Vascular Endothelial Growth Factor-A–Induced Chemotaxis of Monocytes Is Attenuated in Patients With Diabetes Mellitus
A Potential Predictor for the Individual Capacity to Develop Collaterals
Background—Vascular endothelial growth factor-A (VEGF-A) acts on endothelial cells and monocytes, 2 cell types that participate in the angiogenic and arteriogenic process in vivo. Thus far, it has not been possible to identify differences in individual responses to VEGF-A stimulation because of the lack of an ex vivo assay.
Methods and Results—We report a chemotaxis assay using isolated monocytes from individual diabetic patients and from healthy, age-matched volunteers. The chemotactic response of individual monocyte preparations to VEGF-A, as mediated via Flt-1, was quantitatively assessed using a modified Boyden chamber. Although the migration of monocytes from healthy volunteers could be stimulated with VEGF-A (1 ng/mL) to a median of 148.4% of the control value (25th and 75th percentiles, 136% and 170%), monocytes from diabetic patients could not be stimulated with VEGF-A (median, 91.1% of unstimulated controls; 25th and 75th percentiles, 83% and 98%; P<0.0001). In contrast, the response of monocytes to the chemoattractant formylMetLeuPhe remained intact in diabetic patients. The VEGF-A–inducible kinase activity of Flt-1, as assessed by in vitro kinase assays, remained intact in monocytes from diabetic patients. Moreover, the serum level of VEGF-A, as assessed by immunoradiometric assay, was significantly elevated in diabetic patients.
Conclusions—The cellular response of monocytes to VEGF-A is attenuated in diabetic patients because of a downstream signal transduction defect. These data suggest that monocytes are important in arteriogenesis and that their ability to migrate might be critical to the arteriogenic response. Thus, we resolved a fundamental mechanism involved in the problem of impaired collateral formation in diabetic patients.
- collateral circulation
- diabetes mellitus
- signal transduction
- cell movement
- endothelial growth factors
Vascular endothelial growth factor-A (VEGF-A) and its receptors are crucially involved in the process of angiogenesis, as several gene knockout experiments have shown.1 2 3 4 In addition, VEGF-A can induce an angiogenic response and enhance collateral blood flow when administered to an area of regional myocardial or peripheral ischemia.5 6 7 8 9 Therefore, it is important to investigate the cellular responses to VEGF-A stimulation under the pathological conditions that might affect or limit its function and use. Such conditions include diabetes mellitus, a prominent cardiovascular risk factor.10
VEGF-A was initially described as an endothelial cell-specific growth factor that could stimulate endothelial proliferation and migration in vitro11 and angiogenesis and endothelial cell regrowth in vivo.12 VEGF-A recently gained enormous attention in the medical community for the following 2 reasons. (1) It is crucially involved in the pathogenesis of a number of different angiogenic diseases, including diabetic retinopathy, psoriasis, rheumatoid arthritis, and the growth of solid tumors.13 (2) Local or regional VEGF-A application enhances blood flow to areas of regional ischemia, thereby stimulating tissue perfusion. This concept is known as “therapeutic angiogenesis”6 and was initially introduced by Höckel and Burke14 in a noncardiac context. In fact, several clinical trials are currently underway to prove the feasibility and efficiency of improving myocardial perfusion by using VEGF-A.15 16
The concept of therapeutic angiogenesis has been studied in detail in the context of regional ischemia in the heart, in the peripheral circulation, and in the brain. Most recently, the concept of arteriogenesis, ie, the growth of preexisting collaterals, is evolving; this requires discrimination from true angiogenesis.17 In light of this discrimination, the term therapeutic angiogenesis, although extensively used in the past, should no longer be used to describe the stimulation of true collateral growth; the term “therapeutic arteriogenesis” seems to be more appropriate.
The cellular effects of VEGF-A are mediated via 2 distinct receptor tyrosine-kinases11 called Flt-1 (Fms-like tyrosine kinase [VEGFR1]) and KDR (kinase-insert domain-containing receptor [VEGFR2]). In previous studies on the function of KDR, we found in vitro evidence for the regulation of receptor activity under pathological conditions such as hypoxia.
Besides endothelial cells, monocytes specifically respond to VEGF-A. One of the 2 VEGF-receptors (Flt-1) is present on the surface of monocytes and mediates the chemotactic response to VEGF-A and tissue factor induction.18 What makes monocytes especially attractive is the fact that they represent the only cell type in the body that carries receptors for VEGF and, at the same time, can be obtained from individual patients for functional analysis of the VEGF receptor system. Moreover, monocytes play an important role in the angiogenic process and during collateral growth/arteriogenesis.19 20 So far, there was no experimental approach to judge the individual response to VEGF-A stimulation in a defined patient. Given the possibility of obtaining such data, this would be an extremely important piece of information in the context of stimulating collateral formation.
On the basis of these ideas, we developed and established an assay in which peripheral blood monocytes can be isolated from individuals and tested for their chemotactic response to VEGF-A in a modified Boyden chamber. We found that the specific and strong VEGF-A–induced response seen in healthy individuals is completely attenuated in patients with diabetes mellitus. We conclude that monocytes can be used to determine VEGF receptor–mediated cellular function in healthy and diseased individuals. Given the crucial role of monocytes in the development of functional collaterals, the impaired chemotactic response of monocytes to VEGF-A in diabetic patients seems to predict a reduced ability to grow collaterals. The analysis of the VEGF-A–induced migration of monocytes represents the first attempt to study the function of the VEGF system in healthy and diseased individuals.
Characterization of Patients and Healthy Volunteers
Patients with diabetes mellitus (n=16; 6 men and 10 women) and a mean age of 68.3±10.4 years were studied; this group included both patients with insulin-dependent (n=10) and non–insulin-dependent diabetes mellitus (n=6). Patients with underlying inflammatory or malignant disease and those who smoked were not included in this study. The glycosylated fraction of the major component of adult hemoglobin (HbA1c) was elevated in all patients; this elevation ranged from 7.0% to 11.2%, with a median of 8.5% (25th and 75th percentiles, 7.9% and 9.3%, respectively). Healthy volunteers (n=14; 5 men and 9 women) were included in our study as a control group. Their mean age was 56.4±4.0 years. Informed consent was obtained from patients and healthy volunteers according to the requirements of the local ethical committee.
Isolation of Monocytes from Peripheral Venous Blood
Monocytes were isolated from 60 mL of heparinized venous blood samples using a slightly modified version of the method of Denholm and Wolber.21 In brief, density centrifugation was performed using the Ficoll separation solution with a density of 1.077 g/mL (Biochrom) to isolate mononuclear cells. In a second round of centrifugation, monocytes were enriched using Percoll separation solution with a density of 1.139 g/mL (Sigma) before washing and resuspending the cells in DMEM (Biochrom). The purity of the extracted monocytes was up to 93%, as determined by analysis with a fluorescence-activated cell sorter using an antibody recognizing CD14 (M14-FITC, Coulter Electronics). The vitality of the isolated monocytes was assessed by trypan blue exclusion; it was always >90%.
Monocyte chemotaxis was quantitated using a modified 48-well Boyden chamber (Nuclepore) and polycarbonate membranes with a pore diameter of 5 μm (Nuclepore). Monocytes were seeded in a concentration of 5×105 cells/mL in DMEM and allowed to migrate for a total of 3 hours in the humidified incubator (37°C; 5% CO2). Adherent cells on the filter membrane were fixed in 99% ethanol for 10 minutes and stained using Giemsa dye before scraping off cells at the upper side of the filter membrane. For a quantitative assessment of migrated cells, a total of 15 high power fields from 3 different wells (5 each) were counted. Cell migration was stimulated with either VEGF-A165 (0.1 to 10 ng/mL; this was kindly provided by Denis Gospodarowicz, Chiron, Emeryville, Calif) or formylMetLeuPhe (fMLP, 10-8 mol/L; Sigma).
Immunoprecipitation and In Vitro Kinase Assay
Isolated monocytes were preincubated for 5 minutes with 100 μmol/L Na3VO4 to inhibit phosphatase activity. Cells were stimulated for 3 minutes at 37°C with 50 ng/mL VEGF. After washing with ice-cold PBS containing 100 μmol/L Na3VO4, cells were solubilized in a lysis buffer (150 mmol/L NaCl, 20 mmol/L Tris-HCl [pH 7.4], 1% CHAPS [Sigma], 10 mmol/L EDTA, 10% glycerol, 100 μmol/L Na3VO4, 1% Trasylol [Bayer], and 1 mmol/L PMSF). The cell lysates were centrifuged at 10 000g for 15 minutes, and phosphotyrosine-specific immunoprecipitation was performed using the 4G10 monoclonal antibody (UBI) and a rabbit anti-mouse antiserum (Sigma). Immunoprecipitates immobilized on Protein A-Sepharose CL 4B (Pharmacia) were used for the immune complex kinase assay, which was performed for 7 minutes at room temperature in 25 μL of 50 mmol/L HEPES buffer (pH 7.4) containing 10 mmol/L MnCl2, 1 mmol/L dithiothreitol, and 5 μCi of [γ-32P]ATP (Amersham). The samples were separated by SDS-PAGE (5 to 15% gradient) before the gels were incubated for 30 minutes in 2.5% glutaraldehyde, washed 2 times for 15 minutes in 10% acetic acid and 40% methanol, treated for 1 hour at 55°C in 1 mol/L KOH to remove serine-bound phosphate,22 washed 3 times for 20 minutes in 10% acetic acid/40% methanol, dried, and exposed to Hyperfilm MP (Amersham). Radioactive bands were quantitated on a Fuji Phosphorimager.
The VEGF-A concentration was analyzed in serum samples of all diabetic and nondiabetic subjects. Samples were stored at −20°C until analysis. An immunoradiometric assay was performed with 2 monoclonal antibodies specific for VEGF-A, which were generously supplied by Genentech Inc (South San Francisco, Calif). We used the monoclonal antibody B2.6.2 to recognize VEGF-A165 and VEGF-A189 and the monoclonal antibody A4.6.1 to recognize VEGF-A121, VEGF-A165, and VEGF-A189.23 The 96-well plates (Maxisorp, Nunc) were coated with B2.6.2 (5 μg/mL) in 50 mmol/L carbonate buffer (pH 9.6) for 16 hours at 4°C. After washing with 0.03% Tween 80 in PBS (pH 7.4), the plates were blocked for 1 hour at 25°C using PBS (pH 7.4) with 0.5% bovine serum albumin and 0.03% Tween 80. Plates were washed before the addition of serum samples or the VEGF-A165 control (range, 5 pg/mL to 11 ng/mL) and incubated for 2 hours at 25°C; all experiments were performed in triplicate. After a washing step, the monoclonal anti-VEGF antibody A4.6.1, [125I]-labeled using the Chloramin-T method,24 was added to each well (5×104 cpm/well) and incubated for 2 hours at 25°C. Supernatants were then discarded, the plates were washed, and the wells were counted using an automated γ-counter (LKB Wallac 1277 Gammamaster, LKB-Pharmacia). The sensitivity of the assay was 20 pg/mL. No cross-reactivity was found with the closely related platelet-derived growth factor-BB.
Results of the migration assays and the VEGF-A serum levels were analyzed using a 2-sided exact Wilcoxon test for unpaired samples. Data for each group of patients were described as medians and quartiles (25th and 75th percentiles). In the case of the migration assay, testing was primarily performed for a VEGF-A concentration of 1 ng/mL.
VEGF-A–induced monocyte migration was assessed using a modified Boyden chamber assay. The migration of monocytes from healthy volunteers (n=14; age, 56.4±4.0 years) could be significantly stimulated with VEGF-A to 148.4% (25th and 75th percentiles, 136% and 170%) compared with the 100% baseline control value. This value was reached at 1.0 ng/mL VEGF-A (Figure 1A⇓). For a VEGF-A concentration of 0.3 ng/mL, the maximal stimulatory effect was slightly higher, with a median of 153.0% above the unstimulated baseline, but no statistically significant difference was found between this and the value obtained for a VEGF-A concentration of 1.0 ng/mL. Monocytes from diabetic patients (n=16; age, 68.3±10.4 years; HbA1c, 7.0% to 11.2%) could not be stimulated with VEGF-A in this assay. This was true for all VEGF-A concentrations tested (0.1 to 10 ng/mL). Compared with the unstimulated 100% control, VEGF-A–stimulated (1.0 ng/mL) monocyte migration measured 91.1% (25th and 75th percentiles, 83% and 98%). When monocytes were stimulated with VEGF-A at 0.1 ng/mL, this value measured 91.1% (25th and 75th percentiles, 87% and 99%); it measured 91.8% (25th and 75th percentiles, 82% and 100%) at 0.3 ng/mL VEGF-A and 87% (25th and 75th percentiles, 76% and 97%) at 10 ng/mL VEGF-A. All these values were significantly below the results obtained for healthy individuals (P<0.0001). No differences existed between monocytes from patients with insulin-dependent and non–insulin-dependent diabetes mellitus.
To clarify whether the migration of monocytes across the porous filter membranes depended on the presence of a VEGF-A gradient between the lower and the upper compartment, we performed a checkerboard analysis. As shown in the Table⇓, the maximal induction of migration occurred in the presence of a positive concentration gradient between the 2 compartments. In the presence of equal concentrations of VEGF-A, no enhanced migratory response could be observed. These results indicate that VEGF-A can activate a true chemotactic response in monocytes with no appreciable chemokinetic activity.
In contrast to the VEGF-A–induced effects, the VEGF-independent chemotactic response of monocytes to the tripeptide fMLP (10−8 mol/L) remained intact in diabetic patients at 235% (25th and 75th percentiles, 158% and 357%); the control group had a response of 304% (25th and 75th percentiles, 262% and 420%) (Figure 1B⇑). No statistically significant difference existed between the 2 groups.
The VEGF-A–inducible kinase activity of Flt-1 remained fully intact in monocytes from diabetic patients, as assessed by the in vitro kinase assay (Figure 2⇓). VEGF-A could induce similar levels of tyrosine phosphorylation in all monocyte preparations analyzed. No differences existed in monocytes isolated from healthy control subjects.
The VEGF-A serum level was measured in blood samples from all individuals tested. The median VEGF-A serum level in healthy individuals was 98 pg/mL (25th and 75th percentiles, 75 and 137 pg/mL). In contrast, the VEGF-A serum levels of diabetic patients were significantly elevated (median, 153 pg/mL; 25th and 75th percentiles, 106 and 230 pg/mL; P=0.0088) (Figure 3⇓). VEGF-A serum levels of female diabetics were higher than those of male diabetics (230 versus 149 pg/mL; P=NS), just as VEGF-A serum levels in healthy female individuals were higher than those in healthy male individuals (116.5 versus 88 pg/mL; P=NS).
In the present study, we showed for the first time that the cellular response to the angiogenic factor VEGF-A strongly depends on the integrity of the studied cells. Although the Flt-1–mediated chemotactic response to VEGF-A was strong and consistent in all monocytes isolated from healthy volunteers, monocytes from patients with diabetes mellitus did not properly respond, and they lacked the ability to migrate toward VEGF-A in a chemotaxis assay. Moreover, our data on the VEGF-A–induced and Flt-1–mediated signal transduction defect in diabetic patients provide novel insight into the mechanism of how monocyte function is impaired in diabetic patients.
The recruitment of monocytes is thought be an important step during collateral formation secondary to regional myocardial ischemia.20 25 Increased shear stress in preformed epicardial collaterals somewhat distant to the actual area of hypoxia and ischemia paves the way for monocyte accumulation, their maturation to macrophages, and the release of growth factors, which creates an inflammatory environment. In contrast to true angiogenesis, the process of arteriogenesis, which describes the growth of collateral vessels,17 seems to be critically dependent on monocytes and on proper monocyte function.20 25 Therefore, our ex vivo assay using the VEGF-A stimulation of monocytes may have a predictive value for a patient’s ability to develop collateral circulation to an ischemic area in the heart or in the peripheral circulation. In fact, during the initial review process of this article, a study was published in Circulation demonstrating that the ability of diabetic patients with coronary artery disease to develop coronary collaterals is significantly impaired.26 These data are compatible with our hypothesis of growth factor–induced monocyte migration as a predictor of an individual’s ability to develop collateral circulation. Therefore, our novel ex vivo assay is a good candidate for a surrogate assay of the process of arteriogenesis. Likewise, our data may serve as a molecular explanation for the reduced collateralization seen in diabetic patients.
Because VEGF-dependent monocyte function is severely reduced in monocytes from diabetic patients, our data suggest that VEGF-A and its receptor Flt-1 might indeed be critically involved in stimulating the process of arteriogenesis. VEGF-A could stimulate arteriogenesis in the following 2 different and independent ways. (1) Direct VEGF-A action stimulates the endothelium and promotes vascular remodeling. (2) VEGF-A promotes an indirect mode of activation by stimulating monocyte recruitment to the vessel wall. These monocytes and developing macrophages are vehicles for a number of vascular growth factors that are produced by these cells and released at the site of activation, such as vascular growth factors (including VEGF-A),27 28 basic fibroblast growth factor, transforming growth factor-β, and epidermal growth factor.29
Our model, however, does not exclude the functional involvement of other growth factors and cytokines in the process of arteriogenesis. For example, monocytes could be recruited to the vessel wall by monocyte chemoattractant protein-1.19 On the basis of recent data, this protein could also act as a molecular mediator (it can be induced by VEGF-A).30 Because of the limited number of monocytes obtained from each preparation and the need for triplicate analysis and the inclusion of proper controls, we have not been able to test the response of monocytes from diabetic individuals toward other factors. However, this will be the subject of a future study.
In diabetic patients, the decreased chemotactic response of monocytes to VEGF-A is a consequence of impaired VEGF-A–induced and Flt-1–mediated signal transduction. Although the activation of tyrosine phosphorylation seems to remain fully intact, the signal does not reach the (intact) cytoskeletal components responsible for migration. Strong evidence indicates that the impaired VEGF-A–induced and Flt-1–mediated effect is selective and that the investigated monocytes are basically intact; for example, the potent and unspecific tripeptide fMLP can stimulate a proper chemotactic response in these cells. Therefore, our findings suggest a signal transduction defect is responsible for the impaired monocyte migration.
It is presently unclear whether the impaired VEGF-A–induced response of monocytes reflects or predicts an impaired endothelial response to VEGF-A. It may well be that the Flt-1–mediated response of VEGF-A in the endothelial cells of diabetic individuals is impaired as well. However, because most of the VEGF-A–induced responses in the endothelium are mediated by KDR11 and because KDR is not expressed in monocytes,18 our finding of impaired monocyte migration in diabetic individuals does not necessarily predict an impaired endothelial response to VEGF-A. There are many possible explanations for why arteriogenesis might be impaired while angiogenesis, in particular diabetic retinopathy, is stimulated in these patients. Although Flt-1–mediated responses are impaired in patients with diabetes mellitus, which leads to reduced monocyte migration and impaired arteriogenesis (as shown in this article), the angiogenic response of endothelial cells may be enhanced secondary to elevated VEGF-A levels. Another possible explanation might be a different degree of involvement of monocytes in arteriogenesis and angiogenesis. It is tempting to speculate that monocyte migration is the rate-limiting step in arteriogenesis, whereas the involvement of monocytes at sites of angiogenesis reflects the inflammatory nature of this process.
In the present study, we showed that the serum level of VEGF-A is significantly elevated in diabetic patients; this is similar to discoveries in the ocular fluid of patients with proliferative diabetic retinopathy.31 VEGF-A levels are raised under diabetic circumstances as a direct consequence of elevated glucose concentrations.32 In addition, the fact that angiogenesis is promoted in the form of diabetic retinopathy raises questions about differences in the pathogenesis of diabetic retinopathy and other forms of angiogenesis. Proliferative diabetic retinopathy is preceded by a long period of microvascular damage. After decades of chronic changes, microvascular occlusion eventually results in ischemia, which leads to the secretion of VEGF-A from the retina and to the development of abnormal angiogenesis within the isolated compartment of the eye. It is conceivable that elevated VEGF-A levels in diabetic individuals acting on KDR may compensate for the impaired activity of any Flt-1–mediated cellular response.
Taken together, these data indicate that the cellular response of monocytes to VEGF-A is attenuated in diabetic patients due to a downstream signal transduction defect. Therefore, we postulate that the VEGF-A–induced and monocyte-dependent process of collateral formation is severely impaired in diabetic patients and that VEGF-A–based therapeutic strategies to enhance tissue perfusion should give better results in patients not suffering from diabetes mellitus.
This study was supported in part by grants Wa734/2-1 and Wa734/2-4 from the Deutsche Forschungsgemeinschaft and by the Sonderforschungsbereich SFB451, project B1 (all to J.W.). We would like to acknowledge the kind gift of recombinant VEGF-A from Denis Gospodarowicz, Chiron, Emeryville, Calif. We thank Hedwig Frank for performing the checkerboard analysis.
- Received October 29, 1999.
- Revision received February 2, 2000.
- Accepted February 9, 2000.
- Copyright © 2000 by American Heart Association
Banai S, Jaklitsch MT, Shou M, et al. Angiogenic-induced enhancement of collateral blood flow to ischemic myocardium by vascular endothelial growth factor in dogs. Circulation. 1994;89:2183–2189.
Waltenberger J. Modulation of growth factor action: implications for the treatment of cardiovascular diseases. Circulation. 1997;96:4083–4095.
Baumgartner I, Pieczek A, Manor O, et al. Constitutive expression of phVEGF165 after intramuscular gene transfer promotes collateral vessel development in patients with critical limb ischemia. Circulation. 1998;97:1114–1123.
Folkman J. Angiogenic therapy of the human heart. Circulation. 1998;97:628–629.
Waltenberger J. Therapeutic angiogenesis in the heart using peptide growth factors: Angiogenesis research entering clinical trials. Angiogenesis. 1998;2:115–117.
Waltenberger J, Claesson-Welsh L, Siegbahn A, et al. Different signal transduction properties of KDR and Flt1, two receptors for vascular endothelial growth factor. J Biol Chem. 1994;269:26988–26995.
Asahara T, Bauters C, Pastore C, et al. Local delivery of vascular endothelial growth factor accelerates reendothelialization and attenuates intimal hyperplasia in balloon-injured rat carotid artery. Circulation. 1995;91:2793–2801.
Rosengart TK, Lee LY, Patel SR, et al. Angiogenesis gene therapy: phase I assessment of direct intramyocardial administration of an adenovirus vector expressing VEGF121 cDNA to individuals with clinically significant severe coronary artery disease. Circulation. 1999;100:468–474.
Losordo DW, Vale PR, Symes JF, et al. Gene therapy for myocardial angiogenesis: initial clinical results with direct myocardial injection of phVEGF(165) as sole therapy for myocardial ischemia. Circulation. 1998;98:2800–2804.
Schaper W, Buschmann I. Collateral circulation and diabetes. Circulation. 1999;99:2224–2226. Editorial.
Clauss M, Weich H, Breier G, et al. The vascular endothelial growth factor receptor FLT-1 mediates biological activities: implications for a functional role of placenta growth factor in monocyte activation and chemotaxis. J Biol Chem. 1996;271:17629–17634.
Ito WD, Arras M, Winkler B, et al. Monocyte chemotactic protein-1 increases collateral and peripheral conductance after femoral artery occlusion. Circ Res. 1997;80:829–837.
Cooper JA, Hunter T. Changes in protein phosphorylation in rous sarcoma virus-transformed chicken embryo cells. Mol Cell Biol. 1981;1:165–178.
Schaper W. Coronary collateral development: concepts and hypotheses. In: Schaper W, Schaper J, eds. Collateral Circulation: Heart, Brain, Kidney, Limbs. Boston: Kluwer Academic Publishers; 1993:41–64.
Abaci A, Oguzhan A, Kahraman S, et al. Effect of diabetes mellitus on formation of coronary collateral vessels. Circulation. 1999;99:2239–2242.
Berse B, Brown LF, van de Water L, et al. Vascular permeability factor (vascular endothelial growth factor) gene is expressed differentially in normal tissues, macrophages, and tumors. Mol Biol Cell. 1992;3:211–220.
Marumo T, Schini-Kerth VB, Busse R. Vascular endothelial growth factor activates nuclear factor-κB and induces monocyte chemoattractant protein-1 in bovine retinal endothelial cells. Diabetes. 1999;48:1131–1137.
Natarajan R, Bei W, Lanting L, et al. Effects of high glucose on vascular endothelial growth factor expression in vascular smooth muscle cells. Am J Physiol. 1997;273:H2224–H2231.