Vol. 280, Issue 5, H2116-H2125, May 2001
In vivo chemotactic properties and spatial expression of
PDGF in developing mesenteric microvascular networks
Peter J.
Zeller,
Thomas C.
Skalak,
Ana M.
Ponce, and
Richard J.
Price
Department of Biomedical Engineering, University of Virginia,
Charlottesville, Virginia 22908
 |
ABSTRACT |
The recruitment of
perivascular cells to developing microvessels is a key component of
microvessel assembly. Whereas platelet-derived growth factor (PDGF)
signaling is critical for this process during embryonic development,
its role from the postnatal stages through adulthood remains unclear.
We investigated the potential role of PDGF signaling during microvessel
assembly by measuring in vivo the migration of labeled fibroblasts to
PDGF in mesenteric connective tissue and by examining PDGF-B and PDGF
receptor-
(PGDFR-) expression in microvascular networks during
normal maturation. PDGF-B homodimer (PDGF-BB; 30 ng/ml) application
elicited a significant (P < 0.05) increase (7.8 ± 4.1 cells) in labeled fibroblasts within 100 µm of the
source micropipette after 2 h. PDGF-A homodimer (30 ng/ml)
application and control solution did not elicit directed migration.
PDGF-B was expressed in microvessel endothelium and smooth muscle,
whereas PDGFR- was expressed in endothelium, smooth muscle, and
interstitial fibroblasts. Given that PDGF-BB elicits fibroblast
migration in the mesentery and that PDGF-B and PDGFR- are expressed
in a pattern that indicates paracrine signaling from microvessels to
the interstitium, the results are consistent with a role for PDGF-B in
perivascular cell recruitment to microvessels.
platelet-derived growth factor; vascular remodeling; microcirculation; cell migration; arterialization
| |
INTRODUCTION |
THE FORMATION OF PROPERLY
PATTERNED and functional microvascular networks is critically
dependent on the recruitment of perivascular supporting cells to the
abluminal surface of preexisting vessels. Microvessels destined to
become arterioles and venules require subsequent phenotypic modulation
of these precursor cells into smooth muscle (SM). These basic
microvessel assembly processes are tightly regulated by the controlled
spatial and temporal expression of selected growth factors and their
receptors. Members of the platelet-derived growth factor (PDGF) family
of growth factors and their receptors have been implicated in the
formation of arterioles by stimulating the paracrine recruitment of
supporting cells and SM cell progenitors in the in vivo environment.
The PDGF family of growth factors consists of homodimers (PDGF-AA
and PDGF-BB) or heterodimers (PDGF-AB) that are synthesized and
secreted by many cell types, including endothelial cells (7, 28). They are chemotactic for SM cells (16, 2, 10)
and fibroblasts (21, 11, 12) in vitro, and coculture
studies have shown that endothelial expression of PDGF-B elicits the
directed migration of undifferentiated SM precursor cells
(7). Examination of embryos lacking PDGF-B indicates that
throughout many microvascular beds, capillaries lack pericytes and
develop microaneurysms (14). However, the appearance of
normal large arteries such as the aorta has suggested that large artery
vasculogenesis might be unaffected by PDGF disruption
(24). Further evidence that PDGF mediates arteriole
formation was indicated by examining the net advantage conferred by expression of PDGF receptor- (PDGFR-) in specific cell lineages through chimeric analysis (3). Here it was
found that cells lacking PDGFR- composed only 15% of the SM cells
in the aorta, suggesting that PDGFR- is necessary for the SM cell investment associated with large vessel development. This
chimeric study did not, however, differentiate between molecular
mechanisms occurring during embryogenesis from those occurring during
postnatal remodeling. A later study (4) using the same
chimeric mouse model began to address this issue by demonstrating that
endothelial and fibroblast participation in connective tissue formation
is dependent on PDGFR-, thereby providing evidence that PDGF
signaling is recapitulated in the adult during wound healing. The role
of PDGF-B in normal postnatal development remains, however, unclear. Furthermore, despite the implications of PDGF as a perivascular cell
recruitment factor, no direct experiments demonstrating in vivo
chemotaxis of fibroblasts in response to PDGF have been performed.
We investigated the potential role of PDGF in recruiting perivascular
cells to microvessels by studying the chemotactic properties and
spatial expression patterns of PDGF in rat mesenteric microvascular networks that continue to grow during normal maturation (5, 18). To study the in vivo chemotactic properties of PDGF, we employed the novel technique of injecting labeled fibroblasts into the
mesentery to investigate the direct effect of a point source
application of PDGF-AA and PDGF-BB on migration. In a previous study
(23), labeled fibroblasts were observed to be recruited to
abluminal positions on microvessels of various sizes. This approach has
two important advantages over previous studies. The first is that by
injecting fibroblasts into the animal several days before the
experiment, undesirable effects due to the cell culture environment are
reduced. The second is that our assays were performed in the mesentery,
a true physiological matrix compared with synthetically generated
extracellular matrix substrates or polycarbonate filters such as those
employed in Boyden-type chambers. This second consideration becomes
especially important in light of abundant recent evidence suggesting
that the ability of various cell types to migrate in response to the
selected growth factors may be affected by interactions between, and
the expression of, cellular adhesion molecules, growth factor
receptors, and the underlying extracellular matrix substrate (2,
10, 13, 15, 16, 20, 27). To study the spatial patterns of PDGF
expression in vivo, we immunofluorescently labeled whole mount
mesenteries from normally developing animals for PDGF-B and PDGFR-
and subsequently examined them with the use of confocal microscopy.
Markers for endothelial and SM cells were used to establish
cell-specific expression of PDGF-B and PDGFR- in arterioles,
capillaries, and venules. Our results from these two studies indicate
that the chemotactic properties and spatial expression patterns of
PDGF-B and its receptor are consistent with a role for PDGF-B in
microvessel assembly.
| |
MATERIALS AND METHODS |
Cell harvest, culture, and PKH26-labeling methods.
The following animal procedures were approved by the Animal Care and
Use Committee of the University of Virginia. Animals were anesthetized
with an intramuscular injection of ketamine (80 mg/kg body wt) and
xylazine (8 mg/kg body wt), and a midline 3- to 4-cm incision was made
with a sterile scalpel through the abdomen after shaving and
sterilizing the skin with ethanol. Next, the small intestine was gently
exposed, starting at the junction with the large intestine. The small
intestine was then carefully laid out onto a sterile culture dish, and
the mesenteric windows were examined with the use of a Zeiss Stemi 2000 dissecting microscope at high magnification (×50 overall). Windows
that were observed to have no blood vessels were selected for tissue
culture, thereby ensuring that the selected tissue was lacking SM
cells, which could contaminate the cell cultures. Central
portions of each avascular window at least 2 mm from the surrounding
fat were excised using a sterile scalpel under the dissecting
microscope. Tissues were washed in sterile phosphate-buffered saline
(PBS) containing 200 U/ml penicillin, 0.2 mg/ml streptomycin, and 500 ng/ml amphotericin B for a period of 20 min. Dulbecco's modified
Eagle's medium (DMEM) containing 10% fetal bovine serum (FBS), 100 U/ml penicillin, 0.1 mg/ml streptomycin, and 250 ng/ml amphotericin B
(in DMEM) was then slowly instilled over the tissue. Tissue cultures
were placed in a cell culture incubator maintained at 37°C and 5%
CO2 in air. Medium was changed every 3-4 days.
When the cells were nearly confluent (typically 10-14 days after
the start of culture), the cells were released from the culture dishes
with 0.05% trypsin and 0.53 mol/l of EDTA solution and replated in
75-cm2 culture flasks. The cells were then gently washed
with warm complete medium after ~20 min to remove nonadherent cells.
When the replated cells had reached ~80% confluency, they were split
1:3 and thereafter passaged one to three more times. After 3 wk,
~2 × 106 cells were obtained. In preliminary
studies, we found a complete lack of smooth muscle myosin heavy chain
expression in these cultures (data not shown), indicating that there
was no SM cell contamination. Cells were frozen with the use of
standard techniques after the first or second passage and then thawed
as needed for injections because it is undesirable to maintain cells in
culture for extended periods of time.
The manufacturer's (Sigma; St. Louis, MO) instructions were followed
for PKH26 fibroblast labeling and resulted in reproducible bright
fluorescent labeling of the cultured cells. PKH26 is a membrane label
introduced by Horan and Slezak (9) that has been shown to
be useful for cell lineage studies because no cell-to-cell transfer
occurs and the label is visible in daughter cells for up to 40 generations. Primary cultured dermal fibroblasts (107
cells) were first released into suspension using a 0.05% trypsin-0.53 mol/l EDTA solution, and the cells were centrifuged at 1,400 rpm (400 g). After the cells were centrifuged, they were resuspended in DMEM only and centrifuged a second time. The supernatant was then
aspirated, leaving no more than 25 ml of solution. The button of cells
was then tapped in the centrifuge tube to resuspend the cells in the
residual medium. The staining solution was immediately prepared by
adding 5 ml of the PKH26 stock solution (1 mol/l concentration) to 1 ml
of diluent C, which was provided with the staining kit. Diluent C (1 ml) was added to the cells in the centrifuge
tube, and the staining solution was then immediately added to the
cells, yielding a final PKH26 concentration of 2.5 mol/l. The cells
were then gently pipetted to thoroughly mix the cells and dye solution for a staining time of 5 min at room temperature. FBS (2 ml) was then
added to the solution to stop the staining reaction, and the cells were
incubated for 1 min. The serum-stopped solution was then diluted with 4 ml of complete medium. The cells were then centrifuged at 400 g for 10 min at room temperature to remove cells from the
staining solution, and this wash step was repeated three more times
with complete medium. The cells were then resuspended in DMEM only and
prepared for injection.
In vivo fibroblast chemotaxis assays.
PKH26-labeled mesenteric fibroblasts (1.5 × 106
cells) were injected intraperitoneally into 10- to 11-wk-old male
Fischer 344 rats. After 7-14 days had passed, animals were
anesthetized with an intramuscular injection of 1% 
-chloralose and
13.3% urethane in saline at 0.6 ml/100 g body wt. The femoral vein was
cannulated for supplemental anesthesia, and the mesentery was prepared
for intravital microscopy. The mesentery was perfused (150 ml/h) with Ringer solution at 37°C, and the stage was mounted on a Nikon K2 SBIO
microscope connected to a Dage model 104722-01 GenIIsys image
intensifier and a Dage model CCD-72 video camera. The camera output was
connected to a Panasonic model AG-1980 videocassette recorder and Sony
PVM-137 monitor.
Mesenteric windows were scanned with the use of a ×20 objective (0.50 numerical aperture, ×729 overall magnification) and selected if they
had a homogeneous distribution of fluorescently labeled fibroblasts. A
micropipette with a tip diameter of 10-12 µm was embedded in the
tissue at a 30° angle using a micromanipulator. The micropipette was
connected to a Sage Instruments (Cambridge, MA) model 341B infusion
pump, and one of the following solutions was infused at a flow rate of
73 µl/h: 1) control solution (Ringer solution containing
0.6 ng/ml BSA and 18.2 ng/ml acetic acid); 2) PDGF-AA
solution [control solution containing 30 ng/ml recombinant human
PDGF-AA homodimer (Genzyme; Cambridge, MA)]; or 3) PDGF-BB solution [control solution containing 30 ng/ml recombinant human PDGF-BB homodimer (Genzyme)]. The concentration of PDGF was
established by an in vitro migration assay using fibroblasts cultured
from fibrotic lungs (25). The micropipette was centered in
the field of view (FOV), the infusion was initiated, and the area was
videotaped using the ×20 objective under fluorescent illumination and
transillumination to provide reference images to locate the exact
position of the pipette tip. Thereafter, images were videotaped every
15 min for a period of 3 h.
Fluorescent and transilluminated digitized images were acquired,
circles with 50- and 100-µm radii were superimposed on the images
centered on the micropipette tip, and cells within these two circles
were counted at each time point. The total number of cells within the
initial FOV was also recorded. Significance was assessed using repeated
measures analysis of variance, and pairwise comparisons were performed
using the Bonferroni correction to Student's t-test, with
P < 0.05.
To ensure that PKH26-labeled fibroblasts had become incorporated into
the mesenteric tissue and were not attached to the upper or lower
mesothelial layers, separate specimens were created using the same cell
culture and injection methods as described above. Fourteen days after
being injected with PKH26-labeled fibroblasts, rats were anesthetized
with an intramuscular injection of ketamine (80 mg/kg body wt) and
xylazine (8 mg/kg body wt) and then euthanized by an overdose of
anesthesia. Mesenteric windows were dissected free, fixed in 4%
paraformaldehyde in PBS for 30 min at 4°C, washed briefly, and
mounted on slides. Specimens were then examined with the use of a
Bio-Rad MicroRadiance confocal scanner attached to a Nikon TE-300
inverted microscope with a ×60 PlanFluor Nikon objective. The top and
bottom mesothelial layers of the mesentery were first located using
transmitted light microscopy. The confocal microscope was then used to
scan through the thickness of the tissue in 1-µm increments. The
depth of each PKH26-labeled fibroblast nucleus as it came into clear
focus was recorded, and a histogram of PKH26 cell depth was generated
from the data.
PDGF-B and PDGFR- staining.
PDGF-B and PDGFR- expression was assayed in young adult rats
(10-11 wk old) because, at this age, mesenteric microvessel density is ~50% of that in adult rats (16-20 wk), indicating
that these networks are still developing (5). Animals were
anesthetized with an intramuscular injection of ketamine (80 mg/kg body
wt) and xylazine (8 mg/kg body wt) and then euthanized by an overdose of anesthesia. The mesentery was exposed, and the mesenteric vein was
cannulated in a retrogade direction. Heparinized PBS (pH 7.4) was then
infused through the catheter to remove blood from the microvessels.
Mesenteric windows were dissected free, dried as whole mounts on a
gelatin-coated slide, and fixed in either 100% MeOH at 
20°C for 30 min (PDGFR- staining) or 4% paraformaldehyde in PBS (pH 7.4) at
4°C for 3 h (PDGF-B staining). After being washed in PBS,
paraformaldehyde-fixed mesenteries were incubated in trypsin-EDTA at
37°C for 5 min. Methanol-fixed mesenteries did not require
proteolytic pretreatment. Tissues were incubated overnight in rabbit
polyclonal antibodies to PDGFR- (sc-432, Santa Cruz Biotechnology)
at a 1:500 concentration and PDGF-B (sc-7878, Santa Cruz) at a 1:100
concentration in 2% BSA and 5% normal goat serum (NGS) in
PBS. After being washed in PBS, secondary biotinylated goat anti-rabbit
IgG antibodies (Jackson Immunoresearch) were applied for 1 h at
room temperature at a concentration of 1:500 in 2% BSA and 5% NGS in
PBS. CY3-conjugated streptavidin (Jackson Immunoresearch) was then
applied at 1:1,000 concentration in 2% BSA for 1 h. Slides were
washed and then incubated in FITC-conjugated mouse monoclonal antibody
to SM -actin (Sigma) at a concentration of 1:100 in 2% BSA in PBS
or mouse monoclonal antibody to the rat endothelial-specific marker
OX-43 (Serotec) at a concentration of 1:100 in 2% BSA in PBS.
Secondary antibodies for OX-43 labeling were CY2-conjugated goat
anti-mouse IgG (Jackson Immunoresearch). OX-43 has been used previously
to verify that Tie2 is an endothelial-specific marker in rats
(26), and in separate experiments, we have shown that
OX-43 colocalizes with Tie2 and flk1 in the endothelium of mesenteric
microvessels (data not shown). Negative control slides for PDGF-B and
PDGFR- were generated by removing the primary antibodies from the
initial incubation and replacing the primary antibodies with the same
concentration of rabbit IgG. Coverslips were applied to the specimens,
and the whole mount networks were observed with a Bio-Rad MicroRadiance
confocal scanner attached to a Nikon TE-300 inverted microscope.
| |
RESULTS |
PDGF-induced fibroblast migration in vivo.
We investigated the in vivo chemotactic potential of PGDF by measuring
the migration of PKH26-labeled fibroblasts to a point source
application of PDGF in mesenteric connective tissue. To ensure that the
initial distribution of PKH26-labeled fibroblasts was similar for each
treatment group, cells were counted within the entire FOV and within
the 50- and 100-µm radius circles of the micropipette at the start of
each experiment. Table 1 indicates that
the average number of labeled cells within each area was similar
between groups.
Figure 1 illustrates the change in number
of cells within 50 µm of the pipette tip with time compared with the
initial number of PKH26-labeled cells (see Table 1). The PDGF-AA
treatment group had only a slight and insignificant increase in cells.
The PDGF-BB group exhibited an increase in cell number after 30 min and
a steady increase in the number of cells at later times. The increase in cell numbers was significantly different than control at 165 and 180 min after the start of infusion.
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 1.
Changes in the number of PKH26-labeled fibroblasts within
50 µm of the pipette tip. Data are means ± SD and indicate that
platelet-derived growth factor (PDGF)-A homodimer (PDGF-AA) and control
solution exhibited no significant change. The PDGF-B homodimer
(PDGF-BB) values were significantly different than control values after
150 min. *Significantly different than control (P < 0.05).
|
|
Figure 2 indicates the change in number
of PKH26-labeled cells within 100 µm of the pipette tip for each of
the treatment groups. The data were similar to that for the 50-µm
distances, but the trend of an increase in cell numbers for the PDGF-BB
group was more dramatic in this case. Cell numbers increased steadily until ~105 min after the start of the infusion, and the number of cells remained relatively constant for later times. The change in
the number of cells was significantly different than that for control
for the PDGF-BB group at times after 75 min, and the control and
PDGF-AA groups exhibited no significant differences with time.
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 2.
Changes in the number of PKH26-labeled fibroblasts within
100 µm of the pipette tip. Data are means ± SD. PDGF-AA and
control solution exhibited no significant change. The PDGF-BB values
were significantly different than control values after 75 min.
*Significantly different than control (P < 0.05).
|
|
Figure 3 is composed of
representative images for each of the treatment and control groups at
the start of the infusion (A, D, and
G), at 90 min (B, E, and
H), and at 180 min (C, F, and I) after the start of the infusion. The transilluminated
images were superimposed on the red fluorescent PKH26 cell images, and circles of radius 50 and 100 µm centered on the tip of the pipette were drawn and used for counting. Figure 3, J and
K, contains higher magnification images of one of the
PDGF-BB treatment mesenteries and clearly shows the positive change in
number of red PKH26-labeled cells within 50 µm of the pipette tip
with time.
View larger version (193K):
[in this window]
[in a new window]
|
Fig. 3.
Images taken during the chemotaxis assays. PKH26 fibroblasts are
seen as punctate red spots throughout each region and are identified
with 5 arrows in K. A-C: Ringer solution
infusion showed no fibroblast migration at the start of the infusion
(time = 0; A), after 90 min (B), and after
180 min (C). D-F: PDGF-AA showed only slight
movement of cells from the start of the infusion (D) to 90 min (E) and 180 min (F). G-I:
PDGF-BB elicited significant directed migration to the tip of the
micropipette over time, from the start of the infusion (G)
to 90 min (H) and 180 min (I). Here, the change
in the number of cells is evident between 90 and 180 min. Yellow
circles, areas 50 and 100 µm from pipette tip. J and
K: accumulation of labeled fibroblasts after PDGF-BB
infusion at higher magnification from start of infusion (J)
to180 min after infusion (K). Circle, area within 50 µm of
the pipette tip. Scale bars, 50 µm.
|
|
Figures 4 and
5 demonstrate that PKH26-labeled
fibroblasts become embedded in the mesenteric tissue after injection
and do not attach to the upper or lower mesothelial surfaces. Figure 4
is a representative confocal montage acquired through the complete thickness of a region of mesentery containing PKH26-labeled
fibroblasts. In the center of the tissue, but not at the upper and
lower surfaces, PKH26 fluorescence is intense, and the PKH26-labeled
cells are clearly focused. Figure 5 is a histogram quantifying the
position of the PKH26-labeled fibroblasts through the depth of the
tissue. A total of 348 PKH26-labeled fibroblasts were examined in three separate specimens. The distance of each cell from the closest surface
was normalized to total tissue thickness, yielding a histogram in which
a value of 0.0 represents the tissue surfaces and a value of 0.5 represents the tissue midplane. PKH26-labeled fibroblasts were evenly
distributed throughout a normalized tissue depth of 0.25-0.5, with
<3% of the PKH26-labeled fibroblasts seen at or near the upper and
lower tissue surfaces (i.e., depth of 0.0-0.1).
View larger version (157K):
[in this window]
[in a new window]
|
Fig. 4.
Confocal montage of PKH26-labeled fibroblasts within a
mesenteric window. The number in the lower right-hand corner of each
image denotes the depth into the tissue (in µm) at which the image
was acquired. Note that PKH26-positive fibroblasts are bright and
focused within the tissue but absent from the upper and lower surfaces
at 0 and 15 µm, respectively. Each image is 168 × 168 µm.
|
|
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 5.
Histogram illustrating the depth of PKH26-labeled
fibroblast nuclei within mesenteric windows. A normalized cell depth of
0.0 represents the upper or lower tissue surface, whereas a value of
0.5 represents the tissue midplane. Fewer than 3% of PKH26-labeled
fibroblasts were closer than 10% of the total tissue thickness to the
upper or lower surface. PKH26-labeled fibroblast nuclei were evenly
distributed at normalized depths ranging between 0.25 and 0.5.
|
|
PDGF-B and PDGF-R expression in
mesenteric networks.
We studied the spatial patterns of PDGF-B and PDGFR- expression in
vivo by examining immunofluorescently labeled whole mount mesenteries
from normally developing animals with the use of confocal microscopy.
Figure 6 illustrates the expression of
PDGF-B in whole mount microvascular networks by immunofluorescent
staining and confocal microscopy. Figure 6A depicts a region
of the mesentery labeled for the rat endothelial cell-specific marker
OX-43. The capillaries, arteriole, and venule in Fig.
6A are all positive for PDGF-B, as shown in
Fig. 6B. While endothelial colocalization of PDGF-B is
somewhat difficult to discern in the arterioles and venules due to the
presence of surrounding SM cells, capillary endothelial cell expression
of PDGF-B is clearly evident. An additional region is labeled for SM
-actin in Fig. 6C, denoting the presence of an arteriole.
In Fig. 6D, PDGF-B expression is seen in capillaries, a
small arteriole, and a small venule, demonstrating that PDGF-B is
expressed by microvessels without SM cells or SM -actin positive pericytes. Figure 6, E and F, depicts an
arteriole at higher magnification expressing SM -actin and PDGF-B,
respectively. SM expression of PDGF-B is clearly evidenced here because
the pattern of PDGF-B labeling in Fig. 6F matches the
corresponding SM -actin expression in Fig. 6E. To ensure
that the CY3 (red) fluorescent signal attributed to PDGF-B staining in
the SM and endothelium was not being caused by bleed-through from SM
-actin/FITC or OX-43/CY2 fluorescence, we examined separate
specimens that were labeled with only FITC or CY2 using the CY3 filter
settings. These specimens exhibited no detectable bleed-through from
the green spectrum into the red spectrum. Figure 6, G and
H, denotes negative staining controls.
View larger version (120K):
[in this window]
[in a new window]
|
Fig. 6.
Confocal images of PDGF-B expression in mesenteric microvascular
networks. A: region of mesenteric network labeled for the
rat endothelial cell specific marker OX-43. A small arteriole (a), two
capillaries (c), and a venule (v) are present. The same region of
tissue is labeled for PDGF-B in B. PDGF-B expression is
present in each microvessel, with colocalization of OX-43 and PDGF-B in
the capillaries denoting endothelial expression of the growth factor.
C: region of mesentery labeled for smooth muscle -actin.
The same region is labeled for PDGF-B in D. Note that PDGF-B
expression extends beyond the arteriole and into microvessels in which
smooth muscle -actin is not detected. PDGF-B expression by
interstitial fibroblasts is also evident in B and
D. E and F: higher magnification
images of an arteriole labeled for smooth muscle -actin
(E) and PDGF-B (F). PDGF-B expression by smooth
muscle cells is indicated by colocalization of smooth muscle -actin
and PDGF-B. G and H: negative controls for PDGF-B
labeling in which primary antibody was either omitted (G) or
replaced with rabbit IgG (H). Confocal settings in
G and H are identical to that used in
D. Scale bars, 50 µm (A and E) and
100 µm (C, G, and H).
|
|
Figure 7 illustrates the expression of
PDGFR- in whole mount mesenteric microvascular networks by confocal
microscopy. Figure 7, A and C, shows regions of
mesenteric network labeled for SM -actin. In Fig. 7B,
PDGFR- expression is visible in a venule, an arteriole, and the
interstitium. At higher magnification (Fig. 7D), it is
evident that PDGFR- is expressed by SM cells in the arterioles and
fibroblasts in the interstitium. PDGFR- expression in each cell type
is indicated by the absence of nuclear staining in Fig. 7D.
While expression of PDGFR- appears to be greater in the microvessels
than in the interstitium in Fig. 7, A and C, it
is important to note that the interstitium contains only a single layer
of fibroblasts, whereas the microvessels contain multiple layers. Thus
it is possible that the enhanced signal in the microvessels is a
summation of fluorescence from multiple cells and the expression of
PDGFR- in the fibroblasts is no less than in the endothelium and SM
cells. In addition to the SM and interstitial fibroblast nuclei shown
in Fig. 7D, an endothelial cell nucleus is also depicted,
suggesting that endothelial cells may express this receptor.
Endothelial cell expression of PDGFR- is then verified in Fig. 7,
E-H, where it is shown to be colocalized with the
endothelial cell specific marker OX-43. Figure 7, I and J, denotes negative staining controls.
View larger version (131K):
[in this window]
[in a new window]
|
Fig. 7.
Confocal images of PDGF receptor- (PDGFR-)
expression in mesenteric microvascular networks. A and
B: region of mesenteric microvascular network labeled for
smooth muscle -actin (A) and PDGFR- (B).
PDGFR- expression is evident in a venule (v), an arteriole (a), and
in the interstitial fibroblasts. An arteriole is shown at higher
magnification in C [smooth muscle (sm) -actin labeled]
and D (PDGFR- labeled). In D, endothelial
cells (ec), smooth muscle, and interstitial fibroblasts (if) exhibit
PDGFR- expression on the cell membranes and in the cytoplasm,
allowing for clear visualization of cell nuclei. Smooth muscle cells
surrounding the arteriole are depicted by the smooth muscle -actin
label in C. To verify endothelial expression of PDGFR-,
colocalization of the rat endothelial-specific marker OX-43
(E and G) with PDGFR- (F and
H) is depicted at high magnification. Arrows in G
and H denote endothelial cell nuclei. I and
J: negative controls for PDGFR- labeling in which primary
antibody was either omitted (I) or replaced with rabbit IgG
(J). Confocal settings in I and J are
identical to that used in B. Scale bars, 20 µm
(G), 50 µm (C and E), and 100 µm
(A, I, and J).
|
|
| |
DISCUSSION |
The purpose of this study was to investigate the potential role of
PDGF in mediating perivascular cell recruitment during microvessel
assembly in vivo by determining whether a concentration gradient of
PDGF elicits the migration of fibroblasts through native mesenteric
tissue and by describing the spatial expression patterns of PDGF-B and
its receptor PDGFR- in mesenteric microvascular networks that
continue to develop into adulthood. The results indicate that, at 30 ng/ml, PDGF-BB but not PDGF-AA is chemotactic for fibroblasts embedded
in native mesenteric tissue. Additionally, we found that PDGF-B is
expressed by endothelial cells and SM cells in microvessels and
PDGFR- is expressed by SM cells, endothelial cells, and interstitial
fibroblasts. Because PDGF-BB elicits migration of interstitial
fibroblasts in the mesentery and PDGF-B and PDGFR- are spatially
expressed in a pattern that is indicative of paracrine signaling from
the microvessels to the fibroblasts in the extravascular space, these
results are consistent with a role for PDGF-B and its receptor in
mediating the recruitment of perivascular cells to microvessels during
normal development from postnatal stages to adulthood.
PDGF and fibroblast migration.
Cell migration is a coordinated cycle of several processes, including
lamellipodial extension, formation of lamellipod-substratum attachments, cytoskeletal contraction, and release of
cell-substratum attachments. In the current study, fibroblast
migration was observed on a native matrix as it exists in vivo. This is
important because recent studies suggest that a complex interplay
exists between growth factors, their receptors, extracellular matrix
proteins, and adhesion molecule expression and clustering. In
particular, synergies exist between PDGF, PDGF receptors, and the
adhesion molecules that modulate cellular migration (4, 15, 16, 27). Moreover, the ability of cells to migrate in response to PDGF is dependent on integrin-mediated cell-substratum attachments (2, 10, 13, 20, 27). Given this information and the uncertainty in attempting to characterize the extracellular matrix composition of a given tissue, it is difficult to extrapolate the
results of in vitro cell migration studies to the in vivo setting.
Here, although the extracellular matrix components are unknown, we are
essentially recapitulating the in vivo process. Thus we conclude that a
sufficient concentration gradient of PDGF-BB in mesenteric tissue will
elicit fibroblast migration.
Our data and observations from the chemotaxis assays illustrate
two other crucial points. First, the migration of fibroblasts occurred
only in subpopulations of labeled fibroblasts. We counted an increase
of ~8 cells within 100 µm of the pipette tip when there were ~40
PKH26-labeled fibroblasts on average within the FOV. Thus some cells
did not exhibit directed motion, suggesting a heterogeneity of PDGF-BB
responsiveness. This heterogeneity may be due to the fact that the
primary explant technique selects cells that migrate out of the tissue
and are predisposed to enhanced motility. Furthermore, it is inevitable
that the tissues were stretched to varying degrees. Because any slight
stretching of the mesentery affects permeability and extracellular
matrix organization, this may be a source of heterogeneity. Second, the
data from the fibroblast migration experiments indicate that PDGF-BB
but not PDGF-AA elicits the fibroblast migration on a native mesenteric matrix. PDGF-BB has been well established as a chemoattractant for
fibroblasts in vitro, but the chemoattractant potential of PDGF-AA
remains controversial. It has been found that PDGF-AA and PDGF-BB
elicit similar chemotaxis responses in lung fibroblasts (17); however, corneal fibroblasts are significantly more
responsive to PDGF-BB than PDGF-AA (11, 12). Our data
clearly agree with the latter investigation. While we believe our data
are conclusive evidence that PDGF-BB is chemotactic for mesenteric
fibroblasts in vivo, we cannot extend this conclusion to all fibroblast
subpopulations and tissues because differences clearly exist.
Furthermore, our study is limited to a single source concentration of
growth factor. It is possible that, at different source concentrations,
PDGF-AA elicits fibroblast migration on this matrix.
PDGF signaling and microvessel assembly.
In the context of vascular development, PDGF-B and PDGFR- expression
have been well studied in the embryo. However, relatively little is
known of the expression patterns of these molecules during microvessel
development during maturation from postnatal stages to adulthood.
Furthermore, while it has been shown that exogenous application of
PDGF-B elicits angiogenesis in vivo (19) and disrupts
pericyte-endothelial interactions (1), the role of
endogenously produced PDGF-B in these processes in the adult is poorly
defined. To better understand the potential role of PDGF signaling in
microvascular development, we examined PDGF-B and PDGFR- protein
expression in growing mesenteric microvascular networks of young adult
rats (10-11 wk). At this age, an approximate doubling in total
mesenteric microvessel density remains before full adulthood
(16-18 wk) (5); thus these networks were studied in
an actively growing stage.
PDGF-B protein expression patterns in these networks are generally
consistent with previous observations made in the embryonic vasculature. Here, PDGF-B protein was localized to all microvessels, with endothelial cells, SM cells, and (to a lesser extent) interstitial fibroblasts showing immunopositivity for PDGF-B. PDGF-B expression in
SM cells appeared to be greater than in endothelial cells, with
arteriolar SM exhibiting intense expression. It is, however, important
to note that this enhancement may have been due to the greater number
of cell layers in vessels containing SM. During embryonic vascular
development, PDGF-B is expressed by capillary and arteriolar
endothelial cells, but venous endothelial cells lack PDGF-B (6,
14). Postnatal PDGF-B expression is limited to short capillary
sprouts in many tissues (6). Our observations also
indicate that PDGF-B is expressed by capillary sprouts (data not
shown). However, in contrast to our observations, PDGF-B is not located
in mature capillaries in late embryogenesis or early postnatal
development, and nonendothelial PDGF-B expression has not been detected
(6).
PDGFR- protein was localized in all microvessels in mesenteric
networks, with endothelial and SM cells both exhibiting strong immunopositive responses. We also observed intense staining for PDGFR- in the fibroblasts in the intertsitium and immediately surrounding microvessels. Similarly, in the small vessels of the embryonic vasculature, PDGFR- is expressed in pericytes and
mesenchyme immediately surrounding the endothelium (14,
22). SM expression of PDGFR- is also evident during these
developmental stages (6, 14). Endothelial PDGFR-
expression in small embryonic vessels is, however, controversial,
because claims that endothelial cells express PDGFR- (8,
22) have been disputed (14). PDGFR- is required
for endothelial cell participation in wound healing angiogenesis
(4), and our observations support the hypothesis that
microvascular endothelial cells express PDGFR-.
The spatial patterns of PDGF-B and PDGFR- protein expression provide
insight into how PDGF signaling may occur. First, because SM and
endothelial cells express both PDGF-B and PDGFR-, autocrine signaling is possible. Autocrine signaling by PDGF-B has been proposed
as a means by which endothelial cell populations are expanded during
early stages of placental angiogenesis (8). To our
knowledge, PDGF-B expression by SM cells during embryonic or adult
microvascular development has not been observed, and autocrine SM PDGF
signaling has not been proposed as a mechanism of microvessel growth.
While autocrine SM PDGF-B signaling may be proposed here as a means of
enlarging developing microvessels via SM hyperplasia, lineage studies
are needed to explore this hypothesis.
Second, endothelial PDGF-B and SM and interstitial fibroblast PDGFR-
expression support a paracrine signaling hypothesis. In the embryo,
paracrine PDGF signaling from the endothelium to the pericytes and SM
appears to play a role in angiogenic sprouting and vessel enlargement
(6). However, because PDGFR--positive cells are still
found adjacent to developing vessels in PDGF-B-deficient mice, a
PDGF-B-independent induction of PDGFR- cells from the mesenchyme has
been proposed (6). In contrast, we observed PDGFR-
expression well beyond the mesenteric microvessels and throughout the
tissue, indicating that the fibroblasts and SM cells immediately
surrounding these vessels are not unique in their potential ability to
respond to PDGF-B. When coupled with the fact that a concentration
gradient of PDGF-BB will elicit fibroblast migration in this tissue,
the PDGFR- expression pattern raises the possibility that long range
(i.e., greater than one cell diameter) paracrine signaling with PDGF-B
can occur in these networks. Moreover, it has been shown that, during
wound healing angiogenesis, PDGFR- is required for fibroblast
participation (4). This indicates that, in contrast to
embryonic development, induction of PDGFR--positive mesenchymal
cells likely requires PDGF-B in the adult. Ultimately, the answers to
these questions and a clear definition of the role of PDGF signaling in
microvascular development during normal maturation will require
inducible cell-specific gene expression or gene-targeting models for
PDGF and its receptors in the future. At present, our results indicate
that PDGF-BB is able to recruit fibroblasts in mesenteric connective
tissue and that, during normal maturation, PDGF-B and PDGFR- are
expressed in a pattern that is consistent with a role for PDGF in
mediating the microvascular development process during normal maturation.
| |
ACKNOWLEDGEMENTS |
This work was supported by American Heart Association Grant
9730025N (to R. J. Price) and by National Heart, Lung, and Blood Institute Grant HL-52309 (to T. C. Skalak).
| |
FOOTNOTES |
Address for reprint requests and other correspondence: R. J. Price, Dept. of Biomedical Engineering, Univ. of Virginia, Box 800759, Health Sciences Center, Charlottesville, VA 22908-0759 (E-mail:
rprice{at}virginia.edu).
The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement"
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Received 18 September 2000; accepted in final form 2 January 2001.
| |
REFERENCES |
1.
Benjamin, LE,
Hemo I,
and
Keshet E.
A plasticity window for vessel remodeling is defined by pericyte coverage of the preformed endothelial network and is regulated by PDGF-B and VEGF.
Development
252:
1591-1598,
1998.
2.
Bilato, C,
Curto KA,
Monticone RE,
Pauly RR,
White AJ,
and
Crowe MT.
The inhibition of vascular smooth muscle cell migration by peptide and antibody antagonists of the v3 integrin complex is reversed by activated calcium/calmodulin-dependent protein kinase II.
J Clin Invest
100:
693-704,
1997.
3.
Crosby, JR,
Seifert RA,
Soriano P,
and
Bowen-Pope DF.
Chimaeric analysis reveals role of PDGF receptors in all muscle lineages.
Nat Genet
18:
385-388,
1998.
4.
Crosby, JR,
Tappan KA,
Seifert RA,
and
Bowen-Pope DF.
Chimera analysis reveals that fibroblasts and endothelial cells require platelet-derived growth factor receptor expression for particpation in reactive connective tissue formation in adults but not during development.
Am J Pathol
154:
1315-1321,
1999.
5.
Hansen-Smith, FM,
Joswiak GR,
and
Bauster JL.
Regional differences in spontaneously occurring angiogenesis in the adult rat mesentery.
Microvasc Res
47:
369-376,
1994.
6.
Hellström, M,
Kalen M,
Lindahl P,
Abramsson A,
and
Betsholtz C.
Role of PDGF-B and PDGFR- in recruitment of vascular smooth muscle cells and pericytes during embryonic blood vessel formation in the mouse.
Development
126:
3047-3055,
1998.
7.
Hirschi, KK,
Rohovsky SA,
and
D'Amore PA.
PDGF, TGF, and heterotypic cell-cell interactions mediate endothelial cell-induced recruitment of 10T1/2 cells and their differentiation to a smooth muscle fate.
J Cell Biol
141:
805-815,
1998.
8.
Holmgren, L,
Glaser A,
Pfeifer-Ohlsson S,
and
Ohlsson R.
Angiogenesis during human extraembryonic development involves the spatiotemporal control of PDGF ligand and receptor gene expression.
Development
113:
749-754,
1991.
9.
Horan, PK,
and
Slezak SE.
Stable cell membrane labeling.
Nature
340:
167-168,
1989.
10.
Itoh, H,
Nelson PR,
Mureebe L,
Horowitz A,
and
Kent KC.
The role of integrins in spahenous vein vascular smooth muscle cell migration.
J Vasc Surg
25:
1061-1066,
1997.
11.
Kamiyama, K,
Iguchi I,
Wang X,
and
Imanishi J.
Effects of PDGF on the migration of rabbit corneal fibroblasts and epithelial cells.
Cornea
17:
315-325,
1998.
12.
Kim, WJ,
Mohan RR,
and
Wilson SE.
Effect of PDGF, IL-1alpha, and BMP2/4 on corneal fibroblast chemotaxis: expression of the platelet-derived growth factor system in the cornea.
Invest Ophthalmol Vis Sci
40:
1364-1372,
1999.
13.
Klominek, J,
Baskin B,
and
Hauzenburger D.
Platelet-derived growth factor (PDGF) BB acts as a chemoattractant for human malignant mesothelioma cells via PDGF receptor -integrin 31 interaction.
Clin Exp Metastasis
16:
529-539,
1998.
14.
Lindahl, P,
Johansson BR,
Leveen P,
and
Betsholtz C.
Pericyte loss and microaneurysm formation in PDGF-B deficient mice.
Science
277:
242-245,
1997.
15.
Miyamoto, S,
Teramoto H,
Gutkind JS,
and
Yamada KM.
Integrins can collaborate with growth factors for phosphorylation of receptor tyrosine kinases and MAP kinase activation: roles of integrin aggregation and occupancy of receptors.
J Cell Biol
135:
1633-1642,
1996.
16.
Nelson, PR,
Yamamura S,
and
Kent KC.
Platelet-derived growth factor and extracellular matrix proteins provide a synergistic stimulus for human vascular smooth muscle cell migration.
J Vasc Surg
26:
104-112,
1997.
17.
Osornio-Vargas, AR,
Lindroos PM,
Coin PG,
Badgett A,
Hernandez-Rodriguez NA,
and
Bonner JC.
Maximal PDGF-induced lung fibroblast chemotaxis requires PDGF receptor-.
Am J Physiol Lung Cell Mol Physiol
271:
L93-L99,
1996.
18.
Price, RJ,
Owens GK,
and
Skalak TC.
Immunohistochemical identification of arteriolar development using markers of smooth muscle differentiation: evidence that capillary arterialization proceeds from terminal arterioles.
Circ Res
75:
520-527,
1994.
19.
Risau, W,
Drexler H,
Mironov V,
Smits A,
Siegbahn A,
Funa K,
and
Heldin CH.
Platelet-derived growth factor is angiogenic in vivo.
Growth Factors
7:
261-266,
1992.
20.
Sakai, T,
de la Pena JM,
and
Mosher DF.
Synergism among lysophosphatidic acid, 1 integrins, and epidermal growth factor or platelet-derived growth factor in mediation of cell migration.
J Cell Biol
274:
15480-15486,
1999.
21.
Senior, RM,
Griffin GL,
Huang JS,
Walz DA,
and
Deuel TF.
Chemotactic activity of platelet alpha granule proteins for fibroblasts.
J Cell Biol
96:
382-385,
1983.
22.
Shinbrot, E,
Peters KG,
and
Williams LT.
Expression of the platelet-derived growth factor during organogenesis and tissue differentiation in the mouse embryo.
Dev Dyn
199:
169-175,
1994.
23.
Skalak, TC,
Price RJ,
and
Zeller PJ.
Where do new arterioles come from? Mechanical forces and microvessel adaptation.
Microcirculation
5:
91-94,
1998.
24.
Soriano, P.
Abnormal kidney development and hematological disorders in PDGF- receptor mutant mice.
Genes Dev
8:
1888-1896,
1994.
25.
Suganuma, H,
Sato A,
Tamura R,
and
Chida K.
Enhanced migration of fibroblasts derived from lungs with fibrotic lesions.
Thorax
50:
984-989,
1995.
26.
Wong, AL,
Haroon ZA,
Werner S,
Dewhirst MW,
Greenberg CS,
and
Peters K.
Tie2 expression and phosphorylation in angiogenic and quiescent adult tissues.
Circ Res
81:
567-574,
1997.
27.
Woodard, AS,
Garcia-Cardena G,
Leong M,
Madri JA,
Sessa WC,
and
Languino LR.
The synergistic activity of v3 integrin and PDGF receptor increases cell migration.
J Cell Sci
111:
469-478,
1998.
28.
Zerwes, HG,
and
Risau W.
Polarized secretion of a platelet-derived growth factor chemotactic factor by endothelial cells in vitro.
J Cell Biol
105:
2037-2041,
1987.
Am J Physiol Heart Circ Physiol 280(5):H2116-H2125
0363-6135/01 $5.00
Copyright © 2001 the American Physiological Society
TOP
ABSTRACT
INTRODUCTION
