Involvement of the Fibroblast Growth Factor System in Adaptive and Chemokine-Induced Arteriogenesis
Fibroblast growth factors (FGFs) have been applied in a variety of therapeutic and experimental studies to improve collateral blood flow. However, the pathophysiological role and the temporospatial expression of the FGFs and their receptors during arteriogenesis have never been elucidated in vivo. Here, we report that collateral artery growth in its early phase is associated with an increased expression of FGF receptor-1 (FGFR-1) and syndecan-4 on mRNA and protein levels as well as with an increased kinase activity of FGFR-1 in a rabbit model of arteriogenesis. However, the mRNA levels of FGF-1 and -2 remained constant. Our data suggest that these growth factors are supplied by endothelial attracted monocytes that, in turn, produce and deliver the FGFs to growing collateral arteries. Monocyte chemoattractant protein-1-stimulated arteriogenesis was strongly reduced in rabbits by application of the FGF inhibitor polyanetholesulfonic acid, indicating that the monocyte-related arteriogenesis (as well as the unstimulated adaptation proper) is promoted by FGFs. In summary, this study shows that arteriogenesis is associated with an increased expression of the FGFRs at the site of the vessel, whereas the growth-promoting ligands are supplied by monocytes in a paracrine way.
Collateral artery growth (arteriogenesis) occurs spontaneously as an adaptive response to vascular occlusion. However, the speed of growth can be stimulated by factors such as monocyte chemoattractant protein-1 (MCP-1),1 transforming growth factor-β1,2 or granulocyte-macrophage colony-stimulating factor,3,4 which either attract or prolong the lifespan of monocytes. In a variety of studies, fibroblast growth factors (FGFs) have also been applied for their potency to induce vascular growth and to promote the recovery of blood flow after arterial occlusion.
The FGFs are a family of >20 polypeptides that mediate a broad range of biological activities (hematopoiesis, development, and wound repair) in a variety of cell types of mesenchymal and neuroectodermal origin. These activities are initiated through 4 structurally related membrane-associated tyrosine kinase FGF receptors (FGFRs) that are derived from separate genes and exist in a multitude of isoforms.5 Ligand-induced dimerization of these high-affinity receptors, which is mediated by low-affinity receptors, is a key event in transmembrane signaling. It leads to an increase in kinase activity, resulting in the autophosphorylation of the receptor and the induction of diverse biological responses.6–8
Special attention is focused on FGF-1 and FGF-2, potent mitogens playing an important role in cell proliferation, migration, and differentiation.9–11 These FGFs differ from most of the other FGFs because they lack hydrophobic signal peptides and therefore are concentrated within their cell of origin.12 Exogenously administrated, these growth factors have been shown to increase the number of visible collateral arteries and improve collateral blood flow in therapeutic as well as experimental studies.13–16 Despite these positive results, almost nothing is known about the natural temporospatial expression of these genes and their receptors during collateral artery growth, which is the topic of our present study involving a rabbit model of adaptive arteriogenesis.1
Materials and Methods
Femoral artery-ligated New Zealand White rabbits (n=5) were studied as previously described.1 The animals were euthanized at distinct intervals, and tissue samples were collected from the musculus quadriceps vastus intermedius (m. quadriceps, a muscle constantly harboring two arteriolar connections) as well as from control organs and control animals, snap-frozen in liquid nitrogen, and stored at −80°C until further investigation. These procedures were performed in accordance with local guidelines and recommendations.
Postmortem Angiograms and Isolation of Collaterals
Postmortem angiograms were performed as previously described.1 After injection of the contrast medium, collateral arteries from experimental, sham-operated, or control animals were excised, snap-frozen, and stored at −80°C until further investigations. For one RNA preparation of collateral arteries, samples from 4 animals were pooled.
Twenty-four rabbits (10 male and 14 female, body weight 3.7±0.5 kg, aged 3 to 4 months) were randomly assigned to 1 of 4 groups (n=6). After ligation of the femoral artery, 3 groups received PBS, polyanetholesulfonic acid (PAS, 15 mg/kg per day, ICN), or MCP-1 (0.2 μg/kg per day, Peprotech) via an osmotic minipump for 1 week. A fourth group received a local concomitant application of MCP-1 and PAS. Surgical procedures were essentially as described previously.1 During the infusion of PAS, we did not observe any signs of toxicity.
Hemodynamic measurements, the counting of microspheres, and the calculation of conductance were essentially performed as described previously.17 Statistical analysis on number of vessels as well as on collateral conductance was performed by ANOVA, and subsequent multiple comparisons were performed by Bonferroni test (for all experiments shown, n=6). Results are expressed as mean±SEM and are considered statistically significant at P<0.05.
RNA Isolation and Northern Blot
Total RNA was isolated according to the method of Chomczynski and Sacchi.18 Northern hybridization was carried out according to standard procedures.19 For normalization, the data of each hybridization signal were divided by the values of the matching 18S rRNA signal. The average of the data obtained from the control animals was standardized as 100%.
Reverse transcription-polymerase chain reaction (RT-PCR) on total RNA from rabbit heart was performed according to standard procedures.19 The following sense and antisense primers were used: for FGFR-1: a 36-mer oligonucleotide homologous to mouse M33760 position 813 to 848, containing a restriction site for BamHI (CTATCGGAtcCTCCCATCACTCTGCATGGTTGACG), and a 31-mer complementary to mouse M33760 position 1077 to 1047 with a restriction site for HincII (CTGGAGTCAaCTGACACTGTTACCTGTCTGC); for FGF-1: a 30-mer, homologous to rat X14232 position 343 to 372 with a restriction site for PstI (CCAAACTGCaGTACTGCAGCAACGGGGGCC), and a 31-mer, complementary to rat X14232 position 456 to 426, containing a restriction site for SacI (CCGCACTGAGCTcCAGCTGAATGTGCTGGTC). The artificial restriction sites (underlined) were introduced into the oligonucleotides to facilitate the cloning of the PCR products into a Bluescript vector.
For Northern blot analysis, cDNA probes were random prime-labeled using a Rediprime labeling system (Amersham Pharmacia Biotech) according to manufacturer’s procedure. cDNA probes used in the present study were as follows: FGFR-1 (260 bp) and FGF-1 (120 bp) (both RT-PCR products were from rabbit heart; see above); FGF-2 (rabbit, 471 bp; provided by J. Winkles, American Red Cross); syndecan-1 (rat, 800 bp), syndecan-2 (rat, 800 bp), and syndecan-4 (rat, 670 bp) (all provided by G. Cismeci-Smith, University Hospital, Danville, Pa); and 18S (mouse, 770 bp; provided by I. Oberbäumer, Munich, Germany).
FGFR-1 immunohistochemistry was performed by use of the peroxidase method. In brief, 5- or 10-μm-thick cryosections of the m. quadriceps were fixed with cold acetone, dried, and washed with PBS. After incubation for 20 minutes in a blocking solution (0.4% glycine, 0.1% BSA, and 1% normal sheep serum), sections were incubated overnight with an FGFR-1 monoclonal antibody (1:50, Upstate Biotechnology), followed by several PBS washing steps and incubation with peroxidase-conjugated anti-mouse IgG (Sigma) for 1 hour. After the sections were rinsed in PBS, peroxidase activity was assessed by incubation with 0.05% 3′-3 diaminobenzidine (Sigma) plus 0.03% H2O2 in PBS. Tissue sections were counterstained with hematoxylin, mounted in Mowiol (Aventis), and viewed microscopically. Omission of the first antibody was used as a negative control.
Immunofluorescence with the proliferation marker KI-67 (Dianova) was performed as previously described.20
Protein preparation and Western blot were performed according to standard procedures.18 For FGFR-1 and syndecan-4 analyses, equal amounts of the membranous fraction of the total protein preparation were separated on NuPAGE 4% to 12% Bis-Tris PreCast Gel (NOVEX). Immunoreactive bands were visualized by use of an FGFR-1-specific monoclonal antibody (Upstate Biotechnology), a monoclonal anti-phospho-Tyr-specific antibody (Sigma) or a syndecan-4-specific polyclonal antibody (Santa Cruz Biotechnology), and an enhanced chemiluminescence detection system (ECL, Amersham Pharmacia Biotech).
For FGF-2 Western blot, monocytes were isolated from buffy coats of healthy blood donors by density gradient centrifugation and elutriation.21 Cultured cells (2.5×107 per sample) were stimulated for 12 hours with bacterial lipopolysaccharide (LPS, 10 μg/mL). Unstimulated monocytes served as a control. Afterward, supernatants were harvested, and cells were lysed in Laemmli buffer. Equal amounts of protein were separated on a 10% SDS-PAGE and analyzed by Western blot as described above with the use of a polyclonal antibody against FGF-2 (Peprotech).
Recombinant human FGF-2 (Preprotech) and FGFR-1 (Upstate Biotechnology) served as a positive control and a negative control (omission of the first antibody), respectively.
In Vitro Kinase Assay
Equal amounts (280 μg) of the membranous fraction of total protein preparations on collateral arteries isolated 3, 6, 12, and 24 hours after femoral artery ligation or sham operation as well as after treatment with PAS were subjected to immunoprecipitation with the use of a monoclonal FGFR-1 antibody (see above) according to standard procedures. The samples obtained after immunoprecipitation were used for in vitro kinase assay. Precipitates were incubated at 30°C for 15 minutes in a phosphorylation solution (30 mmol/L Tris-HCl, pH 7.4, containing 10 mmol/L MgCl2, 50 μmol/L ATP, and 5 μCi of [γ-32P]ATP). After the addition of a stop solution (Laemmli sample buffer with 2-mercaptoethanol), samples were boiled and separated on 10% SDS-PAGE gels. The results were visualized and quantified with a PhosphorImager (Molecular Dynamics).
Monocytes in whole blood were treated with FGF-2 (200 ng/mL, Peprotech). A monoclonal antibody against rabbit αM or human αM, αL, or β2 integrin (1 μg/mL, all from Dianova) was added, followed by a FITC-conjugated secondary antibody. Finally, a monoclonal antibody against the human monocyte marker CD14 (clone My4, Beckman-Coulter) showing cross reaction with rabbit monocytes was incubated, and red blood cells were lysed with the use of ammonium chloride lysing buffer. Flow cytometry measurement was performed on a FACS-Calibur using CellQuest Software (BD Biosciences).
Monocytes were isolated from buffy coats of healthy blood donors as described above. Human umbilical endothelial cells (HUVECs) were prepared according to the method of Jaffe et al22 and grown to confluence on microtiter plates. An adhesion assay was performed as described previously.21 After stimulation with FGF-1, FGF-2, vascular endothelial growth factor (VEGF), or MCP-1 (each 200 ng per milliliter monocytes, all from Peprotech), the monocytes (105/mL, 100 μL per well) were added to HUVECs. After an incubation period of 90 minutes, adherent monocytes were counted by light microscopy. If the effect of PAS was to be tested, each factor was preincubated with PAS (10 μg/mL, 30 minutes, 37°C) before monocyte incubation.
Signals were quantified with a PhosphorImager (Molecular Dynamics) with the use of ImageQuant software.
FGFR-1 Is Strongly Expressed in SMCs
Immunohistochemical analysis on the m. quadriceps of sham-operated and experimental animals after 1 hour, 3 hours, 6 hours, 12 hours, 1 day, and 3 days of femoral artery ligation showed that FGFR-1 is localized in fibroblasts, some pericytes, and smooth muscle cells (SMCs) of arteries and arterioles (Figures 1A through 1D; data are shown for 12 hours and 1 and 3 days of femoral artery ligation). The endothelial cells (ECs) of capillaries, arteries, and veins, as well as skeletal muscles, showed no staining at any point of time studied. The growth of collateral arteries isolated 3 days after femoral occlusion was verified by KI-67 staining of cryostat sections (data not shown). Mirror sections were immunostained with the FGFR-1 monoclonal antibody (Figure 1D). Strong immunoreactivity was detected in SMCs of growing collateral arteries. The localization of the FGFR-1 was unchanged during the periods of collateral growth studied.
Increased Expression of FGFR-1 mRNA During the Early Phase of Arteriogenesis
Northern blot results revealed a constant level of FGFR-1 mRNA (4.6 kb) in the m. quadriceps of control and sham-operated animals. In tissue samples from experimental animals, we found a significant upregulation of the FGFR-1 mRNA after 3 hours of femoral artery ligation (2.5-fold times control). Peak levels at 6 hours (≈4-fold control) were followed by a continuous fall, reaching values of control at days 3 and 7 (Figures 2A and 2B).
Analyzing the mRNA level of the FGFR-1 in collateral arteries 3 days after femoral occlusion, we found a comparable abundance of FGFR-1 mRNA in experimental and sham-operated animals. However, the total level of the transcript was much higher in the collateral arteries than in the m. quadriceps (Figure 2B). For comparison, we analyzed the mRNA levels of FGFR-1 in different organs. The transcript was expressed at very high levels in the heart, in the uterus of a pregnant rabbit, and in the central nervous system. However, the bone marrow showed levels just above the detection limit (Figure 2B).
Western Blot Analysis
Western blot analysis revealed an upregulation of the FGFR-1 protein 12 hours after femoral artery ligation in the m. quadriceps. At day 3, the protein level of the FGFR-1 was about the same in growing collaterals and in arteriolar connections of sham-operated animals; however, the level of FGFR-1 was much higher in the collaterals than in the m. quadriceps (Figure 3).
Increased Kinase Activity of FGFR-1
The precipitation of FGFR-1 from membranous fractions of proteins prepared from isolated collateral arteries 3, 6, 12, and 24 hours after femoral artery ligation or sham operation was performed using an anti-FGFR-1 antibody. In vitro kinase assays performed with these specific immunoprecipitates evidenced increased levels of FGFR-1 phosphorylation 6 hours after the induction of arteriogenesis (Figure 4). The investigation of changes in the phosphorylation state of FGFR-1 in immunoprecipitates using a specific anti-phospho-Tyr antibody also evidenced increased levels of FGFR-1 phosphorylation at tyrosine residues 6 hours after the induction of arteriogenesis compared with sham operation. However, an infusion of PAS in rabbits completely abolished this phosphorylation of the FGFR-1 (data not shown).
Upregulation of Syndecan-4 During Collateral Artery Growth
Investigating the expression of syndecan-1, -2, and -4 in the m. quadriceps for the first 24 hours and for days 3 and 7 of collateral artery growth, we were not able to demonstrate quantitative amounts of syndecan-1 and -2 in the m. quadriceps via Northern blot analysis (data not shown). For syndecan-4, our results displayed a significant increase of the mRNA in experimental compared with control or sham-operated animals after 3 hours of femoral artery occlusion, followed by an upregulation on the protein level after 6 and 12 hours (Figures 5A and 5B).
Vessel Growth Is Not Associated With Upregulation of FGF-1 and FGF-2 mRNA
The transcript levels of FGF-1 and FGF-2 were analyzed in the m. quadriceps for the first 24 hours and for days 3 and 7 of femoral artery ligation. Northern blot results displayed no significant change of the mRNA level either of FGF-1 or of FGF-2 (Figure 6; data are shown for the first 24 hours). In the heart that was analyzed as a positive control, FGF-1 was expressed at a high level, and FGF-2 was expressed at a low level (Figure 6).
FGF-Stimulated Monocyte Adhesion Is Blocked by PAS
Stimulation of rabbit and human monocytes with FGF-2 resulted in a concentration- and time-dependent significant increase in αM integrin (rabbit) and αL, αM, and β2 integrin (human) expression, as shown by flow cytometry (Figure 7A). Western blot analysis showed that LPS-stimulated monocytes, in contrast to unstimulated monocytes, produce FGF-2 (two isoforms, ≈18 and 19 kDa, forming dimers under nonreducing conditions). However, the supernatants were negative for FGF-2 (Figure 7B). Adhesion assays showed that stimulation of monocytes with FGF-1 and FGF-2, as well as with MCP-1 and VEGF (the latter were analyzed for reasons of control), significantly increased monocyte adhesion to EC monolayers: FGF-1, 50-fold; FGF-2, 4-fold; MCP-1, 14-fold; and VEGF, 9-fold (Figure 7C). Coincubation of FGF-1 and of FGF-2 with PAS significantly reduced growth factor-induced monocyte adhesion but (in both cases) not to the values found with untreated monocytes: in the case of FGF-1, to a 36-fold adhesion; in the case of FGF-2, to a 2-fold adhesion (Figure 7C). However, monocyte adhesion induced by MCP-1 and VEGF was not significantly affected by application of PAS (Figure 7C). Clotting assays revealed that PAS also did not change the activity of VEGF (data not shown).
PAS Treatment Interferes With Collateral Artery Growth
Rabbits were treated for 1 week either with PAS or concomitantly with PAS and MCP-1 or, for reasons of control, with PBS or MCP-1 alone.
Figure 8A shows an angiogram of a rabbit hindlimb after 1 week of occlusion and infusion of PBS. Several collateral arteries spanning the occlusion site show the typical corkscrew morphology, indicating the proliferation and growth of these vessels. The continuous infusion of PAS for 1 week diminished the size of visible vessels displaying the typical signs of growing collateral arteries but not their number (PBS, 16.16±1.40; PAS, 16.83±1.60) (Figure 8B). The infusion of MCP-1 for 1 week significantly increased the number (31.00±3.36) and diameter of visible growing collateral arteries, as shown in Figure 8C. Concomitant application of MCP-1 and PAS in the same setting resulted in a strong reduction of the number (25.66±1.66) and size of visible growing collateral arteries (Figure 8D). Analysis of visible collateral arteries under stereoscopic viewing verified the radiographic findings.
After 1 week of femoral artery occlusion and treatment with PBS, a collateral conductance of 4.10±0.48 mL/min per 100 mm Hg was reached. Intra-arterial MCP-1 significantly increased collateral conductance (33.96±1.76 mL/min per 100 mm Hg) compared with conductance in the PBS-treated group (P<0.005). Compared with the treatment with MCP-1 alone, the concomitant application of PAS and MCP-1 showed a significant reduction of collateral conductance (PAS, 18.29±2.13 mL/min per 100 mm Hg; P<0.01).
In the present study, we showed that arteriogenesis is associated with an increased expression of FGFR-1 and syndecan-4 on mRNA and protein levels as well as with an increased tyrosine kinase activity of FGFR-1. Furthermore, we provide evidence that growth-promoting FGFs are not increased expressed in growing collaterals but are likely to be provided in a paracrine way by monocytes.
Several reports have been published showing the vessel growth-accelerating properties of exogenously supplied FGF-1 and FGF-2 as well as of FGF gene transfer.14–16,23 However, the unstimulated temporal and spatial expression of these genes and their high- and low-affinity receptors during adaptive arteriogenesis have never been elucidated in vivo. To address this question, we performed studies in our rabbit model of arteriogenesis.1 In this model, femoral artery ligation results in redirection of the blood flow to preexisting arteriolar connections in the upper leg (m. quadriceps, musculus adductor), thereby increasing mechanical forces such as shear stress and circumferential wall stress. The endothelium becomes activated, and invading monocytes locally supply growth factors and cytokines that promote collateral artery growth.24 Cell proliferation in these vessels could be demonstrated as early as 24 hours after femoral ligation.25
The expression of the FGFRs and their isoforms (which have unique ligand-binding properties) is temporospatially restricted throughout the body during development and disease, delineating that the mode of action of the FGFs is not only regulated by the expression of the ligands but also by the availability of their receptors. Our results displayed a rapid and pronounced induction of FGFR-1 on RNA and protein levels as well as an increased tyrosine kinase activity of FGFR-1 in the early phase of arteriogenesis. Immunohistochemical studies showed a specific localization of FGFR-1 in SMCs of the arteries in the m. quadriceps. Fibroblasts and pericytes were also stained, but the ECs of capillaries, arteries and veins, or skeletal muscle were not. The following strongly suggest an important role of FGFR-1 for signal transduction in the arterial wall during the early phase of collateral artery growth: (1) the time relationship between FGFR-1 expression and arteriogenesis, (2) the strong expression of FGFR-1 in collateral arteries, (3) the specific immunoreactivity in vascular SMCs, and (4) the fact that FGFR-1 mRNA (which is, like FGF-1 and FGF-2, already expressed in vascular SMCs under normal conditions)26,27 is the major form of the 4 known FGFRs expressed by proliferating (human) arterial SMCs.28
Ligand-induced dimerization of the high-affinity receptor (FGFR) mediated by low-affinity receptors (heparan sulfate proteoglycans) is a key event in transmembrane signaling.6,7 On simultaneously binding of heparan sulfate to several FGFs and release of this complex by proteases, FGFR dimerization and activation are stimulated by multivalent binding of the multimeric FGF complex. Attention is focused on the syndecans, a group of transmembrane heparan sulfate proteoglycans.29 Analyzing the expression of distinct syndecans in our model, we found a pronounced upregulation of syndecan-4 in the m. quadriceps at the early phase of arteriogenesis. Syndecan-4 has been described as a primary response gene induced by FGF-2 in vascular SMCs in vivo.30 Furthermore, overexpression of syndecan-4 resulted in a significant increase in cell growth and migration in response to FGF-2.31 Although heparan sulfate chains are required for syndecan-4 mediation of FGF-2 signal transduction, the increased responsiveness to FGF-2 is not due to an increased number of heparan sulfate-FGF binding sites but rather to an increased expression of the syndecan-4 cytoplasmic domain.31 Besides dimerization of the high-affinity receptor, syndecan-4 has been implicated in various processes, such as the regulation of focal adhesion assembly and protein kinase C activation32; however, which functional properties might be assigned to syndecan-4 during arteriogenesis remains to be clarified.
To define whether arteriogenesis is also naturally associated with an increased expression of FGF-1 and FGF-2, we performed hybridization studies. However, results displayed that neither FGF-1 nor FGF-2 shows a differential expression pattern. In previous studies from our laboratory, we have demonstrated that monocytes play an important role in adaptive (and cytokine induced) collateral artery growth.1 These mononuclear cells that have the capacity to transmigrate the EC layer have been shown to deliver growth factors such as FGF-2 to the vessel during arteriogenesis24 and accumulate in the perivascular space as early as 12 hours after femoral artery ligation.25 Our in vitro results showed that the stimulation of monocytes with LPS leads to an increased level of FGF-2 in these types of cells. Furthermore, flow cytometry results revealed that monocytes (which express FGFR-133) treated with FGF-2 show an increased expression of integrins that are part of the MAC-1 (αM/β2) or leukocyte function-associated antigen-1 (αL/β2) heterodimer. These two heterodimers are receptors that are responsible for monocyte interaction with the endothelium. Our adhesion assays revealed that FGF-1 and FGF-2 have the ability to stimulate monocyte adhesion to EC layers and that FGF-1 is an even more potent factor than MCP-1. These data strongly suggest that during adaptive arteriogenesis, FGFs (which might be released by hemodynamic forces from cells of the vessel wall as described for several in vitro systems previously34,35) attract and activate monocytes, which then in turn supply FGFs to the growing collateral arteries. This would mean that FGFs support arteriogenesis in a paracrine manner that is independent of an increased transcriptional activity in the vascular wall.
To test the role of FGFs in our animal model system, we infused PAS, a nontoxic sulfonic acid polymer that has been described to block the action of FGFs via complex formation36 (and PAS plus MCP-1), into the proximal stump of the occluded femoral artery. To investigate the specificity of PAS, we performed adhesion assays. The results showed that PAS significantly interfered with the action of FGF-1 and FGF-2 to stimulate adhesion of the monocytes to EC layers. However, the influence of MCP-1, a cytokine that strongly promotes arteriogenesis,1,24 as well as of VEGF, another heparin-binding growth factor with strong angiogenic properties, was not affected by PAS.
Our in vivo results demonstrated that 1 week of continuous intra-arterial infusion of PAS markedly reduced the size of growing collateral arteries. The strong immunoreactivity that we found for FGFR-1 in vascular SMCs (but not in ECs) suggests not only that FGFR-1 mediates the signal transduction cascade associated with the proliferation of SMCs but also that the reduced vessel size observed after infusion of PAS is due to a reduced SMC proliferation in growing collateral arteries resulting from an interference of PAS with the action of FGFs. This assumption is corroborated by results from Ueno et al,37 who recently showed that FGF-2 induced DNA synthesis and arterial SMC proliferation was abolished by overexpressing a dominant-negative truncated FGFR-1. Concomitant application of PAS, owing to its property to pass the EC layer of a vessel (oral communication, unpublished data, S. Liekens, 1999), and MCP-1 showed an even stronger negative effect on collateral artery growth than the application of PAS alone, indicating that a significant part of the monocyte-related arteriogenesis is caused by FGFs. Our in vitro results evidenced that PAS can block the action of both FGF-1 and FGF-2 (and maybe of other FGFs too) and that FGF-1 is an even more potent factor than MCP-1 in terms of stimulation of monocyte adhesion. Despite these findings, we consider it likely that the observed in vivo effect of PAS was particularly due to an interference with the action of FGF-2, inasmuch as FGF-1 function has been associated more with vessel differentiation and branching38 than with growing proper.
In summary, our data show that arteriogenesis is associated with an increased expression and activation of FGFR-1 and indicate that growth-promoting FGFs are supplied by monocytes in a paracrine manner. These results suggest that the action of therapeutically applied FGFs might be limited by the availability of their corresponding receptors that become upregulated only under specific conditions.
This work was supported by a grant of the Max-Planck-Gesellschaft. We want to thank F. Pipp for his help in animal treatment and E. Neubauer for technical assistance.
- Received November 13, 2000.
- Revision received November 21, 2002.
- Accepted January 31, 2003.
Ito WD, Arras M, Winkler B, Scholz D, Schaper J, Schaper W. Monocyte chemotactic protein-1 increases collateral and peripheral conductance after femoral artery occlusion. Circ Res. 1997; 80: 829–837.
van Royen N, Piek JJ, Buschmann I, Hoefer I, Voskuil M, Schaper W. Stimulation of arteriogenesis: a new concept for the treatment of arterial occlusive diseases. Cardiovasc Res. 2001; 49: 543–553.
Seiler C, Pohl T, Wustmann K, Hutter D, Nicolet PA, Windecker S, Eberli FR, Meier B. Promotion of collateral growth by granulocyte-macrophage colony-stimulating factor in patients with coronary artery disease: a randomized, double-blind, placebo-controlled study. Circulation. 2001; 23: 2012–2017.
Givol D, Yayon A. Complexity of FGF receptors: genetic basis for structural diversity and functional specificity. FASEB J. 1992; 6: 3362–3369.
Friesel RE, Maciag T. Molecular mechanisms of angiogenesis: fibroblast growth factor signal transduction. FASEB J. 1995; 9: 919–925.
Abraham JA, Mergia A, Whang JL, Tumola A, Gospodarowicz D, Fiddes JC. Nucleotide sequence of a bovine clone encoding the angiogenic protein, basic fibroblast growth factor. Science. 1986; 233: 545–548.
Yang HT, Deschenes MR, Ogilvie RW, Terjung RL. Basic fibroblast growth factor increases collateral blood flow in rats with femoral arterial ligation. Circ Res. 1996; 79: 62–69.
Hoefer IE, van Royen N, Buschmann IR, Piek JJ, Schaper W. Time course of arteriogenesis following femoral artery occlusion in the rabbit. Cardiovasc Res. 2001; 49: 609–617.
Sambrook J, Fritsch EF, Maniatis T. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press; 1989.
Arras M, Ito WD, Scholz D, Winkler B, Schaper J, Schaper W. Monocyte activation in angiogenesis and collateral growth in the rabbit hindlimb. J Clin Invest. 1997; 101: 41–50.
Elenius K, Maatta A, Salmivirta M, Jalkanen M. Growth factors induce 3T3 cells to express bFGF-binding syndecan. J Biol Chem. 1992; 267: 6435–6441.
Cizmeci Smith G, Langan E, Youkey J, Showalter LJ, Carey DJ. Syndecan-4 is a primary-response gene induced by basic fibroblast growth factor and arterial injury in vascular smooth muscle cells. Arterioscler Thromb Vasc Biol. 1997; 17: 172–180.
Volk R, Schwartz JJ, Li J, Rosenberg RD, Simons M. The role of syndecan cytoplasmic domain in basic fibroblast growth factor-dependent signal transduction. J Biol Chem. 1999; 274: 24417–24424.
Couchman RC, Woods A. Syndecan-4 and integrins: combinatorial in cell adhesion. J Cell Sci. 1999; 112: 3415–3420.
Rhoads DN, Eskin SG, McIntire LV. Fluid flow releases fibroblast growth factor-2 from human aortic smooth muscle cells. Arterioscler Thromb Vasc Biol. 2000; 20: 416–421.
Gloe T, Sohn HY, Meininger GA, Pohl U. Shear stress-induced release of basic fibroblast growth factor from endothelial cells is mediated by matrix interaction via integrin αvβ3. J Biol Chem. 2002; 277: 23453–23458.
Ueno H, Li JJ, Masuda S, Qi Z, Yamamoto H, Takeshita A. Adenovirus-mediated expression of the secreted form of basic fibroblast growth factor (FGF-2) induces cellular proliferation and angiogenesis in vivo. Arterioscler Thromb Vasc Biol. 1997; 17: 2453–2460.
Fernandez B, Buehler A, Wolfram F, Kostin S, Espanion G, Franz WM, Niemann H, Doevendans PA, Schaper W, Zimmermann R. Transgenic myocardial overexpression of FGF-1 increases coronary artery density and branching. Circ Res. 2000; 87: 207–213.