| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Cell Biology and Signaling |
and PKC-
in Endothelial CellsFrom the Research Division, Joslin Diabetes Center, Harvard Medical School, Boston, Mass.
Correspondence to George L. King, MD, Joslin Diabetes Center, Harvard Medical School, One Joslin Place, Room 4504, Boston, MA 02215. E-mail george.king{at}joslin.harvard.edu
Abstract
Objective— Vascular endothelial growth factor (VEGF) stimulates proangiogenic signal transduction and cell function in part through activation of protein kinase C (PKC). Our aim was to examine how individual isoforms of PKC affect VEGF action.
Methods and Results— Transfection of bovine aortic endothelial cells with small interfering RNA (siRNA) targeting either PKC-
,
, or
caused a reduction in the cognate PKC protein by 76% to 89% without changing expression of nontargeted isoforms. Downregulation of PKC-
abrogated VEGF-stimulated phosphorylation of Akt at Ser473 and eNOS at Ser1179 and decreased VEGF-stimulated NO synthase activity in intact cells. In contrast, PKC-
knockdown increased Akt and eNOS phosphorylation, whereas PKC
knockdown had no significant effect. PKC-
knockdown also decreased VEGF-stimulated Erk1/2 phosphorylation and abolished VEGF-stimulated DNA synthesis. Consistent with an effect on several pathways of VEGF signaling, VEGF receptor-2 (VEGFR2) tyrosine phosphorylation and expression of VEGFR2 protein and mRNA was decreased by 81, 90, and 84%, respectively, during knockdown of PKC-
, but increased during PKC-
knockdown.
Conclusions— By regulating VEGFR2 expression and activation, PKC-
expression is critical for activation of Akt and eNOS by VEGF and contributes to VEGF-stimulated Erk activation, whereas PKC-
has opposite effects.
Downregulation of PKC
with siRNA decreased Akt, eNOS, and Erk1/2 phosphorylation, NO synthase activity, and DNA synthesis stimulated by VEGF, likely through a dramatic decrease in VEGF receptor-2 (VEGFR2) tyrosine phosphorylation, and VEGFR2 protein and mRNA expression. In contrast, PKC-
knockdown increased VEGFR2 phosphorylation, VEGFR2 expression, and downstream signaling.
Key Words: endothelium vascular vascular endothelial growth factor A receptors vascular growth factor nitric oxide synthase type III protein kinase C
In patients with ischemia in the myocardium and lower limbs, vascular endothelial growth factor (VEGF) is a central regulator of collateral artery formation through arteriogenesis or angiogenesis.1 Expression of VEGF and its receptors are decreased in the myocardium of animal models of type 2 diabetes and in diabetic patients,2,3 and VEGF, VEGF receptors, and capillary density are decreased in animal models of obesity-associated insulin resistance.2,4 This may contribute to a more sinister course of chronic ischemia in patients with diabetes compared with nondiabetic patients or worsen prognosis during recovery from acute myocardial infarction. Therefore, there is a compelling need to identify therapeutic targets that can enhance VEGF action in patients with myocardial or peripheral ischemia, particularly in patients with diabetes.
VEGF-stimulated angiogenesis is mediated by several pathways of intracellular signaling. VEGF phosphorylates Akt through activation of phosphatidylinositol 3-kinase and, in turn, Akt phosphorylates several substrates, many of which have been shown to be important for VEGF-stimulated angiogenesis. Among them are forkhead transcription factors, including FoxO1 and FoxO3a,5–7 and glycogen synthase kinase-3β (GSK3β).8 Furthermore, Akt activates eNOS through phosphorylation on serine S1179.9,10 VEGF-stimulated cell proliferation and angiogenesis involve several other signaling pathways than Akt, among them activation of Erk1/2.11,12
It has long been known that phorbol esters, which are synthetic analogs of diacylglycerol and activators of protein kinase C (PKC), can promote angiogenesis.13 VEGF activates PKC, and PKC activity contributes to VEGF-stimulated endothelial cell proliferation.11,14 However, limited information is available about the isoforms of PKC that are important for VEGF action. In the present study, our aim was to characterize how individual PKC isoforms affect VEGF signaling and VEGF-stimulated cell function. Using RNA interference, we show here that PKC-
is critical for Akt and eNOS activation induced by VEGF, and that VEGF-stimulated Erk phosphorylation and DNA synthesis are significantly dependent on PKC-
expression. The influence of PKC-
expression on both Akt and Erk signal transduction pathways, often presumed to have limited cross-talk, is explained by the finding that downregulation PKC-
dramatically decreases tyrosine phosphorylation and expression of VEGF receptor-2 (VEGFR2) also known as kinase insert domain-containing receptor (VEGFR2/KDR) protein and mRNA. In contrast, PKC-
negatively regulates VEGF signaling and VEGFR2 expression and activation.
Methods
A detailed methods description is included in the supplemental materials (please see http://atvb.ahajournals.org). Experiments were performed in primary bovine aortic endothelial cells (BAECs). Cytosol and membrane fractions of cell lysate were prepared by ultracentrifugation. BAECs were transfected with 21-mer small interfering RNA (siRNA) complexed to Lipofectamine 2000 (Invitrogen, NO synthase activity was assayed in intact BAEC culture as previously described15 with modifications. DNA synthesis was measured by labeling with 5-bromo-2'-deoxyuridine (BrdU) using a commercially available kit (Roche Applied Science). mRNA was measured by real-time polymerase chain reaction (PCR) using TaqMan (Applied Biosystems).
Results
Endothelial cells express several isoforms of PKC, including PKC-
, β1, β2,
,
,
, and
, depending on species and vascular bed.16,17 The known proangiogenic effects of phorbol ester is likely caused by activation of conventional and novel PKC isoforms.18 We therefore characterized the presence of conventional and novel PKC-
, β1, β2,
,
, and
in BAECs used for the present study. PKC-
,
, and
were detected at the protein level in whole cell lysate (Figure 1). When bovine brain was used as a positive control, reverse transcription real-time PCR with primers targeting the sequence common to PKCβ1 and β2 amplified isolated RNA, and Western blotting detected PKCβ1 and β2 protein in tissue lysate (data not shown). However, PKCβ mRNA was not detected in RNA isolated from BAECs, and PKCβ1, β2, or
protein was not detected in whole cell lysate (results not shown).
|
To specifically inhibit individual PKC isoforms, we designed siRNA targeting bovine PKC-
,
, and
. Transfection of BAECs with these siRNAs downregulated PKC-
,
, and
protein levels by 76±3, 89±3, and 76±2%, respectively (P<0.005, Figure 1). Importantly, each siRNA only changed the expression of the targeted isoform and not any of the two other isoforms (Figure 1). Unstimulated Akt phosphorylation was not changed by PKC-
or
knockdown, but decreased by 47±8% during PKC-
knockdown (P<0.001, Figure 2a and 2b). Unexpectedly, PKC-
knockdown completely inhibited VEGF-stimulated Akt phosphorylation (1.7±0.1-fold increase in the control siRNA condition, P<0.001; a nonsignificant 1.1±0.2-fold increase during PKC-
knockdown; Figure 2). In contrast, PKC-
knockdown increased VEGF-stimulated Akt phosphorylation (1.7±0.1 and 2.3±0.4-fold during control and PKC-
knockdown conditions, respectively, P<0.05; Figure 2). PKC
knockdown did not significantly change VEGF-stimulated Akt phosphorylation (Figure 2). Downregulation of PKC isoforms selectively affected VEGF signaling, because PKC-
,
, or
knockdown did not change Akt phosphorylation stimulated by insulin (supplemental Figure I) or insulin-like growth factor (data not shown).
|
We next evaluated the effect of knockdown of individual PKC isoforms on VEGF-stimulated eNOS phosphorylation. Basal levels of eNOS Ser1179 phosphorylation was not significantly changed by knockdown of PKC-
,
, or
(Figure 2a and 2b). However, VEGF-stimulated eNOS phosphorylation was increased during PKC-
knockdown, unchanged during PKC
knockdown, and completely inhibited by PKC-
knockdown (5.2±0.5, 6.1±0.1, 4.2±0.3, and 1.7±0.3-fold increase in the control, PKC-
, PKC
, and PKC-
siRNA conditions, respectively; maximal levels as well as fold-increase of eNOS phosphorylation statistically different from the control condition during knockdown of PKC-
and
(P<0.05), but not
; Figure 2). In turn, changes in VEGF-stimulated NO synthase activity, measured in intact cell cultures by conversion of 3H-L-arginine to 3H-L-citrulline, largely paralleled Akt and eNOS phosphorylation, although with less pronounced effects. Thus, PKC-
knockdown decreased VEGF-stimulated NO synthase activity by 54±12% (P<0.05, Figure 3a). PKC-
knockdown increased (by 14±12%) and PKC
knockdown decreased (by 15±18%) VEGF-stimulated NO synthase activity, although these changes were not statistically significant (Figure 3a).
|
The involvement of PKC-
in VEGF signaling was not confined to Akt/eNOS signaling. VEGF increased phosphorylation of Erk1/2 by 10±1.8-fold after transfection of BAECs with control siRNA (at Thr203/Tyr205 in bovine Erk1/p44 MAPK, homologous to residues Thr202/Tyr204 in human Erk1 and Thr185/Tyr187 in bovine Erk2; Figure 2a and 2b). However, PKC-
knockdown decreased both maximal levels and the increase in Erk1/2 phosphorylation stimulated by VEGF (2.7±0.9-fold increase during PKC-
knockdown, P<0.05 compared to the control siRNA condition, Figure 2). During PKC-
knockdown, unstimulated Erk1/2 phosphorylation decreased to 55±17% of the level during the control siRNA condition (P<0.05, Figure 2). The maximal level of Erk phosphorylation stimulated by VEGF was not changed by PKC-
knockdown, but the relative increase was greater due to the reduction in the unstimulated condition (19.8±2.1-fold increase during PKC-
knockdown). Unstimulated Erk1/2 phosphorylation was not significantly changed during PKC
and
knockdown, and VEGF-stimulated Erk1/2 phosphorylation was not changed by PKC
knockdown (Figure 2).
To evaluate cellular function known to be regulated by Erk signaling, we measured DNA synthesis by BrdU incorporation. PKC-
knockdown increased BrdU incorporation without VEGF stimulation to 239% of the control value (P<0.001, Figure 3b), but the unstimulated BrdU incorporation did not change during PKC
or
knockdown. VEGF increased BrdU incorporation by 34% in the control siRNA condition (P<0.01, Figure 3b) and by 14% and 39% during PKC-
and
knockdown (P<0.05, Figure 3b). In contrast, PKC-
knockdown completely inhibited VEGF-stimulated BrdU incorporation (Figure 3b).
The findings that Akt and Erk signal transduction were both affected by PKC-
knockdown pointed to a role for PKC-
early in VEGF signal transduction. We therefore studied VEGFR2 activation, which is responsible for most known actions of VEGF, including activation of eNOS.1 PKC-
knockdown decreased VEGF-stimulated VEGFR2 tyrosine phosphorylation by 81±6% compared to transfection with control siRNA (P<0.001, Figure 4), whereas PKC-
and
knockdown increased tyrosine phosphorylation by 38±18 and 33±12%, respectively (P<0.05, Figure 4). These changes reflected changes in VEGFR2 protein expression. PKC-
knockdown caused a dramatic reduction in VEGFR2 protein to only 10±2% of the control value (P<0.001, Figure 4). In contrast, PKC-
and
increased VEGFR2 protein by 31±6 and 46±2%, respectively (P<0.05, Figure 4). VEGF stimulation caused a 12±3% decrease in VEGFR2 protein in cells transfected with control siRNA, likely representing ligand-stimulated downregulation of the receptor (P<0.05, Figure 4). A VEGF-stimulated decreased in VEGFR2 protein (by 17±3%) was preserved during PKC-
knockdown, but not detectable during PKC
or
knockdown.
|
In VEGFR2 immunoblots (Figure 4), the less intense slower migrating (200 kDa) band may represent an intermediary cytosolic form, whereas the more intense upper (230 kDa) band may represent the fully N-glycosylated receptor expressed at the cell surface.19 Of note, PKC-
knockdown reduced both bands, suggesting that it affects transcription, translation, or early posttranslational modification rather than promoting internalization and degradation of the surface-bound receptor.20 Therefore, we measured VEGFR2 mRNA expression with real-time PCR during knockdown of PKC-
,
, and
. A second siRNA targeting a different region of PKC-
mRNA was used to minimize the possibility that the effect on VEGFR2 expression was attributable to an off-target effect. Compared with the control siRNA condition, VEGFR2 mRNA expression decreased by 84±7 and 83±6% during PKC-
knockdown with the original and additional siRNA, respectively (P<0.001, Figure 5a). Thus, PKC-
appears to regulate VEGFR2 at the mRNA level. PKC-
knockdown had no significant effect, whereas PKC
increased VEGFR2 mRNA by 16±1% (P<0.001, Figure 5a).
|
VEGF signal transduction during PKC knockdown could change because VEGFR2 expression results in a secondary change in VEGFR1/Fms-like tyrosine kinase-1 (Flt-1) expression. Indeed, PKC-
and PKC
knockdown increased VEGFR1 mRNA expression to 240±20 and 178±1% of the control siRNA condition, respectively (P<0.01, Figure 5b). However, VEGFR1 mRNA did not change significantly during PKC-
knockdown with either of the two siRNAs (Figure 5b). We could not detect VEGFR1 protein or VEGFR1-mediated signal transduction by Western blotting in these cells. VEGFR1 protein was not detectable in whole cell lysate (using two different commercially available antibodies), and VEGFR1 or phosphotyrosine was not detectable in immunoprecipitates with VEGFR1 antibody with or without stimulation with VEGF; furthermore, placental growth factor, which activates VEGFR1 but not VEGFR2, did not increase phosphorylation of eNOS, Akt, Erk, or p38 (data not shown). Thus, VEGFR1 protein is likely expressed at a very low level compared to VEGFR2 in BAECs. Changes seen in VEGF signal transduction during PKC-
and
knockdown could be influenced by upregulation of VEGFR1, but there is no indication that changes in VEGF signaling during PKC-
knockdown can be attributed to changes in VEGFR1 mRNA expression.
Discussion
It is well-recognized that PKC activation is proangiogenic and that PKC mediates several actions of VEGF, including activation of eNOS. However, the involvement of specific isoforms of PKC in VEGF action has been insufficiently described. In the present study, using siRNA-mediated specific downregulation of PKC isoforms, we have demonstrated for the first time that expression of PKC-
is critical for VEGF-stimulated phosphorylation of Akt and eNOS and for catalytic activity of NO synthase in endothelial cells. Furthermore, PKC-
was shown to contribute to VEGF-stimulated Erk phosphorylation and DNA synthesis. In contrast, experiments with RNA interference indicated that PKC-
inhibited VEGF-stimulated Akt and eNOS phosphorylation and negatively regulated DNA synthesis. These changes in signal transduction and cell function can be explained by the finding that VEGFR2 expression and activation appears to be critically dependent on PKC-
, but inhibited by PKC-
. Importantly, PKC-
and
downregulation selectively affected VEGF signaling, as insulin-stimulated Akt phosphorylation was unaffected by PKC knockdown.
Although PKC-
was previously shown to translocate from the cytosol to the membrane fraction in human umbilical vein endothelial cells during VEGF stimulation,11 a role for PKC-
in regulating VEGFR2 expression or downstream VEGF signaling has not been reported before. VEGFR2 downregulation during PKC-
knockdown could be regulated at many different levels, including regulation of transcription21,22 or mRNA stability, or through internalization and degradation of VEGFR2 protein after ubiquitinylation.23–25 For example, activating protein-1 (AP-1), nuclear factor-
B (NF-
B), and Sp1 have been implicated in regulation of VEGFR2 expression, and PKC-
has been shown to regulate all of these transcription factors during hypoxia.26,27 Regulation of VEGFR2 at the level of transcription or mRNA stability seem likely because PKC-
knockdown caused a large reduction in VEGFR2 mRNA, and because a decreased intensity was seen in the 230-kDa band on VEGFR2 immunoblots and in a 200-kDa isoform which may represent a cytosolic immature form of VEGFR2 (Figure 4).19 Studies are ongoing in our laboratory to test these possibilities.
It is likely that PKC-
has significant effects in vivo on vascular function. In PKC-
knockout mice, chronic hypoxia was reported to result in excessive pulmonary hypertension and right ventricular hypertrophy compared to wild-type mice.30 Furthermore, PKC-
knockout mice lacked the induction of eNOS gene expression observed in wild-type mice, suggesting a role for PKC-
in regulating eNOS expression.30 In the present study, although PKC-
knockdown caused a dramatic reduction in VEGFR2 mRNA and protein expression and near-complete inhibition of VEGF-stimulated Akt and eNOS phosphorylation, NO synthase activity and Erk phosphorylation were only moderately affected. However, certain signaling pathways and biological actions of VEGF may be partially preserved despite substantial reduction of VEGFR2 expression. This is suggested by the fact that mice with heterozygous deletion of the VEGFR2 gene are phenotypically normal, whereas homozygous deletion of the VEGFR2 gene is embryonically lethal.31
Our present data demonstrating that PKC-
knockdown increased VEGF-stimulated Erk phosphorylation and DNA synthesis are consistent with a previous publication from our group showing that downregulation of PKC-
protein levels with antisense oligonucleotides increased the VEGF-stimulated growth of BAECs.14 This effect may be specific for VEGF action. Thus, PKC-
had the opposite effect on signal transduction stimulated by fibroblast growth factor-2 (FGF2) in a previous study,33 where FGF2-stimulated eNOS Ser1179 phosphorylation increased during PKC-
overexpression and decreased after PKC-
downregulation.33 We previously reported that adenovirus-mediated expression of PKCβ1 and β2 inhibited VEGF-stimulated Akt activation in endothelial cells.32 It is likely that PKCβ isoforms can inhibit VEGF-stimulated Akt activation if they are present at certain expression levels, but in the present study the BAECs used did not contain detectable levels of the β isoforms.
PKC
knockdown increased expression and VEGF-stimulated tyrosine phosphorylation of VEGFR2, but did not significantly affect downstream effects in the form of phosphorylation of Akt, eNOS, or Erk, or increase in NO synthase activity or DNA synthesis. However, PKC
knockdown inhibited VEGF-stimulated reduction of VEGFR2 protein expression. A previous publication has shown that PMA treatment of endothelial cells causes downregulation of VEGFR2, and that the receptor is internalized and degraded after phosphorylation of serine 1188 and 1191 in its carboxy terminus.28 It is possible that PKC
is necessary for this process. As internalization of the receptor has been shown to be necessary for full effects of VEGF stimulation,29 PKC
knockdown may have complex effects on VEGFR2 expression and downstream signaling, which should be characterized in future studies.
In conclusion, this study has documented that PKC-
expression is critical for VEGF-stimulated phosphorylation and activation of Akt and eNOS, and that VEGF-stimulated Erk phosphorylation and DNA synthesis are significantly dependent on PKC-
expression. In contrast, PKC-
has inhibitory effects on VEGF-stimulated phosporylation of Akt and eNOS and on DNA synthesis. These effects can be explained by a critical role for PKC-
in maintaining VEGFR2 gene expression, and by a negative modulation of VEGFR2 expression by PKC-
. These findings have potential implications for prevention and treatment of disease, because enhancing PKC-
activity may be of use in promoting VEGF-stimulated angiogenesis in patients with ischemia in the myocardium or lower limbs, whereas inhibiting PKC-
may find use in preventing diabetic retinopathy or tumor angiogenesis.
Acknowledgments
Sources of Funding
The Danish Medical Research Council (grant 22-01-0498), the Danish Heart Foundation (grant 01-2-2-79-22946), and a Mary K. Iacocca Research Fellowship supported C.R.-M. This study was also supported by NIH grant NIDDK R01 DK53105 to G.L.K.
Disclosures
None.
Footnotes
Original received September 9, 2007; final version accepted February 25, 2008.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2008 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |