Hypoxic proliferation requires EGFR-mediated ERK activation in human pulmonary microvascular endothelial cells

Abstract

We have previously shown that hypoxic proliferation of human pulmonary microvascular endothelial cells (hPMVECs) depends on epidermal growth factor receptor (EGFR) activation. To determine downstream signaling leading to proliferation, we tested the hypothesis that hypoxia-induced proliferation in hPMVECs would require EGFR-mediated activation of extracellular signal-regulated kinase (ERK) leading to arginase II induction. To test this hypothesis, hPMVECs were incubated in either normoxia (21% O₂, 5% CO₂) or hypoxia (1% O₂, 5% CO₂) and Western blotting was performed for EGFR, arginase II, phosphorylated-ERK (pERK), and total ERK (ERK). Hypoxia led to greater EGFR, pERK, and arginase II protein levels than did normoxia in hPMVECs. To examine the role of EGFR in these hypoxia-induced changes, hPMVECs were transfected with siRNA against EGFR or a scrambled siRNA and placed in hypoxia. Inhibition of EGFR using siRNA attenuated hypoxia-induced pERK and arginase II expression as well as the hypoxia-induced increase in viable cell numbers. hPMVECs were then treated with vehicle, an EGFR inhibitor (AG1478), or an ERK pathway inhibitor (U0126) and placed in hypoxia. Pharmacologic inhibition of EGFR significantly attenuated the hypoxia-induced increase in pERK level. Both AG1478 and U0126 also significantly attenuated the hypoxia-induced increase in viable hPMVECs numbers. hPMVECs were transfected with an adenoviral vector containing arginase II (AdArg2) and overexpression of arginase II rescued the U0126-mediated decrease in viable cell numbers in hypoxic hPMVECs. Our findings suggest that hypoxic activation of EGFR results in phosphorylation of ERK, which is required for hypoxic induction of arginase II and cellular proliferation.

Keywords: arginase; hypoxia; pulmonary hypertension; pulmonary vascular remodeling.

Fig. 1.

Human pulmonary microvascular endothelial cells (hPMVECs) exposed to hypoxia had greater levels of epidermal growth factor receptor (EGFR), phosphorylated extracellular signal-regulated kinase (pERK), and arginase II protein than cells exposed to normoxia. hPMVECs were incubated in either normoxia (21% O₂, 5% CO₂) or hypoxia (1% O₂, 5% CO₂) for 24 h and protein levels of EGFR (A), arginase II (B), or phosphorylated and total ERK (C) were measured. β-Actin was used as a control for protein loading. Representative Western blots are shown from 3 independent experiments. Hypoxia led to easily detectable bands for EGFR and arginase II. Hypoxia increased pERK protein levels without detectably changing total ERK protein levels. D: densitometry data for pERK protein levels normalized to total ERK (n = 3 in each group). Hypoxia led to ~4-fold induction of pERK protein levels in hPVMEC. *P < 0.001, hypoxia different from normoxia.

Fig. 2.

In hypoxia, siRNA-mediated knockdown of EGFR led to lower levels of pERK protein. A: hPMVECs were transfected with either scrambled siRNA or EGFR siRNA and incubated in hypoxia for 24 h and protein levels for EGFR were assessed. Representative Western blots are shown. The bar graph is the densitometry data for EGFR normalized to β-actin (n = 3 in each group) and demonstrates clear knockdown of EGFR protein levels. *P < 0.05, EGFR siRNA different from scrambled. B: hPMVECs were transfected with either scrambled siRNA or EGFR siRNA and incubated in hypoxia for 1 or 24 h and protein levels for pERK were assessed. Representative Western blot for pERK and quantification by densitometry of pERK normalized to total ERK are shown (n = 6–8 for each group). *P < 0.05, EGFR siRNA different from scrambled siRNA.

Fig. 3.

In hypoxia, siRNA-mediated knockdown of EGFR led to fewer viable cells than in scrambled siRNA-treated cells. hPMVECs were transfected with EGFR siRNA or scrambled siRNA and then equal numbers of hPMVECs were plated on 6-well plates and placed in hypoxia for 48 h (n = 5–8 in each group). Viable cells were counted using trypan blue exclusion. *P < 0.005, EGFR siRNA different from scrambled siRNA.

Fig. 4.

Hypoxic hPMVECs treated with either AG1478 or U0126 had lower levels of pERK and arginase II protein than did hypoxic cells treated with vehicle (DMSO). hPMVECs were treated with either DMSO (vehicle), AG1478 (an EGFR inhibitor; 1 µM), or U0126 (an ERK pathway inhibitor; 10 µM) and incubated in hypoxia for 1 h. Protein was harvested and assessed for pERK and total ERK, or for arginase II and β-actin. A: representative Western blot for pERK and quantification of ERK phosphorylation normalized to ERK (n = 4–6 in each group). B: representative Western blot for arginase II (Arg II) and quantification of arginase II expression normalized to β-actin (n = 6 in each group). *P < 0.05, different from DMSO; #P < 0.01, different from AG1478.

Fig. 5.

Pharmacologic inhibition of EGFR attenuated hypoxia-induced viable cell numbers. Equal numbers of hPMVECs were plated on 6-well plates; treated with either DMSO (vehicle), AG1478 (1 µM), or U0126 (10 µM); and placed in hypoxia for 48 h. Viable cells were counted using trypan blue exclusion (n = 6–8 in each group). *P < 0.005, different from DMSO; #P < 0.005, different from AG1478.

Fig. 6.

Only the ERK pathway inhibitor decreased viable cell numbers in hypoxia. Equal numbers of hPMVECs were plated on 6-well plates and treated with either DMSO (vehicle), an ERK pathway inhibitor (U0126; 10 µM), a JNK pathway inhibitor (SP600125; 20 µM), or a p38 pathway inhibitor (SB203580; 10 µM) and placed in hypoxia for 48 h (n = 6–9 in each group). Viable cells were counted using trypan blue exclusion. *P < 0.001, different from all other conditions.

Fig. 7.

Overexpression of arginase II rescued the U0126-mediated decrease in viable cell numbers in hypoxic hPMVECs. Cells were transfected with either an adenoviral vector containing arginase II (AdArg2) or green fluorescent protein (AdGFP). A: Western blot of Arg II protein demonstrating greater Arg II expression in hPMVECs treated with AdArg2 (n = 6) than in those treated with AdGFP (n = 6) in both vehicle-treated or U0126-treated cells. B: equal number of cells were plated onto 6-well plates and treated with either U0126 or DMSO (vehicle) and placed in hypoxia for 48 h (n = 6 in each group). Viable cell numbers were counted by trypan blue exclusion. *P < 0.001, U0126 different from DMSO same transfection condition; #P < 0.001, AdArg2 different from AdGFP same treatment condition.

Fig. 8.

Conditioned media from hypoxic hPMVECs stimulated proliferation in human pulmonary artery smooth muscle cells (hPASMCs). A: hPMVECs were incubated for 24 h in either normoxia (NCM, n = 6) or hypoxia (HCM, n = 6) and the conditioned media were mixed 1:1 with SmGM and placed on equal numbers of hPASMCs (1 × 10⁴ cells per well) in each well of 6-well plates. The hPASMCs were then incubated in normoxia for 120 h and viable cell number determined. As a control conditioned media placed in a cell culture plate with no hPMVECs and incubated for 120 h was used as a control (HM, n = 6). *P < 0.001, HCM different from NCM; #P < 0.001, HCM different from HM. B: to determine the role of ERK in the HCM effects on hPASMCs viable cell numbers after 120 h, the above experiment was repeated except that an additional group of hPMVECs were incubated for 24 h in hypoxia with 10 µM U0126 added (HCM + U0126) to obtain conditioned media; n = 6 for all groups. *P < 0.001, HCM different from NCM; #P < 0.001, HCM + U0126 different from HCM.