Celecoxib, but not indomethacin, ameliorates the hypertensive and perivascular fibrotic actions of cyclosporine in rats: Role of endothelin signaling

Celecoxib, but not indomethacin, ameliorates the hypertensive and perivascular fibrotic actions of cyclosporine in rats: Role of endothelin signaling

Toxicology and Applied Pharmacology 284 (2015) 1–7 Contents lists available at ScienceDirect Toxicology and Applied Pharmacology journal homepage: w...

1MB Sizes 0 Downloads 13 Views

Toxicology and Applied Pharmacology 284 (2015) 1–7

Contents lists available at ScienceDirect

Toxicology and Applied Pharmacology journal homepage: www.elsevier.com/locate/ytaap

Celecoxib, but not indomethacin, ameliorates the hypertensive and perivascular fibrotic actions of cyclosporine in rats: Role of endothelin signaling Mahmoud M. El-Mas a,⁎, Maged W. Helmy b, Rabab M. Ali a, Hanan M. El-Gowelli a a b

Pharmacology and Toxicology, Faculty of Pharmacy, Alexandria University, Egypt Pharmacology and Toxicology, Faculty of Pharmacy, Damanhour University, Egypt

a r t i c l e

i n f o

Article history: Received 28 November 2014 Revised 29 December 2014 Accepted 22 January 2015 Available online 3 February 2015 Keywords: Cyclosporine Celecoxib Indomethacin Endothelin receptors Hypertension Perivascular fibrosis

a b s t r a c t The immunosuppressant drug cyclosporine (CSA) is used with nonsteroidal antiinflammatory drugs (NSAIDs) in arthritic conditions. In this study, we investigated whether NSAIDs modify the deleterious hypertensive action of CSA and the role of endothelin (ET) receptors in this interaction. Pharmacologic, protein expression, and histopathologic studies were performed in rats to investigate the roles of endothelin receptors (ETA/ETB) in the hemodynamic interaction between CSA and two NSAIDs, indomethacin and celecoxib. Tail-cuff plethysmography measurements showed that CSA (20 mg kg−1 day−1, 10 days) increased systolic blood pressure (SBP) and heart rate (HR). CSA hypertension was associated with renal perivascular fibrosis and divergent changes in immunohistochemical signals of renal arteriolar ETA (increases) and ETB (decreases) receptors. While these effects of CSA were preserved in rats treated concomitantly with indomethacin (5 mg kg−1 day−1), celecoxib (10 mg kg−1 day−1) abolished the pressor, tachycardic, and fibrotic effects of CSA and normalized the altered renal ETA/ETB receptor expressions. Selective blockade of ETA receptors by atrasentan (5 mg kg−1 day−1) abolished the pressor response elicited by CSA or CSA plus indomethacin. Alternatively, BQ788 (ETB receptor blocker, 0.1 mg kg−1 day−1) caused celecoxib-sensitive elevations in SBP and potentiated the pressor response evoked by CSA. Together, the improved renovascular fibrotic and endothelin receptor profile (ETA downregulation and ETB upregulation) mediate, at least partly, the protective effect of celecoxib against the hypertensive effect of CSA. Clinically, the use of celecoxib along with CSA in the management of arthritic conditions might provide hypertension-free regimen. © 2015 Elsevier Inc. All rights reserved.

Introduction The clinical use of cyclosporine is often associated with serious cardiovascular disorders such as hypertension (Cauduro et al., 2005; Riva et al., 2013). The mechanisms of the hypertensive effect of CSA include sympathoexcitation (Zhang and Victor, 2000), enhanced renin– angiotensin activity (Nishiyama et al., 2003), arterial baroreflex impairment (El-Mas et al., 2002, 2012b), oxidative stress (El-Mas et al., 2012b), vascular endothelium dysfunction (El-Mas et al., 2003, 2004), and interruption of brainstem nitric oxide synthase/heme oxygenase pathway and downstream guanylate cyclase activity (El-Mas et al., 2012c). Moreover, experimental (Nasser et al., 2014) and clinical (Cauduro et al., 2005) studies suggest the involvement of endothelin Abbreviations: CSA, cyclosporine; NSAIDs, nonsteroidal antiinflammatory drugs; ET, endothelin; SBP, systolic blood pressure; HR, heart rate; COX, cyclooxygenase; TGF-β1, transforming growth factor-beta1. ⁎ Corresponding author at: Department of Pharmacology and Toxicology, Faculty of Pharmacy, Alexandria University, Alazarita, 21521 Alexandria, Egypt. Fax: + 20 3 487 3273. E-mail address: [email protected] (M.M. El-Mas).

http://dx.doi.org/10.1016/j.taap.2015.01.018 0041-008X/© 2015 Elsevier Inc. All rights reserved.

(ET) in the pathogenesis of CSA hypertension. Nonselective blockade of ET receptors with bosentan (Bartholomeusz et al., 1996) or selective blockade of endothelin ETA receptors with BQ-123 (Phillips et al., 1994) abrogates the hypertensive response elicited by CSA. CSA might be used with nonsteroidal antiinflammatory drugs (NSAIDs) in the management of several arthritic conditions including rheumatoid arthritis, chronic polyarthritis, juvenile systemic lupus erythematosus and psoriatic arthritis (Cavalcante et al., 2011; Ash et al., 2012). Like CSA, NSAIDs are believed to adversely affect renal and circulatory functions, which may be followed by elevation in blood pressure (BP) or worsening of pre-existing hypertension (Tavares et al., 2002). However, little information is available regarding the safety profile of the combined CSA/NSAID regimen. For example, the concomitant use of CSA along with naproxen or sulindac elicits additive impairment of renal function (Altman et al., 1992). Diclofenac does not affect blood levels of CSA, but the diclofenac level is nearly doubled and reversible decreases in renal function occasionally occur (Olyaei et al., 1999). The use of tacrolimus, another immunosuppressant with calcineurin inhibitory activity, and nonselective or cyclooxygenase-2 (COX-2) selective NSAIDs in salt-depleted rats impairs the glomerular filtration rate and

2

M.M. El-Mas et al. / Toxicology and Applied Pharmacology 284 (2015) 1–7

reduces tacrolimus blood levels (Soubhia et al., 2005). By contrast, in a recent study we established evidence of a renoprotective effect for celecoxib, selective COX-2 inhibitor, against functional, inflammatory, and structural manifestations of CSA nephrotoxicity (El-Gowelli et al., 2014). The abovementioned studies focused mainly on the renal interaction between CSA and NSAIDs and little or no information is available regarding whether the co-exposure to these two pharmacologic modalities can negatively impact the hemodynamic functions. This issue was addressed in the present study, which investigated the individual and combined hemodynamic effects of chronic CSA, indomethacin (COX-1/ COX-2 inhibitor), and celecoxib (selective COX-2 inhibitor) in rats. The study was then extended to test the hypothesis that the CSA/NSAID interaction is modulated by ET signaling. Two experimental approaches were utilized: (i) the investigation of the effect of selective blockade of ETA (atrasentan) or ETB (BQ788) receptors on the CSA-NSAIDs hemodynamic and fibrotic interactions, and (ii) the measurement of the immunohistochemical expression of ETA and ETB receptors in renal arterioles. The results showed that celecoxib, but not indomethacin, abrogated the hypertensive effect of CSA and underlying renovascular fibrosis and alterations in ETA/ETB receptor expressions. Materials and methods Animals. Male Sprague–Dawley rats (Faculty of Pharmacy, Pharos University, Alexandria, Egypt) weighing 190 to 220 g were used. All experiments were performed in strict accordance with institutional animal care and use guidelines. Materials. Atrasentan (Abbott Laboratories, Illinois, USA), BQ788 [N-[N[N-[(2,6-dimethyl-1-piperidinyl)carbonyl]-4-methyl-L-leucyl]-1(methoxycarbonyl)-D-tryptophyl]-D-norleucine sodium salt] (Peptides International, USA), CSA (Novartis Pharma, AG, Basel, Switzerland), celecoxib (European Egyptian Pharmaceutical Industries, Egypt), cremophore EL [polyoxyethylene castor oil] (Sigma-Aldrich, St. Louis, MO, USA), indomethacin (European Egyptian Pharmaceutical Industries, Egypt) and thiopental sodium (Biochemie GmbH, Vienna, Austria) were purchased from commercial vendors. Cremophore (the vehicle for CSA) was mixed with saline to a final dilution of 40%. CSA was freshly dissolved in 40% cremophore. Indomethacin, celecoxib, BQ788, atrasentan, and thiopental sodium were dissolved in saline. Tail cuff SBP measurements. SBP of rats was measured before, and 5 and 10 days after administration of drugs using a computerized data acquisition system with LabChart-7 pro software (Power Lab 4/30, model ML866/P, AD Instruments, Bella Vista, Australia) as in our previous studies (El-Mas et al., 2012a). Heart rate (HR) was computed from BP waveforms and displayed on another channel of the recording system. A specialized tail cuff and pulse transducer (Pan Lab, Spain) were employed for SBP measurement based on the periodic occlusion of tail blood flow. SBP was measured 3 or 4 times and the values were averaged to get the mean. Histopathology. Masson's trichrome staining was used for detection of periarterial renal fibrosis (Drury and Wallington, 1980; El-Gowelli et al., 2014). Immunostaining. The technique described in previous studies including ours (Chen and Sun, 2006; El-Mas et al., 2006; El-Gowelli et al., 2014) was employed for immunohistochemical determination of the protein expression of ETA and ETB receptors. Kidney sections (5 μm) were placed on positively charged adhesion microscope slides (Thermo Scientific®, Germany), deparaffinized in xylene and rehydrated in a series of decreasing ethanol concentration (100%, 95%, and 70%). Slides were rinsed gently with phosphate-buffered saline (PBS) and drained. The antigenic determinants in the cells were unblocked by incubating the sections at

(95–98 °C) for 20 min in citrate buffer (pH 6, Thermo Scientific® Germany) for heat-induced epitope retrieval. The sections were then rinsed with 1 × TBST (50 mM Tris/HCl, pH 7.4, 150 mM NaCl, 0.1% Tween 20, Thermo Scientific® Germany). Endogenous peroxidases were blocked by adding 3% hydrogen peroxide and then washed with 1× TBST. A universal protein block was applied for 20 min. The appropriate primary monoclonal antibodies (Bioss® USA) were diluted as instructed by the manufacturer and applied to the slides for 45 min at 37 °C. Negative controls were processed without applying the primary antibodies. The slides were then washed with 1 × TBST, rinsed, and incubated for 30 min with the secondary antibody (polyvalent HRP detection kit, Spring Bioscience® Pleasanton, CA). The chromogen 3,3′-diaminobenzidine was prepared and applied as instructed by the manufacturer for protein visualization. Each slide was counterstained with hematoxylin and dipped in ascending concentrations of alcohol and then xylene. The immunohistochemical signals of ETA and ETB receptors were quantified by Image J software (version 1.45s) together with computer-assisted microscopy, which was employed for this purpose using the greyscale thresholding as previously described (Kaczmarek et al., 2004; El-Gowelli et al., 2014). Briefly, the captured colored images were transformed into grey scale by changing image type from colored type into bit-8 type. The area of immunohistochemical reaction to be measured was selected using oval selection tools. Then, thresholding of binary process converted foreground pixels into black color, while background pixels into white color. The developed image was analyzed by obtaining a histogram list of numbers of pixels of black and white areas. The degree of reaction positivity was estimated by the percentage of black pixels in the binary image compared to the total number of pixels of selected area. Protocols and experimental groups CSA–indomethacin hemodynamic interaction A total of 8 groups of rats (n = 6 each) were employed to determine the hemodynamic interaction between CSA and indomethacin and its modulation by endothelin ETA receptors. Rats were randomly assigned to receive one of the following treatments: (i) cremophor (40%, 1 mg kg−1 day−1), (ii) CSA (20 mg kg−1 day−1) (El-Mas et al., 2003, 2011), (iii) indomethacin (5 mg kg− 1 day−1) (Kamalutheen et al., 2009), (iv) CSA + indomethacin, (v) atrasentan (ETA receptor blocker, 5 mg kg− 1 day− 1) (Lima, 2010), (vi) CSA + atrasentan, (vii) indomethacin + atrasentan, or (viii) CSA + indomethacin + atrasentan. All drugs were administered once daily via oral gavage for 10 consecutive days. For rat groups receiving the combined regimens (2 or more drugs), drugs were administered separately and consecutively on daily basis. Tail cuff measurements of SBP were performed as described earlier before, and 5 and 10 days after administration of drugs. On day 10, rats were euthanized with an overdose of thiopental (100 mg kg−1), the abdomen was opened, and the left kidney was quickly removed, fixed in 10% formaldehyde solution, and embedded in a paraffin blocks within 24 h. Kidneys were used for the histopathological identification of renal periarterial fibrosis by Masson's trichrome staining as well as for immunohistochemical measurement of the protein expression of endothelin ETA/B receptors. CSA–celecoxib hemodynamic interaction Two groups of rats (n = 6 each) were used and allocated to receive celecoxib (10 mg kg− 1 day− 1) (El-Gowelli et al., 2014) or CSA + celecoxib for 10 days. Because preliminary findings showed that CSA hypertension was abolished after concurrent administration of celecoxib, evidence was sought to implicate endothelin ETB receptors in the celecoxib effect. The effects of selective ETB blockade by BQ788 on the

M.M. El-Mas et al. / Toxicology and Applied Pharmacology 284 (2015) 1–7

3

hemodynamic responses elicited by celecoxib, CSA, or their combination were evaluated. Four more groups of rats were employed that received BQ788 (0.1 mg kg−1 day−1 for 10 days) (El-Gowelli et al., 2014), CSA + BQ788, celecoxib + BQ788, or CSA + celecoxib + BQ788. All drugs were administered by oral gavage except BQ788 that was given intraperitoneally. Tail cuff SBP measurements and histopathological (renal periarterial fibrosis) and immunohistochemical staining (ETA and ETB) were performed as described above.

Statistical analysis Data are expressed as means ± SEM. Normal distribution was checked using column statistics (modified Kolmogorov–Smirnov test, GraphPad Prism, software release 3.02). Multiple comparisons were analyzed by one way analysis of variance (ANOVA) followed by the Bonferroni post-hoc test. The analysis was performed using GraphPad Prism, software release 3.02. Probability levels less than 0.05 were considered significant.

Results Hemodynamic CSA–NSAID interaction Tail cuff measurements showed that baseline SBP and HR values in all rat groups measured prior to drug administration (Day 0) were not statistically different (Table 1). CSA (20 mg kg−1 day−1 for 10 days) caused significant increases in SBP and HR on the 5th and the 10th day of treatment compared to the vehicle-treated values (Fig. 1). While indomethacin (5 mg kg−1 day−1 for 10 days) caused no hemodynamic changes, the combined CSA/indomethacin treatment increased SBP and HR to levels similar to those caused by CSA alone (Fig. 1). The selective blockade of ETA receptors with atrasentan (5 mg kg− 1 day−1, 10 days) abolished the rise in SBP caused by CSA or the CSA/indomethacin regimen while it had no effect on the associated tachycardia (Fig. 1). Unlike the indomethacin effect, the simultaneous administration of celecoxib (10 mg kg−1 day−1 for 10 days) blunted both the hypertensive and tachycardic effects of CSA (Fig. 2). The blockade of ETB receptors with BQ788 (0.1 mg kg−1 day−1, 10 days) increased SBP and this effect was significantly potentiated in rats receiving BQ788 plus CSA (Fig. 2). The hypertensive responses elicited by BQ788 or by the CSA/BQ788 regimen were significantly reduced upon concurrent administration of celecoxib (Fig. 2).

Table 1 Baseline values of systolic blood pressure (SBP, mm Hg) and heart rate (HR, beats min−1) prior to drug administration (Day 0). Group

SBP

HR

Vehicle CSA Indo CSA + Indo Atrasentan CSA + Atrasentan Indo + Atrasentan CSA + Indo + Atrasentan Celecoxib CSA + Celecoxib BQ788 CSA + BQ788 Celecoxib + BQ788 CSA + Celecoxib + BQ788

97 ± 4 95 ± 2 105 ± 3 96 ± 5 97 ± 3 99 ± 5 91 ± 2 101 ± 4 98 ± 3 101 ± 3 93 ± 4 92 ± 3 94 ± 7 95 ± 4

330 ± 8 340 ± 19 338 ± 5 327 ± 10 348 ± 7 316 ± 5 320 ± 8 310 ± 8 339 ± 9 365 ± 14 345 ± 20 339 ± 20 342 ± 11 351 ± 17

Values are means ± S.E.M. of 6 observations.

Fig. 1. Effects of 10-day treatment with CSA (20 mg kg−1 day−1), indomethacin (5 mg kg−1 day−1), or their combination on systolic blood pressure (SBP) and heart rate (HR) in the absence and presence of atrasentan (endothelin ETA receptor antagonist, 5 mg kg−1 day−1) in male Sprague Dawley rats. Drugs were administered via oral gavage. Values are means ± S.E.M. of 6 observations.*P b 0.05 vs. corresponding vehicle values, + P b 0.05 vs. corresponding CSA values, #P b 0.05 vs. corresponding indomethacin values, % P b 0.05 vs. corresponding atrasentan values, $P b 0.05 vs. corresponding CSA + indomethacin values.

Celecoxib abrogates CSA-evoked renovascular fibrosis and alterations in ET receptor expression Results of Masson's trichrome staining of kidney sections obtained from rats treated with CSA, indomethacin, celecoxib, or their combinations in the absence and presence of selective ETA (atrasentan) or ETB receptor blockade (BQ788) are shown in Fig. 3. Compared with control (vehicle-treated) rats, microscopic examination of stained kidney sections obtained from CSA-treated rats showed perivascular fibrosis in cortical arterioles (Fig. 3B). The simultaneous administration of celecoxib (3G), but not indomethacin (3C), largely abolished the renal fibrotic action of CSA. The latter was also eliminated in preparations co-treated with atrasentan (Fig. 3E) in contrast to no effect for BQ788 (Fig. 3I). Further, little or no perivascular fibrosis was detected in kidney sections obtained from rats treated with the CSA/indomethacin/atrasentan (Fig. 3F) or the CSA/celecoxib/BQ788 regimen (Fig. 3J). Immunohistochemical studies showed that CSA caused significant increases in ETA receptor expression (Fig. 4) and decreases in ETB receptor expression (Fig. 5) in the renal cortical arterioles compared with vehicle-treated rats. These effects of CSA disappeared in rats treated concurrently with celecoxib but remained unaltered in the presence of indomethacin (Figs. 4 and 5).

4

M.M. El-Mas et al. / Toxicology and Applied Pharmacology 284 (2015) 1–7

Discussion

Fig. 2. Effects of 10-day treatment with CSA (20 mg kg−1 day−1), celecoxib (10 mg kg−1 day−1), or their combination on systolic blood pressure (SBP) and heart rate (HR) in the absence and presence of BQ788 (endothelin ETB receptor antagonist, 0.1 mg/kg/day) in male Sprague Dawley rats. Drugs were administered via oral gavage except BQ788 that was given intraperitoneally. Values are means ± S.E.M. of 6 observations. *P b 0.05 vs. corresponding vehicle values, +P b 0.05 vs. corresponding CSA values, # P b 0.05 vs. corresponding celecoxib values, %P b 0.05 vs. corresponding BQ788 values, $ P b 0.05 vs. corresponding CSA + celecoxib values.

This study reports on the hemodynamic interaction between CSA and NSAIDs and its modulation by ET receptors. The most important observations are: (i) the simultaneous administration of celecoxib, but not indomethacin, blunted the CSA-evoked hypertension and associated renal perivascular fibrosis and alterations in protein expression of ETA (increases) and ETB receptors (decreases), (ii) similar favorable effects on the hypertensive and renovascular fibrotic actions of CSA were seen after selective blockade of ETA receptors by atrasentan, (iii) ETB receptor blockade by BQ788 increased SBP and potentiated the hypertensive effect of CSA, and (iv) both effects of BQ788 were reduced upon concurrent treatment with celecoxib. The data suggest that the inhibition of renal perivascular fibrosis and normalization of the altered ET receptor expression profile contribute, at least partly, to the mitigating effect of celecoxib on the hypertensive effect of CSA in rats. Although ET and its receptor sites have been implicated in the CSAevoked hypertension (Phillips et al., 1994; Bartholomeusz et al., 1996; Cauduro et al., 2005), no attempts were made to investigate the relative contributions of ET receptor subtypes (ETA and ETB) in CSA hypertension. In fact, either nonselective ETA/B or selective ETA receptor antagonists were employed in previous studies including ours in order to verify the role of ET receptors in CSA hemodynamics (Phillips et al., 1994; Bartholomeusz et al., 1996; Nasser et al., 2014). Interestingly, immunohistochemical evidence of the current study showed that CSA caused opposite changes in the abundance of ETA (upregulation) and ETB (downregulation) receptors in the renal cortical arterioles. ETA receptors are present mainly on vascular smooth muscle cells and elicit vasoconstriction upon activation, whereas ETB receptors locate in the vascular endothelium and mediate vasodilation when activated by ET through the release of nitric oxide and/or prostacyclin (Davenport and Maguire, 2011). With this in mind, the increased (ETA) and decreased (ETB) ET receptor expression in cortical arterioles of CSA-treated rats might point to enhanced vasoconstrictor propensity of the renal arteriolar system, which perhaps predisposed the rat to the hypertensive effect of CSA. In addition to molecular studies, more direct evidence for the modulatory role of ET receptors in CSA hypertension was obtained from the receptor antagonist studies. The abolition of the hypertensive action of CSA in the presence of atrasentan strongly emphasizes the importance of intact and functional ETA receptors for the development of CSA hypertension. Contrary to ETA receptors, ETB receptors seem to exert tonic restraining influence on the hypertensive action of CSA

Fig. 3. Photomicrographs (100×) of renal cortical arterioles stained with Masson's trichrome from rats treated for 10 consecutive days with vehicle (A), CSA 20 mg kg−1 day−1 (B), CSA + indomethacin 5 mg kg−1 day−1 (C), atrasentan 5 mg kg−1 day−1 (D), CSA + atrasentan (E), CSA + indomethacin + atrasentan (F), CSA + celecoxib 10 mg kg−1 day−1 (G), BQ788 0.1 mg kg−1 day−1 (H), CSA + BQ788 (I), or CSA + celecoxib + BQ788 (J). Black arrows point to the deposition of collagen fibers.

M.M. El-Mas et al. / Toxicology and Applied Pharmacology 284 (2015) 1–7

5

Fig. 4. Bar graphs showing changes in the immunohistochemical ETA receptor expression in renal cortical arterioles of rats treated for 10 consecutive with CSA, (20 mg kg−1 day−1), indomethacin (5 mg kg−1 day−1), celecoxib (10 mg kg−1 day−1) or their combination in Sprague Dawley rats. Representative images of immunostained tissues are also shown (400×). Values are means ± S.E.M. of 6 observations. *P b 0.05 vs. corresponding vehicle values, +P b 0.05 vs. corresponding CSA values.

because the concurrent administration of BQ788, selective ETB receptor blocker, accentuated the hypertensive action of CSA. Further, the increases in BP caused by the sole exposure to BQ788 also favor a physiological downregulatory effect for ETB receptors on BP controlling mechanisms. Together, although the involvement of ET signaling in CSA hypertension has been previously documented (Phillips et al., 1994; Bartholomeusz et al., 1996; Nasser et al., 2014), to our knowledge, the present study is the first to simultaneously report, both pharmacologically

and molecularly, on the role of the two ET receptor subtypes in the BP effect of CSA. Reports favor the concurrent use of CSA and NSAIDs in the control of arthritic conditions (Cavalcante et al., 2011; Ash et al., 2012). Since prostaglandins, the arachidonic acid products of COX enzymatic activity, produce multiple effects on the kidney, vasculature, and sympathetic neurotransmission, the effects of COX inhibition by NSAIDs on blood pressure have been inconsistent (Conlin et al., 2000; Zheng and Du,

Fig. 5. Bar graphs showing changes in the immunohistochemical ETB receptor expression in renal cortical arterioles of rats treated for 10 consecutive with CSA, (20 mg kg−1 day−1), indomethacin (5 mg kg−1 day−1), celecoxib (10 mg kg−1 day−1) or their combination in Sprague Dawley rats. Representative images of immunostained tissues are also shown (400×). Values are means ± S.E.M. of 6 observations. *P b 0.05 vs. corresponding vehicle values, +P b 0.05 vs. corresponding CSA values.

6

M.M. El-Mas et al. / Toxicology and Applied Pharmacology 284 (2015) 1–7

2014). This prompted us to hypothesize that the simultaneous use of NSAIDs such as celecoxib and indomethacin might modify the hypertensive action of CSA and its modulation by ET signaling. In the current study, although the sole use of indomethacin or celecoxib failed to elicit any appreciable changes in the rat hemodynamics, the two NSAIDs variably influenced the hemodynamic responses elicited by the simultaneously administered CSA. The hypertensive and tachycardia actions of CSA were ameliorated upon concomitant use of celecoxib in contrast to no effect for indomethacin. Credibly, the reversal by celecoxib of the CSA-evoked alterations in the ETA/B receptor immunohistochemical profile highlights a key role for normalized renovascular expression ET receptors in the CSA–celecoxib hemodynamic interaction. The importance of ETB receptors in the cardiovascular protective effect of celecoxib is further confirmed by the finding that celecoxib abolished the hypertensive responses elicited by ETB receptor blockade (BQ788). Overall, the similarity in the BP response to CSA and BQ788, the reduction in renovascular ETB receptor expression caused by CSA, and the abolition of all these responses by celecoxib provide ample evidence that implicate renal ETB receptors in the CSA/celecoxib hemodynamic interaction. The inhibition of CSA-induced renal perivascular fibrosis is another mechanism that may explain the favorable effect of celecoxib on CSA hypertension. The increased renovascular fibrosis caused by CSA is consistent with reported studies, which related CSA hypertension to vascular structural derangements including perivascular fibrosis (Lassila et al., 2001; Rezzani et al., 2005). Pertinent observations include the following: (i) reported studies including ours showed that enhanced ET/ transforming growth factor-beta1 (TGF-β1) signaling mediates tissue fibrosis caused by CSA (Chatziantoniou and Dussaule, 2000; El-Gowelli et al., 2014), (ii) ETB receptors reduce TGF-β1 and subsequently attenuate hypertension and renal failure progression in uremic animals (Lavoie et al., 2005), (iii) the ET/ETA receptor signaling promotes the TGF-β1induced renal fibrosis (Chrobak et al., 2013), and (iv) celecoxib prevents and reverses liver fibrosis induced by carbon tetrachloride via decreasing TGF-β1 expression (Chávez et al., 2010). That said, it is plausible to suggest that the renovascular antifibrotic effect of celecoxib, triggered possibly by the increased ETB and decreased ETA receptor expression, contributed to the BP lowering effect of celecoxib in CSA-treated rats. Whether differences in the COX-1/COX-2 selectivity of celecoxib and indomethacin contributed to their discrepant effects on the hypertensive and fibrotic actions of CSA cannot be ascertained from the current study. More studies are obviously needed to investigate this issue. Clinical studies demonstrated inconsistent effects for NSAIDs on blood pressure. Indomethacin may cause increases, decreases, or no changes in blood pressure (Conlin et al., 2000; Zheng and Du, 2014). Moreover, variable BP effects for the “coxibs” family members have been reported. For example, rofecoxib, but not celecoxib, induced a significant rise in SBP (Sowers et al., 2005). Due to the individuality and unpredictability in their hemodynamic profiles, the addition of NSAIDs to an already hypertensive patient should be done carefully and followed by frequent blood pressure measurements to detect unexpected changes in hemodynamics. Such limitation certainly applies to arthritic conditions that necessitate the combined use of CSA and NSAIDs (Cavalcante et al., 2011; Ash et al., 2012). The finding that the blockade of ETA receptors by atrasentan inhibited the CSA-induced pressor, but not tachycardic, effect deserves a comment. This observation raises the possibility that the tachycardia effect of CSA is independent on the ET/ETA pathway. The CSA-induced tachycardia has been attributed to a specific calcineurin-mediated sympathetic overactivation (Zhang and Victor, 2000). Evidence suggests that ETA receptor blockers might increase the HR probably as a compensatory response to the sustained falls in SBP caused by ETA receptor blockade (Reinhart et al., 2002). In summary, pharmacological and molecular studies undertaken in the present study suggest a key role for ET receptors in the hypertensive response elicited by chronic CSA and the preferential amelioration of this response by celecoxib. The hypertensive effect of CSA was

associated with increases and decreases in ETA and ETB receptor expression in the renal cortical arterioles. CSA hypertension was inhibited and accentuated after selective blockade of ETA (atrasentan) and ETB receptors (BQ788), respectively. The blunting effect of celecoxib on CSA hypertension was paralleled, and possibly caused, by the normalization of the altered ETA/B receptor protein expression profile and the inhibition of renal perivascular fibrosis. Clinically, in addition to the potentiated antiarthritic efficacy of the CSA/celecoxib regimen, celecoxib might help minimize or abolish the troublesome rise in BP that associates CSA therapy. Conflict of interest statement The authors declare that there are no conflicts of interest. Acknowledgments This work was supported by the ALEX-REP Grant Fund (Grant # HLTH-13-01) from Alexandria University, Egypt. The authors thank Dr. Reda S. Saad, Adjunct Professor, Department of Pathology, Western University, London, Ontario, Canada, and Hanan Mostafa Menisi, Department of Pathology, Faculty of Medicine, Alexandria University, for their help with the histopathology. We thank Abbott Laboratories (Illinois, USA) for supplying atrasentan. References Altman, R.D., Perez, G.O., Sfakianakis, G.N., 1992. Interaction of cyclosporine A and nonsteroidal anti-inflammatory drugs on renal function in patients with rheumatoid arthritis. Am. J. Med. 93, 396–402. Ash, Z., Gaujoux-Viala, C., Gossec, L., Hensor, E.M., FitzGerald, O., Winthrop, K., van der Heijde, D., Emery, P., Smolen, J.S., Marzo-Ortega, H., 2012. A systematic literature review of drug therapies for the treatment of psoriatic arthritis: current evidence and meta-analysis informing the EULAR recommendations for the management of psoriatic arthritis. Ann. Rheum. Dis. 71, 319–326. Bartholomeusz, B., Hardy, K.J., Nelson, A.S., Phillips, P.A., 1996. Bosentan ameliorates cyclosporin A-induced hypertension in rats and primates. Hypertension 27, 1341–1345. Cauduro, R.L., Costa, C., Lhulier, F., Garcia, R.G., Cabral, R.D., Gonçalves, L.F., Manfro, R.C., 2005. Endothelin-1 plasma levels and hypertension in cyclosporine-treated renal transplant patients. Clin. Transplant. 19, 470–474. Cavalcante, E.G., Aikawa, N.E., Lozano, R.G., Lotito, A.P., Jesus, A.A., Silva, C.A., 2011. Chronic polyarthritis as the first manifestation of juvenile systemic lupus erythematosus patients. Lupus 20, 960–964. Chatziantoniou, C., Dussaule, J.C., 2000. Endothelin and renal vascular fibrosis: of mice and men. Curr. Opin. Nephrol. Hypertens. 9, 31–36. Chávez, E., Segovia, J., Shibayama, M., Tsutsumi, V., Vergara, P., Castro-Sánchez, L., Salazar, E.P., Moreno, M.G., Muriel, P., 2010. Antifibrotic and fibrolytic properties of celecoxib in liver damage induced by carbon tetrachloride in the rat. Liver Int. 30, 969–978. Chen, G.F., Sun, Z., 2006. Effects of chronic cold exposure on the endothelin system. Appl. Physiol. 100, 1719–1726. Chrobak, I., Lenna, S., Stawski, L., Trojanowska, M., 2013. Interferon-γ promotes vascular remodeling in human microvascular endothelial cells by upregulating endothelin (ET)-1 and transforming growth factor (TGF) β2. J. Cell. Physiol. 228, 1774–1783. Conlin, P.R., Moore, T.J., Swartz, S.L., Barr, E., Gazdick, L., Fletcher, C., DeLucca, P., Demopoulos, L., 2000. Effect of indomethacin on blood pressure lowering by captopril and losartan in hypertensive patients. Hypertension 36, 461–465. Davenport, A.P., Maguire, J.J., 2011. Pharmacology of renal endothelin receptors. Contrib. Nephrol. 172, 1–17. Drury, R.A., Wallington, E.A., 1980. Carleton's Histological Technique. 5th ed. Oxford Medical Publications, pp. 36–150. El-Gowelli, H.M., Helmy, M.W., Ali, R.M., El-Mas, M.M., 2014. Celecoxib offsets the negative renal influences of cyclosporine via modulation of the TGF-β1/IL-2/COX-2/ endothelin ET(B) receptor cascade. Toxicol. Appl. Pharmacol. 275, 88–95. El-Mas, M.M., Afify, E.A., Omar, A.G., Sharabi, F.M., 2002. Cyclosporine attenuates the autonomic modulation of reflex chronotropic responses in conscious rats. Can. J. Physiol. Pharmacol. 80, 766–776. El-Mas, M.M., Afify, E.A., Omar, A.G., Mohy El-Din, M.M., Sharabi, F.M., 2003. Testosterone depletion contributes to cyclosporine-induced chronic impairment of acetylcholine renovascular relaxations. Eur. J. Pharmacol. 468, 217–224. El-Mas, M.M., Mohy El-Din, M.M., El-gowilly, S.M., Sharabi, F.M., 2004. Relative roles of endothelial relaxing factors in cyclosporine-induced impairment of cholinergic and β-adrenergic renal vasodilations. Eur. J. Pharmacol. 487, 149–158. El-Mas, M.M., Zhang, J., Abdel-Rahman, A.A., 2006. Upregulation of vascular inducible nitric oxide synthase mediates the hypotensive effect of ethanol in conscious female rats. J. Appl. Physiol. 100, 1011–1018. El-Mas, M.M., El-Gowelli, H.M., Abd-Elrahman, K.S., Saad, E.I., Abdel-Galil, A.G., AbdelRahman, A.A., 2011. Pioglitazone abrogates cyclosporine-evoked hypertension via

M.M. El-Mas et al. / Toxicology and Applied Pharmacology 284 (2015) 1–7 rectifying abnormalities in vascular endothelial function. Biochem. Pharmacol. 81, 526–533. El-Mas, M.M., El-Gowelli, H.M., El-gowilly, S.M., Fouda, M.A., Helmy, M.M., 2012a. Estrogen provokes the depressant effect of chronic nicotine on vagally-mediated reflex chronotropism in female rats. J. Pharmacol. Exp. Ther. 342, 568–575. El-Mas, M.M., Mohy El-Din, M.M., Helmy, M.M., Omar, A.G., 2012b. Redox imbalances incite the hypertensive, baroreflex, and autonomic effects of cyclosporine in rats. Eur. J. Pharmacol. 694, 82–88. El-Mas, M.M., Omar, A.G., Helmy, M.M., Mohy El-Din, M.M., 2012c. Crosstalk between central pathways of nitric oxide and carbon monoxide in the hypertensive action of cyclosporine. Neuropharmacology 62, 1890–1896. Kaczmarek, E., Górna, A., Majewski, P., 2004. Techniques of image analysis for quantitative immunohistochemistry. Ann. Acad. Med. Bialostocensis 49, S55–S158. Kamalutheen, M., Gopalakrishnan, S., Syed-Ismail, T., 2009. Anti-inflammatory and antiarthritic activities of Merremia tridentata (L.). E-J. Chem. 6, 943–948. Lassila, M., Santisteban, J., Finckenberg, P., Salmenperä, P., Riutta, A., Moilanen, E., Virtanen, I., Vapaatalo, H., Nurminen, M.L., 2001. Vascular changes in cyclosporine A-induced hypertension and nephrotoxicity in spontaneously hypertensive rats on high-sodium diet. J. Physiol. Pharmacol. 52, 21–38. Lavoie, P., Robitaille, G., Agharazii, M., Ledbetter, S., Lebel, M., Larivière, R., 2005. Neutralization of transforming growth factor-beta attenuates hypertension and prevents renal injury in uremic rats. J. Hypertens. 23, 1895–1903. Lima, V.V., 2010. O-Glyc-N-Acylation contributes to augmented vascular reactivity induced by endothelin-1. Hypertension 55, 180–188. Nasser, S.A., Elmallah, A.I., Sabra, R., Khedr, M.M., El-Din, M.M., El-Mas, M.M., 2014. Blockade of endothelin ETA, but not thromboxane, receptors offsets the cyclosporineevoked hypertension and interrelated baroreflex and vascular dysfunctions. Eur. J. Pharmacol. 727, 52–59. Nishiyama, A., Kobori, H., Fukui, T., Zhang, G.X., Yao, L., Rahman, M., Hitomi, H., Kiyomoto, H., Shokoji, T., Kimura, S., Kohno, M., Abe, Y., 2003. Role of angiotensin II and reactive oxygen species in cyclosporine A-dependent hypertension. Hypertension 42, 754–760.

7

Olyaei, A.J., de Mattos, A.M., Bennett, W.M., 1999. Immunosuppressant-induced nephropathy: pathophysiology, incidence and management. Drug Saf. 21, 471–488. Phillips, P.A., Rolls, K.A., Burrell, L.M., Casley, D., Aldred, K.L., Hardy, K.J., 1994. Vascular endothelin responsiveness and receptor characteristics in vitro and effects of endothelin receptor blockade in vivo in cyclosporin hypertension. Clin. Exp. Pharmacol. Physiol. 21, 223–226. Reinhart, G.A., Preusser, L.C., Burke, S.E., Wessale, J.L., Wegner, C.D., Opgenorth, T.J., Cox, B.F., 2002. Hypertension induced by blockade of ET(B) receptors in conscious nonhuman primates: role of ET(A) receptors. Am. J. Physiol. Heart Circ. Physiol. 283, H1555–H1561. Rezzani, R., Rodella, L., Buffoli, B., Giugno, L., Stacchiotti, A., Bianchi, R., 2005. Cyclosporine A induces vascular fibrosis and heat shock protein expression in rat. Int. Immunopharmacol. 5, 169–176. Riva, N., Cáceres Guido, P., Rousseau, M., Dip, M., Monteverde, M., Imventarza, O., Mato, G., Schaiquevich, P., 2013. Pharmacovigilance of calcineurin inhibitor in peidatric kidney and liver transplantation. Farm. Hosp. 37, 441–449. Soubhia, R.M., Mendes, G.E., Mendonça, F.Z., Baptista, M.A., Cipullo, J.P., Burdmann, E.A., 2005. Tacrolimus and nonsteroidal anti-inflammatory drugs: an association to be avoided. Am. J. Nephrol. 25, 327–334. Sowers, J.R., White, W.B., Pitt, B., Whelton, A., Simon, L.S., Winer, N., Kivitz, A., van Ingen, H., Brabant, T., Fort, J.G., 2005. Celecoxib Rofecoxib Efficacy and Safety in Comorbidities Evaluation Trial (CRESCENT) Investigators. The effects of cyclooxygenase-2 inhibitors and nonsteroidal anti-inflammatory therapy on 24-hour blood pressure in patients with hypertension, osteoarthritis, and type 2 diabetes mellitus. Arch. Intern. Med. 165, 161–168. Tavares, P., Martinez-Salgado, C., Fontes Ribeiro, C.A., Eleno, N., Lopez-Navoa, J.M., Teixeira, F., 2002. Cyclosporin effects on rat aorta α1-adrenoceptors and their transduction mechanisms. J. Cardiovasc. Pharmacol. 40, 181–188. Zhang, W., Victor, R.G., 2000. Calcineurin inhibitors cause renal afferent activation in rats: a novel mechanism of cyclosporine-induced hypertension. Am. J. Hypertens. 13, 999–1004. Zheng, L., Du, X., 2014. Non-steroidal anti-inflammatory drugs and hypertension. Cell Biochem. Biophys. 69, 209–211.