Protection from cigarette smoke‐induced vascular injury by recombinant human relaxin‐2 (serelaxin)

Abstract Smoking is regarded as a major risk factor for the development of cardiovascular diseases (CVD). This study investigates whether serelaxin (RLX, recombinant human relaxin‐2) endowed with promising therapeutic properties in CVD, can be credited of a protective effect against cigarette smoke (CS)‐induced vascular damage and dysfunction. Guinea pigs exposed daily to CS for 8 weeks were treated with vehicle or RLX, delivered by osmotic pumps at daily doses of 1 or 10 μg. Controls were non‐smoking animals. Other studies were performed on primary guinea pig aortic endothelial (GPAE) cells, challenged with CS extracts (CSE) in the absence and presence of 100 ng/ml (17 nmol/l) RLX. In aortic specimens from CS‐exposed guinea pigs, both the contractile and the relaxant responses to phenylephrine and acetylcholine, respectively, were significantly reduced in amplitude and delayed, in keeping with the observed adverse remodelling of the aortic wall, endothelial injury and endothelial nitric oxide synthase (eNOS) down‐regulation. RLX at both doses maintained the aortic contractile and relaxant responses to a control‐like pattern and counteracted aortic wall remodelling and endothelial derangement. The experiments with GPAE cells showed that CSE significantly decreased cell viability and eNOS expression and promoted apoptosis by sparkling oxygen free radical‐related cytotoxicity, while RLX counterbalanced the adverse effects of CSE. These findings demonstrate that RLX is capable of counteracting CS‐mediated vascular damage and dysfunction by reducing oxidative stress, thus adding a tile to the growing mosaic of the beneficial effects of RLX in CVD.


Introduction
Smoking is regarded as one of the major risk factors for the development of cardiovascular diseases (CVD), accounting for 80% increased risk for coronary artery insufficiency in smokers compared with nonsmokers [1]. Recently, the World Health Organization (WHO) predicted that, by 2030, deaths related to cigarette smoke (CS) will increase up to eight million individuals per year worldwide [2]. This trend is expected to increase in parallel with the earlier onset of tobacco use. Furthermore, clinical studies have shown that young smokers have a higher risk to develop vascular functional and histological abnormalities which are early hallmarks of atherosclerosis [3,4]. Hence, there is a major interest in the development of new therapeutic strategies that can prevent or reduce the adverse effects of CS on the cardiovascular system. Increasing evidence supports the knowledge that oxidative stress and inflammation provide the pathophysiological link between cigarette smoking and CVD [5][6][7][8][9][10]. Endothelial dysfunction is one of the earliest pathological events that occur in the development of CVD [7]. Several clinical and experimental studies indicate that CS oxidants can directly impair endothelial function, primarily through a lack of nitric oxide bioavailability which results both from decreased nitric oxide formation by the dysfunctional endothelium and increased nitric oxide scavenging by oxygen free radicals [10][11][12][13]. There is conflicting evidence regarding regulation of endothelial nitric oxide synthase (eNOS) by CS extract (CSE) at the transcriptional/translational level in cellular models with some suggesting reduced eNOS mRNA/protein levels [14][15][16] while others demonstrating unchanged [17] or even increased eNOS levels [18]: the latter possibly likely represents an attempt of endothelial cells to compensate for eNOS impairment. These functional alterations are accompanied by histological changes in vascular wall consisting in increase in the intima-media wall thickness, a marker of early atherosclerosis, [4,18,19], elastic fibres damage, cytoplasmic vacuolation of smooth muscle cells of the intima-media, and endothelial loss caused by disruption of the adhesion between endothelial cells and their basement membrane [20,21].
On the above grounds, we designed the present study to investigate whether serelaxin (RLX), the recombinant molecule corresponding to the human hormone relaxin-2 [22], shown to possess promising therapeutic properties in cardiac diseases [23,24], may be credited of a protective effect against CS-induced vascular damage and dysfunction. This working hypothesis is based on the following mainstays: (i) blood vessels are a physiological target of RLX and their cells express the specific RLX receptor RXFP1 [25]; (ii) RLX upregulates NOS expression and nitric oxide production in vascular endothelial cells [26][27][28][29][30]; (iii) RLX improves inflammation-induced endothelial dysfunction and NOS fall [31,32]; and (iv) RLX reduces oxidative stress and nitric oxide failure in different animal models of vascular injury [33][34][35][36][37].

Reagents
Serelaxin was kindly provided by the RRCA Relaxin Foundation (Florence, Italy); its use in the present animal model is supported by the recent finding that human relaxin can bind to RXFP1 relaxin receptor of various species, including the guinea pig [38]. Acetylcholine was from Sigma Chemical (St. Louis, MO, USA), phenylephrine from Tocris Bioscience (Bristol, UK). Kentucky Reference cigarettes 3R4F, each containing 11 mg of total particulate matter, 9.4 mg of tar and 0.73 mg of nicotine, were obtained from the Kentucky Tobacco Research Council (Lexington, KY, USA). Unless otherwise specified, the other reagents used for the experiments were from Sigma-Aldrich (Milan, Italy).

In vivo study
Exposure of guinea pigs to CS Male Hartley albino guinea pigs weighing 300-350 g were used for the experiments (Harlan, Correzzana, Italy). Animal handling and use complied with the European Community guidelines for animal care (2010/63/ EU) and were approved by the Committee for Animal Care and Experimental Use of the University of Florence. The animals were housed on a 12 hrs light/dark cycle at 22°C room temperature and had free access to food and water. The experiments were designed to minimize pain and the number of animals used. Sacrifice was carried out by decapitation.
The animals were divided into the following experimental groups (n = 6/group): Group 1: Control untreated animals; Group 2: Animals exposed daily to CS for 8 weeks; Group 3: Animals exposed daily to CS for 8 weeks and treated with RLX given by continuous subcutaneous (s.c.) infusion using osmotic minipumps (Alzet; DURECT Corporation, Cupertino, CA, USA). The pumps were implanted 1 day before starting the exposure to CS on the back upon anaesthesia (i.p. injection of ketamine hydrochloride, 100 mg/kg b.w. and xylazine, 15 mg/kg b.w.) and filled to deliver a daily dose of 1 lg for the whole duration of CS exposure; Group 4: Animals exposed to CS and treated with RLX given by minipumps as above, but delivering a daily RLX dose of 10 lg for the whole duration of CS exposure.
The animals were subjected to CS exposure in a smoke chamber, according to Das et al. [39] with minor modifications. The smoke chamber (2.5 l) was similar to a vacuum desiccator equipped with an open tube for cigarette positioning at one end and a vacuum-connected tube and stopcock at the opposite end. To each group of CS-exposed animals, five 3R4F reference cigarettes were administered daily. Each cigarette was fitted on the inlet tube and lit; then, a puff of CS was introduced into the chamber containing the animals by applying a mild suction of 4 cm water for 20 sec. The guinea pigs were exposed to the accumulated smoke for further 40 sec., for a total duration of CS exposure of 60 sec. After a pause of 60 sec. during which the chamber was opened and ventilated with fresh air, a second puff was administered with the same procedure. The gap between each of the 5 cigarettes/day was 1 hr. At the end of the treatment, the animals were anaesthetized by i.p. injection of ketamine hydrochloride (100 mg/kg b.w.) and xylazine (15 mg/kg b.w.), blood and tissue samples were collected and processed for functional, morphological and biochemical analyses.

Determination of serum RLX levels
The circulating RLX levels were determined in guinea pig serum by ELISA (R&D Systems, Minneapolis, MN, USA) according to the manufacturer's instructions.

Functional studies
Segments of thoracic aortas were cut into helical strips (3 mm width, 20 mm length) and connected to a force displacement transducer (Grass FT03, Grass Technologies, Middleton, WI, USA) coupled to a polygraph (Grass 7K, Grass Technologies) for continuous recording of isometric tension. Strips were mounted in 5-ml organ baths containing Krebs-Henseleit solution, gassed with 95% O 2 -5% CO 2 mixture, of the following composition (mM): NaCl 118, KCl 4.7, MgSO 4 1.2, KH 2 PO 4 1.2, NaHCO 3 25, CaCl 2 2.5 and glucose 10 (pH 7.4) and kept at 37°C. Preparations were allowed to equilibrate for 1 hr under an initial load of 2.5 g, as previously reported [40]. In each experimental group, the contractile responses to phenylephrine (2 9 10 À7 M) were evoked. When contraction reached a stable plateau phase, acetylcholine (Ach, 2 9 10 À6 M) was added. The chosen concentrations are similar to those previously used for similar purposes [40,41]. The interval between two subsequent applications of phenylephrine was ≥15 min., during which repeated washes with Krebs-Henseleit solution were performed. glutaraldehyde and 1% osmium tetroxide and embedded in Epon 812. Ultrathin sections were stained with uranyl acetate and alkaline bismuth subnitrate and examined under a JEM 1010 electron microscope (Jeol, Tokyo, Japan) at 80 kV.

Detection of free carbon monoxide in plasma
Free carbon monoxide (CO) was measured in the plasma of the animals of the different experimental groups as an index of the degree of exposure to CS. The amount of free CO in plasma was measured with a gaseous CO detector (RGA3, Reduction Gas Analyzer; SAES Getters, Milan, Italy) as described [42]. Measurements were obtained by comparison with a CO standard curve prepared immediately before analysis and expressed as parts per million.

In vitro study Cigarette smoke extracts preparation
Cigarette smoke extracts solution was prepared by bubbling smoke from two 3R4F cigarettes, 30 sec. each, in 50 ml of PBS according to Niu et al. [43], with minor modifications. The resulting solution, assumed as 100% CSE, was adjusted to pH 7.4, and filtered through a 0.2 lm filter (Millipore Co.). Cigarette smoke extracts solution was freshly prepared on the day of the experiment and immediately used.
To establish the effective dose of CSE, preliminary concentrationand time-dependent studies were performed (data not shown). Guinea pig aortic endothelial cells were exposed to 1-15% CSE for 4-12 hrs and cell viability was evaluated by trypan blue. Accordingly, 10% CSE solution (vol/vol) for 4 hrs was used in all the experiments because it gave a well-appreciable toxic effect without causing massive cell death.

Isolation and culture of guinea pig aortic endothelial cells
Male Hartley albino guinea pigs (Harlan), 2-to 3-month old, were anaesthetized by i.p. injection of ketamine hydrochloride (100 mg/kg b.w.) and xylazine (15 mg/kg b.w.) and killed by decapitation. The thoracic aorta was quickly removed and guinea pig aortic endothelial (GPAE) cells were isolated based on a previously described method [44].
Characterization of the isolated GPAE cells was performed by immunocytochemistry and cytochemistry, as previously reported [27]. The percentage of immunoreactive cells for endothelial markers, such as vimentin and von Willebrand factor ranged between 96% and 98% (data not shown). For all experiments, cells were used at the first and second passage.
Guinea pig aortic endothelial cells were pre-treated or not with RLX 100 ng/ml (17 nmol/l) for 24 hrs and then exposed to CSE for 4 hrs. This RLX dose was chosen because it was shown to induce significant cell protection from oxidative stress [45]. Control cultures not exposed to CSE were also prepared.

Total RNA extraction and RT-PCR
Expression levels of mRNA for RXFP1, eNOS and b-actin were assayed by RT-PCR in GPAE cells. One microgram of total RNA, extracted with TRIzol Reagent (Invitrogen), was reverse transcribed and amplified with SuperScript One-Step RT-PCR System (Invitrogen). After cDNA synthesis for 30 min. at 55°C, the samples were pre-denatured for 2 min at 94°C and then subjected to 35 cycles of PCR performed at 94°C for 15 sec., alternating with 55°C for 30 sec.

Trypan blue viability assay
The trypan blue exclusion method was used to assess cell viability. Guinea pig aortic endothelial cells (5 9 10 4 /well) were seeded in 24-well plates. At the end of the treatments, the cells were gently detached by trypsin/ethylenediaminetetraacetic acid (EDTA), resuspended in culture medium and mixed 1:1 with 0.4% trypan blue solution. The final cell suspensions were counted under a phase contrast inverted microscope using a Burker chamber. Viable cells were expressed as percentage of the total counted cells.

Mitochondrial membrane potential
Mitochondrial membrane potential (Dw) was assessed as previously reported [46] using tetramethylrhodamine methyl ester perchlorate (TMRM), a lipophilic potentiometric fluorescent dye that distributes between mitochondria and cytosol in proportion to Dw by virtue of its positive charge. Thus, the fluorescence intensity depends on dye accumulation in mitochondria, which is directly related to Dw. For confocal microscope analysis, GPAE cells were cultured on glass coverslips and loaded for 20 min. at 37°C with TMRM, dissolved in 0.

Assessment of Caspase-3 activity
Guinea pig aortic endothelial cells seeded on glass coverslips were incubated with FAM-FLICA TM Caspase 3&7 assay kit (Immunochemistry Technologies, Bloomington, MN, USA) for 30 min. as previously reported [47] and then fixed in 2% paraformaldehyde for 10 min. at room temperature. Fluorescence was detected by a confocal Leica TCS SP5 confocal laser scanning microscope with 488-nm excitation wavelength and 963 oil immersion objective. Caspase-3 activity was also quantified by flow cytometry: single-cell suspensions were incubated with FAM-FLICA TM for 30 min. at 37°C and then analysed using a FACS-Canto flow cytometer (Becton-Dickinson).

Determination of intracellular ROS
Guinea pig aortic endothelial cells seeded on glass coverslips were loaded with the ROS-sensitive fluorescent probe 2 0 ,7 0 -dichlorodihydrofluorescein diacetate (H 2 DCFDA; Invitrogen, 2.5 lmol/l) -dissolved in 0.1% DMSO and Pluronic acid F-127 (0.01% w/v) -added to cell culture media for 15 min. at 37°C [48]. The cells were fixed in 2% paraformaldehyde for 10 min. at room temperature and the H 2 DCFDA fluorescence analysed using a Leica TCS SP5 confocal laser scanning microscope with 488-nm excitation wavelength and 963 oil immersion objective. Intracellular ROS levels were also monitored by flow cytometry: single-cell suspensions were incubated with H 2 DCFDA (1 lmol/l) for 15 min. at 37°C and then analysed using a FACSCanto flow cytometer (Becton-Dickinson).

Evaluation of lipid peroxidation
Lipid peroxidation was investigated by confocal scanning microscopy using BODIPY 581/591 C11 (Life Technologies, Monza, Italy), a lipophilic fluorescent probe that mimics the properties of natural lipids [49]. BODIPY acts as a fluorescent lipid peroxidation reporter as it shifts its fluorescence from red to green in the presence of oxidizing agents. Guinea pig aortic endothelial cells cultured on glass coverslips were loaded with BODIPY, dissolved in 0.1% DMSO (2 lM final concentration in the culture medium), for 30 min. at 37°C. The cells were fixed in 2% paraformaldehyde for 10 min. at room temperature and analysed using a Leica TCS SP5 confocal laser scanning microscope with 581 nm excitation wavelength and 963 oil immersion objective. Lipid peroxidation was also quantified by flow cytometry: single-cell suspensions were incubated, in the dark, for 30 min. at 37°C with BODIPY (1 lM) in cell culture medium and then analysed using a FACSCanto flow cytometer (Becton-Dickinson).
All the flow cytometry data were analysed using FACSDiva software (Becton-Dickinson).

Western blotting
After treatments, GPAE cells were lysed in cold buffer (10 mM Tris/HCl pH 7.4, 10 mM NaCl, 1.5 mM, MgCl 2 , 2 mM Na 2 EDTA, 1% Triton X-100), added with 109 Sigmafast Protease Inhibitor cocktail tablets. Total protein content was measured spectrophotometrically using micro-BCA TM Protein Assay Kit (Thermo Fisher-Pierce, Waltham, MA, USA). Forty microgram of total proteins were electrophoresed by SDS-PAGE and blotted onto nitrocellulose membranes (Amersham, Cologno Monzese, Italy). The membranes were incubated overnight at 4°C with rabbit polyclonal anti-eNOS (1:1000; Abcam) and rabbit polyclonal anti-b-actin antibodies (1:20,000; Sigma-Aldrich), assuming b-actin as control invariant protein. Specific bands were detected using rabbit peroxidase-labelled Fig. 1 Ex vivo contractility of aortic strips from guinea pigs of the different experimental groups. (A) Typical tracings in response to phenylephrine (2 9 10 À7 M) and acetylcholine (Ach, 2 9 10 À6 M). In the CS-exposed animals, the amplitude and slope of both responses are reduced as compared with the untreated controls. RLX reverted the tracing amplitude and slope to a pattern similar to that of the controls. (B) Mean amplitude of contractile and relaxant responses to phenylephrine and Ach, respectively, in the different experimental groups. Amplitudes of the contractile responses to phenylephrine and the relaxant responses to Ach are expressed as per cent increase and decrease, respectively, of the maximal contraction evoked by 2 9 10 À7 M phenylephrine, assumed as 100%. Values are means AE S.E.M. (one-way ANOVA test), n = 6. **P < 0.01, ***P < 0.001 versus controls; † P < 0.05 and † † P < 0.01 versus CS-exposed animals.

Data analysis and statistical tests
The reported data are expressed as the mean AE S.E.M. of six animals per group or at least three independent cell culture experiments. Statistical comparison of differences between groups was carried out using one-way ANOVA followed by Student-Newman-Keuls multiple comparison test. A P-value ≤0.05 was considered significant. GraphPad Prism 2.0 statistical program (GraphPad Software, San Diego, CA, USA) was used for statistical analysis.

Plasma CO levels
The animals of the different experimental groups subjected to chronic CS had significantly elevated CO levels (CS-exposed: 37.9 AE 3.2; CS+RLX 1 lg/day: 32.1 AE 3.1; CS+RLX 10 lg/day: 37.5 AE 5.8) compared with the controls (6.4 AE 0.8: P < 0.001 versus the other groups). No significant differences were detected among the CSexposed groups, suggesting that all the animals were subjected to the same level of CS-induced toxicity.

Plasma RLX levels
The circulating levels of RLX evaluated at the end of the experiment were 308 AE 44 pg/ml and 2.5 AE 0.6 ng/ml upon 1 and 10 lg daily doses respectively. The values measured in the untreated controls and the CS-exposed animals were consistently below the detection threshold.

Ex vivo aortic contractility
In aortic strips from the untreated control animals, addition of phenylephrine to the bath medium caused a rapidly arising contrac-  tion that reached a plateau phase (mean amplitude 0.20 AE 0.05 g) (Fig. 1). In phenylephrine-precontracted preparations, Ach caused a fast relaxation (Fig. 1) that persisted until washout. In aortic strips from the CS-exposed guinea pigs, both the contractile and the relaxant responses to phenylephrine and Ach were significantly reduced in amplitude and delayed (Fig. 1). In strips from the CS-exposed animals treated with RLX at both doses, the amplitude of responses to both phenylephrine and Ach was significantly enhanced as compared with the CS-exposed animals (Fig. 1). No significant differences were observed between 1 and 10 lg/day RLX.

Morphology
Histological examination of transverse sections of aorta ( Fig. 2A) showed that the control guinea pigs had thin wall, showing an intact structure with continuous endothelial lining and alternate layers of elastic lamellae and smooth muscle cells in the tunica media. Cigarette smoke exposure caused extended loss of endothelial cells, significant wall thickening, disarrangement of smooth muscle cells in the tunica media, some of which showed cytoplasmic vacuolation, and accumulation of amorphous material. Conversely, treatment with RLX at both doses during CS exposure resulted in an almost complete disappearance of the noted aortic wall abnormalities. The RLX-induced reduction of aortic wall thickening caused by CS exposure was confirmed by the morphometric analysis (Fig. 2B). Ultrastructural examination (Fig. 3) revealed that, compared with the controls, CS exposure caused diffuse endothelial cell vacuolation and signs of cell demise, such as plasma membrane rupture and detachment from the other cells and the basement membrane; RLX at both doses prevented these degenerative changes.

Fig. 4 (A) RT-PCR showing the expression of the RLX receptor RXFP1 by GPAE cells. (B) Trypan blue viability test performed on GPAE cells:
compared with the controls, exposure to CSE causes a significant decrease in cell viability, while pre-treatment with RLX significantly reduced this effect. RLX per se has no effect on cell viability. Values are means AE S.E.M. (one-way ANOVA test), n = 6. ***P < 0.001 versus controls; † † P < 0.01 versus CSE.

Endothelial cell viability
To investigate the mechanisms of the protection afforded by RLX on the aortic endothelium upon exposure to CS, we performed a series of in vitro experiments on primary GPAE cells. These cells were first shown to express the RLX receptor RXFP1 (Fig. 4A). When exposed to CSE for 4 hrs, GPAE cells showed a marked viability decrease, as assayed with the trypan blue exclusion test. Pre-treatment for 24 hrs with 100 ng/ml RLX significant reduced this adverse effect of CSE (Fig. 4B). Reduced cell viability by CSE was correlated with the induction of apoptosis. Exposure to CSE induced a prominent decrease in mitochondrial membrane polarization, an early apoptotic event, and increase in caspase-3 activity, a marker of ongoing apoptosis, in comparison with the untreated controls ( Fig. 5A and B). Incubation with RLX significantly reduced CSE-induced apoptosis (Fig. 5A and  B). Addition of RLX to untreated control cultures had no effect on cell viability and apoptosis.

Endothelial oxidative injury
Because oxidative injury is a typical detrimental effect of CS (5-10), we then investigated whether the observed endothelial damage was related to oxidative stress. In fact, the serum levels of 8-OHdG, a marker of DNA oxidation, were significantly increased in the guinea pigs exposed to CS in comparison with the untreated controls, while administration of RLX at both doses significantly reduced serum 8-OHdG (Fig. 6A). In keeping with these results, 4-hr exposure of GPAE cells to CSE caused a robust increase in intracellular ROS (Fig. 6B), resulting in significantly higher membrane lipid peroxidation (Fig. 6C), as compared with the untreated controls. The treatment with RLX significantly reduced both these CSE-induced oxidative stress markers ( Fig. 6B and C). Addition of RLX to untreated control cultures had no effect.

Endothelial NOS expression
Cigarette smoke-induced endothelial injury and dysfunction has been associated with impaired eNOS expression [14][15][16]. Accordingly, the current in vivo experiment has shown that CS causes a well-appreciable decrease in eNOS immunoreactivity in the residual endothelial cells of the aortic wall as compared with the untreated controls. This eNOS impairment was reduced in the animals treated with RLX at both doses (Fig. 7A). The supportive effect of RLX on eNOS expression was confirmed in vitro: in fact, eNOS mRNA (Fig. 7B) and protein expression (Fig. 7C) in GPAE cells were markedly reduced in the CSE-treated cultures, while incubation with RLX significantly reduced the CSE-induced eNOS reduction. Addition of RLX to untreated control cultures had no appreciable effect.

Discussion
The results of the present study show that RLX protects against aortic dysfunction and endothelial impairment induced by CS. In fact, longterm administration of RLX to guinea pigs exposed chronically to CS preserved the responsiveness of isolated aortic strips to phenylephrine and Ach, which was compromised by CS. Namely, the decreased amplitude of the vasodilatory responses to Ach accounts for impaired endothelial function, as also reported in previous experimental studies on animals exposed to chronic smoking [12]. On the other hand, the impairment in contractile response to phenylephrine observed in the preparations from CS-exposed animals is likely related to adverse arterial wall remodelling, a typical effect of CSinduced vascular injury [4,20]. The improvement of the contractile response to phenylephrine induced by RLX treatment fits well with the present observation of preserved histology of the aortic wall and is consistent with the report that RLX therapy prevents and even reverses large artery remodelling and fibrosis in a rat model of chronic hypertension [50]. In our experimental model, CS exposure and RLX treatment were concurrent, thus the reported findings support the notion that RLX prevents CS-induced arterial injury and dysfunction but do not provide information to understand whether RLX treatment could revert established vascular damage because of CS.
The observed preservation of the vasodilatory property of aortas in the CS-exposed guinea pigs which received RLX can be chiefly related to the maintenance of the endothelial lining and of adequate levels of eNOS in endothelial cells, as can be argued from the present in vivo and in vitro findings. This property of RLX fits well with previous reports that this hormone can up-regulate NOS expression in vascular endothelial cells [26][27][28][29][30]. However, RLX may also promote arterial dilatation by a direct effect on vascular smooth muscle cells, as reported [29].
Our study indicates that the dysfunctional vasodilatory response of the aortas in CS-exposed guinea pigs is related to endothelial impairment because of oxidative stress, as suggested by the rise in serum 8-OHdG. Indeed, in these animals, the aortic endothelium was lacking in large tracts of the intima and, when present, showed ultrastructural signs of cell injury and a remarkable reduction of eNOS expression. Consistently, exposure of GPAE cells to CSE caused a significant reduction of cell viability and induction of apoptosis, which were closely related to increase in intracellular ROS and lipid peroxidation. The present findings first show that RLX, at physiological circulating levels or nanomolar concentrations in the culture medium,  The cellular mechanisms accounting for this antioxidant effect remain to be elucidated. Preliminary findings from our laboratory obtained with the oxygen radical absorbance capacity assay indicate that RLX has no intrinsic antioxidant properties. On the other hand, RLX administration to GPAE cells markedly increased their endogenous antioxidant capacity: in the presence of CSE, RLX restores it to levels near to the untreated controls (Supporting Information). It can be speculated that RLX may up-regulate one or more cellular defence mechanisms against oxidative stress through activation of PI3 and MAP kinases, typical downstream pathways of RXFP1 activation in blood vessels [30,51]. These signalling pathways are in fact involved in the cell response to ROS, being able to induce the phosphorylation of the transcription factor nuclear factor-E2-related factor 2 (Nrf2), which binds the antioxidant response element within the promoters of genes encoding antioxidant and detoxifying enzymes [52]. This hypothesis for the mechanism of action of RLX on cell resistance to oxidative stress deserves further investigation.
A limitation of this study is that we did not explore in vivo physiological parameters, such as blood pressure or pulse pressure, useful to further assess RLX's therapeutic effects. However, the current observation that RLX is capable of reducing CS-mediated vascular damage and dysfunction may be relevant to human health. In fact, it is known that postmenopausal women who are smokers have double the risk of CVD, a discrepancy which has been related to defective ovarian hormones [53]. However, the protection afforded by hormone replacement therapy (HRT) with oestrogen/progestin is controversial, as some clinical trials failed to confirm that HRT can be beneficial. Two randomized prospective primary or secondary prevention trials showed that HRT may actually increase the risk and events of CVD in post-menopausal women [54]. The failure of HRT in lowering the cardiovascular risk of smoking in post-menopausal women can be explained considering that cessation of ovarian function at menopause implies not only a drop in the secretion of oestrogen and progesterone but also of H2 relaxin, which is mainly secreted by the corpus luteum [55] and has been credited of major protective activities on the cardiovascular system [23,29,56]. On this ground, RLX could be viewed as a new therapeutic tool to prevent or reduce the cardiovascular risk in smokers: this hypothesis is strengthened by the successful use of RLX in the treatment of patients with heart failure demonstrated in a recent clinical trial, which also revealed that the use of RLX in humans is safe and free of adverse side effects [24,57].

Conflict of interest
The authors declare that they have no conflict of interest.

Supporting information
Additional Supporting Information may be found in the online version of this article: Data S1 Oxygen radical absorbance capacity (ORAC) assay. and protein (C) in GPAE cells evaluated by RT-PCR and Western blotting, respectively. Compared with the controls, eNOS is reduced by CSE and maintained by RLX. RLX per se has no appreciable effects.