Therapeutic Effects of Udenafil on Pressure-Overload Cardiac

ORIGINAL ARTICLE

Therapeutic effects of udenaﬁl on pressure-overload cardiac hypertrophy

Hack-Lyoung Kim1,2, Yong-Jin Kim2,3, Kyung-Hee Kim4, Seung-Pyo Lee2,3, Hyung-Kwan Kim2,3, Dae-Won Sohn2,3, Byung-Hee Oh2,3 and Young-Bae Park2,3,5

This study was performed to determine whether the newly developed phosphodiesterase type 5 (PDE5) inhibitor udenafil had beneficial effects on pressure-overload cardiac hypertrophy. Pressure overload cardiac hypertrophy was created by using suprarenal aortic constriction (SAC) in male Sprague–Dawley rats. Rats were divided into three groups: sham (n = 19), SAC (n = 18) and SAC+udenafil(n = 14) groups. Three-week periods of SAC provoked significant left ventricular (LV) hypertrophy. Udenafil was administered (20 mg kg − 1 PO, daily) between the 3rd and 20th weeks after SAC in the SAC+udenafil group. Udenafil improved the survival rate (log-rank P = 0.012) and exercise capacity (maximal exercise duration at the 20th week after surgery: 448 ± 54 s for the SAC+udenafil group versus 317 ± 73 s for the SAC group, Po0.05) of the rats with SAC. Serial echocardiographic examinations showed that udenafil attenuated LV remodeling processes following SAC (mean LV end-diastolic dimension at the 20th week after surgery: 9.84 ± 0.59 mm for SAC and 9.05 ± 0.58 mm for SAC+udenafil group, Po0.05). Invasive hemodynamic studies showed that udenafil improved the LV performance. Udenafil-attenuated myocardial fibrosis and apoptosis. Udenafil also decreased myocardial matrix metalloproteinase-9 expression and augmented serum interleukin-10 concentration. Long-term udenafil use prevented cardiac remodeling and improved exercise capacity and survival in rats exposed to pressure-overload cardiac hypertrophy. Hypertension Research (2015) 38, 597–604; doi:10.1038/hr.2015.46; published online 2 April 2015

Keywords: phosphodiesterase inhibitor; suprarenal aortic constriction; udenaﬁl; ventricular hypertrophy

INTRODUCTION This study was performed to determine whether udenafilexhibitsa Phosphodiesterase (PDE) type 5 selectively hydrolyzes cyclic guanosine protective effect on pressure-overload cardiac hypertrophy. We created monophosphates.1 An inhibitor of PDE5, sildenafil, leads to smooth a chronic pressure-overload cardiac hypertrophy model in rats using muscle relaxation in the corpus cavernosum, which governs penile suprarenal abdominal aorta constriction (SAC) and investigated the erection, and it is widely used to treat erectile dysfunction.2 Recent effects of udenafil on the following factors: (1) prevention of left growing evidence supports the cardio-protective effects of sildenafil ventricular (LV) dysfunction and remodeling, (2) exercise capacity and against ischemia reperfusion3,4 and anthracycline toxicity,5 and PDE5 (3) survival. We also sought to elucidate the mechanisms mediating inhibition by sildenafil blunts responses to adrenergic-stimulated the effects of udenafil. cardiac contractility.6 Sildenafil also suppresses chronic cardiac hypertrophy and dysfunction attributable to pressure or volume MATERIALS AND METHODS overload.7,8 Animals Udenafil is a novel, potent PDE5 inhibitor that exhibits a similar A total of 100 male 8-week-old Sprague–Dawley rats, weighing ~ 300 g, were molecular structure to that of sildenafil. An in vitro study evaluating used for this study. Rats were divided into three groups: sham (n = 24), n = fi n = various PDE isoenzymes demonstrated that the selectivity of udenafil suprarenal aortic constriction (SAC) ( 40) and SAC+udena l( 36). One, 8and7ratsinthesham,SACandSAC+udenafil groups, respectively, died is comparable to sildenafil.9 Udenafil exhibits potent efficacy in erectile within 3 weeks post surgery. The mortality rates in the acute stage post surgery dysfunction with a broad safety margin. Therefore, it has been widely were 4.1% in the sham group and 19.7% in the SAC group. Rats with 10 used in the treatment of erectile dysfunction since 2005. The similar incomplete modeling, which was defined as LV septal wall thickness (LVSWT) fi molecular structures, mechanisms of action and potency of udena l by using transthoracic echocardiography o2.5 mm at the 3rd week of SAC in and sildenafil suggest that udenafil has cardio-protective effects. the SAC (n = 9) and SAC+udenafil(n = 11) groups were excluded. The However, there is a lack of studies on the role of udenafilintheheart. modeling success rate was 67.2%. Four rats in each group were euthanized

1Boramae Medical Center, Seoul, South Korea; 2Department of Internal Medicine, Seoul National University College of Medicine, Seoul, Korea; 3Seoul National University Hospital, Seoul, South Korea; 4Department of Internal Medicine, Sejong General Hospital, Bucheon, South Korea and 5Healthcare System Gangnam Center, Seoul, South Korea Correspondence: Dr Y-J Kim, Seoul National University Hospital, Department of Internal Medicine, Seoul National University College of Medicine, 101 Daehak-ro, Jongno-gu, Seoul 110-744, Korea. E-mail: [email protected] Received 6 August 2014; revised 18 January 2015; accepted 10 February 2015; published online 2 April 2015 Udenaﬁloncardiachypertrophy H-L Kim et al 598

at the 3rd week post operation for baseline analyses, including myocardial histopathological assessment, hemodynamic study and serum cytokine mea- surement. Therefore, the final numbers of animals that had full follow-up between the 3rd and 20th weeks post operation were 19, 18 and 14 in the sham, SAC and SAC+udenafil groups, respectively. All surgeries were performed under isoflurane anesthesia, and all efforts were made to minimize the suffering of the rats. Udenafil (CAS No.; 268203-93-6, 5-[2-propoxy-5-(1-methyl-2-pyrrolidiny- lethylaminosulfonyl)phenyl]-1-methyl-3-propyl-1,6-dihydro-7H-pyrazolo(4,3- d)pyramidine-7-one) was obtained from Dong-A Pharmaceutical Company (Seoul, Korea). The rats in the SAC+udenafil group received daily oral administrations of udenafil (20 mg kg − 1)11,12 via gavage beginning the 3rd week after the SAC surgery until study termination, the 20th week after surgery. The rats in the other groups received vehicle (Trisol buffer solution, citrate sodium hydroxide buffer, pH 5.0, Merck, Darmstadt, Germany) treatment during the same study periods. The treatment was well tolerated by all rats. The rats were housed under standard laboratory conditions, and food and UV- Figure 1 Kaplan–Meier survival curves of the three groups. n = 19, 18 and sterilized tap water were provided. The study was performed in strict 14 in sham, SAC and SAC+UD groups, respectively. SAC, suprarenal aortic accordance with the recommendations of the Institutional Animal Care and constriction; UD, udenafil. Use Committee of Seoul National University Hospital (Seoul, Korea).

Creation of the SAC model LV hypertrophy was induced by using SAC surgery (Supplementary Figure 800 Sham S1).13,14 The abdominal aorta was exposed through a midline abdominal SAC + UD incision under sterile conditions. The abdominal aorta between the diaphragm SAC and the renal artery was partially constricted by using a 22-gauge needle 700 (outside diameter = 0.7 mm). The needle was placed parallel to the aorta and ligated tightly to the aorta. The needle was carefully removed, leaving the vessel 600 constricted to the diameter of the needle. This procedure produces ~ 50% stenosis of the aortic area.13 The rats in the sham group underwent a similar * * * surgical procedure without aortic constriction. 500 * Expanded methods are available in the Supplementary Material online. # * # # 400 RESULTS SAC surgery created cardiac hypertrophy Exercise duration (sec) The suprarenal abdominal aortas were harvested three weeks after 300 SAC, and the luminal cross-sectional area was measured (n = 3ineach group). The mean luminal cross-sectional area of suprarenal abdom- 200 fi inal aortas in the SAC group was signi cantly smaller than in the sham Base 3Wk 6Wk 9Wk 12Wk 15Wk 20Wk 2 operation group (0.51 ± 0.18 versus 1.26 ± 0.37 mm , P = 0.037) Weeks after operation (Supplementary Figure S2). There were no significant differences in o body weight during the study period between the three groups Figure 2 Serial changes in exercise capacities of the three groups. *P 0.05 for difference from sham group; #Po0.05 for difference from SAC group. (Supplementary Figure S3). Supplementary Table S1 summarizes the SAC, suprarenal aortic constriction; UD, udenafil. weights of major organs of normal rats and rats exposed to three weeks of pressure-overload. There were no significant differences in liver weights. In contrast, the rats that underwent the SAC surgery week post surgery in the SAC group. Exercise capacity at the 20th showed significantly increased heart, lung and heart/body weights and week in the SAC group (447 ± 51 s to 317 ± 73 s, Po0.05) decreased lung/body weight ratios (Po0.05). to 70% of the exercise capacity 3 weeks post surgery. Exercise capacity was maintained in the SAC+udenafil group throughout the study Udenafil increased the survival of rats with cardiac hypertrophy period (458 ± 107 s at the 3rd week to 448 ± 54 s at the 20th week post Five of 13 rats (38.5%) in the SAC group and two of 12 rats (16.6%) surgery, P40.05). Exercise capacity was better in the SAC+udenafil in the SAC+udenafil group died during the study period. No deaths group than the SAC group at the 20th week (448 ± 54 s for the SAC occurred in the sham group. The first deaths occurred on the 56th and +udenafil group versus 317 ± 73 s for the SAC group, Po0.05). 128th days post surgery in the SAC and SAC+udenafilgroups, respectively. The survival benefitofudenafil was significant based on Udenafil improved LV systolic function and inhibited LV the Kaplan–Meier survival curve (log-rank P = 0.012) (Figure 1). remodeling in rats with cardiac hypertrophy Pre-surgery echocardiographic examinations showed no significant Udenafil increased the exercise capacity of rats with cardiac differences in LVSWT, LV end-diastolic dimension, LV ejection hypertrophy fraction (LVEF) or LV mass index between the three groups The mean maximal exercise capacities (seconds to exhaustion) at (Supplementary Table S2). Serial echocardiographic changes in baseline were not different between the three groups (P = 0.982). LVSWT, LV end-diastolic dimension, LV mass index and LVEF in Longitudinal changes in maximal exercise capacity are demonstrated the three groups are presented in Figure 3. LVSWT increased in Figure 2. Exercise capacity was progressively impaired from the 6th significantly in the SAC (1.97 ± 0.14 to 2.67 ± 0.13 mm, Po0.05)

Hypertension Research Udenaﬁl on cardiac hypertrophy H-L Kim et al 599

3.5 SAC 11 SAC + UD Sham SAC 10 SAC + UD 3.0 Sham # # # 9 # # 2.5 * # # * * * * * 8

2.0 7 LV septal wall thickness (mm) LV LV end-diastolic dimension (mm) LV

1.5 6 Base 3Wk 6Wk 9Wk 12Wk 15Wk 20Wk Base 3Wk 6Wk 9Wk 12Wk 15Wk 20Wk Weeks after operation Weeks after operation

5 SAC 90 Sham SAC + UD SAC + UD Sham SAC

80 # 4 # # # # * * * * * * 70 * * * * * * # 3 # #

LV mass index (g/kg) LV 60 LV ejection fraction (%) LV

2 50 Base 3Wk 6Wk 9Wk 12Wk 15Wk 20Wk Base 3Wk 6Wk 9Wk 12Wk 15Wk 20Wk Weeks after operation Weeks after operation

Sham or SAC operation

Sham

SAC

SAC + UD

Baseline 3 weeks 9 weeks 20 weeks

Figure 3 The results of serial echocardiographic examinations of the three groups. (a) Left ventricular septal wall thickness, (b) left ventricular end-diastolic dimension, (c) left ventricular mass index, (d) left ventricular ejection fraction, and (e) representative ﬁgures with M-mode tracings of left ventricles. *Po0.05 for difference from sham group; #Po0.05 for difference from SAC group. SAC, suprarenal aortic constriction; UD, udenaﬁl.

Hypertension Research Udenaﬁloncardiachypertrophy H-L Kim et al 600

and SAC+udenafil(2.00± 0.18 to 2.78 ± 0.15 mm, Po0.05) groups at group and the SAC+udenafilgroupweresignificant from the 12th the 3rd week post surgery compared with the baseline. The LVSWT of week after SAC. The mean LVEFs were 58 ± 3% for the SAC group the SAC group increased progressively until the 9th week and slightly and 67 ± 3% for the SAC+udenafil group at the 20th week post decreased thereafter. The LVSWT decreased consistently from the 3rd surgery (Po0.05) (Figure 3d). LV fractional shortening, stroke volume week post surgery in the SAC+udenafil group. The mean LVSWTs and cardiac output showed similar patterns to LVEF in serial were 2.78 ± 0.18 mm for the SAC and 2.38 ± 0.13 mm for SAC echocardiographic examinations (Supplementary Figure S4). Repre- +udenafil groups 20 weeks post surgery (Po0.05) (Figure 3a). LV sentative figures of echocardiographic examinations with M-mode cavity was more rapidly dilated in the SAC group than in the SAC tracings of LV are shown in Figure 3e. +udenafil group. The mean LV end-diastolic dimensions were 9.84 ± 0.59 mm for the SAC and 9.05 ± 0.58 mm for the SAC+udenafil Udenafil improved LV performance as measured using invasive groups (Po0.05) at the 20th week post surgery (Figure 3b). LV mass hemodynamic study in rats with cardiac hypertrophy index increased significantly in the SAC (2.84 ± 0.42 to 3.94 ± 0.46 Hemodynamic studies using an LV micro-catheter were performed at gkg− 1, Po0.05) and SAC+udenafil(2.75± 0.32 to 3.98 ± 0.59 g kg − 1, the 3rd week (n = 4 in each group) and the 20th week (n = 7ineach Po0.05) groups at the 3rd week post surgery compared with baseline. group) post surgery. Mean arterial blood pressure (BP) measured at LV mass index progressively increased in the SAC group throughout the carotid artery was higher in the SAC group at the 3rd week post the study period. Udenafil treatment attenuated the increase in LV surgery than in the sham control group (89.2 ± 10.3 versus 63.0 ± 8.2 mass index (4.61 ± 0.20 g kg − 1 in the SAC group versus 3.84 ± 0.29 mm Hg, Po0.05). LV end-systolic pressure (LVESP) increased in the gkg− 1 in the SAC+udenafil group at the 20th week post surgery, SAC group compared with the sham control group (127 ± 13 versus Po0.05) (Figure 3c). LVEF decreased significantly in the SAC (75 ± 4 97 ± 11 mm Hg, P = 0.05). However, LV end-diastolic pressure to 61 ± 5%, Po0.05) and SAC+udenafil(74± 4to61± 5%, Po0.05) (LVEDP) was not different between the two groups (13.6 ± 2.2 versus groups at the 3rd week post surgery compared with baseline. LVEF 11.4 ± 3.2 mm Hg, P40.05) (Figure 4a). The maximal slope of the progressively decreased throughout the study period in the SAC group, systolic pressure increment (+dP/dt) (6756 ± 654 versus 3648 ± 456 but it increased continuously in the SAC+udenafil group from the 3rd mm Hg s − 1, Po0.001) and the decrease in diastolic pressure (− dP/dt) week post SAC operation. Differences in LVEF between the SAC ( −6568 ± 735 versus − 3348 ± 731 mm Hg s − 1, Po0.001) were better

mmHg mmHg P P = 0.05 = NS

P = NS P = NS

mmHg/sec mmHg/sec P < 0.001

* -dP/dt -dP/dt +dP/dt +dP/dt *

* versus the others, P < 0.05 P < 0.001

Figure 4 Hemodynamic studies of left ventricles. (a) Left ventricular pressures of the sham and SAC groups at the 3rd week post surgery, and (b)left ventricular pressures of the three groups at the 20th week post surgery, (c)+dP/dtand–dP/dt of the sham and SAC groups at the 3rd week post surgery and (d)+dP/dtand–dP/dt of the three groups at the 20th week post surgery. n = 4 in each group. LVEDP, left ventricular end-diastolic pressure; LVESP, left ventricular end-systolic pressure; SAC, suprarenal aortic constriction; Ud, udenaﬁl.

Hypertension Research Udenafil on cardiac hypertrophy H-L Kim et al 601 in the SAC group than the sham control group (Figure 4c). Mean BPs Udenafil-attenuated myocardial fibrosis, apoptosis and MMP-9 of carotid arteries were measured at the 20th week post surgery as expression in rats with cardiac hypertrophy 73.2 ± 12, 92.2 ± 6.7, 85.2 ± 10.3 mm Hg in the sham, SAC and SAC LV wall thickness at the 3rd week post surgery increased significantly +udenafil groups, respectively. The mean carotid artery BP was higher in the SAC group compared with the sham group as shown in gross in the SAC group than in the sham group (Po0.05), but there was no examination (Supplementary Figure S5A) and hematoxylin and eosin ’ significant difference in mean carotid artery BPs between the SAC and stains (Supplementary Figure S5B). Masson s trichrome staining fi SAC+udenafilgroups(P40.05), which suggests that udenafilis revealed that the extent of interstitial and perivascular brosis was fi ineffective in reducing BP. Neither LVEDP nor LV end-systolic signi cantly greater in the SAC group than in the sham group at the 3rd week post surgery (Supplementary Figure S6). PDE5 expression in pressure were different between the three groups at the 20th week the myocardium of the sham group was negligible, but PDE5 post surgery (P40.05) (Figure 4b). The amplitudes of +dP/dt expression increased in the myocardium of the rats exposed to SAC. and –dP/dt signals decreased significantly in the SAC group, which Representative figures of PDE5 expression at the 3rd week post surgery indicates a decrease in cardiac contractility. However, udenafiltreat- are shown in Supplementary Figure S7. ment preserved these amplitudes (+dP/dt, sham versus SAC versus Gross histopathological examinations at the 20th week post surgery fi ± ± ± SAC+udena l, 6694 1505 versus 3540 1544 versus 6518 1306 revealed markedly increased LV wall thickness in the SAC group, but − 1 Po vs fi mm Hg s , 0.05; -dP/dt, sham versus SAC . SAC+udena l, hypertrophy was attenuated in the udenafil-treated SAC group − ± − ± − ± − 1 7306 1639 versus 3549 1790 versus 6575 1560 mm Hg s , (Figure 5a). Masson’s trichrome staining revealed that the fibrotic Po0.05) (Figure 4d). changes in the myocardium were more prominent in the SAC group,

Sham SAC SAC + UD

5mm

Sham SAC SAC + UD

1mm

Sham SAC SAC + UD

1mm Sham SAC SAC + UD

200μm

Figure 5 Histopathological examinations of left ventricles at the 20th week post surgery. (a) Gross examinations, (b)Masson’s trichrome staining showing the extent of ﬁbrosis, (c) immunohistochemistry showing matrix metalloproteinase-9 expression and (d) TUNEL staining showing apoptosis. For TUNEL staining, n = 5 in each group. SAC, suprarenal aortic constriction; TUNEL, terminal deoxynuclotidyl transferase-mediated UTP nick-end labeling; UD, udenaﬁl.

Hypertension Research Udenaﬁloncardiachypertrophy H-L Kim et al 602

and udenafil administration attenuated these changes at the 20th week DISCUSSION post surgery (Figure 5b). Matrix metalloproteinase (MMP)-2 expres- The study results demonstrated that the long-term use of udenafil sion in the myocardium was very weak in the rats in all three groups prevented cardiac remodeling and systolic dysfunction, which (data not shown). MMP-9 was highly expressed in the SAC group, but improved the exercise capacity and survival in rats exposed to udenafil treatment attenuated this expression (Figure 5c). Terminal pressure-overload cardiac hypertrophy. Decreases in fibrosis and deoxynuclotidyl transferase-mediated UTP nick-end labeling staining apoptosis of hypertrophied myocardium and MMP-9 and IL-10 showed a significant increase in myocardial apoptosis in the SAC modulations were proposed as potential mechanisms related to the fi group, which was attenuated in the SAC+udenafil group (apoptotic cardio-protective effects of udena l. cells per high-power field (×200), 3.8 ± 2.3, 40.2 ± 31.9 and 14.6 ± 6.5, The role of PDE5 inhibition in pressure-overload cardiac in sham, SAC and SAC+udenafil groups, respectively, Po0.05) hypertrophy (Figure 5d and Supplementary Figure S8). Cardiac hypertrophy in response to sustained pressure-overload is associated with maladaptive molecular alterations that lead to cardiac failure.15 Accumulating evidence supports the cardio-protective effects Udenafil augmented interleukin (IL)-10 in serum of rats with of PDE5 inhibition in pressure-overload cardiac hypertrophy. PDE5 cardiac hypertrophy expression increased,8,16 and PDE5 inhibition by sildenafil suppressed Four inflammatory cytokines, including interleukin (IL)-1β, IL-6, LV hypertrophy and improved LV systolic function in mice exposed to IL-10 and tumor necrosis factor-α, were measured in rat serum chronic pressure-overload induced by thoracic aorta constriction.8 (n = 4 in each group) at the 3rd and 20th weeks post surgery. IL-6 and Sildenafil inhibited hypertrophic signaling pathways triggered by tumor necrosis factor-α were not reliably measured because of their pressure-overload, such as phosphatidylinositol-3 kinases (PI3K)/Akt low concentrations. Serum IL-1β was increased in the SAC groups and extracellular signal-regulated kinases (ERK)1/2 pathways, which compared with the sham group at the 3rd week post surgery, but it did attenuated cardiac and myocyte hypertrophy and interstitial fibrosis fi P = not reach statistical signi cance ( 0.107). Serum IL-10 levels were and improved cardiac function.8 Intraperitoneal administration of similar between the sham and SAC groups at the 3rd week post sildenafil decreased cardiac hypertrophy and myocyte injury in rats in P = β surgery ( 0.978). Serum IL-1 levels were not different between the an isoproterenol-induced myocardial injury model.17 We demon- P = three groups at the 20th week post surgery ( 0.241). However, strated a marked increase in PDE5 expression in the heart of the rats serum IL-10 levels increased significantly in the SAC+udenafilgroup subjected to SAC for 3 weeks compared with that in the hearts of the − (94 ± 33, 296 ± 206, and 637 ± 120 pg ml 1, in sham, SAC, and SAC controls and a therapeutic effect of udenafil on PDE5 inhibition in this +udenafil groups, respectively, Po0.001) (Figure 6). model, which is consistent with these previous reports. Our results add

P = 0.978 120 P = 0.107 400

90 300

60 200 (pg/mL) β 30 100 IL-1 IL-10 (pg/mL)

0 0 Sham SAC Sham SAC

60 ANOVA P = 0.241800 ANOVA P < 0.001

600 40 * 400 (pg/mL)

β 20 200 IL-10 (pg/mL) IL-1

0 0 Sham SAC + SAC + Sham SAC + SAC + Veh UD Veh UD * versus the others, P < 0.05

Figure 6 Serum IL-1β and IL-10 levels in rat sera. (a and b)Atthe3rdweekpostsurgeryand(c and d) at the 20th week post surgery. n = 4ineachgroup. ANOVA, analysis of variance; IL, interleukin; SAC, suprarenal aortic constriction; UD, udenaﬁl; Veh, vehicle.

Hypertension Research Udenafil on cardiac hypertrophy H-L Kim et al 603 evidence of the important role of PDE5 in pressure-overload cardiac the remodeling process and improve cardiac contractibility of failing hypertrophy and subsequent cardiac failure, which suggests that PDE5 hearts.8,31,32 The regression of pathological cardiac hypertrophy has inhibition is a potential therapeutic target. become a target of cardiac medicine based on these findings.33,34 The exploration of molecular targets that are involved in the regression of Effects of udenafil on IL-10 and MMP-9 modulation pathological cardiac hypertrophy would lead to alternative therapeutic To our knowledge, this is the first study to demonstrate associations of options for heart failure. However, no remarkable progress in the IL-10 and MMP-9 with PDE5 inhibitors in pressure-overload hyper- development of new therapeutics targeting cardiac hypertrophy has fl 18 trophy. IL-10 is a strong anti-in ammatory molecule, and the occurred, despite this effort. The results of this study have clinical fl production of IL-10 increased during SAC-induced pro-in ammatory implications and suggest udenafil to be another therapeutic option for fi fi conditions. We demonstrated that udena ltreatmentsignicantly the regression of established pathological cardiac hypertrophy. Our augmented IL-10 levels 20 weeks after the SAC surgery, which is group is performing clinical trials on the therapeutic effect of udenafil consistent with previous reports that improvements in cardiac in patients with chronic systolic heart failure.35 function after treatment with losartan, growth hormone or immu- 19,20 noglobulins are associated with increased IL-10. The increased Study limitations expression and activation of MMPs were documented in failing heart, This study has several limitations. First, the study groups were not and these changes were implicated in adverse myocardial blinded to the researchers. Second, udenafil demonstrated several remodeling.21 The role of MMP-9 in the LV remodeling process changes in this study, but the mechanisms mediating the action of supports our hypothesis that MMP-9 inhibition by udenafilhasarole udenafil are largely unknown. Third, there were no comparative in delaying the remodeling process. The present results suggest that experiments using sildenafil. Lastly, direct application of our results to the cardio-protective effect of udenafil on LV remodeling and cardiac humans may be difficult because much higher dosages of udenafil function is partially mediated via IL-10 augmentation and MMP-9 were used in this study than those found in clinical use. suppression. CONCLUSION Consideration of BP-dependent effects of udenafil Long-term udenafil use prevents cardiac remodeling, which improves The arterial BP lowering effects of udenafil, which allowed improved LV hemodynamic performance, should also be considered to be exercise capacity and survival in rats exposed to pressure-overload fi another possible mechanism. The BP difference in carotid arteries cardiac hypertrophy. Decreases in brosis and apoptosis of the fl between SAC and SAC + udenafilgroupswasnotsignificant in our hypertrophied myocardium and modulation of MMP and in amma- study, but the result could not exclude the BP lowering effects of tory cytokines are proposed as potential mechanisms related to the fi udenafil. Arterial unloading has a significant impact on LV perfor- cardio-protective effects of udena l. mance and geometry in cardiac hypertrophy.22 Therefore, the BP- dependent effect of udenafil is an important issue. Animal and clinical CONFLICT OF INTEREST fl studies investigated the effects of udenafil on systemic arterial BP.23–25 The authors declare no con icts of interest. These studies demonstrated that the impact of udenafilonsystemic arterial BP was minimal and clinically insignificant, which is consistent ACKNOWLEDGEMENTS with our finding and implies that arterial BP changes are not the main This study was supported by a grant from the Korea Institute of Medicine and the Korea Healthcare Technology Research and Development Project, Ministry factors mediating the therapeutic potential of udenafil. for Health, Welfare and Family Affairs, Republic of Korea (A102169). The fi fi fi authors thank the Dong-A Pharmaceutical Co. for providing udena l. In Strength of udena lcomparedwithsildena l addition, the authors appreciate the technical support of Seong Deok Park and There are major concerns regarding the side effects of PDE5 Jinhee Kim. inhibitors, such as facial flushing, headache, color discrimination abnormality and cardiac events.26 Therefore, efforts are underway to develop new PDE5 inhibitors with higher potencies and a lower fi incidence of adverse drug reactions. Udena l has a number of 1 Maurice DH. Cyclic nucleotide phosphodiesterase-mediated integration of cGMP and strengths compared with sildenafil. Udenafiliscomparabletosildenafil cAMP signaling in cells of the cardiovascular system. Front Biosci 2005; 10: in PDE5 selectivity,9 and it exhibits relatively few side effects and a 1221–1128. 27 2 Rajfer J, Aronson WJ, Bush PA, Dorey FJ, Ignarro LJ. Nitric oxide as a mediator of longer half-life than sildenafil. Udenafil can also be administered relaxation of the corpus cavernosum in response to nonadrenergic, noncholinergic without limitations of meal conditions, and modifications of the neurotransmission. NEnglJMed1992; 326:90–94. 10 3 Das A, Xi L, Kukreja RC. Phosphodiesterase-5 inhibitor sildenafil preconditions adult dosing regimen are unnecessary. Many previous trials demonstrated cardiac myocytes against necrosis and apoptosis. Essential role of nitric oxide signaling. that long-term sildenafil treatment improved exercise capacity in JBiolChem2005; 280:12944–12955. patients with heart failure or pulmonary hypertension.28,29 Udenafil 4 Salloum F, Yin C, Xi L, Kukreja RC. Sildenafil induces delayed preconditioning through fi inducible nitric oxide synthase-dependent pathway in mouse heart. Circ Res 2003; 92: has a longer half-life and action of duration than sildena l, and once- 595–597. daily dosing may be sufficient to obtain favorable effects.30 Our results 5 Fisher PW, Salloum F, Das A, Hyder H, Kukreja RC. Phosphodiesterase-5 inhibition fi with sildenafil attenuates cardiomyocyte apoptosis and left ventricular dysfunction in a suggest that udena l plays an important role in the treatment of – fi chronic model of doxorubicin cardiotoxicity. Circulation 2005; 111:1601 1610. hypertrophic heart disease with practical advantages over sildena l. 6 Borlaug BA, Melenovsky V, Marhin T, Fitzgerald P, Kass DA. Sildenafil inhibit beta- adrenergic-stimulated cardiac contractility in humans. Circulation 2005; 112: 2642–2649. Clinical implications 7 Kim KH, Kim YJ, Ohn JH, Yang J, Lee SE, Lee S, Kim HK, Seo JW, Sohn DW. Long- Pathological cardiac hypertrophy is considered an irreversible pathway. term effects of sildenafil in a rat model of chronic mitral regurgitationbenefits of However, the traditional view of the irreversibility of pathological ventricular remodeling and exercise capacity. Circulation 2012; 125:1390–1401. 8 Takimoto E, Champion HC, Li M, Belardi D, Ren S, Rodriguez ER, Bedja D, Gabrielson cardiac hypertrophy was challenged recently. Several studies reported KL, Wang Y, Kass DA. Chronic inhibition of cyclic GMP phosphodiesterase 5 A prevents that certain devices or drugs can reverse cardiac hypertrophy, prevent and reverses cardiac hypertrophy. Nat Med 2005; 11:214–222.

Hypertension Research Udenaﬁloncardiachypertrophy H-L Kim et al 604

9 Doh H, Shin CY, Son M, Ko JI, Yoo M, Kim SH, Kim WB. Mechanism of erectogenic 22 Garcia D, Pibarot P, Kadem L, Durand LG. Respective impacts of arotic stenosis and effect of the selective phosphodiesterase type 5 inhibitor, DA-8159. Arch Pharm Res systemic hypertension on left ventricular hypertrophy. Hypertens 2007; 40:972–980. 2002; 25:873–878. 23 Choi SM, Kim JE, Kang KK. Chronic treatment of DA-8159, a new phosphodiesterase 10 Kim TE, Kim BH, Kim JR, Lim KS, Hong JH, Kim KP, Kim HS, Shin SG, Jang IJ, type V inhibitor, attenuates endothelial dysfunction in stroke-prone spontaneously Yu KS. Effect of food on the pharmacokinetics of the oral phosphodiesterase 5 hypertensive rat. Life Sci 2006; 78:1211–1216. inhibitor udenafil for the treatment of erectile dysfunction. B J Clin Pharmacol 2009; 24 Chung BH, Lee JY, Lee SH, Yoo SJ, Lee SW, Oh CY. Safety and efficacy of the 68:43–46. simultaneous administration of udenafilandanα-blcoker in men with erectile 11 Shim HJ, Kim YC, Lee JH, Kwon JW, Kim WB, Kim YG, Kim YG, Kim SH, Lee MG. dysfunction concomitant with BPH/LUTS. Int J Import Res 2009; 21:122–128. Interspecies pharmacokinetic scaling of DA-8159, a new erectrogenic, in mice, rats, 25 Kreisel W, Deibert P, Kupcinskas L, Sumskiene J, Appenrodt B, Roth S, Neagu M, rabbits and dogs, and prediction of human pharmacokinetics. Biopharm Drug Dispos Rössle M, Zipprich A, Caca K, Ferlitsch A, Dilger K, Mohrbacher R, Greinwald R, 2005; 26:269–277. Sauerbruch T. The phosphodiesterase-5-inhibitor udenafil lowers portal pressure in 12 Shim HJ, Kim YC, Park KJ, Kim DS, Kwon JW, Kim WB, Lee MG. Pharmacokietics of compensated preascitic liver cirrhosis. A dose-finding phase-II-study. Dig Liver Dis DA-8159, a new electogenic, after intravenous and oral administration to rats: Hepatic 2014; 47:144–150. and intestinal first-pass effects. JPharmSci2003; 92: 2185–2195. 26 Fink HA, Mac Donald R, Rutks IR, Nelson DB, Wilt TJ. Sildenafil for male erectile 13 Buser PT, Wikman-Coffelt J, Wu ST, Derugin N, Parmley WW, Higgins CB. dysfunction: a systematic review and meta-analysis. Arch Intern Med 2002; 162: Postischemic recovery of mechanical performance and energy metabolism in the 1349–1360. presence of left ventricular hypertrophy. A 31 P-MRS study. Circ Res 1990; 66: 27 Kim BH, Lim HS, Chung JY, Kim JR, Lim KS, Sohn DR, Cho JY, Yu KS, Shin SG, Pick 735–746. JS, Jang IJ. Safety, tolerability and pharmacokinetics of udenafil, a novel PDE-5 14 Zhang Y, Shao L, Ma A, Guan G, Wang J, Wang Y, Tian G. Telmisartan delaysmyocardial inhibitor, in healthy young Korean subjects. B J Clin Pharmacol 2008; 65:848–854. fibrosis in rats with hypertensive left ventricular hypertrophy by TGF-β1/Smad signal 28 Michelakis ED, Tymchak W, Noga M, Webster L, Wu XC, Lien D, Wang SH, Modry D, pathway. Hypertens Res 2014; 37:43–49. Archer SL. Long-term treatment with oral sildenafil is safe and improves functional 15 Drazner MH. The progression of hypertensive heart disease. Circulation 2011; 123: capacity and hemodynamics in patients with pulmonary arterial hypertension. Circula- 327–334. tion 2003; 108:2066–2069. 16 Vandenwijngaert S, Pokreisz P, Hermans H, Gillijns H, Pellens M, Bax NA, Coppiello G, 29 Guazzi M, Samaja M, Arena R, Vicenzi M, Guazzi MD. Long-term use of sildenafilinthe Oosterlinck W, Balogh A, Papp Z, Bouten CV, Bartunek J, D'hooge J, Luttun A, Verbeken therapeutic management of heart failure. J Am Coll Cardiol 2007; 50:2136–2144. E, Herregods MC, Herijgers P, Bloch KD, Janssens S. Increased cardiac myocyte PDE5 30 Park JS, Lim HJ, Cho YJ, Lee JH, Yoon HI, Lee CT. Udenafil improves exercise capacity levels in human and murine pressure overload hypertrophy contribute to adverse LV in patients with chronic obstructive pulmonary disease: a prospective study. COPD remodeling. PLoS ONE 2013; 8: e58841. 2012; 9:499–504. 17 Hassan MA, Ketat AF. Sildenafil citrate increases myocardial cGMP content in rat heart, 31 Li YY, Feng Y, McTiernan CF, Pei W, Moravec CS, Wang P, Rosenblum W, Kormos RL, decreases its hypertrophic response to isoproterenol and decreases myocardial leak of Feldman AM. Downregulation of matrix metalloproteinases and reduction in collagen creatine kinase and troponin T. BMC Pharmacol 2005; 5:10. damage in the failing human heart aftersupport with left ventricular assist devices. 18 Moore KW, de Waal Malefyt R, Coffman RL, O'Garra A. Interleukin-10 and the Circulation 2001; 104: 1147–1152. interleukin-10 receptor. Annu Rev Immunol 2001; 19:683–765. 32 Jiang Y, Reynolds C, Xiao C, Feng W, Zhou Z, Rodriguez W, Tyaqi SC, Eaton JW, Saari 19 Adamopoulos S, Parissis JT, Paraskevaidis I, Karatzas D, Livanis E, Georgiadis JT, Kang YJ. Dietary copper supplementation reverses hypertrophic cardiomyopathy Karavolias G, Mitropoulos D, Degiannis D, Kremastinos DT. Effects of growth hormone induced by chronic pressure overload in mice. J ExpMed 2007; 204:657–666. on circulating cytokine network, and left ventricular contractile performance and 33 Shah AM, Mann DL. In search of new therapeutic targets and strategies for heartfailure: geometry in patients with idiopathic dilated cardiomyopathy. Eur Heart J 2003; 24: recent advances in basic science. Lancet 2011; 378:704–712. 2186–2196. 34 Li Y, Jiang H, Ruan C, Zhong J, Gao P, Zhu D, Niu W, Guo S. The interaction of 20 Gullestad L, Aass H, Fjeld JG, Wikeby L, Andreassen AK, Ihlen H, Simonsen S, transient receptor potential melastatin 7 with macrophages promotes vascular adven- Kjekshus J. Immunomodulating therapy with intravenous immunoglobulin in patients titial remodeling in transverse aortic constriction rats. Hypertens Res 2014; 37:35–42. with chronic heart failure. Circulation 2001; 103:220–225. 35 Kim KH, Kim HK, Hwang IC, Lee SP, Cho HJ, Kang HJ, Kim YJ, Sohn DW. ULTIMATE- 21 Li YY, McTiernan CF, Feldman AM. Interplay of matrix metalloproteinases, tissue SHF trial [UdenafiL Therapy to Improve symptoMAtology, exercise Tolerance and inhibitors of metalloproteinases and their regulators in cardiac matrix remodeling. hEmodynamics in patients with chronic systolic heart failure]: study protocol for a Cardiovasc Res 2000; 46:214–224. randomized, placebo-controlled, double-blind trial. Trials 2013; 14: 188.

Supplementary Information accompanies the paper on Hypertension Research website (http://www.nature.com/hr)

Hypertension Research