Angiotensin I-converting enzyme (ACE) inhibitors are widely used to treat patients with cardiovascular and kidney diseases, but inhibition of ACE alone does not fully explain the beneficial effects. We reported that ACE inhibitors directly activate bradykinin B1 receptor at the canonical Zn2+ binding site, leading to prolonged nitric oxide (NO) production in endothelial cells. Protein kinase C (PKC) ϵ, a novel PKC isoform, is up-regulated in myocardium after infarction, suggesting a role in the development of cardiac dysfunction. In cytokine-treated human lung microvascular endothelial cells, B1 receptor activation by ACE inhibitors (enalaprilat, quinaprilat) or peptide ligands (des-Arg10-Lys1-bradykinin, des-Arg9-bradykinin) inhibited PKCϵ with an IC50 = 7 × 10–9 M. Despite the reported differences in binding affinity to receptor, the two peptide ligands were equally active, even when inhibitor blocked the cleavage of Lys1, thus the conversion by aminopeptidase. The synthetic undecapeptide (LLPHEAWHFAR) representing the binding site for ACE inhibitors on human B1 receptors reduced PKCϵ inhibition by enalaprilat but not by peptide agonist. A combination of inducible and endothelial NO synthase inhibitors, 1400W [N-(3(aminomethyl) benzyl) acetamidine dihydrochloride] and Nω-nitro-l-arginine (2 μM), significantly reduced inhibition by enalaprilat (100 nM), whereas the NO donor (Z)-1-[N-(2-aminoethyl)-N-(2-ammonioethyl) amino]diazen-1-ium-1,2-diolate (100 μM) inhibited PKCϵ activity just as the B1 ligands did. In conclusion, NO generated by B1 receptor activation inhibits PKCϵ.
The bradykinin B1 and B2 receptors are G-protein-coupled receptors that mediate the actions of kinins. The B2 receptors are activated by bradykinin and kallidin (Lys-bradykinin) and are constitutively expressed in normal tissues (Bhoola et al., 1992), whereas the B1 receptors are highly induced during inflammatory conditions (McLean et al., 2000). Plasma carboxypeptidase N and tissue carboxypeptidase M cleave the C-terminal Arg of bradykinin and kallidin and generate the endogenous agonists of the B1 receptors, des-Arg9-bradykinin (DBK) and des-Arg10-kallidin (DAKD) (Erdös and Skidgel, 1997).
Angiotensin I-converting enzyme (ACE) inhibitors are beneficial in treating patients with hypertension, myocardial infarction, congestive heart failure, and diabetic nephropathy (Gavras et al., 1974; Yusuf et al., 2000; Gavras and Gavras, 2001). However, inhibition of ACE alone, thus blocking the release of angiotensin II or inactivation of bradykinin, does not fully explain the effects of ACE inhibitors (Erdös and Skidgel, 1997; Corvol et al., 2004). ACE inhibitors potentiate the effect of bradykinin on the B2 receptor and resensitize the receptor previously desensitized by the ligand, apart from blocking peptide hydrolysis (Minshall et al., 1997; Erdös et al., 1999; Marcic et al., 1999). ACE inhibitors also directly activate the B1 receptors, independent of endogenous kinins and ACE (Ignjatovic et al., 2002). In endothelial cells, this activation leads to prolonged nitric oxide (NO) release and increases the uptake of its precursor, l-arginine (Ignjatovic et al., 2002, 2004). ACE inhibitors activate the B1 receptor at a Zn2+ binding pentapeptide sequence (Ignjatovic et al., 2002), similar to the active sites in ACE (Corvol et al., 2004).
ACE inhibitors improve endothelial dysfunction even in normotensive patients with coronary artery disease (Mancini et al., 1996). PKCϵ is a novel PKC isoform that is up-regulated in the myocardium after infarction. In a rat model of myocardial infarction, treatment with ACE inhibitors reduces the expression of PKCϵ after infarction (Simonis et al., 2003; Wang et al., 2003), indicating that PKCϵ may play a role in the development of cardiac dysfunction and that ACE inhibitors can influence PKCϵ expression. We wanted to determine whether direct activation of the B1 receptors by ACE inhibitors affects PKCϵ activity and whether two endogenous B1 receptor agonists, DAKD and DBK, are equally active despite big differences in their affinity for the B1 receptors (Hess et al., 1996). We have found that B1 receptor activation by ACE inhibitors (enalaprilat, quinaprilat) or peptide agonists blocks PKCϵ activity in human endothelial cells and that the PKCϵ inhibition is mediated by NO release. The two endogenous ligands, DAKD and DBK, are almost equally active, even in the presence of an aminopeptidase inhibitor that blocks the conversion of DAKD to DBK by removal of its N-terminal Lys1 (Erdös et al., 1963; Webster and Pierce, 1963).
Materials and Methods
Materials. Enalaprilat, the active form of the ACE inhibitor enalapril, was from Toronto Research Chemicals, Inc. (North York, ON, Canada). Quinaprilat was from Parke-Davis Pharmaceutical Research Division (Ann Arbor, MI). Fetal bovine serum was from Atlanta Biologicals (Norcross, GA). Interleukin (IL)-1β and interferon (IFN)-γ were purchased from Calbiochem (San Diego, CA) and Invitrogen (Carlsbad, CA). The NO donor DETA-NONOate (t1/2 = 20 h at 37°C) was from Cayman Chemical (Ann Arbor, MI). PKCϵ assay kit, PKCαβγ assay kit, and PKCϵ polyclonal antibody directed against residues 726 to 737 (KGFSYFGEDLMP), along with Jurkat cell lysate (a positive control) were obtained from Upstate Biotechnology (Lake Placid, NY). All other reagents were obtained from Sigma-Aldrich (St. Louis, MO).
Cell Culture. Human lung microvascular endothelial (HLMVE) cells (Cell Applications, San Diego, CA; Cambrex Bio Science Baltimore, Inc., Baltimore, MD) were cultured in endothelial basal medium-2 (Cambrex Bio Science, Walkersville, MD) containing 10 ng/ml human epidermal growth factor, 5 ng/ml vascular endothelial growth factor, 2 ng/ml human fibroblast growth factor, 2 ng/ml insulin-like growth factor, 0.2 μg/ml ascorbic acid, 50 ng/ml gentamicin-amphotericin B, and 10% fetal bovine serum. HLMVE cells were used in passages 6 to 8 and routinely treated with 5 ng/ml IL-1β and 50 ng/ml IFN-γ for 18 h before the experiments to induce B1 receptors (11, 12). The experiments were performed using phenol-red free Dulbecco's modified Eagle's medium/F12 medium (Invitrogen) containing no fetal bovine serum.
PKC Assay. HLMVE cells were treated for 20 min with medium alone (control) or medium containing B1 receptor agonists. Cells were then scraped into ice-cold PBS, washed twice with PBS, and then lysed by sonication (two times for 10 s on ice). Cell lysates were assayed for PKCϵ activity following the manufacturers' instructions for the phosphorylation of the specific substrate peptide (ERMRPRKRQGSVRRRV) in the presence of [γ-32P] ATP. The phosphorylated substrate is separated from residual [γ-32P] ATP on phosphocellulose paper and quantified in a scintillation counter. Other serine/threonine kinases such as protein kinase A and calmodulin-dependent kinases were blocked during assay by an inhibitor cocktail. The kit assay buffer contains 1 mM Ca2+, but in selected control experiments, Ca2+-free buffer containing 1 mM EDTA was used to eliminate any contribution of conventional PKC isoforms. The same procedure was used to measure PKCα activity using buffer containing 1 mM Ca2+, except the peptide substrate for conventional PKC isoforms (QKRPSQRSKYL) was used.
DAKD Conversion to DBK. The hydrolysis of DAKD by HLMVE cells in culture was assayed by high-performance liquid chromatography (HPLC). HLMVE cells were incubated with DAKD (50 μM) for 20 min at 37°C (pH = 7.2). Two groups of the cells were pretreated for 10 min with bestatin (150 μM), an aminopeptidase inhibitor, and then exposed to DAKD (50 μM) for 20 min. The reactions were stopped with 5% trifluoroacetic acid, centrifuged at 14,000g for 10 min, and peptide products were separated and quantitated by HPLC.
NO Release. NO was measured electrochemically with porphyrinic microsensor (Ignjatovic et al., 2002, 2004). Cells were plated in 12-well plates and used upon reaching confluence in 2 days. Cells were preincubated for a few minutes at 37°C until a stable baseline was established. Ligands were added, and the responses (current versus time) were recorded continuously. The current generated on the porphyrinic electrode was proportional to the NO released and was quantitated with a standard NO solution.
Western Blot. Western blot of PKCϵ expression was performed with HLMVE cell lysates and as a positive control, with Jurkat cell lysate employing polyclonal anti-PKCϵ antibody. Cell lysate proteins were separated by 10% SDS-polyacrylamide gel electrophoresis and transferred to polyvinylidene diflouride membranes, incubated overnight with polyclonal PKCϵ primary antibody (1:1000 v/v) at 4°C, and then with secondary anti-rabbit goat serum (1:300,000).
PKCϵ Immunoprecipitation. PKCϵ was immunoprecipitated by primary PKCϵ antibody produced in rabbits. In brief, cytokine-treated HLMVE cells were washed twice with ice-cold PBS and then scraped into ice-cold homogenization buffer (pH 7.5, 20 mM Tris, 2.5 mM EDTA, 2.5 mM EGTA, 1 mM dithiothreitol, 1 mM phenylmethylsulphonyl fluoride, 1% protease cocktail inhibitor, and 0.1% Triton X-100). Cells were subsequently sonicated twice for 15 s and then centrifuged for 15 min at 14,000g. Supernatants were collected and incubated overnight or for 2 h with polyclonal anti-PKCϵ antibody or nonimmune rabbit IgG at 4°C. After mixing with protein A-Sepharose for 2 h at 4°C and centrifugation, PKCϵ activity was measured in Ca2+-free buffer in the resuspended pellet and supernatant.
Statistics. Data in the figures are expressed as mean ± S.E. of n observations. Statistical evaluation was performed by one-way analysis of variance. Values of p < 0.05 were considered statistically significant.
PKCϵ Expression in HLMVE Cells. To investigate the effect of the B1 receptor activation on PKCϵ activity, we stimulated HLMVE cells with proinflammatory cytokines, IL-1β and IFN-γ, for 18 h at 37°C to induce B1 receptors. Western blotting demonstrated that PKCϵ was indeed expressed in HLMVE cells (Fig. 1).
PKCϵ Immunoprecipitation from HLMVE Cells. To determine PKCϵ activity, we measured phosphorylation of a PKCϵ peptide substrate. To confirm the specificity of the assay in HLMVE cells, PKCϵ was immunoprecipitated from HLMVE cell supernatants with polyclonal PKCϵ antibody and as a control, with nonimmune rabbit IgG. As a result, 69.3 ± 1.5% of the total PKC activity was immunoprecipitated by primary PKCϵ antibody versus 8.2 ± 0.8% by nonimmune rabbit IgG (n = 4).
Inhibition of PKCϵ. The B1 receptor agonists, DAKD (10 nM) and enalaprilat (10 nM), inhibited 65 and 70% of the PKCϵ activity measured in HLMVE cells (Fig. 2A). The B1 receptor antagonist, des-Arg10-Leu9-kallidin (100 nM), inhibited the response to both DAKD and enalaprilat, confirming this conclusion. Enalaprilat and the two peptide agonists of the B1 receptor had about the same low IC50 value (7 × 10–9 M) (Fig. 2B).
Effect of the Undecapeptide. The human B1 receptor contains in its second extracellular loop (residues 195–199) the HEAWH sequence, a canonical Zn2+ binding pentamer, which represents the suggested site of the activation by ACE inhibitors (Ignjatovic et al., 2002). A synthetic undecapeptide (LLPHEAWHFAR) (residues 192–202 of the human B1 receptor), which includes this pentamer, blocked B1 receptor activation by enalaprilat but not by DAKD (Ignjatovic et al., 2002). To confirm that enalaprilat inhibits PKCϵ due to B1 receptor activation, we pretreated HLMVE cells with 100 μM peptide before adding either DAKD or enalaprilat (Fig. 2C). The undecapeptide reduced the effect of enalaprilat on PKCϵ inhibition (26% inhibition versus 63% in the absence of the undecapeptide), whereas DAKD was not significantly affected, distinguishing the actions of the B1 receptor agonists.
The Time Course of PKCϵ Inhibition. We treated the cells with 100 nM DAKD, enalaprilat, or quinaprilat for 1, 10, or 20 min. PKCϵ activity did not change after 1 min; however, after 10 or 20 min, the peptide and both ACE inhibitors similarly inhibited PKCϵ (Fig. 3A). This result indicates that B1 receptor-mediated PKCϵ inhibition is slower than the increase in intracellular [Ca2+]i that is almost immediate in response to the B1 agonists (Ignjatovic et al., 2002).
Effect of NOS Inhibitors and NO Donor. Both DAKD and enalaprilat release NO from human endothelial cells (Ignjatovic et al., 2004). We explored whether the inhibition of PKCϵ could be NO-dependent (Fig. 3B). HLMVE cells were pretreated for 10 min with both inducible nitric oxide synthase and endothelial nitric oxide synthase inhibitors (2 μM 1400W and 2 μM Nω-nitro-l-arginine) and then stimulated for 20 min with either DAKD (100 nM) or enalaprilat (100 nM). The NOS inhibitors significantly reduced the effect of the B1 receptor agonists on PKCϵ inhibition (Fig. 3B). The NO donor DETA-NONOate (100 μM) alone inhibited PKCϵ activity similar to the effects of DAKD and enalaprilat, indicating that the B1 receptor activation reduces PKCϵ activity via an NO-dependent mechanism. To determine whether direct protein modification by NO or peroxynitrite is involved, cells were pretreated for 20 min with dithiothreitol (20 mM), a mercapto compound, or ebselen (10 μM), a peroxynitrate scavenger. In these experiments, 100 nM DAKD or enalaprilat reduced PKCϵ activity to 36 ± 2% or 34% ± 3% of control, respectively, when used alone to 39 ± 3% or 37 ± 2% in the presence of dithiothreitol and to 38 ± 1% or 34 ± 2% in the presence of ebselen (mean values ± S.E. for n = 3). The lack of effect of dithiothreitol or ebselen suggests that tyrosine nitration or cysteine S-nitrosylation did not cause the inhibition of PKCϵ.
PKCα Activity. We further investigated whether the B1 receptor agonists would affect another PKC isoform, PKCα, a conventional PKC (Wang et al., 2003). DAKD (100 nM) or enalaprilat (100 nM) were incubated with HLMVE cells for 20 min, and the phosphorylation of the PKC substrate for conventional PKC isoforms (α, β, and γ) was measured. PKCα activity in the presence of 100 nM DAKD or enalaprilat was 100 ± 2% or 104 ± 3% of control (mean values ± S.E.; n = 3). Because the standard PKCϵ assay buffer contained calcium, we wondered whether the residual activity remaining after B1 receptor inhibition of PKCϵ might be due to PKCα. Indeed, when cells were preincubated with the specific PKCα/β1 inhibitor, Gö6976 (100 nM for 20 min), 100 nM DAKD inhibited the remaining PKC activity by 92% (mean value for n = 2, assayed in triplicate). By itself, 100 nM Gö6976 inhibited 33% of the activity (mean value for n = 2, assayed in triplicate) measured with the PKCϵ substrate. Furthermore, when the PKCϵ assay was done in the absence of Ca2+, 100 nM DAKD inhibited PKCϵ activity by 93% (mean value for n = 2, assayed in triplicate). These data indicate that B1 receptor activation almost completely (93–95%) inhibits PKCϵ and that the lack of complete inhibition in the studies above was due to the residual activity of a conventional PKC (most likely α) that can phosphorylate the PKCϵ substrate approximately 25%.
NO Release. B1 agonists inhibited PKCϵ, which was mediated by NO, with a similar IC50. As a follow-up, we measured NO release from cytokine-treated HLMVE cells to determine whether the three B1 agonists would be equally effective. Indeed, this is so (Fig. 4A); DAKD and DBK (100 nM) released NO similarly, despite the large differences in binding affinity to B1 receptors (16). The ACE inhibitor enalaprilat (100 nM) also stimulated the release of NO, at the levels produced by the peptides.
DAKD Hydrolysis. Kallidin (Lys-bradykinin) is converted to bradykinin when an aminopeptidase of plasma (Erdös et al., 1963; Webster and Pierce, 1963) or cell membrane (Bhoola et al., 1992; Erdös and Skidgel, 1997) cleaves its Lys1-Arg2 bond, which applies to the conversion of DAKD to DBK. This could potentially explain the equal potency of DAKD and DBK in our experiments (Fig. 2B). To investigate that, cells were pretreated with aminopeptidase inhibitor bestatin (10 μM for 10 min), but DAKD and DBK still similarly inhibited PKCϵ (Fig. 4B). To confirm whether DAKD can be converted to DBK by endothelial aminopeptidases, HLMVE cells were incubated with 50 μM DAKD for 20 min, and then the reaction products were analyzed by HPLC using standards to identify and quantify product peaks (Fig. 5). Cytokine-treated HLMVE cells, within 20 min, generated 11.6 ± 0.95 nmol DBK/2 × 105 cells at 37°C. In the presence of bestatin (150 μM), an aminopeptidase inhibitor, only 3.4 ± 0.72 nmol DBK/2 × 105 cells was generated for the same incubation time (n = 5). Similar results were obtained in control HLMVE cells, indicating that endothelial aminopeptidase was not induced by IL-1β and IFN-γ. Thus, the two B1 agonist peptides are equally active in stimulating NO release and inhibiting PKCϵ in HLMVE cells.
When peptides or ACE inhibitors activate the B1 receptor, the resulting NO release inhibits PKCϵ in human endothelial cells. The two endogenous ligands, DAKD and DBK, were almost equally active even without the conversion of DAKD to DBK by aminopeptidase. HLMVE cells clearly expressed PKCϵ, as shown in Western blot (Fig. 1), and 69% of PKCϵ activity was immunoprecipitated from lysed cells by primary PKCϵ antibody.
In endothelial cells, NO suppresses shear stress-induced activation of PKCϵ and ERK1/2 (Ni et al., 2003). In our study, NOS inhibitors blocked the inhibition of PKCϵ by DAKD and enalaprilat. The NO donor DETA-NONOate induced a response similar to those of DAKD and enalaprilat, suggesting that B1 receptor activation inhibits PKCϵ by enhancing NO synthesis. Although the inhibition could depend on post-translational modification of the enzyme by nitration of tyrosine or S-nitrosylation of cysteine residues, the mercapto compound, dithiothreitol, and the peroxynitrate scavenger, ebselen, did not affect PKCϵ inhibition, suggesting that tyrosine nitration and cysteine S-nitrosylation are not involved in the PKCϵ inhibition.
Macrophages from PKCϵ knockout mice release far less cytokines (Castrillo et al., 2001), and in rabbit cardiomyocytes, PKCϵ activates NF-kB (Li et al., 2000), leading to production of cytokines and adhesion molecules (Matsumori, 2004). Here, we found that B1 receptor activation decreases PKCϵ activity in human endothelial cells. The B1 receptors are highly expressed under inflammatory conditions, including atheromatous disease, in endothelial cells or smooth muscle cells (Raidoo et al., 1997) and in rat glomeruli in diabetes (Christopher and Jaffa, 2002). Thus, by reducing PKCϵ activity, ACE inhibitors can indirectly decrease NF-κB activity, lowering the production of proinflammatory cytokines and adhesion molecules. Because inflammation induces B1 receptor expression (Bhoola et al., 1992), inhibition of PKCϵ by B1 agonists may represent a negative feedback of NF-κB activation, contributing to beneficial effects of ACE inhibitors in atheromatous disease, with endothelial dysfunction.
Angiotensin II plays an important role in cardiac remodeling and fibrosis and increases the adhesion of cardiac fibroblasts to collagen via activation of PKCϵ, which subsequently phosphorylates and activates β1 integrins (Stawowy et al., 2005). Thus, ACE inhibitors could have a dual beneficial effect in this pathway by reducing generation of angiotensin II by ACE as well as by activating B1 receptors to inhibit PKCϵ activity.
Overexpression of constitutively active PKCϵ in the hearts of transgenic mice causes development of concentric cardiac hypertrophy (Takeishi et al., 2000). In heart failure, overexpression of PKCϵ alters myofilaments (Goldspink et al., 2004). These results indicate that PKCϵ may play a role in the development of cardiac dysfunction and may provide a basis for ACE inhibitor actions.
DAKD has three orders of magnitude higher affinity for the human B1 receptor than DBK in binding studies (Hess et al., 1996). However, in our experiments, both peptides were almost equally active. When HLMVE cells were pretreated with the aminopeptidase inhibitor bestatin to block the conversion of DAKD to DBK, both peptides still similarly inhibited PKCϵ and released NO. In published reports, DBK or DAKD were shown to have both similar and different efficacies as B1 agonists, depending on the species and/or tissue source used. For example, on the rat ileum (Ueno et al., 2002) or in RAW 264.7 macrophages (Burch and Kyle, 1992), the two peptides had the same affinity or activity on B1 receptors, whereas in rabbit vascular smooth muscle cells, the two agonists exhibited a 100-fold difference in potency for stimulation of phosphoinositide hydrolysis (Schneck et al., 1994).
The reason for the apparent discrepancy between the reported affinity and the activity we found in our studies is not clear, but it is quite possible that the two peptides have different efficacies (i.e., effectiveness of the peptide-receptor complex). Early studies on receptors and theories of drug-receptor interactions already revealed the fact that equal biologic responses did not necessarily mean equal degrees of receptor occupancy (for example, see Goldstein et al., 1974; Limbird, 2004). Thus, a peptide such as DBK may have a higher efficacy than DAKD, leading to a similar response at a lower level of receptor occupancy. Alternatively, there may exist in HLMVE cells complexes of B1 receptors with another protein or proteins that could alter receptor binding and equalize the affinities of the two peptides. It should also be noted that binding studies are usually done with cellular plasma membrane preparations, frequently at 4°C to minimize peptide degradation and dissociation after equilibrium binding. This may not accurately reflect the affinity to and efficacy on intact cells in a more physiological environment at 37°C. Consequently, although Lys1 greatly enhances the affinity of DAKD for the human B1 receptors over DBK in binding assays, they activate the receptors in the human endothelial cells used similarly. That they are similarly effective may be of significance in cardiovascular functions and extends the potential importance of the B1 receptor by having one more effective agonist, a stable ligand (DABK), instead of only a metabolically less stable one (DAKD) that is rapidly converted to a purportedly inactive metabolite.
ACE and its inhibitors are in a central position to regulate both the kallikrein-kinin and renin-angiotensin systems (Bhoola et al., 1992; Erdös and Skidgel, 1997; Corvol et al., 2004). A working kallikrein-kinin system is necessary for normal cardiac and arterial function to counterbalance the renin-angiotensin system (Meneton et al., 2001; Schmaier, 2003) and may protect the heart against remodeling after myocardial infarction (Xu et al., 2005). In aging rat hearts, the B2 receptor expression is decreased, whereas B1 receptor expression is increased, indicating a possible compensatory reaction (Kintsurashvili et al., 2005). B1 receptor activation can result in noxious consequences, e.g., pain and inflammation, which warrant the interest in development of therapeutically useful B1 receptor blockers (Marceau and Regoli, 2004). However, ACE inhibitors are beneficial in many pathological conditions where B1 receptors are likely to be induced, such as after myocardial infarction. The release of NO is considered to be one of the favorable effects of ACE inhibitors and besides potentiating bradykinin, it is also achieved by a direct stimulation of the B1 receptors (Ignjatovic et al., 2002, 2004). Clearly, not all proteins induced during inflammation have only deleterious functions as the recent problems with COX2 inhibitors have shown. These considerations suggest caution in applying B1 receptor blockers to the treatment of patients with chronic cardiovascular diseases, such as hypertension and congestive heart failure.
In conclusion, ACE inhibitors and peptide ligands directly activate the bradykinin B1 receptors in human cells and, as a consequence, inhibit PKCϵ via an NO-dependent mechanism. The two endogenous peptide agonists of the B1 receptors are equally active in stimulating this response. This inhibition of PKCϵ, independent of blocking Ang II generation or bradykinin metabolism by ACE, can contribute to the therapeutic usefulness of these compounds.
- Received August 5, 2005.
- Accepted November 9, 2005.
This work was supported by National Institute of Health Grants HL 36473, HL 68580, and HL 60678.
ABBREVIATIONS: DBK, des-Arg9-bradykinin; DAKD, des-Arg10-kallidin; ACE, angiotensin I-converting enzyme; NO, nitric oxide; PKC, protein kinase C; IL, interleukin; IFN, interferon; DETA-NONOate, (Z)-1-[N-(2-aminoethyl)-N-(2-ammonioethyl) amino]diazen-1-ium-1,2-diolate; HLMVE, human lung microvascular endothelial; PBS, phosphate-buffered saline; HPLC, high-performance liquid chromatography; 1400W, N-(3(aminomethyl) benzyl) acetamidine dihydrochloride; Gö6976, 12-(2-cyanoethyl)-6,7,12,13-tetrahydro-13-methyl-5-oxo-5H-indolo(2,3-a)pyrrolo(3,4-c)-carbazole; NF, nuclear factor; EPT, enalaprilat.
- The American Society for Pharmacology and Experimental Therapeutics