| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the Departments of Cardiovascular Research,*
Protein Chemistry,
Molecular
Biology,§
and Pathology,¶
Genentech, South San Francisco, California; and Curagen
Corporation,
New Haven, Connecticut
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
vß3
integrins, have been implicated in various steps of angiogenesis. These
advances have enabled the development of new therapeutic strategies for
inhibiting angiogenesis (eg, to inhibit tumor growth) or promoting
angiogenesis (coronary and peripheral ischemia, wound healing). Understanding the molecular events that direct angiogenesis and the order in which they occur and identifying new pathways that are required for this process are of fundamental importance for all researchers who study angiogenesis. The present study was undertaken to identify the alterations in gene expression that occur in an in vitro model of angiogenesis. In this model, endothelial cells are suspended in a three-dimensional gel composed of type I collagen and incubated with a mixture of stimuli (phorbol myristate acetate (PMA), basic fibroblast growth factor (bFGF), and vascular endothelial cell growth factor (VEGF)). Previous studies by our laboratory demonstrated that this combination of stimuli resulted in the optimal formation of a three-dimensional tubular network of endothelial cells with interconnecting lumenal structures.1 In this model, endothelial differentiation into tubelike structures is completely blocked by inhibitors of new mRNA (actinomycin D) or protein synthesis (cycloheximide). Furthermore, the cells progress through this differentiation process in a coordinated and synchronized manner, thus optimizing the profile of gene expression.
The goal of the present study was to identify a molecular fingerprint or transcriptional profile of endothelial differentiation into tubelike structures, using amplification and an imaging approach called GeneCalling.2 This method was previously shown to provide a comprehensive sampling of cDNA populations in conjunction with the sensitive detection of quantitative differences in mRNA abundance for both known and novel genes.2 We describe the identification of 115 differentially expressed cDNA fragments, which corresponded to 90 previously identified genes. The identification and differential expression of these genes was confirmed by a second independent method employing real-time quantitative polymerase chain reaction (PCR). Although some of the cDNA fragments identified were genes previously known to play some role in the process of angiogenesis, many other differentially expressed genes were unexpected and suggest possible roles for these additional genes in endothelial differentiation and vessel assembly.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Human umbilical vein endothelial cells (HUVECs) were purchased from Clonetics (San Diego, CA) and maintained in endothelial growth medium (EGM) media supplemented to a final concentration of 10% fetal bovine serum. Type I rat tail collagen was from Upstate Biotechnology (Lake Success, NY), and recombinant bFGF was purchased from Collaborative Biomedical Products (Becton Dickinson Labware, Bedford, MA). Recombinant VEGF was from Genentech (South San Francisco, CA). Medium 199 (10x) (M199, M0650), PMA, ITS (insulin, transferrin, and selenium-A), trypsin, actinomycin D, and cycloheximide were from Gibco-BRL (Gaithersburg, MD).
Formation of Three-Dimensional Collagen Gels
Collagen gels were formed by mixing together ice-cold gelation solution (10x M199, H2O, 0.53 mol/L NaHCO3, 200 mmol/L L-glutamine, type I collagen, 0.1 mol/L NaOH, 100:27.7:50:10:750:62.5 v/v) and cells in 1x basal medium (see below) at a concentration of 3 x 106 cells/ml at a ratio of four volumes gelation solution to one volume of cells. The gels were allowed to form by incubation in a CO2-free incubator at 37°C for 30 minutes to 1 hour. The gels were then overlaid with 1x basal medium consisting of M199 supplemented with 1% FBS, 1x ITS, 2 mmol/L L-glutamine, 50 µg/ml ascorbic acid, 26.5 mmol/L NaHCO3, 100 U/ml penicillin, and 100 U/ml streptomycin. In the tube-forming experiments, the culture medium was supplemented with 80 nmol/L PMA, 40-ng/ml bFGF, and 40 ng/ml VEGF.
mRNA Isolation and cDNA Synthesis
Medium was aspirated from the surface of the collagen gels, and the gels were scraped into a 50-ml polypropylene tube containing three volumes of Tri-Reagent-LS (Molecular Research Center, Cincinnati, OH). The tubes were incubated for 10 minutes at 23°C with intermittent gentle agitation. The tubes were stored at -80°C until all experimental samples had been collected. The tubes were then thawed at room temperature, and the RNA was extracted following the manufacturer’s specifications. The RNA pellets were resuspended in diethyl-pyrolidine-carbonate-treated water, and the RNA content was quantified spectroscopically at 260 nm. RNA samples were stored at -20°C. Samples used for GeneCalling analysis were shipped on dry ice to CuraGen (New Haven, CT). Samples from time points of 4, 24, and 48 hours were used for the GeneCalling analysis, and in separate experiments, samples from additional time points of 30 minutes and 2, 4, 8, 16, 24, 38, and 46.5 hours were prepared for TaqMan confirmation. For the quantitative expression analysis, contaminating DNA was removed by treatment of the isolated RNA with DNase I (Promega, Madison, WI). PolyA+ RNA was prepared by fractionation of total RNA with an mRNA purification kit that uses the biotinylated oligo-dT-streptavidin magnetic bead method (MPG, Lincoln Park, NJ), followed by cDNA synthesis by reverse transcription of oligo-dT-primed mRNA (Superscript II; Life Technologies) and second-strand synthesis. Terminal phosphate removal is achieved by treatment with arctic shrimp alkaline phosphatase (Amersham Life Sciences, Piscataway, NJ), followed by purification of cDNA by phenol-chloroform extraction. Yield of cDNA was quantitated by fluorometry using PicoGreen dye (Molecular Probes, Eugene, OR). Double-stranded DNA was digested using pairs of restriction enzymes with 6-bp recognition sites. More than 48 enzyme pairs were used and were chosen such that a representative coverage of most of the possible sequences in a given DNA sample was achieved.2 PCR amplification using specific linkers was carried out as described previously.2 The final DNA products were denatured by heating to 96°C and electrophoresed on ultrathin polyacrylamide gels under denaturing conditions in 6 mol/L urea. PCR products were visualized by the presence of 6-carboxy fluorescein (FAM) label on the product, using a multicolor laser excitation (Niagara; CuraGen, New Haven CT) imaging system.
Data Interpretation
The data obtained from Niagara gels were queried (ie, "GeneCalled") against public and proprietary databases.2 GeneCalling is the process that takes the restriction enzyme pair recognition site information and the cDNA fragment size determined from the migration of the labeled fragment on Niagara gels and uses that information (the size of the fragment and the relative position of the terminal sequences defined by the restriction enzyme pairs) to search public and proprietary databases for likely gene matches, using statistical and mathematical criteria. A GeneCall is defined as the probability of a cDNA fragment belonging to a known gene.2 The cDNA fragment data were compiled as a list of likely genes to which that cDNA fragment might belong. If a provisional identification of a cDNA fragment could not be obtained by querying databases, the cDNA fragment was designated as belonging to a putative novel gene.
Confirmation of Gene Calls
GeneCalls were confirmed in a competitive PCR reaction, "GeneCall poisoning," in which the known sequence of the likely gene of interest is used to design poisoning primers as previously described.2 Ablation of the cDNA fragment of interest confirmed that the cDNA fragment belonged to the gene for which the specific poisoning primer was designed.
Novel cDNA Fragments
If no GeneCall was obtained for a cDNA fragment, the cDNA fragment was eluted and subcloned into Escherichia coli with the standard TA-cloning vector (Invitrogen, Palo Alto, CA). The cDNA fragment was then sequenced, and the resulting sequence was used to design poisoning primers for confirmation as described above.
Validation and Confirmation of Gene Expression by Quantitative Reverse Transcriptase-Polymerase Chain Reaction (TaqMan)
To confirm the expression data from GeneCalling by an independent
technique, gene-specific PCR oligonucleotide primer pairs and an
oligonucleotide probe labeled with a reporter fluorescent dye at the 5'
end and quencher fluorescent dye at the 3' end were designed using
Oligo 4.0 software (National Bioscience, Plymouth, MN). Table 1
provides the sequences for
the primers and probes used in this study. Total RNA (50 ng) was added
to a 50 µl reverse transcriptase-polymerase chain reaction (RT-PCR)
reaction mixture according to the manufacturer’s protocol (Roche
Molecular Systems, Branchburg, NJ). The thermal cycling conditions
included one cycle at 48°C for 30 minutes, one cycle at 95°C for 10
minutes, 40 cycles at 95°C for 15 s, annealing at 60°C for 1
minute, and a final hold at 25°C for 2 minutes. Standard curves for
the expression of each gene were generated by serial dilution of a
standard preparation of total RNA isolated from quiescent HUVECs grown
in monolayer culture. Data are expressed as the fold induction
normalized to the same gene from quiescent HUVEC RNA.
|
mRNA was harvested from endothelial cells incubated 4 and 24 hours
in the absence (control) or presence of 10 µmol/L of the PPAR
ligand 15-deoxy-
2,14–15-prostaglandin J2
(15d-PGJ2). In both groups the cells were
incubated with the mixture of growth stimuli (ie, PMA, VEGF, and bFGF),
and the cells were incorporated in the collagen gels as described
above.
In Situ Hybridization of Tissue Specimens
Formalin-fixed, paraffin-embedded human tissues were investigated for in situ mRNA expression. Tissues included first-trimester (14–15-week) placenta, adult adrenal cortex, aorta, muscular artery with atherosclerosis, brain, gall bladder, heart, pancreas, prostate, stomach, eye with age-related macular degeneration (AMD), inflamed appendix, pulmonary adenocarcinoma, ductal mammary adenocarcinoma, kidney with renal cell carcinoma, hepatocellular carcinoma, squamous cell carcinoma, osteosarcoma, and chondrosarcoma. In vitro transcription and 33P labeling of sense and antisense riboprobes were performed as described previously.3 Briefly, stanniocalcin, osteonidogen, podocalyxin, and ADAMTS-4 sequences were PCR-amplified from plasmid DNA, using gene-specific primers that encoded T3 or T7 RNA polymerase initiation sites. Sense and antisense riboprobes were prepared by in vitro transcription from the PCR-amplified templates and diluted in hybridization buffer to a specific activity of 1 x 106 cpm/ml. Tissue sections 5 µm thick were deparaffinized, deproteinated in 4 µg/ml of proteinase K for 30 minutes at 37°C, hybridized at 55°C overnight, then washed at high stringency (55°C in 0.1x standard saline citrate for 2 hours). Glass slides were dipped in NBT2 nuclear track emulsion (Eastman Kodak), exposed in sealed plastic slide boxes containing desiccant for 4 weeks at 4°C, developed, and counterstained with hematoxylin and eosin.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
As reported previously,1
incubation of endothelial
cells in 3D gels in the absence of the growth factors resulted in rapid
induction of apoptosis. Therefore, no comparison was made of mRNA from
cells in 3D gels in the absence of growth factors. Instead we evaluated
temporal changes in gene expression in the 3D gel environment in the
presence of PMA, VEGF, and bFGF, by comparison of the RNA harvested at
4, 24, and 48 hours. A summary of the differences observed can be found
in Table 2
.
The differentially expressed cDNA fragments in the 24-hour
versus the 4-hour data set were examined in more detail. As
shown in Table 3
, the identities of 115
cDNA fragments were determined by oligonucleotide poisoning or cloning
of the gene fragments, resulting in the identification of 90 distinct
genes. In addition (not shown), 80 cDNA fragments were identified as
totally novel or as corresponding to expressed sequence tags
(ESTs) of unknown function. Full-length cloning of these genes
is currently under way.
|
|
TaqMan probes were prepared to confirm by an independent method
the identification of 67 of the genes identified by GeneCalling. As
shown in Table 3
, there was a reasonable agreement in the direction of
the fold induction as predicted by GeneCalling and as analyzed by
TaqMan. Disagreement between the GeneCalling and TaqMan results are
likely due to small differences in the temporal sequence of events in
the two separate experiments (ie, the RNA harvested for the GeneCalling
experiment versus the RNA harvested for later TaqMan
analysis), which could readily account for the discrepancies in the
fold induction. A more complete time course analysis of 26 of the genes
is depicted in Figure 1, A–D
.
|
According to the classification schema used, the genes identified
fell into most of the major role categories, including cell division,
cell signaling, cell adhesion, hormone/growth factors, receptors,
cytoskeleton, extracellular matrix, protein turnover, protein
modification, and metabolism (Table 3)
. There was no apparent bias in
the identification of any given class of gene.
The mRNA changes were clustered based on four basic patterns of
expression. Group I, Early Transient (Figure 1A)
, which included
interleukin-8 (IL-8), binding protein A1, plasminogen activator
inhibitor-2 (PAI-2), growth-related oncogene (GRO-), and
cyclooxygenase-2 (COX-2), was characterized by mRNA levels that were
rapidly and highly induced then declined to the initial levels within
24 hours. Peak mRNA levels were observed at 2–4 hours. Not shown in
Figure 1A
because of the small magnitude of fold induction, is an EST
with homology to the fibrinogen domain of angiopoeitin-2 (AI79861). The
message levels for this EST increased by twofold by 4 hours then
returned to baseline levels between 8 and 12 hours. Group II, Delayed
Transient, which included the genes white protein homolog, fibroblast
growth factor-16 (FGF16), KIAA0188, ADAMTS-4 (aggrecanase-1), tissue
factor pathway inhibitor-2 (TFPI-2), podocalyxin-like protein,
cathepsin B, and epithelial tyrosine kinase (ECK), was characterized by
mRNA levels that peaked somewhat later than those of Group I (8–12
hours), then fell back to near-baseline levels by 46.5 hours. Group
III, Stable Induction (Figure 1C)
, was characterized by genes whose
mRNA levels rise somewhat later than those of Group I or Group II,
peaked at 12–46.5 hours, and remained markedly above baseline levels,
even at 46.5 hours. This group included placental growth factor, yLAT1,
clone 23689, osteonidogen, matrix metalloproteinase 9 (MMP-9), CXC
chemokine receptor 4 (CXCR4), and stanniocalcin precursor (STC). Group
IV, Rapid Repression (Figure 1D)
, was quite different from Groups
I–III. mRNA levels for genes in Group IV declined from the initial
value observed at 30 minutes and, by 16–24 hours, were below the mRNA
levels observed in the mRNA controls obtained from quiescent HUVEcs
(see Materials and Methods). Genes in Group IV included extracellular
protein S1–5 (S1–5), axl, polo-like kinase, and mesothelial keratin.
PPAR Modulation of Endothelial Gene Expression
Treatment of endothelial cells with 10 µmol/L
15-d-PGJ2 completely blocks endothelial tube
formation in response to bFGF, VEGF, and PMA.4
As shown in
Table 4
, 15d-PGJ2
treatment reduced or, in some cases, abrogated the fold increase (24
versus 4 hours) of PLGF, clone 23689, STC, OX-40, TFPI-2,
MMP-9, KIAA0188, +yLAT1, laminin 2, and
PLA2. In addition,
15d-PGJ2 reduced the fold decrease (24
versus 4 hours) of IL-8, axl, PAI-2, white protein homolog,
and keratin K-7 observed during tube formation. The ratios of other
mRNAs (A1, FGF16, eck, podocalyxin, and osteonidogen, for example) were
not markedly affected by 15d-PGJ2.
|
As an additional test of the biological relevance of the genes
identified, we evaluated the in situ expression of four of
the genes identified in this study, ie, STC, podocalyxin, osteonidogen,
and ADAMTS-4, by examining their expression in a number of different
tumors as well as sections prepared from a variety of human organs (see
Materials and Methods). All four genes were detected at sites of
endothelial activation or new blood vessel formation. For example, STC
demonstrated strong but variable expression in the vasculature in and
around mammary adenocarcinoma and squamous cell carcinoma (Figure 2)
and, to a lesser extent, in
chondrosarcoma and renal cell carcinoma (not shown), but there was no
significant expression seen in normal vessels (not shown). Detectable
expression of podocalyxin in normal adult tissue was limited to
glomerular urinary epithelial cells (podocytes) and some endothelial
cells in the adventitia around large vessels. Podocalyxin expression
was expressed in the endothelium of small vessels associated with
chondrosarcoma, squamous and renal cell carcinomas, and ductal mammary
adenocarcinoma. (not shown), as well as in arteriolar endothelium in
inflamed appendix (Figure 2)
. Osteonidogen expression was absent in
normal adult vessels but was observed in endothelial cells of inflamed
appendix (not shown) and in peritumor stromal (Figure 2)
endothelium
and nonendothelial cell types, as well as in osteosarcoma,
chondrosarcoma, and squamous cell CA tumors. ADAMTS-4 expression in
adult tissue was intensely expressed in vascular endothelium and smooth
muscle in areas of inflammation (appendices, around tumors, in inflamed
lung) (Figure 2)
, as well as in scleral and corneal limbic endothelium
in an age-related macular degeneration eye (not shown). No detectable
expression of ADAMTS-4 was observed in blood vessels of normal adult
tissues.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. The
EST AI79861 identified from the GeneCalling analysis is 78% identical
to the COOH-terminal region of human angiopoietin-2, a naturally
occurring antagonist for the tie-2 receptor kinase,5
suggesting that yet another member of the angiopoietin family may play
some role in the regulation of new vessel formation.
TFPI-2, which was a highly up-regulated gene in this study (Figure 1B)
,
is a 32-kd serine protease and is associated with the extracellular
matrix that inhibits the activation of matrix metalloproteinase
zymogens, pro-MMP-1 and MMP-3.6,7
TFPI has also been
reported to be a smooth muscle mitogen.8
Thus the
up-regulation of TFPI-2 by the differentiation of endothelial cells
might have a dual role of limiting the extent of matrix degradation and
recruiting or promoting the proliferation of mural cells, leading to
the assembly of the new vessel wall.
The mRNA levels for a number of proteases increased substantially
during the initial 8–10 hours of incubation in the gel, notably the
matrix metalloproteinase, MMP-9 (Figure 1C)
, and cathepsin B (Figure 1B)
. MMP-9 is an established participant in angiogenesis, playing
important roles in the degradation of basement membrane/matrix in both
in vitro and in vivo models of
angiogenesis.9,10
MMP inhibitors reduce the elongation of
endothelial cells into tubelike structures in
vitro11
and inhibit tumor angiogenesis in in
vivo animal models12
and in cancer
patients.13
Cathepsin B, a lysosomal cysteine protease,
has been observed in many different tumor types and is highly expressed
in tumor blood vessels as compared to normal
vasculature.14
One of the newest members of "disintegrin
and metalloproteinase family members with thrombospondin
motifs," ADAMTS-4 was identified as a differentially expressed gene
in this study. Although relatively little information is available
concerning ADAMTS-4, Tortorella et al15
recently reported
that this protein has aggrecanase activity and suggested that ADAMTS-4
played an important role in the turnover of the proteoglycan aggrecan
in diseases such as osteoarthritis. Aggrecan has not previously been
reported in vascular tissues, and we were unable to detect the
expression of aggrecan mRNA in our endothelial cells under a variety of
experimental conditions (data not shown). Intriguingly, however, two
other members of the ADAM-TS family, METH-1 (human ADAMTS-1) and METH-2
(ADAMTS-8), were recently reported to be potent antiangiogenic
agents.16
The up-regulation of ADAMTS-4 during endothelial
tube formation suggests a potential role of this enzyme in the
metabolism of vascular proteoglycans, such as versican or other
components of the basement membrane.
The mRNAs for the chemokines IL-8 and Gro- are up-regulated early in
the time course of endothelial differentiation into tubelike structures
in the 3D gel environment (Figure 1A)
. These chemokines have previously
been reported to have angiogenic activity.17
mRNA levels
for CXCR4, the chemokine receptor for SDF-1,18
increased by nearly 40-fold over a time period between 30 minutes and
16 hours, then remained elevated for the duration of the experiment
(Figure 1C)
. Deletion studies have shown that both CXCR4 and SDF-1 null
mice have defective formation of large blood vessels supplying the
gastrointestinal tract.19
The mRNA level for the receptor "ECK" or epithelial tyrosine kinase
(EphA2), a Eph receptor kinase family, increased modestly between 4 and
8 hours, and then slowly declined toward the levels observed at 30
minutes (Figure 1B)
. The ligand for ECK is a gene known as B61 (ephrin
A1), initially identified as a tumor necrosis factor-
(TNF)-induced gene in endothelial cells.20
Antibodies
to B61 block TNF-induced angiogenesis in a corneal
neovascularization assay.21
The mRNA for the laminin 2 increased by almost 1000-fold between 0.5
and 24 hours (Figure 1B)
. Laminins are known to play key roles in
angiogenesis22
and are an important component of the
basement membrane. Although the precise role of laminin in the
differentiation process is unclear, laminin might be important for the
establishment of the endothelial apical:basal polarity that may be a
prerequisite for the formation of a vascular lumen. The message levels
for another basement membrane protein, osteonidogen (also known as
Nidogen-2), increased throughout the duration of the HUVEC incubation
in the 3D gels (Figure 1C)
. Nidogens can interact with laminin,
collagen, and proteoglycans, and there are suggestions that this family
of molecules may play some role in angiogenesis.23
In contrast to the genes discussed above, which have previously been implicated in various steps of angiogenesis, the study also identified a number of genes not previously associated with endothelial cell proliferation, differentiation, or angiogenesis.
The mRNA, designated KIAA0188, codes for a novel hypothetical protein.
Domain analysis of the predicted amino acid sequence revealed a
putative domain, aa 724–732, with homology to the consensus
Kunitz-type serine protease inhibitor, and a domain, 455–467, with
homology to the subtilase family of serine proteases, suggesting that
this gene might code for a proform of a serine protease. KIA0188 mRNA
levels increased modestly during the initial 8–10 hours of incubation
in the gel, then declined to the initial levels (Figure 1B)
.
STC was first isolated from bony fishes,24
where this
glycoprotein is synthesized and secreted by the corpuscles of Stannius
and regulates blood calcium levels through its inhibitory action on
calcium ion uptake in the gill, a highly vascularized
tissue.24
Mammalian cDNAs encoding STC have been reported
for the human and mouse and are highly homologous to those of the
fish,25
although the role of this protein in mammals is
not known. mRNA levels for STC increased over 100-fold between 30
minutes and 16 hours and remained well above baseline levels out to
46.5 hours (Figure 1C)
. What role this gene plays in angiogenesis is
unknown, but this observation suggests that further study in the
context of angiogenesis is warranted.
The mRNA for FGF-16 demonstrated a biphasic profile, increasing
substantially during the initial 8 hours and then declining for the
remainder of the experiment (Figure 2B
. This recent member of the FGF
family of growth factors was originally cloned from human heart
cDNA.26
FGF-16 weakly stimulates NIH 3T3 fibroblast
proliferation and but is a reasonably potent stimulus of primary rat
oligodendrocyte proliferation.27
The mRNA levels of a number of membrane receptors were also increased over the time course of endothelial differentiation into tubelike structures, including the signaling component of the IL-6 receptor gp130, the TNFR-related protein OX40,28 and the sialomucin, podocalyxin-like protein. There is no known role for any of these receptors in angiogenesis. Indeed, this study documents for the first time the expression of OX40 by endothelial cells. Previously this TNFR family member was thought to be restricted to cells of the lymphocyte lineage.29 Increased protein expression of OX40 was also confirmed by fluorescence-activated cell sorter analysis (data not shown).
Podocalyxin-like protein, a well-known constituent of the endothelial plasma membrane,30 was recently shown protein to function as an L-selectin receptor in inflamed lymph nodes,31 suggesting a role in cell-cell interactions or adhesion.
The mRNA levels for the antiapoptotic bcl-2-related protein A1 were
elevated at the early time points and then declined (Figure 1A)
. In the
three-dimensional gel environments, HUVECs do not survive well in the
absence of growth factors, and they cannot be rescued by
supplementation with VEGF or bFGF.1
However, PMA treatment
will induce endothelial survival and tubule formation. The induction of
A1 expression may thus be related to inhibition of apoptosis in the 3D
gel environment.
The protein designated "white protein homolog" (also known as
ATP-binding cassette (ABC) 8) is 84% identical to the
Drosophila gene white protein, which codes for a transporter
protein whose expression results in white eye color. Many members of
the ABC family of proteins function as transporters or channels. The
mRNA levels for this gene increased by ~10-fold between 0.5 and 8
hours (Figure 1B)
. There is little information relating to its
expression or function in mammalian cells, although ESTs containing
white protein sequence from a variety of tissue sources and tumors can
be found in GenBank. The mRNA for y+LAT-1, a new member of a family of
polytopic transmembrane proteins,32
increased by about
eightfold between 0.5 and 24 hours (Figure 1C)
. Little is known about
the function of this permease, although it undoubtedly plays a role in
amino acid transport and protein synthesis, two activities critical to
altered endothelial protein expression.
Cyclooxygenase 2 (COX-2), a rate-limiting enzyme in the prostaglandin
biosynthesis pathway, was detected in our experiment system. COX2 mRNA
levels rose abruptly during the initial few hours of the experiment,
then declined (Figure 1A)
. The transient expression we observed for
COX2 is consistent with published reports proposing a role for
COX2-regulated prostanoid responses after vascular
injury.33
Xin et al4
recently reported that agonists of PPAR
receptors specifically blocked endothelial tube formation in
vitro and VEGF-driven angiogenesis in vivo.
Furthermore, Xin et al4
found that treatment of
endothelial cells with the PPAR ligand
15d-PGJ2 inhibited the induction of kdr, flt-1,
and uPA in a three-dimensional collagen gel model identical to the
system used in the present investigation. The effects of
15d-PGJ2 on mRNA levels for a number of genes
identified in this study were therefore examined for the purpose of
identifying genes modulated during tube formation specifically
responsive to treatment with the PPAR ligand. Those genes so
identified might represent important targets for therapeutic
intervention. As shown in Table 4
, there were different classes of
response to the PPAR ligand (supermodulation, countermodulation, and
no change in modulation), suggesting that the effect of treatment with
15d-PGJ2 was not simply a general phenomenon. The
selective and marked effects of 15d-PGJ2
treatment on endothelial gene expression in the three-dimensional
collagen system provides further support for the potential roles in
angiogenesis of many of the genes identified in this study.
In summary, GeneCalling successfully identified 115 differentially
expressed cDNA fragments corresponding to 90 known genes from the study
of collagen matrix-driven endothelial cell gene expression. In addition
to the known genes identified, 80 fragments considered totally novel,
or belonging to ESTs of unknown function, were identified in this
study. The identity and expression of 67 of the known genes were
confirmed by a second independent method (TaqMan). For the initial
confirmations, we focused on membrane proteins and secreted proteins
and only confirmed the identity of a few of the other cDNA fragments by
this independent technique. However, in every instance the gene
identified by the GeneCalling method was confirmed by TaqMan to be
expressed by HUVECs in 3D gels, and the magnitude and direction of the
changes in expression agreed reasonably well with the GeneCalling
estimates. Because the method also identified a number of potentially
new genes, we have not, at this time, pursued TaqMan confirmation
of the remaining 23 genes, choosing to focus, instead, on the
identification of new genes that might play a role in the process of
endothelial differentiation into tubelike structures. Most importantly,
the biological relevance of many of these newly identified
"angiogenesis-associated" genes is strongly supported by the
selective abrogation of their differential expression by the
PPARligand, 15d-PGJ2, as well as by the
in situ demonstration of selective expression of some of the
genes at sites of new blood vessel formation. Although many of the
identified genes have previously been associated with angiogenesis or
tumor vasculature (eg, cathepsin B, MMP-9, PLGF, IL-8, GRO-, CXCR4)
or have reported roles or expression patterns consistent with a
function in the differentiation process (A1, TFPI-2, laminin 2), a
number of genes identified in this study had never previously been
associated with angiogenesis (stanniocalcin precursor, OX40, white
protein homolog, the angiopoietin-2 homolog, ADAMTS-4, FGF16,
KIAA0188), suggesting the need for the further evaluation of the
potential biological roles of these genes in the process of new blood
vessel formation
|
|
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Ms. Kahn and Dr. Mehraban contributed equally to this study.
Accepted for publication February 9, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  E. Wallgard, E. Larsson, L. He, M. Hellstrom, A. Armulik, M. H. Nisancioglu, G. Genove, P. Lindahl, and C. Betsholtz Identification of a Core Set of 58 Gene Transcripts With Broad and Specific Expression in the Microvasculature Arterioscler. Thromb. Vasc. Biol., August 1, 2008; 28(8): 1469 - 1476. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Chakraborty, H. Brooks, P. Zhang, W. Smith, M. R. McReynolds, J. B. Hoying, R. Bick, L. Truong, B. Poindexter, H. Lan, et al. Stanniocalcin-1 regulates endothelial gene expression and modulates transendothelial migration of leukocytes Am J Physiol Renal Physiol, February 1, 2007; 292(2): F895 - F904. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. H. Chidlow Jr., W. Langston, J. J.M. Greer, D. Ostanin, M. Abdelbaqi, J. Houghton, A. Senthilkumar, D. Shukla, A. P. Mazar, M. B. Grisham, et al. Differential Angiogenic Regulation of Experimental Colitis Am. J. Pathol., December 1, 2006; 169(6): 2014 - 2030. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. C. Aplin, M. Gelati, E. Fogel, E. Carnevale, and R. F. Nicosia Angiopoietin-1 and vascular endothelial growth factor induce expression of inflammatory cytokines before angiogenesis Physiol Genomics, October 3, 2006; 27(1): 20 - 28. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. M. Borthwick, A. S. Johnson, M. Partington, J. Burn, R. Wilson, and H. M. Arthur Therapeutic levels of aspirin and salicylate directly inhibit a model of angiogenesis through a Cox-independent mechanism FASEB J, October 1, 2006; 20(12): 2009 - 2016. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. W.M. van Hinsbergh, M. A. Engelse, and P. H.A. Quax Pericellular Proteases in Angiogenesis and Vasculogenesis Arterioscler. Thromb. Vasc. Biol., April 1, 2006; 26(4): 716 - 728. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Ryschich, P. Lizdenis, C. Ittrich, A. Benner, S. Stahl, A. Hamann, J. Schmidt, P. Knolle, B. Arnold, G. J. Hammerling, et al. Molecular Fingerprinting and Autocrine Growth Regulation of Endothelial Cells in a Murine Model of Hepatocellular Carcinoma Cancer Res., January 1, 2006; 66(1): 198 - 211. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Bohman, T. Matsumoto, K. Suh, A. Dimberg, L. Jakobsson, S. Yuspa, and L. Claesson-Welsh Proteomic Analysis of Vascular Endothelial Growth Factor-induced Endothelial Cell Differentiation Reveals a Role for Chloride Intracellular Channel 4 (CLIC4) in Tubular Morphogenesis J. Biol. Chem., December 23, 2005; 280(51): 42397 - 42404. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. C.-M. Chang, J. Cha, F. Koentgen, and R. R. Reddel The Murine Stanniocalcin 1 Gene Is Not Essential for Growth and Development Mol. Cell. Biol., December 1, 2005; 25(23): 10604 - 10610. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Y. Yeung, K. P. Lai, H. Y. Chan, N. K. Mak, G. F. Wagner, and C. K. C. Wong Hypoxia-Inducible Factor-1-Mediated Activation of Stanniocalcin-1 in Human Cancer Cells Endocrinology, November 1, 2005; 146(11): 4951 - 4960. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Paciga, E. R. Hirvi, K. James, and G. F. Wagner Characterization of big stanniocalcin variants in mammalian adipocytes and adrenocortical cells Am J Physiol Endocrinol Metab, August 1, 2005; 289(2): E197 - E205. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. N. Hahn, Z. J. Su, C. J. Drogemuller, A. Tsykin, S. R. Waterman, P. J. Brautigan, S. Yu, G. Kremmidiotis, A. Gardner, P. J. Solomon, et al. Expression profiling reveals functionally important genes and coordinately regulated signaling pathway genes during in vitro angiogenesis Physiol Genomics, June 16, 2005; 22(1): 57 - 69. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Yin, R. G. Russell, L. E. Dettin, R. Bai, Z.-L. Wei, A. P. Kozikowski, L. Kopleovich, and R. I. Glazer Peroxisome Proliferator-Activated Receptor {delta} and {gamma} Agonists Differentially Alter Tumor Differentiation and Progression during Mammary Carcinogenesis Cancer Res., May 1, 2005; 65(9): 3950 - 3957. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. A.J. Kempf, M. Lebiedziejewski, K. Alitalo, J.-H. Walzlein, U. Ehehalt, J. Fiebig, S. Huber, B. Schutt, C. A. Sander, S. Muller, et al. Activation of Hypoxia-Inducible Factor-1 in Bacillary Angiomatosis: Evidence for a Role of Hypoxia-Inducible Factor-1 in Bacterial Infections Circulation, March 1, 2005; 111(8): 1054 - 1062. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Manalo, A. Rowan, T. Lavoie, L. Natarajan, B. D. Kelly, S. Q. Ye, J. G. N. Garcia, and G. L. Semenza Transcriptional regulation of vascular endothelial cell responses to hypoxia by HIF-1 Blood, January 15, 2005; 105(2): 659 - 669. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Print, R. Valtola, A. Evans, K. Lessan, S. Malik, and S. Smith Soluble factors from human endometrium promote angiogenesis and regulate the endothelial cell transcriptome Hum. Reprod., October 1, 2004; 19(10): 2356 - 2366. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Palmeri, F.-R. Zuo, S. D. Rosen, and S. Hemmerich Differential gene expression profile of human tonsil high endothelial cells: implications for lymphocyte trafficking J. Leukoc. Biol., May 1, 2004; 75(5): 910 - 927. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Zlot, G. Ingle, J. Hongo, S. Yang, Z. Sheng, R. Schwall, N. Paoni, F. Wang, F. V. Peale Jr., and M. E. Gerritsen Stanniocalcin 1 Is an Autocrine Modulator of Endothelial Angiogenic Responses to Hepatocyte Growth Factor J. Biol. Chem., November 28, 2003; 278(48): 47654 - 47659. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Liu, H. Jia, D. I. R. Holmes, A. Stannard, and I. Zachary Vascular Endothelial Growth Factor-Regulated Gene Expression in Endothelial Cells: KDR-Mediated Induction of Egr3 and the Related Nuclear Receptors Nur77, Nurr1, and Nor1 Arterioscler. |
