Abstract
Objective Endothelin (ET) modulates cellular processes relevant to vascular remodeling, but there is still some debate as to the potential of ET to be a trophic factor or a mitogen. Moreover, the signaling of ET in vivo to produce these effects is largely unknown.
Methods3H-leucine and 3H-thymidine incorporation in rat small mesenteric arteries was studied with several doses of ET-1 (0.1–10pmol/kg/min) administered for 26h in vivo.
Results The EC50 for protein synthesis was four times lower than that of DNA synthesis, with maximal effects around 1 and 3pmol/kg/min, respectively. At 5pmol/kg/min, ET enhanced CDK2 activity by reducing the binding of its inhibitor p27Kip1. In contrast, the binding was enhanced at 0.5pmol/kg/min. The reduced binding observed at 5pmol/kg/min could not be explained by changes of p27Kip1 or CDK2 content. Phosphorylation of p27Kip1 on serine 10 was significantly reduced at 5pmol/kg/min ET. Although the phosphoinositide 3-kinase pathway was activated, it did not contribute to the protein or DNA synthesis responses. Administration of 1 or 5pmol/kg/min ET-1 for 28days increased the thickness and cross-sectional area of the small mesenteric artery due to hypertrophy and hyperplasia, respectively, thus confirming the results obtained in acute conditions.
Conclusion ET modulates p27Kip1 binding to CDK2, producing hypertrophy at low and hyperplasia at higher concentrations. Taken together, these results suggest that ET can act both as a trophic factor and as a mitogen in an in vivo environment, depending on its local concentration.
This article is referred to in the Editorial by M. Takahashi (pages 4–5) in this issue.
1 Introduction
Endothelin (ET)-1, a 21-amino-acid peptide, was initially identified as a potent and long lasting vasoconstrictor substance isolated from endothelial cells [1]. In mammalian tissues, ET exerts its vascular physiological actions by stimulating two receptors subtypes: ETA located on vascular smooth muscle cells (VSMCs) and ETB mainly situated on endothelial cells [2]. Beside its vasoactive effects, ET has been suggested to possess a direct mitogenic action, as it can increase the expression of cellular proto-oncogenes [3,4]. Moreover, in vitro experiments have illustrated the ability of ET to stimulate the DNA incorporation of 3H-thymidine as well as 3H-leucine into newly synthesized proteins [5–7]. Furthermore, ET seems to be a key factor in the maintenance of norepinephrine (NE)-induced hyperplasia in large arteries [8]. In contrast, chronic studies in vivo have demonstrated that ET represents the most important mediator of cellular hypertrophy in resistance arteries (internal diameter <300μm) undergoing hypertrophic remodeling [9]. Indeed, ET-receptor antagonists blunt the development of small arterial hypertrophy in different experimental models of hypertension, including chronic NE or Angiotensin II (Ang II) administration and DOCA-salt [10–12]. Thus, there is evidence that ET can act as both a trophic and a mitogenic factor in cultured cells in vitro and in large arteries in vivo, but mainly as a trophic mediator in small arteries in vivo.
In eukaryotes, growth factors stimulate signaling cascades which eventually lead to the sequential activation of cyclin-dependent kinases (CDKs), allowing an adequate and ordered progression of the cell cycle [13,14]. Their kinase activity requires the binding of specific positive regulatory subunits known as cyclins [13]. Cyclin E binds to CDK2 and allows the entry into S phase and initiation of DNA replication, whereas cyclin A promotes the S phase transition and control of DNA replication by binding to the same kinase [15]. The activity of cyclin/CDK complexes is inhibited by interactions with proteins known as CDK inhibitors (CKIs) [14,16]. Among them, p27Kip1 is a potent inhibitor of cyclin E/CDK2 and cyclin A/CDK2 complexes [14,17]. Its level of expression is appropriately high in quiescent cells and decreases upon re-entry into the cell cycle [18], and its over-expression limits vascular injury-induced intimal hyperplasia in vivo [19]. There are conflicting evidences regarding the effect of ET on cell cycle proteins in culture conditions. Indeed, using the same dose of exogenous ET, one study reported CDK2 activation and DNA synthesis in NIH3T3 cells [6], while another showed no cell cycle activation in VSMC [7].
In the present study, our aim was to determine the trophic and mitogenic response of small mesenteric arteries to increasing doses of locally delivered ET-1 in vivo. In addition, we studied the in vivo signaling events activated by ET to mediate small artery responses, with emphasis on cell cycle regulation by CKIs p27Kip1 and p21Cip1.
2 Methods
2.1 Acute experiments
For the estimation of DNA and protein synthesis, control Sprague-Dawley rats (300–325g, Charles River) were compared to rats receiving 0.1, 0.3, 1, 3 or 10pmol/kg/min of ET-1 (05-23-3800, Calbiochem; n=5–6 per dose and per analysis) for 26h (ET-1d) using Alzet osmotic pump (model 1003D; Durect) implanted intra-peritoneally (i.p.). We have previously confirmed that an i.p. pump releasing saline does not affect protein synthesis (unpublished). Rats were instrumented simultaneously to the pump implantation, as previously described [20]. DNA and protein synthesis were determined by 3H-thymidine and 3H-leucine incorporation. Briefly, freely moving rats received an infusion of 3.75μCi/kg/min of 3H-thymidine or 0.6μCi/kg/min of 3H-leucine intravenously for 4h, i.e. from hours 22 to 26 after the end of surgery and start of ET-1 infusion [20]. LY294002 (4mg/kg; a potent and specific PI-3K inhibitor; Calbiochem) was given 5h before the sacrifice, i.e. 21h after the beginning of ET infusion. Rats were then killed by decapitation under anesthesia and small mesenteric arteries (segments extending from, but excluding, the superior mesenteric artery to the intestine were pooled for each rat) were rapidly collected at 4°C and stored at − 80°C. After appropriate DNA or protein extraction, small mesenteric artery levels of radioactive thymidine and leucine incorporated into DNA and proteins were counted. Data are presented as cpm/μg DNA or cpm/mg protein, respectively.
For Western blot analysis, rats (n=5/group) were treated for 26h with 5pmol/kg/min of ET-1 and did not receive labeled leucine or thymidine. In additional rats (n=5), an i.p. injection of LY294002 or tamoxifen (3.5mg/kg; a PKC inhibitor; Sigma) was given 5h before the sacrifice. Proteins were extracted from small mesenteric arteries by using Triton X-100 lysis [21]. Afterwards, equal amounts (20–40μg) of proteins were resolved by electrophoresis on 10% or 12% PAGE and transferred to PVDF membrane. Thereafter, membranes were incubated at 4°C overnight with antibodies against p27Kip1 (554069, 1:1000, BD Biosciences), p21Cip1 (sc-397, 1:500, Santa-Cruz Biotechnology), CDK2 (sc-163, 1:1000), PKB/Akt (9275; 1:1000, New England Biolabs) and Serine 10 (Ser 10) p-p27Kip1 (34-6300, 1:250, Zymed Labs, San Francisco, CA USA). Immunoreactive bands were then revealed by chemiluminescence after an appropriate incubation with secondary antibodies. Protein loading was corrected by p27 content for ser10-p27, and p27 for p-p27. Optical density of 3 control samples was averaged on each gel and other samples expressed as a % change from this control value set as 100%.
For immunoprecipitation and CDK2 kinase assay experiments, rats were treated with ET-1 at the dose of 5 (all experiments) and 0.5pmol/kg/min (p27 binding to CDK2 only) and compared to controls (n=3–5/group). Following an extraction with Triton X-100 buffer, total proteins lysate (200–400μg) were incubated for 4h at 4°C with polyclonal anti-CDK2 antibody (2μg/sample) and the immune complexes were collected with protein A-Sepharose beads. Binding of p27Kip1 and p21Cip1 to CDK2 was assayed by Western blots corrected by CDK2 protein content and the kinase activity of CDK2 was evaluated by immune complex kinase assay using histone H1 as substrate, as described previously [21]. Binding of p27 to CDK2 was also tested in the presence of a high dose of tamoxifen (3.5mg/kg i.p., 5h before sacrifice), which has been shown to inhibit protein kinase C in vivo [22].
2.2 Chronic study
For the determination of arterial structure during chronic ET stimulation, age-matched Sprague-Dawley rats (175–200g) were compared to rats receiving 1 or 5pmol/kg/min of ET-1 administered via an i.p. Alzet osmotic pump (model 2004) for a period of 4weeks (n=6/group). Rats were weighed before ET-1 administration and weekly thereafter. Prior to sacrifice, a short polyethylene-50 catheter was inserted under anesthesia (Pentobarbital, 65mg/kg i.p.) into the left femoral artery to allow measurement of systolic (SBP), diastolic blood pressure (DBP) and heart rate (HR). The structure of small mesenteric arteries was determined in perfused and pressurized conditions as previously described [11]. Afterwards, the number of VSMCs was determined histologically by using the tri-dimensional dissector technique [11]. The studies at 1 and 5pmol/kg/min ET-1 were performed by different investigators and compared only to their respective controls, since the size of the arteries harvested differed between the two protocols.
2.3 Measurement of ET levels in small mesenteric arterial bed
For ET-1 level measurements in mesenteric arterial beds, controls Sprague-Dawley rats (300–325g) were compared to rats treated with ET-1 (5pmol/kg/min) for 3days, to compare these levels to previous results [11]. Indeed, local ET levels were not measured at all doses of ET, as they are expected to be proportional to dosing. Small mesenteric arteries were harvested and levels of ET were determined by RIA [11]. All the animal study protocols were approved by the Animal Care and Use Committee at the University of Montréal, and conform to NIH guidelines.
2.4 Statistics
Data are presented as mean±SEM. Groups were compared by unpaired Student's t-test. When more than 2 groups were compared (Fig. 3), a one-way ANOVA with Bonferonni's correction was used and a priori comparisons were: endothelin vs. control and endothelin+LY294002 vs. endothelin. Comparisons were significant when P<0.05.
Effects of 26-h endothelin-1 (ET-1) administration alone or in association with the inhibitor of the PI-3 kinase pathway LY294002 (+LY) on small mesenteric artery Akt phosphorylation. In a representative western blot, lanes 1–3 are from control rats, lanes 4–6 from rats treated with ET-1 at 5pmol/kg/min and lanes 7–8 from rats treated with ET-1 and LY294002. IB: immunoblot; IP: immunoprecipitation. *P<0.05 vs. control; † P<0.05 vs. ET-1 (ANOVA+Bonferroni's correction).
3 Results
3.1 Endothelin dose–response curves
Endothelin-1 administration for 26h produced a sigmoid increase of protein and DNA synthesis (Fig. 1). The EC50 for protein synthesis was 0.33pmol/kg/min, whereas ET had an EC50 of 1.28pmol/kg/min for DNA synthesis. This fourfold increase was significant when analyzed by non-linear fitting in Prism software. Protein synthesis reached a maximum around 1pmol/kg/min, whereas DNA synthesis plateaud around 3pmol/kg/min. From these experiments, a dose of 0.5, producing mainly protein synthesis, and 5pmol/kg/min, inducing both protein and DNA synthesis, were compared in the signaling studies (represented by dashed lines in Fig. 1).
Effects of increasing doses of endothelin-1 (ET) administered for 26h on DNA synthesis (left axis, open circles) and protein synthesis (right axis, closed circles) in small mesenteric arteries. The scales of Y-axes were adjusted to have the minimum and maximum responses coincide to better appreciate the shift in EC50. Values of protein synthesis were slightly offset to the right to improve clarity. Dashed lines represent the doses used in other acute studies.
3.2 Signaling
To identify potential mechanisms leading to DNA synthesis, signaling events were studied at the higher dose of ET (5pmol/kg/min), since the low dose did not modify DNA synthesis. Endothelin-1 did not modify significantly the mesenteric artery protein content of two important CKI: p27Kip1 and p21Cip1 (Fig. 2A). ET-1 stimulated CDK2 kinase activity on its substrate histone H1 (Fig. 2B), and this response was not associated with modifications of CDK2 protein content in small mesenteric arteries (Fig. 2A). Since the activity of CDK2 depends on a reduction of CKI binding, CDK2 was immunoprecipitated and the amount of CKI bound to the kinase determined by Western blot analysis. No alteration in the binding of p21Cip1 to CDK2 was noticed following ET-1 stimulation in small mesenteric arteries (not shown). However, a significant decrease in the complex formation between CDK2 and p27Kip1 was observed at the end of the 26-h ET stimulation (40% reduction, Fig. 2C). In contrast to what was found with the higher dose, 0.5pmol/kg/min of ET-1 actually enhanced the binding of p27Kip1 to CDK2, thus strengthening the inhibitory effect on the cell cycle (Fig. 2C). The relative Ser10 phosphorylation corrected by the total amount of p27kip1 was not significantly modified at 0.5pmol/kg/min ET-1, but was reduced at 5pmol/kg/min (Fig. 2D).
Effects of 26-h in vivo endothelin-1 administration on small mesenteric artery (A) protein content of p27Kip1 (Kip1), p21Cip1 (Cip1) and CDK2, (B) CDK2 activity, (C) binding of p27Kip1 to CDK2 and (D) p27Kip1 phorphorylation on serine 10. In this panel, the dashed line delineates the control response. All experiments were performed with 5pmol/kg/min (ET-1 5, black bars), except the right portion of panels C and D, which were performed at 0.5pmol/kg/min (ET-1 0.5, hatched bars). See Methods for further details. *P<0.05 vs. control, Student's t-test.
Although ET (5pmol/kg/min) stimulated Akt phosphorylation in vivo (Fig. 3, lanes 4–6 versus lanes 1–3), suggesting PI-3K activation, the specific inhibitor of this signaling cascade, LY294002, did not alter ET-induced protein or DNA synthesis (data not shown), although it reduced Akt phosphorylation by ≈40% (Fig. 3, lanes 7 and 8). Tamoxifen, which was used to block protein kinase C, did not improve p27 binding to CDK2 at 5pmol/kg/min ET-1 (59.8±18.3% vs. 73.0±13% in the ET-1 group).
3.3 Vascular remodelling and local ET concentration
Chronic ET-1 treatment for 28days via an i.p. osmotic pump did not alter any hemodynamic parameter or weight gain. Data for the 5pmol/kg/min ET-1 group are the following: mean blood pressure: 93.3±7.8mm Hg vs. 104.5±6.8 in controls; heart rate: 338±20bpm vs. 362±12 in controls; weight gain: 43.2±2.2g/week vs. 47.2±2.1 in controls. Both 1 and 5pmol/kg/min ET-1 produced an enlargement of small mesenteric arteries, as indicated by the significant increase of both internal and external diameters, which was associated with a marked medial thickening of the arteries (P<0.05, Table 1 and Fig. 4A). Altogether, these structural alterations contributed to a significant increase of CSA in both groups (Fig. 4B). At a dose of 0.5pmol/kg/min, ET-1 did not produce vascular hypertrophy (data not shown). To determine the nature of cellular changes (hypertrophy or hyperplasia), we performed stereological analysis of the small mesenteric arterial wall, by using the tri-dimensional dissector method. Our results show that the number of VSMCs doubled in the 5pmol/kg/min group but was unchanged at the 1pmol/kg/min dose (Fig. 4C). In the high dose group, doubling of VSMC number with an unchanged cell density (Table 1) suggests that hyperplasia likely explains the increase in CSA. In the low dose group, unchanged cell number with a trend for decreased density and a significant increase in cross-sectional area suggests that cellular hypertrophy occurred.
Values of media thickness (A), cross-sectional area (CSA, B) and number of vascular smooth muscle cells (C) in the 4th branch of small mesenteric arteries in control rats (Ctl) and rats treated with endothelin-1 for 28days at 1 (ET-1 1) and 5pmol/kg/min (ET-1 5, n=8/group). Table 1 presents the data used for calculations. See Methods for further details. *P<0.05 vs. control rats, Student's t-test.
Small mesenteric artery characteristics of control and rats treated for 28 days with an i.p. infusion of ET-1 (1 or 5pmol/kg/min).
| Control | ET-1 (5) | Control | ET-1 (1) | |
|---|---|---|---|---|
| Structure | ||||
| ID (μm) | 230±10 | 288±12* | 278±12 | 377±29* |
| ED (μm) | 270±10 | 363±13* | 329±13 | 433±31* |
| 3D-disector | ||||
| Volume (× 103 μm3) | 22.1±5.3 | 42.5±14.5 | 19.4±4.6 | 13.2±2.3 |
| Nt | 41±10 | 58±7 | 15±2 | 16±2 |
| Nb | 25±7 | 27±4 | 14±2 | 15±2 |
| Nv (μm−3 × 10−5) | 73.2±10 | 76.7±8 | 25.0±3.0 | 13.9±6.0 |
| Control | ET-1 (5) | Control | ET-1 (1) | |
|---|---|---|---|---|
| Structure | ||||
| ID (μm) | 230±10 | 288±12* | 278±12 | 377±29* |
| ED (μm) | 270±10 | 363±13* | 329±13 | 433±31* |
| 3D-disector | ||||
| Volume (× 103 μm3) | 22.1±5.3 | 42.5±14.5 | 19.4±4.6 | 13.2±2.3 |
| Nt | 41±10 | 58±7 | 15±2 | 16±2 |
| Nb | 25±7 | 27±4 | 14±2 | 15±2 |
| Nv (μm−3 × 10−5) | 73.2±10 | 76.7±8 | 25.0±3.0 | 13.9±6.0 |
Values are mean±SEM (n=8/group). ET-1: endothelin-1; MBP: mean blood pressure; HR: heart rate; ID, internal diameter; ED, external diameter; Nt, number of nuclei present in the top section; Nb, number of nuclei still present in the bottom section; Nv, cell numerical density. Structural characteristics calculated from these values are presented in Fig. 4. *P<0.05 vs. control rats, Student's t-test.
Small mesenteric artery characteristics of control and rats treated for 28 days with an i.p. infusion of ET-1 (1 or 5pmol/kg/min).
| Control | ET-1 (5) | Control | ET-1 (1) | |
|---|---|---|---|---|
| Structure | ||||
| ID (μm) | 230±10 | 288±12* | 278±12 | 377±29* |
| ED (μm) | 270±10 | 363±13* | 329±13 | 433±31* |
| 3D-disector | ||||
| Volume (× 103 μm3) | 22.1±5.3 | 42.5±14.5 | 19.4±4.6 | 13.2±2.3 |
| Nt | 41±10 | 58±7 | 15±2 | 16±2 |
| Nb | 25±7 | 27±4 | 14±2 | 15±2 |
| Nv (μm−3 × 10−5) | 73.2±10 | 76.7±8 | 25.0±3.0 | 13.9±6.0 |
| Control | ET-1 (5) | Control | ET-1 (1) | |
|---|---|---|---|---|
| Structure | ||||
| ID (μm) | 230±10 | 288±12* | 278±12 | 377±29* |
| ED (μm) | 270±10 | 363±13* | 329±13 | 433±31* |
| 3D-disector | ||||
| Volume (× 103 μm3) | 22.1±5.3 | 42.5±14.5 | 19.4±4.6 | 13.2±2.3 |
| Nt | 41±10 | 58±7 | 15±2 | 16±2 |
| Nb | 25±7 | 27±4 | 14±2 | 15±2 |
| Nv (μm−3 × 10−5) | 73.2±10 | 76.7±8 | 25.0±3.0 | 13.9±6.0 |
Values are mean±SEM (n=8/group). ET-1: endothelin-1; MBP: mean blood pressure; HR: heart rate; ID, internal diameter; ED, external diameter; Nt, number of nuclei present in the top section; Nb, number of nuclei still present in the bottom section; Nv, cell numerical density. Structural characteristics calculated from these values are presented in Fig. 4. *P<0.05 vs. control rats, Student's t-test.
In vivo infusion of 5pmol/kg/min of ET-1 in the peritoneal cavity during 72h produced a significant (0.7-fold) increase in plasma ET levels (from 15.3±2.7 in control to 25.4±2.3pg/mL in ET-treated rats), whereas levels in small mesenteric arteries were more markedly elevated (≈2.5-fold, from 1.48±0.13 to 3.71±0.76pg/mg tissue, P<0.05).
4 Discussion
The major novel finding of the present study is the in vivo demonstration that ET can act both as a trophic and a mitogenic factor in the vascular wall depending on its local concentration. Indeed, intra-peritoneal infusion of ET-1 induced significant and dose-dependent leucine incorporation in the mesenteric arterial bed, supporting extensive evidence that ET is involved in small artery hypertrophy [23,24]. However, at higher doses, ET led to an acute increase in both protein and DNA synthesis. In line with these acute findings, chronic administration of a high dose of ET produced marked small artery hyperplasia, while a low dose of ET resulted in hypertrophy. These chronic observations provide pathophysiological relevance to the acute investigations and help to position previous reports into a more global perspective. Indeed, chronic Ang II or NE administration leads to an ET-dependent development of small artery hypertrophy [10,11]. During chronic NE administration, ET levels increased transiently with a peak delta increase of 0.6pg/mg (control values: 1.5±0.1, NE values: 2.1±0.1pg/mg, P<0.05) at day 3 [11]. In the present study, the delta increase from control values was 2.2pg/mg in rats treated for 3days with 5pmol/kg/min ET-1 (time (3days) was matched to facilitate the comparison between studies). Thus, the elevation of local ET concentrations with the 5pmol/kg/min dose was nearly four times higher than with exogenous NE. Interestingly, the in vivo dose–response curves to endothelin have revealed that the EC50 for hypertrophy and hyperplasia are different from a factor of approximately 4. Accordingly, chronic exposition to a dose of ET 5 times lower (1pmol/kg/min) lead to hypertrophy of small mesenteric arteries. Therefore, it appears that small variations of local ET concentrations are sufficient to induce different cellular responses in an in vivo environment.
The doses of ET were selected to be within the physiopathological range and they should not be considered irrelevant pharmacological doses. Indeed, the relative increase of ET levels was 2.5-fold (at 5pmol/kg/min) and greater than fourfold elevations were reported in lungs during congestive heart failure [25] and pulmonary hypertension [26]. In accordance with our data, pulmonary hypertension is associated with VSMC hyperplasia [26]. Moreover, the elevation of plasma concentrations was modest as compared to reports in pathological conditions. Our results also suggest that ET can achieve its maximal effect on protein and DNA synthesis within physiopathological variations of its local content, as dose–response curves were maximal around 1 and 3pmol/kg/min, respectively. In addition, our study may help to reconcile conflicting data obtained in cell culture conditions, with some studies showing hypertrophic responses (no DNA synthesis) and others reporting mitogenic responses [6,7]. Indeed, relatively small changes in experimental conditions could produce these different results.
According to our observations, at a certain level of ET receptor activation, the binding of the inhibitory element p27Kip1 to CDK2 decreases. This phenomenon is known to lead to CDK2 activation, cell cycle progression and increase in DNA synthesis, which were confirmed in the present study. Previous studies suggested that changes of CKI content could explain the regulation of the cell cycle [18,21]. In line with this, over-expression of p27Kip1 in an in vivo model of vascular injury halved intimal VSMC proliferation [19]. In our conditions, the reduced binding between CDK2 and p27Kip1 was not due to a reduction in the cellular levels of p27Kip1. The binding of another CKI, p21Cip1, was also not modified by ET. In line with our results, Suzuki et al. reported increased CDK2 activity upon ET stimulation without significant modification of p27Kip1 levels in NIH3T3 cells [6]. In addition, there was no change in the vascular expression of CDK2 by ET-1, again suggesting that transcriptional or translational regulation cannot explain our findings. Since p27Kip1 phosphorylation has been suggested to modulate its function, it was evaluated with a low and a high dose of ET. The reduced phosphorylation on Ser10 at the higher dose of ET is in line with a study by Ishida et al. suggesting that dephosphorylation of p27Kip1 at Ser10 might play an important role in progression of the cell cycle from G0–G1 to S phase, by the reducing its stability[27]. The mechanisms by which this regulation occurs is still a matter of intensive research and beyond the scope of our study [28,29]. Nevertheless, as described for classical mitogens, Ser10 phosphorylation of p27Kip1 was significantly reduced, confirming that the cells were actively engaged in the S phase of the cell cycle with 5pmol/kg/min ET-1 [29]. It is possible that other phosphorylation sites, such as threonin 198 [30], have regulatory functions in vivo, a possibility that should be explored in future studies. Nonetheless, our results provide strong in vivo evidence that p27Kip1 binding to CDK2 acts as a molecular switch to prevent or promote DNA synthesis in an ET-1 dose-dependant manner.
In an effort to determine upstream signaling elements responsible for the altered p27Kip1-CDK2 binding, we studied the PI-3K pathway, as recent evidences indicate that Akt may phosphorylate p27Kip1[30,31]. In addition, there were recent reports to suggest that ET stimulates the PI-3K/Akt pathway in many cell types, although direct evidence for VSMC is lacking [32,33]. Exogenous ET-1 increased Akt phosphorylation in small mesenteric arteries, suggesting an in vivo activation of this pathway by ET. It was then logical to test the effect of the PI-3K inhibitor (LY294002) on ET-induced protein and DNA synthesis. Although LY294002, at the dose selected, reduced ET-1 induced Akt phosphorylation in vivo, it did not blunt ET-induced protein or DNA synthesis. Thus, although the pathway is activated, it is not involved in the proliferative response to ET-1. We have previously reported that the Src-family tyrosine kinases and ERK 1/2 are not involved in the trophic effect of ET, but act as signaling molecules to induce ET-1 production when Ang II is used as the agonist [34]. In order to determine the potential involvement of another signaling pathway (protein kinase C), a high dose of tamoxifen was used. We could not detect any change in binding of p27Kip1 to CDK2 with this drug at 5pmol/kg/min ET-1. It is possible that several pathways contribute to the final response, as proposed in other conditions [35], and that the inhibition of only one may not be sufficient to affect the global response. This will deserve further investigation.
4.1 Perspectives
We cannot discard the involvement of intermediary substances that could be stimulated by ET-1 to act as co-mitogens. Indeed, it is possible that a local substance stimulated by higher concentrations of ET is producing the additional effect on p27Kip1 binding to CDK2 and DNA synthesis. It is noteworthy that the local cellular environment was most likely similar between low and high dose groups and the only change was the local concentration of ET-1. Thus, it is expected that, if applicable, a similar integration of local influences would operate in pathological conditions presenting enhanced local ET production, such as pulmonary hypertension and some forms of systemic hypertension. For that reason, more in vivo work will be necessary to improve our understanding of cellular signaling triggered by ET during vascular remodeling.
In conclusion, ET-1 is a trophic factor at low concentrations but leads to vascular hyperplasia with increasing concentrations that can be found in some pathological conditions. The switch between the two responses appears to be mediated by the regulation of p27Kip1 binding to CDK2, by mechanisms that appear PI-3K-independent and that have yet to be identified. Thus, cellular responses to vasoactive peptides, such as ET, may not be exclusive, but may diverge depending on the level of activation within a rich in vivo environment. These findings may be important to devise appropriate therapeutic strategies depending on the type of cellular response elicited by variable pathological ET over-expression.
Acknowledgements
The authors acknowledge the financial support of Canadian Institutes of Health Research (CIHR-14380). HHD, CB and PB receive a stipend from the Fonds de Recherche en santé du Québec (FRSQ), Rx and D/CIHR and Canadian Hypertension Society/CIHR, respectively. MS and PM are scholars form the FRSQ and JdeC is a JC Edwards career investigator. The authors would like to acknowledge the skilled technical assistance of Louise Ida Grondin.
References
Author notes
Time for primary review 23 days
