Vascular endothelial growth factor (VEGF), also known as vascular permeability factor (VPF), is a basic, heparin-binding, homodimeric glycoprotein of 45,000 daltons that is specifically mitogenic for endothelial cells and that induces vascular permeability (1-4) (5-7) 165 is the predominant isoform secreted by a variety of normal and transformed cells (5-7) (8,9) (10) (11-13) (14) (15)
In vitro studies showed that VEGF induces endothelium-dependent relaxation in isolated canine coronary arteries (16)
MATERIALS AND METHODS
Male Sprague-Dawley (SD) rats (Charles River Breeding Laboratories) aged 8 weeks were acclimated to the facility for at least 1 week before operation, fed a pelleted rat chow and water ad libitum, and housed in a light- and temperature-controlled room. The experimental procedures, which were approved by Genentech's Institutional Animal Care and Use Committee, conform to the guiding principles of the American Physiological Society. The 165-amino acid homodimeric species of recombinant human VEGF was purified from transfected Chinese hamster ovary cells as previously described (17)
Measurements of arterial pressure and HR
Rats were anesthetized by intraperitoneal injection of ketamine 80 mg/kg (Aveco, Fort Dodge, IA, U.S.A.) and xylazine 10 mg/kg (Rugby Laboratories, Rockville Center, NY, U.S.A.). A catheter (PE-10 fused with PE 50) filled with heparin/saline (50 U/ml) for measurement of mean arterial pressure (MAP) and HR was implanted in the abdominal aorta through the right femoral artery. A second catheter for VEGF administration was inserted in the right femoral vein. Catheters were exteriorized and fixed at the back of the neck. After catheter implantation, all rats were housed individually.
One day after catheterization, MAP and HR were measured in conscious, unrestrained rats with a model CP-10 pressure transducer (Century Technology, Inglewood, CA, U.S.A.) coupled to a Grass model 7 polygraph (Grass Instruments, Quincy, MA, U.S.A.). After a 45-min stabilization period, the rats received an intravenous (i.v.) 150- to 200-μl bolus of VEGF (1, 50, 250, or 500 μg/kg). MAP and HR were then monitored for 60 min. In a pilot study, intravenous injection of normal saline solution alone (200 μl) did not affect MAP, HR, or cardiac output.
Administration of N ω -nitro-L-arginine-methylester (L-NAME) Rats were anesthetized and catheterized as already described, and MAP and HR were measured as before. After stabilization, the rats were pretreated with L-NAME (Sigma Chemical, St. Louis, MO, U.S.A.), an inhibitor of NO formation, given as a 30-mg/kg bolus in normal saline solution, followed by intravenous infusion (4 mg/kg/min) for 8 min. Control animals received an equal volume of saline alone. A pilot study in conscious rats had shown that administration of L-NAME at this dose increased MAP maximally at 4-5 min and that MAP thereafter remained increased. Immediately after the infusion was discontinued, VEGF 50 μg/kg, which had been shown to be a submaximal dose for the dose-dependent depressor response, was injected intravenously. In addition, because L-NAME increased MAP, rats were pretreated with intravenous infusion of phenylephrine 7.5 μg/kg/min for 8 min, the dose that produced a MAP response similar to that produced by L-NAME, as an additional control.
Measurements of cardiac performance in conscious rats
After anesthesia and cannulation of the femoral artery and vein as already described, rats were intubated through a tracheotomy and ventilated with a respirator (Harvard Apparatus model 683, South Natick, MA, U.S.A.). Through a right-sided thoracotomy, the ascending aorta was exposed and gently separated from the pulmonary artery. An ultrasonic perivascular flowprobe (no. 2S165, Transonic Systems, Ithaca, NY, U.S.A.) was placed around the ascending aorta, and sterile K-Y jelly (Johnson & Johnson Medical, Arlington, TX, U.S.A.) was injected into the space between the vessel and the flowprobe. The flowprobe cable was exteriorized at the back of the neck, and the cable connector was sutured and fixed in place. The chest was closed, and the tracheal incision was sutured after extubation.
One day after operation, the arterial catheter was connected for measurement of MAP and HR in conscious rats as already described. For measurement of cardiac output, the flowprobe cable was connected to a model T 201 flowmeter (Transonic Systems). The mean blood flow curve of the ascending aorta was recorded on a chart recorder (Kipp & Zonen, Holland); cardiac output, as shown by the mean flow, was also digitally obtained by the flowmeter. Stroke volume was calculated as cardiac output divided by HR, and systemic vascular resistance was calculated as MAP divided by cardiac output. After hemodynamic stabilization, VEGF (250 μg/kg) was injected intravenously. This is a therapeutic dose as a single bolus for stimulation of angiogenesis in an animal model of limb ischemia (14)
To account for the possible influence of a left ventricular (LV) catheter on cardiac output, we measured the maximal first derivation of LV pressure (dP/dt ), an index of myocardial inotropic state, in a separate group of rats. Animals were anesthetized as before, and a catheter (PE 50) filled with heparin-saline was implanted in the left ventricle through the right carotid artery. Another catheter was implanted in the right femoral vein for VEGF administration as already described. Both catheters were exteriorized on the back of the neck. One day after catheterization, dP/dt was measured through the ventricular catheter with a pressure transducer connected to a model 7P20 differentiator (Grass Instruments). After intravenous injection of VEGF (250 μg/kg), LV dP/dt was measured. Although a fluid-filled catheter system may over-estimate dP/dt, we and other investigators have demonstrated that the values of dP/dt obtained are useful for making relative comparisons before and after treatment within or between groups of anesthetized and conscious rats (18-20)
In vitro studies in the isolated rat heart
Male SD rats weighing ≈300 g were anesthetized with 60 mg/kg intraperitoneally (i.p.) pentobarbital sodium (Fort Dodge Laboratories, Fort Dodge, IA, U.S.A.). After thoracotomy, hearts were removed and placed in 4 °C Krebs-Henseleit solution containing (in mM ): NaCl 118, KCl 4.7, CaCl2 2.5, MgSO4 1.6, NaHCO3 25, KH2 PO4 1.2, and glucose 11 and were gassed with 95% O2 /5% CO2 . After cannulation of the ascending aorta, the heart was perfused at constant pressure (100 mm Hg) with Krebs-Henseleit solution at 37 °C. A small incision was made in the left atrium, and a saline-filled balloon (Hugo Sachs Elektronik, March-Hugstetten, Germany), connected to a pressure transducer (Gould Electronics, Valley View, OH, U.S.A.) was placed in the left ventricle for the measurement of LV developed pressure (LVDP). The end-diastolic pressure was set at 0 mm Hg. From this pressure recording, HR and contractile force, measured as dP/dt (mm Hg/s), were electronically derived on a Grass polygraph.
After 30-min equilibration, the flow rate through each preparation was measured gravimetrically. A single 12-min infusion of VEGF was administered through a port adjacent to the aortic inflow, followed by a 12-min infusion of excipient. The dose was derived from in vivo experiments with a 250-μg/kg dose. Assuming a plasma volume of 40 ml/kg, we calculated the maximal possible concentration of VEGF in the circulation to be 6.25 μg/ml. The infusion of a 380-μg/ml stock solution into each isolated heart preparation was adjusted to give a final concentration of 6.25 μg/ml in the perfusate. The rate of the excipient infusion was identical in each case. These experiments were repeated in three separate preparations.
Hematocrit measurements
To determine whether VEGF induces a decrease in venous return as a result of vascular hyperpermeability, we examined the effect of VEGF on hematocrit in conscious rats. Catheters were placed in the femoral artery and vein of rats as already described. One day later, blood (0.3 ml) was collected in tubes containing EDTA through the arterial catheter as pretreatment controls. VEGF (250 μg/kg) or equal volume of vehicle was then injected intravenously. Blood was collected 10 min after injection (the time when maximal reduction in cardiac output or stroke volume was observed) and 4 h after injection (the time for recovery) as additional controls. Hematocrit was measured by centrifugation.
Statistical analysis
Results are mean ± SEM. Parameters before and after treatment in the same group were compared by a paired Student's t test, and parameters at the same timepoint between two groups were compared by an unpaired Student's t test; p < 0.05 was considered statistically significant.
RESULTS
Intravenous injection of VEGF at the dose of 1-250 μg/kg caused a rapid dose-dependent decrease in MAP that reached a nadir at 2-5 min and lasted 20 min (Fig. 1A) . In conjunction with the decrease in MAP, VEGF produced dose-related increases in HR that were rapid, reached a peak level at 2-5 min, and lasted 20 min at the lower doses and >60 min at the higher doses (Fig. 1B) . The depressor and tachycardic responses to VEGF at 500 μg/kg (data not shown in Fig. 1 for the sake of clarity) were not different from those to the dose of 250 μg/kg.
Intravenous administration of L-NAME had significant pressor and bradycardic effects (p < 0.01) (Fig. 2) . The maximal increase in MAP and decrease in HR were ≈42 and 20% of baseline, respectively. Injection of the saline vehicle did not alter MAP or HR. Pretreatment with L-NAME significantly attenuated the depressor and tachycardic responses to VEGF in conscious rats (p < 0.05) (Fig. 2) . In contrast, pretreatment with phenylephrine at the dose that induced a pressor effect similar to that induced by L-NAME did not significantly affect the hemodynamic responses to VEGF (Fig. 2) .
The basal level of cardiac output measured by the ultrasonic flowmeter was 115.2 ± 6.9 ml/min in conscious rats. Intravenous administration of VEGF (250 μg/kg) resulted in a significant decrease in cardiac output (Fig. 3A) . Actually, cardiac output was transiently increased immediately after intravenous injection of VEGF and rapidly decreased to the basal level at 3 min. However, this increase was not statistically significant. A significant reduction (p < 0.01) in cardiac output in response to VEGF began 5 min after injection, reached a nadir at 10 min, and lasted at least 60 min. VEGF also significantly reduced stroke volume (p < 0.01) (Fig. 3B) . The pattern of the reduction in stroke volume was similar to that in cardiac output. The maximal reduction in cardiac output and stroke volume was 23 and 32%, respectively.
Intravenous injection of VEGF (250 μg/kg) caused a significant decrease in MAP and increase in HR as before. MAP began to decrease almost immediately after injection, reached a nadir at 3 min, and returned to the pretreatment level at 15-20 min (Fig. 4A) . Systemic vascular resistance was initially reduced for 3 min (p < 0.05), returned to the basal level at 5 min, and then was increased (p < 0.01) at 10-40 min (Fig. 4B) . There was no significant alteration in maximum LVdP/dt after VEGF administration as compared with the pretreatment level (Fig. 5) .
Administration of VEGF, at a dose of 6.25 μg/ml for 12 min, to the rat isolated heart preparation had no direct effect on either the LVDP, HR, or dP/dt (Fig. 6) . A slight but transient depressor effect was observed consistently on initiation of the infusion, together with an equally transient stimulatory effect on completion of the infusion. Similar responses were observed during excipient infusions.
Intravenous injection of VEGF 250 μg/kg resulted in a significant increase in hematocrit (p < 0.01) at 10 min, the time when a maximal decrease in cardiac output or stroke volume was observed (Fig. 7) . The hematocrit at 4 h returned to the pretreatment level. Intravenous injection of vehicle did not cause significant alterations in hematocrit. There were no significant differences in hematocrit before treatment or 4 h after treatment between vehicle- and VEGF-treated groups. However, a significant difference (p < 0.01) in hematocrit at 10 min was observed between the two groups.
DISCUSSION
This is the first systematic investigation of the effects of VEGF on the cardiovascular system. Our results demonstrated that intravenous administration of VEGF caused a dose-related decrease in MAP and a compensatory increase in HR in conscious SD rats. Pretreatment with an inhibitor of NO formation significantly attenuated the depressor and tachycardic responses to VEGF, suggesting that the depressor effect of VEGF is mediated, at least in part, by NO. Further, intravenous administration of VEGF resulted in a reduction in cardiac output and stroke volume in conscious rats. Our in vitro and in vivo studies also showed that VEGF had no direct depressant effect on the myocardium, suggesting that the VEGF-induced decrease in cardiac output and stroke volume may be due to a decrease in venous return.
Our finding that intravenous administration of VEGF decreased arterial pressure is consistent with the observation of Ku and colleagues (16) (16) N- mono-methyl-L-arginine (L-NMMA), another NO synthase inhibitor, indicating that the vasodilatory effect of VEGF is at least partly endothelium-dependent and is mediated by NO release. These data are consistent with the previous observation that endothelial cells are the biologically relevant target of VEGF, both in vitro and in vivo (21)
We observed that the depressor response to VEGF was associated with an increase in HR in conscious animals. Furthermore, pretreatment with a NO synthase inhibitor that attenuated the depressor response to VEGF attenuated the tachycardic response. VEGF, however, did not alter HR in the isolated heart preparation. Together, the data indicate that VEGF-induced tachycardia in conscious animals is probably a reflex response to a decrease in arterial pressure rather than a direct effect on the cardiac pacemaker.
To avoid the effects of anesthesia on hemodynamics and cardiac function, in the present study we measured cardiac output using a flowprobe around the ascending aorta in conscious, unrestrained rats. After intravenous injection of VEGF, cardiac output and stroke volume were significantly reduced despite the increase in HR, indicating that the decrease in cardiac output is due to reduced stroke volume. There are three main factors that could influence stroke volume: myocardial contractility, afterload, and preload. Our results suggest that VEGF did not affect myocardial contractility, as manifested either in maximum dP/dt in the isolated rat heart in vitro or in LVdP/dt in conscious animals in vivo. The data argue against the possibility that the decrease in stroke volume is due to a direct cardiac depressant action of VEGF.
The pattern of systemic vascular resistance observed after VEGF administration showed two opposite phases: (a) an early reduction phase, beginning almost immediately, lasting 3 min, and recovering at 5 min; and (b) a late elevation phase, beginning at 10 min, lasting 30 min, and recovering after 50-60 min (Fig. 5) . The pattern of the early reduction phase was consistent with that of MAP reduction, suggesting that the reduction in systemic vascular resistance and in arterial pressure may be due to the vasodilatory effect of VEGF. The late elevation phase, however, corresponded to the decrease in cardiac output (Fig. 4) , suggesting that the increase in systemic vascular resistance may be secondary to a reflex response to decreased cardiac output. Therefore, the change in peripheral vascular resistance or afterload may not be the primary reason for the VEGF-induced decrease in stroke volume.
Finally, a reduction in venous return or preload might be expected to be related to decreased stroke volume caused by VEGF. One mechanism for the reduction in venous return is venodilation. Saturable binding sites for VEGF have been identified on venous endothelial cells (22-24) (16) (25) (26,27) (26) (2,4) (28,29)
The acute hemodynamic effects of VEGF include (a) a decrease in arterial pressure which may be mediated by NO; (b) a decrease in cardiac output, caused by reduced stroke volume, despite an increase in HR; (c) a reduction in stroke volume probably due to a decrease in preload rather than a direct effect on myocardial contractility; and (d) an increase in vascular permeability and/or venodilation, leading to the decrease in venous return and preload. Whether the hemodynamic and angiogenic effects of VEGF are mediated by the same receptor or receptors or are in any way linked remains to be seen. Nevertheless, the hemodynamic effects should be taken into account when VEGF is used for angiogenic therapy.
FIG. 1:
. Dose-related responses of mean arterial pressure (MAP) (A) and heart rate (HR) (B) to intravenous injection of vascular endothelial growth factor (VEGF) in conscious rats. Data are the mean of 6-9 animals (the number in parentheses is the animal number in each group). * p < 0.05 and ** p < 0.01 as compared with the other two doses at the same timepoint. # p < 0.05 and ## p < 0.01 as compared with the 1-μg/kg dose at the same timepoint.
FIG. 2:
. Effects of N ω -nitro-L-arginine methyl ester (L-NAME 30 mg/kg followed by 4 mg/kg/min for 8 min), a specific inhibitor of nitric oxide formation, or phenylephrine (PE 7.5 μg/kg/min) on the depressor and tachycardic responses to vascular endothelial growth factor (VEGF 50 μg/kg). * p < 0.05 as compared with the vehicle + VEGF or PE + VEGF group (n = 5 in each group). HR, heart rate.
FIG. 3:
. Effects of vascular endothelial growth factor (VEGF) on cardiac output (CO) (A) and stroke volume (SV) (B) in conscious rats. * p < 0.05 and ** p < 0.01 as compared to the 0 timepoint.
FIG. 4:
. Effects of vascular endothelial growth factor (VEGF) on mean arterial pressure (MAP) (A) and systemic vascular resistance (SVR) (B) in conscious rats. * p < 0.05 and ** p < 0.01 compared to the 0 timepoint.
FIG. 5:
. Effects of vascular endothelial growth factor (VEGF) on left ventricular (LV) maximum dP/dt in conscious rats.
FIG. 6:
. Trace representing a recording from a rat isolated heart preparation perfused at a constant pressure of 100 mm Hg with Krebs-Henseleit solution gassed with 95% O2 /5% CO2 . The addition of vascular endothelial growth factor (VEGF 6.25 μg/ml) to the perfusate had no depressor effect on left ventricular developed pressure (LVDP), heart rate (HR), or ventricular contractility (dP/dt ).
FIG. 7:
. Effects of vascular endothelial growth factor (VEGF) on hematocrit in conscious rats. ** p < 0.01 as compared with the respective vehicle-treated group at the same time. ## p < 0.01 as compared to the 0 timepoint in the same group. Vehicle (open columns, n = 5); VEGF 250 μg (hatched columns, n = 5).
REFERENCES
1. Ferrara N, Henzel WJ. Pituitary follicular cells secrete a novel heparin-binding
growth factor specific for vascular endothelial cells.
Biochem Biophys Res Commun 1989;161:851-8.
2. Connolly DT, Heuvelman D, Nelson R, et al. Tumor vascular permeability factor stimulates endothelial cell growth and angiogenesis.
J Clin Invest 1989;84:1470-8.
3. Connolly DT, Olander JV, Heuvelman D, et al. Human vascular permeability factor. Isolation from U937 cells.
J Biol Chem 1989;264:20017-24.
4. Keck PJ, Hauser SD, Krivi G, et al. Vascular permeability factor, an endothelial mitogen related to PDGF.
Science 1989;246:1309-12.
5. Tischer E, Mitchell R, Hartmen T, et al. The human gene for vascular endothelial
growth factor : multiple protein forms are encoded through alternative exon splicing.
J Biol Chem 1991;266:11947-54.
6. Leung DW, Cachianes G, Kuang WJ, Goeddel DV, Ferrara N. Vascular endothelial
growth factor is a secreted angiogenic mitogen.
Science 1989;246:1306-9.
7. Houck KA, Ferrara N, Winer J, Cachianes G, Li B, Leung DW. The vascular endothelial
growth factor family: identification of a fourth molecular species and characterization of alternative splicing of RNA.
Mol Endocrinol 1991;5:1806-14.
8. Phillips HS, Hains J, Lenug DW, Ferrara N. Vascular endothelial
growth factor is expressed in rat corpus luteum.
Endocrinology 1990;127:965-8.
9. Breier G, Albrecht U, Sterrer S, Risau W. Expression of vascular endothelial
growth factor during embryonic angiogenesis and endothelial cell differentiation.
Development 1992;114:521-32.
10. Jakeman LB, Winer J, Bennett GL, Altar CA, Ferrara N. Binding sites for vascular endothelial
growth factor are localized on endothelial cells in adult rat tissues.
J Clin Invest 1992;89:244-53.
11. de Vries C, Escobedo JA, Ueno H, Houck KA, Ferrara N, Williams LT. The fms-like tyrosine kinase, a receptor for vascular endothelial
growth factor .
Science 1992;255:989-91.
12. Terman BI, Vermazen MD, Carrion ME, et al. Identification of the KDR tyrosine kinase as a receptor for vascular endothelial
growth factor .
Biochem Biophys Res Commun 1992;34:1578-86.
13. Quinn T, Peters KG, de Viries C, Ferrara N, Williams LT. Fetal liver kinase 1 is a receptor for vascular endothelial
growth factor and is selectively expressed in vascular endothelial.
Proc Natl Acad Sci USA 1993;90:7533-7.
14. Takeshita S, Zheng LP, Brogi E, et al. Therapeutic angiogenesis. A single intraarterial bolus of vascular endothelial
growth factor augments revascularization in a rabbit ischemic hind limb model.
J Clin Invest 1994;93:662-70.
15. Banai S, Jaklitsch MT, Shou M, et al. Angiogenic-induced endothelial enhancement of collateral blood flow to ischemic myocardium by vascular endothelial
growth factor in dogs.
Circulation 1994;89:2183-9.
16. Ku DD, Zaleski JK, Liu S, Brock TA. Vascular endothelial
growth factor induces EDRF-dependent relaxation in coronary arteries.
Am J Physiol 1993;265 (
Heart Circ Physiol 34):H585-92.
17. Ferrara N, Leung DW, Cachianes G, Winer J, Henzel WJ. Purification and cloning of vascular endothelial
growth factor secreted by pituitary folliculostellate cells.
Methods Enzymol 1991;198:391-404.
18. Pfeffer JM, Pfeffer MA, Braunwald E. Influence of chronic captopril therapy on the infarcted left ventricle of the rat.
Circ Res 1985;57:84-95.
19. Buttrick P, Perla C, Malhotra A, Geenen D, Lahorra M, Scheuer J. Effects of chronic dobutamine on cardiac mechanics and biochemistry after myocardial infarction.
Am J Physiol 1991;260 (
Hear Circ Physiol 29):H473-9.
20. Yang RH, Bunting S, Gillett N, Clark RG, Jin H. Effects of growth hormone in rats with postinfarction left ventricular dysfunction.
Cardiovasc Drugs Ther 1995;9:125-31.
21. Ferrara N, Houck K, Jakeman L, Leung DW. Molecular and biological properties of the vascular endothelial
growth factor family of proteins.
Endocrinol Rev 1992;13:18-32.
22. Bikfalvi A, Sauzeau C, Moukadiri H, et al. Interaction of vasculotropin/vascular endothelial
growth factor with human umbilical endothelial cells: binding, internalization, degradation, and biological effects.
J Cell Physiol 1991;149:50-9.
23. Gitay-Goren H, Soker S, Vlodasky I, Neufeld G. The binding of vascular endothelial
growth factor to its receptors is dependent on cell surface associated heparin-like molecules.
J Biol Chem 1992;267:6093-8.
24. Myoken Y, Kayada Y, Okamoto T, Kan M, Sato GH, Sato JD. Vascular endothelial cell
growth factor (VEGF) produced by A-431 human epidermoid carcinoma cells and identification of VEGF membrane binding sites.
Proc Natl Acad Sci USA 1991;88:5819-23.
25. Moncada S, Palmer RMJ, Higgs EA. Nitric oxide: physiology, pathophysiology, and pharmacology.
Pharmacol Rev 1991;43:109-42.
26. Thomas GR, Thiemermann C, Walder C, Vane JR. The effects of endothelium-dependent vasodilators on cardiac output and their distribution in the anesthetized rat: a comparison with sodium nitroprusside.
Br J Pharmacol 1988;95:986-92.
27. Pouler H, Covell JW, Ross J. Effects of nitroprusside on venous return and central blood volume in absence and presence of acute heart failure.
Circulation 1980;61:328-36.
28. Senger DR, Galli SJ, Dvorak AM, Peruzzi CA, Harvey VS, Dvorak HF. Tumor cells secrete a vascular permeability factor that promotes accumulation of ascites fluid.
Science 1983;219:983-5.
29. Senger DR, Perruzzi CA, Feder J, Dvorak HF. A highly conserved vascular permeability factor secreted by a variety of human and rodent tumor cell lines.
Cancer Res 1986;46:5629-32.
