| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
RESEARCH COMMUNICATION |
a Molecular/Cancer Biology Laboratory, Haartman Institute and Department of Biomedicine, University of Helsinki, Helsinki, Finland;
b NEB Bone and Joint Institute, Beth Israel Deaconess Medical Center, Harvard Institutes of Medicine, Boston, Massachusetts 02115, USA; and
c Biocenter Oulu and Department of Biochemistry, University of Oulu, Linnanmaa, 90570 Oulu, Finland
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: gene therapy • receptor tyrosine kinase • vascular endothelium
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In contrast to cells contributing to the skeletal muscle and hematopoietic lineages, little is known about the mechanisms of differentiation and maturation of endothelial cells. Regulatory regions of several endothelial cell-specific genes have been identified and partially characterized only recently 2-13) . The promoter elements controlling endothelial-specific gene expression are of special interest because they may be useful in making targeting vectors for gene therapy directed against angiogenesis associated with tumors or other disease processes involving the vascular system. Tie (tyrosine kinase with immunoglobulin and epidermal growth factor homology domains)2 was the first one of the five receptor tyrosine kinases found to be endothelial cell specific (14) . To understand the molecular mechanisms underlying specific expression of genes in endothelial cells, we have studied the elements that control Tie transcription.
Tie has an essential role in the development of the vascular system. It is first expressed on embryonic day 8.5 (E8.5) in angioblasts of the head mesenchyme, in endothelial cells of the dorsal aorta, and in the blood islands of the yolk sac, and continues in endothelial cells throughout embryogenesis (15) . Expression of Tie is down-regulated in many adult tissues after development, but in those where it persists, it is endothelial cell-specific (15) . However, Tie mRNA is enhanced in the endothelium of adult mice at sites of neovascularization associated with corpus luteum formation during the menstrual cycle and with wound healing (16) . Increased levels of Tie mRNA and protein have also been found in some pathological conditions such as in the vasculature of certain tumors, suggesting that Tie is needed for tumor angiogenesis (17 , 18 ). To date, no ligand for Tie has been identified, but it is possible that Tie modulates angiopoietin signals via the related Tek/Tie-2 receptor (19) .
The promoters of both human and mouse Tie gene have been isolated and partially characterized (4) . Neither promoter contains classical promoter elements such as TATA or CAAT boxes. Significant homology between the mouse and human promoters was seen in the region of about 0.9 kb upstream of the translational initiation codon. This promoter region contains conserved putative binding motifs for transcription factors conforming to Ets, AP-2, a conserved octamer factor binding DNA element, and GT repeat sequences, also described in the mouse Tek/Tie-2 endothelial-specific enhancer (10) . The 0.8 kb AflII-ApaI fragment of the mouse Tie promoter containing all these elements could target gene expression specifically to endothelial cells in vivo, indicating that the elements relevant for endothelial cell specificity are contained within this fragment (4) .
Our aim was to determine which of these elements are required for the activity and cell type specificity of the mouse Tie promoter. Mutations were made to the promoter by site-directed mutagenesis, and activities of the mutant promoters were tested both in vitro and in vivo by using reporter gene constructs. In this study, the Tie promoter showed endothelial cell-specific activity in transfected cells. Mutations made to the octamer region and to the Ets binding sites strongly decreased promoter activity in vitro. The results from transgenic embryos showed that all the sequences necessary for endothelial-specific expression are within a 333 bp region, which contains several putative binding sites for Ets-factors and the octameric sequence.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-32P]UTP. Twenty
micrograms of poly(A+) RNA were incubated with labeled probe at 50°C
overnight. Unhybridized RNA was digested with RNAse A (10 U/ml) and T1
(1 µg/ml) at 37°C for 15 min and the samples were analyzed in 8%
sequencing gels.
|
Primer extension analysis
Primer extension analysis was carried out using the AMV reverse
transcriptase primer extension system (Promega, Madison, Wis.). A
synthetic oligonucleotide complementary to the 5' end of mouse Tie cDNA
(5'-ACTTACCAACATGAGAGGCCAAGAAAAGAGTG-3') was labeled according to the
manufacturer's protocol. The primer (10 pmol) was incubated with 1x
forward exchange buffer, 10 µCi/ml [-32P]ATP, and 10 U T4
polynucleotide kinase at 37°C for 1 h. The kinase was
inactivated by heating at 90°C for 2 min and the labeled primer was
ethanol precipitated.
About 20 µg of poly(A+) RNA and 5 x 105 cpm of labeled primer were annealed in hybridization buffer (40 mM Pipes, pH 6.4, 1 mM EDTA (pH 8.0) 0.4 M NaCl and 80% formamide) by heating at 95°C for 12 min, cooled slowly to room temperature, and ethanol precipitated. The dried annealing mixture was suspended in primer extension buffer (50 mM Tris-HCl, pH 8.3, at 42°C, 50 mM KCl, 10 mM MgCl2, 10 mM DTT, 2 mM each dNTP and 0.5 mM spermidine), and 20 U RNasin and 40 U AMV reverse transcriptase were added. After a 2 h incubation at 42°C, template RNA was digested by 20 µg/ml RNAse A in 100 mM NaCl, 10 mM Tris-HCl, and 1 mM EDTA (pH 7.5) at 37°C for 15 min. The mixture was phenol extracted and ethanol precipitated; the pellet was resuspended in loading dye (98% formamide, 10 mM EDTA, 0.1% xylene cyanol, and 0.1% bromophenol blue) and loaded to a 9% acrylamide/7 M urea gel. After electrophoresis, the dried gels were autoradiographed for 2–4 days.
Plasmids and mutagenesis
In transfection assays of cultured cells, the promoter
activities were analyzed using luciferase reporter gene constructs.
Deletions of the AflII-ApaI promoter fragment in
the pGL2 reporter vector were made using the appropriate primers and
polymerase chain reaction (PCR), followed by recloning to the same
vector and confirmed by sequencing. pGL3 basic and control vectors
contained the firefly luciferase gene without promoter and driven by
the SV40 promoter/enhancer, respectively. Additional deleted and mutant
fragments of the mouse Tie promoter were inserted into the pGL3 basic
vector. Schematic pictures of the resulting constructs appear in
Fig. 2
and Fig. 3
A. For site-directed mutagenesis, the 1.2 kb
HindIII-BamHI fragment of mouse Tie promoter
containing the AflII-ApaI fragment, which was
previously shown to be endothelial cell specific in vivo
(4)
, was cloned to the pALTER site-directed mutagenesis
vector. Because this fragment contains the translational initiation
codon (ATG) of the Tie open reading frame, an NcoI site was
introduced the sequence GT ATG G by using the Altered Sites
II in vitro mutagenesis system (Promega) and the primer
5'-TCCCCACCAGACC ATG GTCTGGAAGCTGCCCT-3'. The ATG codon embedded in the
NcoI site (CC ATG G), would then be in the
correct reading frame with the luciferase open reading frame of the
pGL3 vector.
|
|
Upstream deletions of the GT repeat sequences and the Ets1 site were similarly made using primers 5'-TGAGAAGGTTTGGAGGCAGGACATTGTTCTTTTTACTGAGATTTTATTATGTTGTTATCT-3' and 5'-CACGGGAGACAGGAAAGGGAGGAAGCCAGTTGAGCACTTACAGTTCTCTGACTAAC GGTC-3', respectively. Mutations were also made to the 3' cluster of Ets sites (primer 3'-CCTCGGTGCTGGGGATTAAGATTAAATGATGGTTAGGG-3'), the 5' cluster (primer 5'-ACGGGAGACAGGAAATTGATTAAGCCAGTGGCTTCC-3'), the octamer binding sequence (primer 5'-AGGTTGTGACAAGGACGCGTCTGGGGATGGTTGG-3'), and the AP-2 binding sequence (primer 5'-TTTTGGCGTCTTCCATGGTCTGGAAGCTGCCACCTCGGTGCTGGGGAGGAAGAGGAAATG-3') using as template the HindIII-NcoI fragment with a deletion of the GT repeat sequences. These latter mutations were made by using the GeneEditor in vitro mutagenesis kit (Promega). All constructs were confirmed by sequencing and named according to the mutated DNA sequences (see Fig. 3A ).
For studies in transgenic mouse embryos, HindIII-NcoI digested Tie promoter fragments were ligated to the pSDKLacZpA reporter vector (a kind gift from Dr. Janet Rossant, Mount Sinai Research Institute, Toronto, Canada; see ref 20 ) containing the ß-galactosidase open reading frame, followed by a simian virus 40 poly(A) addition sequence.
Cell culture and transfection assays
Mouse lung endothelial cells (LE-II; ref 21
) and brain
endothelial cells (BEND; ref 22
) were grown in minimal essential medium
(MEM) and NIH3T3 murine fibroblasts in Dulbecco's modified Eagles'
medium (DMEM) containing 10% fetal calf serum and antibiotics. Mouse
keratinocytes (MK-2) were grown in low (2 mM) CaCl2 MEM
containing 10% fetal calf serum, antibiotics, and epidermal growth
factor (4 ng/ml). The spontaneously immortalized human umbilical vein
endothelial cells (ECV304, 23) were grown in Media 199. The human
embryonic kidney cells (HEK293, ATCC) and the murine endothelial cell
line PY41 (24)
were grown in DMEM-supplemented 10% fetal
bovine serum and antibiotics, as described previously
(25)
.
LE-II, BEND, NIH3T3, and MK-2 cells were transfected by the calcium phosphate-mediated transfection method. The minimal amount of plasmid registering full transcriptional activity was titrated prior to the experiments. In subsequent cases, 5 µg of the appropriate reporter construct was transfected along with 0.2 µg of pRL-TK to correct for variability in transfection efficiency. The pRL-TK vector contains the herpes simplex virus thymidine kinase promoter region upstream of the renilla luciferase gene. Cell extracts were prepared 48 h after transfection in the passive lysis buffer (Promega). Luciferase activity was measured using Digene DCR-1 luminometer and Promega Dual Luciferase Assay System.
The ratio of firefly luciferase activity to renilla luciferase activity in each sample served as a measure of the normalized luciferase activity, which was divided by the activity of the pGL3 control vector and expressed as relative luciferase activity. Each construct was transfected at least five times, and data for each construct are presented as the mean ±SE. Relative luciferase activity among constructs was compared by a factorial analysis of variance, followed by the Student's t test. Statistical significance was accepted at P <0.05.
Cotransfections of 1.5–2 x 105 ECV and HEK293 cells were performed using 1.75 µg of the reporter gene construct DNA and 0.75 µg of the expression vector DNA with 6.25 µl of lipofectamine (Gibco BRL, Paisley, U.K.). Cells were washed with serum-free DMEM. A total of 0.8 ml of serum-free DMEM was added per well. Liposomes were incubated with the DNA in 200 µl of serum-free DMEM for 15 min at room temperature and then with the cells for 4 h at 37°C. One milliliter of DMEM containing 20% fetal calf serum was added; the cells were harvested 16 h after transfection and assayed for luciferase (26) . Transfections with the PY41 cells were performed similarly, but in Optimem. Minimally, transfections were in duplicate or triplicate with similar results. Cotransfection of a second plasmid for determination of transfection efficiency was omitted because potential artifacts with this technique have been reported (27) and because many commonly used viral promoters contain potential binding sites for Ets factors.
Production and analysis of Tie promoter-LacZ embryos
The transgene was isolated from the pSDKLacZpA vector by
SalI digestion and transgenic embryos were
produced by microinjection. The injected C57/B16xDBA/2
zygotes were transferred into the oviducts of pseudopregnant CD 1
foster mothers. Embryos were stained on day 11.5 and were genotyped by
PCR analysis of yolk sac and amnion DNA prepared by the salt
precipitation method (28)
, using a primer that annealed
within the Tie promoter (5'-AGGGAGAGAGGGTGGGAAG-3') and another one
specific for the multiple cloning site of the SDKLacZpA vector
(5'-GCTCTAGAACTAGTGGATC-3').
Whole embryos were rinsed with phosphate buffer (0.1 M Na2HPO4, 0.1 M NaH2PO4, pH 7.3), fixed in 0.2% glutaraldehyde, 5 mM EGTA, 2 mM MgCl2 in phosphate buffer for 20 min, washed three times for 15 min with the washing buffer (2 mM MgCl2, 0.01% desoxycholate, 0.02% Nonidet 40 in phosphate buffer), and stained in 5-bromo-4-chloro-3-indolyl ß-D-galactopyranoside for 1 day. After staining, embryos were incubated in washing buffer for 1 day, postfixed with 4% paraformaldehyde for 1 day, and dried with methanol. Embryos were embedded with paraffin; 7 µm sections were cut and counterstained with nuclear fast red.
Production and analysis of Tie promoter-hGH transgenic mice
The AflII-ApaI fragment of the mouse Tie
promoter was cloned into the p|$$/OGH vector (Nichols Institute, Los
Angeles, Calif.) upstream of the human growth hormone (hGH) cDNA. The
transgene was isolated by EcoRI-XbaI digestion.
DNA was extracted from mouse tail samples by the salt precipitation
method. Positive pups were screened by PCR using primers annealing to
human Tie promoter (5'-GAGACAGGGGATGGGAAAAA-3') and to the hGH cDNA
(primer 5'-AATGGTTGGGAAGGCACTGC-3'). The copy number of the transgene
was studied by Southern blotting. DNA samples were digested with
AvaI enzyme, and the labeled
BglII-ApaI fragment of the Tie promoter was used
as the probe. Blood samples were taken twice from mice at the age of 8
and 10 months. The hGH concentration was determined by a clinical hGH
immunoradiometric assay. Serum from a wild-type mouse was used as a
negative control.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Activity of the Tie promoter in cultured endothelial and
nonendothelial cells
To define the minimal sequences of the mouse Tie promoter
responsible for its activity, deletions were made to the
AflII-ApaI promoter fragment found to direct
endothelial cell-specific expression in transgenic mice
(4)
. The deleted fragments were cloned into the luciferase
reporter vector, transfected to LE-II mouse endothelial cells, and the
cell extracts were later assayed for luciferase activity. Figure 2
shows the deletions schematically and relative promoter activities of
the constructs. A 86 bp deletion of the 3' sequences of the
AflII-ApaI fragment, including the major
transcription initiation sites, abolished all promoter activity. A
promoter deleted of the 541 bp AflII-BglII
fragment from the 5' end yielded nearly one-third the activity of the
strong Rous sarcoma virus promoter used as a positive control. A
further 3' deletion of the 86 bp from the
BglII-ApaI fragment abolished this activity,
whereas minimal activity was registered in a construct truncated at the
RNA initiation sites. Only minimal promoter activity was recovered upon
deletion of 96 bp of the 5' sequences from the
BglII-ApaI fragment.
Further mutations were made to the promoter region by site-directed mutagenesis (see Fig. 3A ). Although previous analysis showed that Tie expression is restricted to endothelial cells in vivo, Tie promoter activity was found both in endothelial cells and cervical carcinoma cells in culture (4) . To determine whether a cell culture system is suitable for studying endothelial cell-specific elements of the Tie promoter constructs, we carefully compared promoter activity in endothelial and nonendothelial mouse cells. The results of this analysis are shown in Fig. 3B . In LE-II and BEND mouse endothelial cells, the 1.1 kb HindIII-NcoI Tie promoter fragment showed 42% and 70% of the activity, respectively, of the SV40 promoter construct used as the positive control. In NIH3T3 fibroblasts and MK-2 epithelial cells, the promoter was also active but at a markedly lower level (12 and 7%, respectively, of the SV40 promoter/enhancer), indicating that the activity of the promoter is relatively specific for endothelial cells. The deletion of an upstream 0.6 kb fragment, including the GT repeats (Fig. 3A ), reduced significantly the promoter activity in BEND cells, whereas in LE-II cells the activity was somewhat increased (Fig. 3B ). Additional deletion of the upstream Ets binding site resulted in an ~10% decrease of promoter activity in both cell types when compared to the GT-deleted construct.
To further analyze the promoter, the individual transcription factor consensus DNA binding sites in the mTie prom-GT construct were subjected to site-directed mutagenesis as described in Materials and Methods and in Fig. 3A . Mutation of the P1 doublet of Ets sites or the octamer sequence resulted in about a 60% reduction of the activity in both endothelial cell lines. The AP-2 site was also essential for strong Tie promoter activity in the endothelial cells in comparison with the -GT construct. Mutation to the P2 doublet of Ets sites just upstream of the major transcription initiation site had the strongest effect on promoter activity, reducing the activity by 83% in LE-II cells and by about 92% in MK-2 cells. Mutations affecting Tie promoter activity only in nonendothelial cells were not found.
Ets factor transactivation
Because a number of Ets binding sites are conserved between the
mouse and human Tie promoters, we compared the ability of seven
transcription factors of the Ets family to transactivate the promoter
in human and murine endothelial cells (Fig. 4
). In human ECV endothelial cells, Tie promoter activity increased
only between 1.5- and 3-fold upon cotransfection with expression
vectors for the NERF-2, ETS1, and ETS2 transcription factors, whereas
no stimulation was obtained with other Ets factors tested (data not
shown). To avoid the possible high endogenous background caused by Ets
factors such as NERF-2, which are expressed in endothelial cells, the
human embryonic kidney cell line HEK293 was used in a subsequent
experiment. In these cells, NERF-2, ETS1, and ETS2 were able to
transactivate the Tie promoter constructs by about 8.5-, 4-, and
3-fold, respectively, whereas TEL, SAP, NERF-1, and ELK yielded little
or no activity (Fig. 4A
).
|
Deletion of the GT-rich region and the upstream Ets consensus binding site reduced the degree of NERF-2- and ETS1-specific transactivation by 45% and 25%, respectively, but did not abolish the induction. In murine PY41 endothelial cells, ETS2 and NERF-2 were the best transactivators, although again the degree of stimulation was weaker, probably because of endogenous Ets activity in these cells. In the endothelial cells, ELK, SAP, and Tel apparently down-regulate the basal activity of the promoter when compared with HEK293 cells. Deletion of the GT-rich region enhanced the induction and transactivation by ETS1, which occurred only when the GT-rich area was deleted (Fig. 4B ). Further deletion of the upstream Ets binding site decreased promoter activity back to wild-type levels (Fig. 4B ). Mutations of the P1 or P2 Ets doublet binding sites resulted in an almost complete loss of basal activity and transactivation capacity (Fig 4C ), indicating the importance of these binding sites in the Ets-induced activity of the Tie promoter.
Activity of Tie promoter constructs in transgenic mouse embryos
We showed earlier that the AflII-ApaI
fragment of the mouse Tie promoter targets expression of heterologous
genes specifically to endothelial cells in transgenic mice
(4)
. Deletion of the 541 bp
AflII-BglII fragment from the 5' end of this
promoter (cf. Fig. 2
) was found to restrict the activity of the
promoter to the bone marrow and lungs of transgenic mice (data not
shown), indicating that some critical regions reside in the
AflII-BglII fragment. To determine more
specifically the DNA sequences critical for endothelial cell
specificity in vivo, transgenic mouse embryos were produced
and analyzed for ß-galactosidase activity at 11.5 days postcoitum.
Deletion of the 5' region containing the GT repeats of Tie promoter did not affect endothelial cell specificity of the promoter. However, when the mTie prom-GT-Ets construct was tested in transgenic embryos, one of five injected embryos showed ß-galactosidase activity throughout the endothelium (Fig. 5 A). The expression level in the endothelium of the positive embryo was comparable to that of ß-galactosidase activity in the Tie/Tie-LacZ heterozygous knock-in embryos (Fig. 5B ; see ref 20 ), except in the heart endocardium, where only minor signals were obtained. Staining patterns in the various organs were confirmed by analysis of tissue sections. Intense blue staining was seen in the developing perineural/meningeal capillaries (Fig. 5C, D ) as well as in the intersomitic arteries (Fig. 5E, F ) in both mTie prom-GT-Ets transgenic and Tie/Tie-LacZ knock-in embryos. Blue staining was also observed in the endocardium of the heart ventricles in both cases (Fig. 5G, H ), although the staining was clearly much stronger in the heterozygous knock-in embryos. Mutation of the P2 doublet of Ets sites abolished most of the promoter activity. Five transgenic embryos made with this construct did not stain for ß-galactosidase; only one embryo was found displaying weak staining of part of the vessels positive in mTie prom-GT-Ets embryo (Fig. 5A , inset).
|
Tie promoter-driven secretion of hGH into blood circulation
Transgenic mice expressing the human growth hormone cDNA under the
0.8 kb Tie promoter were generated and analyzed as a model of Tie
promoter-driven secretion of protein into the circulation, because the
hGH levels can be easily quantitated in mice. The level of hGH in the
serum from four mice was determined twice at the ages of 8 and 10
months by using hGH immunoradiometric assay, which does not cross-react
with mouse GH. Results are presented in Table 1
. The amount of hGH in blood correlated with the copy number of the
transgene, being between 0.25 and 1.04 mg/ml per one copy of the
transgene. This suggested that the integration site of the transgene
did not have a strong effect on the activity of the 0.8 kb promoter.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Although the Tie promoter does not contain a TATA box, a single major transcription start site was obtained in RNase protection and primer extension analysis of embryonic and adult kidney RNA. Deletions made to the endothelial cell-specific AflII-ApaI fragment resulted in an inactive promoter, when the RNA start sites were included in the deleted fragments. These deletions also suggested that the P2 sequence containing putative Ets binding sites is responsible for much of the basal activity of the promoter in endothelial cells. Deletion of 0.6 kb containing the GT-rich region from the 5' end of the HindIII-NcoI Tie promoter fragment had opposite effects on promoter activity in two different endothelial cell lines, but it did not affect the endothelial cell specificity of the Tie promoter in vivo. The result that the GT element is not required for Tie promoter activity in endothelial cells is consistent with the finding that a similar fragment of the human Tie promoter does not contain a GT repeat (4) . Additional 5' deletion, including the putative binding site for an Ets factor, did not markedly change the promoter activity in vitro, which suggests that Ets either does not bind to this site or that the binding is dispensable for promoter activity. The fact that this site is not conserved in the human sequence suggests that it is not a critical Ets binding site. Studies of Tie promoter-LacZ transgenic embryos gave further support for this view, as some transgenic embryos for mTie prom-GT-Ets/lacZ construct showed ß-galactosidase activity specifically in the endothelial cells. However, a greater integration site dependency was seen with this construct, since ectopic expression was observed more frequently. In contrast, a mutation made in the octamer sequence, and especially the one in the P2 doublet of Ets sites in the 3' promoter region, reduced promoter activity dramatically in all cell lines studied, indicating the importance of these sequences for Tie promoter activity. Embryos transgenic for the mTie prom-GT-P2/lacZ construct had a strongly reduced reporter gene expression, confirming the critical role of these Ets sites for Tie promoter activity also in vivo.
Recently, an autonomous endothelial-specific enhancer in the first intron of the mouse Tek gene was characterized. In combination with the Tek promoter, this enhancer allowed targeting of gene expression to virtually all endothelial cells in mouse embryos (10) . The Tek enhancer contains many DNA elements that also occur in the Tie promoter. A mutation destroying the palindromic octamer and the putative NF-S site in the Tek enhancer caused greater integration site dependency of the corresponding construct. However, endothelial expression was only slightly weakened by this mutation. The same octamer sequence was also identified in the intercellular adhesion molecule 2 (ICAM-2) promoter region (12) . Human ICAM-2 expression in tissues is restricted largely to vascular endothelial cells and megakaryocytes (29) . A mutation of the octamer sequence reduced ICAM-2 promoter activity in bovine aortic endothelial cells by 70% (12) . Deletion of a Tek enhancer fragment containing solo and doublet Ets sites completely abolished in vivo activity of the Tek enhancer (10) . Because this enhancer contains only one Ets doublet element, the solo Ets site in this case may correspond to the P2 element rather than to the dispensable Ets site in the Tie promoter. The critical region of ICAM-2 promoter for endothelial cell-specific activity also contains a tandem Ets motif, but the significance of this region for promoter function remains to be determined. The essential function of Ets doublet elements in Tie and Tek promoters and their presence in the ICAM-2 promoter suggest that these sites are important for endothelial cell-specific gene transcription.
To date, more than 30 Ets family members have been isolated from a variety of different species. Ets proteins are implicated in the regulation of gene expression during many important biological processes such as cell growth, differentiation and transformation. They all share a conserved 85 amino acid-long Ets domain. The Ets domain binding site contains a centrally located purine-rich sequence GGA (30) . Many different Ets proteins can bind to the same Ets binding site, and it has been suggested that the expression of specific Ets target genes in vivo is regulated by tissue-specific expression of particular Ets proteins. Among seven different members of the Ets transcription factor family tested, NERF-2 showed the strongest transactivation of the Tie promoter. NERF-2 is one of the three isoforms of NERF, which is a recently isolated member of the Ets factor family (31) . Among the Ets family, NERF has highest amino acid sequence homology to ELF-1, which is involved in the regulation of several T and B cell-specific genes (31) . NERF-2 has also been shown to transactivate Tek reporter constructs 15 to 20 fold (A. N. Dube et al., unpublished results), whereas the maximal transactivation of the Tie reporter construct was found to be eight- to ninefold in the present study. As NERF-2 is expressed in endothelial cells, it may also regulate Tie promoter activity in vivo. NERF-1a isoform, which is only weakly expressed in endothelial cells compared to NERF-2, was found to have no effect on Tie promoter activity.
SAP1 and ELK-1 genes form a subgroup of the Ets family; they both interact with the serum response factor. However, neither SAP-1 nor ELK-1 transactivated Tie gene reporter constructs. ETS1 has a somewhat similar expression pattern as Tie (4, 32). This proto-oncogene is transcribed in endothelia during angiogenesis in human and chick embryos, granulation tissue, and especially the endothelium of tumor vessels (32, 33). Both ETS1 and ETS2 proteins can activate transcription through doublet Ets motifs and regulate transcription of genes encoding matrix-degrading proteases. In our experiments, ETS1 increased Tie promoter activity maximally by only three- to fourfold. Although one of the Ets factor gene knockouts (TEL) was recently reported to result in an extraembryonic vascular phenotype with abnormal vitelline vein development (34) , it did not transactivate the Tie promoter. In conclusion, our results show that among the Ets factors tested, NERF-2 and ETS2 are the strongest transactivators of the Tie promoter, and that mutations made to all three putative Ets factor binding sites reduce significantly this transactivation capacity; the most striking reduction of both basal and Ets factor-induced activity was seen when either one of the Ets doublet elements was mutated. The contribution of other Ets factors in vascular development and Tie gene regulation remains to be determined, but it is highly likely that members of the Ets factor family play an important role given the conservation of Ets binding sites in several vascular-specific genes.
Angiogenesis is involved in many disease processes, including diabetes, psoriasis, and rheumatoid arthritis, as well as tumor growth and metastasis. One goal of studies of endothelial cell-specific promoters is to define elements that might be useful for pro- or antiangiogenic therapy. Another potentially useful application of endothelial cell-specific gene control elements is in the construction of vectors for endothelial cell-specific secretion of proteins replacing missing or nonfunctional gene products in the circulation. In this report, we tested the potential of the AflII-ApaI fragment of the mouse Tie promoter for such purposes, using a gene encoding the hGH marker protein that occurs in serum and can easily be quantitated. The Tie promoter-driven secretion of hGH was found to be relatively independent of the integration site, but dependent on transgene copy number. As high amounts of human growth hormone were detected in the circulation of transgenic mice, the Tie promoter may thus have potential for use in gene therapy.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Abbreviations: ATG, translational initiation codon; BEND, brain endothelial cells; DMEM, Dulbecco's modified Eagles' medium; E, embryonic day; HEK293, human embryonic kidney cells; hGH, human growth hormone; ICAM-2, intercellular adhesion molecule 2; LE-II, mouse lung endothelial cells; MEM, minimal essential medium; MK-2, mouse keratinocytes; NERF-2, a novel transcription factor related to the ets factor ELF-1; PCR, polymerase chain reaction; Tie, tyrosine kinase with immunoglobulin and epidermal growth factor homology domains.
Received for publication August 24, 1998.
Revision received October 2, 1998.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-induced negative promoter regulation. Nucl. Acids Res. 1993;21:5636-5643.
This article has been cited by other articles:
|
|
 |
|
  Y. Okada, K. Yano, E. Jin, N. Funahashi, M. Kitayama, T. Doi, K. Spokes, D. L. Beeler, S.-C. Shih, H. Okada, et al. A Three-Kilobase Fragment of the Human Robo4 Promoter Directs Cell Type-Specific Expression in Endothelium Circ. Res., June 22, 2007; 100(12): 1712 - 1722. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Oettgen Regulation of Vascular Inflammation and Remodeling by ETS Factors Circ. Res., November 24, 2006; 99(11): 1159 - 1166. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Huang, C. Brown, W. Ni, E. Maynard, A. C. Rigby, and P. Oettgen Critical role for the Ets transcription factor ELF-1 in the development of tumor angiogenesis Blood, April 15, 2006; 107(8): 3153 - 3160. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Hegen, S. Koidl, K. Weindel, D. Marme, H. G. Augustin, and U. Fiedler Expression of Angiopoietin-2 in Endothelial Cells Is Controlled by Positive and Negative Regulatory Promoter Elements Arterioscler. Thromb. Vasc. Biol., October 1, 2004; 24(10): 1803 - 1809. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Wang, Y. Peng, Y. Ma, and N. Jahroudi Histone H1-like protein participates in endothelial cell-specific activation of the von Willebrand factor promoter Blood, September 15, 2004; 104(6): 1725 - 1732. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Minami, J. A. Kuivenhoven, V. Evans, T. Kodama, R. D. Rosenberg, and W. C. Aird Ets Motifs Are Necessary for Endothelial Cell-Specific Expression of a 723-bp Tie-2 Promoter/Enhancer in Hprt Targeted Transgenic Mice Arterioscler. Thromb. Vasc. Biol., November 1, 2003; 23(11): 2041 - 2047. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-H. Cheng and S. Handwerger Identification of an Enhancer of the Human Activating Protein-2{alpha} Gene That Contains a Critical Ets1 Binding Site J. Clin. Endocrinol. Metab., July 1, 2003; 88(7): 3305 - 3311. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Peng and N. Jahroudi The NFY Transcription Factor Inhibits von Willebrand Factor Promoter Activation in Non-endothelial Cells through Recruitment of Histone Deacetylases J. Biol. Chem., February 28, 2003; 278(10): 8385 - 8394. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Elvert, A. Kappel, R. Heidenreich, U. Englmeier, S. Lanz, T. Acker, M. Rauter, K. Plate, M. Sieweke, G. Breier, et al. Cooperative Interaction of Hypoxia-inducible Factor-2alpha (HIF-2alpha ) and Ets-1 in the Transcriptional Activation of Vascular Endothelial Growth Factor Receptor-2 (Flk-1) J. Biol. Chem., February 21, 2003; 278(9): 7520 - 7530. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. ILJIN, T. V. PETROVA, T. VEIKKOLA, V. KUMAR, M. POUTANEN, and K. ALITALO A fluorescent Tie1 reporter allows monitoring of vascular development and endothelial cell isolation from transgenic mouse embryos FASEB J, November 1, 2002; 16(13): 1764 - 1774. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Hadri, A. Ozog, F. Soncin, and A.-M. Lompre Basal Transcription of the Mouse Sarco(endo)plasmic Reticulum Ca2+-ATPase Type 3 Gene in Endothelial Cells Is Controlled by Ets-1 and Sp1 J. Biol. Chem., September 20, 2002; 277(39): 36471 - 36478. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Lelievre, F. Lionneton, V. Mattot, N. Spruyt, and F. Soncin Ets-1 Regulates fli-1 Expression in Endothelial Cells. IDENTIFICATION OF ETS BINDING SITES IN THE fli-1 GENE PROMOTER J. Biol. Chem., July 5, 2002; 277(28): 25143 - 25151. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Gaspar, S. Thai, C. Voland, A. Dube, T. A. Libermann, M.L. Iruela-Arispe, and P. Oettgen Opposing Functions of the Ets Factors NERF and ELF-1 During Chicken Blood Vessel Development Arterioscler. Thromb. Vasc. Biol., July 1, 2002; 22(7): 1106 - 1112. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Marchetti, C. Gimond, K. Iljin, C. Bourcier, K. Alitalo, J. Pouyssegur, and G. Pages Endothelial cells genetically selected from differentiating mouse embryonic stem cells incorporate at sites of neovascularization in vivo J. Cell Sci., May 15, 2002; 115(10): 2075 - 2085. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Oettgen Transcriptional Regulation of Vascular Development Circ. Res., August 31, 2001; 89(5): 380 - 388. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. ILJIN, M. J. KARKKAINEN, E. C. LAWRENCE, M. A. KIMAK, M. UUTELA, J. TAIPALE KATRI PAJUSOLA LEENA ALHONEN, M. HALMEKYTO, D. N. FINEGOLD, R. E. FERRELL, and K. ALITALO VEGFR3 gene structure, regulatory region, and sequence polymorphisms FASEB J, April 1, 2001; 15(6): 1028 - 1036. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Etoh, H. Inoue, S. Tanaka, G. F. Barnard, S. Kitano, and M. Mori Angiopoietin-2 Is Related to Tumor Angiogenesis in Gastric Carcinoma: Possible in Vivo Regulation via Induction of Proteases Cancer Res., March 1, 2001; 61(5): 2145 - 2153. [Abstract] [Full Text]
|
|
|
|
|
|
A. Dube, S. Thai, J. Gaspar, S. Rudders, T. A. Libermann, L. Iruela-Arispe, and P. Oettgen ELF-1 Is a Transcriptional Regulator of the Tie2 Gene During Vascular Development Circ. Res., February 2, 2001; 88(2): 237 - 244. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Ishikawa, S. Baba, T. Katsuya, N. Iwai, T. Asai, M. Fukuda, S. Takiuchi, Y. Fu, T. Mannami, J. Ogata, et al. T+31C Polymorphism of Angiotensinogen Gene and Essential Hypertension Hypertension, February 1, 2001; 37(2): 281 - 285. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E Gustafsson, C Brakebusch, K Hietanen, and R Fassler Tie-1-directed expression of Cre recombinase in endothelial cells of embryoid bodies and transgenic mice J. Cell Sci., January 2, 2001; 114(4): 671 - 676. [Abstract] [PDF]
|
|
|
|
|
|
A. Kappel, T. M. Schlaeger, I. Flamme, S. H. Orkin, W. Risau, and G. Breier |
