Erythropoietin (EPO) regulates the production of red blood cells primarily by preventing apoptosis of erythroid progenitors. More recently, however, EPO has emerged as a major cytoprotective cytokine in several nonhemopoietic tissues in the setting of stress or injury. The underlying mechanisms of the protective responses of EPO have not been fully defined. Here we show that EPO triggers a phosphatidylinositol 3-kinase-(PI3K)-dependent survival pathway that counteracts endothelial cell death. The protection conferred by PI3K relies on the subsequent induction of Bcl-xL, a prosurvival member of the Bcl-2 protein family. In addition, EPO counteracts the upregulation of the pro-apoptotic BH3-only protein BIM, which is induced by serum withdrawal. EPO also activates extracellular signal-regulated kinase 1 and 2 (ERK1/2), which are involved in a Bcl-xL-independent cytoprotective pathway. EPO caused a prolonged activation of nuclear factor (NF)-κB, which was blocked by inhibition of PI3K, but not by inhibition of mitogen-activated protein (MAP)/ERK kinase (MEK), suggesting that EPO-activated NF-κB requires PI3K activity. However, the activation of the NF-κB pathway was not required for the ability of EPO to counteract endothelial apoptosis. Thus EPO promotes survival of endothelial cells through PI3K-dependent Bcl-xL-induction and BIM regulation, as well as through a separate mechanism involving the ERK pathway.
- human umbilical vein endothelial cells
- extracellular signal-regulated kinase
- pro-apoptotic BH3-only protein
erythropoietin (EPO) is a glycoprotein hormone which has a prominent role in erythropoiesis. EPO is produced mainly by the adult kidney in response to hypoxia. Although, it has been documented to effectively promote survival, proliferation, and differentiation of erythroid progenitors, its role is to prevent apoptosis of late erythroid progenitors, which is now considered the primary mechanism of EPO action (46). The anti-apoptotic function of EPO has been the subject of many studies performed in the past few years (20, 24). Recent findings indicate that the anti-apoptotic activity of EPO is not restricted to erythroid progenitor cells but extends to other tissues such as the brain, retina, spinal cord, peripheral nerves, kidney, heart, and vasculature (3, 10, 11, 17, 26, 43, 45). EPO has therefore emerged as a remarkable anti-apoptotic cytokine against potential stress or toxic stimuli.
With respect to the vasculature, EPO acts on both endothelial and smooth muscle cells. EPO receptors are expressed on human vascular endothelial cells from coronary, pulmonary, and cerebral arteries, as well as umbilical vein and dermal vessels (2, 4). The endothelium acts as a barrier and interface between blood and tissue and provides the route for systemically administered EPO to pass into organs. It is therefore important to define the effect of EPO on the endothelium.
Several signal transduction pathways, including the phosphatidylinositol 3-kinase (PI3K) and extracellular signal-regulated kinase 1 and 2 (ERK1/2) pathways, are known to be involved in the anti-apoptotic functions of EPO in erythroblasts (9, 21, 28, 46). In contrast, EPO-mediated neuronal survival involves cross talk between the nuclear factor (NF)-κB and Janus kinase 2 signaling pathways (15). However, the major anti-apoptotic pathway in endothelial cells has not been elucidated. In light of this tissue-protective function of EPO, we set out to determine the mechanisms signaling survival that are evoked by EPO in endothelial cells. Our results show that EPO inhibits apoptosis in endothelial cells by PI3K-mediated upregulation of Bcl-xL and by an independent mechanism involving the activity of the ERK pathway. In addition, EPO represses the induction of the pro-apoptotic Bcl-2 family member BIM in response to serum withdrawal. Our findings implicate a critical role for EPO in balancing Bcl-2 family member expression to prevent cell death.
MATERIALS AND METHODS
Human recombinant EPO was purchased from Roche Applied Science (Laval, QC, Canada). Rabbit anti-Akt, anti-phospho-Akt (Ser-473), anti-Bcl-xL anti-phospho-ERK1/2 antibodies, LY-294002, and U0126 were from Cell Signaling Technology (Beverly, MA). Antibodies recognizing ERK1/2 were purchased from Stressgen (Victoria, BC, Canada). Lipofectin, human endothelial serum-free medium (SFM) basal growth medium, Opti-MEM I, l-glutamine, streptomycin, and penicillin were purchased from Invitrogen (Carlsbad, CA). MCDB131 medium and heparin came from Sigma (St. Louis, MO). Recombinant tumor necrosis factor-α (TNF) was purchased from R&D Systems (Minneapolis, MN). The NF-κB inhibitor SN50 and the control peptide SN50M were obtained from Biomol (Plymouth Meeting, PA). The dual-luciferase reporter assay kit and the Renilla luciferase control reporter plasmid pRL-CMV were purchased from Promega (Madison, WI). The NF-κB reporter plasmid pNF-κB-luc, which contains five tandem NF-κB-binding sites upstream of the luciferase gene, was a gift from Dr. Frank Jirik (University of Calgary, Calgary, AB, Canada).
Human umbilical vein endothelial cells (HUVEC) were obtained from collagenase-digested umbilical veins and cultured in MCDB131 medium supplemented with 20% heat-inactivated bovine calf serum (HyClone Laboratories, Logan, UT), endothelial cell growth supplement (20 μg/ml, Calbiochem, La Jolla, CA), heparin (90 μg/ml), l-glutamine (2 mM) in the presence of penicillin (50 U/ml), and streptomycin (50 μg/ml) in a 5% CO2 environment at 37°C as previously described (27).
Transient transfection of plasmid vectors was performed with Lipofectin in accordance with the manufacturer's instructions. Briefly, 60,000 cells were seeded per well of a 24-well plate 1 day before transfection. For each well, 4 ng of the plasmid pRL-CMV and 210 ng of pNF-κB-luc were diluted to 20 μl with OPTI-MEM I. One microliter of Lipofectin was diluted to 20 μl with OPTI-MEM I. Both solutions were incubated for 45 min at room temperature and mixed. After 20 min, 160 μl OPTI-MEM I was added and the mixture added to cells. After 3 h, the transfection mixture was replaced with human endothelial-SFM basal growth medium containing 0.5% calf serum. Sixteen hours later, cells were used for various experiments. Cell extracts were prepared, and the relative luciferase activity was determined according to the manufacturer's instructions.
Western blot analysis.
Two hundred and fifty thousand cells were seeded per well of a 12-well plate. After attachment, the growth medium was replaced with MCDB131 medium containing 1% calf serum and 0.5% bovine serum albumin. Sixteen hours later cells were subjected to various treatments. Cells were washed twice with cold phosphate-buffered saline and lysed in SDS sample buffer containing 2% SDS and 100 mM dithiothreitol. Samples were sonicated and electrophoresed on 10% polyacrylamide gels. After transfer of proteins to nitrocellulose, membranes were blocked with 5% nonfat milk and incubated with various antibodies according to the manufacturer's instructions. Detection was performed using horseradish peroxidase-coupled secondary antibodies and SuperSignal West Pico chemiluminescent system (Pierce, Rockford, IL).
For determination of viability, HUVEC were seeded at a density of 10,000 cells per well of a 96-well plate. After attachment, cells were switched to human endothelial-SFM basal growth medium containing 0.5% calf serum for 16 h and subjected to various treatments. Medium was removed and replaced with medium containing 0.0025% neutral red (Sigma) and incubated for 4 h. The medium was removed, and after solubilization with 1% acetic acid in 50% ethanol, the absorbance at 560 nm was determined for each well.
Mitochondrial membrane potential.
The mitochondrial membrane potential was assessed using the fluorometric dye tetramethylrhodamine ethyl ester (TMRE). Cells were grown in 12-well plates at a density of 2.5 × 105 cells/well. The next day, cells were changed to human endothelial-SFM basal growth medium containing 1% calf serum in the presence or absence of EPO for 6 h. In some experiments, inhibitors (SN50, 50 μg/ml; LY-294002, 50 μM; wortmannin, 100 nM; U0126, 10 μM) were added 30 min before EPO exposure. TMRE (20 nM) was added during the last 30 min. Cells were harvested and analyzed by flow cytometry.
Student's t-test was used for statistical analysis between two groups. Data are presented as means ± SE. Statistical significance was taken at a P value of ≤0.05.
EPO promotes endothelial survival.
HUVEC were serum-starved in medium containing 1% calf serum (CS) for 16 h. Cell viability was measured using the neutral red assay. As shown in Fig. 1A, the viability of cells grown in 1% CS decreased by 40% compared with cells grown in complete growth medium containing 20% CS. EPO treatment reversed the loss of HUVEC viability associated with CS withdrawal in a dose-dependent manner (Fig. 1A). To validate the effects of EPO, we analyzed other parameters of apoptosis. As shown in Fig. 1B, serum withdrawal also caused the dissipation of the mitochondrial transmembrane potential (Δψm). EPO prevented the loss of Δψm induced by serum withdrawal (Fig. 1B). We induced apoptosis in HUVEC using ceramide to determine whether the antiapoptotic effects of EPO were dependent on the death trigger. As shown in Fig. 1C, EPO also prevented the loss of viability triggered by ceramide, suggesting more widespread cytoprotective activity against various triggers.
EPO induces activation of PI3K in endothelial cells.
We next determined the pathways involved in the cytoprotective effects of EPO in HUVEC. We first determined the effects of EPO on PI3K. The serine-threonine kinase Akt is activated downstream of PI3K in response to a number of growth factors and is also known to induce potent cell survival pathways (18, 19, 23, 35). Activation of Akt involves dual phosphorylation on threonine at position 308 and serine at position 473 downstream of PI3K (1). Because of the absolute requirement of phosphorylation for Akt activation, we used a phosphospecific antibody that only recognizes the phosphorylated form of the enzyme to determine its activation. Figure 2A shows the time course of EPO-dependent Akt activation. Akt phosphorylation appeared within 5 min of treatment and remained elevated up to 60 min. The phosphorylation of Akt was also determined in the presence of a PI3K inhibitor. Pretreatment of cells with LY-294002 blocked the phosphorylation of Akt induced by EPO, confirming that Akt is a downstream target of PI3K in EPO signaling (Fig. 2B).
EPO induces activation of ERK1/2 in endothelial cells.
We next determined whether EPO activates the mitogen-activated protein kinase (MAPK) pathway in HUVEC. We focused on ERK1/2 MAPK because of their role in promoting cell survival (7, 25, 30). Activation of these kinases is dependent on dual phosphorylation on tyrosine and threonine, which is catalyzed by their upstream activator MEK (44). Accordingly, detection of phosphorylation of ERK1/2 is a convenient assay for ERK activation. Serum-starved HUVEC were treated with 5 U/ml EPO for various times, and Western blot analysis was performed using a phosphospecific ERK antibody. As shown in Fig. 3A, EPO induced phosphorylation of both ERK1 and ERK2 within 5 min of treatment. EPO-induced phosphorylation was sustained up to 90 min. In contrast, we did not detect phosphorylation/activation of other MAPKs such as p38 and c-Jun NH2-terminal kinase by EPO (data not shown). We next evaluated the effect of U0126, a MEK inhibitor on the EPO-induced phosphorylation of ERK1/2. U0126 completely diminished the ability of EPO to induce phosphorylation of ERK1/2 indicating the absolute requirement of the upstream activator MEK for activation of ERK1/2 by EPO (Fig. 3B).
EPO activates NF-κB in endothelial cells.
We also assessed the role of EPO in the induction of NF-κB in HUVEC. NF-κB transcriptionally activates multiple genes involved in cellular immune response, growth, and survival (33). In unstimulated cells, NF-κB resides mainly in the cytoplasm bound to inhibitory κB proteins. After stimulation, the inhibitory κB proteins are phosphorylated and then degraded liberating NF-κB to translocate into the nucleus to initiate gene transcription. Cells were transiently transfected with the reporter plasmid pNF-κB-luc, which contains five tandem NF-κB binding sites upstream of a luciferase gene, to quantitate NF-κB transcriptional activity. Twenty-four hours following transfection, cells were treated with various doses of EPO for 4 h, and luciferase activity was determined. As shown in Fig. 4A, EPO caused a dose-dependent increase in NF-κB activity. The activity was maximal at 5 U/ml of EPO, reaching a threefold induction compared with untreated cells. This activation was comparable to that achieved by TNF, a strong NF-κB inducer. To further assess the kinetics of NF-κB activation by EPO, a time course of activation was performed using 5 U/ml EPO. The activation of NF-κB by EPO was apparent within 2 h and reached maximal levels between 4 and 8 h. The activity remained elevated for 8 h and slowly declined by 10 h (Fig. 4B). To further assess the activation of NF-κB activity by EPO, SN50, a cell-permeable inhibitor of NF-κB, was used. As expected, SN50 significantly blocked the ability of EPO to activate NF-κB, whereas a related control compound SN50M had no effect (Fig. 4C).
In different contexts both MAPK and PI3K have been reported to be required for NF-κB activation (25, 30, 37, 41, 42). To characterize further the pathways that lead to NF-κB activation, cells were pretreated with U0126, LY-294002, or SN50 and then stimulated with 5 U/ml EPO for 4 h. The activation of NF-κB by EPO was suppressed in the presence of SN50 as expected. Likewise, the PI3K inhibitor LY-294002 diminished activation of NF-κB by EPO. In contrast, inhibition of the ERK pathway by use of the MEK inhibitor U0126 was without effect (Fig. 4D). These results indicate that PI3K, but not the ERK pathway, is involved in EPO-mediated effects leading to NF-κB activation in endothelial cells.
PI3K and MEK provide EPO-mediated cytoprotective signals in endothelial cells.
Next, we examined whether the PI3K/Akt, MEK/ERK, and NF-κB pathways are involved in cytoprotection by EPO. We assessed cell viability in the presence of inhibitors of the relevant pathways (LY-294002, U0126, and SN50). Inhibition of PI3K by LY-294002 diminished the ability of EPO to promote HUVEC survival after serum withdrawal, indicating that EPO-provides a prosurvival signal through activation of the PI3K/Akt pathway (Fig. 5A). Furthermore, inhibition of the MEK/ERK pathway using U0126 also blocked the cytoprotective effects of EPO implying that EPO provides yet another survival signal by activation of the MEK/ERK pathway. In contrast, inhibition of NF-κB activation was unable to reverse the cytoprotective effects of EPO (Fig. 5A). The pathway inhibitors were also used to assess the effects of EPO on preserving mitochondrial transmembrane potential. As shown in Fig. 5B, inhibition of PI3K by LY-294002 or the structurally unrelated compound wortmannin abrogated the effects of EPO on preserving the mitochondrial transmembrane potential. Likewise, inhibition of MEK by U0126 also reversed cytoprotection by EPO. In contrast, inhibition of NF-κB by SN50 was without effect (Fig. 5B).
EPO regulates the expression of Bcl-2 family members in endothelial cells.
Given the effect of EPO on Δψm, we next investigated expression of various anti- and pro-apoptotic Bcl-2 family members, including Bcl-2, Bcl-x, BAX, BAD, and BIM. Serum-starved HUVEC were incubated in the absence or presence of 5 U/ml EPO, and the expression of the various Bcl-2 family members was determined by immunoblotting. No changes in the expression of Bcl-2 were detected under these conditions (data not shown). However, EPO led to a time-dependent increase in Bcl-xL protein levels (Fig. 6A). At 4 h EPO induced a 2.5-fold increase in Bcl-xL protein level, and by 16 h this had increased to eightfold (Fig. 6A). No consistent change in the level of BAX, BID Bcl-xS, and BAD was detected, and we did not observe cleavage of BID after serum withdrawal (data not shown). In addition, phosphorylation of BAD was not detected (data not shown). However, there was a large increase in the level of BIM protein after serum withdrawal, and this was counteracted by EPO (Fig. 6B). We next used endothelial cells derived from precursors present in the umbilical cord blood [CBEC, as previously described (31)] to determine whether the induction of Bcl-xL by EPO occurred specifically in HUVEC or whether it also occurred in other types of endothelial cells. As shown in Fig. 6C, EPO also induced Bcl-xL in CBEC in a similar temporal fashion.
PI3K is involved in the upregulation of Bcl-xL by EPO.
We next determined the signaling pathways involved in the induction of Bcl-xL by EPO by examining the effects of inhibitors of the PI3K, MEK, and NF-κB pathways. Inhibition of PI3K in the presence of EPO abolished the effects of EPO on Bcl-xL expression (Fig. 7A). In contrast, neither the blockade of NF-κB nor that of MEK activation attenuated the effects of EPO on Bcl-xL expression (Fig. 7A). Hence, these results suggest that EPO-mediated upregulation of Bcl-xL protein in HUVEC is mediated by PI3K and independent of ERK and NF-κB pathways.
In healthy cells, BAK activity is regulated directly and specifically by Bcl-xL (52). Therefore, to assess the relative importance of Bcl-xL to the antiapoptotic effects of EPO in HUVEC, we treated cells with a peptide bearing the BH3 domain of the pro-apoptotic protein BAK (BAK-BH3). BAK-BH3 binds tightly to Bcl-xL (52). Thus by binding to Bcl-xL, BAK-BH3 should reverse EPO-mediated cytoprotection and inhibit its anti-apoptotic activity. HUVEC were pretreated with BAK-BH3 peptide and then treated with or without EPO for 6 h. Apoptosis was determined by measuring Δψm. As shown in Fig. 7B, preincubation of cells with BAK-BH3 reversed the effects of EPO on maintaining the transmembrane mitochondrial potential, thus implicating a critical role for Bcl-xL in EPO-dependent endothelial protection.
Endothelial cell damage is a critical component in the pathogenesis of a number of vascular pathologies such as endotoxic shock and atherosclerosis (6, 14). Identification of biochemical pathways that can signal anti-apoptotic responses is therefore important because this may provide a basis for implementing a clinical intervention. In this study, we explored apoptosis-inhibiting pathways mediated by EPO in endothelial cells subjected to serum withdrawal. Our data indicate that EPO confers cytoprotection through the activation of the PI3K/Akt pathway with subsequent upregulation of Bcl-xL as well as through the activity of the ERK pathway independent of Bcl-xL. Serum starvation also led to an increase in the level of pro-apoptotic BH3-only protein BIM, and this effect was reverted by EPO.
The Bcl-2 family consists of pro- and anti-apoptotic proteins that play a critical role in regulating cell survival. The pro-apoptotic BH3-only proteins respond to apoptotic stimuli and are thought to activate multi-domain pro-apoptotic members (BAX and BAK), which trigger the intrinsic pathway of apoptosis. In contrast, anti-apoptotic Bcl-2 family members bind and sequester BH3-only molecules and thus prevent activation of BAX and BAK.
As a survival factor, EPO regulates the activities of some of the Bcl-2 family of proteins. EPO maintains levels of anti-apoptotic molecules (Bcl-2 and Bcl-xL) and decreases the amount of pro-apoptotic BAX in various other cell types (43, 46, 49). In this study, we show that EPO induces the expression of Bcl-xL in HUVEC following serum withdrawal. Bcl-2 was detected at low levels or not at all in HUVEC, and EPO treatment did not induce Bcl-2 expression. Bcl-xL, but not Bcl-2, therefore, appears to be important for EPO-mediated cell survival in endothelial cells. EPO is also required for optimal expression of Bcl-xL in erythroid cells (21, 46). Similar results in neurons indicate that upregulation of Bcl-xL may be necessary for EPO to confer its anti-apoptotic effects (50). Therefore, the regulation of Bcl-xL expression is likely to play an important role in the anti-apoptotic effects of EPO.
Although EPO upregulates Bcl-xL expression in a variety of cells, the signaling pathways that promote this response appear to vary in different cell types. In hematopoietic cells, for example, EPO upregulates the expression of Bcl-xL through activation of Janus kinases but independently of PI3K, Ras, and STAT5 (40). However, in another study, EPO was shown to induce expression of the Bcl-xL gene through activation of STAT5 by binding to the Bcl-xL promoter (48). In the present study, we found that EPO upregulates Bcl-xL in endothelial cells through activation of the PI3K pathway but not the MEK/ERK pathway. This finding also differs from EPO-mediated Bcl-xL regulation in erythroid progenitor cells where activation of ERKs is reported to be involved (36).
The BH3 domain of the BH3-only proteins is an interaction domain that enables them to bind to multidomain anti- or pro-apoptotic members antagonizing the survival or activating the pro-apoptotic activities (12, 13, 29). To further clarify the role of Bcl-xL in EPO-mediated endothelial cell survival, we determined whether inhibition of Bcl-xL with a BAK-BH3 peptide induces apoptosis in the presence of EPO (52). We demonstrate that treatment of cells with a BAK-BH3 peptide that specifically binds to Bcl-xL reversed the anti-apoptotic effect of EPO. By inhibiting apoptosis, the anti-death Bcl-2 proteins preserve the integrity of the mitochondria. Consistent with this role, we also show that Bcl-xL inhibition by the BAK-BH3 peptide caused a reduction in the mitochondrial potential. Furthermore, inhibition of PI3K, which is required for the upregulation of Bcl-xL by EPO, also reduced the mitochondrial potential in the presence of EPO.
To determine further the mechanism by which EPO exerts its anti-apoptotic effects, we measured the expression of some of the pro-apoptotic members of the Bcl-2 family. The expression of the pro-apoptotic Bcl-2 family member BIM increased following serum withdrawal in HUVEC. Furthermore, this upregulation of BIM as with the induction of apoptosis was blocked by EPO. Increased expression of BIM has been observed following withdrawal of survival factors in a variety of cells, including fibroblasts, epithelial cells, lymphocytes, and neurons (8, 16, 39, 51), and appears to be a conserved mechanism for inducing apoptosis under these conditions.
In contrast, expression of BAX and BID was not changed following serum withdrawal or exposure to EPO. In contrast, BAX expression was upregulated in an in vivo rat model of renal injury and this was counteracted by EPO (49). BID is activated by proteolytic cleavage following ligation of death receptors, but no loss of full length BID or cleavage products were detected after serum deprivation of endothelial cells. BID may therefore not be involved in apoptosis of HUVEC following serum withdrawal. BAD is regulated through phosphorylation at Ser136 and Ser112, and this promotes its sequestration in the cytoplasm by binding to 14-3-3 proteins. There was no evidence of BAD phosphorylation following exposure of HUVEC to EPO, suggesting that BAD phosphorylation is not one of the mechanisms by which EPO exerts its anti-apoptotic effects in HUVEC. So far there is no evidence to support a role of BAD in growth factor withdrawal-induced apoptosis using targeted gene deletion studies in mice. In contrast to BAD, resistance of BIM-deficient lymphocytes and neurons to cytokine withdrawal indicates that BIM is a critical initiator of this pathway to apoptosis. EPO is able to suppress BIM induction in endothelial cells, thereby providing an important additional survival mechanism.
The NF-κB signaling pathway is critical in regulating the apoptotic response. NF-κB is commonly associated in apoptosis suppression by transactivating the expression of anti-apoptotic genes. EPO activated NF-κB in HUVEC, and this activation was blocked by SN50, a peptide inhibitor derived from p50 subunit of NF-κB, which inhibits translocation of active NF-κB into the nucleus (34). Activation of both the PI3K/Akt and MEK/ERK pathways has been shown to be required for NF-κB activation under various conditions (25, 30, 47, 53). The lack of a role for the MEK/ERK pathway in NF-κB activation has also been reported (5). Here we show that PI3K, but not MEK/ERK, is required for EPO-induced activation of NF-κB in endothelial cells. These observations suggest an absence of cross talk between ERK and NF-κB pathways in the EPO signaling cascade. Furthermore, selective inhibition of NF-κB did not attenuate EPO-mediated endothelial survival. Therefore, in this study, the NF-κB pathway can be dissociated from the survival pathways required to confer cytoprotection by EPO.
EPO activated the MEK/ERK pathway, and suppression of this pathway reversed the anti-apoptotic effects of EPO. One mechanism by which the MEK/ERK pathway protects cells from apoptosis is through ribosomal S6 kinase (RSK), which inactivates the proapoptotic protein BAD through phosphorylation (7). ERK5 in particular has been shown to protect endothelial cells from apoptosis by stimulating phosphorylation of BAD (38). We did not detect BAD phosphorylation after exposure of HUVEC to EPO, therefore, this pathway leading to BAD phosphorylation is likely not required for anti-apoptotic effects of EPO. Another mechanism by which the MEK/ERK pathway confers cytoprotection is through phosphorylation of BIM leading to BIM degradation by the proteasome (32). In addition to promoting BIM turnover, phosphorylation of BIM by ERK1/2 may also prevent BIM from interacting with BAX, presumably preventing activation of BAX and cell death (22). In line with the aforementioned studies, serum withdrawal in HUVEC led to the upregulation of BIM protein. Treatment with EPO reversed BIM upregulation. A schema of the mechanism of the anti-apoptotic action of EPO is illustrated in Fig. 8. Taken together our findings indicate that EPO-mediated survival of endothelial cells is mediated by the PI3K and MEK/ERK pathways as well as through regulation of BIM expression.
This work was supported by grants from the Canadian Institutes of Health Research and the Heart and Stroke Foundation of BC and the Yukon, and funds from Ortho Biotech. A. Karsan is a Senior Scholar of the Michael Smith Foundation for Health Research.
We thank Fred Wong for technical support.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 by the American Physiological Society