- Email: [email protected]
Erythropoietin receptor-operated Ca2+ channels: Activation by phospholipase C-γ1
Kidney International, Vol. 53 (1998), pp. 1259 –1268
Erythropoietin receptor-operated Ca21 channels: Activation by phospholipase C-g1 MARIO B. MARRERO, RICHARD C. VENEMA, HEPING MA, BRIAN N. LING, and DOUGLAS C. EATON Renal Division and The Center for Cell and Molecular Signaling, Departments of Pathology and Medicine, Emory University School of Medicine and Veterans Affairs Medical Center, Atlanta; and Vascular Biology Center, Medical College of Georgia, Augusta, Georgia, USA
Erythropoietin receptor-operated Ca21 channels: Activation by phospholipase C-g1. Erythropoietin (EPO) increases Ca21 influx in vascular smooth muscle cells and acts both as a direct vasoconstrictor and vascular growth factor (that is, angiogenesis). However, the mechanism by which EPO promotes extracellular Ca21 entry in contractile cells has not been elucidated. In hematopoietic cells, EPO induces tyrosine kinase (TK)dependent activation of phospholipase C (PLC)-g1 and Ca21 influx via a voltage-independent Ca21 conductance. In contractile mesangial cells, we have recently characterized a voltage-independent, 1 pS Ca21 channel that is dependent on both TK and PLC-g1 activity. Therefore, we examined cultured rat glomerular mesangial cells after timed exposure to recombinant human EPO (20 U/ml). Erythropoietin increased the tyrosine phosphorylation of PLC-g1, promoted membrane complex formation between PLC-g1 and the EPO receptor itself, and raised the levels of intracellular inositol 1,4,5-trisphosphate and intracellular Ca21. Consistent with our previous studies, 1 pS Ca21 channel activity was extremely low under basal, unstimulated conditions in cell-attached patches, but was dramatically increased when EPO was present in the patch pipette. Tyrosine kinase inhibition with 100 mM genistein or 1 mM PP1 (Src; selective tyrosine kinase inhibitor) prevented all of these EPO-induced responses. We conclude that: (1) EPO-induced stimulation of 1 pS Ca21 channels is mediated via a cytosolic Src TK in glomerular mesangial cells. (2) Stimulation of this Ca21-activated, Ca21-permeable channel is dependent on the tyrosine phosphorylation/activation of PLC-g1. (3) This cascade provides a possible mechanism for the vasoconstriction and hypertension observed with clinical EPO use for the treatment of chronic anemias.
Human recombinant erythropoietin (EPO) therapy for the treatment of chronic anemias is associated with an increased incidence of new onset hypertension and a worsening of preexisting hypertension [1–5]. Erythropoietin acts as both a direct vasoconstrictor and a vascular growth factor (that is, angiogenesis) [2, 5–7]. In platelets from hypertensive patients and hypertensive rat models, but not in platelets from normotensive patients or rats, EPO elevates cytosolic Ca21 levels [8, 9]. Similarly, EPO increases intracellular Ca21 concentrations in vascular smooth muscle cells [1]. However, the mechanism by which EPO mediates
Key words: tyrosine kinase, intracellular Ca21, renal failure, hypertension, anemia. Received for publication May 2, 1997 and in revised form November 25, 1997 Accepted for publication November 26, 1998
© 1998 by the International Society of Nephrology
increases in intracellular Ca21 levels in contractile cells has not been elucidated. In hematopoietic cells, EPO induces tyrosine kinase-dependent activation of phospholipase C (PLC)-g1 and extracellular Ca21 influx via a voltage-independent, EPO receptor-operated Ca21 conductance pathway [8, 10 –15]. The g isoforms of PLC are classically activated after tyrosine phosphorylation by growth factor receptor-linked tyrosine kinase activity [such as plateletderived growth factor (PDGF), epidermal growth factor, fibroblast growth factor]. However, unlike classic growth factor receptors, EPO and other similar cytokine receptors do not possess intrinsic tyrosine kinase activity [16]. Despite this fact, EPO has been shown to stimulate rapid protein tyrosine phosphorylation in hematopoietic cells [11, 16 –21]. Cytosolic protein tyrosine kinases, which are activated by receptors lacking intrinsic kinase activity (such as interleukins, interferons, angiotensin II), include members of the Src family (such as p60src, p59fyn, p56lck, p62yes) [22]. We have shown that p60src indeed does promote the tyrosine phosphorylation and activation of PLC-g1, leading to the hydrolysis of phosphatidylinositol 4,5-bisphosphate (PIP2) and the generation of inositol 1,4,5-trisphosphate (IP3) [23–25]. This PLC-g1 cascade stimulates a biphasic rise in intracellular Ca21 consisting of an initial release from IP3-sensitive intracellular Ca21 stores followed by an influx of extracellular Ca21. In non-excitable tissues, including glomerular mesangial and vascular smooth muscle cells, the influx of extracellular Ca21 is an important prerequisite for contraction and the initiation of cellular DNA synthesis and proliferation [26, 27]. Evidence suggests that this extracellular Ca21 influx occurs via voltage-independent, receptor-operated Ca21 channels [28 –30]. In glomerular mesangial cells and vascular smooth muscle cells, we have recently identified PLC-g1 and have shown that PLC-g1 is tyrosine phosphorylated and activated by both receptors possessing (for example, PDGF-b receptor) and lacking (such as angiotensin II receptor) intrinsic tyrosine kinase activity [23–25, 29]. Our group has also characterized a voltage-independent, receptor-operated Ca21 channel in rat glomerular mesangial cells [27–29, 31]. There are approximately 1000 of these low conductance (1 pS) Ca21-permeable channels channels per cell (or 4 3 107 channels/cm2). The channels are inactive under basal conditions; they are strongly activated by growth factors (such as PDGF); and, to be active, they require activation of at least one tyrosine kinase and activation of PLC-g1. All of these features
1259
1260
Marrero et al: Ca21 channel activation by PLC-g1
make this 1 pS Ca21 channel a potential candidate for mediating the previously reported voltage-independent, EPO receptor-operated Ca21 entry [8, 11–15]. In the present study, we investigated the role of tyrosine kinases, PLC-g1 tyrosine phosphorylation, PLC-g1/EPO receptor complex formation, and intracellular Ca21 in the regulation of the 1 pS Ca21 channel by EPO in cultured rat glomerular mesangial cells. METHODS Preparation of rat glomerular mesangial cell cultures Glomerular mesangial cells were cultured using previously described methods [24, 28, 31–33]. Briefly, renal cortices were dissected from male Sprague-Dawley rats (75 to 150 g). Mesangial cell-enriched glomerular cores were isolated by differential sieving and incubated for 45 to 60 minutes with collagenase (1200 U/ml) in Ca21/Mg21 free Hank’s balanced salt solution (Irvine Scientific, Santa Ana, CA, USA). The mesangial cell suspension was washed and plated in standard RPMI-1640 medium supplemented with 17% (vol/vol) fetal bovine serum, 2 mM glutamine, 5 mg/ml insulin, 5 mg/ml transferrin, 5 ng/ml selenium, 100 U/ml penicillin, 100 mg/ml streptomycin, and 2 mg/ml amphotericin B at 37°C in 5% CO2/95% air. For immunoblot analysis and measurement of intracellular IP3 levels, mesangial cell subpassages 3 to 5 were grown to near confluence in 100 mm culture dishes (Fisher Scientific, Pittsburgh, PA, USA). To growth arrest the cells, medium was replaced with serum-free medium for 24 to 48 hours prior to each experiment. For patch clamp experiments, mesangial cell subpassages 3 to 5 were grown to confluency on glass coverslips. Immunoprecipitation of phospholipase C-g1 in mesangial cell lysates Mesangial cells, grown in 100 mm dishes to near confluence, were growth-arrested in 5 ml of serum-free RPMI-1640 for 24 to 48 hours. Cultures were then exposed to EPO for timed periods. Incubations were terminated by aspirating the medium and washing two to three times in ice-cold PBSV (phosphate-buffered saline with 1 mM Na3VO4). Each dish was then treated with 2 ml of ice-cold lysis buffer (25 mM Tris-HCl, pH 7.4, 1% Triton X-100, 10% glycerol, 10 mM Na3P2O7, 50 mM NaF, 1 mM Na3VO4, 1 mM PMSF and 10 mg/ml aprotonin) and placed on ice for 30 minutes. The thawed cells were harvested by scraping and sonicated for five seconds. with a Cole Parmer ultrasonic homogenizer, and then the homogenate was centrifuged at 6,000 3 g for 20 minutes. The soluble extracts were precleared with 20 to 30 ml/ml cell lysate of Immunoprecipitin (Life Technologies, Inc., Grand Island, NY, USA), which are membranes of formalin-fixed, heat-killed Staphylococcus aureus bacteria. To immunoprecipitate phosphotyrosine proteins from the cleared lysate, we used 5 mg/ml antiphosphotyrosine (PY20) monoclonal antibody (Transduction Laboratories, Lexington, KY, USA). To immunoprecipitate PLC-g1 directly, we used 5 mg/ml of a mixed monoclonal anti-PLC-g1 antibody (Upstate Biotechnology Inc., Lake Placid, NY, USA) as previously described [24]. The immunoprecipitates were then recovered by centrifugation and washed four times with ice-cold wash buffer (25 mM Tris-HCl, pH 7.4, 0.1% Triton X-100, 150 mM NaCl, and 2 mM Na3V04). Immunoprecipitated proteins were dissolved in 80 ml of Laemmli buffer, boiled for five minutes at 95°C, and analyzed on 10% gels by SDS-PAGE. Proteins were
then transferred to nitrocellulose membranes for 16 hours at 100 mA. The membrane was blotted with either anti-PLC-g1 monoclonal antibody or antiphosphotyrosine (4G10) monoclonal antibody (Upstate Biotechnology). The proteins were detected using Enhanced Chemiluminescense (Amersham Corp., Arlington Heights, IL, USA). Immunoprecipitation of EPO receptor/PLC-g1 complex in mesangial cell membrane fractions Mesangial cells, grown to near confluence, were growth-arrested in 5 ml of serum-free RPMI-1640 for 24 hours and then exposed to EPO. To terminate the incubation, the cells were washed three times with ice cold PBS plus 1 mM Na3V04. Hypotonic buffer (50 mM Tris-HCl, pH 7.4, 2 mM Na3VO4, 50 mM NaF, 10 mM Na2P2O7, 1 mM phenylmethylsulfonyl fluoride, 10 mg/ml leupeptin, 10 mg/ml aprotinin) was added and the cells were then placed on ice to thaw for 30 minutes. The thawed cells were scraped from the plate and centrifuged for 20 minutes at 6,000 3 g at 4°C. The supernatant was discarded and to the pellet was added 1 ml of ice-cold lysis buffer, followed by incubation for one hour at 4°C. After centrifugation for 20 minutes at 6,000 3 g at 4°C, the supernatant was collected, anti-EPO receptor antibody (5 ml/ml lysate) was added, followed by incubation overnight at 4°C. The next day, Protein A/G plus 30 ml of buffer (2 mM Na3VO4, 50 mM NaF, 10 mM Na2P2O7, 1 mM phenylmethylsulfonyl fluoride, 10 mg/ml leupeptin, 10 mg/ml aprotinin) was added while rocking at 4°C for two to three hours. The immunoprecipitates were then recovered by centrifugation and washed three times with ice-cold wash buffer (50 mM Tris-HCl, pH 7.4, 0.1% NP-40, 150 mM NaCl, 1 mM Na3VO4). Immunoprecipitated membrane proteins were dissolved in 80 ml of Laemmli buffer, boiled for five minutes at 95°C, and analyzed on 10% gels by SDS-PAGE. Proteins were then transferred to nitrocellulose membranes and blotted with anti-PLC-g1 antibody. The proteins were detected using enhanced chemiluminescense. Measurement of inositol 1,4,5-trisphosphate Following mesangial cell exposure to 20 U/ml EPO, the incubation was stopped by the addition of 1 ml of 100% ice-cold trichloroacetic acid to the plates. The plates were placed on ice for 10 minutes. The cells were then harvested by scraping and transferring to polyethylene tubes. The cell extract was sonicated on ice with a Cole Parmer ultrasonic homogenizer for approximately one minute and then centrifuged for 10 minutes at 6,000 3 g. The supernatant was removed and warmed to room temperature for 15 minutes. Levels of IP3 in each supernatant were determined using a radioimmunoassay based kit (Dupont NEN, Boston, MA, USA) as previously described [24, 29]. Confocal intracellular Ca21 fluorescent imaging To measure intracellular Ca21 levels, fluorescence was measured with mesangial cells grown on glass cover slips [34]. Cells were preincubated with 5 mM Indo-1 AM (Molecular Probes, Eugene, OR, USA) for 30 minutes at room temperature in darkness and then washed with saline solution three times. Cover slips were then placed in a chamber secured to the stage of a Meridian ACAS 570/Ultima laser scanning confocal microscope (Laser Cytofluorimeter Working Station; Meridian Instruments, MI, USA). Cells loaded with Indo-1 AM were excited by UV light
Marrero et al: Ca21 channel activation by PLC-g1
at 351 to 364 nm and the cytoplasmic free calcium signal was read as the fluorescence ratio at 405 nm and 530 nm wave lengths. Intracellular Ca21 concentrations were calculated by using the standard working curves of fluorescence ratio versus free Ca21 in Ca21-EGTA buffer. Patch-clamp recording TW150F-6 glass pipettes (World Precision Instruments, New Haven, CT, USA) were prepared with a PP-83 patch-pipette puller and a MF-83 pipette polisher (Narishige, Tokyo, Japan) [31–33]. Pipettes with resistances of . 10 megaohms were used for the experiments. Single-channel patch clamp configurations, including the cell-attached and inside-out modes, were established on the membrane of mesangial cells cultured on glass coverslips. In the cell-attached mode, single channel currents were recorded without applying potential to the pipette (that is, resting membrane potential). In the excised inside-out mode, single channel recordings were obtained with 160 mV applied pipette voltage (Vm 5 260 mV). All experiments were conducted with an Axopatch-1D amplifier (Axon Instruments, Foster, CA, USA) at room temperature, 22 to 23°C. Only patches with a seal resistance . 30 gigaohms and containing currents without baseline drift were used for experiments. Data were digitally recorded with a DAS 601 Pulse Code Modulator (Dagan Corp., MN, USA) and a SL-HF860D VCR (Sony Corp. of America, NJ, USA), and acquired using a 902LPF 8-pole Bessel filter (Frequency Devices Inc., MA, USA), TL-2 acquisition hardware and Axotape software (Axon Instruments). Single channel traces were sampled at 2 msec/point and software filtered at 100 Hz, and inward Ca21 or Mn21 current (pipette to cell) was represented by downward channel transitions. Single channel analysis was performed on IBM PC class computers using both Pclamp 6.0 and locallydeveloped software [31–33]. NPo (number of channels times the open probability) was used as a measure of channel activity. All NPo values were calculated from three minutes of single channel recording and are reported as mean values 6 SEM. When cellattached patch experiments were conducted in a paired fashion (that is, data from each patch membrane serving as the control for an experimental manipulation), the average change in NPo for a group of patches, compared before and after an experimental treatment, was analyzed using a paired t-test or analysis of variance (ANOVA) for multiple comparisons. For unpaired experiments, the traditional t-test or ANOVA was used. Results were considered significant if P , 0.05. Statistical analysis was performed with SigmaPlot and SigmaStat software (Jandel Scientific, CA, USA). Solutions and chemicals For cell-attached patch experiments, the standard extracellular bath solution (outside the patch pipette) contained (in mM) 140 NaCl (after adjusting pH to 7.4 with NaOH), 4 KCl, 1 CaCl2, 1 MgCl2 and 10 HEPES. When excised inside-out patch experiments were performed, the “cytoplasmic” bath solution contained (in mM) 140 KCl (after adjusting pH to 7.4 with KOH), 10 NaCl, 1 MgCl2, 2 EGTA and 10 HEPES. The “cytoplasmic” bath CaCl2 was varied for different experiments and was reported as the final free Ca21 activity. Since we were primarily interested in the function of the specific EPO- (and PDGF-) activated calcium channel and since the unit conductance of this channel was quite low conductance, we used a solution in the patch pipette which
1261
was high in calcium to the exclusion of other cations similar to the solution that had allowed us to first identify the channel [31]. The solution contained (in mM) 110 CaCl2, 4 KCl (after adjusting pH to 7.4 with KOH), 1 MgCl2, and 10 HEPES. Tween 20, acrylamide, N-N9-methylene-bisacrylamide, N,N,N9,N9-tetramethylenediamine, and SDS were purchased from Bio-Rad Labs (Hercules, CA, USA). Prestained molecular weight standards, genistein, and Immunoprecipitin were obtained from Life Technologies, Inc. (Grand Island, NY, USA). Penicillin, streptomycin, insulin, PP1 and Indo-1 were purchased from Calbiochem Corp. (San Diego, CA, USA) and Sigma Chemical Corp. (St. Louis, MO, USA). Human recombinant erythropoietin was obtained from Amgen (Thousand Oaks, CA, USA). Nitrocellulose membranes and sheep anti-mouse IgG were purchased from Amersham Corp. Monoclonal antibodies to PLC-g1 and antiphosphotyrosine (4G10) were obtained form Upstate Biotechnology, Inc. Goat anti-mouse IgG was obtained from Jackson ImmunoResearch Laboratories, Inc. (West Grove, PA, USA). Antiphosphotyrosine (PY20) and EPO receptor antibodies were obtained from Transduction Laboratories and protein A/G plus agarose came from Santa Cruz Biotechnologies, Inc. (Santa Cruz, CA, USA). RESULTS Effect of erythropoietin on phospholipase C-g1 in glomerular mesangial cells Ren et al [10] have shown that EPO stimulates the tyrosine phosphorylation and activation of PLC-g1 in a hematopoietic cell line. In glomerular mesangial and vascular smooth muscle cells, we have previously shown that classic growth factors, such as PDGF, promote the tyrosine phosphorylation and activation of PLC-g1 [25, 27, 28, 31]. However, we have also shown that PLC-g1 can be activated by receptors that, like the EPO receptor, lack intrinsic tyrosine kinase activity, but rather operate via cytosolic tyrosine kinases (such as angiotensin II) [23–25, 35]. Therefore, we examined EPO-induced tyrosine phosphorylation of PLC-g1 in cultured rat glomerular mesangial cells by immunoprecipitation. Tyrosine phosphorylated proteins were first immunoprecipitated using antiphosphotyrosine antibodies and then probed with anti-PLC-g1 antibodies. When compared to basal levels, exposure to 20 U/ml EPO produced an increase in the tyrosine phosphorylation of PLC-g1 within 30 seconds (N 5 3; Fig. 1). Peak phosphorylation was observed within five minutes of EPO exposure and then returned to baseline by 30 minutes. We also performed the experiment reversing the order of antibody addition, first immunoprecipitating mesangial cell lysates with anti-PLC-g1 antibodies and then probing with antiphosphotyrosine antibody. Again, a significant increase in the tyrosine phosphorylation of PLC-g1 was observed upon exposure to 20 U/ml EPO (N 5 3; data not shown). The peak phosphorylation response was again observed within five minutes. of EPO exposure. In mesangial cells which were first preincubated with tyrosine kinase inhibitors, 100 mM genistein for one hour (N 5 3) or 1 mM PP1 for 15 minutes (N 5 3), the usual EPO-induced tyrosine phosphorylation of PLC-g1 was abolished (Fig. 1). Membrane complex formation between phospholipase C-g1 and the erythropoietin receptor We and other groups have previously shown that growth factor-mediated tyrosine phosphorylation of PLC-g1 promotes
1262
Marrero et al: Ca21 channel activation by PLC-g1
Fig. 1. Erythropoietin (EPO)-induced tyrosine phosphorylation of phospholipase C (PLC)-g1. (Western blot, inset) Mesangial cells exposed to 20 U/ml EPO for various timed periods. Mesangial cell lysates were then immunoprecipitated with antiphosphotyrosine antibody (PY-20) and probed with anti-PLC-g1 antibody. The PLC-g1 bands were quantitated by densitometry using a La Cie scanner interfaced with a personal computer. Each band was scanned in two dimensions and the density corrected for the background present in the lane. The results shown in the graph represent the corrected density for each time point and are expressed as arbitrary units plotted against time of EPO exposure in mesangial cells that were either pre-treated with tyrosine kinase inhibitors, 100 mM genistein (‚) for one hour or 1 mM PP1 (F) for 15 minutes, or vehicle (E). Each data point represents mean 6 SE for N 5 3 separate plates of mesangial cells.
the formation of a membrane-bound complex between PLC-g1 and the growth factor receptors themselves (such as PDGF-b receptor/PLC-g1 complex) [27, 28, 31, 36, 37]. Therefore, we investigated the ability of EPO to promote the membrane association of PLC-g1 with the EPO receptor itself in glomerular mesangial cells. Mesangial cells were incubated in EPO (20 U/ml) for various time periods followed by isolation of the membrane fractions. We first immunoprecipitated the isolated membrane fraction with an anti-EPO receptor antibody. Probing the immunoprecipitated proteins with an anti-PLC-g1 antibody revealed the formation of a membrane-associated PLC-g1/EPO receptor complex (N 5 3; Fig. 2). Membrane fractions were also examined from mesangial cells preincubated with either 100 mM genistein for one hour (N 5 3) or 1 mM PP1 for 15 minutes (N 5 3) prior to timed exposures to EPO. Proteins were again immunoprecipitated with anti-EPO receptor antibodies and then probed with anti-PLC-g1 antibodies. Under these conditions, the 145 kDa band (molecular weight of PLC-g1) was no longer observed, as shown in Figure 2 for cells that were preincubated with PP1. These studies suggested that tyrosine kinase activity, perhaps involving cytosolic Src kinases, was essential for EPO-induced membrane recruitment of PLC-g1 and for membrane complex formation between PLC-g1 and the EPO receptor itself in glomerular mesangial cells. Effect of erythropoietin on 1 pS Ca21 channels in glomerular mesangial cells In glomerular mesangial cells, we have previously shown that PLC-g1 tyrosine phosphorylation and PLC-g1/growth factor re-
ceptor complex formation promotes the activation of receptoroperated, 1 pS Ca21 channels [27, 28, 31]. Erythropoietin induces tyrosine kinase-dependent extracellular Ca21 influx via a voltageindependent, EPO receptor-operated Ca21 conductance pathway in hematopoietic cells [8, 11–15]. However, this voltage-independent, EPO receptor-operated Ca21 conductance has not been characterized at a single channel level. Therefore, we investigated the effect of EPO on the 1 pS Ca21 channel in cultured rat glomerular mesangial cells. Consistent with our previous studies, 1 pS Ca21 channel activity was extremely low under basal, unstimulated conditions in cell-attached patches; NPo (number of channels times the open probability) was 0.06 6 0.02 (N 5 8; Fig. 3). The addition of recombinant human EPO (20 U/ml) to the extracellular bath outside the cell-attached patch pipette did not increase channel activity recorded from the cell-attached membrane patch (NPo 5 0.09 6 0.04; N 5 7) from basal values (P 5 0.55). In contrast, channel activity was dramatically increased when recombinant human EPO (20 U/ml) was present inside the patch pipette; NPo was 0.73 6 0.09 (N 5 10). Intrapipette EPO significantly increased channel activity over basal values (P , 0.00002). We also examined the time course of EPO’s activation of channels. We presumed that the formation of the seal resulted in a step change in EPO concentration and calculated the open probability for three second windows starting from the time of seal formation. Three such records of open probability were averaged to produce the mean open probability shown in the lower left panel of Figure 3. These results demonstrate that the time course of activation is consistent with the time course of PLC
Marrero et al: Ca21 channel activation by PLC-g1
1263
Fig. 2. Erythropoietin (EPO)-induced membrane complex formation between phospholipase C (PLC)-g1 and the EPO receptor. Mesangial cells were treated for the times indicated with 20 U/ml EPO in cells that were or were not pre-treated with PP1 (1 mM) for 15 minutes prior to the addition of the EPO. The membrane fraction was then separated after hypotonic cellular lysis. Proteins were first immunoprecipitated with an anti-EPO receptor antibody and developed with an anti-PLC-g1 antibody. A representative experiment shows a significant increase in EPO/PLC-g1 complex formation within 30 seconds of EPO exposure.
phosphorylation. However, since channel opening is a stochastic process, a precise determination of the time course would require summing of many more records. Thus, we feel that, while the time course of channel activation is similar to that of PLC phosphorylation, the channel activation time should be interpreted with caution. We then investigated the effect of tyrosine kinase inhibition on EPO-induced 1 pS Ca21 channel activation. Channel activity was first recorded with cell-attached patch pipettes containing 20 U/ml EPO (Fig. 4). Genistein (100 mM) was then added to the extracellular bath outside the patch pipette [24, 28, 29]. Despite the continued presence of intrapipette EPO, tyrosine kinase inhibition abolished Ca21 channel activity, decreasing NPo from 0.56 6 0.14 to 0.08 6 0.02 (P , 0.01; N 5 8). Another set of mesangial cells were pretreated with PP1 (1 mM), an inhibitor [38], for 15 minutes prior to patching. Basal Ca21 channel NPo after PP1 pretreatment was 0.04 6 0.01 (N 5 7; Fig. 4). NPo under the latter condition was not significantly different from basal Ca21 channel activity recorded in the absence of PP1 (Fig. 3). However, PP1 prevented the usual channel activation observed when EPO was present in the cell-attached patch pipette (NPo 0.06 6 0.03; N 5 7; P 5 0.60; Fig. 4). These results suggested that EPO-induced activation of the 1 pS Ca21 channel was dependent on protein tyrosine phosphorylation likely involving a member of the Src family of cytosolic tyrosine kinases. Role of intracellular Ca21 in 1 pS Ca21 channel activation by erythropoietin Pholipase C promotes the release of Ca21 from IP3-dependent intracellular stores, which in turn triggers voltage-independent
Ca21 entry in multiple cell types, including glomerular mesangial cells [24, 25, 29, 30]. In hematopoietic and vascular cell types, EPO elevates cytosolic Ca21 levels [1, 8, 9, 11–15]. Therefore, we measured intracellular IP3 levels in glomerular mesangial cells (Fig. 5) and found that exposure to 20 U/ml EPO significantly increased IP3 levels with a time course paralleling EPO-induced tyrosine phosphorylation of PLC-g1 and membrane complex formation between PLC-g1 and the EPO receptor. Using confocal microscopy, we also measured intracellular Ca21 levels in Indo1-loaded glomerular mesangial cells. In response to 20 U/ml EPO exposure, intracellular Ca21 levels did indeed increase from basal values (Fig. 6). The peak intracellular Ca21 response occurred approximately 20 minutes after EPO exposure. Erythropoietininduced increases in intracellular IP3 and Ca21 levels were markedly attenuated by pretreatment with 100 mM genistein (1 hr) or 1 mM PP1 (15 min). Other groups have shown that intracellular Ca21 release plays an important role in the regulation of receptor-operated Ca21 entry in many non-excitable cell types [30]. In particular, we have shown that growth factor-induced increases in intracellular Ca21 leads to activation of the 1 pS Ca21 channel in glomerular mesangial cells, and that in the absence of intracellular Ca21 the channels were completely inactive [29]. Therefore, we examined the role of intracellular Ca21 on 1 pS channel activity in cellattached patches. Without EPO present in the pipettes, cellattached patches were established and excised, exposing the inside-out patch membranes to a “cytoplasmic” bath containing 1028 M Ca21; NPo was 0.007 6 0.012 (N 5 5). Channel activity increased significantly when “cytoplasmic” bath Ca21 was raised to 1027 M (NPo 5 0.25 6 0.08; P , 0.01) and then to 1026 M (NPo
1264
Marrero et al: Ca21 channel activation by PLC-g1
Fig. 3. Erythropoietin (EPO)-induced 1 pS Ca21 channel activity. (Left) Representative traces depict 1 pS Ca21 channel activity in cell-attached patches on glomerular mesangial cells. Channel activity was measured in the absence of EPO (Control), in the presence of 20 U/ml in the extracellular bath outside the patch pipette, and in the presence of 20 U/ml inside the patch pipette. Inward current is represented by downward deflections. “C” marks the zero current level (closed state) and O1 marks channel openings. (Lower left) The time course of EPO’s activation of channels. We presumed that the formation of the seal resulted in a step change in EPO concentration and calculated the open probability for three seconds windows starting from the time of seal formation. Three such records of open probability were averaged to produce the mean open probability. These results demonstrate that despite the significant variability in the open probability (as expected for any single channel recording), the time course of activation is consistent with the time course of PLC phosphorylation. (Right) Summary plot of NPo for all cell-attached patch experiments.
5 0.65 6 0.10; P , 0.02). Combined with our biochemical results, these findings suggested that EPO-induced activation of 1 pS Ca21 channels occurs via the tyrosine phosphorylation and activation of PLC-g1 leading to the localized release of IP3-dependent intracellular Ca21 stores. DISCUSSION Erythropoietin-induced 1 pS Ca21 channel activation is dependent on Src-specific tyrosine kinases Erythropoietin has been shown to stimulate voltage-independent Ca21 influx in multiple cell types, including vascular smooth muscle and hematopoietic cells [8, 11–15]. Our previous patch clamp studies have characterized voltage-independent, PDGF-b receptor-operated 1 pS Ca21 channels in glomerular mesangial cells [27–29, 31]. In our present study, we find that EPO also activates the same 1 pS Ca21 channel in cell-attached patches on cultured rat glomerular mesangial cells. Channel activation by EPO is significantly reduced by two structurally different tyrosine kinase inhibitors: genistein and PP1. We have previously shown
that tyrosine kinase inhibition (such as with genistein and tryphostin 9) blocks the activation of 1 pS Ca21 channels by PDGF in glomerular mesangial cells [28, 29]. Genistein has also been shown to block receptor-operated Ca21 influx in rat pancreatic acinar cells, human fibroblasts, and mouse osteoblast-like cells [28]. While genistein is a nonspecific tyrosine kinase inhibitor that competes for the ATP binding site, PP1 is a Src-specific tyrosine kinase inhibitor [38]. Cytokine receptors related to the EPO receptor (such as interleukin, interferon) lack intrinsic receptor tyrosine kinase activity, but have been shown to activate cytosolic Src tyrosine kinases in multiple cell types [16, 24, 28]. Erythropoietin promotes tyrosine phosphorylation and activation of phospholipase C-g1 in mesangial cells In multiple cell types, a number of proteins are tyrosine phosphorylated in response to EPO. These tyrosine phosphorylated proteins include phospholipase C-g1, the EPO receptor itself, DNA binding proteins (p84, p91, p93), phosphatidylinositol 3 kinase (PI3K), p120 GTPase-activating protein (GAP), p52shc,
Marrero et al: Ca21 channel activation by PLC-g1
1265
Fig. 4. Effect of tyrosine kinase inhibition on erythropoietin (EPO)-induced Ca21 channel activity. (Upper left) Channel activity induced by intrapipette EPO (20 U/ml) was recorded in cell-attached patches (upper trace). Within five minutes of adding 100 mM genistein to the extracellular bath, channel activity was significantly inhibited (lower trace). (Lower left) Cell-attached patches were established on mesangial cells after 15 minutes pretreatment with 1 mM PP1. Channel activity was measured in the absence (upper trace) and presence of intrapipette EPO (20 U/ml) (lower trace). (Right) Summary plot of NPo for all cell-attached patch experiments. Symbols connected by lines represent the change in channel activity for the same cell-attached patch before and after exposure to genistein.
STAT (signal transducers and activators of transcription) 1 and 5, and several as yet unidentified proteins ranging in molecular weight from 25 to 145 kDa [7, 10, 16 –21, 39, 40]. Phospholipase C-g1 contains four tyrosine residue sites for phosphorylation [24]. In glomerular mesangial cells, our biochemical data show that EPO binding to its respective EPO receptor leads to the tyrosine phosphorylation of PLC-g1. This in turn promotes the formation of a membrane-associated complex between PLC-g1 and the EPO receptor itself. We find that EPOinduced tyrosine phosphorylation of PLC-g1, PLC-g1/EPO receptor membrane complex formation, and increases in intracellular Ca21 are abolished by the tyrosine kinase inhibitors, genistein and PP1. Phospholipase C-mediated increases in intracellular IP3 levels promotes the mobilization of intracellular Ca21 stores [24, 29]. In mesangial cells, intracellular Ca21 concentrations are 1027 M under basal conditions and increase to 1026 M in response to IP3-mediated signal transduction pathways [29] [24]. Mene and associates [30] have recently shown that the release of intracellu-
lar Ca21 stores or ionophore-mediated Ca21 entry promotes voltage-independent Ca21 (or Mn21) entry in human glomerular mesangial cells. They hypothesized that “the InsP3-sensitive Ca21 pool may serve as a trigger for a sustained response, thus initiating a cascade amplification signaling mechanism.” In the absence of intrapipette EPO, we do indeed find that the 1 pS Ca21 channel is itself directly activated by increasing “cytoplasmic” Ca21 concentrations (1028 to 1026 M) in excised inside-out patches. Our results suggest that the Ca21 channels are initially triggered by intracellular Ca21, released from IP3-dependent stores. Sustained channel activation may be maintained by Ca21 entry through the channel itself. Ca21-permeable channels allow Ca21 entry at rates . 106 ions/sec [41]. Local Ca21 in the immediate vicinity of the channel pore can reach high concentrations [41, 42], providing a positive feedback loop for the Ca21-activated, 1 pS Ca21 channel. Finally, we find that 1 pS Ca21 channel activity in excised patches, at a given “cytoplasmic” Ca21 concentration, is higher with intrapipette EPO present than without intrapipette EPO
1266
Marrero et al: Ca21 channel activation by PLC-g1
Fig. 6. Erythropoietin (EPO)-induced intracellular Ca21 concentrations. Mesangial cells with and without pretreatment with tyrosine kinase inhibitors, 100 mM genistein for one hour or 1 mM PP1 (a tyrosine kinase inhibitor) for 15 minutes, were exposed to 20 U/ml EPO for various timed periods. Intracellular Ca21 was assessed by measuring the Indo-1 fluorescence of individual mesangial cells by confocal microscopy. Each datum point represents mean 6 SE for N individual mesangial cells. Fig. 5. Erythropoietin (EPO)-induced intracellular inositol 1,4,5triphosphate (IP3) production. Mesangial cells with and without (E) pretreatment with tyrosine kinase inhibitors, 100 mM genistein (F) for one hour or 1 mM PP1 (a tyrosine kinase inhibitor; ‚) for 15 minutes, were exposed to 20 U/ml EPO for various timed periods. IP3 was determined by radioimmunoassay. Each datum point represents mean 6 SE for N 5 3 separate plates of mesangial cells.
present (data not shown). This is consistent with the activation of EPO receptor-operated Ca21 channels by another signal transduction pathway besides PLC-g1. In mesangial cells, we have recently shown that in addition to the PLC-g1 cascade, PDGF stimulates the 1 pS Ca21 channel via the GRB-2 (growth factor receptor binding protein 2)/SOS (son of sevenless)/Ras pathway. Interestingly, several groups have shown that EPO can also stimulate GRB-2 and the Ras pathway [19, 40]. These results suggest that convergent signaling pathways are responsible for EPO receptor-operated Ca21 influx and involve cytosolic Src tyrosine kinases, the PLC-g1 cascade, and perhaps Ras oncogenes. Pathological and physiological significance of the erythropoietin-induced Ca21 channel The administration of recombinant human EPO has become an important clinical therapeutic adjunct to the treatment of patients with chronic anemias (such as renal failure, myelodysplasia, chemotherapy). However, the bone marrow is not the only EPO target tissue and as a result, EPO administration has been associated with complications that are independent of its hematologic effects, most notably an increased incidence of hypertension [1–5]. Our present study suggests that a potential mechanism for clinically observed EPO-induced vasoconstriction involves extracellular Ca21 influx via EPO receptor-operated, 1 pS Ca21
channels. However, other mechanisms may be involved in EPOinduced hypertension. Erythropoietin is a well-characterized cytokine growth factor responsible for the proliferation of erythroid progenitor cells [7, 10, 16 –21, 39, 40]. Therefore, it is perhaps not surprising that in addition to being a direct vasoconstrictor, EPO has been shown to act as a vascular growth factor mediating the proliferation of endothelial cells and aortic rings [6, 7]. In the present study, we have shown that EPO stimulates Src tyrosine kinase, PLC-g1 and 1 pS Ca21 channel activity in glomerular mesangial cells, which are phenotypically similar to vascular smooth muscle cells. Predictably, we have found that EPO also stimulates the tyrosine phosphorylation of PLC-g1 and PLC-g1/EPO receptor complex formation in rat aortic vascular smooth muscle cells (unpublished data). The importance of tyrosine kinases, including the cytosolic Src family, and PLC-g1 in controlling cellular growth and differentiation is now well established [35]. DNA synthesis and PLC activity is blocked by tyrosine kinase inhibition (genistein or herbimycin A) and stimulated by tyrosine phosphatase inhibition (vanadate) in human glomerular mesangial cells [26, 27]. In fibroblasts, microinjection of PLC-g1 or PLC-g1 antibodies has been shown to induce or inhibit DNA synthesis, respectively [26, 27]. Higher levels of PLC-g are found in cancer cells and may play a role in abnormal cellular proliferation and mitogenesis [26, 27]. The initiation of cellular DNA synthesis and proliferation requires the influx of extracellular Ca21 [26, 27]. Conversely, inhibition of Ca21 entry blocks mitogenic responses [26, 27]. Finally, EPO activates multiple mitogenic signaling pathways in addition to PLC (such as Ras/mitogen-activated protein kinase, JAK/STAT), which promote the activation of early growth response genes in hematopoietic cells [7, 10, 16 –21, 39, 40].
Marrero et al: Ca21 channel activation by PLC-g1
1267
Fig. 7. Model for 1 pS Ca21 channel activation by erythropoietin (EPO) in glomerular mesangial cells. (Left) Binding of EPO to its receptor stimulates tyrosine phosphorylation of phospholipase C (PLC)-g1, translocation of PLC-g1 from the cytosol to the plasma membrane, and membrane complex formation between PLC-g1 an the EPO receptor itself. Subsequent activation of PLC-g1, promotes PIP2 hydrolysis and the generation of IP3. (Right) Ca21 is released from IP3-dependent intracellular stores and triggers 1 pS Ca21 channel activity. Ca21 entry through the channel allows for sustained activation of this Ca21-sensitive channel. Increases in intracellular Ca21 levels ultimately leads to mesangial cell contraction and proliferation.
Activation of all these same mitogenic factors by classic growth factors and angiotensin II has been strongly implicated in the clinical pathogenesis of maladaptive cellular growth in pathologic disease processes (such as hypertension, cardiac failure, renal failure) [27–29, 35]. Our study raises the concern that exogenous EPO administration for the treatment of anemia may trigger maladaptive growth in non-hematopoietic cells (such as mesangial and vascular smooth muscle cells), potentially leading to the acceleration of glomerulosclerosis, arteriosclerosis, and atherosclerosis. In rats, anemia retards the progression of proteinuria and glomerulosclerosis after five-sixths nephrectomy [43]. However, glomerular injury and renal failure was accelerated when these animals were treated with EPO. Conclusions Other investigators have shown that the EPO-induced increase in intracellular Ca21 in vascular smooth muscle and hematopoietic cells is due to extracellular Ca21 influx via a voltageindependent Ca21 conductance. Our studies provide a candidate pathway involving: (1) EPO binding to EPO receptors, which leads to tyrosine phosphorylation of the -g1 isoenzyme of PLC and membrane translocation of PLC-g1, where it forms a complex with the EPO receptor itself. (2) PLC-g1-mediated hydrolysis of PIP2 increases intracellular IP3. (3) Stimulation of Ca21-activated 1 pS Ca21 channels is initially triggered by intracellular Ca21 release from IP3-dependent stores and is sustained by extracellular Ca21 entry via the channel itself (Fig. 7). Mene and associates [30] have shown that the release of intracellular Ca21 stores on extracellular Ca21 entry, promote voltage-independent Ca21 entry in human glomerular mesangial cells. They proposed that “the InsP3 sensitive Ca21 pool may serve as a trigger for a sustained response, thus initiating a cascade amplification signaling mechanism.” Our results provide a potential mechanism for the common
hypertensive complications associated with the therapeutic administration of EPO. ACKNOWLEDGMENTS Support was provided by National Institutes of Health Grants K08DK02111 and P01-DK50268, a Veterans Affairs Merit Review Award, an American Diabetes Association Research Award, National and Georgia Affiliate American Heart Association (AHA) Grant-In-Aid Awards, and an AHA Minority Scientist Developmental Award (M.B. Marrero). Dr. Shun M. Ling provided insightful scientific and clinical input during this project. Preliminary work was presented at the American Society of Nephrology Annual Meeting in New Orleans, LA (November 3– 6, 1996). Ms. Amy Ling and Ms. Elaine O’Dell provided editorial assistance. Reprint requests to Mario B. Marrero, Ph.D., Emory University School of Medicine, The Center for Cell and Molecular Signaling, Physiology Building, 1638 Pierce Drive, N.E., Atlanta, Georgia 30322, USA. E-mail: [email protected]
REFERENCES 1. NEUSSER M, TEPEL M, ZIDEK W: Erythropoietin increases cytosolic free calcium concentration in vascular smooth muscle cells. Cardiovasc Res 27:1233–1236, 1993 2. VAZIRI ND, ZHOU XJ, SMITH J, OVEISI F, BALDWIN K, PURDY RE: In vivo and in vitro pressor effects of erythropoietin in rats. Am J Physiol 269(Renal Fluid Electrolyte Physiol):F838 –F845, 1995 3. KOKOT F, WIECEK A: Arterial hypertension in uraemic patients treated with erythropoietin. Nephron 71:127–132, 1995 4. LEBEL M, KINGMA I, GROSE JH, LANGLOIS S: Effect of recombinant human erythropoietin therapy on ambulatory blood pressure in normotensive and in untreated borderline hypertensive hemodialysis patients. Am J Hypertens 8:545–551, 1995 5. HEIDENREICH S, RAHN K-H, ZIDEK W: Direct vasopressor effect of recombinant human erythropoietin on renal resistance vessels. Kidney Int 39:259 –265, 1991 6. CARLINI RG, REYES AA, ROTHSTEIN M: Recombinant human erythropoietin stimulates angiogenesis in vitro. Kidney Int 47:740 –745, 1995
1268
Marrero et al: Ca21 channel activation by PLC-g1
7. HALLER H, CHRISTEL C, DANNENBERG L, THIELE P, LINDSCHAU C, LUFT FC: Signal transduction of erythropoietin in endothelial cells. Kidney Int 50:481– 488, 1996 8. TEPEL M, WISCHNIOWSKI H, ZIDEK W: Erythropoietin increases cytosolic free calcium concentration and thrombin induced changes in cytosolic free calcium in platelets from spontaneously hypertensive rats. Biochem Biophys Res Comm 776:313–316, 1984 9. TEPEL M, WISCHNIOWSKI H, ZIDEK W: Erythropoietin induced transmembrane calcium influx in essential hypertension. Life Sci 51:161– 167, 1992 10. REN H-Y, KOMATSU N, SHIMIZU R, OKADA K, MIURA Y: Erythropoietin induces tyrosine phosphorylation and activation of phospholipase C-g1 in a human erythropoietin-dependent cell line. J Biol Chem 269:19633–19638, 1994 11. MILLER BA, BELL LL, LYNCH CJ, CHEUNG JY: Erythropoietin modulation of intracellular calcium: A role for tyrosine phosphorylation. Cell Calcium 16:481– 490, 1994 12. MILLER BA, SCADUTO RC JR, TILLOTSON DL, BOTTI JJ, CHEUNG JY: Erythropoietin stimulates a rise in intracellular free calcium concentration in single early human erythroid precursors. J Clin Invest 82:309 –315, 1988 13. MILLER BA, BELL L, HANSEN CA, ROBISHAW JD, LINDER ME, CHEUNG JY: G-protein a subunit Gia2 mediates erythropoietin signal transduction in human erythroid precursors. J Clin Invest 98:1728 – 1736, 1996 14. CHEUNG JY, ELENSKY M, BRAUNEIS U, SCADUTO RC JR, BELL LL, TILLOTSON DL, MILLER BA: Ion channels in human erythroblasts: Modulation by erythropoietin. J Clin Invest 90:1850 –1856, 1992 15. MILLER BA, CHEUNG JY: Mechanisms of erythropoietin signal transduction: Involvement of calcium channels. Proc Soc Exper Biol Med 206:263–267, 1994 16. FINBLOOM DS, PETRICOIN EFI, HACKETT RH, DAVID M, FELDMAN GM, IGARASHI K-I, FIBACH E, WEBER MJ, THORNER MO, SILVA CM, LARNER AC: Growth hormone and erythropoietin differentially activate DNA-binding proteins by tyrosine phosphorylation. Mol Cell Biol 14:2113–2118, 1994 17. DAMEN JE, CUTLER RL, JIAO HY, YI TL, KRYSTAL G: Phosphorylation of tyrosine 503 in the erythropoietin receptor (EpR) is essential for binding the P85 subunit of phosphatidylinositol (PI) 3-Kinase and for EpR-associated PI 3- kinase activity. J Biol Chem 270:23402– 23408, 1995 18. SAWYER ST, PENTA K: Association of JAK2 and STAT5 with erythropoietin receptors-Role of receptor phosphorylation in erythropoietin signal transduction. J Biol Chem 271:32430 –32437, 1996 19. DAMEN JE, LIU L, CUTLER RL, KRYSTAL G: Erythropoiein stimulates the tyrosine phosphorylation of Shc and its association with Grb2 and a 145-Kd tyrosine phosphorylated protein. Blood 82:2296 –2303, 1993 20. MAYEUX P, DUSANTER-FOURT I, MULLER O, MAUDUIT P, SABBAH M, DRUKER B, VAINCHENKER W, FISCHER S, LACOMBE C, GISSELBRECHT S: Erythropoietin induces the association of phosphatidylinositol 39-kinase with a tyrosine-phosphorylated protein complex containing the erythropoietin receptor. Eur J Biochem 216:821– 828, 1993 21. MIURA O, NAKAMURA N, IHLE JN, AOKI N: Erythropoietin-dependent association of phosphatidylinositol 3-kinase with tyrosine-phosphorylated erythropoietin receptor. J Biol Chem 269:614 – 620, 1994 22. PAXTON WG, MARRERO MB, KLEIN JD, DELAFONTAINE P, BERK BC, BERNSTEIN KE: The angiotensin II AT1 receptor is tyrosine and serine phosphorylated and can serve as a substrate for the src family of tyrosine kinases. Biochem Biophys Res Comm 200:260 –267, 1994 23. MARRERO MB, SCHIEFFER B, PAXTON WG, SCHIEFFER E, BERNSTEIN KE: Electroporation of pp60c-src antibodies inhibits angiotensin II activation of phospholipase C-g1 in rat aortic smooth muscle cells. J Biol Chem 270:15734 –15738, 1995 24. MARRERO MB, SCHIEFFER B, MA H, BERNSTEIN KE, LING BN: Ang II-induced tyrosine phosphorylation stimulates phospholipase C-g1
25.
26. 27. 28.
29. 30. 31. 32. 33. 34. 35. 36.
37.
38.
39. 40.
41. 42. 43.
and Cl2 channels in glomerular mesangial cells. Am J Physiol 270: C1834 –C1842, 1996 MARRERO MB, PAXTON WG, DUFF JL, BERK BC, BERNSTEIN KE: Angiotensin II stimulates tyrosine phosphorylation of phospholipase C-g1 in vascular smooth muscle cells. J Biol Chem 269:10935–10939, 1994 MOGAMI H, KOJIMA I: Stimulation of calcium entry is prerequisite for DNA synthesis induced by platelet-derived growth factor in vascular smooth muscle cells. Biochem Biophys Res Comm 196:650 – 658, 1993 LING BN, MATSUNAGA H, MA H, EATON DC: Role of growth factors in mesangial cell ion channel regulation. Kidney Int 48:1158 –1166, 1995 MA H, MATSUNAGA H, LI B, SCHIEFFER B, MARRERO MB, LING BN: Ca21 channel activation by PDGF-induced tyrosine phosphorylation and Ras GTP-binding proteins in rat glomerular mesangial cells. J Clin Invest 97:2332–2341, 1996 MA H, MATSUNAGA H, LI B, MARRERO MB, LING BN: Regulation of PDGF b receptor-operated Ca21 channels by phospholipase C-g1 in glomerular mesangial cells. Am J Physiol 271:F994 –F1003, 1996 MENE P, TETI A, PUGLIESE F, CINOTTI GA: Calcium release-activated calcium influx in cultured human mesangial cells. Kidney Int 46:122– 128, 1994 MATSUNAGA H, LING BN, EATON DC: Ca21-permeable channel associated with platelet-derived growth factor receptor in mesangial cells. Am J Physiol 267:C456 –C465, 1994 LING BN, SEAL EE, EATON DC: Regulation of mesangial cell ion channels by insulin and angiotensin II: Possible role in diabetic glomerular hyperfiltration. J Clin Invest 92:2141–2151, 1993 SEAL EE, EATON DC, GOMEZ LM, MA H, LING BN: Extracellular glucose reduces mesangial cell ion channel responsiveness to angiotensin II. Am J Physiol 269:F389 –F397, 1995 WAHL M, LUCHERINI MJ, GRUENSTEIN E: Intracellular Ca21 measurement with Indo-1 in substrate-attached cells: Advantages and special considerations. Cell Calcium 11:487–500, 1990 SCHIEFFER B, DREXLER H, LING BN, MARRERO MB: G proteincoupled receptors control vascular smooth muscle cell proliferation via pp60c-src and p21ras. Am J Physiol 272:C2019 –C2030, 1997 KUMJIAN DA, WAHL MI, RHEE SG, DANIAL TO: Platelet-derived growth factor (PDGF) binding promotes physical association of PDGF receptor with phospholipase C. Proc Natl Acad Sci USA 86:8232– 8236, 1989 MORRISON DK, KAPLAN DR, RHEE SG, WILLIAMS LT: Plateletderived growth factor (PDGF)-dependent association of phospholipase C-gamma with the PDGF receptor signaling complex. Mol Cell Biol 10:2359 –2366, 1990 HANKE JH, GARDNER JP, DOW RL, CHANGELIAN PS, BRISSETTE WH, WERINGER EJ, POLLOK BA, CONNELLY PA: Discovery of a novel, potent, and Src family-selective tyrosine kinase inhibitor. J Biol Chem 271:695–701, 1996 PENTA K, SAWYER ST: Erythropoietin induces the tyrosine phosphorylation, nuclear translocation, and DNA binding of STAT1 and STAT5 in erythroid cells. J Biol Chem 270:31282–31287, 1995 TORTI M, MARTI KB, ALTSCHULER D, YAMAMOTO K, LAPETINA EG: Erythropoietin induces p21ras activation and p120GAP tyrosine phosphorylation in human erythroleukemia cells. J Biol Chem 267:8293, 1992 ALLBRITTON NL, MEYER T, STRYER L: Range of messenger action of calcium ion and inositol 1,4,5-trisphosphate. Science 258:1812–1815, 1992 KARGACIN GJ: Calcium signaling in restricted diffusion spaces. Biophys J 67:262–272, 1994 GARCIA DL, ANDERSON S, RENNKE HG, BRENNER BM: Anemia lessens and its prevention with recombinant human erythropoietin worsens glomerular injury and hypertension in rats with reduced renal mass. Proc Natl Acad Sci USA 85:6142– 6146, 1988
Erythropoietin receptor-operated Ca21 channels: Activation by phospholipase C-g1 MARIO B. MARRERO, RICHARD C. VENEMA, HEPING MA, BRIAN N. LING, and DOUGLAS C. EATON Renal Division and The Center for Cell and Molecular Signaling, Departments of Pathology and Medicine, Emory University School of Medicine and Veterans Affairs Medical Center, Atlanta; and Vascular Biology Center, Medical College of Georgia, Augusta, Georgia, USA
Erythropoietin receptor-operated Ca21 channels: Activation by phospholipase C-g1. Erythropoietin (EPO) increases Ca21 influx in vascular smooth muscle cells and acts both as a direct vasoconstrictor and vascular growth factor (that is, angiogenesis). However, the mechanism by which EPO promotes extracellular Ca21 entry in contractile cells has not been elucidated. In hematopoietic cells, EPO induces tyrosine kinase (TK)dependent activation of phospholipase C (PLC)-g1 and Ca21 influx via a voltage-independent Ca21 conductance. In contractile mesangial cells, we have recently characterized a voltage-independent, 1 pS Ca21 channel that is dependent on both TK and PLC-g1 activity. Therefore, we examined cultured rat glomerular mesangial cells after timed exposure to recombinant human EPO (20 U/ml). Erythropoietin increased the tyrosine phosphorylation of PLC-g1, promoted membrane complex formation between PLC-g1 and the EPO receptor itself, and raised the levels of intracellular inositol 1,4,5-trisphosphate and intracellular Ca21. Consistent with our previous studies, 1 pS Ca21 channel activity was extremely low under basal, unstimulated conditions in cell-attached patches, but was dramatically increased when EPO was present in the patch pipette. Tyrosine kinase inhibition with 100 mM genistein or 1 mM PP1 (Src; selective tyrosine kinase inhibitor) prevented all of these EPO-induced responses. We conclude that: (1) EPO-induced stimulation of 1 pS Ca21 channels is mediated via a cytosolic Src TK in glomerular mesangial cells. (2) Stimulation of this Ca21-activated, Ca21-permeable channel is dependent on the tyrosine phosphorylation/activation of PLC-g1. (3) This cascade provides a possible mechanism for the vasoconstriction and hypertension observed with clinical EPO use for the treatment of chronic anemias.
Human recombinant erythropoietin (EPO) therapy for the treatment of chronic anemias is associated with an increased incidence of new onset hypertension and a worsening of preexisting hypertension [1–5]. Erythropoietin acts as both a direct vasoconstrictor and a vascular growth factor (that is, angiogenesis) [2, 5–7]. In platelets from hypertensive patients and hypertensive rat models, but not in platelets from normotensive patients or rats, EPO elevates cytosolic Ca21 levels [8, 9]. Similarly, EPO increases intracellular Ca21 concentrations in vascular smooth muscle cells [1]. However, the mechanism by which EPO mediates
Key words: tyrosine kinase, intracellular Ca21, renal failure, hypertension, anemia. Received for publication May 2, 1997 and in revised form November 25, 1997 Accepted for publication November 26, 1998
© 1998 by the International Society of Nephrology
increases in intracellular Ca21 levels in contractile cells has not been elucidated. In hematopoietic cells, EPO induces tyrosine kinase-dependent activation of phospholipase C (PLC)-g1 and extracellular Ca21 influx via a voltage-independent, EPO receptor-operated Ca21 conductance pathway [8, 10 –15]. The g isoforms of PLC are classically activated after tyrosine phosphorylation by growth factor receptor-linked tyrosine kinase activity [such as plateletderived growth factor (PDGF), epidermal growth factor, fibroblast growth factor]. However, unlike classic growth factor receptors, EPO and other similar cytokine receptors do not possess intrinsic tyrosine kinase activity [16]. Despite this fact, EPO has been shown to stimulate rapid protein tyrosine phosphorylation in hematopoietic cells [11, 16 –21]. Cytosolic protein tyrosine kinases, which are activated by receptors lacking intrinsic kinase activity (such as interleukins, interferons, angiotensin II), include members of the Src family (such as p60src, p59fyn, p56lck, p62yes) [22]. We have shown that p60src indeed does promote the tyrosine phosphorylation and activation of PLC-g1, leading to the hydrolysis of phosphatidylinositol 4,5-bisphosphate (PIP2) and the generation of inositol 1,4,5-trisphosphate (IP3) [23–25]. This PLC-g1 cascade stimulates a biphasic rise in intracellular Ca21 consisting of an initial release from IP3-sensitive intracellular Ca21 stores followed by an influx of extracellular Ca21. In non-excitable tissues, including glomerular mesangial and vascular smooth muscle cells, the influx of extracellular Ca21 is an important prerequisite for contraction and the initiation of cellular DNA synthesis and proliferation [26, 27]. Evidence suggests that this extracellular Ca21 influx occurs via voltage-independent, receptor-operated Ca21 channels [28 –30]. In glomerular mesangial cells and vascular smooth muscle cells, we have recently identified PLC-g1 and have shown that PLC-g1 is tyrosine phosphorylated and activated by both receptors possessing (for example, PDGF-b receptor) and lacking (such as angiotensin II receptor) intrinsic tyrosine kinase activity [23–25, 29]. Our group has also characterized a voltage-independent, receptor-operated Ca21 channel in rat glomerular mesangial cells [27–29, 31]. There are approximately 1000 of these low conductance (1 pS) Ca21-permeable channels channels per cell (or 4 3 107 channels/cm2). The channels are inactive under basal conditions; they are strongly activated by growth factors (such as PDGF); and, to be active, they require activation of at least one tyrosine kinase and activation of PLC-g1. All of these features
1259
1260
Marrero et al: Ca21 channel activation by PLC-g1
make this 1 pS Ca21 channel a potential candidate for mediating the previously reported voltage-independent, EPO receptor-operated Ca21 entry [8, 11–15]. In the present study, we investigated the role of tyrosine kinases, PLC-g1 tyrosine phosphorylation, PLC-g1/EPO receptor complex formation, and intracellular Ca21 in the regulation of the 1 pS Ca21 channel by EPO in cultured rat glomerular mesangial cells. METHODS Preparation of rat glomerular mesangial cell cultures Glomerular mesangial cells were cultured using previously described methods [24, 28, 31–33]. Briefly, renal cortices were dissected from male Sprague-Dawley rats (75 to 150 g). Mesangial cell-enriched glomerular cores were isolated by differential sieving and incubated for 45 to 60 minutes with collagenase (1200 U/ml) in Ca21/Mg21 free Hank’s balanced salt solution (Irvine Scientific, Santa Ana, CA, USA). The mesangial cell suspension was washed and plated in standard RPMI-1640 medium supplemented with 17% (vol/vol) fetal bovine serum, 2 mM glutamine, 5 mg/ml insulin, 5 mg/ml transferrin, 5 ng/ml selenium, 100 U/ml penicillin, 100 mg/ml streptomycin, and 2 mg/ml amphotericin B at 37°C in 5% CO2/95% air. For immunoblot analysis and measurement of intracellular IP3 levels, mesangial cell subpassages 3 to 5 were grown to near confluence in 100 mm culture dishes (Fisher Scientific, Pittsburgh, PA, USA). To growth arrest the cells, medium was replaced with serum-free medium for 24 to 48 hours prior to each experiment. For patch clamp experiments, mesangial cell subpassages 3 to 5 were grown to confluency on glass coverslips. Immunoprecipitation of phospholipase C-g1 in mesangial cell lysates Mesangial cells, grown in 100 mm dishes to near confluence, were growth-arrested in 5 ml of serum-free RPMI-1640 for 24 to 48 hours. Cultures were then exposed to EPO for timed periods. Incubations were terminated by aspirating the medium and washing two to three times in ice-cold PBSV (phosphate-buffered saline with 1 mM Na3VO4). Each dish was then treated with 2 ml of ice-cold lysis buffer (25 mM Tris-HCl, pH 7.4, 1% Triton X-100, 10% glycerol, 10 mM Na3P2O7, 50 mM NaF, 1 mM Na3VO4, 1 mM PMSF and 10 mg/ml aprotonin) and placed on ice for 30 minutes. The thawed cells were harvested by scraping and sonicated for five seconds. with a Cole Parmer ultrasonic homogenizer, and then the homogenate was centrifuged at 6,000 3 g for 20 minutes. The soluble extracts were precleared with 20 to 30 ml/ml cell lysate of Immunoprecipitin (Life Technologies, Inc., Grand Island, NY, USA), which are membranes of formalin-fixed, heat-killed Staphylococcus aureus bacteria. To immunoprecipitate phosphotyrosine proteins from the cleared lysate, we used 5 mg/ml antiphosphotyrosine (PY20) monoclonal antibody (Transduction Laboratories, Lexington, KY, USA). To immunoprecipitate PLC-g1 directly, we used 5 mg/ml of a mixed monoclonal anti-PLC-g1 antibody (Upstate Biotechnology Inc., Lake Placid, NY, USA) as previously described [24]. The immunoprecipitates were then recovered by centrifugation and washed four times with ice-cold wash buffer (25 mM Tris-HCl, pH 7.4, 0.1% Triton X-100, 150 mM NaCl, and 2 mM Na3V04). Immunoprecipitated proteins were dissolved in 80 ml of Laemmli buffer, boiled for five minutes at 95°C, and analyzed on 10% gels by SDS-PAGE. Proteins were
then transferred to nitrocellulose membranes for 16 hours at 100 mA. The membrane was blotted with either anti-PLC-g1 monoclonal antibody or antiphosphotyrosine (4G10) monoclonal antibody (Upstate Biotechnology). The proteins were detected using Enhanced Chemiluminescense (Amersham Corp., Arlington Heights, IL, USA). Immunoprecipitation of EPO receptor/PLC-g1 complex in mesangial cell membrane fractions Mesangial cells, grown to near confluence, were growth-arrested in 5 ml of serum-free RPMI-1640 for 24 hours and then exposed to EPO. To terminate the incubation, the cells were washed three times with ice cold PBS plus 1 mM Na3V04. Hypotonic buffer (50 mM Tris-HCl, pH 7.4, 2 mM Na3VO4, 50 mM NaF, 10 mM Na2P2O7, 1 mM phenylmethylsulfonyl fluoride, 10 mg/ml leupeptin, 10 mg/ml aprotinin) was added and the cells were then placed on ice to thaw for 30 minutes. The thawed cells were scraped from the plate and centrifuged for 20 minutes at 6,000 3 g at 4°C. The supernatant was discarded and to the pellet was added 1 ml of ice-cold lysis buffer, followed by incubation for one hour at 4°C. After centrifugation for 20 minutes at 6,000 3 g at 4°C, the supernatant was collected, anti-EPO receptor antibody (5 ml/ml lysate) was added, followed by incubation overnight at 4°C. The next day, Protein A/G plus 30 ml of buffer (2 mM Na3VO4, 50 mM NaF, 10 mM Na2P2O7, 1 mM phenylmethylsulfonyl fluoride, 10 mg/ml leupeptin, 10 mg/ml aprotinin) was added while rocking at 4°C for two to three hours. The immunoprecipitates were then recovered by centrifugation and washed three times with ice-cold wash buffer (50 mM Tris-HCl, pH 7.4, 0.1% NP-40, 150 mM NaCl, 1 mM Na3VO4). Immunoprecipitated membrane proteins were dissolved in 80 ml of Laemmli buffer, boiled for five minutes at 95°C, and analyzed on 10% gels by SDS-PAGE. Proteins were then transferred to nitrocellulose membranes and blotted with anti-PLC-g1 antibody. The proteins were detected using enhanced chemiluminescense. Measurement of inositol 1,4,5-trisphosphate Following mesangial cell exposure to 20 U/ml EPO, the incubation was stopped by the addition of 1 ml of 100% ice-cold trichloroacetic acid to the plates. The plates were placed on ice for 10 minutes. The cells were then harvested by scraping and transferring to polyethylene tubes. The cell extract was sonicated on ice with a Cole Parmer ultrasonic homogenizer for approximately one minute and then centrifuged for 10 minutes at 6,000 3 g. The supernatant was removed and warmed to room temperature for 15 minutes. Levels of IP3 in each supernatant were determined using a radioimmunoassay based kit (Dupont NEN, Boston, MA, USA) as previously described [24, 29]. Confocal intracellular Ca21 fluorescent imaging To measure intracellular Ca21 levels, fluorescence was measured with mesangial cells grown on glass cover slips [34]. Cells were preincubated with 5 mM Indo-1 AM (Molecular Probes, Eugene, OR, USA) for 30 minutes at room temperature in darkness and then washed with saline solution three times. Cover slips were then placed in a chamber secured to the stage of a Meridian ACAS 570/Ultima laser scanning confocal microscope (Laser Cytofluorimeter Working Station; Meridian Instruments, MI, USA). Cells loaded with Indo-1 AM were excited by UV light
Marrero et al: Ca21 channel activation by PLC-g1
at 351 to 364 nm and the cytoplasmic free calcium signal was read as the fluorescence ratio at 405 nm and 530 nm wave lengths. Intracellular Ca21 concentrations were calculated by using the standard working curves of fluorescence ratio versus free Ca21 in Ca21-EGTA buffer. Patch-clamp recording TW150F-6 glass pipettes (World Precision Instruments, New Haven, CT, USA) were prepared with a PP-83 patch-pipette puller and a MF-83 pipette polisher (Narishige, Tokyo, Japan) [31–33]. Pipettes with resistances of . 10 megaohms were used for the experiments. Single-channel patch clamp configurations, including the cell-attached and inside-out modes, were established on the membrane of mesangial cells cultured on glass coverslips. In the cell-attached mode, single channel currents were recorded without applying potential to the pipette (that is, resting membrane potential). In the excised inside-out mode, single channel recordings were obtained with 160 mV applied pipette voltage (Vm 5 260 mV). All experiments were conducted with an Axopatch-1D amplifier (Axon Instruments, Foster, CA, USA) at room temperature, 22 to 23°C. Only patches with a seal resistance . 30 gigaohms and containing currents without baseline drift were used for experiments. Data were digitally recorded with a DAS 601 Pulse Code Modulator (Dagan Corp., MN, USA) and a SL-HF860D VCR (Sony Corp. of America, NJ, USA), and acquired using a 902LPF 8-pole Bessel filter (Frequency Devices Inc., MA, USA), TL-2 acquisition hardware and Axotape software (Axon Instruments). Single channel traces were sampled at 2 msec/point and software filtered at 100 Hz, and inward Ca21 or Mn21 current (pipette to cell) was represented by downward channel transitions. Single channel analysis was performed on IBM PC class computers using both Pclamp 6.0 and locallydeveloped software [31–33]. NPo (number of channels times the open probability) was used as a measure of channel activity. All NPo values were calculated from three minutes of single channel recording and are reported as mean values 6 SEM. When cellattached patch experiments were conducted in a paired fashion (that is, data from each patch membrane serving as the control for an experimental manipulation), the average change in NPo for a group of patches, compared before and after an experimental treatment, was analyzed using a paired t-test or analysis of variance (ANOVA) for multiple comparisons. For unpaired experiments, the traditional t-test or ANOVA was used. Results were considered significant if P , 0.05. Statistical analysis was performed with SigmaPlot and SigmaStat software (Jandel Scientific, CA, USA). Solutions and chemicals For cell-attached patch experiments, the standard extracellular bath solution (outside the patch pipette) contained (in mM) 140 NaCl (after adjusting pH to 7.4 with NaOH), 4 KCl, 1 CaCl2, 1 MgCl2 and 10 HEPES. When excised inside-out patch experiments were performed, the “cytoplasmic” bath solution contained (in mM) 140 KCl (after adjusting pH to 7.4 with KOH), 10 NaCl, 1 MgCl2, 2 EGTA and 10 HEPES. The “cytoplasmic” bath CaCl2 was varied for different experiments and was reported as the final free Ca21 activity. Since we were primarily interested in the function of the specific EPO- (and PDGF-) activated calcium channel and since the unit conductance of this channel was quite low conductance, we used a solution in the patch pipette which
1261
was high in calcium to the exclusion of other cations similar to the solution that had allowed us to first identify the channel [31]. The solution contained (in mM) 110 CaCl2, 4 KCl (after adjusting pH to 7.4 with KOH), 1 MgCl2, and 10 HEPES. Tween 20, acrylamide, N-N9-methylene-bisacrylamide, N,N,N9,N9-tetramethylenediamine, and SDS were purchased from Bio-Rad Labs (Hercules, CA, USA). Prestained molecular weight standards, genistein, and Immunoprecipitin were obtained from Life Technologies, Inc. (Grand Island, NY, USA). Penicillin, streptomycin, insulin, PP1 and Indo-1 were purchased from Calbiochem Corp. (San Diego, CA, USA) and Sigma Chemical Corp. (St. Louis, MO, USA). Human recombinant erythropoietin was obtained from Amgen (Thousand Oaks, CA, USA). Nitrocellulose membranes and sheep anti-mouse IgG were purchased from Amersham Corp. Monoclonal antibodies to PLC-g1 and antiphosphotyrosine (4G10) were obtained form Upstate Biotechnology, Inc. Goat anti-mouse IgG was obtained from Jackson ImmunoResearch Laboratories, Inc. (West Grove, PA, USA). Antiphosphotyrosine (PY20) and EPO receptor antibodies were obtained from Transduction Laboratories and protein A/G plus agarose came from Santa Cruz Biotechnologies, Inc. (Santa Cruz, CA, USA). RESULTS Effect of erythropoietin on phospholipase C-g1 in glomerular mesangial cells Ren et al [10] have shown that EPO stimulates the tyrosine phosphorylation and activation of PLC-g1 in a hematopoietic cell line. In glomerular mesangial and vascular smooth muscle cells, we have previously shown that classic growth factors, such as PDGF, promote the tyrosine phosphorylation and activation of PLC-g1 [25, 27, 28, 31]. However, we have also shown that PLC-g1 can be activated by receptors that, like the EPO receptor, lack intrinsic tyrosine kinase activity, but rather operate via cytosolic tyrosine kinases (such as angiotensin II) [23–25, 35]. Therefore, we examined EPO-induced tyrosine phosphorylation of PLC-g1 in cultured rat glomerular mesangial cells by immunoprecipitation. Tyrosine phosphorylated proteins were first immunoprecipitated using antiphosphotyrosine antibodies and then probed with anti-PLC-g1 antibodies. When compared to basal levels, exposure to 20 U/ml EPO produced an increase in the tyrosine phosphorylation of PLC-g1 within 30 seconds (N 5 3; Fig. 1). Peak phosphorylation was observed within five minutes of EPO exposure and then returned to baseline by 30 minutes. We also performed the experiment reversing the order of antibody addition, first immunoprecipitating mesangial cell lysates with anti-PLC-g1 antibodies and then probing with antiphosphotyrosine antibody. Again, a significant increase in the tyrosine phosphorylation of PLC-g1 was observed upon exposure to 20 U/ml EPO (N 5 3; data not shown). The peak phosphorylation response was again observed within five minutes. of EPO exposure. In mesangial cells which were first preincubated with tyrosine kinase inhibitors, 100 mM genistein for one hour (N 5 3) or 1 mM PP1 for 15 minutes (N 5 3), the usual EPO-induced tyrosine phosphorylation of PLC-g1 was abolished (Fig. 1). Membrane complex formation between phospholipase C-g1 and the erythropoietin receptor We and other groups have previously shown that growth factor-mediated tyrosine phosphorylation of PLC-g1 promotes
1262
Marrero et al: Ca21 channel activation by PLC-g1
Fig. 1. Erythropoietin (EPO)-induced tyrosine phosphorylation of phospholipase C (PLC)-g1. (Western blot, inset) Mesangial cells exposed to 20 U/ml EPO for various timed periods. Mesangial cell lysates were then immunoprecipitated with antiphosphotyrosine antibody (PY-20) and probed with anti-PLC-g1 antibody. The PLC-g1 bands were quantitated by densitometry using a La Cie scanner interfaced with a personal computer. Each band was scanned in two dimensions and the density corrected for the background present in the lane. The results shown in the graph represent the corrected density for each time point and are expressed as arbitrary units plotted against time of EPO exposure in mesangial cells that were either pre-treated with tyrosine kinase inhibitors, 100 mM genistein (‚) for one hour or 1 mM PP1 (F) for 15 minutes, or vehicle (E). Each data point represents mean 6 SE for N 5 3 separate plates of mesangial cells.
the formation of a membrane-bound complex between PLC-g1 and the growth factor receptors themselves (such as PDGF-b receptor/PLC-g1 complex) [27, 28, 31, 36, 37]. Therefore, we investigated the ability of EPO to promote the membrane association of PLC-g1 with the EPO receptor itself in glomerular mesangial cells. Mesangial cells were incubated in EPO (20 U/ml) for various time periods followed by isolation of the membrane fractions. We first immunoprecipitated the isolated membrane fraction with an anti-EPO receptor antibody. Probing the immunoprecipitated proteins with an anti-PLC-g1 antibody revealed the formation of a membrane-associated PLC-g1/EPO receptor complex (N 5 3; Fig. 2). Membrane fractions were also examined from mesangial cells preincubated with either 100 mM genistein for one hour (N 5 3) or 1 mM PP1 for 15 minutes (N 5 3) prior to timed exposures to EPO. Proteins were again immunoprecipitated with anti-EPO receptor antibodies and then probed with anti-PLC-g1 antibodies. Under these conditions, the 145 kDa band (molecular weight of PLC-g1) was no longer observed, as shown in Figure 2 for cells that were preincubated with PP1. These studies suggested that tyrosine kinase activity, perhaps involving cytosolic Src kinases, was essential for EPO-induced membrane recruitment of PLC-g1 and for membrane complex formation between PLC-g1 and the EPO receptor itself in glomerular mesangial cells. Effect of erythropoietin on 1 pS Ca21 channels in glomerular mesangial cells In glomerular mesangial cells, we have previously shown that PLC-g1 tyrosine phosphorylation and PLC-g1/growth factor re-
ceptor complex formation promotes the activation of receptoroperated, 1 pS Ca21 channels [27, 28, 31]. Erythropoietin induces tyrosine kinase-dependent extracellular Ca21 influx via a voltageindependent, EPO receptor-operated Ca21 conductance pathway in hematopoietic cells [8, 11–15]. However, this voltage-independent, EPO receptor-operated Ca21 conductance has not been characterized at a single channel level. Therefore, we investigated the effect of EPO on the 1 pS Ca21 channel in cultured rat glomerular mesangial cells. Consistent with our previous studies, 1 pS Ca21 channel activity was extremely low under basal, unstimulated conditions in cell-attached patches; NPo (number of channels times the open probability) was 0.06 6 0.02 (N 5 8; Fig. 3). The addition of recombinant human EPO (20 U/ml) to the extracellular bath outside the cell-attached patch pipette did not increase channel activity recorded from the cell-attached membrane patch (NPo 5 0.09 6 0.04; N 5 7) from basal values (P 5 0.55). In contrast, channel activity was dramatically increased when recombinant human EPO (20 U/ml) was present inside the patch pipette; NPo was 0.73 6 0.09 (N 5 10). Intrapipette EPO significantly increased channel activity over basal values (P , 0.00002). We also examined the time course of EPO’s activation of channels. We presumed that the formation of the seal resulted in a step change in EPO concentration and calculated the open probability for three second windows starting from the time of seal formation. Three such records of open probability were averaged to produce the mean open probability shown in the lower left panel of Figure 3. These results demonstrate that the time course of activation is consistent with the time course of PLC
Marrero et al: Ca21 channel activation by PLC-g1
1263
Fig. 2. Erythropoietin (EPO)-induced membrane complex formation between phospholipase C (PLC)-g1 and the EPO receptor. Mesangial cells were treated for the times indicated with 20 U/ml EPO in cells that were or were not pre-treated with PP1 (1 mM) for 15 minutes prior to the addition of the EPO. The membrane fraction was then separated after hypotonic cellular lysis. Proteins were first immunoprecipitated with an anti-EPO receptor antibody and developed with an anti-PLC-g1 antibody. A representative experiment shows a significant increase in EPO/PLC-g1 complex formation within 30 seconds of EPO exposure.
phosphorylation. However, since channel opening is a stochastic process, a precise determination of the time course would require summing of many more records. Thus, we feel that, while the time course of channel activation is similar to that of PLC phosphorylation, the channel activation time should be interpreted with caution. We then investigated the effect of tyrosine kinase inhibition on EPO-induced 1 pS Ca21 channel activation. Channel activity was first recorded with cell-attached patch pipettes containing 20 U/ml EPO (Fig. 4). Genistein (100 mM) was then added to the extracellular bath outside the patch pipette [24, 28, 29]. Despite the continued presence of intrapipette EPO, tyrosine kinase inhibition abolished Ca21 channel activity, decreasing NPo from 0.56 6 0.14 to 0.08 6 0.02 (P , 0.01; N 5 8). Another set of mesangial cells were pretreated with PP1 (1 mM), an inhibitor [38], for 15 minutes prior to patching. Basal Ca21 channel NPo after PP1 pretreatment was 0.04 6 0.01 (N 5 7; Fig. 4). NPo under the latter condition was not significantly different from basal Ca21 channel activity recorded in the absence of PP1 (Fig. 3). However, PP1 prevented the usual channel activation observed when EPO was present in the cell-attached patch pipette (NPo 0.06 6 0.03; N 5 7; P 5 0.60; Fig. 4). These results suggested that EPO-induced activation of the 1 pS Ca21 channel was dependent on protein tyrosine phosphorylation likely involving a member of the Src family of cytosolic tyrosine kinases. Role of intracellular Ca21 in 1 pS Ca21 channel activation by erythropoietin Pholipase C promotes the release of Ca21 from IP3-dependent intracellular stores, which in turn triggers voltage-independent
Ca21 entry in multiple cell types, including glomerular mesangial cells [24, 25, 29, 30]. In hematopoietic and vascular cell types, EPO elevates cytosolic Ca21 levels [1, 8, 9, 11–15]. Therefore, we measured intracellular IP3 levels in glomerular mesangial cells (Fig. 5) and found that exposure to 20 U/ml EPO significantly increased IP3 levels with a time course paralleling EPO-induced tyrosine phosphorylation of PLC-g1 and membrane complex formation between PLC-g1 and the EPO receptor. Using confocal microscopy, we also measured intracellular Ca21 levels in Indo1-loaded glomerular mesangial cells. In response to 20 U/ml EPO exposure, intracellular Ca21 levels did indeed increase from basal values (Fig. 6). The peak intracellular Ca21 response occurred approximately 20 minutes after EPO exposure. Erythropoietininduced increases in intracellular IP3 and Ca21 levels were markedly attenuated by pretreatment with 100 mM genistein (1 hr) or 1 mM PP1 (15 min). Other groups have shown that intracellular Ca21 release plays an important role in the regulation of receptor-operated Ca21 entry in many non-excitable cell types [30]. In particular, we have shown that growth factor-induced increases in intracellular Ca21 leads to activation of the 1 pS Ca21 channel in glomerular mesangial cells, and that in the absence of intracellular Ca21 the channels were completely inactive [29]. Therefore, we examined the role of intracellular Ca21 on 1 pS channel activity in cellattached patches. Without EPO present in the pipettes, cellattached patches were established and excised, exposing the inside-out patch membranes to a “cytoplasmic” bath containing 1028 M Ca21; NPo was 0.007 6 0.012 (N 5 5). Channel activity increased significantly when “cytoplasmic” bath Ca21 was raised to 1027 M (NPo 5 0.25 6 0.08; P , 0.01) and then to 1026 M (NPo
1264
Marrero et al: Ca21 channel activation by PLC-g1
Fig. 3. Erythropoietin (EPO)-induced 1 pS Ca21 channel activity. (Left) Representative traces depict 1 pS Ca21 channel activity in cell-attached patches on glomerular mesangial cells. Channel activity was measured in the absence of EPO (Control), in the presence of 20 U/ml in the extracellular bath outside the patch pipette, and in the presence of 20 U/ml inside the patch pipette. Inward current is represented by downward deflections. “C” marks the zero current level (closed state) and O1 marks channel openings. (Lower left) The time course of EPO’s activation of channels. We presumed that the formation of the seal resulted in a step change in EPO concentration and calculated the open probability for three seconds windows starting from the time of seal formation. Three such records of open probability were averaged to produce the mean open probability. These results demonstrate that despite the significant variability in the open probability (as expected for any single channel recording), the time course of activation is consistent with the time course of PLC phosphorylation. (Right) Summary plot of NPo for all cell-attached patch experiments.
5 0.65 6 0.10; P , 0.02). Combined with our biochemical results, these findings suggested that EPO-induced activation of 1 pS Ca21 channels occurs via the tyrosine phosphorylation and activation of PLC-g1 leading to the localized release of IP3-dependent intracellular Ca21 stores. DISCUSSION Erythropoietin-induced 1 pS Ca21 channel activation is dependent on Src-specific tyrosine kinases Erythropoietin has been shown to stimulate voltage-independent Ca21 influx in multiple cell types, including vascular smooth muscle and hematopoietic cells [8, 11–15]. Our previous patch clamp studies have characterized voltage-independent, PDGF-b receptor-operated 1 pS Ca21 channels in glomerular mesangial cells [27–29, 31]. In our present study, we find that EPO also activates the same 1 pS Ca21 channel in cell-attached patches on cultured rat glomerular mesangial cells. Channel activation by EPO is significantly reduced by two structurally different tyrosine kinase inhibitors: genistein and PP1. We have previously shown
that tyrosine kinase inhibition (such as with genistein and tryphostin 9) blocks the activation of 1 pS Ca21 channels by PDGF in glomerular mesangial cells [28, 29]. Genistein has also been shown to block receptor-operated Ca21 influx in rat pancreatic acinar cells, human fibroblasts, and mouse osteoblast-like cells [28]. While genistein is a nonspecific tyrosine kinase inhibitor that competes for the ATP binding site, PP1 is a Src-specific tyrosine kinase inhibitor [38]. Cytokine receptors related to the EPO receptor (such as interleukin, interferon) lack intrinsic receptor tyrosine kinase activity, but have been shown to activate cytosolic Src tyrosine kinases in multiple cell types [16, 24, 28]. Erythropoietin promotes tyrosine phosphorylation and activation of phospholipase C-g1 in mesangial cells In multiple cell types, a number of proteins are tyrosine phosphorylated in response to EPO. These tyrosine phosphorylated proteins include phospholipase C-g1, the EPO receptor itself, DNA binding proteins (p84, p91, p93), phosphatidylinositol 3 kinase (PI3K), p120 GTPase-activating protein (GAP), p52shc,
Marrero et al: Ca21 channel activation by PLC-g1
1265
Fig. 4. Effect of tyrosine kinase inhibition on erythropoietin (EPO)-induced Ca21 channel activity. (Upper left) Channel activity induced by intrapipette EPO (20 U/ml) was recorded in cell-attached patches (upper trace). Within five minutes of adding 100 mM genistein to the extracellular bath, channel activity was significantly inhibited (lower trace). (Lower left) Cell-attached patches were established on mesangial cells after 15 minutes pretreatment with 1 mM PP1. Channel activity was measured in the absence (upper trace) and presence of intrapipette EPO (20 U/ml) (lower trace). (Right) Summary plot of NPo for all cell-attached patch experiments. Symbols connected by lines represent the change in channel activity for the same cell-attached patch before and after exposure to genistein.
STAT (signal transducers and activators of transcription) 1 and 5, and several as yet unidentified proteins ranging in molecular weight from 25 to 145 kDa [7, 10, 16 –21, 39, 40]. Phospholipase C-g1 contains four tyrosine residue sites for phosphorylation [24]. In glomerular mesangial cells, our biochemical data show that EPO binding to its respective EPO receptor leads to the tyrosine phosphorylation of PLC-g1. This in turn promotes the formation of a membrane-associated complex between PLC-g1 and the EPO receptor itself. We find that EPOinduced tyrosine phosphorylation of PLC-g1, PLC-g1/EPO receptor membrane complex formation, and increases in intracellular Ca21 are abolished by the tyrosine kinase inhibitors, genistein and PP1. Phospholipase C-mediated increases in intracellular IP3 levels promotes the mobilization of intracellular Ca21 stores [24, 29]. In mesangial cells, intracellular Ca21 concentrations are 1027 M under basal conditions and increase to 1026 M in response to IP3-mediated signal transduction pathways [29] [24]. Mene and associates [30] have recently shown that the release of intracellu-
lar Ca21 stores or ionophore-mediated Ca21 entry promotes voltage-independent Ca21 (or Mn21) entry in human glomerular mesangial cells. They hypothesized that “the InsP3-sensitive Ca21 pool may serve as a trigger for a sustained response, thus initiating a cascade amplification signaling mechanism.” In the absence of intrapipette EPO, we do indeed find that the 1 pS Ca21 channel is itself directly activated by increasing “cytoplasmic” Ca21 concentrations (1028 to 1026 M) in excised inside-out patches. Our results suggest that the Ca21 channels are initially triggered by intracellular Ca21, released from IP3-dependent stores. Sustained channel activation may be maintained by Ca21 entry through the channel itself. Ca21-permeable channels allow Ca21 entry at rates . 106 ions/sec [41]. Local Ca21 in the immediate vicinity of the channel pore can reach high concentrations [41, 42], providing a positive feedback loop for the Ca21-activated, 1 pS Ca21 channel. Finally, we find that 1 pS Ca21 channel activity in excised patches, at a given “cytoplasmic” Ca21 concentration, is higher with intrapipette EPO present than without intrapipette EPO
1266
Marrero et al: Ca21 channel activation by PLC-g1
Fig. 6. Erythropoietin (EPO)-induced intracellular Ca21 concentrations. Mesangial cells with and without pretreatment with tyrosine kinase inhibitors, 100 mM genistein for one hour or 1 mM PP1 (a tyrosine kinase inhibitor) for 15 minutes, were exposed to 20 U/ml EPO for various timed periods. Intracellular Ca21 was assessed by measuring the Indo-1 fluorescence of individual mesangial cells by confocal microscopy. Each datum point represents mean 6 SE for N individual mesangial cells. Fig. 5. Erythropoietin (EPO)-induced intracellular inositol 1,4,5triphosphate (IP3) production. Mesangial cells with and without (E) pretreatment with tyrosine kinase inhibitors, 100 mM genistein (F) for one hour or 1 mM PP1 (a tyrosine kinase inhibitor; ‚) for 15 minutes, were exposed to 20 U/ml EPO for various timed periods. IP3 was determined by radioimmunoassay. Each datum point represents mean 6 SE for N 5 3 separate plates of mesangial cells.
present (data not shown). This is consistent with the activation of EPO receptor-operated Ca21 channels by another signal transduction pathway besides PLC-g1. In mesangial cells, we have recently shown that in addition to the PLC-g1 cascade, PDGF stimulates the 1 pS Ca21 channel via the GRB-2 (growth factor receptor binding protein 2)/SOS (son of sevenless)/Ras pathway. Interestingly, several groups have shown that EPO can also stimulate GRB-2 and the Ras pathway [19, 40]. These results suggest that convergent signaling pathways are responsible for EPO receptor-operated Ca21 influx and involve cytosolic Src tyrosine kinases, the PLC-g1 cascade, and perhaps Ras oncogenes. Pathological and physiological significance of the erythropoietin-induced Ca21 channel The administration of recombinant human EPO has become an important clinical therapeutic adjunct to the treatment of patients with chronic anemias (such as renal failure, myelodysplasia, chemotherapy). However, the bone marrow is not the only EPO target tissue and as a result, EPO administration has been associated with complications that are independent of its hematologic effects, most notably an increased incidence of hypertension [1–5]. Our present study suggests that a potential mechanism for clinically observed EPO-induced vasoconstriction involves extracellular Ca21 influx via EPO receptor-operated, 1 pS Ca21
channels. However, other mechanisms may be involved in EPOinduced hypertension. Erythropoietin is a well-characterized cytokine growth factor responsible for the proliferation of erythroid progenitor cells [7, 10, 16 –21, 39, 40]. Therefore, it is perhaps not surprising that in addition to being a direct vasoconstrictor, EPO has been shown to act as a vascular growth factor mediating the proliferation of endothelial cells and aortic rings [6, 7]. In the present study, we have shown that EPO stimulates Src tyrosine kinase, PLC-g1 and 1 pS Ca21 channel activity in glomerular mesangial cells, which are phenotypically similar to vascular smooth muscle cells. Predictably, we have found that EPO also stimulates the tyrosine phosphorylation of PLC-g1 and PLC-g1/EPO receptor complex formation in rat aortic vascular smooth muscle cells (unpublished data). The importance of tyrosine kinases, including the cytosolic Src family, and PLC-g1 in controlling cellular growth and differentiation is now well established [35]. DNA synthesis and PLC activity is blocked by tyrosine kinase inhibition (genistein or herbimycin A) and stimulated by tyrosine phosphatase inhibition (vanadate) in human glomerular mesangial cells [26, 27]. In fibroblasts, microinjection of PLC-g1 or PLC-g1 antibodies has been shown to induce or inhibit DNA synthesis, respectively [26, 27]. Higher levels of PLC-g are found in cancer cells and may play a role in abnormal cellular proliferation and mitogenesis [26, 27]. The initiation of cellular DNA synthesis and proliferation requires the influx of extracellular Ca21 [26, 27]. Conversely, inhibition of Ca21 entry blocks mitogenic responses [26, 27]. Finally, EPO activates multiple mitogenic signaling pathways in addition to PLC (such as Ras/mitogen-activated protein kinase, JAK/STAT), which promote the activation of early growth response genes in hematopoietic cells [7, 10, 16 –21, 39, 40].
Marrero et al: Ca21 channel activation by PLC-g1
1267
Fig. 7. Model for 1 pS Ca21 channel activation by erythropoietin (EPO) in glomerular mesangial cells. (Left) Binding of EPO to its receptor stimulates tyrosine phosphorylation of phospholipase C (PLC)-g1, translocation of PLC-g1 from the cytosol to the plasma membrane, and membrane complex formation between PLC-g1 an the EPO receptor itself. Subsequent activation of PLC-g1, promotes PIP2 hydrolysis and the generation of IP3. (Right) Ca21 is released from IP3-dependent intracellular stores and triggers 1 pS Ca21 channel activity. Ca21 entry through the channel allows for sustained activation of this Ca21-sensitive channel. Increases in intracellular Ca21 levels ultimately leads to mesangial cell contraction and proliferation.
Activation of all these same mitogenic factors by classic growth factors and angiotensin II has been strongly implicated in the clinical pathogenesis of maladaptive cellular growth in pathologic disease processes (such as hypertension, cardiac failure, renal failure) [27–29, 35]. Our study raises the concern that exogenous EPO administration for the treatment of anemia may trigger maladaptive growth in non-hematopoietic cells (such as mesangial and vascular smooth muscle cells), potentially leading to the acceleration of glomerulosclerosis, arteriosclerosis, and atherosclerosis. In rats, anemia retards the progression of proteinuria and glomerulosclerosis after five-sixths nephrectomy [43]. However, glomerular injury and renal failure was accelerated when these animals were treated with EPO. Conclusions Other investigators have shown that the EPO-induced increase in intracellular Ca21 in vascular smooth muscle and hematopoietic cells is due to extracellular Ca21 influx via a voltageindependent Ca21 conductance. Our studies provide a candidate pathway involving: (1) EPO binding to EPO receptors, which leads to tyrosine phosphorylation of the -g1 isoenzyme of PLC and membrane translocation of PLC-g1, where it forms a complex with the EPO receptor itself. (2) PLC-g1-mediated hydrolysis of PIP2 increases intracellular IP3. (3) Stimulation of Ca21-activated 1 pS Ca21 channels is initially triggered by intracellular Ca21 release from IP3-dependent stores and is sustained by extracellular Ca21 entry via the channel itself (Fig. 7). Mene and associates [30] have shown that the release of intracellular Ca21 stores on extracellular Ca21 entry, promote voltage-independent Ca21 entry in human glomerular mesangial cells. They proposed that “the InsP3 sensitive Ca21 pool may serve as a trigger for a sustained response, thus initiating a cascade amplification signaling mechanism.” Our results provide a potential mechanism for the common
hypertensive complications associated with the therapeutic administration of EPO. ACKNOWLEDGMENTS Support was provided by National Institutes of Health Grants K08DK02111 and P01-DK50268, a Veterans Affairs Merit Review Award, an American Diabetes Association Research Award, National and Georgia Affiliate American Heart Association (AHA) Grant-In-Aid Awards, and an AHA Minority Scientist Developmental Award (M.B. Marrero). Dr. Shun M. Ling provided insightful scientific and clinical input during this project. Preliminary work was presented at the American Society of Nephrology Annual Meeting in New Orleans, LA (November 3– 6, 1996). Ms. Amy Ling and Ms. Elaine O’Dell provided editorial assistance. Reprint requests to Mario B. Marrero, Ph.D., Emory University School of Medicine, The Center for Cell and Molecular Signaling, Physiology Building, 1638 Pierce Drive, N.E., Atlanta, Georgia 30322, USA. E-mail: [email protected]
REFERENCES 1. NEUSSER M, TEPEL M, ZIDEK W: Erythropoietin increases cytosolic free calcium concentration in vascular smooth muscle cells. Cardiovasc Res 27:1233–1236, 1993 2. VAZIRI ND, ZHOU XJ, SMITH J, OVEISI F, BALDWIN K, PURDY RE: In vivo and in vitro pressor effects of erythropoietin in rats. Am J Physiol 269(Renal Fluid Electrolyte Physiol):F838 –F845, 1995 3. KOKOT F, WIECEK A: Arterial hypertension in uraemic patients treated with erythropoietin. Nephron 71:127–132, 1995 4. LEBEL M, KINGMA I, GROSE JH, LANGLOIS S: Effect of recombinant human erythropoietin therapy on ambulatory blood pressure in normotensive and in untreated borderline hypertensive hemodialysis patients. Am J Hypertens 8:545–551, 1995 5. HEIDENREICH S, RAHN K-H, ZIDEK W: Direct vasopressor effect of recombinant human erythropoietin on renal resistance vessels. Kidney Int 39:259 –265, 1991 6. CARLINI RG, REYES AA, ROTHSTEIN M: Recombinant human erythropoietin stimulates angiogenesis in vitro. Kidney Int 47:740 –745, 1995
1268
Marrero et al: Ca21 channel activation by PLC-g1
7. HALLER H, CHRISTEL C, DANNENBERG L, THIELE P, LINDSCHAU C, LUFT FC: Signal transduction of erythropoietin in endothelial cells. Kidney Int 50:481– 488, 1996 8. TEPEL M, WISCHNIOWSKI H, ZIDEK W: Erythropoietin increases cytosolic free calcium concentration and thrombin induced changes in cytosolic free calcium in platelets from spontaneously hypertensive rats. Biochem Biophys Res Comm 776:313–316, 1984 9. TEPEL M, WISCHNIOWSKI H, ZIDEK W: Erythropoietin induced transmembrane calcium influx in essential hypertension. Life Sci 51:161– 167, 1992 10. REN H-Y, KOMATSU N, SHIMIZU R, OKADA K, MIURA Y: Erythropoietin induces tyrosine phosphorylation and activation of phospholipase C-g1 in a human erythropoietin-dependent cell line. J Biol Chem 269:19633–19638, 1994 11. MILLER BA, BELL LL, LYNCH CJ, CHEUNG JY: Erythropoietin modulation of intracellular calcium: A role for tyrosine phosphorylation. Cell Calcium 16:481– 490, 1994 12. MILLER BA, SCADUTO RC JR, TILLOTSON DL, BOTTI JJ, CHEUNG JY: Erythropoietin stimulates a rise in intracellular free calcium concentration in single early human erythroid precursors. J Clin Invest 82:309 –315, 1988 13. MILLER BA, BELL L, HANSEN CA, ROBISHAW JD, LINDER ME, CHEUNG JY: G-protein a subunit Gia2 mediates erythropoietin signal transduction in human erythroid precursors. J Clin Invest 98:1728 – 1736, 1996 14. CHEUNG JY, ELENSKY M, BRAUNEIS U, SCADUTO RC JR, BELL LL, TILLOTSON DL, MILLER BA: Ion channels in human erythroblasts: Modulation by erythropoietin. J Clin Invest 90:1850 –1856, 1992 15. MILLER BA, CHEUNG JY: Mechanisms of erythropoietin signal transduction: Involvement of calcium channels. Proc Soc Exper Biol Med 206:263–267, 1994 16. FINBLOOM DS, PETRICOIN EFI, HACKETT RH, DAVID M, FELDMAN GM, IGARASHI K-I, FIBACH E, WEBER MJ, THORNER MO, SILVA CM, LARNER AC: Growth hormone and erythropoietin differentially activate DNA-binding proteins by tyrosine phosphorylation. Mol Cell Biol 14:2113–2118, 1994 17. DAMEN JE, CUTLER RL, JIAO HY, YI TL, KRYSTAL G: Phosphorylation of tyrosine 503 in the erythropoietin receptor (EpR) is essential for binding the P85 subunit of phosphatidylinositol (PI) 3-Kinase and for EpR-associated PI 3- kinase activity. J Biol Chem 270:23402– 23408, 1995 18. SAWYER ST, PENTA K: Association of JAK2 and STAT5 with erythropoietin receptors-Role of receptor phosphorylation in erythropoietin signal transduction. J Biol Chem 271:32430 –32437, 1996 19. DAMEN JE, LIU L, CUTLER RL, KRYSTAL G: Erythropoiein stimulates the tyrosine phosphorylation of Shc and its association with Grb2 and a 145-Kd tyrosine phosphorylated protein. Blood 82:2296 –2303, 1993 20. MAYEUX P, DUSANTER-FOURT I, MULLER O, MAUDUIT P, SABBAH M, DRUKER B, VAINCHENKER W, FISCHER S, LACOMBE C, GISSELBRECHT S: Erythropoietin induces the association of phosphatidylinositol 39-kinase with a tyrosine-phosphorylated protein complex containing the erythropoietin receptor. Eur J Biochem 216:821– 828, 1993 21. MIURA O, NAKAMURA N, IHLE JN, AOKI N: Erythropoietin-dependent association of phosphatidylinositol 3-kinase with tyrosine-phosphorylated erythropoietin receptor. J Biol Chem 269:614 – 620, 1994 22. PAXTON WG, MARRERO MB, KLEIN JD, DELAFONTAINE P, BERK BC, BERNSTEIN KE: The angiotensin II AT1 receptor is tyrosine and serine phosphorylated and can serve as a substrate for the src family of tyrosine kinases. Biochem Biophys Res Comm 200:260 –267, 1994 23. MARRERO MB, SCHIEFFER B, PAXTON WG, SCHIEFFER E, BERNSTEIN KE: Electroporation of pp60c-src antibodies inhibits angiotensin II activation of phospholipase C-g1 in rat aortic smooth muscle cells. J Biol Chem 270:15734 –15738, 1995 24. MARRERO MB, SCHIEFFER B, MA H, BERNSTEIN KE, LING BN: Ang II-induced tyrosine phosphorylation stimulates phospholipase C-g1
25.
26. 27. 28.
29. 30. 31. 32. 33. 34. 35. 36.
37.
38.
39. 40.
41. 42. 43.
and Cl2 channels in glomerular mesangial cells. Am J Physiol 270: C1834 –C1842, 1996 MARRERO MB, PAXTON WG, DUFF JL, BERK BC, BERNSTEIN KE: Angiotensin II stimulates tyrosine phosphorylation of phospholipase C-g1 in vascular smooth muscle cells. J Biol Chem 269:10935–10939, 1994 MOGAMI H, KOJIMA I: Stimulation of calcium entry is prerequisite for DNA synthesis induced by platelet-derived growth factor in vascular smooth muscle cells. Biochem Biophys Res Comm 196:650 – 658, 1993 LING BN, MATSUNAGA H, MA H, EATON DC: Role of growth factors in mesangial cell ion channel regulation. Kidney Int 48:1158 –1166, 1995 MA H, MATSUNAGA H, LI B, SCHIEFFER B, MARRERO MB, LING BN: Ca21 channel activation by PDGF-induced tyrosine phosphorylation and Ras GTP-binding proteins in rat glomerular mesangial cells. J Clin Invest 97:2332–2341, 1996 MA H, MATSUNAGA H, LI B, MARRERO MB, LING BN: Regulation of PDGF b receptor-operated Ca21 channels by phospholipase C-g1 in glomerular mesangial cells. Am J Physiol 271:F994 –F1003, 1996 MENE P, TETI A, PUGLIESE F, CINOTTI GA: Calcium release-activated calcium influx in cultured human mesangial cells. Kidney Int 46:122– 128, 1994 MATSUNAGA H, LING BN, EATON DC: Ca21-permeable channel associated with platelet-derived growth factor receptor in mesangial cells. Am J Physiol 267:C456 –C465, 1994 LING BN, SEAL EE, EATON DC: Regulation of mesangial cell ion channels by insulin and angiotensin II: Possible role in diabetic glomerular hyperfiltration. J Clin Invest 92:2141–2151, 1993 SEAL EE, EATON DC, GOMEZ LM, MA H, LING BN: Extracellular glucose reduces mesangial cell ion channel responsiveness to angiotensin II. Am J Physiol 269:F389 –F397, 1995 WAHL M, LUCHERINI MJ, GRUENSTEIN E: Intracellular Ca21 measurement with Indo-1 in substrate-attached cells: Advantages and special considerations. Cell Calcium 11:487–500, 1990 SCHIEFFER B, DREXLER H, LING BN, MARRERO MB: G proteincoupled receptors control vascular smooth muscle cell proliferation via pp60c-src and p21ras. Am J Physiol 272:C2019 –C2030, 1997 KUMJIAN DA, WAHL MI, RHEE SG, DANIAL TO: Platelet-derived growth factor (PDGF) binding promotes physical association of PDGF receptor with phospholipase C. Proc Natl Acad Sci USA 86:8232– 8236, 1989 MORRISON DK, KAPLAN DR, RHEE SG, WILLIAMS LT: Plateletderived growth factor (PDGF)-dependent association of phospholipase C-gamma with the PDGF receptor signaling complex. Mol Cell Biol 10:2359 –2366, 1990 HANKE JH, GARDNER JP, DOW RL, CHANGELIAN PS, BRISSETTE WH, WERINGER EJ, POLLOK BA, CONNELLY PA: Discovery of a novel, potent, and Src family-selective tyrosine kinase inhibitor. J Biol Chem 271:695–701, 1996 PENTA K, SAWYER ST: Erythropoietin induces the tyrosine phosphorylation, nuclear translocation, and DNA binding of STAT1 and STAT5 in erythroid cells. J Biol Chem 270:31282–31287, 1995 TORTI M, MARTI KB, ALTSCHULER D, YAMAMOTO K, LAPETINA EG: Erythropoietin induces p21ras activation and p120GAP tyrosine phosphorylation in human erythroleukemia cells. J Biol Chem 267:8293, 1992 ALLBRITTON NL, MEYER T, STRYER L: Range of messenger action of calcium ion and inositol 1,4,5-trisphosphate. Science 258:1812–1815, 1992 KARGACIN GJ: Calcium signaling in restricted diffusion spaces. Biophys J 67:262–272, 1994 GARCIA DL, ANDERSON S, RENNKE HG, BRENNER BM: Anemia lessens and its prevention with recombinant human erythropoietin worsens glomerular injury and hypertension in rats with reduced renal mass. Proc Natl Acad Sci USA 85:6142– 6146, 1988