Associations of Plasma Pentraxin 3 and Monocyte Chemoattractant Protein-1 Concentrations with Cardiovascular Disease in Patients with Chronic Kidney Disease

Abstract

We examined the associations of circulating levels of pentraxin 3 (PTX3), monocyte chemoattractant protein-1 (MCP-1), and some other inflammatory mediators with cardiorenal syndrome. In advanced chronic kidney disease (CKD) patients (estimated glomerular filtration rate <30 mL/min/1.73 m²), the values of area under the curve of PTX3, tumor necrosis factor α, and high-sensitivity C-reactive protein for the detection of the association of cardiovascular disease (CVD) were 0.664, 0.507, and 0.318, respectively. In contrast, serum levels of MCP-1 were significantly higher in CKD patients than in control subjects independently of association with CVD.

Abstract

Both pentraxin 3 (PTX3) and monocyte chemoattractant protein-1 (MCP-1) are mediators of inflammation. They also appear to play critical roles in vascular endothelial dysfunction but their associations with cardiorenal syndrome remain largely unknown. The objective of this study was to examine their associations with cardiorenal syndrome. Circulating levels of PTX3, MCP-1, and some other biomarkers were evaluated in 134 patients with chronic kidney disease (CKD) and/or cardiovascular disease (CVD) and 55 age- and gender-matched subjects without CKD or CVD. Levels of PTX3, high-sensitivity C-reactive protein (hsCRP), and tumor necrosis factor α (TNFα) were significantly higher in CKD patients with CVD than in those without CVD. In advanced CKD patients (estimated glomerular filtration rate <30 mL/min/1.73 m²), the values of area under the curve of PTX3, TNFα, and hsCRP for the detection of the association of CVD were 0.664, 0.507, and 0.318, respectively. In contrast, serum levels of MCP-1 were significantly higher in CKD patients than in control subjects independently of association with CVD. PTX3, hsCRP, and TNFα, but not MCP-1 could predict the presence of CVD as a complication associated with CKD. Additionally, PTX3 might be a more sensitive marker for the association of CVD than hsCRP and TNFα in patients with advanced CKD.

• Chronic kidney disease
• cardiovascular disease
• pentraxin 3
• monocyte chemoattractant protein-1
• cardiorenal syndrome

INTRODUCTION

Cardiovascular morbidity and mortality is greatly increased in patients with chronic kidney disease (CKD). Besides traditional risk factors, including hypertension, hyperlipidemia, and diabetes mellitus, chronic inflammation observed in CKD patients is considered to be one of the important nontraditional risk factors.^1,2 Inflammatory biomarkers such as high-sensitivity C-reactive protein (hsCRP), interleukin-6, tumor necrosis factor α (TNFα), serum amyloid P‐component (SAP), and fibrinogen have been well studied; especially, there are several associations with levels of hsCRP and cardiovascular disease (CVD) outcomes in patients with CKD.^2,3

Although CRP and SAP are the prototypes of the short pentraxin family and mainly produced in the liver, pentraxin 3 (PTX3) is the prototype of the long pentraxin family and produced in peripheral tissues (e.g., endothelial cells, macrophages, T cells, smooth muscle cells) and is modulated partly by inflammatory mediators such as TNFα.^4–7 There are several evidences showing the link between PTX3 levels and ischemic heart disease. PTX3 levels increase rapidly in acute myocardial infarction and exhibit high levels in patients with unstable angina and those undergoing stenting.^8–10 In addition, CKD patients with or without hemodialysis treatment have higher plasma levels of PTX3 than those without CKD.^11,12 There have been several studies reporting PTX3 levels with regard to cardiorenal syndrome. Plasma levels of PTX3 in patients on hemodialysis therapy are higher with coronary artery disease and peripheral artery disease than without these diseases.¹² Moreover, CKD patients with CVD who are not on hemodialysis treatment have higher concentrations of plasma PTX3 levels than their counterparts, and they have higher all-cause and cardiovascular mortality.^12,13 Monocyte chemoattractant protein-1 (MCP-1) is one of the important chemotactic factors for monocyte and is released under inflammatory conditions. MCP-1 is regulated by some stimuli including TNFα^14–16 and is expressed by a variety of inflammatory activated cells such as endothelial cells, smooth muscle cells, and monocytes/macrophages and develops vascular atherosclerotic lesions.¹⁷ An increase in plasma levels of MCP-1 in patients with unstable angina¹⁸ and an independent association between plasma levels of MCP-1 and clinical outcomes in patients with acute coronary syndrome¹⁹ have been reported. On the other hand, circulating levels of MCP-1 are elevated in CKD patients including hemodialysis patients.^20,21 However, unlike PTX3, the association between circulating levels of MCP-1 and cardiorenal syndrome is not well investigated. Both PTX3 and MCP-1 are associated with inflammation and atherosclerosis and their levels increase in both CKD and CVD patients. Thus, these two biomarkers are expected to have predictive value of cardiorenal syndrome, but the association and the role of them in CKD and/or CVD patients have not been well studied. The aim of this study was to investigate the circulating levels of PTX3 and MCP-1 and evaluate which of them could predict the presence of CVD in CKD patients.

MATERIALS AND METHODS

This study was performed at University of Miyazaki Hospital and Miyazaki Konan Hospital. Recruitment of the patients with CKD or CVD occurred from April 2008 through March 2009, who had admitted in the First Department of Internal Medicine, University of Miyazaki Hospital. A total of 134 patients (mean age 61.9 ± 14.8 years, 95 males) with CKD and/or CVD were enrolled. The exclusion criteria were clinical signs of acute inflammatory or infectious disease, malignancy, and patients on dialysis therapy at the time of evaluation. Blood samples of the patients were taken 7.9 ± 4.6 days after their admission.

These participants were classified as suffering from CKD if the estimated glomerular filtration rate (eGFR) was less than 60 mL/min/1.73 m² (N = 86) or if they have proteinuria without decrease of eGFR (N = 21). Primary kidney disease of CKD patients were chronic glomerulonephritis in 31 patients (29.0%), diabetic nephropathy in 9 patients (8.4%), nephrosclerosis in 36 patients (33.6%), and others including secondary renal insufficiency in 15 patients (14.0%) and unknown in 16 patients (15.0%).

A total of 76 patients had CVD at the time of evaluation. Presence of CVD was defined as the following: ischemic heart disease diagnosed in cases where there were coronary artery stenoses by coronary angiography (N = 47, 61.8%), congestive heart failure or systolic dysfunction with ejection fraction below 50% (N = 11, 14.5%), valvular disease that needs surgical treatment (N = 11, 14.5%), history of cardiac arrest or shock (N = 5, 6.6%), aortic aneurysm or dissection (N = 3, 4.0%), stroke (N = 4, 5.3%), peripheral artery disease (N = 5, 6.6%), and a composite of these. The control subjects were matched to the patients with respect to age and gender (N = 55, age 58.6 ± 4.4 years, 44 males) to be used for comparative analyses of biochemical parameters.

All the participants were divided into four groups as follows: CKD(−)/CVD(−) as Group C (control group, N = 55), CKD(−)/CVD(+) as Group V (CVD group, N = 27), CKD(+)/CVD(−) as Group K (CKD group, N = 58), and CKD(+)/CVD(+) as Group KV (CKD group associated with CVD, N = 49).

Blood samples were immediately separated to plasma and serum within 30 min and samples were kept frozen at −80°C until assay. Plasma concentration of PTX3 was measured by enzyme-linked immunosorbent assay (ELISA) method (Perseus Proteomics, Inc., Tokyo, Japan). Serum concentrations of TNFα and MCP-1 were also determined by ELISA method (R&D Systems, Inc., Minneapolis, MN, USA). Plasma hsCRP levels were measured using high-sensitivity assays (Denka Seiken, Tokyo, Japan). eGFR was calculated according to the Japanese Society of Nephrology formula in 2008.²² The ethical committees of University of Miyazaki Hospital and Miyazaki Konan Hospital approved this study and informed consent was obtained from all patients and controls.

Statistical Analysis

Data are expressed as mean ± SD. p < 0.05 was considered to be statistically significant. Comparison between two groups was assessed for continuous variables with the Student's unpaired t-test, Mann–Whitney test, or chi-square test, as appropriate. Differences among three or four groups were analyzed by analysis of variance (ANOVA) using one-way ANOVA, followed by Tukey honestly significant difference (HSD) test as a post hoc test if ANOVA was significant. Pearson product moment correlation was used to determine correlations of eGFR with other variables. For evaluation of the sensitivity and specificity of plasma levels of PTX3, TNFα, and hsCRP as predictor of the presence of CVD in CKD patients, a receiver operator characteristics (ROC) analysis was performed. Multivariate analysis was carried out on age, gender, and variables that were significantly different between CKD patients with and without CVD. Nonnormally distributed variables were log transformed before entering analysis. The statistical analysis was performed using JMP statistical software (version 8.0; SAS Institute, Japan).

RESULTS

Baseline characteristics of control subjects and patients with CKD and/or CVD are shown in Table 1. Plasma levels of PTX3 and serum levels of TNFα, hsCRP, and MCP-1 were significantly higher in patients with CKD and/or CVD compared with control subjects. Overall, in univariate analysis, TNFα (r = −0.49, p < 0.0001), PTX3 (r = −0.26, p < 0.001), hsCRP (r = −0.24, p < 0.001), and MCP-1 (r = −0.38, p < 0.0001) were significantly and negatively correlated with eGFR. Baseline characteristics of the four groups such as Groups C, V, K, and KV were shown in Table 2. As for traditional risk factors of CVD, the percentage of hypertension was significantly higher in the CVD group than in the control group. In the patients with CKD, age, the percentage of diabetes mellitus, and hypertension were significantly higher in patients with CVD than in those without CVD. The value of eGFR was significantly lower in CKD patients with CVD than in those without CVD. Contrary to our expectation, total cholesterol was lower in the subjects with CVD independently of CKD compared with the controls.

Table 1.  Baseline characteristics of subjects without CKD or CVD and with CKD and/or CVD

	Control	CKD and/or CVD	p-Value
N	55	134	NS
Male (%)	80.0	70.9	NS
Age (years)	58.6 ± 4.4	61.9 ± 14.8	NS
BMI (kg/m^2)	23.9 ± 3.9	23.4 ± 3.8	<0.0001
eGFR (mL/min/1.73 m^2)	78.8 ± 14.4	52.8 ± 23.5	NS
T-cho (mg/dL)	203.4 ± 33.5	209.2 ± 68.5	NS
HDL-cho (mg/dL)	57.3 ± 13.2	56.4 ± 16.3	<0.0001
HT (%)	18.2	58.2	<0.01
DM (%)	7.3	24.6	NS
smoking (%)	41.8	50.0	<0.0001
TNFα (pg/mL)	0.98 ± 0.58	3.52 ± 4.70	<0.0001
PTX3 (ng/mL)	2.15 ± 0.93	3.80 ± 2.35	<0.0001
hsCFP (mg/mL)	0.10 ± 0.13	0.45 ± 0.85	<0.01
MCP-1 (pg/mL)	122.4 ± 43.3	146.8 ± 72.8	<0.05

Notes: TNFα, PTX3, hsCRP, and MCP-1 were log transformed before entering analysis. Significantly different (p < 0.05) as assessed by Student's unpaired t-test or chi-square test.

CKD, chronic kidney disease; CVD, cardiovascular disease; BMI, body mass index; eGFR, estimated glomerular filtration rate; T-cho, total cholesterol; HDL-cho, high-density lipoprotein cholesterol; HT, hypertension; DM, diabetes mellitus; TNFα, tumor necrosis factor α; PTX3, pentraxin 3; hsCRP, high-sensitivity C-reactive protein; MCP-1, monocyte chemoattractant protein-1; NS, not significant.

Table 2.  Baseline characteristics of four groups

	CKD(−)			CKD(+)
	CVD(−) Group C N = 55	CVD (+) Group V N = 27	p-Value	CVD (−) Group KN = 58	CVD (+) Group KVN = 58	p-Value
Male (%)	80	81.5	NS	63.8	73.5	NS
Age (years)	58.6 ± 4.4	61.8 ± 13.7	NS	56.1 ± 15.1	68.9 ± 11.9	<0.0001
BMI (kg/m^2)	23.9 ± 3.9	22.5 ± 3.8	NS	23.8 ± 4.1	23.5 ± 3.5	NS
eGFR (mL/min/1.73 m^2)	78.8 ± 14.4	73.7 ± 10.2	NS	54.0 ± 25.2	39.7 ± 17.3	<0.001
T-cho (mg/dL)	203.4 ± 33.5	182.9 ± 32.7	<0.001	232.0 ± 87.0	196.9 ± 47.6	<0.05
HDL-cho (mg/dL)	57.3 ± 13.2	56.9 ± 17.3	NS	58.6 ± 16.3	53.5 ± 15.7	NS
HT (%)	18.2	56.3	<0.001	48.3	69.4	<0.05
DM (%)	7.3	22.2	NS	15.5	36.7	<0.05
smoking (%)	41.8	59.3	NS	43.1	53.1	NS

Notes: Significantly different (p < 0.05) as assessed by Student's unpaired t-test or chi-square test.

CKD, chronic kidney disease; CVD, cardiovascular disease; BMI, body mass index; eGFR, estimated glomerular filtration rate; T-cho, total cholesterol; HDL-cho, high-density lipoprotein cholesterol; DM, diabetes mellitus; HT, hypertension; NS, not significant.

Meanwhile, as for biomarkers, crude data of PTX3, TNFα, MCP-1, and hsCRP are shown in Table 3. Levels of TNFα and PTX3 were significantly higher in the CVD group and the CKD group compared with the control group. They were also higher in the CKD group with CVD than in those without CVD. Levels of hsCRP were not different between the CVD group and the control group, but were significantly higher in the CKD group with CVD than in those without CVD. In contrast, serum levels of MCP-1 were significantly higher in the CKD group than in the control group independently of association with CVD.

Table 3.  Comparison of biomarker levels in the investigated participants

	Group C	Group V	Group K	Group KV
TNFα (pg/mL)	0.98 ± 0.58	5.25 ± 8.56^a	2.43 ± 2.33^a,b	3.87 ± 3.44^a,c
PTX3 (ng/mL)	2.15 ± 0.93	3.84 ± 2.68^a	3.31 ± 1.95^d	4.34 ± 2.53^a,e
hsCRP (mg/dL)	0.10 ± 0.13	0.27 ± 0.37	0.28 ± 0.87	0.74 ± 0.93^a,c,f
MCP-1 (pg/mL)	122.4 ± 43.3	114.2 ± 46.0	156.1 ± 80.1^f,g	153.8 ± 71.9^f,g

Notes: Data were log transformed before entering analysis. Significantly different (p < 0.05) as assessed by HSD test as a post hoc test if ANOVA was significant. Values are expressed as mean ± SD. TNFα, tumor necrosis factor α; PTX3, pentraxin 3; hsCRP, high-sensitivity C-reactive protein; MCP-1, monocyte chemoattractant protein-1.

^aDenotes p < 0.0001 versus group C.

^bDenotes p < 0.001 versus group V.

^cDenotes p < 0.001 versus group K.

^dDenotes p < 0.01 versus group C.

^eDenotes p < 0.05 versus group K.

^fDenotes p < 0.01 versus group V.

^gDenotes p < 0.05 versus group C.

Moreover, considering the degree of eGFR in CKD patients with CVD, levels of TNFα and PTX3 were significantly higher in patients with eGFR < 30 mL/min/1.73 m² than in those with eGFR ≥ 30mL/min/1.73 m² (5.06 ± 2.85 vs. 3.40 ± 3.57 pg/mL, p < 0.01; 5.70 ± 3.36 vs. 3.80 ± 1.91 ng/mL, p < 0.05, respectively) but not hsCRP nor MCP-1 (Figure 1). The ROC curve analysis showed that the area under the curve (AUC) for levels of PTX3 in advanced CKD patients (eGFR < 30 mL/min/1.73 m²) with CVD was 0.664. The plasma PTX3 cutoff value was 4.05 ng/mL, and its sensitivity and specificity were 71.4% and 41.4%, respectively (Figure 2A). On the contrary, AUC for levels of TNFα and hsCRP in advanced CKD patients with CVD were 0.507 and 0.318, respectively (Figure 2A and C). In the multivariate analysis, older age (odds ratio 1.24), greater body mass index (odds ratio 1.62), and greater levels of PTX3 (odds ratio 1.56) were independently associated with the presence of CVD in advanced CKD patients (Table 4).

Figure 1. Circulating levels of TNFα, PTX3, hsCRP, and MCP-1 in CKD patients with CVD with their eGFR ≥ 30 mL/min/1.73 m² (N = 35) and <30 mL/min/1.73 m² (N = 14).

Notes: TNFα, tumor necrosis factor α; PTX3, pentraxin 3; hsCRP, high-sensitivity C-reactive protein; MCP-1, monocyte chemoattractant protein-1; CKD, chronic kidney disease; CVD, cardiovascular disease; eGFR, estimated glomerular filtration rate; NS, not significant.

Figure 2. Receiver characteristic operating curves for differentiating the presence of CVD in advanced CKD patients (eGFR < 30 mL/min/1.73 m²) on the basis of the circulating levels of (A) PTX3, (B) TNFα, and (C) hsCRP (area under the curve = 0.664, 0.507, and 0.318, respectively).

Notes: CVD, cardiovascular disease; CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; PTX3, pentraxin 3; TNFα, tumor necrosis factor α; hsCRP, high-sensitivity C-reactive protein.

Table 4.  Factors independently associated with the presence of CVD in advanced CKD patients (eGFR < 30 mL/min/1.73 m²)

Varriable	Odds ratio	95% CI	p-Value
Age	1.24	1.05–1.74	<0.01
BMI	1.62	1.06–3.55	<0.05
PTX3	1.56	1.01–3.64	<0.05

Notes: Multivariate logistic regression, including gender, age, body mass index, levels of TNFα, PTX3, and hsCRP. CI, confidence interval; BMI, body mass index; PTX3, pentraxin 3.

DISCUSSION

In agreement with previous studies, PTX3 and MCP-1 levels were negatively correlated with eGFR.^13,20 One of the main reasons of this result is probably due to the higher concentration of them in advanced CKD patients. These negative correlations could be explained by inadequate clearance of large molecular weight of them and enhanced synthesis and release upon stimulation in peripheral tissues including renal epithelial cells and mesangial cells.^13,20,23,24 There might be another mechanism that explains our results in the inflammatory point of view.

This study shows that levels of PTX3 but not MCP-1 increase in CKD patients with CVD. Previous studies reported that plasma PTX3 levels were significantly higher in CKD patients with CVD than in those without CVD.¹³ In addition, this study showed that levels of PTX3 were significantly higher in CKD patients with eGFR < 30 mL/min/1.73 m², which corresponds to CKD stages 4 and 5 according to KDOQI Clinical Practice Guidelines for Chronic Kidney Disease than those with eGFR ≥ 30mL/min/1.73 m². It is plausible that similar result was seen in levels of TNFα, as TNFα has been proved to be proinflammatory mediator that upregulates the expression of PTX3. As shown in ROC, in CKD patients with eGFR < 30 mL/min/1.73 m², circulating levels of PTX3 might play a more important role in detecting the presence of CVD than levels of hsCRP, which have been reported to be an independent predictor of cardiorenal syndrome.³ PTX3 is the prototype of the long pentraxin family produced in a variety of tissues and cells at sites of inflammation.^4,25,26 Inflammatory response of PTX3 is closely linked to vascular endothelial dysfunction. Recent studies reported that PTX3 is strongly expressed in atherosclerotic lesions, primarily in monocytes, macrophages, endothelial cells, and smooth muscle cells.^27,28 Moreover, several studies demonstrated that PTX3 is induced in vascular smooth muscle cells by atherogenic-modified low-density lipoprotein and is present in atherosclerotic lesions.^27,29 PTX3 inhibits FGF2-dependent activation of smooth muscle cells.³⁰ Therefore, it could be said that PTX3 plays a role in the defense mechanism of CVD. On the other hand, PTX3 induced in endothelial cells might promote thrombogenesis and vascular ischemia.³¹ In advanced CKD patients, higher concentration of plasma PTX3 might potentiate CVD to occur through these inflammatory responses mentioned above. Our study suggests that in advanced CKD patients, PTX3 might have more information and role in predicting CVD than hsCRP.

On the other hand, unlike PTX3 levels, there was no difference of MCP-1 levels between CKD patients with and without CVD. MCP-1 is expressed in endothelial cells, smooth muscle cells, and macrophages in atherosclerotic lesions and induces endothelial dysfunction, thrombosis, and angiogenesis and develops atherosclerosis and plaque destabilization.^17,32 Elimination of the MCP-1 pathway decreases atheroma formation and inhibits the development of atherosclerosis in mice.³³ These findings support that MCP-1 might progress in a positive feedback loop to enhance atherosclerotic changes at the lesion of vascular inflammation.³² MCP-1 acts as vascular inflammatory mediator by developing atherosclerosis. In addition, uremic toxicity induces the expression of cytokines and adhesion molecules including MCP-1.²⁰ Thus, increasing levels of MCP-1 in CKD patients with CVD could be induced by uremic milieu more than by vascular inflammation. In other words, cardiovascular inflammation has a less effect on serum MCP-1 concentration in CKD patients with CVD.

We report here for the first time the comparison of the levels of PTX3 and MCP-1 in the presence of CVD in CKD patients. Compared with hsCRP, which has been considered to be reliable inflammatory marker of CVD in CKD patients, the higher levels of PTX3 might indicate the comorbid condition of CVD in advanced CKD patients or in patients with CKD stages 4 and 5. On the other hand, the comorbid condition of CVD in CKD patients does not affect the levels of MCP-1.

ACKNOWLEDGMENTS

We express appreciation to Drs. Hiroko Fukae and Yurika Kawano and Mses. Kumiko Kawakita and Mina Sagara for their help in measuring circulating levels of mediators, and to Drs. Keiichi Fukudome and Osamu Ogawa for their collaborative assistance.

Declaration Of Interest:

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