Apelin: A Peptide Involved in Cardiovascular Risk in Hemodialysis Patients?

Abstract

Inflammation, oxidative stress, and obesity are important features associated with pathogenesis of cardiovascular disease, a major contributor to the mortality of hemodialysis (HD) patients. Apelin is an adipokine involved in a variety of physiological functions; however, little is known about apelin in chronic kidney disease (CKD). Thus, the purpose of this study was to analyze apelin plasma levels in HD patients and verify whether there is any relationship with inflammation, oxidative markers, and obesity. Twenty-four HD patients [53.6 ± 14.4 years, 14 men, and body mass index (BMI) of 25.0 ± 4.2 kg/m²] were studied and compared with 15 healthy subjects (51.3 ± 13.5 years, 7 men, and BMI of 26.3 ± 3.7 kg/m²). Plasma apelin-12 and -36 were measured using the enzyme immunometric assay method. Plasma electronegative low-density lipoprotein [LDL(–)] levels were measured using ELISA method, and tumor necrosis factor-α, interleukin-6, leptin, and plasminogen activator inhibitor-1 levels were measured by a multiplex assay kit. C-Reactive protein (CRP) was determined by immunoturbidimetry. Anthropometric data were also evaluated. There was no difference between apelin-36 levels in HD patients (0.82 ± 0.60 ng/mL) and healthy subjects (0.83 ± 0.23 ng/mL). In contrast, apelin-12 levels were significantly higher in patients (0.34 ± 0.15 ng/mL vs. 0.24 ± 0.13 ng/mL in healthy subjects). TNF-α, CRP, and LDL(–) levels were higher in patients; however, there was no correlation among apelin-12 or -36 and inflammatory or oxidative markers. The adiposity parameters were also not associated with apelin-12 or -36. In conclusion, plasma apelin seems to be not associated with cardiovascular risk in HD patients.

• apelin
• inflammation
• oxidative stress
• hemodialysis

INTRODUCTION

Apelin, a newly discovered peptide, is known as an endogenous ligand of the human G-protein-coupled receptor APJ.¹ Apelin is synthesized as a preprotein of 77 amino acids that is sequentially cleaved into at least four circulating active peptides, apelin-36, -17, -13, and -12.² In cells, the order of potency in inhibiting forskolin-induced increased adhenosine monophosphate cyclic (AMPc)is inversely proportional to peptide size. Thus, apelin-12 seems to be the most potent apelin peptide.³

Apelin and its receptor are widely distributed in kidneys, heart, lung mammary gland, and central nervous system.⁴ This peptide seems to exert positive effects on cardiovascular physiology^5,6 and insulin sensitivity.^7,8 Besides, in obese and insulin-resistant mice, the peripheral administration of apelin is involved in the reduction of body weight and adiposity.^9,10

In contrast to anti-obesity and anti-diabetic properties, apelin also has been related to oxidative stress and inflammation. Daviaud et al.¹¹ demonstrated a tight positive correlation between apelin and tumor necrosis factor-α (TNF-α) expression in adipose tissue of massively obese individuals. In 2007, García-Díaz et al.¹² observed a possible role for apelin in increased adiposity, insulin resistance, liver oxidative stress, and inflammation. Hashimoto et al.¹³ observed that apelin may be a critical factor in atherogenesis under high-cholesterol dietary conditions. Recently, Melgar-Lesmes et al.¹⁴ related that inflammatory factors could play a role in the activation of the hepatic apelin system. Then, apelin–APJ system can be a mediator of oxidative stress in many tissues.

Systemic inflammation and oxidative stress are associated with progressive cardiovascular disease in chronic kidney disease (CKD)^15,16 and, in those patients undergoing hemodialysis (HD), oxidative stress seems to be the link between inflammation and cardiovascular risk.¹⁷ To date, no study measured isolated apelin-12 and its possible roles in HD patients. Thus, this study was performed to investigate the plasma apelin-36 and -12 concentrations in HD patients and their relationships with oxidative stress, inflammation, and obesity.

MATERIALS AND METHODS

Twenty-four HD patients treated at a private clinic (RenalCor Clinic) in Rio de Janeiro, Brazil, were studied and compared with 15 healthy individuals. The inclusion criteria were patients on HD for at least 6 months and age between 18 and 75 years. Patients with vascular access by catheter, inflammatory diseases, acute illness, heart failure, or known malignancies were excluded.

The etiology of CKD was hypertensive nephrosclerosis (9), diabetic nephropathy (7), chronic glomerulonephritis (3), and other diseases or unknown cause (5). To obtain all relevant clinical data, a researcher reviewed each patient’s medical chart. As antihypertensive medication, five patients (20.8%) received angiotensin-converting enzyme (ACE) inhibitors and one patient received calcium channel blocker. The average arterial blood pressure at the end of HD session was 122/74 mmHg. The dialysis sessions were 3.5–4.5 h three times per week, with a blood flow greater than 300 mL/min, a dialysate flow around 600 mL/min, and bicarbonate buffer.

The Ethics Committee of the School of Medicine of Universidad Federal Fluminense approved the study protocol. The HD patients and healthy subjects were aware of the study and signed an informed consent after reading required documents.

Nutritional Assessment

Anthropometric measurements were obtained immediately after the HD session by a trained researcher. Body mass index (BMI) was calculated as body weight divided by squared stature. The International Society for Renal Nutrition and Metabolism recommends that a BMI lower than 23 kg/m² is a marker of protein energy wasting.¹⁸ Triceps skinfold (TSF) was measured to the nearest millimeter using a Lange Skinfold Caliper (Beta Technology Incorporated, Cambridge, MA, USA). TSF and upper mid-arm circumference (MAC) were measured on the nonfistula side using standard techniques.¹⁹ Mid-arm muscle area (MMA) was calculated using the following equation: MMA = [[MAC (cm) – π × TSF (cm)]²/4π] – n, where n = 10 for male and 6.5 for female.²⁰ The MMA values less than the 15th percentile were considered as muscle wasting.²¹

Waist circumference (WC) was measured at a level midway between the lowest lateral border of the ribs and the uppermost lateral iliac crest. High WC values were classified according to NCEP guidelines²² (≥94 cm for men and ≥80 cm for women). Abdominal fat deposition was assessed by means of a conicity index (Ci), which estimates fat accumulation in the abdomen as the deviation of body shape from a cylindrical toward a double-cone shape. The Ci was calculated using the equation proposed by Valdez:²³ Ci = WC/[0.109 × square root of (weight × height)].

Body fat percentage (%BF) of HD patients was evaluated by dual-energy X-ray absorptiometry (Prodigy Advance Plus, Lunar Corp., Madison, WI, USA). The %BF values were classified according to Lohman et al.¹⁹ (above of mean, 16–24% for men and 24–31% for women; metabolic derangements associated with obesity, ≥25% for men and ≥32% for women).

Biochemical Variables

Blood samples were obtained from an arterial line before the start of the HD session after the patients had fasted overnight, and plasma was immediately frozen at −80°C until analyzed. Apelin-36 and -12 were measured by microplate EIA assay kit (Phoenix Pharmaceuticals, Burlingame, CA, USA).

Plasma TNF-α, plasminogen activator inhibitor type-1 factor, and leptin were measured by a multiplex assay kit manufactured by R&D Systems® (Minneapolis, MN, USA) and high-sensitivity C-reactive protein (hsCRP) by immunoturbidimetry. Electronegative low-density lipoprotein, an important oxidative stress marker, was determined as described by Lobo et al.²⁴

Plasma albumin, blood urea nitrogen (BUN), and creatinine were measured using standard laboratory methods. The adequacy of dialysis treatment was estimated by single-pool Kt/V (Kt/V_sp).²⁵

Statistical Analysis

All data were analyzed for normality of distribution using the Kolgomorov–Sminorv test. The data are expressed as mean ± SD (standard deviation) or medians (interquartile range), as applicable. The Student’s t or Mann–Whitney tests were used to compare means. The associations between variables were measured by Pearson or Spearman correlations. Statistical significance was accepted as p < 0.05, and the statistical analyses were performed through the SPSS 17.0 program (SPSS Inc., Chicago, IL, USA).

RESULTS

Demographic and anthropometric characteristics of HD patients and healthy subjects are summarized in Table 1. The time on dialysis and Kt/V_sp were 74.3 ± 39.3 months and 1.49 ± 0.22, respectively. Mean serum albumin was 3.77 ± 0.33 g/dL and four HD patients (16.6%) had albumin <3.8 g/dL. The BUN and creatinine values were 146.3 ± 28.8 mg/dL and 10.9 ± 2.0 mg/dL, respectively.

Table 1.  Demographic and anthropometric characteristics of hemodialysis patients and healthy subjects.

	HD patients	Healthy subjects
Sample size (n)	24	15
Age (years)	53.6 ± 14.4	51.3 ± 13.5
Male gender (n)	14	7
BMI (kg/m^2)	25.0 ± 4.2	26.3 ± 3.7
MMA (cm^2) – males	40.6 ± 12.1	46.8 ± 3.6
MMA (cm^2) – females	32.5 ± 11.0	37.9 ± 5.6
WC (cm) – males	97.4 ± 13.3	98.7 ± 8.9
WC (cm) – females	85.3 ± 15.7	80.2 ± 9.2
Conicity index	1.32 ± 0.1	1.22 ± 0.1
Body fat (%) – males	28.5 ± 12.5	27.5 ± 5.6
Body fat (%) – females	34.6 ± 8.6	31.0 ± 5.7

Notes: HD, hemodialysis; BMI, body mass index; MMA, mid-arm muscle area; WC, waist circumference.

*p < 0.05.

In HD group, nine (37.5%) patients presented BMI < 23.0 kg/m² but, according to MMA, 50% presented muscle wasting. Fourteen (58.3%) patients had %BF compatible with metabolic derangements associated with obesity, and 11 (45.8%) patients had high WC values. The Ci values were higher in HD patients when compared with healthy subjects.

The apelin-36 levels were not different between HD patients and healthy subjects. However, the HD patients presented higher apelin-12 levels when compared with healthy subjects (Table 2). In HD patients, apelin-36 and -12 were not different according to gender, BMI, %BF, or WC, but it is possible to observe a trend for lower apelin-36 values on high BMI, central obesity, and high fat mass (Table 3).

Table 2.  Apelin, inflammation, and oxidative markers levels in hemodialysis patients and healthy subjects.

	HD patients	Healthy subjects
Apelin-12 (ng/mL)	0.34 ± 0.15	0.24 ± 0.13
Apelin-36 (ng/mL)	0.82 ± 0.6	0.83 ± 0.23
TNF-α (pg/mL)	5.4 ± 1.3	2.2 ± 1.3
CRP (mg/dL)	0.15 (0.33)	0.02 (0.13)
IL-6 (pg/mL)	3.14 ± 1.0	2.72 ± 0.27
Leptin (ng/mL)	9.8 (39.8)	5.2 (7.0)
PAI-1 (ng/mL)	7.9 ± 2.9	6.9 ± 2.2
LDL(−) (U/L)	0.19 ± 0.13	0.10 ± 0.06

Notes: HD, hemodialysis; TNF-α, tumor necrosis factor-alpha; CRP, C-reactive protein; IL-6, interleukin-6; PAI-1, plasminogen activator inhibitor-1; LDL(−), electronegative LDL.

*p < 0.05.

Table 3.  Apelin levels in hemodialysis patients according to various clinical conditions.

	Apelin-36 (pg/mL)		Apelin-12 (pg/mL)
	Yes	No	Yes	No
Gender (male)	0.87 ± 0.69	0.75 ± 0.38	0.32 ± 0.15	0.37 ± 0.14
Diabetes	0.88 ± 0.64	0.77 ± 0.56	0.30 ± 0.19	0.35 ± 0.13
BMI > 23 kg/m^2	0.70 ± 0.33	0.98 ± 0.87	0.34 ± 0.12	0.34 ± 0.19
Central obesity	0.65 ± 0.33	1.06 ± 0.89	0.33 ± 0.13	0.37 ± 0.19
High body fat mass	0.77 ± 0.50	0.99 ± 0.89	0.34 ± 0.12	0.39 ± 0.27
Inflammation	0.79 ± 0.44	0.73 ± 0.57	0.36 ± 0.19	0.32 ± 0.14

Notes: BMI, body mass index.

Central obesity, waist circumference ≥94 cm for men and ≥80 cm for women²²; high body fat mass, % body fat ≥25% for men and ≥32% for women¹⁹; inflammation, C-reactive protein >0.3 mg/dL.

Significant correlation was not found among the levels of apelin-36 and -12 and inflammation and oxidative markers. However, when the HD patients and healthy subjects were treated as a whole group, the apelin-12 levels positively correlated with leptin (r = 0.36; p = 0.02); however, the apelin levels did not correlate with adiposity markers (WC, %BF, and Ci).

DISCUSSION

Although apelin seems to be a beneficial peptide,²⁶ some studies have related it with inflammatory and oxidative status.^11,12,14 In HD patients, both inflammation and oxidative stress are associated with premature death from cardiovascular disease.¹⁶ However, as presented in this study, apelin-36 and -12 are not involved in inflammation or oxidative stress in these patients.

There are only six studies about the apelin levels in HD patients. Recently, Malyszko et al.²⁷ related that apelin levels were higher in patients with proximal arteriovenous fistula (AVF) than those with distal AVF, probably because the arm localization of AVF might contribute to chronic heart failure in HD patients. In 2006, the same work group showed the apelin levels are lower in HD patients with coexisting coronary artery disease.²⁸ El-Shehaby et al.²⁹ also observed that apelin was related to echocardiographic features and its levels were lower in HD patients when compared with healthy individuals. However, Codognotto et al.³⁰ found lower apelin-36 levels in HD patients with dilated cardiomyopathy (DCM) when compared to those with idiopathic DCM and normal glomerular filtration rate, indicating that the uremic status, and not heart involvement, was the determinant for apelin reduction.

On the contrary, Zhang et al.³¹ observed higher apelin-13 levels in HD patients (without heart failure) when compared with healthy subjects. This work also showed that predialysis apelin-13 was positively correlated with asymmetric dimethyl-arginine, a major endogenous inhibitor of nitric oxide synthesis. In contrast, our results showed that apelin-36 levels were similar between HD patients and healthy subjects. Malyszko et al.³² also did not observe differences between apelin levels in HD (and peritoneal dialysis) patients and healthy individuals.

Many factors can contribute to the plasma apelin concentrations. In HD patients, specifically, the apelin removal by dialysis must be considered. El-Shehaby et al.²⁹ related no difference in apelin levels before and after dialysis session. However, Zhang et al.³¹ showed that apelin-13 can be removed by dialysis procedure. Interestingly, in our study, the apelin-12 levels (and not apelin-36) were higher in HD patients when compared with healthy subjects. Thus, apelin-12 concentration may be reduced after HD probably, because of the small molecular size of this apelin isoform.

In this context, the apelin production/release is also important. The endothelium of cardiac and pulmonary vasculature seems to be an important source of plasma apelin.³³ Thus, the vascular endothelial dysfunction by uremia can be involved in the altered apelin levels in HD patients.²⁹ Although in our study apelin was not associated with adiposity markers, adipose tissue is also related to apelin production and secretion.³⁴ Furthermore, Daviaud et al.¹¹ observed a parallel expression between apelin and TNF-α in human and mouse adipocytes. In addition, administration of an APJ antagonist decreases inflammatory cytokines including TNF-α.³⁵ In the gastrointestinal tract, inflammation also increases apelin expression.³⁶

In HD patients, the association between apelin and inflammatory status are dissonant. In fact, Malyszko et al.²⁷ found a positive correlation between apelin and CRP. In another study, this same research group³² showed a positive correlation between apelin and visfatin, an important pro-inflammatory mediator. However, El-Shehaby et al.²⁹ found an inverse association between apelin and interleukin-6 and CRP. On the other hand, in our study, apelin-36 and -12 were not correlated with inflammatory markers, in agreement with Codognotto et al.³⁰

Obesity is related to inflammation and hypoxia, two factors able to induce expression and secretion of apelin by both humans and murine adipocytes.^37,38 However, curiously, apelin can also have opposite effects on adiposity. The apelin peripheral administration in obese and insulin-resistant mice is able to reduce adiposity and body weight,^9,10 but apelin intracerebroventricular administration in standard mice was associated with increased adiposity.^39,40 Thus, it is difficult to establish if this lack of association between apelin and inflammation is due to the interaction between any apelin isoform and adipose tissue.

Additionally, the oxidative stress can also be modulated by apelin. The blockade of renin–angiotensin system ameliorates apelin expression and secretion and, consequently, prevents excessive lipid accumulation and the generation of reactive oxygen species (ROS) in differentiating adipocytes.⁴¹ Apelin also behaves as a catalase activator in cardiomyocytes.⁴² In HD patients, several factors are involved in ROS generation, such time on HD, uremia, and bioincompatibility of membranes and solutions.⁴³ Thus, the apelin modulation cannot be sufficient to contribute for improving oxidative status.

Our study has some limitations. First, the cross-sectional nature of our study does not permit to infer causative process between apelin and obesity or inflammation. Second, apelin can be degraded by ACE-related carboxypeptidase, an enzyme homolog of ACE,⁴⁴ and 20.0% of patients were receiving ACE inhibitors.

In summary, different apelin isoforms can be determined. Many studies measured the total apelin or a mix of apelin-12, -13, and -36. Our study was the first to determine apelin-12, apparently the most potent apelin peptide,³ in HD patients. However, no relationship was found between the apelin peptide and obesity, inflammatory, or oxidative markers. Therefore, further studies are needed to investigate the role of apelin in HD patients.

ACKNOWLEDGMENT

This study was supported by Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), and Unidade de Pesquisa Clínica (UCP)–HUAP-UFF.

Declaration of interest:The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.