Short Communication: Paraoxonase 1 (PON1) in French HIV-Infected Patients Under Antiretroviral Therapy: Relationship with the Metabolic Syndrome and Inflammation

Abstract

Our goal was to determine if paraoxonase 1 (PON1) activity relates to the presence of metabolic syndrome (MS) and inflammation in HIV patients treated with highly active antiretroviral therapy (HAART). This was a prospective, multicenter study including 269 patients receiving HAART for at least 1 year and a maximum of 4 years. PON1 and inflammatory markers [C reactive protein (CRP), interleukin-6 (IL-6), serum amyloid A (SAA), and soluble tumor necrosis factor receptors 2 (sTNF-R2)] were compared between patients with or without MS and the association between inflammatory markers and PON1 was assessed by logistic regression analyses. MS was found in 18.2% of the patients. Inflammatory markers, with the exception of sTNF-R2, were significantly higher, while PON1 activity was significantly lower in the presence of metabolic syndrome. PON1 activity was significantly related to apolipoprotein C3, CD4 count, and sTNF-R2. It may be concluded that PON1 appears to be a marker for the metabolic syndrome in HIV-infected subjects. PON1 activity is related to dyslipidemia and the immunological status of the patients but is not fully determined by inflammation.

W e have recently shown that metabolic syndrome (MS) is present in 16%¹ to 18%² of HIV-infected subjects under therapy. In plasma, paraoxonase 1 (PON1) is carried by high-density lipoproteins (HDL) and protects low-density lipoprotein (LDL) from oxidation.³ PON1 activity was recently shown to be reduced in HIV-infected patients.⁴ However, remodeling of the HDL particles, including changes in apolipoprotein composition and the addition of acute phase proteins such as serum amyloid A (SAA), may affect PON1 concentration or activity.⁵

This study addressed two questions: (1) Is PON1 activity associated with MS in HIV-infected patients under therapy? (2) Does PON1 activity in these patients depend on inflammation?

This study was approved by the French ethical committee (CCPPRB) from Pays de Loire (No. 86/03 dated December 2, 2003) and all subjects signed an informed consent. The study design and main results were detailed previously.² Briefly, this was a prospective, multicenter study including HIV-infected patients receiving highly active antiretroviral therapy (HAART) for at least 1 year and a maximum of 4 years without treatment interruption. At the time of the study, MS, as defined by NCEP-ATPIII criteria, was assessed⁶ and was found in 18.2% of the 269 patients,² fitting with previous results.^7,8 The present results consist of a substudy focusing on PON1 activity and inflammation markers.

All analyses described here were run on −80°C frozen samples in a central laboratory. Apolipoprotein A1 and B, highly sensitive C reactive protein (CRP), and SAA were determined by immunonephelometry (Dade Behring, Rueil Malmaison, France). Apolipoproteins (apo) C3 and E were determined by immunoturbidimetry (Diasys, Bouffémont, France). Interleukin-6 (IL-6) and soluble tumor necrosis factor receptors 2 (sTNF-R2) were determined by ELISA (R and D Systems, Minneapolis, MN). PON1 activity was assessed by the rate of enzymatic hydrolysis of 3 mM paraoxon (Sigma Aldrich, St. Louis, MO) to p-nitrophenol in 2 mM CaCl₂ (Sigma Aldrich, St. Louis, MO) in 0.1 M Tris-HCl (Bio-Rad Laboratories, Hercules, CA) (pH 8.0). Serum samples were diluted 1:2. The amount of p-nitrophenol generated was monitored with a continuously recording spectrophotometer (Spectramax M250) by the increase in absorbance at 405 nm and 25°C between 1 and 7 min. Enzymatic activity was calculated using a molar extinction coefficient of 18,450/mol/cm. All analyses were done on EDTA plasma with the exception of PON1, which was run on serum.

All statistics were run on SAS software version 9.1 (Chapel Hill, NC). Descriptive statistics consist of means, standard deviation, and median. A logistic regression analysis was performed to identify a potential association between PON1 or inflammatory markers and MS. The association of various biological and clinical parameters with PON1 was determined using univariate and multivariate logistic regression models. In these models, PON1 was introduced as the explained variable at two levels of activity (above or below the median value) and log-transformation or categorization (quartile or known cut-off value) was used if the log-linearity assumption was not met for the other variables. The multivariate model was run including independent variables with an association at the p<0.25 level in univariate analysis. A descending selection procedure was performed on select variables. The final model included variables with a significant association (p<0.05). Odds ratios with 95% confidence interval and p-values were calculated.

MS was found in 49 out of 269 patients (18.2%). The association of PON1 and inflammation markers with MS is shown in Table 1. All inflammation markers, with the exception of sTNF-R2, were significantly increased in the presence of MS, while PON1 was significantly decreased.

Table 1.

Inflammatory Markers and Paraoxonase 1 Activity According to the Presence or Absence of Metabolic Syndrome

	MS (n=49)	No MS (n=220)	p
CRP (mg/liter)	4.8±5.2 (3.3)	3.7±6.9 (1.4)	0.0016
IL-6 (ng/liter)	3.8±15.6 (1.3)	1.5±2.8 (0.9)	0.0322
sTNF-R2 (μg/liter)	2.1±1.3 (1.9)	2.0±1.0 (1.9)	0.4961
SAA (mg/liter)	8.7±10.4 (5.7)	9.2±24.8 (3.6)	0.0303
PON (μmol/min/liter)	18.1±21.1 (9.9)	43.8±72.6 (20.7)	0.0133

Mean±SD (median) and p value (logistic regression analysis).

MS, metabolic syndrome; CRP, C reactive protein; IL-6, interleukin 6; sTNF-R2, tumor necrosis factor soluble receptor 2; SAA, serum amyloid A; PON, paraoxonase.

As shown in Table 2, PON1 activity was positively associated with apo C3 [OR: 1.01 (1.00; 1.01), p=0.0042] and apo E [OR: 1.02 (1.00; 1.03), p=0.0262], while it was negatively associated with sTNF-R2 [OR: 0.36 (0.18; 0.74) for quartile 2 vs. quartile 1; OR: 1.07 (0.51; 2.24) for quartile 3 vs. quartile 1; and OR: 0.33 (0.16; 0.71) for quartile 4 vs. quartile 1; p=0.0011] and CD4 count [0.52 (0.30; 0.89); p=0.0161 for CD4 above vs. below the median]. In a stepwise descending multivariate logistic regression analysis, apo C3, sTNF-R2, and CD4 count were the only variables that remained independently associated with PON1 activity. While apo C3 [OR: 1.01 (1.00; 1.01), p=0.0140] was associated with an increase in PON1 activity, higher levels of sTNF-R2 and CD4 count were associated with lower levels of PON1.

Table 2.

Factors Associated with Paraoxonase Activity (Above or Below the Median), as Assessed by Logistic Regression Analysis

	Univariate analysis		Multivariate analysis
	Odds ratio	p	Odds ratio	p
Non-HDL	0.91 [0.72; 1.14]	0.4089	—
HDL (ref. > normal)^a	0.88 [0.44; 1.77]	0.7233	—
Triglycerides	0.88 [0.69; 1.13]	0.3261	—
Apo A1	1.64 [0.78; 3.46]	0.1893	—
Apo B	0.71 [0.29; 1.74]	0.4524	—
Apo C3	1.01 [1.00; 1.01]	0.0042	1.01 [1.00; 1.01]	0.0140
Apo E	1.02 [1.00; 1.03]	0.0262	—
CRP	1.01 [0.97; 1.05]	0.5609	—
Insulin (ref. <15 mU/liter)	0.88 [0.38; 1.86]	0.6667	—
CD4 [ref. <403 cells/mm³ (median)]	0.52 [0.30; 0.89]	0.0161	0.49 [0.27; 0.86]	0.0137
Coinfection HCV (ref yes)	0.83 [0.25; 2.79]	0.7584	—
IL-6: ref. Q1		0.5583
Q2	0.70 [0.33; 1.46]		—
Q3	0.87 [0.42; 1.80]		—
Q4	1.18 [0.57; 2.42]		—
sTNF-R2: ref. Q1		0.0011		0.0043
Q2	0.36 [0.18; 0.74]		0.41 [0.20; 0.87]
Q3	1.07 [0.51; 2.24]		1.33 [0.61; 2.89]
Q4	0.33 [0.16; 0.71]		0.42 [0.19; 0.96]
SAA	1.15 [0.89; 1.49]	0.2699	—

Since log linearity was not met, HDL was introduced as categories, below vs. above the threshold value for MS (1.03 mmol/liter for men and 1.30 mmol/liter for women). Multivariate logistic regression consists of a stepwise descending procedure starting with variables linked to paraoxonase with a p value<0.25.

Odd ratio [5th; 95th].

Our goal in the present study was to relate MS to PON1 activity and inflammation. Although it would be interesting to take lipodystrophy into consideration in this context, we limited ourself to the usual definition of MS, in which the only criterion for fat redistribution was waist circumference.

The relationship between inflammation and MS has been acknowledged for a long time. Several studies have demonstrated an association between adipose tissue or MS and CRP level in plasma.⁹ Here we confirm that CRP is significantly higher in the presence of MS in HIV patients. It has been suggested that IL-6 is involved in insulin sensitivity in mice and influences glucose tolerance in humans.¹⁰ Plasma levels of IL-6 have been associated with cardiovascular disease risk.¹¹ In a recent study¹² it was suggested that IL-6, together with plasminogen activator inhibitor 1 (PAI-1), CRP, and fibrinogen, may contribute to the risk of developing MS. It was also suggested that IL-6 and CRP, but not TNF, represent useful markers of obesity-related inflammation.¹³ However, TNF soluble receptors (sTNFR) emerge as better indicators of TNF production than TNF itself, because of their longer half-life.¹⁴ Recent data suggest that circulating levels of sTNFR are directly linked with lipoprotein disturbance, insulin resistance, and obesity.¹⁵ In this study, we chose to measure sTNF-R2 rather than TNF itself and we failed to show any difference between HIV subjects with and without MS in this marker. SAA is a marker of acute inflammation, which was already correlated with CRP, obesity, triglycerides, and insulin resistance, as well as MS itself.¹⁶ Here we confirm that HIV subjects with MS have higher levels of SAA. Although the mean value appeared slightly lower in subjects with MS, the median value was much higher and the logistic regression analysis demonstrated a positive association between SAA and the presence of MS [OR: 1.38 (CI: 1.03–1.85, p=0.03)].

The main finding of our study is that PON1 activity is significantly decreased in subjects with MS. In a previous study,⁴ it was shown that HIV patients have lower PON1 activity than controls. Our data suggest that these results are related to the relatively high frequency of MS in HIV patients. It was suggested that oxidative stress explains the relative increase in cardiovascular risk in HIV-infected subjects,¹⁷ while PON1 activity did not show any relationship. Therefore the clinical significance of our findings needs to be further explored. Because PON1 is carried by HDL, and HDL structure may be influenced by protein changes linked to MS and/or inflammation, we measured the relationship between various components of lipid metabolism and inflammation and PON1 activity in our patients.

Apolipoprotein A1, as well as HDL, was not associated with PON1 activity, suggesting that PON1 does not depend on the concentration of HDL, its carrier. However, a positive correlation between apolipoproteins C3 and E and PON1 was observed. These proteins are both components of triglyceride-rich lipoproteins and HDL. On the one hand, they might have a synergistic action with PON1 to protect against oxidation. On the other hand, it may be that these apolipoproteins are associated with lipoprotein oxidation and that, in parallel, PON1 expression is stimulated by oxidative stress. Surprisingly, the only significant association between PON1 and inflammation markers was found with sTNF-R2. However, it should be kept in mind that this association does not seem to be linear, since the odds ratios clearly differ between quartiles of sTNF-R2. Although it may be speculated that PON1 and TNF receptors share regulating factors, the exact significance of this isolated association remains to be elucidated. This would be of particular importance considering that sTNF-R2 is significantly associated with PON1 in multiple regression analysis. Nevertheless, the fact that sTNF-R2 is related to PON1 activity in a nonlinear manner, together with the absence of any difference in this parameter between MS and non-MS, should encourage further studies on the mechanistic aspects of the relationship between this marker and MS in this context. In particular, the role of various types of HAART in this situation should be evaluated.

It could have been speculated that during the acute phase response, the parallel HDL enrichment in SAA and impoverishment in apo A1 would lead to lower PON1 activity, since paraoxonase requires apo A1 to function.¹⁸ The absence of any significant association between SAA or apo A1 and PON1 does not argue in favor of this hypothesis. In addition, because the patients are under therapy for 1 to 4 years, we cannot really consider that they are in an acute phase situation. In addition to these associations, a high level of CD4 was related to lower PON1 activity. This observation does not fit with previous results in which a high CD4 count was associated with high PON1 activity.⁴ Although it is difficult to explain this discrepancy, it may be postulated that in the present study oxidative stress is less pronounced when the immunological status of the patients is improved. While HCV coinfection has been previously shown to affect PON1 activity,⁴ we did not find any association in the present study.

When interpreting our results, it should be kept in mind that gene polymorphism may influence PON concentration and/or activity. An important study¹⁹ has clearly indicated that a haplotype distribution may differ between HIV-infected patients and controls. Although our goal was not to compare HIV-infected subjects with controls, it could be that MS and non-MS patients may have different gene polymorphism distributions. This gene polymorphism could partly explain the observed differences in PON1 activity. Nevertheless, the clinical significance of our findings should be further evaluated.

In summary, PON1 appears to be a marker of the metabolic syndrome in HIV-infected subjects. Its activity is related to dyslipidemia and the immunological status of the patients but is not fully determined by inflammation. The clinical consequence of this observation remains to be evaluated in terms of follow-up of the patients and prevention of metabolic disturbances.

Footnotes

Acknowledgments

This study was made possible by a grant from SIDACTION. The authors wish to thank Tanguy Le Néel and Biofortis for their help in managing the data.

Author Disclosure Statement

No competing financial interests exist.

References

Bard

, Lassalle

, Capeau

et al. Association of apolipoproteins C3 and E with metabolic changes in HIV-infected adults treated with a protease-inhibitor-containing antiretroviral therapy. Antivir Ther, 2006; 11:361–370.

Biron

, Bobin-Dubigeon

, Volteau

et al. Metabolic syndrome in French HIV-infected patients: Prevalence and predictive factors after 3 years of antiretroviral therapy. AIDS Res Hum Retroviruses, 2012; 28:1672–1678.

Mackness

, Arrol

, Durrington

. Paraoxonase prevents accumulation of lipoperoxides in low-density lipoprotein. FEBS Lett, 1991; 286:152–154.

Parra

, Alonso-Villaverde

, Coll

et al. Serum paraoxonase-1 activity and concentration are influenced by human immunodeficiency virus infection. Atherosclerosis, 2007; 194:175–181.

Banka

, Yuan

, de Beer

et al. Serum amyloid A (SAA): Influence on HDL-mediated cellular cholesterol efflux. J Lipid Res, 1995; 36:1058–1065.

Grundy

, Brewer HB

, Cleeman

et al. Definition of metabolic syndrome: Report of the National Heart, Lung, and Blood Institute/American Heart Association conference on scientific issues related to definition. Circulation, 2004; 109:433–438.

Jerico

, Knobel

, Montero

et al. Metabolic syndrome among HIV-infected patients: Prevalence, characteristics, and related factors. Diabetes Care, 2005; 28:132–137.

Samaras

, Wand

, Law

et al. Prevalence of metabolic syndrome in HIV-infected patients receiving highly active antiretroviral therapy using International Diabetes Foundation and Adult Treatment Panel III criteria: Associations with insulin resistance, disturbed body fat compartmentalization, elevated C-reactive protein, and [corrected] hypoadiponectinemia. Diabetes Care, 2007; 30:113–119.

Lemieux

, Pascot

, Prud'homme

et al. Elevated C-reactive protein: Another component of the atherothrombotic profile of abdominal obesity. Arterioscler Thromb Vasc Biol, 2001; 21:961–967.

10.

Fasshauer

, Kralisch

, Klier

et al. Interleukin-6 is a positive regulator of tumor necrosis factor alpha-induced adipose-related protein in 3T3-L1 adipocytes. FEBS Lett, 2004; 560:153–157.

11.

Luc

, Bard

, Juhan-Vague

et al. C-reactive protein, interleukin-6, and fibrinogen as predictors of coronary heart disease: The PRIME Study. Arterioscler Thromb Vasc Biol, 2003; 23:1255–1261.

12.

Kraja

, Province

, Arnett

et al.

Do inflammation and procoagulation biomarkers contribute to the metabolic syndrome cluster?

Nutr Metab (Lond), 2007; 4:28.

13.

Browning

, Krebs

, Magee

et al. Circulating markers of inflammation and their link to indices of adiposity. Obes Facts, 2008; 1:259–265.

14.

Aderka

. The potential biological and clinical significance of the soluble tumor necrosis factor receptors. Cytokine Growth Factor Rev, 1996; 7:231–240.

15.

Dzienis-Straczkowska

, Straczkowski

, Szelachowska

et al. Soluble tumor necrosis factor-alpha receptors in young obese subjects with normal and impaired glucose tolerance. Diabetes Care, 2003; 26:875–880.

16.

Liu

, Young

, Zinman

et al. Lifestyle variables, non-traditional cardiovascular risk factors, and the metabolic syndrome in an Aboriginal Canadian population. Obesity (Silver Spring), 2006; 14:500–508.

17.

Parra

, Coll

, Aragones

et al. Nonconcordance between subclinical atherosclerosis and the calculated Framingham risk score in HIV-infected patients: Relationships with serum markers of oxidation and inflammation. HIV Med, 2010; 11:225–231.

18.

Van Lenten

, Hama

, de Beer

et al. Anti-inflammatory HDL becomes pro-inflammatory during the acute phase response. Loss of protective effect of HDL against LDL oxidation in aortic wall cell cocultures. J Clin Invest, 1995; 96:2758–2767.

19.

Parra

, Marsillach

, Aragones

et al. Paraoxonase-1 gene haplotypes are associated with metabolic disturbances, atherosclerosis, and immunologic outcome in HIV-infected patients. J Infect Dis, 2010; 201:627–634.