topeeng.gif (8383 bytes)

[ Scientific Activity - Actividad CientÝfica ] [ Brief Communications - Temas Libres ]

Inhibition of agonist-induced rises in internal calcium in human endothelial cells by Ginkgo biloba extract (EGb 761) and rolipram

Campos-Toimil Manuel; Lugnier Claire; Takeda Kenneth

Pharmacologie et Physico-Chimie des Interactions Moleculaires et Cellulaires CNRS, UMR 653491
UniversitÚ Louis Pasteur
Strasbourg, France

Abstract
Introduction
Objectives
Material and Methods
Results
Discussion
Conclusions
References

Abstract
EGb 761 is a standardized extract which has been shown to have complex and multiple cardiovascular effects, including increases in peripheral and cerebral blood flow and microcirculation, inhibition of platelet aggregation and a decrease in capillary permeability. It also relaxes spasmotic blood vessels and contracts those that are abnormally dilated. In the present work, the effects of EGb 761 on five isolated vascular cyclic nucleotide phosphodiesterase (PDE) isoforms were evaluated. EGb 761 preferentially inhibits PDE4 (IC50 = 25▒3 mg/L), the isoform which is mainly present in endothelial cells, in a competitive manner (Ki = 12.5 mg/L). Since changes in cyclic nucleotide levels may affect intracellular calcium levels ([Ca2+]i) in endothelial cells, we examined the effects of EGb 761 on both resting [Ca2+]i levels and agonist-induced rises in [Ca2+]i in single human umbilical vein endothelial cells (HUVEC) in culture. The effects of EGb 761 were compared to those of rolipram, a selective PDE4 inhibitor which increases cellular cAMP levels, and the cAMP analogue dibutyryl-cAMP (db-cAMP). EGb 761 (20, 100 mg/L), rolipram (50 Ámol/L) and db-cAMP (100 Ámol/L) significantly inhibit histamine-, ATP- and thrombin-induced [Ca2+]i increases in HUVEC, without modifying resting [Ca2+]i levels. Similar results were obtained using a Ca2+-free bathing solution. EGb 761 (100 mg/L), but not rolipram (50 Ámol/L) or db-cAMP (100 Ámol/L), also inhibits the capacitative Ca2+ entry in cells having thapsigargin-depleted internal Ca2+ stores bathed in Ca2+-free external solution. Our results are consistent with an inhibition of PDE activity causing a reduction of agonist-induced increases in [Ca2+]i in HUVEC, mainly by inhibition of Ca2+ mobilization from internal stores. Thus, it may also be that the cardiovascular effects of EGb 761 involve an inhibition of PDE4 activity, and a subsequent modification of Ca2+ signaling in endothelial cells.

Top

Introduction:

EGb 761 a highly standardized extract prepared from the leaves of Ginkgo biloba L. has been shown to have complex and multiple cardiovascular effects1. It relaxes spasmotic blood vessels and contracts those that are abnormally dilated1,2. Contractions elicited by EGb 761 appear to involve a release of catecholamines2 and its vasorelaxant effect seems to depend upon an intact endothelium1,2 and is mediated at least in part by scavenging of free-radicals3, thereby protecting nitric oxide from oxidation. However, it is unclear whether EGb 761 interferes with the generation and release of endothelial vasoactive factors. One important mechanism of action underlying the vascular effects of EGb 761, in addition to free-radical scavenging, may be inhibition of cyclic nucleotide phosphodiesterase (PDE) activity, since an inhibitory effect of EGb 761 on cAMP-PDE activity has been previously described4. It is generally accepted that an important crosstalk between cyclic nucleotide levels and intracellular Ca2+ levels ([Ca2+]i) occurs in various cell types and that many endothelium-dependent processes which contribute to the regulation of vascular tone and blood cell activation depend on [Ca2+]i in endothelial cells. Thus, substances which influence endothelial cell [Ca2+]i homeostasis potentially influence the production of endothelial factors and underlying vascular tone. The present study was conducted to examine whether some of the cardiovascular effects of EGb 761 might be explained by an interference with the [Ca2+]i handling in endothelial cells and whether this interference could be mediated by an inhibition of cyclic nucleotide activity.

Top

Material and Methods:

PDE activities were measured by radioenzymatic assay as described previously5. EGb 761 dose-effect curves of PDE activity were made using six concentrations of the extract. Human umbilical endothelial cells (HUVEC) were cultured as described previously6. The cells were subcultured in 75 cm2 flasks and for experiments in 35 mm Petri dishes in which a 20 mm diameter hole had been cut in the base and replaced by a thin (0.07 mm) glass coverslip. Cells were seeded at low density (2 x 103 cells/cm2), and kept in culture (37░C, 5% CO2 in air) for 2-4 days before experiments. [Ca2+]i levels were determined used by single-cell measurements of [Ca2+]i using a Fura-2-based digital imaging setup, following the method described in Lynch et al.7 Results are expressed as means ▒ s.e.mean. Significant differences between means (P<0.05) were determined by Student’s t-test for unpaired data. The area under the Ca2+ curves from individual cells was determined by the trapezoid rule (Prism 2 software; Graphpad). The IC50 values of EGb 761 against the different PDE isoforms were calculated by non linear regression. Apparent inhibitory constant (Ki) values for EGb 761 were determined by using substrate concentrations ranging from 0.1 to 20 Ámol/L. Lineweaver-Burk plots were analyzed by least squares linear regression analysis.

Top

Results:

EGb 761 preferentially inhibited PDE4 and to a lesser extent PDE5, having negligible effects on PDE1 and PDE2 in their basal states (Table 1). However, EGb 761 significantly inhibited the Ca2+-calmodulin activated state of PDE1, indicating a possible interaction with calmodulin. Moreover, PDE2 was more sensitive to EGb 761 in its cGMP-stimulated state. Kinetic studies showed that EGb 761 inhibits competitively PDE4, PDE5 and activated PDE2 with Ki values of 12.5▒1.1 mg/L, 14.0▒1.6 mg/L and 62.7▒5.9 mg/L, respectively (Figure 1). The apparent Km values for PDE4, PDE5 and activated PDE2 were 0.89▒0.07 Ámol/L cAMP, 0.44▒0.03 Ámol/L cGMP and 14.28▒0.12 Ámol/L cAMP, respectively. The mean basal [Ca2+]i level in HUVEC was 79.9▒2.3 nmol/L (n=72), and was unchanged over the experimental time course. Resting [Ca2+]i levels were not significantly affected by a 2 min pre-incubation with EGb 761 (1-100 mg/L) rolipram (10, 50 Ámol/L) or db-cAMP (100 Ámol/L). A 60 s application of histamine (10 Ámol/L) or ATP (10 Ámol/L) or thrombin (5000 U/L) caused a biphasic increase in [Ca2+]i, which was significantly decreased by a 2 min pre-incubation of cells with 20-100 mg/L EGb 761 . Lower doses of EGb 761 were without effect (Figure 2a). The effects of EGb 761 against agonist-induced increases in [Ca2+]i were mimicked by a 2 min pre-incubation of cells with 50 Ámol/L rolipram, a selective PDE4 inhibitor (Figure 3a-c), but not by 10 Ámol/L rolipram. Similarly, 2 min pre-incubation with 100 Ámol/L dibutyryl-cAMP (db-cAMP) also significantly reduced the [Ca2+]i rise induced by histamine (Figure 3d). Similar results were obtained using a Ca2+-free bathing solution (Figure 4a-e)   EGb 761 (100 mg/L), but not rolipram (50 Ámol/L) or db-cAMP (100 Ámol/L), also inhibits the capacitative Ca2+ entry in cells having thapsigargin-depleted internal Ca2+ stores bathed in Ca2+-free external solution (Figure 4f).

Table  1

table1.gif (5677 bytes)

 

fig1.gif (7050 bytes)
Fig. 1: Inhibitory effects of EGb 761 on isolated vascular phosphodiesterase (PDE) isoforms.
(a) Dose-effect curves for PDE4 and PDE2 in its basal and activated states.
Lineweaver- Burke representation of the inhibitory effects of EGb 761 against
PDE4 (b). PDE5 (c) and activated PDE2 (d). Data are representative of 3 experiments.

Top

 

fig2.gif (9335 bytes)
Fig. 2: Effects of 2 min pre-incubation of HUVEC with different concentrations of EGb 761
(a) on the maximal [Ca
2+]i increase induced by histamine. Effects of 2 min pre-incubation
of HUVEC with EGb 761 (100mg/L) on  [Ca
2+]i increases induced by histamine (b), ATP
(c) and thrombin (d) in a 2 mM Ca2+ - containing external solution and *P <0.05 with
respect to control values.

Top

fig3.gif (8888 bytes)
Fig. 3: The selective PDE4 inhibitor rolipram (50 Ámol/L) reduces increases in [Ca2+]i evoked
by histamine (a), ATP (b) and thrombin (c) in a 2 mmol/L Ca2+-containing external solution.
(d) Inhibitory effect of 100 Ámol/L db-cAMP on histamine-induced increase in [Ca2+]i.
Each value is the mean ▒ s.e.mean of at least 14 cells. *P<0.05 with respect to control values.

Top

fig4.gif (11014 bytes)
Fig. 4: Effects of 2 min pre-incubation with EGb 761 (100 mg/L) on [Ca2+]i increases in HUVEC
induced by histamine (a), ATP (b) and thrombin (c) in Ca2+-free external solution. Inhibitory
effects of 50 Ámol/L rolipram (d) and 100 Ámol/L db-cAMP (e) on histamine-induced increases
in [Ca2+]i in Ca2+-free external solution. Capacitative entry of Ca2+ provoked by application
of 2 mmol/L Ca2+-containing external solution is inhibited by 100 mg/L EGb 761 (f) in thapsigargin
pre-treated HUVEC bathed in Ca2+-free external solution. Each value is the mean ▒ s.e.mean of
at least 11 cells. *P<0.05 with respect to control values.

Top

Discussion:

Agonist-induced [Ca2+]i rises in our preparation were largely accounted for by mobilization from thapsigargin-sensitive internal stores and in a minor way, from Ca2+ influx, as removal of extracellular Ca2+ did not affect the maximal [Ca2+]i achieved and the measured capacitative entry of Ca2+ was relatively weak. We found that EGb 761 preferentially inhibits PDE4 in a competitive manner. As vascular endothelial cells contain mostly PDE2 and PDE48,9, the inhibitory effects of a 2 min pre-incubation of HUVEC with EGb 761 on agonist-induced [Ca2+]i mobilization observed here might be associated with PDE inhibition and subsequent increased cAMP levels. This hypothesis is supported by: i) the cAMP elevating agents rolipram and db-cAMP also inhibiting agonist-induced [Ca2+]i rises in HUVEC; ii) EGb 761 inhibiting both increases in [Ca2+]i and PDE activity at the same range of concentrations; and iii) EGb 7611,2 and PDE4 inhibitors10 being endothelium-dependent relaxing agents. Taken together, our results suggest that elevated cAMP levels interfere with Ca2+ mobilization and that the inhibitory effects of rolipram and EGb 761 on agonist-induced [Ca2+]i increases are both linked to an inhibition of PDE4. However, the effect of elevating cAMP on [Ca2+]i in endothelial cells is presently somewhat controversial11. These discrepancies might be explained by differences in the degree of cAMP accumulation and functional compartmentalization of the cAMP cascade, allowing different physiological responses to various cAMP increasing agents. In fact, the high cAMP hydrolytic activity of PDE2 overcame global cAMP increase to a local PDE4 inhibition8. Thus, cAMP assay failed to detect local cAMP increases induced by PDE4 inhibition specifically implicated in physiological responses9, 12. At least one effect of EGb 761 (at 100 mg/L, close to the IC50 against PDE2) in reducing capacitative Ca2+ entry was not shared by rolipram or db-cAMP. This suggests that inhibition of activated PDE2 may also be involved in the modulation of [Ca2+]i homeostasis by EGb 761 in HUVEC.

Top

Conclusions:

EGb 761 significantly reduces agonist-induced [Ca2+]i increases in HUVEC, most probably via inhibition of Ca2+ mobilization from internal stores. Given the measured inhibitory action of EGb 761 on PDE4 activity and the similar effects of rolipram and db-cAMP, this effect likely involves an elevation of cAMP levels. Our data indicate that the complex cardiovascular effects of EGb 761 involve, at least in part, an inhibition of PDE4 activity and a subsequent modification of calcium signaling in endothelial cells.

Top

References

1. DeFeudis FV (1994) In Advances in Ginkgo biloba Extract Research, Vol. 3. Clostre F, DeFeudis FV (eds) pp. 105-119. Paris: Elseiver.
2. Auguet M et al (1994) In Advances in Ginkgo biloba Extract Research, Vol. 3. Clostre F, DeFeudis FV (eds) pp. 19-26. Paris: Elseiver.
3. Maitra I et al (1995) Biochem Pharmacol, 49, 1649-1655.
4. Macovschi O et al (1987) J Neurochem, 49, 107-114.
5. Keravis TM et al. (1980) Biochim Biophys Acta, 613, 116-129.
6. Klein-Soyer C et al (1986) Thromb Haemost, 56, 232-235.
7. Lynch JW et al (1994) J Biol Chem, 269, 8226-8233.
8. Lugnier C, Schini VB (1990) Biochem Pharmacol, 39, 75-84.
9. Souness JE et al (1990) Biochem J, 266, 127-132.
10. Komas N et al (1991) Br J Pharmacol, 104, 495-503.
11. Moore TM et al (1998) Am J Physiol, 275, L203-L222.
12. Kessler T, Lugnier C (1995) Eur J Pharmacol, 290, 163-167.

Top

Questions, contributions and commentaries to the Authors: send an e-mail message (up to 15 lines, without attachments) to pharma-pcvc@pcvc.sminter.com.ar , written either in English, Spanish, or Portuguese.

Top


ę CETIFAC
Bioengineering
UNER
Update
Mar/05/2000


These Companies contributed to the Congress

bago.gif (384 bytes)bagople6.gif (2759 bytes)

klchlogo4.gif (6328 bytes)