HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) All Versions of this Article: 17/3/449 02-0398fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by SCHÄFER, M. Articles by NOLL, TH. Search for Related Content PubMed PubMed Citation Articles by SCHÄFER, M. Articles by NOLL, TH.

Signaling of hypoxia-induced autonomous proliferation of endothelial cells¹

Institute of Physiology, Justus-Liebig-University, D-35392 Giessen, Germany

²Correspondence: Physiologisches Institut, Justus-Liebig-Universität, Aulweg 129, D-35392 Giessen, Germany. E-mail: Michael.Piper{at}physiologie.med.uni-giessen.de

SPECIFIC AIMS

The present study was designed to analyze the ability of endothelial cells to proliferate on transient exposure to hypoxia by an autonomous response mechanism, i.e., without influence of paracrine effectors such as VEGF. Existence of such an autonomous proliferative response has been identified in cell cultures, but its signaling mechanism is unknown. We investigated the roles of the MEK/MAPK pathway and of cytosolic Ca²⁺ in the signaling mechanism.

PRINCIPAL FINDINGS

1. Transient hypoxia or metabolic inhibition induce endothelial cell proliferation
After serum-free incubation for 24 h, transient hypoxia for 1 h resulted in increased proliferation of endothelial cells, quantified after another 24 h normoxic serum-free postincubation period (Fig. 1 ). This proliferative effect was accompanied by an increase in ³H-thymidine incorporation, which indicates increased DNA synthesis (not shown). Rotenone (10 µM), an inhibitor of complex I of the mitochondrial respiratory chain, mimicked the proliferative effect of hypoxia. This indicates that the proliferative response to hypoxia is not due to depletion of oxygen per se but to inhibition of mitochondrial electron transport. To alter the energy balance of the cells, we stimulated glycolysis by adding glucose (15 mM) 30 min before and during exposure to hypoxia or rotenone. This treatment totally inhibited hypoxia- or rotenone-induced cell proliferation and DNA synthesis (n.s. vs. normoxic control). The finding indicates that the proliferative response to hypoxia depends on the loss of energy.

View larger version (26K): [in this window] [in a new window]   Figure 1. Effect of hypoxia (open bars) or rotenone (filled bars; 10 µM) on endothelial cell proliferation. Cell numbers are given as % of controls not exposed to hypoxia or rotenone. Preincubation of monolayers with 8-phenyl-theophillin (8-PT; 10 µM) or a VEGF-neutralizing antibody (0.2 µg/mL) had no effect; preincubation with the MEK inhibitor PD 98059 (PD; 20 µM) diminished the proliferative response. Data are means ± SE of n = 16 culture dishes of 4 independent cell preparations. *P < 0.05 vs. control. #P < 0.05 vs. hypoxia or rotenone alone.

To exclude side effects of HEPES-buffered media, the influence of hypoxia or rotenone was analyzed under bicarbonate-buffered conditions. The proliferative response was virtually identical to that observed in HEPES-buffered media (not shown).

Metabolically inhibited endothelial cells may release adenosine or VEGF and these may act as autocrine stimuli. First, we tested whether hypoxia-induced cell proliferation is sensitive to 8-phenyl-theophyllin (8-PT, 10 µM), an adenosine receptor antagonist. Whereas 8-PT abolished endothelial cell proliferation due to exogenously applied adenosine (10 µM for 1 h) in normoxic control experiments (not shown), 8-PT had no effect on hypoxia- or rotenone-induced proliferation (Fig. 1) . Second, we tested whether the presence of a polyclonal neutralizing anti-human-VEGF-antibody (0.2 µg/mL) affected hypoxia- or rotenone-induced proliferation (Fig. 1) . It did not, but the antibody blocked completely the proliferative effect of exogenously supplied VEGF in normoxic controls (10 ng/mL administered for 1 h; not shown). This indicates that the autonomous proliferative response of endothelial cells to hypoxia is independent of autocrine actions of either adenosine or VEGF. We then analyzed the role of the MEK/MAPK pathway. We administered PD 98059 (PD, 20 µM) or UO 126 (UO, 10 µM), both inhibitors of MEK, the kinase upstream of p42/p44 MAPK. PD (Fig. 1) or UO (UO: hypoxia+UO: 99±1% of cells/dish in controls, n=16; n.s. vs. control) diminished hypoxia- or rotenone-induced proliferation, thus indicating involvement of the MEK/MAPK pathway.

2. Hypoxia leads to activation and nuclear translocation of p42 MAPK
Since the inhibitor of MEK diminished the proliferative response, we analyzed activation of a downstream target of MEK, i.e., p42 MAPK (ERK 2). As a parameter of activation, phosphorylation of the protein was determined by Western blot analysis. Before hypoxia, only a small percentage of p42 MAPK was phosphorylated (11±5%; n=6 experiments). During 5 min hypoxia it became rapidly phosphorylated (51+7%; n=6; P<0.05 vs. control). Phosphorylation was transient and declined to baseline level within 1 h. This activation pattern was also observed when rotenone was applied. Administration of PD (20 µM) totally abolished p42 MAPK activation in either case. Activation of p42 MAPK in the presence of hypoxia or rotenone was not suppressed by addition of the anaerobic glycolytic substrate glucose (15 mM; not shown). Subcellular localization of p42 MAPK was monitored by immunocytochemistry. In the unstimulated control situation, the kinase was located in the perinuclear region. After 30 min of oxygen deprivation, nuclei exhibited a marked staining for p42 MAPK, indicating a nuclear translocation. Administration of rotenone induced a similar intracellular distribution of p42 MAPK. Nuclear translocation of p42 MAPK was verified by differential centrifugation of subcellular fractions (see below).

3. Hypoxia evokes an early release and subsequent influx of Ca²⁺
Hypoxia elicited a biphasic rise of cytosolic Ca²⁺ determined by Fura-2 fluorescence. Cytosolic Ca²⁺ (normoxic control [Ca²⁺]_i: 69.2±6.5 nM, n=60 cells) peaked at 2.5 min after onset of hypoxia ([Ca²⁺]_i: 153.8±11.6 nM, n=60; P<0.05 vs. control), then declined transiently before it rose again steadily ([Ca²⁺]_i after 1 h hypoxia: 335.3±28.6 nM, n=60; P<0.05 vs. control). In normoxic controls, cytosolic Ca²⁺ was unchanged throughout. The presence of glucose (15 mM) abolished the hypoxia-induced cytosolic Ca²⁺ rise ([Ca²⁺]_i after 1 h of hypoxia: 66.7±11 nM, n=60; n.s. vs. normoxic control), indicating its energy dependence. Removal of extracellular Ca²⁺ or addition of Ni²⁺, an inhibitor of capacitative Ca²⁺ influx, did not affect the early increase in cytosolic Ca²⁺ vs. hypoxia alone, but reduced the delayed rise of cytosolic Ca²⁺ ([Ca²⁺]_i after 1 h of hypoxia: Ca²⁺ free: 10.5±17 nM; with Ni²⁺: 114.9±19.4 nM; n=60; P<0.05 vs. hypoxia alone). Inhibition of IP₃-dependent Ca²⁺ release from endoplasmic reticulum with use of the specific inhibitor xestospongin C (XeC; 3 µM) abolished the first and attenuated the delayed Ca²⁺ rise ([Ca²⁺]_i after 1 h of hypoxia; 136.0±15.9 nM; n=60; P<0.05 vs. hypoxia alone). These results indicate that under hypoxia the early transient rise of Ca²⁺ is due to IP₃-sensitive Ca²⁺ release from endoplasmic reticulum; the later rise is evoked by Ca²⁺ influx from the extracellular space. The same behavior was observed in experiments with rotenone.

4. Hypoxic Ca²⁺elevation is not involved in activation of p42 MAPK activation but permits nuclear translocation and proliferation
We analyzed p42 MAPK activation under each condition where the hypoxic rise of cytosolic Ca²⁺ was suppressed. We found that p42 MAPK phosphorylation in hypoxia was not affected by the Ca²⁺-modulating protocols. The results indicate that an increase in cytosolic Ca²⁺ is not required for activation of the p42 MAPK pathway under hypoxia or rotenone application. However, the described Ca²⁺-modulating interventions diminished the proliferative response to transient hypoxia or presence of rotenone, indicating the contribution of a Ca²⁺-sensitive step (Fig. 2 ). Comparison with the pattern of p42 MAPK activation reveals that this Ca²⁺-sensitive step is not upstream of p42 MAPK. We evaluated nuclear vs. cytosolic distribution of p42 MAPK by differential centrifugation. Only 17 ± 4% (n=4) of p42 MAPK was found in the nuclear fraction in untreated control cultures. After 30 min of hypoxia, the nuclear content of p42 MAPK was doubled (32.6±6.3%; n=4; P<0.05 vs. control). Conditions reducing the delayed cytosolic Ca²⁺ rise during hypoxia abolished the translocation of p42 MAPK into the nuclear fraction (n=4; n.s. vs. control). The results indicate, therefore that the delayed but sustained rise in Ca²⁺ triggers nuclear translocation of p42 MAPK.

View larger version (24K): [in this window] [in a new window]   Figure 2. Effect of modulation of the hypoxia- (open bars) or rotenone-induced (10 µM, filled bars) Ca2+ response on endothelial cell proliferation. Cell numbers are given as % of controls not exposed to hypoxia or rotenone. Removal of extracellular Ca2+ (-Ca2+), incubation with xestospongin C (XeC; 3 µM) or Ni2+ (1 mM) diminished endothelial cell proliferation. Data are means ± SE of n = 16 culture dishes of 4 independent cell preparations. *P < 0.05 vs. control. #P < 0.05 vs. hypoxia or rotenone alone.

CONCLUSIONS

We have investigated the autonomous proliferative response of endothelial cells to transient hypoxia or administration of rotenone under normoxic conditions. We found it to depend on activation of p42 MAPK, which is Ca²⁺ independent, and nuclear translocation of p42 MAPK, which requires a sustained rise of cytosolic Ca²⁺. The scheme of signaling is given in Fig. 3 .

View larger version (12K): [in this window] [in a new window]   Figure 3. Scheme of intracellular signaling promoting the autonomous proliferative response of endothelial cells to hypoxia.

All investigated effects of hypoxia could be mimicked by application of the blocker of mitochondrial complex I, rotenone, under normoxic conditions. This shows that these effects are not due to nonmitochondrial oxygen sensing but to mitochondrial inhibition. Experiments in which the potential roles of VEGF or adenosine were tested showed that these known autocrine/paracrine activators of endothelial cell growth are not involved in the autonomous proliferative response to hypoxia in the investigated model.

During hypoxia, p42 MAPK was transiently activated, which could be inhibited by the MEK inhibitor PD 98059 or UO 126. Blockade of the MEK/MAPK pathway abolished the proliferative response to hypoxia and demonstrated that this pathway is essential. p42 MAPK activation was not inhibited by application of the anaerobic energy substrate glucose nor was it dependent on changes in cytosolic Ca²⁺ homeostasis. p42 MAPK activation must therefore be due to a factor produced by inhibition of mitochondrial respiratory chain other than energy loss or subsequent Ca²⁺ overload.

Nuclear translocation but not the activation of p42 MAPK itself was suppressed when the hypoxic rise of cytosolic Ca²⁺ was prevented. The changes in cytosolic Ca²⁺ observed in the present study reveal an early release of Ca²⁺ from the endoplasmic reticulum, followed by a delayed but progressive Ca²⁺ influx of external Ca²⁺. It was common to all Ca²⁺-modulating interventions that prevented p42 MAPK translocation that the delayed Ca²⁺ rise was suppressed. Therefore, it can be concluded that this part of the hypoxic Ca²⁺ rise triggers nuclear p42 MAPK translocation. The presence of glucose prevented hypoxic Ca²⁺ changes and p42 MAPK translocation. Both effects are thus dependent on energy loss under hypoxia.

Some details of the present work deserve further analysis: the exact mechanism by which hypoxic Ca²⁺ overload promotes nuclear translocation of p42 MAPK and the nature of the mitochondrial factor promoting activation of the MEK/MAPK pathway under hypoxia. Concerning the latter, one may speculate that reactive oxygen species are involved. In summary, the data show that the investigated autonomous proliferative response of endothelial cells to hypoxia depends on two different signaling mechanisms, both initiated by mitochondrial inhibition: one leads to p42 MAPK activation and is independent of energy loss and Ca²⁺ rise; a second causes p42 MAPK translocation and is dependent on the cytosolic Ca²⁺ rise induced by energy loss.

FOOTNOTES

¹ To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.02-0398fje; to cite this article, use FASEB J. (January 21, 2003) 10.1096/fj.02-0398fje

This article has been cited by other articles:

				Y. Abdallah, D. Gligorievski, S. A. Kasseckert, L. Dieterich, M. Schafer, C. R. Kuhlmann, T. Noll, H. Sauer, H.M. Piper, and C. Schafer The role of poly(ADP-ribose) polymerase (PARP) in the autonomous proliferative response of endothelial cells to hypoxia Cardiovasc Res, February 1, 2007; 73(3): 568 - 574. [Abstract] [Full Text] [PDF]

				S. M. O'Reilly, M. O. Leonard, N. Kieran, K. M. Comerford, E. Cummins, M. Pouliot, S. B. Lee, and C. T. Taylor Hypoxia induces epithelial amphiregulin gene expression in a CREB-dependent manner Am J Physiol Cell Physiol, February 1, 2006; 290(2): C592 - C600. [Abstract] [Full Text] [PDF]

				Y. Ben-Yosef, A. Miller, S. Shapiro, and N. Lahat Hypoxia of endothelial cells leads to MMP-2-dependent survival and death Am J Physiol Cell Physiol, November 1, 2005; 289(5): C1321 - C1331. [Abstract] [Full Text] [PDF]

				D. J. Manalo, A. Rowan, T. Lavoie, L. Natarajan, B. D. Kelly, S. Q. Ye, J. G. N. Garcia, and G. L. Semenza Transcriptional regulation of vascular endothelial cell responses to hypoxia by HIF-1 Blood, January 15, 2005; 105(2): 659 - 669. [Abstract] [Full Text] [PDF]

				S.-J. Gao, J.-H. Deng, and F.-C. Zhou Productive Lytic Replication of a Recombinant Kaposi's Sarcoma-Associated Herpesvirus in Efficient Primary Infection of Primary Human Endothelial Cells J. Virol., September 15, 2003; 77(18): 9738 - 9749. [Abstract] [Full Text] [PDF]

This Article Full Text (PDF) All Versions of this Article: 17/3/449 02-0398fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by SCHÄFER, M. Articles by NOLL, TH. Search for Related Content PubMed PubMed Citation Articles by SCHÄFER, M. Articles by NOLL, TH.