozygotic mice may result from increased stability of these ectopic transcripts. Together, our results indicate that the lack of fluorescence in most of the adult tissues of Tie1-Z/EGFP mice is due mainly to lack of reporter gene expression driven by the mouse 0.8 kb Tie1 promoter.
Tie1-LacZ expression in the adult skin
Whole-mount tissue preparations of ear skin from the Tie1+/LacZ heterozygotic and Tie1-LacZ transgenic mice were studied after vascular perfusion with biotin-labeled lectin and ß-galactosidase staining. As Lycopersicon esculentum lectin binds to the surface of the endothelial cells after i.v. injection, all blood vessels can be visualized. The large veins and arteries are located adjacent to each other and can be differentiated on the basis of vessel morphology. ß-Galactosidase activity was seen mainly in the capillaries and arterioles of the Tie1+/LacZ mice, although a scattered, faint blue signal was detected in large arteries and veins (Fig. 6
A). Deep lymphatic vessels of the Tie1+/LacZ mice were ß-galactosidase positive, indicating that Tie1 is expressed in the lymphatic vessels of adult skin (Fig. 6A, B
). The ß-galactosidase staining pattern in the skin of Tie1-LacZ transgenic mice was similar, showing that in addition to the blood vasculature, the 0.8 kb Tie1 promoter fragment drives gene expression to the lymphatic vessels of adult tissues (Fig. 6C
).
|
To ensure that the lectin-negative and ß-galactosidase-positive vessels were lymphatic vessels, Tie1+/LacZ and Tie1-LacZ mice were mated with the K14-VEGFR-3-Ig and K14-VEGF-C156S mice having lymphatic phenotypes (26
, 27)
. The K14-VEGFR-3-Ig adult mice lack lymphatic vessels of the skin whereas the lymphatic vessels in the skin of the K14-VEGF-C156S transgenic mice are hyperplastic. No blue vessels without lectin staining were seen in the skin of Tie1-LacZ x K14-VEGFR-3-Ig mice, whereas in the Tie1+/LacZ x K14-VEGF-C156S mice the lectin-negative and ß-galactosidase-positive vessels were dilated, confirming that Tie1 is transcribed in the lymphatic endothelium (Fig. 6E, F
). In addition to lymphatic capillaries, the collecting lymphatic vessels of the skin of adult mice stained blue in Tie1+/LacZ mice (Fig. 6G
). Lymphatic Tie1 expression was not restricted to the skin, as the lymphatic vessels of the diaphragm and on the surface of heart and small intestine, mesenterium, and esophagus stained blue in Tie1+/LacZ and Tie1-LacZ mice (Fig. 6H
and data not shown).
Isolation and culture of endothelial cells from Tie1-EGFP embryos
To confirm the endothelial cell type-specific expression of EGFP in Tie1-Z/EGFP embryos, cells isolated at E9.5 were allowed to attach on glass coverslips and stained with antibodies against CD31, VEGFR-2, and VE-cadherin. All EGFP-positive cells expressed these endothelial markers, though expression was detected in some EGFP-negative cells (data not shown). Flow cytometric analysis indicated that only 2–3% of all embryonic cells were fluorescent (Fig. 7
A). All EGFP-positive cells were positive for CD31, although the latter were more numerous (5–7% of the embryonic cells in E11.5 embryos). After fluorescence-activated cell sorting, 93% of cells in the EGFP-positive pool were fluorescent (Fig. 7B
) whereas cells in the negative pool did not display fluorescence. In cell culture, the cells gradually formed monolayers, where all cells were EGFP positive at the sixth passage (Fig. 7C
). CD31, VEGFR-2, and VEGFR-3 were expressed in such cultures at passage 2 (Fig. 7D-I
), although their expression decreased on repeated subculturing. Results of Northern blot confirmed that the cells at passages 4 and 8 express EGFP, Tie1 (Fig. 7J
) and Tie2 (data not shown). The cellular morphology was endothelial-like with a typical cobblestone morphology at confluency (data not shown). The cells retained EGFP positivity at least until passage 12 (data not shown). To produce immortalized endothelial cell lines, cells prepared from Tie1-ZEGFP embryos at E12.5 were immortalized by retroviral infection transfecting the polyoma middle T antigen and the neomycin resistance selection marker into the cells. Cells infected with the retrovirus survived and could be passaged continuously, but did not survive in Zeosin containing culture medium beyond 4 wk.
|
Tie1 promoter activation in tumor neovasculature
To analyze possible activation of the Tie1 promoter in pathological angiogenesis, Lewis lung carcinoma cells were injected s.c. into the backs of the Tie1-Z/EGFP and Tie1-LacZ transgenic mice. For comparison, we analyzed tumors grown in the Tie1+/LacZ mice. No fluorescence was observed in the Tie1-Z/EGFP tumors, although in whole-mount analysis, the peripheral blood vessels of tumors grown in Tie1+/LacZ mice were strongly ß-galactosidase positive (data not shown). Blue staining was detected in blood vessels within the tumors transplanted into Tie1+/LacZ mice whereas no staining could be detected in tumors grown in either Tie1-LacZ transgenic or wild-type mice. The ß-galactosidase staining patterns were confirmed by analysis of tissue sections. Intense blue staining was seen in the blood vessels of the skin and in the tumor endothelium of Tie1+/LacZ mice (Fig. 8
A, D). In Tie1-LacZ transgenic mice, staining was detected in the skin endothelium whereas no staining was seen in the tumor vasculature (Fig. 8B, E
).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The use of GFP variants as lineage markers in mice has been studied by using astrocyte-specific glial fibrillary acidic protein promoter and vascular endothelial-specific, Tie2 promoter/enhancer-driven reporter expression (29
, 30)
. In this study, we analyzed the suitability of the EGFP as a reporter for the vascular development in mice having Tie1 promoter-targeted transgene expression.
In several EGFP transgenic founder lines, bright fluorescence was detected starting at E8, and the expression pattern was consistent with previous results obtained with transgenic mice expressing the LacZ reporter driven either by the Tie1 promoter or the endogenous Tie1 locus as well as with in situ hybridization results for the Tie1 mRNA (9
, 11
, 19
, 25
, 31
, 32)
. Fluorescence was high in structures formed by de novo differentiation of angioblasts in the dorsal aorta and heart endocardium, as well as in angiogenically derived vessels such as intersomitic vessels and limb bud vasculature. The outflow track of the heart was highly fluorescent in the Tie1-Z/EGFP mice throughout embryogenesis.
Although Tie1 transcription is abundant during early embryogenesis, previous studies with chimeric mice having a proportion of cells without functional Tie1 gene have indicated that Tie1 does not have a significant function in vasculogenesis or in early angiogenic processes for up to E10.5 (11)
. Whether Tie1 has a role in the proper development and function of these blood vessels later in embryogenesis is not known, though results from studies of embryonic chimeras indicate this is not the case at least for the aortic endothelium at E15.5 (11)
. Tie1-deficient mice die after E13.5 due to loss of vascular integrity leading to edema and hemorrhages (25
, 32)
. Analysis of chimeric mice has shown that Tie1 is needed for capillary endothelial cell proliferation and migration during late embryogenesis and in adults (11)
. After midgestation, the Tie1 promoter was no longer active in all blood vessels. No fluorescence was seen in the veins in the abdominal, head, or tail regions of transgenic embryos at E13.5, whereas the signal was strong in arterial endothelia (e.g., in the endothelium of developing dorsal aorta and the carotid arteries).
Several genes are known to be differentially expressed between the developing arteries and veins. These include the Eph family transmembrane ligand ephrin-B2, expressed only in the arteries, and its receptor, EphB4, which is expressed only in veins (33)
. The activin receptor like kinase-1 (ACVRL1) apparently participates in the determination of arteries and veins, as a loss of function mutation in this gene causes arteriovenous malformations (34
, 35)
. Tie1 is up-regulated in arteriovenous malformations in humans, supporting the idea that Tie1 may have a role in the process of arterial-venous specification during embryonic development (36)
. However, some veins expressed Tie1 on the basis of in situ hybridization results using E12.5 mouse embryos (12)
. In addition, ß-galactosidase staining of E18.5 Tie1+/LacZ embryos revealed Tie1 promoter activity in the umbilical veins and veins of the yolk sac, although at lower levels than in the arteries (data not shown).
The lack of fluorescence signals in adult Tie1-Z/EGFP mice is partly due to reduced Tie1 promoter activity and increased autofluorescence of adult tissues. Because each EGFP protein molecule represents one fluorophore, the fluorescence signal is not amplified as in the case of the chromophore signals produced by the enzymatic ß-galactosidase reporter. Therefore, we studied Tie1 transcription in adult mice using the LacZ reporter. A comparison of LacZ expression patterns in various tissues of Tie1-LacZ transgenic and Tie1+/LacZ heterozygous mice indicated that some regulatory elements needed for the endogenous Tie1 transcription are missing from the promoter fragment. We do not know whether additional Tie1 enhancer elements needed to recapitulate the endogenous Tie1 expression pattern in adults are located in the first intron, as in the case of the endothelial cell-specific Tie2 and Vegfr2 genes (37
, 38)
. Lectin-perfused and ß-galactosidase-stained skin from the ears of Tie1+/LacZ mice indicated Tie1 transcription mainly in the arterioles and capillaries, though very faint and scattered reporter gene expression could be detected in large arteries and veins. A similar blood vascular expression pattern was seen in the Tie1-LacZ transgenic mice, indicating that the Tie1 promoter targets mainly arterioles and capillaries in the adult mice.
Tie1 transcription occurred in the lymphatic endothelium of adult skin. Again, similar reporter gene expression was seen in the transgenic and knock-in mice. The lymphatic nature of the stained vessels was confirmed by mating Tie1-LacZ transgenic and Tie1+/LacZ heterozygous mice with transgenic mice having lymphatic hypoplasia or hyperplasia (26
, 27)
. Lymphatic Tie1 transcription was not restricted to the adult skin, as ß-galactosidase expression was seen in the lymphatic endothelium of the diaphragm and esophagus and in the lymphatic vessels on the surface of the heart and small intestine as well as in mesenteric lymphatic vessels. Tie1 is already expressed in the lymphatic endothelium during embryogenesis at E13.5, when the first lymphatic vessels can be morphologically distinguished from the blood vessels (K. Iljin et al., unpublished results). Whereas no differences were seen between the blood vessels of Tie1-LacZ transgenic and Tie1-LacZxK14-VEGFR-3-Ig embryos (26)
, the blue-stained lymphatic vessels were missing from double transgenic mice at E17.5 and P0 (K. Iljin et al., unpublished results). Further analyses are needed to characterize Tie1 expression and its function in the lymphatic endothelium in detail.
A recent study shows that mice deficient of the Tie2 ligand Ang2 die within a couple of weeks after birth and have defects in the lymphatic vessels of the skin and gut (39)
. Ang2 was first characterized as an antagonist of Tie2 in endothelial cells, whereas in nonendothelial cells, Ang2 induced phosphorylation of ectopically expressed Tie2 (5)
. During embryogenesis, Ang2 is expressed in the smooth muscle layer of dorsal aorta and major aortic branches (5)
. In adults, Ang2 expression is seen in tissues subject to physiological angiogenesis—for example in the ovary, uterus, and placenta (5)
. As Tie1 is expressed in the endothelium of the same structures as Ang2 during embryogenesis and enhanced Tie1 mRNA expression occurs together with Ang2 mRNA in the adults, it is tempting to speculate that Tie1 might be the endothelial factor modulating Ang2 function via the Tie2 receptor. So far, little is known about molecules interacting with the Tie1 receptor. Despite great efforts, no ligand for Tie1 has been identified and only ligand-independent signal transduction pathways have been characterized (40
41
42)
. However, double null mouse embryos for Tie1 and Tie2 receptors have a more severe phenotype than single knockout phenotypes, suggesting that these receptors do interact in vivo (10)
. Recently Tie1 and Tie2 receptor heterodimers have been shown to exist in cultured endothelial cells, but Tie1 is not transphosphorylated on Tie2 receptor activation and the cellular function of heterodimerization remains to be determined (43)
.
Large-scale isolation of primary endothelial cells is of great interest for vascular gene profiling, isolation of endothelial cells for tissue engineering, and proangiogenic treatment of pathological conditions, such as tissue ischemia. Several protocols have been described for isolation of endothelial cells from murine tissues with magnetic beads conjugated with antibodies against endothelial cell markers (44
, 45)
. We describe here an efficient way to isolate primary murine endothelial cells by fluorescence-activated cell sorting from the transgenic embryos expressing Z/EGFP under control of the Tie1 promoter. The relatively small number of endothelial cells in the starting material can thus be enriched and expanded in culture. The morphology, fluorescence, and expression of endothelial cell-specific receptors indicate that these cells maintain at least some of their endothelial characteristics in culture, establishing Tie1-Z/EGFP mice as a convenient source for isolating primary endothelial cells for a variety of experimental purposes.
We were unable to produce stable endothelial cell cultures by retroviral infection of polyoma middle T antigen from the Tie1-ZEGFP embryos. Some studies indicate that cells expressing high levels of GFP are selected against in culture (46)
. Reports showing a link between Aequorea EGFP expression and a high degree of cell death have been published (47
, 48)
. However, other studies have indicated that the expression of EGFP does not confer a growth disadvantage to mammalian cells (49)
. Also, in transgenic mice expressing high levels of EGFP in all tissues with the exception of erythrocytes and hair, no toxic effects were seen (28)
.
The regulatory regions driving gene expression specifically to the angiogenic endothelial cells may have potential in the development of new therapeutic approaches to vascular diseases. TIE1 is expressed in the vascular endothelium of human breast cancer, Kaposi’s sarcoma, and cutaneous angiosarcoma (16
, 18)
. TIE1 expression has been shown to be inversely correlated with the survival of gastric cancer patients, and elevated TIE1 mRNA expression has been detected in brain tumors and melanomas of the skin (13
, 15
, 17)
. We analyzed the regulation of Tie1 during tumor neovascularization by transplanting Lewis lung carcinoma into the Tie1-LacZ transgenic or knock-in mice. Our results indicate that although endogenous Tie1 transcription is activated in tumor endothelium, the mouse Tie1 promoter fragment used did not contain the regulatory elements needed for the expression in tumor vasculature, as no fluorescence or ß-galactosidase activity was seen in tumors grown in the promoter-reporter transgenic mice.
Tie1-Z/EGFP transgenic mice provide a useful tool for imaging vascular morphogenesis during embryonic development, as the promoter is highly active and the tissues are transluminescent. However, LacZ was more sensitive than Z/EGFP as a reporter when analyzing newborn and adult mouse tissues. We believe that Tie1-Z/EGFP transgenic mice are very useful in angiogenesis research, as these mice can be mated with other transgenic or knockout mice and endothelial cells from the double transgenic embryos can easily be isolated for further characterization in vitro.
|
ACKNOWLEDGMENTS |
|---|
Received for publication February 4, 2002. Accepted for publication July 22, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. He, H. Hu, R. Braren, S.-Y. Fong, A. Trumpp, T. R. Carlson, and R. A. Wang c-myc in the hematopoietic lineage is crucial for its angiogenic function in the mouse embryo Development, July 15, 2008; 135(14): 2467 - 2477. [Abstract] [Full Text] |
