EURO PEAN SO CIETY O F CARDIOLOGY ®

Original scientific paper

Exercise rescued chronic kidney disease by attenuating cardiac hypertrophy through the cardiotrophin-1 ! LIFR/gp 130 ! JAK/ STAT3 pathway

European Journal of Preventive Cardiology 2014, Vol. 21(4) 507–520 ! The European Society of Cardiology 2012 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/2047487312462827 ejpc.sagepub.com

Kuan-Chou Chen1, Chiu-Lan Hsieh2, Chiung-Chi Peng1 and Robert Y Peng3

Abstract Background: Chronic kidney disease (CKD) is usually associated with cardiac apoptosis and/or cardiac hypertrophy. We hypothesized that exercise can reduce the CKD-induced cardiac damage. Methods and results: The doxorubicin-induced CKD (DRCKD) model was used in rats to compare two exercise models: 60-min running and 60-min swimming. Results indicated that in healthy normal groups, the signals cardiotrophin1 (CT-1), interleukin 6 (IL-6), leukaemia inhibitory factor receptor (LIFR), and gp130 were upregulated and janus kinase (JAK) and signal transducer and activation of transcription (STAT) were downregulated by both exercises. In contrast, all signals were highly upregulated in CKD. After exercise training, all signals (CT-1, IL-6, LIFR, gp130, and STAT) were downregulated, with JAK being only slightly upregulated in the running group but not in the swimming group. The myocyte death pathway (CT-1/IL-6 ! LIFR/gp130 ! PI3K ! Akt ! Bad) was excluded due to no change found for Bad. Nitric oxide (NO; normal, 15.63  0.86 mmol/l) was significantly suppressed in CKD rats (2.95  0.32 mmol/l), and both running and swimming training highly upregulated the NO level to 30.33  1.03 mmol/l and 27.82  2.47 mmol/l in normal subjects and 24.0  3.2 mmol/l and 22.69  3.79 mmol/l in the DRCKD rats, respectively. The endothelial progenic cells CD34 were significantly suppressed in DRCKD rats, which were not rescued significantly by exercise. In contrast, the CD 34 cells were only slightly suppressed in the healthy subjects by exercise. Conclusion: Both exercise regimens were beneficial by rescuing cardiac function in CKD victims. Its action mechanism was by way of inhibiting myocyte death and rescuing cardiac hypertrophy.

Keywords Cardiac hypertrophy, chronic kidney disease (CKD), doxorubicin Received 4 June 2012; accepted 7 September 2012

Introduction The level of kidney dysfunction is an independent risk factor for cardiovascular disease (CVD) and cardiorenal syndrome (CRS).1,2 Prolonged anaemia associated with CKD causes left ventricular hypertrophy.3–6 The reduction of CVD risk factors has been suggested to be the best strategy for CKD patients.7 CRS usually involves the prevalence of coronary heart disease, congestive heart failure (CHF), CVD, dilated cardiomyopathy, and coronary artery disease.3–10 Both CHF and CKD usually exhibit states of chronic inflammation with elevated levels of circulating inflammatory mediators.11,12 Animal studies on CHF

show increased mRNA expression of the pro-inflammatory cytokines tumour necrosis factor alpha (TNFa) and interleukin (IL) 6 and 1b in myocardial cells and cardiac blood vessels.13 TNFa and IL-1b also contribute structurally to adverse ventricular remodelling in 1

Taipei Medical University, Taipei, Taiwan National Changhua University of Education, Changhua City, Taiwan 3 Hungkuang University, New Taichung City, Taiwan 2

Corresponding author: Chiung-Chi Peng, Graduate Institute of Clinical Medicine, College of Medicine, Taipei Medical University, 250 Wu-shing St., Taipei, 11031, Taiwan. Email: [email protected]

508 CHF by increasing apoptosis14–17 and disrupting the vasomotor balance by enhancing nitric oxide (NO) degradation.18,19 Venous congestion also has triggered markers of inflammation such as cyclooxygenase-2 (COX-2) and inducible nitric oxide synthase (iNOS).20–23 Cardiotrophin-1 (CT-1) shares the glycoprotein (gp) 130 receptor and has been implicated as a factor involved in myocardial remodelling.24,25 CT-1 has been shown involved in many pathological changes typical of cardiovascular diseases.25 Higher level of plasma CT1 in human is reported to be with hypertension, more elevated in CKD with left ventricular hypertrophy and aortic stenosis,26–29 and playing an important role in the structural remodelling of CHF.30 Doxorubicin (DR) has been used in many CKD/ CRS models.31–34 DR is a well-known cytotoxic drug causing cardiomyopathy, but the experimental dose required to induce CKD is far less than that usually required for the induction of DR cardiomyopathy,34 i.e. the factors relating to the occurrence of diffuse cardiac fibrosis would be minimum in this present paper. Under such a consideration, both cardiac and renal dysfunctions resulting from DR administration are characterized by many features observed in humans with renocardiac failure – both on the functional as well as on the histological level.34,35 Epidemiologically, CVD is already well established at the onset of ESRD.36–38 The indications could involve a reduced muscle mass and a significant decrease in exercise tolerance.38 Muscle function can be affected by a number of direct and indirect treatments,39 and the risk of CVD in CKD can be reduced by many physical activities.40 Patients receiving rehabilitation after myocardial infarction have improved 20–25% of the pathological symptoms compared with those without rehabilitation.40 In CAD patients, exercise improves the content of skeletal muscle to allow a more efficient use of oxygen.40 We proposed that exercise may ameliorate cardiac hypertrophy through the CT-1 ! LIFR/gp130 ! JAK/ STAT3 pathway rather than the phosphatidylinositol 3-OH kinase (PI3K)/Akt ! Bad pathway. We carried out two types of exercise programmes: treadmill running and swimming exercises to confirm our hypothesis.

Materials and methods Chemicals and kits Collagen was assayed by Sirius Red staining (SigmaAldrich, USA). Doxorubicin was a product of Pfizer (Milano, Italia). pro-PREP lysis buffer was purchased from iNTRON Biotechnology (Seongnam, Korea). The kit sources for other determinations included SOD and

European Journal of Preventive Cardiology 21(4) TBARS from Cayman (MI, USA), Rat IL-6 EIA KIT from PeproTech (NJ, USA). The sources of antibodies used in this experiment were: LIFR (1:500), phosphoPI3K (1:1000) and gp130 (1:500) from Santa Cruz (CA, USA); phospho-Akt (1:1000) from Cell Signaling (MA, USA); Bad (1:1000), phospho-STAT (1:500), phospho-JAK (1:500) from Epitomics (CA, USA); CD34 (1:400) from Leica (Germany); b-actin and chemiluminescent HRP substrate from Novus Biologicals (CO, USA).

Experimental design The experimental design flowchart for this paper is shown in Figure 1. The doxorubicin-induced CKD (DRCKD) model was adopted to induce CKD, which as mentioned could indirectly cause cardiac damage.33,34 To determine whether exercise can be beneficial in alleviating cardiac damage, we examined the PI3K ! Akt ! Bad survival pathway, the CT1 ! IL-6 ! LIFR/gp130 ! JAK ! STAT3 hypertrophy pathway, and the oxidative stress-antioxidative pathway that involves NO, superoxide dismutase (SOD), and malondialdehyde (MDA). Two types of exercises – treadmill running and swimming – were examined.

Animal CKD model This experimental protocol was approved by the China Medical University Institutional Animal Care and Use Committee (IACUC; Taichung, Taiwan). Principles of laboratory animal care (NIH) were followed. Thirty-six 4-week-old Sprague-Dawley male rats (BioLASCO Taiwan) weighing 225–250 g were used in the study. These rats were acclimated during the first week. The rats were housed in animal room maintained at a relative humidity 50–60% at 23  1 C with a 12/12 light/ dark cycle. The animals were allowed free access to water and ordinary laboratory pellet chow containing 1.8–2.2% calcium, 1.1% phosphorus, and 2650 kcal/kg energy. These rats were randomly assigned to six groups: (1) normal sedentary, (2) normal 60-min running, (3) normal 60-min swimming, (4) doxorubicininduced CKD (DRCKD) sedentary, (5) DRCKD 60min running, and (6) DRCKD 60-min swimming groups. These six groups were separately housed in 12 colony cages, three rats in each. CKD was induced by a single s.c. injection of DR 7.5 mg/kg.32

Exercise protocols Treadmill exercise. Initially, pre-exercise training was tried by increasing gradually the running exercise from 5 to 10 min, and then to 50 min (30 m/min) and

Chen et al.

509

Doxorubicin

CKD Exercises GFR

Heart weight Cardiomyopathy Body weight

Blood test Exercises Serum creatinine

Heart wt/Body wt

Heart tissues Sirus stain CD34

Survival pathway

Hypertrophy pathway

P-PI3K

CT-1

P-Akt

IL-6

Bad

LIFR/gp130 JAK

Oxidative stress/ Antioxidant

NO SOD

Serum albumin

Urinary protein Urinary BUN Urinary creatinine

Serum Cholesterol Serum triglyceride BUN

MDA

STAT3

Figure 1. Experimental design flow chart.

administered with DR after a 2-week pre-exercise training. From the fourth week onwards, the rats were subjected to formal running exercise 60-min/day, 3 days/ week for a total period of 11 weeks. During the whole course, the sedentary were remained in the cages under the same environmental condition and inspected daily.32 Swimming exercise. In the second week, the pre-swimming exercise acclimation was started in an experimental swimming pool (30 C, water depth 44 cm, radius 120 cm). A gradual progression protocol was applied beginning with swimming for 5–10 min, and then gradually extended to 50 min/day. CKD was induced after the pre-swimming training. From the fourth week on, the rats were subjected to formal swimming exercise 60min/day, 3 days/week for a total period of 11 weeks.

Express Plus; Ciba-Corning, USA). The urinary protein concentration was measured using a ELISA reader (EZ Read 400 Microplate Reader, Biochrom, UK). Blood collection. After urine collection, the blood samples were immediately withdrawn from the abdominal aorta under either intraperitoneal ketamine and xylasine anaesthesia. The sample blood was centrifuged at 3000 g to separate the serum. The serum obtained was used for measurement of parameters, including albumin, cholesterol, triglyceride, BUN, and creatinine, as above. Tissue collection. After euthanasia, the organs were inspected by vision for external morphological changes. The hearts were desiccated and immediately frozen with liquid nitrogen and stored in 80 C.

Blood, urine, and tissue sample collection

Glomerular filtration rate

Urine collection. On finishing treatment at week 11, rats were moved to the metabolic cage 2 days before the end of week 11. The urine was collected from 8.00 a.m. to 8.00 a.m. of the next day. The total volume of urine/day for each rat was taken. The urine samples were freshly analysed for urinary protein, creatinine, and blood urea nitrogen (BUN) or immediately stored in the freezer at 0–4 C when not in use. Urinary BUN and creatinine were measured by reagents (Siemens, Bakersfield, CA, USA) and an automatic analyser (Ciba-Corning

The glomerular filtration rate (GFR) is typically recorded in units of volume/time, e.g. ml/min, by the expression GFR ¼ (urine concentration  urine flow)/ plasma concentration.41 GFR was measured by method of creatinine clearance (CCr). Briefly, 24-h urine was collected to determine the amount of creatinine that was removed from the blood over a 24-h interval. Ccr was calculated from the creatinine concentration in the collected urine sample (Ucr), urine flow rate (V),

510 and the plasma concentration (Pcr). Since the product of urine concentration and urine flow rate yields creatinine excretion rate, which is the rate of removal from the blood, creatinine clearance is calculated as removal rate/min (Ucr  V) divided by the plasma creatinine concentration.42 This is commonly represented mathematically as Ccr ¼ (Ucr  V)/Pcr.

Histochemical examinations Hearts were fixed by immersion with 10% formalin in PBS (pH 7.4) at 4 C for 24 h and processed for paraffin embedding. Paraffin sections were dewaxed in xylene and rehydrated in a series of ethanol washes. The nuclei of these specimens were subjected to Weigert’s haematoxylin-eosin stain; the capillary cells were immunostained with anti-CD34. Otherwise, the collagen content was stained with Sirius Red. The integrated optical density was measured and a mean value calculated.43 ELISA of IL-6, cardiotrophin, NO, MDA, and SOD. Serum levels of IL-6, cardiotrophin, NO, MDA, and SOD and were measured by the Rat IL-6 EIA Kit (PeproTech, NJ, USA), Cardiotrophin 1 Enzymelinked Immunosorbent Assay Kit (USCN Life Science, Wuhan, PR China), and NO, SOD, and TBARS assay kits (Cayman, Michigan, USA). All ELISA protocols were performed according to the manufacturer’s instructions. Results were obtained using an EZ Read 400 Microplate Reader (Biochrome Ltd., Cambridge, UK). Western blotting. Frozen heart tissue samples (approximately 100 mg) were homogenized with the homogenizer (T10 basic; The IKA Company, Germany) in 1 ml pro-PREP lysis buffer (pH 7.2). The homogenate was centrifuged at 12,000 g for 20 min at 4 C, and the supernatant was collected as tissue sample lysate. The sample protein lysates were heated at 100 C for 10 min before loading and separated on precasted 7.5% SDS-PAGE. The proteins were electrotransferred onto the PVDF membrane in transfer buffer for 1 h. The nonspecific binding to the membrane was blocked for 1 h at room temperature with 5% nonfat milk in TBS buffer. The membranes were then incubated for 16 h at 4 C with various primary antibodies. After extensive washing in TBS buffer, the membranes were incubated with secondary antibody in blocking buffer containing 5% nonfat milk for 1 h at room temperature. Membranes were then washed with TBS buffer and the signals were visualized using an Luminescent Image Analyzer LAS4000 (Fujifilm, Tokyo, Japan). Levels of LIFR, gp130, phospho-STAT, phospho-JAK, PI3K, phospho-PI3K,

European Journal of Preventive Cardiology 21(4) Akt, phospho-Akt, Bad, and b-actin were analysed by immunoassay according as above. b-Actin was used as the reference protein.

Statistical analysis Data obtained in the same group were treated with ANOVA and then analysed by the Duncan’s multiple range tests with SPSS version 10.0 (SPSS, Chicago, IL, USA). Significance of difference was judged by a confidence level of p < 0.05.

Results and discussion Heart weight to body weight ratio was alleviated by exercise The ratio of heart weight/body weight (Rh/b) reached 0.37  0.04, 0.40  0.03, and 0.42  0.06 in the DR sedentary, DRCKD 60-min running, and DRCKD 60-min swimming groups, respectively. The corresponding normal control values were 0.29  0.00 for the sedentary, 0.33  0.01 for the 60-min running, and 0.32  0.03 for the 60-min-swimming, respectively (Figure 2). Relatively, the differences among these Rh/b values were very tiny. Such slight differences among values of Rh/b between the normal and the CKD groups implied that the pathway for cardiac hypertrophy might be prevail over that of myocyte apoptosis, even in victims severely affected by CKD.35 The reason may be ascribed to the fact that normal physiological cardiac hypertrophy is also inducible by exercise. Adams and Vaziri44 reported that voluntary running can result in apparently nonpathological left ventricular hypertrophy in both the sham-operated and CRF rats. Thus the overall ratio change in Rh/b can be affected at least by CKD and/or exercise.

Glomerular filtration rate mostly was restored by exercise The normal glomerular filtration rate is reported to be 0.74  0.03 ml/min/100 g.45 All the GFR data obtained were normalized with respect to this reference value and are shown in Table 1. DR severely damaged renal function, leading to a GFR 0.30  0.02 ml/min/100 g, while the 60-min running and the 60-min swimming exercises were able to rescue most part of such renal dysfunction, although the effect of running programme was slightly greater (Table 1). DR had caused renal dysfunction of approximately 60.0%, while DRCKD 60-min running and DRCKD 60-min swimming groups were suffering renal dysfunction of 12% and 24%, respectively (Table 1), pointing to the beneficial effect

Chen et al.

511

0.6 Normal DRCKD

Heart/Body weight ratio (%)

0.5

d d

c 0.4

b

b a 0.3

0.2

0.1

0.0 Sedentary

60min Run

60min Swim

Group

Figure 2. Ratio of heart weight to body weight affected by treatment with doxorubicin and exercise. Data obtained in the same group were analysed by Duncan’s multiple range tests. Significance of difference was judged by a confidence level of p < 0.05. Different superscripts above each column indicate a significant difference; DRCKD, doxorubicin-induced CKD.

Table 1. Glomerular filtration rate used as the renal function indexing in normal and doxorubicin-induced CKD rats Normal rats

Doxorubicin-induced CKD rats

Items

Sedentary

60-min run

60-min swim

Sedentary

60-min run

60-min swim

Serum creatinine (mg/dl) Urinary creatinine (mg/dl) Glomerular filtration rate (ml/min/100 g)

0.6  0.1a 172  15c 0.74  0.03a

0.5  0.1a 177  14c 0.77  0.02a

0.8  0.1b 175  18c 0.72  0.03a

1.0  0.1c 9  3a 0.30  0.02d

0.7  0.1b 73  11b 0.65  0.02b

1.1  0.1c 78  13b 0.56  0.02c

Values are mean  SD (n ¼ 6); Serum creatinine normal range: 0.2–0.8 mg/dl; Glomerular filtration rate values have been normalized with that reported45; Data obtained in the same group were treated with ANOVA and then by Duncan multiple analysis. Significance level p < 0.05. Different superscripts in each row indicate a significant difference.

of exercise to CKD patients, with running slightly more beneficial.

Serum and urinary biochemical parameters DRCKD reduced serum albumin level from 4.2  0.1 to 2.8  0.0 g/dl, but significantly increased cholesterol (681  65 mg/dl), triglyceride (532  82 mg/dl), urinary protein (163  16 mg/dl), and BUN (247  15 mg/dl), compared with the corresponding control values of 68  11, 47  11, 21  14, and 61  6 mg/dl, respectively (Table 2). Running exercise was shown to ameliorate completely serum cholesterol (72  8 mg/dl) and BUN (21  1 mg/dl) but only partially serum albumin

(3.1  0.4 g/dl), triglyceride (207  28 mg/dl), urinary protein (138  10 mg/dl), and BUN (180  6 mg/dl), and swimming was inferior to running in the extent of its beneficial effects. In contrast, swimming showed a better effect on urinary BUN reduction than running, but the effect was only partial compared with the normal sedentary group (Table 2).

Nitric oxide level In normal groups, the production of NO ranges within 15–20 mmol/l (Figure 3A). Both exercises increased the production of NO to 30.33  1.03 and 27.82  2.47 mmol/l (Figure 3A), corresponding to

512

European Journal of Preventive Cardiology 21(4)

Table 2. Serum and urinary biochemical parameters in normal and doxorubicin-induced CKD rats Normal rats

Serum Albumin (g/dl) Cholesterol (mg/dl) Triglyceride (mg/dl) BUN (mg/dl) Urine Protein (mg/dl) BUN (mg/dl)

Doxorubicin-induced CKD rats

Normal range

Sedentary

60-min run

60-min swim

Sedentary

60-min run

60-min swim

3.8–4.8 40–130 26–145 15–21

4.2  0.1a 68  11d 47  11d 16  2d

4.4  0.2a 53  14e 45  10d 17  1d

4.1  0.1a 50  2e 41  2e 18  2d

2.8  0.0c 681  65a 532  82a 35  2b

3.1  0.4b 72  8c 207  28c 21  1c

3.2  0.3b 348  13b 255  11b 48  3a

21  14e 61  6d

21  5e 50  4e

113  25d 58  16d

163  16b 247  15a

138  10c 180  6b

258  32a 140  14c

– –

Values are mean  SD (n ¼ 6); Data obtained in the same group were treated with ANOVA and then by Duncan multiple analysis. Significance level p < 0.05. Different superscripts in each row indicate a significant difference. BUN: blood urea nitrogen.

(A) 35 30

(B) 25

a

Normal DRCKD

b c

Normal DR-CKD

d

20

b

20

SOD (U/mL)

NO (μΜ)

25 e

15

a b c

c

15 d

10

10 5

5

f

0 Sedentary

60 min Run

0

60 min Swim

Sedentary

Group

60 min Run Group

60 min Swim

(C) 12

MDA (μM)

Normal DR-CKD

a

10 8

b

6 4

c

c

b c

2 0 Sedentaty

60min Run 60min Swim Group

Figure 3. Expression levels of nitric oxide (NO; A), superoxide dismutase (SOD; B), and malondialdehyde (MDA, C) affected by treatment with doxorubicin and exercise. Data obtained in the same group were treated with ANOVA and then by Duncan multiple analysis. Significance level p < 0.05. Different superscripts in each row indicate a significant difference; DRCKD, doxorubicin-induced CKD.

Chen et al. 1.7- and 1.6-fold increases, respectively (p < 0.05). A similar trend was found in the DRCKD groups, NO was also highly upregulated by both exercises to 24.00  3.19 and 22.69  3.79 mmol/l, respectively, compared with the highly suppressed NO in the DRCKD sedentary group (2.95  0.32 mmol/l; p < 0.05; Figure 3A), reaching overall increases of 8.0- and 7.6-fold, respectively. The results were consistent with Husain.46 A recent report has indicated that deficiency of either neuronal nitric oxide synthase (NOS1) or endothelial nitric oxide synthase (NOS3) leads to cardiac hypertrophy in mice.47 A combined deficiency of NOS caused increased mortality, myocyte hypertrophy, and an ageassociated increase in ventricular stiffness.47 The data of the present study evidently showed that exercise training would significantly increase cardiac NO levels and supposedly the NOS activity in rats (Figure 3A). NO regulates both adenylyl cyclase and guanylyl cyclase in the cardiovascular system and regulates the blood pressure (BP).48,49 In the sedentary group, the normal endothelial release of NO through eNOS reaction mediates vasodilation, excess release through iNOS induction may play a role in lowering the BP.46 We showed that the normal NO level ranged 15–20 mmol/l and the level was severely suppressed down to only 2.95  0.32 mmol/l in the DRCKD rats (Figure 3A). The interaction of DR with ferric ions tends to produce ROS,35 which in turn will deplete the NO. Recently, an inhibitory effect of NO with MDA and thiobarbituric acid reactive substance (TBARs) formation has been reported.50 Either physical activity was seen to stimulate huge amount of NO production in rats (Figure 3A). As consequence, the oxidative stress would have been attenuated to a level being relatively low and safe for cardiac health. To summarize, both types of physical conditioning were effectively beneficial in restoring sufficient NO levels to ameliorate the effect of DR (Figure 3A).

Superoxide dismutase activity was effectively restored by exercise Both running and swimming elevated the serum level of SOD to 21.01  0.73 and 19.17  0.53 U/ml, respectively, in the two control groups when compared with 17.20  0.59 U/ml of the sedentary (p < 0.05; Figure 3B). DR suppressed the serum SOD level to 7.56  1.63 U/ml (p < 0.05), which was alleviated, respectively, to 15.54  1.32 and 14.52  1.14 U/ml by running and swimming (p < 0.05; Figure 3B), providing evidence that exercise not only could effectively ameliorate the in vivo oxidative stress of the DRCKD patients, but also strengthen the antioxidative capability of healthy subjects.51

513

Serum malondialdehyde level was only partially suppressed by exercise In the sedentary normal group, the MDA level was very low (2.1  0.6 mmol/l). The MDA levels in the other controls were also rather comparable (Figure 3C). DR stimulated the low-density lipoprotein oxidation to yield huge amount of MDA (8.8  0.9 mmol/l; p < 0.05; Figure 3C). Both exercises effectively reduced the overproduced MDA to 4.7  0.6 and 5.1  0.1 mmol/l, respectively (p < 0.05; Figure 3C). Interestingly, although exercise did not have any effect on the serum MDA levels in the three control groups, but was able to ameliorate the elevation of MDA in DRCKD victims, implying that exercise behaved differently with respect to physiological and pathological status.

Serum levels of cardiotrophin and IL-6 were affected by exercise CT-1 is a IL-6 family cytokine with known protective and hypertrophic effects in the heart.25 CT-1 induces hypertrophy more potently than other members of the IL-6 family of cytokines or other known mediators of cardiomyocyte hypertrophy.52 CT-1 treatment increases levels of heat shock proteins 70 and 90 in cardiac cells, which may be involved in the protective effects of CT-1.53 In the sedentary group, 60-min running and swimming exercises increased CT-1 to 350.5  38.0 and 331.2  24.9 pg/ml. We showed that CT-1 normally ranged 200–250 pg/ml. In normal subjects, both exercises upregulated CT-1 to 350.538.0 and 331.2  24.9 pg/ml, respectively, compared with the control value 232.9  20.9 pg/ml (p < 0.05; Figure 4A), indicating the resistant exercise may elicit beneficial effects through upregulation of CT1 in normal subjects.25 On the contrary, the CT-1 levels in DRCKD sedentary rats, originally ranging highly within 440–500 pg/ml, was suppressed to 238  46 pg/ ml and 27433 pg/ml, respectively, completely by the running exercise, but only almost completely by swimming (Figure 4A). Thus, downregulation of CT-1 could be beneficial to CKD victims and CT-1 signalling could exhibit a physiopathological duality regarding the pathogenesis of cardiac hypertrophy (Figure 4A), consistent with Calabro` et al.25 Early expression of CT-1 in the ischaemic myocardium may represent an adaptive, protective phenomenon that is beneficial in reducing myocyte loss and inducing hypertrophy of remaining myocytes so that overall ventricular function is maintained.54 Moreover, CT-1 is a chemoattractant for cardiac fibroblasts.55 It is worth noting that the sustaining period can be an alternative determinant factor.56 CT-1 expression in

514

European Journal of Preventive Cardiology 21(4)

(A) 600

(B) 1800 a

Normal DR-CKD

e

1600

400

c d

300

e

e

200

1200 1000

Normal DR-CKD

d

1400 b

IL-6 (pg/mL)

Cardiotrophin-1 (pg/ml)

500

c

b ab

a

800 600 400

100

200 0

0 Sedentary

60 min Run

60 min Swim

Group

Sedentary

60min Run

60min Swim

Group

Figure 4. Levels of cardiotrophin-1 (A) and interleukin 6 (IL-6, B) affected by treatment with doxorubicin and exercise. Data obtained in the same group were treated with ANOVA and then by Duncan multiple analysis. Significance level p < 0.05. Different superscripts in each row indicate a significant difference; DRCKD, doxorubicin-induced CKD.

the late phase of the onset of heart failure may contribute to ventricular dilation by inducing eccentric hypertrophy.54 Based on the value Rh/b, more severe hypertrophy was seen in the DRCKD groups (Figure 2). The CT-1 levels were shown to be more downregulated in the running and the swimming groups than the normal exercise groups (1.53- and 1.28-fold, respectively; Figure 4A), pointing to the different interaction between CT-1 and DR. Previously, a significant correlation was found between the plasma CT-1 and log IL-6,52 and similar results are seen in Figure 4A and B. High plasma levels of CT-1, brain natriuretic peptide (BNP), and IL-6 are independent predictors of mortality of chronic heart failure on stepwise multivariate analysis. CT-1 is increased in patients with CHF.57 There is a significant negative correlation between plasma CT-1 and left ventricular ejection fraction.57 The hazard ratio for mortality in patients with plasma BNP >170 pg/ml and CT-1 >658 fmol/ml was 2.48 (95% confidence interval 1.217–5.060) compared with those with plasma BNP >170 pg/ml and CT-1