Down-regulation of neprilysin (EC3.4.24.11) expression in vascular endothelial cells by laminar shear stress involves NADPH oxidase-dependent ROS production

The International Journal of Biochemistry & Cell Biology 41 (2009) 2287–2294

Contents lists available at ScienceDirect

The International Journal of Biochemistry & Cell Biology journal homepage: www.elsevier.com/locate/biocel

Down-regulation of neprilysin (EC3.4.24.11) expression in vascular endothelial cells by laminar shear stress involves NADPH oxidase-dependent ROS production Paul A. Fitzpatrick a , Anthony F. Guinan a , Tony G. Walsh a , Ronan P. Murphy b , Maria T. Killeen a , Nicholas P. Tobin a , Adrian R. Pierotti c , Philip M. Cummins a,∗ a

School of Biotechnology, Dublin City University, Glasnevin, Dublin 9, Ireland School of Health & Human Performance, Dublin City University, Dublin, Ireland c Department of Biological & Biomedical Sciences, Glasgow Caledonian University, Glasgow, United Kingdom b

a r t i c l e

i n f o

Article history: Received 17 February 2009 Received in revised form 14 May 2009 Accepted 16 May 2009 Available online 21 May 2009 Keywords: Neprilysin Endothelium Laminar shear stress NADPH oxidase Reactive oxygen species

a b s t r a c t Neprilysin (NEP, neutral endopeptidase, EC3.4.24.11), a zinc metallopeptidase expressed on the surface of endothelial cells, influences vascular homeostasis primarily through regulated inactivation of natriuretic peptides and bradykinin. Earlier in vivo studies reporting on the anti-atherosclerotic effects of NEP inhibition and on the atheroprotective effects of flow-associated laminar shear stress (LSS) have lead us to hypothesize that the latter hemodynamic stimulus may serve to down-regulate NEP levels within the vascular endothelium. To address this hypothesis, we have undertaken an investigation of the effects of LSS on NEP expression in vitro in bovine aortic endothelial cells (BAECs), coupled with an examination of the signalling mechanism putatively mediating these effects. BAECs were exposed to physiological levels of LSS (10 dynes/cm2 , 24 h) and harvested for analysis of NEP expression using real-time PCR, Western blotting, and immunocytochemistry. Relative to unsheared controls, NEP mRNA and protein were substantially down-regulated by LSS (≥50%), events which could be prevented by treatment of BAECs with either N-acetylcysteine, superoxide dismutase, or catalase, implicating reactive oxygen species (ROS) involvement. Employing pharmacological and molecular inhibition strategies, the signal transduction pathway mediating shear-dependent NEP suppression was also examined, and roles implicated for G␤␥, Rac1, and NADPH oxidase activation in these events. Treatment of static BAECs with angiotensin-II, a potent stimulus for NADPH oxidase activation, mimicked the suppressive effects of shear on NEP expression, further supporting a role for NADPH oxidase-dependent ROS production. Interestingly, inhibition of receptor tyrosine kinase signalling had no effect. In conclusion, we confirm for the first time that NEP expression is down-regulated in vascular endothelial cells by physiological laminar shear, possibly via a mechanotransduction mechanism involving NADPH oxidase-induced ROS production. © 2009 Elsevier Ltd. All rights reserved.

1. Introduction Neprilysin (NEP, neutral endopeptidase, EC3.4.24.11) is an integral plasma membrane zinc metallopeptidase (93 kDa) belonging to the M13 family of peptidases (Barrett et al., 2001). NEP inactivates a range of neuropeptides throughout the mammalian cardiovascular, nervous, renal, and immune systems. Related NEP-like enzymes include endothelin-converting enzyme (ECE1/2), KELL (blood group antigen), and PEX (metalloproteinase fragment with integrin-binding activity) (Turner et al., 2001), whilst structural and functional parallels between NEP and angiotensin-converting

∗ Corresponding author. Tel.: +3531 700 7857; fax: +3531 700 5412. E-mail address: [email protected] (P.M. Cummins). 1357-2725/$ – see front matter © 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.biocel.2009.05.011

enzyme (ACE1/2) have also been described (Skidgel and Erdös, 2004; Corti et al., 2001). Within the vasculature, NEP is expressed on the surface of endothelial cells where it contributes to vascular homeostasis through the regulated inactivation of natriuretic (e.g. ANP, atrial natriuretic peptide) and vasodilatory (e.g. BK, bradykinin) peptides (Graf et al., 1995). Consequently, inhibition of NEP has been shown to induce elevated natriuresis and vasodilation (Quaschning et al., 2003), to promote anti-proliferative and anti-migratory effects on vascular smooth muscle cells (Barber et al., 2005), and to decrease peripheral vascular resistance and blood pressure. Consistent with these observations, several animal model studies have also reported on the anti-atherosclerotic effects of NEP inhibition (Kugiyama et al., 1996; Grantham et al., 2000; Arnal et al., 2001; Jandeleit-Dahm et al., 2005). With respect to this latter point, the

2288

P.A. Fitzpatrick et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 2287–2294

established role of blood flow-associated shearing forces in atherogenic lesion development is potentially relevant. Laminar shear stress (LSS) is known to induce an “atheroprotective” endothelial phenotype, which disfavours pathological remodelling of the vessel wall, whilst attenuated or oscillatory shear (OSS) conditions common to vessel curvatures and bifurcations, can lead to endothelial dysfunction and lesion formation (Traub and Berk, 1998). In view of the anti-atherosclerotic effects of pharmacological NEP inhibition reported above, we therefore hypothesized that the atheroprotective influence of physiologic laminar shear in vivo may entail down-regulation of NEP levels. The regulatory impact of LSS on ACE1 and ECE1 has previously been addressed to some extent (reviewed in Cummins et al., 2004). Studies by Rieder et al. (1997) and Masatsugu et al. (1998, 2003) report shear-dependent suppression of vascular endothelial ACE1 and ECE1 mRNA expression, respectively. Moreover, possible roles for reactive oxygen species (ROS) (Masatsugu et al., 2003) and nitric oxide (Pertrini et al., 2003) in these events have been proposed, although the signalling pathways involved remain largely undefined. To date, nothing is known of the dynamic relationship between NEP and shear stress within the vascular endothelium, although inspection of the NEP promoter reveals a shear stress response element (–GAGACC–) (Sezaki et al., 2003), a common feature of shear-sensitive genes (including ACE1 and ECE1) (Shinoki et al., 1998). In the present study, we have addressed our hypothesis through an investigation of the effects of LSS on vascular endothelial NEP expression in vitro. Our findings demonstrate for the first time that chronic laminar shear substantially down-regulates NEP mRNA and protein expression in vascular endothelial cells. The signal transduction pathway mediating these events has also been examined to some extent, and a role implicated for NADPH oxidase-dependent ROS production. 2. Experimental 2.1. Materials All reagents used in this study were of the highest purity and unless otherwise stated, were mostly obtained from Sigma–Aldrich (Dorset, UK). LipofectamineTM reagent was purchased from Invitrogen (Groningen, The Netherlands). For protein assays and Western blotting purposes, bicinchoninic acid (BCA) reagent and West Pico SuperSignal reagent, respectively, were purchased from Pierce Chemicals (Northumberland, UK). Rat anti-NEP monoclonal IgG was obtained from Alpha Diagnostic International (San Antonio, TX), whilst HRP-conjugated goat anti-rat IgG was obtained from Amersham Biosciences (Buckinghamshire, UK). Alexafluor 488-conjugated goat anti-rat IgG, used for immunocytochemical purposes, was acquired from Molecular Probes (Eugene, OR). All primers were purchased from MWG Biotech (Buckinghamshire, UK).

harvested for analysis of NEP expression by real-time PCR, Western immunoblotting, and immunocytochemistry as described below. A number of agents were also examined for their ability to influence shear-dependent changes in NEP expression. These included: (i) antioxidants – 5 mM N-acetyl-l-cysteine (NAC), 1000 U/mL catalase (CAT), 100 U/mL superoxide dismutase (SOD); (ii) pharmacological inhibitors – 10 mM apocynin, 50 ␮M NSC23766, and 50 mM genistein; (iii) oxidants – 10 mM hydrogen peroxide (H2 O2 ) and 0.1 mM angiotensin-II (Ang-II); and (iv) molecular inhibitors – ␤ adrenergic receptor kinase C-terminus (␤ARK-ct). For antioxidant and inhibitor studies, compounds were prepared in RPMI 1640 complete media and incubated with BAECs approximately 1 h prior to initiation of shear until the end of the experiment. For oxidant studies, compounds were prepared in RPMI 1640 complete media and incubated with static BAECs for 24 h (H2 O2 ) and 8 h (Ang-II), after which, cells were harvested for mRNA analysis. Delivery of ␤ARK-ct into BAECs was achieved via transfection with LipofectamineTM reagent as previously described (Cotter et al., 2004). Assessment of transfection efficiency was routinely determined by co-transfection with green fluorescent protein (GFP), and mock transfections were included in all ␤ARK-ct experiments. 2.3. Cell lysate preparation and protein assay Following trypsinization and washing in PBS, cell suspensions were pelleted by centrifugation, the supernatant removed, and cells resuspended in 1× lysis buffer (20 mM Tris, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% (v/v) Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM ␤-glycerophosphate, 1 mM sodium orthovanadate, and 1 mg/mL leupeptin) in a microfuge tube. Suspensions were then subjected to three cycles of rapid freeze–thaw, followed by continuous tube rotation at 4 ◦ C overnight. Resulting lysates were stored at −20 ◦ C (short-term) and −80 ◦ C (long-term). Protein was routinely quantified by BCA microplate assay (Smith et al., 1985). 2.4. Western immunoblotting Following shearing experiments, BAECs were harvested and total lysate samples resolved by 10% SDS-PAGE under reducing conditions according to the method of Laemmli (1970). Gels were electroblotted onto nitrocellulose membrane using an ATTO semidry transfer system (1 h, 100 V) and membranes blocked for 2 h in Tris-buffered saline (TBS: 10 mM Tris pH 8.0, 150 mM NaCl) containing 5% (w/v) BSA. Immunostaining for NEP was as follows; primary antisera: overnight incubation in 1:50 rat anti-NEP monoclonal IgG. Secondary antisera: 3 h incubation in 1:500 HRPconjugated goat anti-rat IgG. Antibody–antigen complexes were detected by incubation in West Pico SuperSignal reagent, followed by exposure to autoradiographic film (Amersham Hyperfilm ECL). For quantitative comparisons between bands, scanning densitometry was performed using NIH Image v1.61 software. Membranes were routinely stained with Ponceau S to normalize for protein loading/transfer.

2.2. Cell culture and shear stress

2.5. Real-time PCR

BAECs (Coriell Cell Repository, Camden, NJ) were cultured in RPMI 1640 complete media (supplemented with 10% fetal bovine serum, 100 U/mL penicillin, and 100 ␮g/mL streptomycin). Cells were grown in a humidified atmosphere of 5% CO2 /95% air at 37 ◦ C. For all experiments, cells between passages 8 and 14 were used. For shear stress studies, BAECs were seeded at 1 × 104 cells/cm2 into 6well plates and grown to confluency. Media was then replenished (4 mL) and cells exposed to either 0 or 10 dynes/cm2 of laminar shear stress for 24 h on an orbital rotator as previously described (Hendrickson et al., 1999; Colgan et al., 2007). Post-shear, cells were

Following shearing experiments, extraction of total BAEC RNA and performance of real-time PCR were conducted as previously described (von Offenberg Sweeney et al., 2005). To monitor changes in NEP mRNA levels, gene-specific primers for NEP were as follows: forward 5 -caaagccaaagaagaaacag-3 , reverse 5 catctcttaaaatgtcaaag-3 . Cycle conditions (×45) were as follows: denature at 95 ◦ C for 20 s, annealing at 55 ◦ C for 30 s, elongation at 72 ◦ C for 30 s. Gene-specific primers and cycling conditions for glyceraldehyde phosphate dehydrogenase (GAPDH, included for normalization purposes) and eNOS have been published previously

P.A. Fitzpatrick et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 2287–2294

2289

Fig. 1. BAEC response to LSS. BAECs were exposed to an applied shear of 10 dynes/cm2 for 24 h. (A) Morphological alignment of cells in the direction of the shear vector as shown by phase contrast microscopy (i and ii). Actin cytoskeletal realignment as shown by rhodamine-phalloidin staining with DAPI nuclear staining also shown in blue (iii and iv). Down-regulation of NEP expression as shown by immunofluorescence microscopy (v and vi). Shear vector is shown by dotted arrow. Blue arrow in (iv) highlights cortical actin formation. White arrows in (v) highlight cellular NEP staining. (B) Representative Western blots showing impact of shear stress on NEP protein levels. (C) Impact of shear stress on NEP mRNA levels. (D) Impact of shear stress on eNOS mRNA and nitrite levels. Histograms represent fold change to unsheared control and are averaged from three independent experiments ±SEM. *P ≤ 0.05 versus unsheared control. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

(Coen et al., 2004). All primer pairs used were routinely screened for non-specific primer-dimer products by melt curve analysis and by standard PCR in conjunction with agarose gel electrophoresis. 2.6. Immunocytochemistry Following shearing experiments, BAECs were prepared for immunocytochemical analysis according to the method of Groarke et al. (2001), with minor modifications. Briefly, cells in 6-well plates

were washed twice with PBS and fixed with 3% formaldehyde for 15 min. Cells were then washed, permeabilized with 0.2% Triton X100 for 15 min, and blocked in 5% BSA for 30 min. Immunostaining for NEP was as follows; primary antisera: 3 h incubation in 1:50 rat anti-NEP monoclonal IgG. Secondary antisera: 1 h incubation in 1:400 Alexafluor 488-conjugated goat anti-rat IgG. The bottom of the well was then removed and sealed using Dako mounting media (Dako Cytomation, Cambridgeshire) and cover slips for visualization by fluorescence microscopy (Olympus BX50). Appropriate antibody controls were included for all experiments. Nuclear (DAPI,

Fig. 2. Role of ROS in LSS-dependent NEP suppression. (A) Impact of NAC, SOD, and CAT on shear-dependent decrease in NEP mRNA levels following BAEC exposure to an applied shear of 10 dynes/cm2 for 24 h; (B) impact of H2 O2 and Ang-II on NEP mRNA levels in unsheared BAECs after 24 and 8 h treatments, respectively. Histograms represent fold change to unsheared control and are averaged from three independent experiments ±SEM. *P ≤ 0.05 versus unsheared control. ␾ P ≤ 0.05 versus shear.

2290

P.A. Fitzpatrick et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 2287–2294

Fig. 3. Role of NADPH oxidase in LSS-dependent NEP suppression. BAECs were exposed to an applied shear of 10 dynes/cm2 for 24 h in the absence and presence of NADPH oxidase inhibitors. Impact of apocynin and NSC23766 on the shear-dependent reduction in NEP expression levels as monitored by: (A) immunofluorescence microscopy, (B) Western blotting, and (C) real-time PCR. Histograms represent fold change to unsheared control and are averaged from three independent experiments ±SEM. *P ≤ 0.05 versus unsheared control. ␾ P ≤ 0.05 versus shear. White arrows in (A) highlight cellular NEP staining. Blots shown in (B) are representative.

500 ng/mL, 3 min), F-actin (Rhodamine-Phalloidin, 1:400, 10 min), and superoxide (10 mM dihydroethidium, 15 min – DeKeulenaer et al., 1998) staining were also conducted. 2.7. Statistical analysis Results are expressed as mean ± SEM. Experimental points were performed in triplicate with a minimum of three independent experiments (n = 3). Statistical comparisons between control versus treated groups were made by Student’s unpaired t-test and Wilcoxon Signed-Rank test (for non-parametric comparisons). A value of *P ≤ 0.05 (or ␾ P ≤ 0.05) was considered significant. 3. Results 3.1. BAEC response to LSS The effect of LSS on BAEC morphology and biochemical characteristics was examined. Following exposure of cells to LSS (10 dynes/cm2 , 24 h), morphological realignment of cells was clearly observed in parallel with reduced actomyosin stress fibre formation and enhanced cortical actin (Fig. 1A(i–iv)). A significant shear-dependent increase in levels of eNOS mRNA (Fig. 1D, LHS histogram) and nitrite production (Fig. 1D, RHS histogram) were also observed, and were fully consistent with elevated eNOS protein levels previously reported by this group under identical experimental conditions (Tobin et al., 2008). The effect of LSS on BAEC ROS production was also examined by fluorescence microscopy using superoxide-specific dihydroethidium staining (DeKeulenaer et al.,

1998). Following exposure of cells to LSS (0–3 h, 10 dynes/cm2 ), a transient surge in superoxide (O2 −• ) production was observed, peaking between 1 and 2 h of shear (data not shown). In parallel with these observations, levels of NEP immunoreactivity (Fig. 1A(v and vi)), protein (Fig. 1B), and mRNA (Fig. 1C) were all substantially reduced (≥50% reduction) under shear. 3.2. Involvement of ROS in LSS-dependent NEP suppression The effects of blocking ROS production on LSS-dependent NEP suppression in BAECs was investigated. BAECs were exposed to LSS (10 dynes/cm2 , 24 h) in the absence and presence of either NAC, SOD, or CAT. In the presence of either of these antioxidants, shear-dependent suppression of NEP mRNA levels was completely prevented (Fig. 2A). Moreover, when either H2 O2 or Ang-II were employed to induce ROS elevation in BAECs under static (unsheared) conditions, NEP levels were significantly decreased relative to untreated controls (Fig. 2B). 3.3. Involvement of NADPH oxidase in LSS-dependent NEP suppression The effects of selectively blocking NADPH oxidase activation on LSS-dependent NEP suppression in BAECs was investigated. Cells were exposed to LSS (10 dynes/cm2 , 24 h) in the absence and presence of either of the NADPH oxidase inhibitors, apocynin (Johnson et al., 2002) or NSC23766, the latter blocking activation of Rac1 GTPase, and consequently activation of NAPDH oxidase (Miyano and Sumimoto, 2007). In the presence of either of these

P.A. Fitzpatrick et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 2287–2294

2291

Fig. 4. Role of receptor tyrosine kinases in LSS-dependent NEP suppression. BAECs were exposed to an applied shear of 10 dynes/cm2 for 24 h in the absence and presence of genistein, an RTK signalling inhibitor. Impact of genistein on shear-dependent reduction in NEP expression levels as monitored by: (A) Western blotting, (B) real-time PCR, and (C) immunocytochemistry. Histograms represent fold change to unsheared control and are averaged from three independent experiments ±SEM. *P ≤ 0.05 versus unsheared control. White arrows in (C) highlight cellular NEP staining. Blots shown in (A) are representative.

inhibitors, shear-dependent suppression of NEP immunoreactivity (Fig. 3A(i–vi)), protein (Fig. 3B), and mRNA (Fig. 3C) were observed to be completely reversed. 3.4. Involvement of receptor tyrosine kinases in LSS-dependent NEP suppression The effects of selectively blocking receptor tyrosine kinase (RTK) activation on LSS-dependent NEP suppression in BAECs was also investigated. Cells were exposed to LSS (10 dynes/cm2 , 24 h) in the absence and presence of genistein, an RTK inhibitor. Inhibitor treatment did not prevent the shear-dependent suppression of NEP protein levels, as monitored by Western blotting (Fig. 4A), real-time PCR (Fig. 4B), and immunocytochemistry (Fig. 4C). 3.5. Involvement Gˇ in LSS-dependent NEP suppression The effects of selectively blocking G␤␥ dimer activation on LSSdependent NEP suppression in BAECs was also investigated using ␤ARK-ct, a G␤␥ sequestering agent. Mock transfected and ␤ARKct transfected BAECs were exposed to LSS (10 dynes/cm2 ) for 24 h. We observed that blockade of G␤␥ led to complete reversal of the shear-dependent suppression of NEP protein and mRNA levels, as monitored by Western blotting (Fig. 5A) and real-time PCR (Fig. 5B). 4. Discussion In view of the widely reported anti-atherosclerotic effects of pharmacological NEP inhibition (Kugiyama et al., 1996; Grantham

et al., 2000; Arnal et al., 2001; Jandeleit-Dahm et al., 2005), in conjunction with the atheroprotective influence of blood flowassociated laminar shear stress (Traub and Berk, 1998), we hypothesized that physiologic laminar shear may serve to downregulate NEP levels within the vascular endothelium in vivo, with putative consequences for vessel homeostasis. With this paper we have begun to address this hypothesis, starting with a comprehensive investigation of the effects of LSS on NEP expression in BAECs, coupled with an examination of the signal transduction pathway putatively mediating these effects. We now confirm for the first time that NEP mRNA and protein expression levels are substantially down-regulated in endothelial cells in vitro by laminar shear. In parallel investigations, we also observed sheardependent up-regulation of eNOS expression and NO production, known atheroprotective features of laminar shear (Traub and Berk, 1998). Our findings are consistent with earlier papers reporting LSSdependent suppression of ACE1 and ECE1 expression (Rieder et al., 1997; Masatsugu et al., 1998, 2003). Like NEP, these metallopeptidases have a profound impact on endothelial function and vascular tone, and also possess a shear stress response element(s) within their promoter region (Sezaki et al., 2003; Cummins et al., 2004; Rieder et al., 1997; Masatsugu et al., 1998). The contribution of ROS to vascular pathology is well established. Endothelial dysfunction and associated diseases such as atherosclerosis and hypertension for example, are characterized by excess vascular oxidant production (Schulz et al., 2008; Lassègue and Clempus, 2003). Numerous studies also indicate that the intracellular ROS milieu is an important physiological modulator of signalling cascades, gene transcription, cell growth, angiogenesis, and apoptosis (Chen and Keaney, 2004; Ullrich and

2292

P.A. Fitzpatrick et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 2287–2294

Fig. 5. Role of G␤␥ in LSS-dependent NEP suppression. BAECs were exposed to an applied shear of 10 dynes/cm2 for 24 h in the absence and presence of an inhibitor of G␤␥ activation. Impact of ␤ARK-ct on shear-dependent reduction in NEP expression levels as monitored by: (A) Western blotting and (B) real-time PCR. Histogram represents fold change to unsheared control and is averaged from three independent experiments ±SEM. *P ≤ 0.05 versus unsheared control. ␾ P ≤ 0.05 versus shear. Blot shown in (A) is representative.

Bachschmid, 2000; Irani, 2000; Ushio-Fukai, 2006). Within this context, shear-induced intracellular ROS production leading to modulation of vascular endothelial gene expression has been reported (DeKeulenaer et al., 1998; Lehoux, 2006; Chiu et al., 1997), allowing us to consider a possible role for shear-dependent ROS signalling in our current BAEC/NEP model. Our subsequent investigations using BAECs demonstrated that laminar shear can induce a transient increase (0–3 h) in intracellular superoxide levels (as monitored via dihydroethidium assay). Moreover, we determined that the shear-dependent reduction in NEP expression can be completely prevented by treatment of BAECs with either N-acetylcysteine (an antioxidant), or with either superoxide dismutase or catalase. Collectively, these findings strongly implicate the involvement of superoxide and its dismutated hydrogen peroxide end-product (the primary forms of ROS in vascular cells) in the LSS-induced suppression of endothelial NEP levels. Upstream of oxidant production, all layers of the vascular wall have ROS-generating enzyme systems. In particular, the multisubunit enzyme, NADPH oxidase, is known to be a major superoxide source in vascular cells. Endothelial cells have been shown to express all of the essential subunits for this enzyme (p22phox , p47phox , p67phox , and Rac1) (Jones et al., 1996), as well as the Nox4 (and to a lesser extent, Nox2) isoform of the gp91phox catalytic subunit (Sorescu et al., 2002). Our investigations clearly demonstrated

that BAEC treatment of cells with the NADPH oxidase inhibitor, apocynin (Johnson et al., 2002), prevented shear-dependent NEP suppression. Rac1 activation, an established pre-requisite for association of the cytosolic and membrane components of the NADPH oxidase complex (Miyano and Sumimoto, 2007), when blocked using NSC23766, yielded similar results. Moreover, treatment of “static” BAECs with Ang-II, a potent stimulus for NADPH oxidase activation and superoxide production (Griendling and Ushio-Fukai, 2000), mimicked the suppressive effects of laminar shear on NEP expression levels. Based on these observations, we can conclude that shear-dependent down-regulation of endothelial NEP expression putatively involves NADPH oxidase-mediated superoxide production. Consistent with this conclusion, growing evidence implicates NADPH oxidase as an integral participant in flowinduced vascular remodelling (Yeh et al., 1999; Castier et al., 2005), whilst another recent study demonstrates transient activation of endothelial NADPH oxidase in response to laminar shear (Duerrschmidt et al., 2006), analogous to the transient ROS flux observed in our study. The possible contribution of elevated eNOS towards potentiation of ROS production in this model is also deserving of some discussion. It is now well known that under chronic pathological stimulus (e.g. oscillatory shear), NADPH oxidase-derived superoxide may react with eNOS-derived NO to produce peroxynitrite, directly leading to a nitrosative stress that can “uncouple” eNOS and contribute to endothelial dysfunction (Schulz et al., 2008). By contrast, under physiologic laminar shear, NADPH oxidase (and superoxide production) is only transiently up-regulated during the early shear onset period, with the shear-induced long-term up-regulation of eNOS (and NO) directly contributing to the long-term down-regulation of NADPH oxidase via both transcriptional and functional mechanisms (Duerrschmidt et al., 2006; Selemidis et al., 2007). In our view, it is unlikely that elevated eNOS potentiates ROS production during the early shear onset period for a number of reasons: (i) appreciable increases in shear-induced eNOS expression would not be fully realized during this short time period; and (ii) down-regulation of NEP expression by shear-dependent ROS production could be fully prevented by treatment with SOD (superoxide dismutase would compete with NO for superoxide), suggesting the absence and/or lack of effect of residual NO-derived peroxynitrite. Interestingly, blockade of RTK activation using genistein had no observable effect on shear-dependent NEP suppression. In parallel experiments, integrin blockade using both linear and cyclic RGD peptides also had no effect (data not shown). However, in view of the potential ambiguity with respect to RGD peptide specificity for different integrin ␤-tails, more detailed integrin blockade studies will have to be performed before we can definitively conclude lack of an integrin role in these shear-mediated events. Also noteworthy is our observation that transfection of BAECs with the functional G␤␥ scavenger, ␤ARK-ct, could prevent shear-dependent suppression of NEP. This points to a role for heterotrimeric Gprotein signalling upstream of Rac1 activation and ROS production in this cascade. Moreover, this is consistent with recent studies which demonstrate that G␤␥, through stimulation of a p114RhoGEF guanine exchange factor, can regulate RhoGTPase (RhoA, Rac1) activation, with consequences for ROS production and endothelial cell functions (Zeng et al., 2002; Niu et al., 2003). Downstream of oxidant production, signalling components mediating NEP regulation are of obvious interest. Mitogenactivated protein kinase (MAPK) signalling for example is highly redox sensitive (Chen and Keaney, 2004; Griendling et al., 2000). Indeed, peroxide-mediated activation of p38 MAPK (as well as upstream MAPKK regulators, MKK3 and MKK6) has been reported in human pulmonary endothelial cells (HPECs) and human umbilical vein endothelial cells (HUVECs) (Hashimoto et al., 2001; Huot et al., 1997). Both stretch- and shear-dependent modulation of ERK1/2

P.A. Fitzpatrick et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 2287–2294

signalling have also been reported to involve ROS production (Yeh et al., 1999; Wung et al., 1999), whilst the mechanoactivation of endothelial JNK by physiologic shear has been shown to require nitric oxide and superoxide (Go et al., 1999). Finally, a recent study by Masatsugu and co-workers indicates that shear-dependent ECE1 suppression in HUVECs involves ROS-mediated activation of a p38 MAPK/NF␬B pathway (Masatsugu et al., 2003). In view of these reports therefore, participation of MAPK signalling in the sheardependent regulation of NEP in endothelial cells is highly likely, and will undoubtedly be a focus of future investigations. In summary, this study confirms for the first time that NEP mRNA and protein expression are substantially down-regulated in vascular endothelial cells in vitro by physiological laminar shear stress. Selective inhibition strategies also implicate roles for G␤␥, Rac1, and NADPH oxidase activation, likely leading to transient superoxide production, in the mechanotransduction of these events. Although beyond the scope of the present study, future work will correlate flow-dependent changes in NEP levels with the temporal nature of NADPH oxidase activation under laminar versus oscillatory shear conditions, as various studies demonstrate “sustained” up-regulation of NADPH oxidase (and ROS production) under circumstances of pathological flow (Hwang et al., 2003; Jo et al., 2006). This will enable us to accurately model the dynamics of NEP regulation with respect to varying shear rates and pathological conditions. Moreover, whilst several studies clearly infer an as yet undefined role for NEP in shear-dependent pathologies such as atherosclerosis (Kugiyama et al., 1996; Grantham et al., 2000; Arnal et al., 2001; Jandeleit-Dahm et al., 2005), future investigations will need to delineate the precise functional consequences of shear-dependent NEP regulation on endothelial properties. Whilst the value of NEP inhibition as a desirable therapeutic strategy has been established in specific vascular diseases (e.g. intimal hyperplasia, resistance hypertension, and chronic heart failure) (Quaschning et al., 2003; Barber et al., 2005; Azizi et al., 2006), the present study suggests a potentially broader, more exploitable role for NEP in pathologies manifesting flow-dependent endothelial dysfunction and remodelling (e.g. atherosclerosis). We therefore anticipate that a fuller understanding of the dynamic regulatory relationship between blood flow-associated shear and NEP, and of its functional outcomes at the cellular level, will help clarify the full pathophysiologic relevance and therapeutic potential of this metallopeptidase. Acknowledgements This research was supported through funding from the Science Foundation Ireland Research Frontiers Programme and Enterprise Ireland Basic Research Grant Programme (P.M. Cummins). We also wish to acknowledge Dublin City University for internal support through the Postgraduate Scholarship (P.A. Fitzpatrick), Overhead Allocation (R.P. Murphy), Equipment Maintenance (P.M. Cummins), and International Visitors Programme (A.R. Pierotti) schemes. The authors also acknowledge the Higher Education Authority Programme for Research in Third Level Institutes (HEA PRTLI Cycle IV) Targeted Therapeutics & Theranostics (T3 ) Programme (P.M. Cummins, R.P. Murphy). References Arnal JF, Castano C, Maupas E, Mugniot A, Darblade B, Gourdy P, et al. Omapatrilat, a dual angiotensin-converting enzyme and neutral endopeptidase inhibitor, prevents fatty streak deposit in apolipoprotein E-deficient mice. Atherosclerosis 2001;155:291–5. Azizi M, Bissery A, Peyrard S, Guyene TT, Ozoux ML, Floch A, et al. Pharmacokinetics and pharmacodynamics of the vasopeptidase inhibitor AVE7688 in humans. Clin Pharmacol Ther 2006;79:49–61.

2293

Barber MN, Gaspari TA, Kairuz EM, Dusting GJ, Woods RL. Atrial natriuretic peptide preserves endothelial function during intimal hyperplasia. J Vasc Res 2005;42:101–10. Barrett AJ, Rawlings ND, O’Brien EA. The MEROPS database as a protease information system. J Struct Biol 2001;134:95–102. Castier Y, Brandes RP, Leseche G, Tedgui A, Lehoux S. p47phox-dependent NADPH oxidase regulates flow-induced vascular remodelling. Circ Res 2005;97:533–40. Chen K, Keaney Jr JF. Reactive oxygen species-mediated signal transduction. Endothelium 2004;11:109–21. Chiu JJ, Wung BS, Shyy JY, Hsieh HJ, Wang DL. Reactive oxygen species are involved in shear stress-induced intercellular adhesion molecule-1 expression in endothelial cells. Arterioscler Thromb Vasc Biol 1997;17:3570–7. Coen P, Cummins P, Birney Y, Devery R, Cahill P. Modulation of PGF-1␣ and nitric oxide production in bovine aortic endothelial cells by conjugated linoleic acid. Endothelium 2004;11:211–20. Colgan OC, Ferguson G, Collins NT, Murphy RP, Meade G, Cahill PA, et al. Regulation of bovine brain microvascular endothelial tight junction assembly and barrier function by laminar shear stress. Am J Physiol: Heart Circ Physiol 2007;292:H3190–7. Corti R, Burnett Jr JC, Rouleau JL, Ruschitzka F, Lüscher TF. Vasopeptidase inhibitors: a new therapeutic concept in cardiovascular disease? Circulation 2001;104:1856–62. Cotter EJ, von Offenberg Sweeney N, Coen PM, Birney YA, Glucksman MJ, Cahill PA, et al. Regulation of endopeptidases EC3.4.24.15 and EC3.4.24.16 in vascular endothelial cells by cyclic strain: role of Gi protein signalling. Arterioscler Thromb Vasc Biol 2004;24:457–63. Cummins PM, Cotter EJ, Cahill PA. Hemodynamic regulation of metallopeptidases within the vasculature. Protein Peptide Lett 2004;5:433–42. DeKeulenaer GW, Chappell DC, Ishizaka N, Nerem RM, Wayne-Alexander R, Griendling KK. Oscillatory and steady laminar shear stress differentially affect human endothelial redox state: role of a superoxide-producing NADH oxidase. Circ Res 1998;82:1094–101. Duerrschmidt N, Stielow C, Muller G, Pagano PJ, Morawietz H. NO-mediated regulation of NAD(P)H oxidase by laminar shear stress in human endothelial cells. J Physiol 2006;576:557–67. Go YM, Park H, Maland MC, Darley-Usmar VM, Stoyanov B, Wetzker R, et al. Evidence for peroxynitrite as a signalling molecule in flow-dependent activation of c-Jun NH(2)-terminal kinase. Am J Physiol 1999;277:H1647–53. Graf K, Koehne P, Grafe M, Zhang M, Auch-Schwelk W, Fleck E. Regulation and differential expression of neutral endopeptidase 24.11 in human endothelial cells. Hypertension 1995;26:230–5. Grantham JA, Schirger JA, Wennberg PW, Sandberg S, Heublein DM, Subkowski T, et al. Modulation of functionally active endothelin-converting enzyme by chronic neutral endopeptidase inhibition in experimental atherosclerosis. Circulation 2000;101:1976–81. Griendling KK, Sorescu D, Lassegue B, Ushio-Fukai M. Modulation of protein kinase activity and gene expression by reactive oxygen species and their role in vascular physiology and pathophysiology. Arterioscler Thromb Vasc Biol 2000;20:2175–83. Griendling KK, Ushio-Fukai M. Reactive oxygen species as mediators of angiotensin II signalling. Regul Pept 2000;91:21–7. Groarke DA, Drmota T, Bahia DS, Evans NA, Wilson S, Milligan G. Analysis of the C-terminal tail of the rat thyrotropin-releasing hormone receptor-1 in interactions and co-internalization with beta-arrestin 1 green fluorescent protein. Mol Pharmacol 2001;59:375–85. Hashimoto S, Gon Y, Matsumoto K, Takeshita I, Horie T. N-acetylcysteine attenuates TNF-alpha-induced p38 MAP kinase activation and p38 MAP-kinase-mediated IL-8 production by human pulmonary vascular endothelial cells. Br J Pharmacol 2001;132:270–6. Hendrickson RJ, Cahill PA, Sitzmann JV, Redmond EM. Ethanol enhances basal and flow-stimulated nitric oxide synthase activity in vitro by activating an inhibitory guanine nucleotide binding protein. J Pharmacol Exp Ther 1999;289:1293– 300. Huot J, Houle F, Marceau F, Landry J. Oxidative stress-induced actin reorganization mediated by the p38 mitogen-activated protein kinase/heat shock protein 27 pathway in vascular endothelial cells. Circ Res 1997;80:383–92. Hwang J, Ing MH, Salazar A, Lassègue B, Griendling K, Navab M, et al. Pulsatile versus oscillatory shear stress regulates NADPH oxidase subunit expression: implication for native LDL oxidation. Circ Res 2003;93:1225–32. Irani K. Oxidant signalling in vascular cell growth, death, and survival: a review of the roles of reactive oxygen species in smooth muscle and endothelial cell mitogenic and apoptotic signalling. Circ Res 2000;87:179–83. Jandeleit-Dahm K, Lassila M, Davis BJ, Candido R, Johnston CI, Allen TJ, et al. Antiatherosclerotic and renoprotective effects of combined angiotensin-converting enzyme and neutral endopeptidase inhibition in diabetic apolipoprotein Eknockout mice. J Hypertens 2005;23:2071–82. Johnson DK, Schillinger KJ, Kwait DM, Hughes CV, McNamara EJ, Ishmael F, et al. Inhibition of NADPH oxidase activation in endothelial cells by ortho-methoxysubstituted catechols. Endothelium 2002;9:191–203. Jo H, Song H, Mowbray A. Role of NADPH oxidases in disturbed flow- and BMP4-induced inflammation and atherosclerosis. Antioxid Redox Signal 2006;8:1609–19. Jones SA, O’Donnell VB, Wood JD, Broughton JP, Hughes EJ, Jones OT. Expression of phagocyte NADPH oxidase components in human endothelial cells. Am J Physiol 1996;271:H1626–34. Kugiyama K, Sugiyama S, Matsumura T, Ohta Y, Doi H, Yasue H. Suppression of atherosclerotic changes in cholesterol-fed rabbits treated with an oral inhibitor

2294

P.A. Fitzpatrick et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 2287–2294

of neutral endopeptidase 24.11 (EC 3.4.24.11). Arterioscler Thromb Vasc Biol 1996;16:1080–7. Laemmli U. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680–5. Lassègue B, Clempus RE. Vascular NAD(P)H oxidases: specific features, expression, and regulation. Am J Physiol Regul Integr Comp Physiol 2003;285:R277–97. Lehoux S. Redox signalling in vascular responses to shear and stretch. Cardiovasc Res 2006;71:269–79. Masatsugu K, Itoh H, Chun TH, Ogawa Y, Tamura N, Yamashita J, et al. Physiologic shear stress suppresses endothelin-converting enzyme-1 expression in vascular endothelial cells. J Cardiovasc Pharmacol 1998;31:S42–5. Masatsugu K, Itoh H, Chun T, Saito T, Yamshita Y, Doi K, et al. Shear stress attenuates endothelin and endothelin-converting enzyme expression through oxidative stress. Regul Pept 2003;111:13–9. Miyano K, Sumimoto H. Role of the small GTPase Rac in p22phox-dependent NADPH oxidases. Biochimie 2007;89:1133–44. Niu J, Profirovic J, Vaiskunaite R, Vovno-Yasenetskaya T. G-protein betagamma subunits stimulate p114RhoGEF, a guanine nucleotide exchange factor for RhoA and Rac1: regulation of cell shape and reactive oxygen species production. Circ Res 2003;93:848–56. Pertrini CM, Miyakawa AA, Laurindo FR, Krieger JE. Nitric oxide regulates angiotensin-I converting enzyme under static conditions but not under shear stress. Braz J Med Biol Res 2003;36:1175–8. Quaschning T, Galle J, Wanner C. Vasopeptidase inhibition: a new treatment approach for endothelial dysfunction. Kidney Int Suppl 2003;84:S54–7. Rieder MJ, Carmona R, Krieger JE, Pritchard KA, Greene AS. Suppression of angiotensin-converting enzyme expression and activity by shear stress. Circ Res 1997;80:312–9. Schulz E, Jansen T, Wenzel P, Daiber A, Münzel T. Nitric oxide, tetrahydrobiopterin, oxidative stress, and endothelial dysfunction in hypertension. Antioxid Redox Signal 2008;10:1115–26. Selemidis S, Dusting GJ, Peshavariya H, Kemp-Harper BK, Drummond GR. Nitric oxide suppresses NADPH oxidase-dependent superoxide production by S-nitrosylation in human endothelial cells. Cardiovasc Res 2007;75:349– 58. Sezaki N, Ishimaru F, Tabayashi T, Kataoka I, Nakase K, Fujii K, et al. The type 1 CD10/neutral endopeptidase 24.11 promoter: functional characterization of the 5 -untranslational region. Br J Hematol 2003;123:177–83.

Shinoki N, Kawasaki T, Minamino N, Okahara K, Ogawa A, Ariyoshi H, et al. Shear stress down-regulates gene transcription and production of adrenomedullin in human aortic endothelial cells. J Cell Biochem 1998;71:109–15. Skidgel RA, Erdös EG. Angiotensin converting enzyme (ACE) and neprilysin hydrolyze neuropeptides: a brief history, the beginning and follow-ups to early studies. Peptides 2004;25:521–5. Smith PK, Krohn RI, Hermanson GT, Mallia AK, Gartner FH, Provenzano MD, et al. Measurement of protein using bicinchoninic acid. Anal Biochem 1985;150:76–85. Sorescu D, Weiss D, Lassègue B, Clempus RE, Szöcs K, Sorescu GP, et al. Superoxide production and expression of nox family proteins in human atherosclerosis. Circulation 2002;105:1429–35. Traub O, Berk BC. Laminar shear stress: mechanisms by which endothelial cells transduce an atheroprotective force. Arterioscler Thromb Vasc Biol 1998;18:677–85. Tobin N, Henehan GT, Murphy RP, Atherton JC, Guinan AF, Kerrigan S, et al. Helicobacter pylori-induced inhibition of endothelial cell functions: a role for VacA-dependent nitric oxide reduction. Am J Physiol Heart Circ Physiol 2008;295:H1403–13. Turner AJ, Isaac RE, Coates D. The neprilysin (NEP) family of zinc metalloendopeptidases: genomics and function. Bioessays 2001;23:261–9. Ullrich V, Bachschmid M. Superoxide as a messenger of endothelial function. Biochem Biophys Res Commun 2000;278:1–8. Ushio-Fukai M. Redox signalling in angiogenesis: role of NADPH oxidase. Cardiovasc Res 2006;71:226–35. von Offenberg Sweeney N, Cummins PM, Cotter EJ, Fitzpatrick PA, Birney YA, Redmond EM, et al. Cyclic strain-mediated regulation of vascular endothelial cell migration and tube formation. Biochem Biophys Res Commun 2005;329:573–82. Wung BS, Cheng JJ, Chao YJ, Hsieh HJ, Wang DL. Modulation of Ras/Raf/extracellular signal-regulated kinase pathway by reactive oxygen species is involved in cyclic strain-induced early growth response-1 gene expression in endothelial cells. Circ Res 1999;84:804–12. Yeh LH, Park YJ, Hansalia RJ, Ahmed IS, Deshpande SS, Goldschmidt-Clermont PJ, et al. Shear-induced tyrosine phosphorylation in endothelial cells requires Rac1dependent production of ROS. Am J Physiol: Cell Physiol 1999;276:C838–47. Zeng H, Zhao D, Mukhopadhyay D. KDR stimulates endothelial cell migration through heterotrimeric G protein Gq/11-mediated activation of a small GTPase RhoA. J Biol Chem 2002;277:46791–8.