Platelet-Derived Growth Factor BBeCMediated Normalization of Dermal Interstitial Fluid Pressure After Mast Cell Degranulation Depends on 3 b
http://www.100md.com
《循环研究杂志》
the Department of Medical Biochemistry and Microbiology (A.L., K.R.), University of Uppsala, Sweden
the Department of Biomedicine (A.B., T.N., R.K.R.), Division of Physiology, University of Bergen, Norway.
Abstract
Interstitial fluid pressure (PIF) is one of the determinants of transcapillary fluid flux and thereby interstitial fluid volume. Cell-mediated control of PIF regulates fluid content in the loose interstitial connective tissues that surround the capillary bed. To maintain a normal PIF in dermis, 1 integrins mediate the tensile strength applied by connective tissue cells on the extracellular matrix. Platelet-derived growth factor (PDGF)-BB normalizes anaphylaxis-induced reduction of PIF. AntieC3 integrin IgG and a cyclic RGD peptide that inhibits the V3 integrin blocked the ability of PDGF-BB to normalize the lowered PIF resulting from mast cell degranulation. PDGF-BB was unable to normalize PIF lowered as a result of mast cell degranulation in 3-negative mice. Monoclonal antieC3 integrin IgG had no effect on PIF in normal mouse dermis. In contrast, administration of antieC1 integrin IgM lowered PIF in normal dermis but had no effect on PDGF-BBeCinduced normalization of PIF after anaphylaxis. Furthermore, collagen gel contraction mediated by wild-type mouse embryonal fibroblasts were only marginally affected by function-blocking antieC1 integrin antibodies, especially in the presence of PDGF-BB. In contrast, contraction mediated by V-negative mouse embryonic fibroblasts was completely blocked by antieC1 integrin antibodies, even after stimulation with PDGF-BB. These results show a previously unrecognized in vivo function for the V3 integrin, as a participant in the control of PIF during inflammatory reactions. Furthermore, our data demonstrate that PDGF-BB induces connective tissue cells to generate tensile forces via V3 during such reactions.
Key Words: anaphylaxis V integrins collagen gel contraction edema fluid homeostasis
Introduction
Fluid constantly filters across the capillary wall into the surrounding loose interstitial connective tissues. The rate of fluid filtration is determined by the net pressure difference across the capillary, ie, the difference in the colloid osmotic pressures in plasma (COPPL) and interstitial fluid (COPIF), and between hydrostatic capillary pressure (HPC) and interstitial fluid pressure (PIF) and a constant expressing capillary area and permeability (for a review see, Aukland and Reed1). Normally, PIF will prevent changes in interstitial volume because increased capillary filtration will raise interstitial volume and thereby PIF, which in turn acts across the capillary to limit further capillary filtration. Reduced capillary filtration will cause opposite changes in interstitial fluid volume and PIF, but again changes in PIF act to limit changes in interstitial volume.1 PIF has commonly been described as a function of interstitial volume and compliance such that PIF is increasing with increasing interstitial volume, but not necessarily in a linear fashion.1 In contrast to the role in maintaining a constant interstitial volume and the relationship normally described by compliance, a reduction in PIF has been observed to provide the driving force for edema formation during inflammation and burn injuries, with PIF falling in some cases to levels as low as eC150 mm Hg.2 Through a series of studies, we have demonstrated that cells in loose connective tissue actively control PIF and thereby fluid fluxes across capillary walls,2 and this study presents further details on this cellular control of PIF and thereby on the control of interstitial volume and edema formation.
Injection of antieC1 integrin IgG in rat dermis lowers PIF and leads to rapid edema formation.3 Monoclonal IgG specific for the collagen/laminin-binding integrin 21, but not monoclonal anti-11 IgG, induces a reduction in PIF in rat dermis, suggesting that 21 is of particular importance among the 1 integrins. Several proinflammatory mediators such as interleukin (IL)-1, tumor necrosis factor-, IL-6, and the prostaglandins (PGs) E1 and I2, as well as the phosphatidyl inositol 3-kinase (PI3K) inhibitor Wortmannin, all act to reduce dermal PIF.2,4eC6 A reduced PIF, resulting from blockade of 1 integrin function or inflammatory reactions can be normalized by a prostaglandin F analog and by platelet-derived growth factor (PDGF)-BB, but not PDGF-AA or fibroblast growth factor.5,7 By using mice with a knockout lesion for PDGF receptoreCinduced activation of PI3K, we demonstrated that the normalization of PIF that normally can be induced by PDGF-BB was completely abolished. Consequently, PI3K is necessary to induce the PIF-normalizing effects of PDGF-BB.8 We have proposed a model for how connective tissue cells participate in the control of PIF. According to this model, connective tissue cells, ie, fibrocytes and/or pericytes, exert a tension on the collagen/microfibrillar network via integrins, eg, 21.2,9,10 This collagen/microfibrillar network in turn restrains the intrinsic swelling properties of the hyaluronan/proteoglycan ground substance of the extracellular matrix (ECM).11 If the ground substance is allowed to swell, PIF will be reduced and if fluid is supplied by an intact blood circulation edema will form. Conversely, an increased cellular tension will increase PIF and filter fluid back across the capillaries or into the lymphatics.
Fibroblast-mediated collagen gel contraction12 depends on 1 integrins and is stimulated by PDGF.13eC15 PDGF-BBeCstimulated collagen gel contraction is mediated by V3 integrins in experimental conditions when 1 integrins are absent or functionally perturbed.16,17 Furthermore, forced expression of the V subunit in human osteosarcoma cells that lack endogenous expression of the collagen-binding integrin 21 promotes collagen gel contraction by these cells.18 These findings are consistent with an earlier report suggesting that the V3 integrin, under certain conditions, may function as a collagen receptor that mediates collagen gel contraction.19 Consistent similarities in the control of cell-mediated collagen-gel contraction in vitro and PIF and edema formation in vivo have been observed.3eC5,7,8 Against this background, it seemed reasonable to hypothesize that V3 integrins are involved in PDGF-BBeCdirected control of PIF in vivo. The present study was performed to test this hypothesis.
Materials and Methods
Animals
Female C57BL/6 and BALB/c mice were from Mllegaard (Lille Skensved, Denmark). Mice were fed ad libitum before experiments and anesthetized by an SC injection of a 1:1 mixture of Midazolam (Dormicum, Roche, Basel, Switzerland) combined with fentanyl/fluanisone (Hypnorm, Janssen Pharmaceutica, Beerse, Belgium) (0.1 mL/10 g body weight). 3-Null mice on C57BL/6eC129S4 background20,21 were backcrossed for 7 generations against BALB/c mice (Jackson Laboratories, Bar Harbor, Me) in the Massachusetts Institute of Technology facility and donated by Dr Richard Hynes (Massachusetts Institute of Technology). These mice are viable and fertile. The mice were catheterized in the right carotid artery to monitor arterial blood pressure and in the external jugular vein for IV injections. Circulatory arrest was induced by an IV injection of saturated KCl. Experiments were performed with the approval of and in accordance with the recommendations laid down by the Norwegian State Commission for Laboratory Animals.
PCR Analysis
Three-primer PCR was performed to genotype mouse tail DNA samples for screening the 3 integrin genotype. Common forward primer 1 (5' CTTAGACACCTGCTACGGGC 3') lay 5' of the pgk-neo cassette; reverse primer 2 (5' CACGAGACTAGTGAGACGTG 3') was neo specific and reverse primer 3 (5' CCTGCCTGAGGCTGAGTG 3') was wild-type specific. The PCR conditions (95°C, 10 minutes; followed by 30 cycles of 95°C, 1 minute; 66°C, 2 minutes; 72°C, 2 minutes; and finally 72°C, 10 minutes) yielded products of 538 bp (mutant) and 446 bp (wild-type).
Collagen Gel Contraction
Mouse embryonic fibroblasts were established by standard protocols22 from embryonic day (E) 9 to E10 integrin V-negative and wild-type littermate embryos.23 Cells were grown in DMEM supplemented with 10% FBS and antibiotics in culture dishes and flasks precoated overnight at &8°C with 50 e/mL native collagen (Vitrogen 100, Cohesion, Palo Alto, Calif) in PBS. Collagen gel contraction was quantified as described elsewhere.14 Briefly, 96-well plates were blocked in BSA overnight at 37°C and then washed in PBS. A collagen solution was made up from 5 parts double-concentrated DMEM, 1 part 0.2 mol/L HEPES (pH 8.0), and 4 parts collagen type I (Vitrogen 100, 3 mg/mL). Cells were washed 2 times in DMEM and diluted to a final concentration of 106 cells/mL in the same medium. One part cell suspension was mixed with 9 parts collagen solution. Cell-collagen suspension (100 e蘈) was added to each well, and gels were allowed to form for 1.5 hour at 37°C. Gels were detached by the ejection of 100 e蘈 of DMEM, with or without 50 ng/mL PDGF-BB and with or without 800 e/mL polyclonal anti-rat integrin 1 IgG24 into the wells. The relaxed, floating gels were further incubated at 37°C and gel diameters were measured microscopically at the indicated time points. The degree of contraction is presented as the gel area as a percentage of the original area. Each experiment was performed with a minimum of 3 samples per condition.
PIF Measurements
PIF was measured using sharpened glass capillaries (tip diameter, 3 to 5 e) filled with 0.5 mol/L NaCl colored with Evans Blue and connected to a servocontrolled counterpressure system.25,26 The punctures were performed through intact skin using a stereomicroscope (Wild M5, Heerbrugg, Switzerland) with the pipette tip located 0.3 to 0.5 mm below the skin. Care was taken not to cause any compression or retraction of the skin while puncturing. The animal was placed in a supine position and the left hind paw was carefully fixed to the table with surgical tape. Control PIF was measured with the circulation still intact. C48/80 (200 e, Sigma, St Louis, Mo) in 0.1 mL of 0.9% NaCl was injected IV and allowed to circulate for 2 minutes before circulatory arrest was induced. C48/80 causes a reduction of the PIF through degranulation of mast cells. Circulatory arrest was induced to prevent a potential underestimation of the lowered PIF caused by an increase in interstitial fluid volume as a result of increased transcapillary fluid flux. The lowering of the PIF was monitored for 30 minutes and test substances were then injected subdermally in a volume of 1 e蘈 of buffer (0.14 mol/L NaCl, 4.7 mmol/L KCl, 0.65 mmol/L MgSO4, 1.2 mmol/L CaCl2, 10 mmol/L HEPES, pH 7.4) using a 10-e蘈 chromatography syringe (Hamilton, Bonaduz, Switzerland) with a 33- or 34-gauge needle. Measurements were then continued for another 60 minutes. Test substances used were PDGF-BB, anti-rat integrin 1 (Ha2/5) monoclonal IgM (BD Pharmingen, San Diego, Calif), anti-mouse integrin 3 (HM3) monoclonal IgG (BD Pharmingen), and cyclo (Arg-Gly-Asp-D-Phe-Val) (Bachem, Bubendorf, Switzerland). The test substances were administered either separately or in combination. The pressure measurements were averaged in the following periods: 0 to 10, 11 to 20, 21 to 30, 31 to 40, 41 to 50, 51 to 60, 61 to 75, and 76 to 90 minutes after C48/80 injection. For a measurement to be accepted, the following criteria had to be fulfilled: (1) feedback gain could be changed without changing the pressure; (2) applying suction to the pipette by the pump increased the resistance in the pipette (this ensured contact between the pipette and the interstitial fluid, ie, the pipette was open); and (3) zero pressure did not change during the measurement.
PDGF-BB in Serum and Tissue
Eight wild-type BALB/c mice were anesthetized with an SC injection of 0.2 mL Ketalar/Dormicum and supplied with a catheter in the right jugular vein. The mice received an IV injection of 0.1 mL of NaCl (control, n=4) or 0.1 mL 200 e C48/80 (experiment, n=4). Blood samples were obtained via the intravenous catheter or heart puncture 2 minutes later. Circulatory arrest was then induced by an IV injection of saturated KCl and the skin on one hind paw was removed. Blood (&200 e蘈) was immediately added to 200 e蘈 of 3.8% citrate buffer containing 0.1 mol/L benzamidine, 103 U/mL Trasylol, and protease inhibitors (Complete, Roche Diagnostics, Mannheim, Germany). The mixture was centrifuged at 3000g for 20 minutes in a centrifuge without a brake. The skin samples (&100 mg) were minced with a scalpel and placed into tubes containing 200 e蘈 of extraction buffer (50 mmol/L Tris-HCl pH 7.4, 5 mmol/L EDTA, 1% NP-40, 1% Na-deoxycholate, 50 mmol/L NaF, 50 mmol/L -glycerophosphate and protease inhibitors [Complete, Roche Diagnostics]) for 24 hours. Tissue samples were freeze thawed twice during this period. PDGF-BB was determined in the supernatants of serum and tissue extracts using a Mouse/Rat PDGF-BB Immunoassay (R&D Systems, Minneapolis, Minn). The procedure described in the kit was followed and Calibrator diluent RD6eC3 was used to dilute the samples. The detection limit for the assay given by the manufacturer was 7.7 pg/mL (range 4.0 to 19.3 pg/mL).
Statistical Analysis
All values are mean±SD. Data were recorded for each 10-minute period within each experimental group up to 60 minutes and thereafter in 15 minutes periods. Each experimental group was compared for control and 21 to 30 minutes and 51 to 60 minutes using 1-way ANOVA followed by post hoc Bonferroni and StudenteCNewmaneCKeul tests. A value of P<0.05 was considered statistically significant.
Results
A subdermal injection of 0.9 e of anti-1 IgM in a volume of 1 e蘈 lowered PIF in C57BL/6 mouse dermis. Control PIF averaged eC0.96±0.51 mm Hg before subdermal injection and averaged eC3.08±1.63 mm Hg 60 minutes after IgM injection (P<0.05 compared with control). This is in agreement with earlier results showing that 1 integrins are involved in controlling PIF in naive dermis.3,7 A subdermal injection of 0.25 e of monoclonal antieC3 integrin IgG had, however, no significant effect on PIF during the investigated time period (Table) (P>0.05), indicating that 3 integrin(s) is not necessary to maintain dermal PIF. In the present experimental protocol, the test substance was injected some distance from the site where the micropuncture needle was inserted. A higher dose of the anti-1 antibody than the anti-3 IgG was therefore used because IgM could be assumed to diffuse less rapidly than the IgG to reach effective concentrations in the tissue segment where the PIF was recorded.
The reduction of PIF in mouse dermis resulting from systemic administration of C48/80 is reversed by a local instilment of PDGF-BB (Figure 1 and the Table) as previously reported.8 The PIF-normalizing effect of 0.2 ng PDGF-BB was not affected in C57BL/6 mouse dermis by simultaneous instilment of 0.9 e of monoclonal antieC1 integrin IgM (Figure 1 and the Table). Instilment of 0.25 e of antieC3 integrin IgG, together with the PDGF-BB, however, abolished the capability of the latter to normalize PIF (Figure 1 and the Table). Furthermore, the PIF normalizing effect of PDGF-BB was abolished when it was instilled together with 500 eol/L of the cyclic V3 inhibitor cyclo (Arg-Gly-Asp-D-Phe-Val) in a volume of 1 e蘈 (Figure 1 and the Table). The IC50 of this peptide for inhibition of 125I-vitronectin binding to immobilized recombinant V3 integrin is 24 nmol/L and 2000 nmol/L for V5,27 and in the range of 0.6 to 4.4 eol/L for V3 and >100 eol/L for 51-mediated cell adhesion.28 The test substance was injected at some distance from the site where the micropuncture needle was inserted, leading to at least a 20-fold dilution of the test substance during its diffusion to the tissue segment in which the PIF was recorded. Thus, the effective concentration of the inhibitor can be calculated to be <25 eol/L in the present experiments, suggesting that the effect of the inhibitor on PIF is caused by blockage of V3 and not 51.
To further investigate the possibility that the PIF-normalizing effect of PDGF-BB in mouse dermis was dependent on 3 integrins, we took advantage of 3-deficient mice.20 These mice carry a null mutation of the 3 integrin gene but are fertile and develop normally, except for disturbances of platelet and osteoclast functions leading to thrombastenia20 and osteosclerosis.21 Systemic administration of C48/80 in wild-type and 3 null BALB/c mice reduced dermal PIF (Figure 2 and the Table). Local instilment of PDGF-BB normalized the reduced PIF in wild-type, but not in the dermis of 3 integrin null, mice (Figure 2 and the Table). Together, our results strongly suggest that PDGF-BB acts on 3, but not on 1, integrins during the restoration of a normal dermal PIF that had been reduced by mast cell degranulation.
To validate the in vivo data suggesting a role of the V3 integrin in contractile processes, we took advantage of mouse embryonic fibroblasts (MEF) isolated from V-negative mice23 and compared the ability of these cells to contract collagen gels in vitro with wild-type MEF. Wild-type and V-negative MEF both contracted collagen gels (Figure 3A and 3B). Collagen-gel contraction mediated by the V-negative cells was completely inhibited by antieC1 integrin IgG, whereas the same IgG was unable to completely inhibit the contraction mediated by wild-type MEF (Figure 3A). PDGF-BB only marginally stimulated contraction by the wild-type MEF, but restored the ability of these cells to contract collagen gels in the presence of antieC1 integrin IgG (Figure 3B). The latter effect of PDGF-BB, ie, to partly override the blockade by antieC1 integrin IgG was not observed in V-negative MEF (Figure 3B). These findings demonstrate that PDGF-BB stimulates contractile process in vitro by a process that depends on V integrins.
Discussion
Normal interstitial fluid balance is a function of fluid flux across the capillaries and the outflux via the lymphatic system. The balance between these fluxes is such that the interstitium is normally "dry" and this will also keep PIF below ambient pressure. When the fluid flux across the capillaries into the interstitium exceeds lymph flow, interstitial volume will increase and visible edema will eventually occur. Increased transcapillary flux is caused by increased capillary permeability and/or changes in the pressures acting across the capillary wall enhancing filtration.1 The hydrostatic and colloid osmotic pressures that determine the pressure imbalance across the capillary creates the capillary filtration and also the ability to "autoregulate" the filtration across the capillary. When edema occurs, PIF will reach positive values. In contrast, we have demonstrated that inflammatory edemas are commonly associated with a lowering of PIF,2 which then becomes an active force in generating, rather than opposing, edema formation. However, it should be noted that the lowering of PIF can only function during the initial stages of the edema formation. As the edema is being established, the increasing interstitial fluid volume will eventually raise PIF to positive values. Furthermore, once the edema has been established, and PIF is positive, the increased capillary filtration must be caused by increased capillary filtration pressure and/or capillary permeability.
The present study was performed to further elucidate the events in the connective tissue cells associated with the lowering of PIF. In particular, we have investigated the phenomenon that the lowering of PIF can be reversed, ie, the capillary net filtration pressure normalized by PDGF-BB.7,8 The present study demonstrates that 3 integrins are necessary for this effect of PDGF-BB. The 3 integrin subunit associates with the IIb and V integrin subunits to form 2 RGD-directed integrins.29 Because the expression of IIb3 is restricted to megakaryocytes and platelets, it may be inferred that the 3 integrineCdependent effect on PIF is mediated by the integrin V3. Although a subdermal injection of PDGF-BB into naive dermis does not change PIF, PDGF-BB is able to normalize PIF that has been lowered after mast cell degranulation or blockade of 1 integrins.7,8 Similarly, PIF is not affected by perturbation of V3 integrin function in naive dermis and PIF is normal in undisturbed dermis of 3-null mice (present findings). In order for PDGF-BB to normalize PIF after anaphylaxis, 3 must be unperturbed (present findings). Our data thus provide evidence that PDGF-BBeCdirected usage of V3 for control of PIF occurs only after induction of anaphylaxis or perturbation of 1 integrin function. A role for V3 in concert with PDGF in counteracting edema formation during inflammation is consistent with the reported upregulation of activated PDGF receptors30eC33 and V334 in various inflammatory conditions. Based on the present findings, it is proposed that these receptors at least partly protect the inflamed tissue from the adverse effects of the inflammatory response on tissue integrity.
The tissue concentration of PDGF-BB is not known and we could not detect PDGF-BB in plasma or dermal extracts neither from normal mice nor during the acute phase of anaphylaxis (data not shown). The detection limit for the assay used was 7.7 pg/mL. In a previous immunohistochemical study, we reported that, in normal human skin, PDGF-AB/BB is confined to peripheral nerve fibers and to solitary cells of the epidermis and of the superficial dermis.31 PDGF-BB is believed to act in a paracrine fashion where the producer cell delivers the growth factor in intercellular contacts.35 The local concentration in these contacts is likely to be high. In the present experimental system, we injected 0.2 ng of PDGF-BB that diffused to the site where PIF was measured. Although the PDGF-BB became diluted during the diffusion, the cells at this site were exposed to a PDGF-BB concentration higher than what is likely to be present overall in normal dermis. In spite of this, we consider our experimental system physiologically relevant because the endogenous PDGF-BB is likely to have its effects locally and then in relatively high concentrations, as discussed above. It can be anticipated that during an acute systemic anaphylaxis that is a lethal condition the V3-directed control of PIF stimulated by endogenous PDGF BB is insufficient to counteract edema formation. Rather, the presently described edema-counteracting system involving PDGF-BB and V3 is likely to be relevant during later phases of anaphylaxis and acute inflammation. In line with this is previously reported data showing that during wound healing, as well as in carcinoma and inflammatory lesions, PDGF-AB/BB is upregulated and also detected in infiltrating macrophages and vascular cells.31,32,35
The injection into a tissue, with or without adding a substance or protein normally present or not, will inflict a trauma to the tissue. In the present study, as well as previous studies on the control of PIF, we have made several attempts to control for the injection trauma as such, including vascular reactions with mobilization of leukocytes, as well as to validate the effect of the specificity of the injected protein. Although there is an effect of the injection trauma, it has never been observed that injection of saline or unspecific proteins including nonspecific IgG elicits a lowering of PIF.2 In contrast, IgG, cytokines, and other agents that modulate collagen gel contraction in vitro also has an in vivo action on PIF.3eC5,7,8 Because we have demonstrated parallel effects of a number of substances in in vivo experiments and in fibroblast-mediated collagen gel contraction in vitro, we have used the latter system as a model for PIF control in vivo. We show here that mouse embryonic fibroblasts lacking V integrins, exclusively depended on 1 integrins to contract collagen gels even when the cells were stimulated by PDGF-BB. Wild-type fibroblasts, in contrast, contracted collagen gels when 1 integrins were blocked, and this contraction was stimulated by PDGF-BB. These findings show that V3 is necessary for collagen-gel contraction when 1 integrins are blocked and that this process is stimulated by PDGF-BB. Several reports have shown that under certain conditions the V3 integrin is able to mediate collagen gel contraction in vitro.16eC19 Many inflammatory agents, including PGE15 and IL-1,6 induce a reduction in PIF in normal dermis and inhibit cell-mediated collagen gel contraction. The inhibitory effects on collagen gel contraction by both these agents can at least partly be overcome by stimulation with PDGF-BB.5,15 Further studies will be needed to investigate the hypothesis that these proinflammatory agents block 1 integrins but allow PDGF-BB stimulated V3-mediated collagen-gel contraction in vitro and control of PIF in vivo.
The ligand for V3 that is involved in PIF control in vivo is not defined and candidates include the collagen type I fibers and microfibrils.36 Available data suggest that native triple helical interstitial collagens are not recognized by the RGD-directed integrin V3.37 The nature of the PDGF-BBeCstimulated celleCcollagen interaction mediated by V3 that is needed for collagen gel contraction in vitro and potentially in vivo is therefore not apparent. Further studies are needed to characterize the in vivo ECM ligand that is recognized by V3 and functions to restrain dermal tissue swelling. Regardless of the ECM ligand that is recognized by V3 in connective tissues and/or in collagen gels in vitro, the fact that stimulation with PDGF-BB was needed for V3-mediated contraction in vitro and in vivo suggests a requirement for activation of the integrin through inside-out signaling.29 Such activation involves affinity modulation by conformational changes of the integrin.38,39 PDGF-BBeCregulated activation of V3 may thus be a suitable target for the development of pharmaceutical agents for the treatment of uncontrolled edema, eg, in septic shock.
In conclusion, the results from this study support the hypothesis that PDGF-BB counteracts a tendency to form edema by stimulating the activity of V3-integrins. Furthermore, our data suggest that under normal conditions, tension between connective tissue cells and the dermal fibers is maintained by 1-integrineCmediated contraction. Proinflammatory mediators are likely to block the 1 integrins, causing a reduction in PIF and edema formation, a process that is opposed by PDGF-BBeCdirected V3-mediated contraction.
Acknowledgments
This study was supported by grants from the Swedish Cancer Foundation (to K.R.), the Gustaf V:s 80-rsfond (to K.R.), and the Norwegian Research Council (to R.K.R.). Ann-Marie Gustafson and Gerd Salvesen are gratefully acknowledged for technical assistance. We thank Dr Richard O. Hynes for helpful comments and for the 3-negative mice.
References
Aukland K, Reed RK. Interstitial-lymphatic mechanisms in the control of extracellular fluid volume. Physiol Rev. 1993; 73: 1eC78.
Wiig H, Rubin K, Reed RK. New and active role of the interstitium in control of interstitial fluid pressure: potential therapeutic consequences. Acta Anaesthesiol Scand. 2003; 47: 111eC121.
Reed RK, Rubin K, Wiig H, Rodt S. Blockade of 1-integrins in skin causes edema through lowering of interstitial fluid pressure. Circ Res. 1992; 71: 978eC983.
hlen K, Berg A, Stiger F, Tengholm A, Siegbahn A, Gylfe E, Reed RK, Rubin K. Cell interactions with collagen matrices in vivo and in vitro depend on phosphatidylinositol 3-kinase and free cytoplasmic calcium. Cell Adhes Commun. 1998; 5: 461eC473.
Berg A, Ekwall AK, Rubin K, Stjernschantz J, Reed RK. Effect of PGE1, PGI2, and PGF2 analogs on collagen gel compaction in vitro and interstitial pressure in vivo. Am J Physiol. 1998; 274: H663eCH671.
Nedreb T, Berg A, Reed RK. Effect of tumor necrosis factor-, IL-1, and IL-6 on interstitial fluid pressure in rat skin. Am J Physiol. 1999; 277: H1857eCH1862.
Rodt S, hlen K, Berg A, Rubin K, Reed RK. A novel physiological function for platelet-derived growth factor-BB in rat dermis. J Physiol (Lond). 1996; 495: 193eC200.
Heuchel R, Berg A, Tallquist M, hlen K, Reed RK, Rubin K, Claesson-Welsh L, Heldin C-H, Soriano P. Platelet-derived growth factor receptor regulates interstitial fluid homeostasis through phosphatidylinositol-3' kinase signaling. Proc Natl Acad Sci U S A. 1999; 96: 11410eC11415.
Rubin K, Gullberg D, Tomasini-Johansson B, Reed RK, Rydeen C, Borg TK. Molecular recognition of the extracellular matrix by cell surface receptors. In: Comper WD, ed. Extracellular Matrix, Volume 2: Molecular Components and Interactions. Reading, UK: Harwood Academic Publishers; 1996: 262eC309.
Reed RK, Berg A, Gjerde EA, Rubin K. Control of interstitial fluid pressure: role of 1-integrins. Semin Nephrol. 2001; 21: 222eC230.
Meyer FA, Laver-Rudich Z, Tanenbaum R. Evidence for a mechanical coupling of glycoprotein microfibrils with collagen fibrils in Wharton’s jelly. Biochim Biophys Acta. 1983; 755: 376eC387.
Grinnell F. Fibroblast biology in three-dimensional collagen matrices. Trends Cell Biol. 2003; 13: 264eC269.
Clark RA, Folkvord JM, Hart CE, Murray MJ, McPherson JM. Platelet isoforms of platelet-derived growth factor stimulate fibroblasts to contract collagen matrices. J Clin Invest. 1989; 84: 1036eC1040.
Gullberg D, Tingstrm A, Thuresson AC, Olsson L, Terracio L, Borg TK, Rubin K. 1 integrin-mediated collagen gel contraction is stimulated by PDGF. Exp Cell Res. 1990; 186: 264eC272.
Tingstrm A, Heldin C-H, Rubin K. Regulation of fibroblast-mediated collagen gel contraction by platelet-derived growth factor, interleukin-1 and transforming growth factor-1. J Cell Sci. 1992; 102: 315eC322.
Cooke ME, Sakai T, Mosher DF. Contraction of collagen matrices mediated by 21A and V3 integrins. J Cell Sci. 2000; 113: 2375eC2383.
Grundstrm G, Mosher DF, Sakai T, Rubin K. Integrin V3 mediates platelet-derived growth factor-BB-stimulated collagen gel contraction in cells expressing signaling deficient integrin 21. Exp Cell Res. 2003; 291: 463eC473.
Levinson H, Hopper JE, Ehrlich HP. Overexpression of integrin V promotes human osteosarcoma cell populated collagen lattice contraction and cell migration. J Cell Physiol. 2002; 193: 219eC224.
Agrez MV, Bates RC, Boyd AW, Burns GF. Arg-Gly-Asp-containing peptides expose novel collagen receptors on fibroblasts: implications for wound healing. Cell Regul. 1991; 2: 1035eC1044.
Hodivala-Dilke KM, McHugh KP, Tsakiris DA, Rayburn H, Crowley D, Ullman-Cullere M, Ross FP, Coller BS, Teitelbaum S, Hynes RO. 3-integrin-deficient mice are a model for Glanzmann thrombasthenia showing placental defects and reduced survival. J Clin Invest. 1999; 103: 229eC238.
McHugh KP, Hodivala-Dilke K, Zheng MH, Namba N, Lam J, Novack D, Feng X, Ross FP, Hynes RO, Teitelbaum SL. Mice lacking 3 integrins are osteosclerotic because of dysfunctional osteoclasts. J Clin Invest. 2000; 105: 433eC440.
Todaro GJ, Green H. Quantitative studies of the growth of mouse embryo cells in culture and their development into established lines. J Cell Biol. 1963; 17: 299eC313.
Bader BL, Rayburn H, Crowley D, Hynes RO. Extensive vasculogenesis, angiogenesis, and organogenesis precede lethality in mice lacking all V integrins. Cell. 1998; 95: 507eC519.
Gullberg D, Terracio L, Borg TK, Rubin K. Identification of integrin-like matrix receptors with affinity for interstitial collagens. J Biol Chem. 1989; 264: 12686eC12694.
Wiederhielm CA, Woodbury JW, Kirk S, Rushmer RF. Pulsatile pressures in the microcirculation of frog’s mesentery. Am J Physiol. 1964; 207: 173eC176.
Wiig H, Reed RK, Aukland K. Micropuncture measurement of interstitial fluid pressure in rat subcutis and skeletal muscle: comparison to wick-in-needle technique. Microvasc Res. 1981; 21: 308eC319.
Nisato RE, Tille JC, Jonczyk A, Goodman SL, Pepper MS. V3 and V5 integrin antagonists inhibit angiogenesis in vitro. Angiogenesis. 2003; 6: 105eC119.
Srivatsa SS, Fitzpatrick LA, Tsao PW, Reilly TM, Holmes DR Jr, Schwartz RS, Mousa SA. Selective V3 integrin blockade potently limits neointimal hyperplasia and lumen stenosis following deep coronary arterial stent injury: evidence for the functional importance of integrin V3 and osteopontin expression during neointima formation. Cardiovasc Res. 1997; 36: 408eC428.
Hynes RO. Integrins: bidirectional, allosteric signaling machines. Cell. 2002; 110: 673eC687.
Rubin K, Tingstrm A, Hansson GK, Larsson E, Rnnstrand L, Klareskog L, Claesson-Welsh L, Heldin C-H, Fellstrm B, Terracio L. Induction of B-type receptors for platelet-derived growth factor in vascular inflammation: possible implications for development of vascular proliferative lesions. Lancet. 1988; 1: 1353eC1356.
Reuterdahl C, Sundberg C, Rubin K, Funa K, Gerdin B. Tissue localization of -receptors for platelet-derived growth factor and platelet-derived growth factor B chain during wound repair in humans. J Clin Invest. 1993; 91: 2065eC2075.
Sundberg C, Ljungstrm M, Lindmark G, Gerdin B, Rubin K. Microvascular pericytes express platelet-derived growth factor- receptors in human healing wounds and colorectal adenocarcinoma. Am J Pathol. 1993; 143: 1377eC1388.
Rajkumar VS, Sundberg C, Abraham DJ, Rubin K, Black CM. Activation of microvascular pericytes in autoimmune Raynaud’s phenomenon and systemic sclerosis. Arthritis Rheum. 1999; 42: 930eC941.
Wilder RL. Integrin V3 as a target for treatment of rheumatoid arthritis and related rheumatic diseases. Ann Rheum Dis. 2002; 61 (suppl 2): ii96eCii99.
Heldin CH, Westermark B. Mechanism of action and in vivo role of platelet-derived growth factor. Physiol Rev. 1999; 79: 1283eC1316.
Kielty CM, Shuttleworth CA. Microfibrillar elements of the dermal matrix. Microsc Res Tech. 1997; 38: 413eC427.
Davis GE. Affinity of integrins for damaged extracellular matrix: V3 binds to denatured collagen type I through RGD sites. Biochem Biophys Res Commun. 1992; 182: 1025eC1031.
Takagi J, Petre BM, Walz T, Springer TA. Global conformational rearrangements in integrin extracellular domains in outside-in and inside-out signaling. Cell. 2002; 110: 599eC511.
Springer TA, Wang JH. The three-dimensional structure of integrins and their ligands, and conformational regulation of cell adhesion. Adv Protein Chem. 2004; 68: 29eC63.(sa Lideen, Ansgar Berg, T)
the Department of Biomedicine (A.B., T.N., R.K.R.), Division of Physiology, University of Bergen, Norway.
Abstract
Interstitial fluid pressure (PIF) is one of the determinants of transcapillary fluid flux and thereby interstitial fluid volume. Cell-mediated control of PIF regulates fluid content in the loose interstitial connective tissues that surround the capillary bed. To maintain a normal PIF in dermis, 1 integrins mediate the tensile strength applied by connective tissue cells on the extracellular matrix. Platelet-derived growth factor (PDGF)-BB normalizes anaphylaxis-induced reduction of PIF. AntieC3 integrin IgG and a cyclic RGD peptide that inhibits the V3 integrin blocked the ability of PDGF-BB to normalize the lowered PIF resulting from mast cell degranulation. PDGF-BB was unable to normalize PIF lowered as a result of mast cell degranulation in 3-negative mice. Monoclonal antieC3 integrin IgG had no effect on PIF in normal mouse dermis. In contrast, administration of antieC1 integrin IgM lowered PIF in normal dermis but had no effect on PDGF-BBeCinduced normalization of PIF after anaphylaxis. Furthermore, collagen gel contraction mediated by wild-type mouse embryonal fibroblasts were only marginally affected by function-blocking antieC1 integrin antibodies, especially in the presence of PDGF-BB. In contrast, contraction mediated by V-negative mouse embryonic fibroblasts was completely blocked by antieC1 integrin antibodies, even after stimulation with PDGF-BB. These results show a previously unrecognized in vivo function for the V3 integrin, as a participant in the control of PIF during inflammatory reactions. Furthermore, our data demonstrate that PDGF-BB induces connective tissue cells to generate tensile forces via V3 during such reactions.
Key Words: anaphylaxis V integrins collagen gel contraction edema fluid homeostasis
Introduction
Fluid constantly filters across the capillary wall into the surrounding loose interstitial connective tissues. The rate of fluid filtration is determined by the net pressure difference across the capillary, ie, the difference in the colloid osmotic pressures in plasma (COPPL) and interstitial fluid (COPIF), and between hydrostatic capillary pressure (HPC) and interstitial fluid pressure (PIF) and a constant expressing capillary area and permeability (for a review see, Aukland and Reed1). Normally, PIF will prevent changes in interstitial volume because increased capillary filtration will raise interstitial volume and thereby PIF, which in turn acts across the capillary to limit further capillary filtration. Reduced capillary filtration will cause opposite changes in interstitial fluid volume and PIF, but again changes in PIF act to limit changes in interstitial volume.1 PIF has commonly been described as a function of interstitial volume and compliance such that PIF is increasing with increasing interstitial volume, but not necessarily in a linear fashion.1 In contrast to the role in maintaining a constant interstitial volume and the relationship normally described by compliance, a reduction in PIF has been observed to provide the driving force for edema formation during inflammation and burn injuries, with PIF falling in some cases to levels as low as eC150 mm Hg.2 Through a series of studies, we have demonstrated that cells in loose connective tissue actively control PIF and thereby fluid fluxes across capillary walls,2 and this study presents further details on this cellular control of PIF and thereby on the control of interstitial volume and edema formation.
Injection of antieC1 integrin IgG in rat dermis lowers PIF and leads to rapid edema formation.3 Monoclonal IgG specific for the collagen/laminin-binding integrin 21, but not monoclonal anti-11 IgG, induces a reduction in PIF in rat dermis, suggesting that 21 is of particular importance among the 1 integrins. Several proinflammatory mediators such as interleukin (IL)-1, tumor necrosis factor-, IL-6, and the prostaglandins (PGs) E1 and I2, as well as the phosphatidyl inositol 3-kinase (PI3K) inhibitor Wortmannin, all act to reduce dermal PIF.2,4eC6 A reduced PIF, resulting from blockade of 1 integrin function or inflammatory reactions can be normalized by a prostaglandin F analog and by platelet-derived growth factor (PDGF)-BB, but not PDGF-AA or fibroblast growth factor.5,7 By using mice with a knockout lesion for PDGF receptoreCinduced activation of PI3K, we demonstrated that the normalization of PIF that normally can be induced by PDGF-BB was completely abolished. Consequently, PI3K is necessary to induce the PIF-normalizing effects of PDGF-BB.8 We have proposed a model for how connective tissue cells participate in the control of PIF. According to this model, connective tissue cells, ie, fibrocytes and/or pericytes, exert a tension on the collagen/microfibrillar network via integrins, eg, 21.2,9,10 This collagen/microfibrillar network in turn restrains the intrinsic swelling properties of the hyaluronan/proteoglycan ground substance of the extracellular matrix (ECM).11 If the ground substance is allowed to swell, PIF will be reduced and if fluid is supplied by an intact blood circulation edema will form. Conversely, an increased cellular tension will increase PIF and filter fluid back across the capillaries or into the lymphatics.
Fibroblast-mediated collagen gel contraction12 depends on 1 integrins and is stimulated by PDGF.13eC15 PDGF-BBeCstimulated collagen gel contraction is mediated by V3 integrins in experimental conditions when 1 integrins are absent or functionally perturbed.16,17 Furthermore, forced expression of the V subunit in human osteosarcoma cells that lack endogenous expression of the collagen-binding integrin 21 promotes collagen gel contraction by these cells.18 These findings are consistent with an earlier report suggesting that the V3 integrin, under certain conditions, may function as a collagen receptor that mediates collagen gel contraction.19 Consistent similarities in the control of cell-mediated collagen-gel contraction in vitro and PIF and edema formation in vivo have been observed.3eC5,7,8 Against this background, it seemed reasonable to hypothesize that V3 integrins are involved in PDGF-BBeCdirected control of PIF in vivo. The present study was performed to test this hypothesis.
Materials and Methods
Animals
Female C57BL/6 and BALB/c mice were from Mllegaard (Lille Skensved, Denmark). Mice were fed ad libitum before experiments and anesthetized by an SC injection of a 1:1 mixture of Midazolam (Dormicum, Roche, Basel, Switzerland) combined with fentanyl/fluanisone (Hypnorm, Janssen Pharmaceutica, Beerse, Belgium) (0.1 mL/10 g body weight). 3-Null mice on C57BL/6eC129S4 background20,21 were backcrossed for 7 generations against BALB/c mice (Jackson Laboratories, Bar Harbor, Me) in the Massachusetts Institute of Technology facility and donated by Dr Richard Hynes (Massachusetts Institute of Technology). These mice are viable and fertile. The mice were catheterized in the right carotid artery to monitor arterial blood pressure and in the external jugular vein for IV injections. Circulatory arrest was induced by an IV injection of saturated KCl. Experiments were performed with the approval of and in accordance with the recommendations laid down by the Norwegian State Commission for Laboratory Animals.
PCR Analysis
Three-primer PCR was performed to genotype mouse tail DNA samples for screening the 3 integrin genotype. Common forward primer 1 (5' CTTAGACACCTGCTACGGGC 3') lay 5' of the pgk-neo cassette; reverse primer 2 (5' CACGAGACTAGTGAGACGTG 3') was neo specific and reverse primer 3 (5' CCTGCCTGAGGCTGAGTG 3') was wild-type specific. The PCR conditions (95°C, 10 minutes; followed by 30 cycles of 95°C, 1 minute; 66°C, 2 minutes; 72°C, 2 minutes; and finally 72°C, 10 minutes) yielded products of 538 bp (mutant) and 446 bp (wild-type).
Collagen Gel Contraction
Mouse embryonic fibroblasts were established by standard protocols22 from embryonic day (E) 9 to E10 integrin V-negative and wild-type littermate embryos.23 Cells were grown in DMEM supplemented with 10% FBS and antibiotics in culture dishes and flasks precoated overnight at &8°C with 50 e/mL native collagen (Vitrogen 100, Cohesion, Palo Alto, Calif) in PBS. Collagen gel contraction was quantified as described elsewhere.14 Briefly, 96-well plates were blocked in BSA overnight at 37°C and then washed in PBS. A collagen solution was made up from 5 parts double-concentrated DMEM, 1 part 0.2 mol/L HEPES (pH 8.0), and 4 parts collagen type I (Vitrogen 100, 3 mg/mL). Cells were washed 2 times in DMEM and diluted to a final concentration of 106 cells/mL in the same medium. One part cell suspension was mixed with 9 parts collagen solution. Cell-collagen suspension (100 e蘈) was added to each well, and gels were allowed to form for 1.5 hour at 37°C. Gels were detached by the ejection of 100 e蘈 of DMEM, with or without 50 ng/mL PDGF-BB and with or without 800 e/mL polyclonal anti-rat integrin 1 IgG24 into the wells. The relaxed, floating gels were further incubated at 37°C and gel diameters were measured microscopically at the indicated time points. The degree of contraction is presented as the gel area as a percentage of the original area. Each experiment was performed with a minimum of 3 samples per condition.
PIF Measurements
PIF was measured using sharpened glass capillaries (tip diameter, 3 to 5 e) filled with 0.5 mol/L NaCl colored with Evans Blue and connected to a servocontrolled counterpressure system.25,26 The punctures were performed through intact skin using a stereomicroscope (Wild M5, Heerbrugg, Switzerland) with the pipette tip located 0.3 to 0.5 mm below the skin. Care was taken not to cause any compression or retraction of the skin while puncturing. The animal was placed in a supine position and the left hind paw was carefully fixed to the table with surgical tape. Control PIF was measured with the circulation still intact. C48/80 (200 e, Sigma, St Louis, Mo) in 0.1 mL of 0.9% NaCl was injected IV and allowed to circulate for 2 minutes before circulatory arrest was induced. C48/80 causes a reduction of the PIF through degranulation of mast cells. Circulatory arrest was induced to prevent a potential underestimation of the lowered PIF caused by an increase in interstitial fluid volume as a result of increased transcapillary fluid flux. The lowering of the PIF was monitored for 30 minutes and test substances were then injected subdermally in a volume of 1 e蘈 of buffer (0.14 mol/L NaCl, 4.7 mmol/L KCl, 0.65 mmol/L MgSO4, 1.2 mmol/L CaCl2, 10 mmol/L HEPES, pH 7.4) using a 10-e蘈 chromatography syringe (Hamilton, Bonaduz, Switzerland) with a 33- or 34-gauge needle. Measurements were then continued for another 60 minutes. Test substances used were PDGF-BB, anti-rat integrin 1 (Ha2/5) monoclonal IgM (BD Pharmingen, San Diego, Calif), anti-mouse integrin 3 (HM3) monoclonal IgG (BD Pharmingen), and cyclo (Arg-Gly-Asp-D-Phe-Val) (Bachem, Bubendorf, Switzerland). The test substances were administered either separately or in combination. The pressure measurements were averaged in the following periods: 0 to 10, 11 to 20, 21 to 30, 31 to 40, 41 to 50, 51 to 60, 61 to 75, and 76 to 90 minutes after C48/80 injection. For a measurement to be accepted, the following criteria had to be fulfilled: (1) feedback gain could be changed without changing the pressure; (2) applying suction to the pipette by the pump increased the resistance in the pipette (this ensured contact between the pipette and the interstitial fluid, ie, the pipette was open); and (3) zero pressure did not change during the measurement.
PDGF-BB in Serum and Tissue
Eight wild-type BALB/c mice were anesthetized with an SC injection of 0.2 mL Ketalar/Dormicum and supplied with a catheter in the right jugular vein. The mice received an IV injection of 0.1 mL of NaCl (control, n=4) or 0.1 mL 200 e C48/80 (experiment, n=4). Blood samples were obtained via the intravenous catheter or heart puncture 2 minutes later. Circulatory arrest was then induced by an IV injection of saturated KCl and the skin on one hind paw was removed. Blood (&200 e蘈) was immediately added to 200 e蘈 of 3.8% citrate buffer containing 0.1 mol/L benzamidine, 103 U/mL Trasylol, and protease inhibitors (Complete, Roche Diagnostics, Mannheim, Germany). The mixture was centrifuged at 3000g for 20 minutes in a centrifuge without a brake. The skin samples (&100 mg) were minced with a scalpel and placed into tubes containing 200 e蘈 of extraction buffer (50 mmol/L Tris-HCl pH 7.4, 5 mmol/L EDTA, 1% NP-40, 1% Na-deoxycholate, 50 mmol/L NaF, 50 mmol/L -glycerophosphate and protease inhibitors [Complete, Roche Diagnostics]) for 24 hours. Tissue samples were freeze thawed twice during this period. PDGF-BB was determined in the supernatants of serum and tissue extracts using a Mouse/Rat PDGF-BB Immunoassay (R&D Systems, Minneapolis, Minn). The procedure described in the kit was followed and Calibrator diluent RD6eC3 was used to dilute the samples. The detection limit for the assay given by the manufacturer was 7.7 pg/mL (range 4.0 to 19.3 pg/mL).
Statistical Analysis
All values are mean±SD. Data were recorded for each 10-minute period within each experimental group up to 60 minutes and thereafter in 15 minutes periods. Each experimental group was compared for control and 21 to 30 minutes and 51 to 60 minutes using 1-way ANOVA followed by post hoc Bonferroni and StudenteCNewmaneCKeul tests. A value of P<0.05 was considered statistically significant.
Results
A subdermal injection of 0.9 e of anti-1 IgM in a volume of 1 e蘈 lowered PIF in C57BL/6 mouse dermis. Control PIF averaged eC0.96±0.51 mm Hg before subdermal injection and averaged eC3.08±1.63 mm Hg 60 minutes after IgM injection (P<0.05 compared with control). This is in agreement with earlier results showing that 1 integrins are involved in controlling PIF in naive dermis.3,7 A subdermal injection of 0.25 e of monoclonal antieC3 integrin IgG had, however, no significant effect on PIF during the investigated time period (Table) (P>0.05), indicating that 3 integrin(s) is not necessary to maintain dermal PIF. In the present experimental protocol, the test substance was injected some distance from the site where the micropuncture needle was inserted. A higher dose of the anti-1 antibody than the anti-3 IgG was therefore used because IgM could be assumed to diffuse less rapidly than the IgG to reach effective concentrations in the tissue segment where the PIF was recorded.
The reduction of PIF in mouse dermis resulting from systemic administration of C48/80 is reversed by a local instilment of PDGF-BB (Figure 1 and the Table) as previously reported.8 The PIF-normalizing effect of 0.2 ng PDGF-BB was not affected in C57BL/6 mouse dermis by simultaneous instilment of 0.9 e of monoclonal antieC1 integrin IgM (Figure 1 and the Table). Instilment of 0.25 e of antieC3 integrin IgG, together with the PDGF-BB, however, abolished the capability of the latter to normalize PIF (Figure 1 and the Table). Furthermore, the PIF normalizing effect of PDGF-BB was abolished when it was instilled together with 500 eol/L of the cyclic V3 inhibitor cyclo (Arg-Gly-Asp-D-Phe-Val) in a volume of 1 e蘈 (Figure 1 and the Table). The IC50 of this peptide for inhibition of 125I-vitronectin binding to immobilized recombinant V3 integrin is 24 nmol/L and 2000 nmol/L for V5,27 and in the range of 0.6 to 4.4 eol/L for V3 and >100 eol/L for 51-mediated cell adhesion.28 The test substance was injected at some distance from the site where the micropuncture needle was inserted, leading to at least a 20-fold dilution of the test substance during its diffusion to the tissue segment in which the PIF was recorded. Thus, the effective concentration of the inhibitor can be calculated to be <25 eol/L in the present experiments, suggesting that the effect of the inhibitor on PIF is caused by blockage of V3 and not 51.
To further investigate the possibility that the PIF-normalizing effect of PDGF-BB in mouse dermis was dependent on 3 integrins, we took advantage of 3-deficient mice.20 These mice carry a null mutation of the 3 integrin gene but are fertile and develop normally, except for disturbances of platelet and osteoclast functions leading to thrombastenia20 and osteosclerosis.21 Systemic administration of C48/80 in wild-type and 3 null BALB/c mice reduced dermal PIF (Figure 2 and the Table). Local instilment of PDGF-BB normalized the reduced PIF in wild-type, but not in the dermis of 3 integrin null, mice (Figure 2 and the Table). Together, our results strongly suggest that PDGF-BB acts on 3, but not on 1, integrins during the restoration of a normal dermal PIF that had been reduced by mast cell degranulation.
To validate the in vivo data suggesting a role of the V3 integrin in contractile processes, we took advantage of mouse embryonic fibroblasts (MEF) isolated from V-negative mice23 and compared the ability of these cells to contract collagen gels in vitro with wild-type MEF. Wild-type and V-negative MEF both contracted collagen gels (Figure 3A and 3B). Collagen-gel contraction mediated by the V-negative cells was completely inhibited by antieC1 integrin IgG, whereas the same IgG was unable to completely inhibit the contraction mediated by wild-type MEF (Figure 3A). PDGF-BB only marginally stimulated contraction by the wild-type MEF, but restored the ability of these cells to contract collagen gels in the presence of antieC1 integrin IgG (Figure 3B). The latter effect of PDGF-BB, ie, to partly override the blockade by antieC1 integrin IgG was not observed in V-negative MEF (Figure 3B). These findings demonstrate that PDGF-BB stimulates contractile process in vitro by a process that depends on V integrins.
Discussion
Normal interstitial fluid balance is a function of fluid flux across the capillaries and the outflux via the lymphatic system. The balance between these fluxes is such that the interstitium is normally "dry" and this will also keep PIF below ambient pressure. When the fluid flux across the capillaries into the interstitium exceeds lymph flow, interstitial volume will increase and visible edema will eventually occur. Increased transcapillary flux is caused by increased capillary permeability and/or changes in the pressures acting across the capillary wall enhancing filtration.1 The hydrostatic and colloid osmotic pressures that determine the pressure imbalance across the capillary creates the capillary filtration and also the ability to "autoregulate" the filtration across the capillary. When edema occurs, PIF will reach positive values. In contrast, we have demonstrated that inflammatory edemas are commonly associated with a lowering of PIF,2 which then becomes an active force in generating, rather than opposing, edema formation. However, it should be noted that the lowering of PIF can only function during the initial stages of the edema formation. As the edema is being established, the increasing interstitial fluid volume will eventually raise PIF to positive values. Furthermore, once the edema has been established, and PIF is positive, the increased capillary filtration must be caused by increased capillary filtration pressure and/or capillary permeability.
The present study was performed to further elucidate the events in the connective tissue cells associated with the lowering of PIF. In particular, we have investigated the phenomenon that the lowering of PIF can be reversed, ie, the capillary net filtration pressure normalized by PDGF-BB.7,8 The present study demonstrates that 3 integrins are necessary for this effect of PDGF-BB. The 3 integrin subunit associates with the IIb and V integrin subunits to form 2 RGD-directed integrins.29 Because the expression of IIb3 is restricted to megakaryocytes and platelets, it may be inferred that the 3 integrineCdependent effect on PIF is mediated by the integrin V3. Although a subdermal injection of PDGF-BB into naive dermis does not change PIF, PDGF-BB is able to normalize PIF that has been lowered after mast cell degranulation or blockade of 1 integrins.7,8 Similarly, PIF is not affected by perturbation of V3 integrin function in naive dermis and PIF is normal in undisturbed dermis of 3-null mice (present findings). In order for PDGF-BB to normalize PIF after anaphylaxis, 3 must be unperturbed (present findings). Our data thus provide evidence that PDGF-BBeCdirected usage of V3 for control of PIF occurs only after induction of anaphylaxis or perturbation of 1 integrin function. A role for V3 in concert with PDGF in counteracting edema formation during inflammation is consistent with the reported upregulation of activated PDGF receptors30eC33 and V334 in various inflammatory conditions. Based on the present findings, it is proposed that these receptors at least partly protect the inflamed tissue from the adverse effects of the inflammatory response on tissue integrity.
The tissue concentration of PDGF-BB is not known and we could not detect PDGF-BB in plasma or dermal extracts neither from normal mice nor during the acute phase of anaphylaxis (data not shown). The detection limit for the assay used was 7.7 pg/mL. In a previous immunohistochemical study, we reported that, in normal human skin, PDGF-AB/BB is confined to peripheral nerve fibers and to solitary cells of the epidermis and of the superficial dermis.31 PDGF-BB is believed to act in a paracrine fashion where the producer cell delivers the growth factor in intercellular contacts.35 The local concentration in these contacts is likely to be high. In the present experimental system, we injected 0.2 ng of PDGF-BB that diffused to the site where PIF was measured. Although the PDGF-BB became diluted during the diffusion, the cells at this site were exposed to a PDGF-BB concentration higher than what is likely to be present overall in normal dermis. In spite of this, we consider our experimental system physiologically relevant because the endogenous PDGF-BB is likely to have its effects locally and then in relatively high concentrations, as discussed above. It can be anticipated that during an acute systemic anaphylaxis that is a lethal condition the V3-directed control of PIF stimulated by endogenous PDGF BB is insufficient to counteract edema formation. Rather, the presently described edema-counteracting system involving PDGF-BB and V3 is likely to be relevant during later phases of anaphylaxis and acute inflammation. In line with this is previously reported data showing that during wound healing, as well as in carcinoma and inflammatory lesions, PDGF-AB/BB is upregulated and also detected in infiltrating macrophages and vascular cells.31,32,35
The injection into a tissue, with or without adding a substance or protein normally present or not, will inflict a trauma to the tissue. In the present study, as well as previous studies on the control of PIF, we have made several attempts to control for the injection trauma as such, including vascular reactions with mobilization of leukocytes, as well as to validate the effect of the specificity of the injected protein. Although there is an effect of the injection trauma, it has never been observed that injection of saline or unspecific proteins including nonspecific IgG elicits a lowering of PIF.2 In contrast, IgG, cytokines, and other agents that modulate collagen gel contraction in vitro also has an in vivo action on PIF.3eC5,7,8 Because we have demonstrated parallel effects of a number of substances in in vivo experiments and in fibroblast-mediated collagen gel contraction in vitro, we have used the latter system as a model for PIF control in vivo. We show here that mouse embryonic fibroblasts lacking V integrins, exclusively depended on 1 integrins to contract collagen gels even when the cells were stimulated by PDGF-BB. Wild-type fibroblasts, in contrast, contracted collagen gels when 1 integrins were blocked, and this contraction was stimulated by PDGF-BB. These findings show that V3 is necessary for collagen-gel contraction when 1 integrins are blocked and that this process is stimulated by PDGF-BB. Several reports have shown that under certain conditions the V3 integrin is able to mediate collagen gel contraction in vitro.16eC19 Many inflammatory agents, including PGE15 and IL-1,6 induce a reduction in PIF in normal dermis and inhibit cell-mediated collagen gel contraction. The inhibitory effects on collagen gel contraction by both these agents can at least partly be overcome by stimulation with PDGF-BB.5,15 Further studies will be needed to investigate the hypothesis that these proinflammatory agents block 1 integrins but allow PDGF-BB stimulated V3-mediated collagen-gel contraction in vitro and control of PIF in vivo.
The ligand for V3 that is involved in PIF control in vivo is not defined and candidates include the collagen type I fibers and microfibrils.36 Available data suggest that native triple helical interstitial collagens are not recognized by the RGD-directed integrin V3.37 The nature of the PDGF-BBeCstimulated celleCcollagen interaction mediated by V3 that is needed for collagen gel contraction in vitro and potentially in vivo is therefore not apparent. Further studies are needed to characterize the in vivo ECM ligand that is recognized by V3 and functions to restrain dermal tissue swelling. Regardless of the ECM ligand that is recognized by V3 in connective tissues and/or in collagen gels in vitro, the fact that stimulation with PDGF-BB was needed for V3-mediated contraction in vitro and in vivo suggests a requirement for activation of the integrin through inside-out signaling.29 Such activation involves affinity modulation by conformational changes of the integrin.38,39 PDGF-BBeCregulated activation of V3 may thus be a suitable target for the development of pharmaceutical agents for the treatment of uncontrolled edema, eg, in septic shock.
In conclusion, the results from this study support the hypothesis that PDGF-BB counteracts a tendency to form edema by stimulating the activity of V3-integrins. Furthermore, our data suggest that under normal conditions, tension between connective tissue cells and the dermal fibers is maintained by 1-integrineCmediated contraction. Proinflammatory mediators are likely to block the 1 integrins, causing a reduction in PIF and edema formation, a process that is opposed by PDGF-BBeCdirected V3-mediated contraction.
Acknowledgments
This study was supported by grants from the Swedish Cancer Foundation (to K.R.), the Gustaf V:s 80-rsfond (to K.R.), and the Norwegian Research Council (to R.K.R.). Ann-Marie Gustafson and Gerd Salvesen are gratefully acknowledged for technical assistance. We thank Dr Richard O. Hynes for helpful comments and for the 3-negative mice.
References
Aukland K, Reed RK. Interstitial-lymphatic mechanisms in the control of extracellular fluid volume. Physiol Rev. 1993; 73: 1eC78.
Wiig H, Rubin K, Reed RK. New and active role of the interstitium in control of interstitial fluid pressure: potential therapeutic consequences. Acta Anaesthesiol Scand. 2003; 47: 111eC121.
Reed RK, Rubin K, Wiig H, Rodt S. Blockade of 1-integrins in skin causes edema through lowering of interstitial fluid pressure. Circ Res. 1992; 71: 978eC983.
hlen K, Berg A, Stiger F, Tengholm A, Siegbahn A, Gylfe E, Reed RK, Rubin K. Cell interactions with collagen matrices in vivo and in vitro depend on phosphatidylinositol 3-kinase and free cytoplasmic calcium. Cell Adhes Commun. 1998; 5: 461eC473.
Berg A, Ekwall AK, Rubin K, Stjernschantz J, Reed RK. Effect of PGE1, PGI2, and PGF2 analogs on collagen gel compaction in vitro and interstitial pressure in vivo. Am J Physiol. 1998; 274: H663eCH671.
Nedreb T, Berg A, Reed RK. Effect of tumor necrosis factor-, IL-1, and IL-6 on interstitial fluid pressure in rat skin. Am J Physiol. 1999; 277: H1857eCH1862.
Rodt S, hlen K, Berg A, Rubin K, Reed RK. A novel physiological function for platelet-derived growth factor-BB in rat dermis. J Physiol (Lond). 1996; 495: 193eC200.
Heuchel R, Berg A, Tallquist M, hlen K, Reed RK, Rubin K, Claesson-Welsh L, Heldin C-H, Soriano P. Platelet-derived growth factor receptor regulates interstitial fluid homeostasis through phosphatidylinositol-3' kinase signaling. Proc Natl Acad Sci U S A. 1999; 96: 11410eC11415.
Rubin K, Gullberg D, Tomasini-Johansson B, Reed RK, Rydeen C, Borg TK. Molecular recognition of the extracellular matrix by cell surface receptors. In: Comper WD, ed. Extracellular Matrix, Volume 2: Molecular Components and Interactions. Reading, UK: Harwood Academic Publishers; 1996: 262eC309.
Reed RK, Berg A, Gjerde EA, Rubin K. Control of interstitial fluid pressure: role of 1-integrins. Semin Nephrol. 2001; 21: 222eC230.
Meyer FA, Laver-Rudich Z, Tanenbaum R. Evidence for a mechanical coupling of glycoprotein microfibrils with collagen fibrils in Wharton’s jelly. Biochim Biophys Acta. 1983; 755: 376eC387.
Grinnell F. Fibroblast biology in three-dimensional collagen matrices. Trends Cell Biol. 2003; 13: 264eC269.
Clark RA, Folkvord JM, Hart CE, Murray MJ, McPherson JM. Platelet isoforms of platelet-derived growth factor stimulate fibroblasts to contract collagen matrices. J Clin Invest. 1989; 84: 1036eC1040.
Gullberg D, Tingstrm A, Thuresson AC, Olsson L, Terracio L, Borg TK, Rubin K. 1 integrin-mediated collagen gel contraction is stimulated by PDGF. Exp Cell Res. 1990; 186: 264eC272.
Tingstrm A, Heldin C-H, Rubin K. Regulation of fibroblast-mediated collagen gel contraction by platelet-derived growth factor, interleukin-1 and transforming growth factor-1. J Cell Sci. 1992; 102: 315eC322.
Cooke ME, Sakai T, Mosher DF. Contraction of collagen matrices mediated by 21A and V3 integrins. J Cell Sci. 2000; 113: 2375eC2383.
Grundstrm G, Mosher DF, Sakai T, Rubin K. Integrin V3 mediates platelet-derived growth factor-BB-stimulated collagen gel contraction in cells expressing signaling deficient integrin 21. Exp Cell Res. 2003; 291: 463eC473.
Levinson H, Hopper JE, Ehrlich HP. Overexpression of integrin V promotes human osteosarcoma cell populated collagen lattice contraction and cell migration. J Cell Physiol. 2002; 193: 219eC224.
Agrez MV, Bates RC, Boyd AW, Burns GF. Arg-Gly-Asp-containing peptides expose novel collagen receptors on fibroblasts: implications for wound healing. Cell Regul. 1991; 2: 1035eC1044.
Hodivala-Dilke KM, McHugh KP, Tsakiris DA, Rayburn H, Crowley D, Ullman-Cullere M, Ross FP, Coller BS, Teitelbaum S, Hynes RO. 3-integrin-deficient mice are a model for Glanzmann thrombasthenia showing placental defects and reduced survival. J Clin Invest. 1999; 103: 229eC238.
McHugh KP, Hodivala-Dilke K, Zheng MH, Namba N, Lam J, Novack D, Feng X, Ross FP, Hynes RO, Teitelbaum SL. Mice lacking 3 integrins are osteosclerotic because of dysfunctional osteoclasts. J Clin Invest. 2000; 105: 433eC440.
Todaro GJ, Green H. Quantitative studies of the growth of mouse embryo cells in culture and their development into established lines. J Cell Biol. 1963; 17: 299eC313.
Bader BL, Rayburn H, Crowley D, Hynes RO. Extensive vasculogenesis, angiogenesis, and organogenesis precede lethality in mice lacking all V integrins. Cell. 1998; 95: 507eC519.
Gullberg D, Terracio L, Borg TK, Rubin K. Identification of integrin-like matrix receptors with affinity for interstitial collagens. J Biol Chem. 1989; 264: 12686eC12694.
Wiederhielm CA, Woodbury JW, Kirk S, Rushmer RF. Pulsatile pressures in the microcirculation of frog’s mesentery. Am J Physiol. 1964; 207: 173eC176.
Wiig H, Reed RK, Aukland K. Micropuncture measurement of interstitial fluid pressure in rat subcutis and skeletal muscle: comparison to wick-in-needle technique. Microvasc Res. 1981; 21: 308eC319.
Nisato RE, Tille JC, Jonczyk A, Goodman SL, Pepper MS. V3 and V5 integrin antagonists inhibit angiogenesis in vitro. Angiogenesis. 2003; 6: 105eC119.
Srivatsa SS, Fitzpatrick LA, Tsao PW, Reilly TM, Holmes DR Jr, Schwartz RS, Mousa SA. Selective V3 integrin blockade potently limits neointimal hyperplasia and lumen stenosis following deep coronary arterial stent injury: evidence for the functional importance of integrin V3 and osteopontin expression during neointima formation. Cardiovasc Res. 1997; 36: 408eC428.
Hynes RO. Integrins: bidirectional, allosteric signaling machines. Cell. 2002; 110: 673eC687.
Rubin K, Tingstrm A, Hansson GK, Larsson E, Rnnstrand L, Klareskog L, Claesson-Welsh L, Heldin C-H, Fellstrm B, Terracio L. Induction of B-type receptors for platelet-derived growth factor in vascular inflammation: possible implications for development of vascular proliferative lesions. Lancet. 1988; 1: 1353eC1356.
Reuterdahl C, Sundberg C, Rubin K, Funa K, Gerdin B. Tissue localization of -receptors for platelet-derived growth factor and platelet-derived growth factor B chain during wound repair in humans. J Clin Invest. 1993; 91: 2065eC2075.
Sundberg C, Ljungstrm M, Lindmark G, Gerdin B, Rubin K. Microvascular pericytes express platelet-derived growth factor- receptors in human healing wounds and colorectal adenocarcinoma. Am J Pathol. 1993; 143: 1377eC1388.
Rajkumar VS, Sundberg C, Abraham DJ, Rubin K, Black CM. Activation of microvascular pericytes in autoimmune Raynaud’s phenomenon and systemic sclerosis. Arthritis Rheum. 1999; 42: 930eC941.
Wilder RL. Integrin V3 as a target for treatment of rheumatoid arthritis and related rheumatic diseases. Ann Rheum Dis. 2002; 61 (suppl 2): ii96eCii99.
Heldin CH, Westermark B. Mechanism of action and in vivo role of platelet-derived growth factor. Physiol Rev. 1999; 79: 1283eC1316.
Kielty CM, Shuttleworth CA. Microfibrillar elements of the dermal matrix. Microsc Res Tech. 1997; 38: 413eC427.
Davis GE. Affinity of integrins for damaged extracellular matrix: V3 binds to denatured collagen type I through RGD sites. Biochem Biophys Res Commun. 1992; 182: 1025eC1031.
Takagi J, Petre BM, Walz T, Springer TA. Global conformational rearrangements in integrin extracellular domains in outside-in and inside-out signaling. Cell. 2002; 110: 599eC511.
Springer TA, Wang JH. The three-dimensional structure of integrins and their ligands, and conformational regulation of cell adhesion. Adv Protein Chem. 2004; 68: 29eC63.(sa Lideen, Ansgar Berg, T)