CLB-09087; No. of pages: 3; 4C: Clinical Biochemistry xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Clinical Biochemistry journal homepage: www.elsevier.com/locate/clinbiochem
Paired measurement of serum amyloid A (SAA) and paraoxonase 1 (PON1) as useful markers in breast cancer recurrence Christine Bobin-Dubigeon a,b, Armelle Lefrançois a, Jean-Marc Classe c, Marie-Pierre Joalland a, Jean-Marie Bard a,b,⁎ a b c
Institut de Cancérologie de l'Ouest-Biopathologie, Boulevard Jacques Monod 44805 Saint-Herblain, France Université de Nantes, MMS - EA 2160 - Mer Molécules Santé, IUML - Institut Universitaire Mer et Littoral - FR3473 CNRS, France Institut de Cancérologie de l'Ouest-Oncologie Chirurgicale, Boulevard Jacques Monod 44805 Saint-Herblain, France
a r t i c l e
i n f o
Article history: Received 3 June 2015 Received in revised form 7 July 2015 Accepted 13 July 2015 Available online xxxx Keywords: Paraoxonase SAA Inflammation Breast cancer
a b s t r a c t Objectives: Paraoxonase 1 (PON1) and serum amyloid A (SAA) are carried by HDL. In case of inflammation, SAA and PON1 tend to change in opposite direction. In this study we determined if inflammation leads to altered PON1 activity using three different substrate hydrolysis rates, paraoxonase (PON), arylesterase (ARE) and lactonase (LAC) in breast cancer recurrence. Design and methods: 49 patients with a recurrence of breast cancer were analyzed for SAA, CRP, lipids, oxidized LDL, PON, ARE and LAC. Distribution of PON1 activities across the quartiles of CRP and SAA were compared by the Kruskal Wallis test. Non-parametric estimates of the survivor function were computed with Kaplan–Meier method. The association of SAA and ARE with short term death was assessed by logistic regression models. Results: HDL and ARE decrease significantly across the quartiles of CRP. No significant differences were observed across SAA quartiles. The survival time was significantly related to the level of SAA (log rank: p b 0.001) as well as the level of ARE (log rank: p = 0.039). SAA and ARE were independently related to survival time below one year. Conclusions: PON1 does not seem to be directly affected by SAA, for any of the tested substrates, PON, ARE and LAC. The combined measurement of SAA and ARE could be a useful tool in this clinical situation, since they are independently related to short term death. © 2015 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
1. Introduction Paraoxonase 1 (PON1) has paraoxonase, arylesterase and lactonase activity [1] and is involved in the protection against xenobiotic toxicity [2]. In plasma, PON1 is carried by high-density lipoproteins (HDL) and plays a role in the oxidative stress, mainly because it protects lowdensity lipoproteins (LDL) from oxidation [2]. Serum amyloid A (SAA) protein is another HDL-associated apolipoprotein, well known as an acute-phase protein during inflammation [3]. Inflammation and oxidative stress are coexisting conditions underlying various chronic diseases, including cancer [4]. It has been demonstrated that serum PON1 activity is lower in patients with various cancer diseases when compared to control groups [5] and we have previously shown that PON1 could represent a marker of short term death in breast cancer recurrence [6]. However, clinical studies in various inflammatory situations suggest that SAA and PON1 tend to change in opposite directions [7]. ⁎ Corresponding author at: Institut de Cancérologie de l'Ouest-Biopathologie, Boulevard Jacques Monod 44805 Saint-Herblain, France. E-mail address: [email protected] (J.-M. Bard).
It is therefore suggested that the combined determination of PON1 and SAA may represent a useful tool in clinical practice [8]. The goal of the present study is to determine how PON1 activity is influenced by inflammation in breast cancer disease and if SAA and PON1 represent markers of mortality in breast cancer recurrence. 2. Material and methods 2.1. Patients Fifty patients who were previously treated by surgery and standard therapy for breast cancer in our center and who experienced metastatic disease progression and cancer recurrence were included in this study. Informed consent was obtained from patients to use their biological specimens and clinicopathological data for research purposes, as required by the French regulation and the French Committee for the Protection of Human Subjects. The main clinical characteristics of the patients at the time of diagnosis were extensively described in a previous paper [6]. Data were extracted from a prospective database (Berenis®). All patients had surgery at diagnosis but at the time of the
http://dx.doi.org/10.1016/j.clinbiochem.2015.07.020 0009-9120/© 2015 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
Please cite this article as: C. Bobin-Dubigeon, et al., Paired measurement of serum amyloid A (SAA) and paraoxonase 1 (PON1) as useful markers in breast cancer recurrence, Clin Biochem (2015), http://dx.doi.org/10.1016/j.clinbiochem.2015.07.020
2
C. Bobin-Dubigeon et al. / Clinical Biochemistry xxx (2015) xxx–xxx
study, all experienced clinical metastasis. All patients were treated with anti-estrogen therapy at the time of the study. Most of the patients were addressed to our center for an invasive ductal carcinoma (92%). 66% were post-menopausal women at the time of diagnosis. Tumor characteristics were as follows: 48% expressed human epidermal growth factor receptor-2 (Her2), 56% were estrogen receptor (ER) positive and 38% were progesterone receptor (PR) positive, 44% were classified grade 3 in the grading system of Scarff Bloom and Richardson (SBR), 36% were SBR grade 2 and 4% were SBR grade 1. The SBR grade was unknown for 16% of patients. Blood samples were drawn for routine laboratory analyses at the time of recurrence. Venous blood was collected into tubes containing heparin. Whole blood samples were centrifuged immediately at 1500 g, 4 °C for 5 min, and plasma was divided into aliquots and immediately stored at − 70 °C prior to analysis. One sample did not raise enough volume, therefore 49 samples were analyzed. Follow up of the patients was done during one year and delay between sampling time and death was recorded during one year. Biological analyses were run in frozen samples without knowledge of the vital status of the patients.
2.2. Biological analyses PON1 activity was determined using paraoxon (PON) and phenyl acetate (ARE) as substrates, as previously described [6]. The lactonase (LAC) activity was also determined by an adaptation of a method previously published by others [9]. Briefly, a substrate solution of 60 mM Thiobutyrolactone was prepared in a buffer containing Tris 50 mM, CaCl2 1 mM, pH 8.0. Plasma samples were diluted 1/125 in DTNB (5.5, Dithiobis (2-nitrobenzoic acid)) 1 mM. The reaction was followed in microplates at 412 nm for 5 min at 25 °C, after mixing 100 μL of diluted plasma with 100 μL substrate solution. All reagents were obtained from Sigma Aldrich (Saint-Louis, MO, USA), except for Tris which was from Bio Rad Laboratories, Hercules, CA, USA. The mean CVs from 4 different controls were 3.9% and 7.9% for intra and inter-assays, respectively. C Reactive Protein (CRP), cholesterol, HDL Cholesterol and triglycerides were measured in a routine analyzer (Beckman AU 800) and LDL cholesterol was calculated by the Friedewald formula, except for samples with triglycerides above 4 mmol/L, for which LDL cholesterol was not calculated [10]. OxLDL (Mercodia, Uppsala, Sweden) and SAA (Human SAA, Invitrogen, Saint-Aubin, France) were determined using ELISA.
2.3. Statistical analyses All statistics were run using the SAS software version 9.1 (Chapell Hill, NC, USA). The distribution of various biological variables across quartiles of CRP or SAA was compared using the Kruskal Wallis test. Non-parametric estimates of the survivor function was computed with Kaplan–Meier method and log-rank test of the equality of survival distributions across quartiles was performed with the LIFETEST procedure of SAS. Univariate and multivariate logistic regression models were performed to identify potential relationships of biomarkers with early death as defined by death occurring within the follow up time of one year (n = 28). A univariate model was used to determine the individual associations and a bivariate analysis was used to detect any independent association between SAA or ARE and survival below one year. 3. Results Table 1 shows the distribution of the three paraoxonase activities and lipids across the quartiles of CRP and SAA. HDL and ARE significantly decrease across the quartiles of CRP. No significant difference was observed for any lipids or paraoxonase activity across SAA quartiles. A tendency was observed for HDL and LAC but this did not reach significance. A shown in Fig. 1, the survival time was significantly related to the level of SAA (log rank: p b 0.001) as well as the level of ARE (log rank: p = 0.039). Quartile 1 differed from quartile 4 with a hazard ratio of 0.206 [0.072–0.588], p = 0.0032 for SAA and with a hazard ratio of 5.48 [1.48–20.36], p = 0.011 for ARE. In addition the bivariate logistic regression analysis indicated that both SAA (odds ratio: 2.24 [1.21–4.15], p = 0.01) and ARE (odds ratio: 0.49 [0.26–0.92], p = 0.03) were independently related to survival time of less than one year (n = 28). 4. Discussion The present study was undertaken to determine if inflammation observed in breast cancer recurrence, leads to altered PON1 activity. The aim of the study was not to determine if patients with breast cancer differ from healthy subjects for PON1 activity and we did not run a control group. It is therefore not possible to conclude that all PON1 activities are decreased in our patients. In vitro, it was shown that PON1 mRNA was reduced following stimulation with inflammatory cytokines [11].
Table 1 Distribution of paraoxonase activities and lipids, across quartiles of CRP and SAA. CRP (mg/L)
[0.8–1.8[
[1.8–8.6[
[8.6–25.8[
[25.8–279.2]
P⁎
PON (μmol/L/min) ARE (mmol/L/min) LAC (μmol/L/min) HDL-C (mmol/L) LDL-C (mmol/L) TG (mmol/L) oxLDL (AU) oxLDL/LDL-C (AU)
114 [36–308] 31.0 [20.1–51.3] 0.21 [0.06–0.32] 1.65 [1.32–2.29] 3.01 [2.43–4.88] 1.19 [0.50–2.26] 48.6 [39.4–95.1] 16.6 [11.6–25.3]
122 [51–225] 27.9 [11.7–51.3] 0.18 [0.06–0.25] 1.59 [0.69–1.98] 3.51 [1.59–4.93] 1.26 [0.87–2.34] 51.2 [36.7–91.2] 15.6 [11.4–26.9]
60 [31–374] 21.9 [13.7–33.0] 0.19 [0.08–0.25] 1.54 [1.00–2.02] 3.23 [2.60–4.67] 1.10 [0.73–2.53] 50.2 [36.1–74.1] 15.3 [11.4–19.6]
64 [37–148] 24.5 [11.7–33.0] 0.18 [0.10–0.28] 1.16 [0.93–1.66] 3.31 [1.94–5.74] 1.31 [0.73–4.84] 50.8 [30.8–137.8] 18.9 [12.9–35.9]
0.30 0.02 0.21 0.01 0.67 0.82 0.98 0.31
SAA (μg/L)
[6.8–28.6[
[28.6–54.5[
[54.5–149.3[
[149.3–1200]
P⁎
PON (μmol/L/min) ARE (mmol/L/min) LAC (μmol/L/min) HDL-C (mmol/L) LDL-C (mmol/L) TG (mmol/L) oxLDL (AU) oxLDL/LDL-C (AU)
100 [44–207] 25.7 [11.7–51.3] 0.18 [0.06–0.32] 1.43 [0.69–2.09] 3.21 [1.59–4.20] 1.19 [0.50–2.34] 47.9 [36.7–72.4] 16.1 [11.6–23.1]
113 [35–374] 25.8 [15.3–43.4] 0.20 [0.08–0.25] 1.66 [1.00–2.01] 3.36 [2.88–4.88] 1.34 [0.73–2.53] 58.6 [36.1–95.1] 15.0 [11.4–26.9]
61 [37–249] 23.5 [18.9–43.4] 0.21 [0.16–0.29] 1.51 [0.99–2.29] 3.18 [1.94–5.74] 1.66 [0.73–4.84] 53.6 [43.6–137.8] 19.1 [11.4–30.0]
93 [31–235] 25.2 [11.7–33.0] 0.15 [0.10–0.26] 1.16 [0.93–1.66] 3.31 [1.94–5.72] 1.20 [0.80–1.90] 47.2 [30.8–81.7] 14.8 [12.3–35.9]
0.69 0.56 0.07 0.08 0.66 0.66 0.23 0.33
Median [5th–95th]. ⁎ Kruskal–Wallis test.
Please cite this article as: C. Bobin-Dubigeon, et al., Paired measurement of serum amyloid A (SAA) and paraoxonase 1 (PON1) as useful markers in breast cancer recurrence, Clin Biochem (2015), http://dx.doi.org/10.1016/j.clinbiochem.2015.07.020
C. Bobin-Dubigeon et al. / Clinical Biochemistry xxx (2015) xxx–xxx
3
no change in oxidized LDL was observed, suggesting that the PON1 anti-oxidative activity was not really affected. The reason why ARE activity would be the only affected substrate hydrolysis function of the enzyme is unclear. We could raise the hypothesis that some cytokines or substrates generated during inflammation would decrease the hydrolysis rate by competing with the substrate or by modifying the enzyme conformation. This hypothesis would need further in vitro studies to be confirmed. However, our study has some limitations and definitive conclusion is not possible. The number of patients is relatively low and it should be kept in mind that the absence of significance for some substrate hydrolysis activity could just be related to a lack of power. In this particular group of patients with a high short term mortality rate, we show that high levels of SAA and low levels of ARE are clearly associated with shorter survival time. As a previous study by Tanimoto et al. indicated that the ratio between SAA and PON1 could be increased in case of inflammatory disease [7] it was suggested by others [8] that the combined determination of PON1 and SAA may represent a useful tool in clinical practice. However, if SAA and ARE were strongly negatively correlated, it would be expected that the association between these two parameters and short term death would not be independent. On the contrary, in our bivariate logistic regression analysis, we observed that both SAA and ARE are independently related to death occurring within one year. This would suggest that the combined measurement of these two parameters could be of clinical interest. However, PON1 enzyme activity is currently measured with “home made” methods and SAA is available only as ELISA kits. This represents a limitation for routine application yet. In summary, this study indicates that not all substrate hydrolysis rates of PON1 are inversely related to inflammation and that the combined measurement of SAA and PON1 (arylesterase activity) could be useful for clinical assessment of patients with breast cancer recurrence. References
Fig. 1. Survival time curves according to SAA (panel A) and ARE (panel B) quartiles and odds ratio and [5th–95th] Wald confidence intervals for one quartile of SAA and ARE in the bivariate logistic regression analysis with death occurring within one year of follow up as the explained variable (panel C).
However, acute phase proteins, such as CRP and SAA do not seem to exert any direct effect on PON1 expression. Nevertheless, it was suggested that the increase of SAA in inflammatory HDL and the following decrease of apolipoprotein (apo) AI could lead to a reduction in circulating PON1 [12]. Our results indicate that not all substrate hydrolysis activities of PON1 are affected by inflammation. Only the ARE activity decreases in presence of inflammation. Therefore, we cannot argue in favor of any direct effect of inflammation on the enzyme expression. In addition, the relationship between ARE and inflammation markers is observed when CRP but not SAA is considered. This observation confirms what we have already shown in a population of HIV infected patients [13]. It does not fit with the hypothesis of a direct effect of SAA which would take the place of apo AI in HDL and subsequently reduce PON1. Moreover, although HDL decreases with inflammation,
[1] O. Khersonsky, D.S. Tawfik, Structure-reactivity studies of serum paraoxonase PON1 suggest that its native activity is lactonase, Biochemistry 44 (2005) 6371–6382. [2] M.I. Mackness, S. Arrol, P.N. Durrington, Paraoxonase prevents accumulation of lipoperoxides in low-density lipoprotein, FEBS Lett. 286 (1991) 152–154. [3] C.M. Uhlar, A.S. Whitehead, Serum amyloid A, the major vertebrate acute-phase reactant, Eur. J. Biochem. 265 (1999) 501–523. [4] B.N. Ames, Dietary carcinogens and anticarcinogens. Oxygen radicals and degenerative diseases, Science 221 (1983) 1256–1264. [5] M.N. Akcay, M.F. Polat, I. Yilmaz, G. Akcay, Serum paraoxonase levels in pancreatic cancer, Hepatogastroenterology 50 (Suppl. 2) (2003) ccxxv–ccxxvii. [6] C. Bobin-Dubigeon, I. Jaffré, M.-P. Joalland, et al., Paraoxonase 1 (PON1) as a marker of short term death in breast cancer recurrence, Clin. Biochem. 45 (2012) 1503–1505. [7] N. Tanimoto, Y. Kumon, T. Suehiro, et al., Serum paraoxonase activity decreases in rheumatoid arthritis, Life Sci. 72 (2003) 2877–2885. [8] K. Kotani, T. Yamada, A. Gugliucci, Paired measurements of paraoxonase 1 and serum amyloid A as useful disease markers, BioMed Res. Int. 2013 (2013) 481437. [9] J. Marsillach, G. Aragones, R. Beltran, et al., The measurement of the lactonase activity of paraoxonase-1 in the clinical evaluation of patients with chronic liver impairment, Clin. Biochem. 42 (2009) 91–98. [10] W.T. Friedewald, R.I. Levy, D.S. Fredrickson, Estimation of the concentration of lowdensity lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge, Clin. Chem. 18 (1972) 499–502. [11] Y. Kumon, Y. Nakauchi, T. Suehiro, et al., Proinflammatory cytokines but not acute phase serum amyloid A or C-reactive protein, downregulate paraoxonase 1 (PON1) expression by HepG2 cells, Amyloid 9 (2002) 160–164. [12] B.J. Van Lenten, S.Y. Hama, F.C. de Beer, et al., Anti-inflammatory HDL becomes proinflammatory during the acute phase response. Loss of protective effect of HDL against LDL oxidation in aortic wall cell cocultures, J. Clin. Investig. 96 (1995) 2758–2767. [13] C. Bobin-Dubigeon, C. Biron, C. Volteau, et al., Short communication: paraoxonase 1 (PON1) in French HIV-infected patients under antiretroviral therapy: relationship with the metabolic syndrome and inflammation, AIDS Res. Hum. Retrovir. 29 (2013) 1571–1574.
Please cite this article as: C. Bobin-Dubigeon, et al., Paired measurement of serum amyloid A (SAA) and paraoxonase 1 (PON1) as useful markers in breast cancer recurrence, Clin Biochem (2015), http://dx.doi.org/10.1016/j.clinbiochem.2015.07.020
Contents lists available at ScienceDirect
Clinical Biochemistry journal homepage: www.elsevier.com/locate/clinbiochem
Paired measurement of serum amyloid A (SAA) and paraoxonase 1 (PON1) as useful markers in breast cancer recurrence Christine Bobin-Dubigeon a,b, Armelle Lefrançois a, Jean-Marc Classe c, Marie-Pierre Joalland a, Jean-Marie Bard a,b,⁎ a b c
Institut de Cancérologie de l'Ouest-Biopathologie, Boulevard Jacques Monod 44805 Saint-Herblain, France Université de Nantes, MMS - EA 2160 - Mer Molécules Santé, IUML - Institut Universitaire Mer et Littoral - FR3473 CNRS, France Institut de Cancérologie de l'Ouest-Oncologie Chirurgicale, Boulevard Jacques Monod 44805 Saint-Herblain, France
a r t i c l e
i n f o
Article history: Received 3 June 2015 Received in revised form 7 July 2015 Accepted 13 July 2015 Available online xxxx Keywords: Paraoxonase SAA Inflammation Breast cancer
a b s t r a c t Objectives: Paraoxonase 1 (PON1) and serum amyloid A (SAA) are carried by HDL. In case of inflammation, SAA and PON1 tend to change in opposite direction. In this study we determined if inflammation leads to altered PON1 activity using three different substrate hydrolysis rates, paraoxonase (PON), arylesterase (ARE) and lactonase (LAC) in breast cancer recurrence. Design and methods: 49 patients with a recurrence of breast cancer were analyzed for SAA, CRP, lipids, oxidized LDL, PON, ARE and LAC. Distribution of PON1 activities across the quartiles of CRP and SAA were compared by the Kruskal Wallis test. Non-parametric estimates of the survivor function were computed with Kaplan–Meier method. The association of SAA and ARE with short term death was assessed by logistic regression models. Results: HDL and ARE decrease significantly across the quartiles of CRP. No significant differences were observed across SAA quartiles. The survival time was significantly related to the level of SAA (log rank: p b 0.001) as well as the level of ARE (log rank: p = 0.039). SAA and ARE were independently related to survival time below one year. Conclusions: PON1 does not seem to be directly affected by SAA, for any of the tested substrates, PON, ARE and LAC. The combined measurement of SAA and ARE could be a useful tool in this clinical situation, since they are independently related to short term death. © 2015 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
1. Introduction Paraoxonase 1 (PON1) has paraoxonase, arylesterase and lactonase activity [1] and is involved in the protection against xenobiotic toxicity [2]. In plasma, PON1 is carried by high-density lipoproteins (HDL) and plays a role in the oxidative stress, mainly because it protects lowdensity lipoproteins (LDL) from oxidation [2]. Serum amyloid A (SAA) protein is another HDL-associated apolipoprotein, well known as an acute-phase protein during inflammation [3]. Inflammation and oxidative stress are coexisting conditions underlying various chronic diseases, including cancer [4]. It has been demonstrated that serum PON1 activity is lower in patients with various cancer diseases when compared to control groups [5] and we have previously shown that PON1 could represent a marker of short term death in breast cancer recurrence [6]. However, clinical studies in various inflammatory situations suggest that SAA and PON1 tend to change in opposite directions [7]. ⁎ Corresponding author at: Institut de Cancérologie de l'Ouest-Biopathologie, Boulevard Jacques Monod 44805 Saint-Herblain, France. E-mail address: [email protected] (J.-M. Bard).
It is therefore suggested that the combined determination of PON1 and SAA may represent a useful tool in clinical practice [8]. The goal of the present study is to determine how PON1 activity is influenced by inflammation in breast cancer disease and if SAA and PON1 represent markers of mortality in breast cancer recurrence. 2. Material and methods 2.1. Patients Fifty patients who were previously treated by surgery and standard therapy for breast cancer in our center and who experienced metastatic disease progression and cancer recurrence were included in this study. Informed consent was obtained from patients to use their biological specimens and clinicopathological data for research purposes, as required by the French regulation and the French Committee for the Protection of Human Subjects. The main clinical characteristics of the patients at the time of diagnosis were extensively described in a previous paper [6]. Data were extracted from a prospective database (Berenis®). All patients had surgery at diagnosis but at the time of the
http://dx.doi.org/10.1016/j.clinbiochem.2015.07.020 0009-9120/© 2015 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
Please cite this article as: C. Bobin-Dubigeon, et al., Paired measurement of serum amyloid A (SAA) and paraoxonase 1 (PON1) as useful markers in breast cancer recurrence, Clin Biochem (2015), http://dx.doi.org/10.1016/j.clinbiochem.2015.07.020
2
C. Bobin-Dubigeon et al. / Clinical Biochemistry xxx (2015) xxx–xxx
study, all experienced clinical metastasis. All patients were treated with anti-estrogen therapy at the time of the study. Most of the patients were addressed to our center for an invasive ductal carcinoma (92%). 66% were post-menopausal women at the time of diagnosis. Tumor characteristics were as follows: 48% expressed human epidermal growth factor receptor-2 (Her2), 56% were estrogen receptor (ER) positive and 38% were progesterone receptor (PR) positive, 44% were classified grade 3 in the grading system of Scarff Bloom and Richardson (SBR), 36% were SBR grade 2 and 4% were SBR grade 1. The SBR grade was unknown for 16% of patients. Blood samples were drawn for routine laboratory analyses at the time of recurrence. Venous blood was collected into tubes containing heparin. Whole blood samples were centrifuged immediately at 1500 g, 4 °C for 5 min, and plasma was divided into aliquots and immediately stored at − 70 °C prior to analysis. One sample did not raise enough volume, therefore 49 samples were analyzed. Follow up of the patients was done during one year and delay between sampling time and death was recorded during one year. Biological analyses were run in frozen samples without knowledge of the vital status of the patients.
2.2. Biological analyses PON1 activity was determined using paraoxon (PON) and phenyl acetate (ARE) as substrates, as previously described [6]. The lactonase (LAC) activity was also determined by an adaptation of a method previously published by others [9]. Briefly, a substrate solution of 60 mM Thiobutyrolactone was prepared in a buffer containing Tris 50 mM, CaCl2 1 mM, pH 8.0. Plasma samples were diluted 1/125 in DTNB (5.5, Dithiobis (2-nitrobenzoic acid)) 1 mM. The reaction was followed in microplates at 412 nm for 5 min at 25 °C, after mixing 100 μL of diluted plasma with 100 μL substrate solution. All reagents were obtained from Sigma Aldrich (Saint-Louis, MO, USA), except for Tris which was from Bio Rad Laboratories, Hercules, CA, USA. The mean CVs from 4 different controls were 3.9% and 7.9% for intra and inter-assays, respectively. C Reactive Protein (CRP), cholesterol, HDL Cholesterol and triglycerides were measured in a routine analyzer (Beckman AU 800) and LDL cholesterol was calculated by the Friedewald formula, except for samples with triglycerides above 4 mmol/L, for which LDL cholesterol was not calculated [10]. OxLDL (Mercodia, Uppsala, Sweden) and SAA (Human SAA, Invitrogen, Saint-Aubin, France) were determined using ELISA.
2.3. Statistical analyses All statistics were run using the SAS software version 9.1 (Chapell Hill, NC, USA). The distribution of various biological variables across quartiles of CRP or SAA was compared using the Kruskal Wallis test. Non-parametric estimates of the survivor function was computed with Kaplan–Meier method and log-rank test of the equality of survival distributions across quartiles was performed with the LIFETEST procedure of SAS. Univariate and multivariate logistic regression models were performed to identify potential relationships of biomarkers with early death as defined by death occurring within the follow up time of one year (n = 28). A univariate model was used to determine the individual associations and a bivariate analysis was used to detect any independent association between SAA or ARE and survival below one year. 3. Results Table 1 shows the distribution of the three paraoxonase activities and lipids across the quartiles of CRP and SAA. HDL and ARE significantly decrease across the quartiles of CRP. No significant difference was observed for any lipids or paraoxonase activity across SAA quartiles. A tendency was observed for HDL and LAC but this did not reach significance. A shown in Fig. 1, the survival time was significantly related to the level of SAA (log rank: p b 0.001) as well as the level of ARE (log rank: p = 0.039). Quartile 1 differed from quartile 4 with a hazard ratio of 0.206 [0.072–0.588], p = 0.0032 for SAA and with a hazard ratio of 5.48 [1.48–20.36], p = 0.011 for ARE. In addition the bivariate logistic regression analysis indicated that both SAA (odds ratio: 2.24 [1.21–4.15], p = 0.01) and ARE (odds ratio: 0.49 [0.26–0.92], p = 0.03) were independently related to survival time of less than one year (n = 28). 4. Discussion The present study was undertaken to determine if inflammation observed in breast cancer recurrence, leads to altered PON1 activity. The aim of the study was not to determine if patients with breast cancer differ from healthy subjects for PON1 activity and we did not run a control group. It is therefore not possible to conclude that all PON1 activities are decreased in our patients. In vitro, it was shown that PON1 mRNA was reduced following stimulation with inflammatory cytokines [11].
Table 1 Distribution of paraoxonase activities and lipids, across quartiles of CRP and SAA. CRP (mg/L)
[0.8–1.8[
[1.8–8.6[
[8.6–25.8[
[25.8–279.2]
P⁎
PON (μmol/L/min) ARE (mmol/L/min) LAC (μmol/L/min) HDL-C (mmol/L) LDL-C (mmol/L) TG (mmol/L) oxLDL (AU) oxLDL/LDL-C (AU)
114 [36–308] 31.0 [20.1–51.3] 0.21 [0.06–0.32] 1.65 [1.32–2.29] 3.01 [2.43–4.88] 1.19 [0.50–2.26] 48.6 [39.4–95.1] 16.6 [11.6–25.3]
122 [51–225] 27.9 [11.7–51.3] 0.18 [0.06–0.25] 1.59 [0.69–1.98] 3.51 [1.59–4.93] 1.26 [0.87–2.34] 51.2 [36.7–91.2] 15.6 [11.4–26.9]
60 [31–374] 21.9 [13.7–33.0] 0.19 [0.08–0.25] 1.54 [1.00–2.02] 3.23 [2.60–4.67] 1.10 [0.73–2.53] 50.2 [36.1–74.1] 15.3 [11.4–19.6]
64 [37–148] 24.5 [11.7–33.0] 0.18 [0.10–0.28] 1.16 [0.93–1.66] 3.31 [1.94–5.74] 1.31 [0.73–4.84] 50.8 [30.8–137.8] 18.9 [12.9–35.9]
0.30 0.02 0.21 0.01 0.67 0.82 0.98 0.31
SAA (μg/L)
[6.8–28.6[
[28.6–54.5[
[54.5–149.3[
[149.3–1200]
P⁎
PON (μmol/L/min) ARE (mmol/L/min) LAC (μmol/L/min) HDL-C (mmol/L) LDL-C (mmol/L) TG (mmol/L) oxLDL (AU) oxLDL/LDL-C (AU)
100 [44–207] 25.7 [11.7–51.3] 0.18 [0.06–0.32] 1.43 [0.69–2.09] 3.21 [1.59–4.20] 1.19 [0.50–2.34] 47.9 [36.7–72.4] 16.1 [11.6–23.1]
113 [35–374] 25.8 [15.3–43.4] 0.20 [0.08–0.25] 1.66 [1.00–2.01] 3.36 [2.88–4.88] 1.34 [0.73–2.53] 58.6 [36.1–95.1] 15.0 [11.4–26.9]
61 [37–249] 23.5 [18.9–43.4] 0.21 [0.16–0.29] 1.51 [0.99–2.29] 3.18 [1.94–5.74] 1.66 [0.73–4.84] 53.6 [43.6–137.8] 19.1 [11.4–30.0]
93 [31–235] 25.2 [11.7–33.0] 0.15 [0.10–0.26] 1.16 [0.93–1.66] 3.31 [1.94–5.72] 1.20 [0.80–1.90] 47.2 [30.8–81.7] 14.8 [12.3–35.9]
0.69 0.56 0.07 0.08 0.66 0.66 0.23 0.33
Median [5th–95th]. ⁎ Kruskal–Wallis test.
Please cite this article as: C. Bobin-Dubigeon, et al., Paired measurement of serum amyloid A (SAA) and paraoxonase 1 (PON1) as useful markers in breast cancer recurrence, Clin Biochem (2015), http://dx.doi.org/10.1016/j.clinbiochem.2015.07.020
C. Bobin-Dubigeon et al. / Clinical Biochemistry xxx (2015) xxx–xxx
3
no change in oxidized LDL was observed, suggesting that the PON1 anti-oxidative activity was not really affected. The reason why ARE activity would be the only affected substrate hydrolysis function of the enzyme is unclear. We could raise the hypothesis that some cytokines or substrates generated during inflammation would decrease the hydrolysis rate by competing with the substrate or by modifying the enzyme conformation. This hypothesis would need further in vitro studies to be confirmed. However, our study has some limitations and definitive conclusion is not possible. The number of patients is relatively low and it should be kept in mind that the absence of significance for some substrate hydrolysis activity could just be related to a lack of power. In this particular group of patients with a high short term mortality rate, we show that high levels of SAA and low levels of ARE are clearly associated with shorter survival time. As a previous study by Tanimoto et al. indicated that the ratio between SAA and PON1 could be increased in case of inflammatory disease [7] it was suggested by others [8] that the combined determination of PON1 and SAA may represent a useful tool in clinical practice. However, if SAA and ARE were strongly negatively correlated, it would be expected that the association between these two parameters and short term death would not be independent. On the contrary, in our bivariate logistic regression analysis, we observed that both SAA and ARE are independently related to death occurring within one year. This would suggest that the combined measurement of these two parameters could be of clinical interest. However, PON1 enzyme activity is currently measured with “home made” methods and SAA is available only as ELISA kits. This represents a limitation for routine application yet. In summary, this study indicates that not all substrate hydrolysis rates of PON1 are inversely related to inflammation and that the combined measurement of SAA and PON1 (arylesterase activity) could be useful for clinical assessment of patients with breast cancer recurrence. References
Fig. 1. Survival time curves according to SAA (panel A) and ARE (panel B) quartiles and odds ratio and [5th–95th] Wald confidence intervals for one quartile of SAA and ARE in the bivariate logistic regression analysis with death occurring within one year of follow up as the explained variable (panel C).
However, acute phase proteins, such as CRP and SAA do not seem to exert any direct effect on PON1 expression. Nevertheless, it was suggested that the increase of SAA in inflammatory HDL and the following decrease of apolipoprotein (apo) AI could lead to a reduction in circulating PON1 [12]. Our results indicate that not all substrate hydrolysis activities of PON1 are affected by inflammation. Only the ARE activity decreases in presence of inflammation. Therefore, we cannot argue in favor of any direct effect of inflammation on the enzyme expression. In addition, the relationship between ARE and inflammation markers is observed when CRP but not SAA is considered. This observation confirms what we have already shown in a population of HIV infected patients [13]. It does not fit with the hypothesis of a direct effect of SAA which would take the place of apo AI in HDL and subsequently reduce PON1. Moreover, although HDL decreases with inflammation,
[1] O. Khersonsky, D.S. Tawfik, Structure-reactivity studies of serum paraoxonase PON1 suggest that its native activity is lactonase, Biochemistry 44 (2005) 6371–6382. [2] M.I. Mackness, S. Arrol, P.N. Durrington, Paraoxonase prevents accumulation of lipoperoxides in low-density lipoprotein, FEBS Lett. 286 (1991) 152–154. [3] C.M. Uhlar, A.S. Whitehead, Serum amyloid A, the major vertebrate acute-phase reactant, Eur. J. Biochem. 265 (1999) 501–523. [4] B.N. Ames, Dietary carcinogens and anticarcinogens. Oxygen radicals and degenerative diseases, Science 221 (1983) 1256–1264. [5] M.N. Akcay, M.F. Polat, I. Yilmaz, G. Akcay, Serum paraoxonase levels in pancreatic cancer, Hepatogastroenterology 50 (Suppl. 2) (2003) ccxxv–ccxxvii. [6] C. Bobin-Dubigeon, I. Jaffré, M.-P. Joalland, et al., Paraoxonase 1 (PON1) as a marker of short term death in breast cancer recurrence, Clin. Biochem. 45 (2012) 1503–1505. [7] N. Tanimoto, Y. Kumon, T. Suehiro, et al., Serum paraoxonase activity decreases in rheumatoid arthritis, Life Sci. 72 (2003) 2877–2885. [8] K. Kotani, T. Yamada, A. Gugliucci, Paired measurements of paraoxonase 1 and serum amyloid A as useful disease markers, BioMed Res. Int. 2013 (2013) 481437. [9] J. Marsillach, G. Aragones, R. Beltran, et al., The measurement of the lactonase activity of paraoxonase-1 in the clinical evaluation of patients with chronic liver impairment, Clin. Biochem. 42 (2009) 91–98. [10] W.T. Friedewald, R.I. Levy, D.S. Fredrickson, Estimation of the concentration of lowdensity lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge, Clin. Chem. 18 (1972) 499–502. [11] Y. Kumon, Y. Nakauchi, T. Suehiro, et al., Proinflammatory cytokines but not acute phase serum amyloid A or C-reactive protein, downregulate paraoxonase 1 (PON1) expression by HepG2 cells, Amyloid 9 (2002) 160–164. [12] B.J. Van Lenten, S.Y. Hama, F.C. de Beer, et al., Anti-inflammatory HDL becomes proinflammatory during the acute phase response. Loss of protective effect of HDL against LDL oxidation in aortic wall cell cocultures, J. Clin. Investig. 96 (1995) 2758–2767. [13] C. Bobin-Dubigeon, C. Biron, C. Volteau, et al., Short communication: paraoxonase 1 (PON1) in French HIV-infected patients under antiretroviral therapy: relationship with the metabolic syndrome and inflammation, AIDS Res. Hum. Retrovir. 29 (2013) 1571–1574.
Please cite this article as: C. Bobin-Dubigeon, et al., Paired measurement of serum amyloid A (SAA) and paraoxonase 1 (PON1) as useful markers in breast cancer recurrence, Clin Biochem (2015), http://dx.doi.org/10.1016/j.clinbiochem.2015.07.020
