Atherosclerosis 232 (2014) 403e409
Contents lists available at ScienceDirect
Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis
Supplementation with carnosine decreases plasma triglycerides and modulates atherosclerotic plaque composition in diabetic apo E/ mice Bronwyn E. Brown a, f,1, Christine H.J. Kim a,1, Fraser R. Torpy b, Christina A. Bursill c, Lucinda S. McRobb d, 2, Alison K. Heather d, e, Michael J. Davies a, f, David M. van Reyk e, * a
Free Radical Group, Heart Research Institute, 7 Eliza Street, Newtown, NSW 2042, Australia School of the Environment, University of Technology, Sydney, PO Box 123, Broadway, NSW 2007, Australia Immunobiology Group, Heart Research Institute, 7 Eliza Street, Newtown, NSW 2042, Australia d Gene Regulation Group, Heart Research Institute, 7 Eliza Street, Newtown, NSW 2042, Australia e School of Medical and Molecular Biosciences, University of Technology, Sydney, PO Box 123, Broadway, NSW 2007, Australia f Faculty of Medicine, University of Sydney, NSW 2006, Australia b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 23 April 2013 Received in revised form 8 November 2013 Accepted 20 November 2013 Available online 18 December 2013
Objective: Carnosine has been shown to modulate triglyceride and glycation levels in cell and animal systems. In this study we investigated whether prolonged supplementation with carnosine inhibits atherosclerosis and markers of lesion stability in hyperglycaemic and hyperlipidaemic mice. Methods: Streptozotocin-induced diabetic apo E/ mice were maintained for 20 weeks, post-induction of diabetes. Half of the animals received carnosine (2 g/L) in their drinking water. Diabetes was confirmed by significant increases in blood glucose and glycated haemoglobin, plasma triglyceride and total cholesterol levels, brachiocephalic artery and aortic sinus plaque area; and lower body mass. Results: Prolonged carnosine supplementation resulted in a significant (w20-fold) increase in plasma carnosine levels, and a significant (w23%) lowering of triglyceride levels in the carnosine-supplemented groups regardless of glycaemic status. Supplementation did not affect glycaemic status, blood cholesterol levels or loss of body mass. In the diabetic mice, carnosine supplementation did not diminish measured plaque area, but reduced the area of plaque occupied by extracellular lipid (w60%) and increased both macrophage numbers (w70%) and plaque collagen content (w50%). The area occupied by a-actin-positive smooth muscle cells was not significantly increased. Conclusions: These data indicate that in a well-established model of diabetes-associated atherosclerosis, prolonged carnosine supplementation enhances plasma levels, and has novel and significant effects on atherosclerotic lesion lipid, collagen and macrophage levels. These data are consistent with greater lesion stability, a key goal in treatment of existing cardiovascular disease. Carnosine supplementation may therefore be of benefit in lowering triglyceride levels and suppressing plaque instability in diabetesassociated atherosclerosis. Ó 2013 Elsevier Ireland Ltd. All rights reserved.
Keywords: Carnosine Diabetes Atherosclerosis Cholesterol Triglycerides
1. Introduction
* Corresponding author. Tel.: þ61 2 95142221; fax: þ61 2 95148206. E-mail addresses: [email protected] (B.E. Brown), [email protected] (C.H.J. Kim), [email protected] (F.R. Torpy), [email protected] (C.A. Bursill), [email protected] (L.S. McRobb), [email protected] (A.K. Heather), [email protected] (M.J. Davies), [email protected] (D.M. van Reyk). 1 These authors contributed equally to the work of the project. 2 Present address: Centre for Vascular Research, University of New South Wales, NSW 2052, Australia. 0021-9150/$ e see front matter Ó 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.atherosclerosis.2013.11.068
The impact of diabetes on cardiovascular health [1] has resulted in significant interest in therapies designed to prevent, inhibit or reverse diabetes-associated atherosclerosis. The apolipoprotein E knock out (apo E/) mouse model [2] has been used extensively to evaluate such approaches and is characterised by hyperglycaemia and dyslipidaemia that contribute to atherosclerosis development. Hyperglycaemia may act through multiple mechanisms including increased production of advanced glycation end products (AGE) mediated by reactive aldehydes and reactive oxidants [3].
404
B.E. Brown et al. / Atherosclerosis 232 (2014) 403e409
Carnosine is an endogenous b-alanyl-L-histidine dipeptide present at high concentrations in several organs including skeletal muscle. Carnosine has been suggested to have multiple functions including as a buffer; as a regulator of Ca2þ sensitivity; as an antioxidant; or by preventing or reversing glycation- and glycoxidation-induced modifications [4,5]. In vitro, carnosine can block both pro-atherosclerotic aldehyde-mediated modifications and copper-mediated oxidation of low-density lipoproteins [6,7]. The inhibitory effect of intact carnosine on aldehyde-mediated LDL modification has been reported as being greater than for its constituents b-alanine and histidine [6,7]. The potential of carnosine and analogues to regulate lipid and lipoprotein metabolism has been investigated, with studies using fat-fed apo E/ mice demonstrating an anti-atherosclerotic activity of D-carnosine octyl ester [6,8]. Carnosine also blocks the increases in plasma triglycerides and cholesterol in fat-fed, nondiabetic, C57BL/6 mice [9]. As an endogenous peptide, carnosine is safe for oral delivery. It is absorbed as the intact dipeptide with hydrolysis in the intestinal mucosa readily saturable [4]. However, human plasma carnosine levels are low due to rapid hydrolysis by carnosinase [4]. Synthetic derivatives are less susceptible to hydrolysis [10], and may have therapeutic potential. In contrast, chronic supplementation of rodents results in persistent and significant elevations of both plasma and tissue (e.g. aorta, heart, kidney, liver and spleen) levels [6,8,9,11,12]; this therefore provides a suitable model to examine the potential of carnosine as an anti-glycation/anti-dyslipidaemic agent. Given the role of enhanced glycation and dyslipidaemia in diabetes-associated atherosclerosis [3], the current study assesses, for the first time, the in vivo anti-atherogenic potential and plaquemodulating potential of chronic supplementation of diabetic animals with carnosine. 2. Methods 2.1. Reagents Chemicals and proteins were purchased from the following sources: aprotinin, butylated hydroxytoluene, EDTA, o-phthaldialdehyde, paraformaldehyde, soybean trypsin inhibitor and streptozotocin (Sigma, St Louis, MO, USA); carnosine (Triton Chemtech Co. Ltd, Shanghai, China); D-phenylalanyl-L-prolyl-L-arginine chloromethyl ketone (PPACK, Calbiochem, KGaA, Darmstadt, Germany); methoxyflurane (Medical Developments International Ltd, Springvale, VIC, Australia); phosphate buffered saline (PBS, Amresco Inc. (Solon, OH, USA)). Solutions were prepared using nanopure water (Milli Q system, Millipore Waters, Billerica, MA, USA) except for injections where pyrogen-free 0.9% w/v saline was used (Baxter Health Care, Old Toongabbie, NSW, Australia). 2.2. Mice Eighty male apo E/ mice (50% C57BL/6, 50% 129SvJ; aged 5e9 weeks) were maintained and monitored in accordance with the Sydney South West Area Health Service Animal Welfare Committee (approval #2007/002). Diabetes was induced in half of the mice via a single intraperitoneal injection of streptozotocin (STZ; 55 mg/kg body mass) on five consecutive days [13]. Control mice received matched volumes of vehicle (sodium citrate buffer, pH 4.5). Post-treatment, half of each group (diabetes and controls) received carnosine (2 g/L) in drinking water to increase plasma carnosine levels. Previous studies have reported increases in plasma carnosine at 1.0 but not 0.5 g carnosine/L drinking water over four weeks [12]. Food (meat-free rat and mice diet, Speciality Feeds, Glen Forrest, WA, Australia: protein (20.0 %),
total fat (4.8%), total carbohydrate (59.4%)) and water were available ad libitum. 2.3. Sample collection and storage At 20 weeks after diabetes induction, and following a 6-h premorning fast, all mice were weighed, anaesthetised using methoxyflurane delivered by nose cone, and euthanised by cardiac exsanguination. Blood was immediately transferred to vials containing a protease inhibitor cocktail (1 mL/mL aprotinin, 0.04 mmoles/L PPACK, 20 mg/mL soybean trypsin inhibitor, 2 mmoles/L EDTA), and used to determine blood glucose and glycated haemoglobin levels (see below). The remaining material was centrifuged (800g, 10 min, 4 C) to obtain plasma that was snapfrozen in liquid N2 and stored at 80 C until analysed. The heart and arterial tree were collected for histological analysis by perfusion through the left ventricle, at physiological pressure, with PBS, containing 2 mmoles/L EDTA and 20 mmoles/L butylated hydroxytoluene, for 4 min, followed with fixation by perfusion with 4 % (w/v) paraformaldehyde (pH 7.5) for 6 min [14]. The tissue was then dissected and stored in paraformaldehyde overnight, then rinsed in PBS and stored in 70% v/v ethanol. 2.4. Determination of blood glucose and glycated haemoglobin Blood glucose was quantified using a True TrackÒ smart system glucometer calibrated using a Roche Modular P800 auto-analyser (the Automated Laboratory, Biochemistry Dept., Royal Prince Alfred Hospital, Sydney). Glycated haemoglobin was determined using A1cNowþ kits (Metrika Inc., Sunnyvale, CA, USA). 2.5. Determination of blood cholesterol and triglycerides Total cholesterol and triglycerides were quantified using commercial kits (Total Cholesterol Kit, Roche Diagnostics, Basel, Switzerland; TG Assay, Novachem, from Merck KGaA, Darmstadt, Germany) in accordance with the manufacturer’s instructions. 2.6. Determination of plasma carnosine levels Plasma-derived low molecular mass filtrates ( 0.05) indicated no linear relationship between plasma carnosine and plaque area for mice of the same glycaemic status. Diabetes was associated with a significant decrease in plaque collagen content and an increase in area occupied by extracellular
Table 1 The measured group parameters at time of sacrifice are presented. For mean body mass, total plasma cholesterol and triglyceride plasma levels of the non-diabetic (control) and diabetic apo E / mice (with or without carnosine (CN) supplementation) groups at the time of sacrifice: results are expressed as mean standard error of the mean (SEM) for 12e19 mice per group. Plasma carnosine levels (mmoles/L), brachiocephalic artery and aortic sinus plaque areas (mm2) are presented as the 95% confidence interval together with the mean (n ¼ 12e18). * Significantly different (p < 0.05) from control group at the same level of carnosine supplementation; and # significantly different (p < 0.05) from the non-supplemented group matched for glycaemic status with data transformed so that it conformed to a normal distribution when necessary.
Mass (g) Total cholesterol (mmoles/L) Triglycerides (mmoles/L) Plasma carnosine (mmoles/L) Brachiocephalic artery plaque area (mm2) Aortic sinus plaque area (mm2)
Control
Control with CN
Diabetic
Diabetic with CN
27.8 0.6 9.6 0.5 1.1 0.1 1.2 0.3 0.045 0.009 0.138 0.013
29.9 0.5 9.0 0.3 0.8 0.1# 8.7 3.4# 0.040 0.009 0.134 0.012
24.6 0.9* 14.6 2.1* 1.7 0.2* 0.7 0.2 0.062 0.010* 0.152 0.016*
24.2 0.5* 15.2 0.7* 1.3 0.1*# 23.6 11.6# 0.059 0.006* 0.150 0.015*
406
B.E. Brown et al. / Atherosclerosis 232 (2014) 403e409
Fig. 1. Blood glucose levels (A) and glycated haemoglobin levels (B), at the time of sacrifice, for control (CTL) and diabetic (DIAB) apo E/ mice with ( þ CN) or without carnosine supplementation. Individual determinations (n ¼ 12e18) are presented, together with the median value (horizontal bar). Out of range results (>33 mmoles/L for blood glucose; and 13% for glycated haemoglobin) are represented as markers placed on the dashed line that delimits the range of detection. A number of animals with diabetes have superimposed data points. * Significantly different (p < 0.05) from control group at the same level of carnosine supplementation using rank-transformed data
lipid in the diabetic animals. Carnosine supplementation blocked these changes. That is, the collagen content and extracellular lipid content of the plaques from the diabetic carnosine-supplemented mice was found to be no different from those of the plaques from their non-diabetic littermates (Figs. 2 and 3). No significant differences were detected in plaque area occupied by a-actin-positive cells at either site (% of total plaque area {median, 1st and 3rd quartile}: CTL {3, 0, 0}; CTL þ CB {2, 1, 4}; DIAB {1, 0, 2}; DIAB þ CN {3, 2, 5}). For the brachiocephalic arteries there was a significant reduction in the area occupied by F4/80-positive cells with diabetic mice but this was unaffected by carnosine
supplementation, while at the aortic sinus there was a significant reduction for non-supplemented diabetic mice only (Fig. 4). 3.4. Multivariate analysis The analysis detected a clear differentiation between the four groups based on simultaneous covariation in the data variables, although there was some overlap between control and control þ carnosine animals. ANOSIM detected a significant difference amongst treatment groups, based on the simultaneous variance in the 16 data variables (Global R ¼ 0.265; p ¼ 0.001). All pairwise
Fig. 2. Panel A: representative images of atherosclerotic plaques stained with picrosirius red in the aortic sinus cross-sections of control (CTL), control þ carnosine (CTL þ CN), diabetes (DIAB) and diabetes þ carnosine (DIAB þ CN) mice. Panel B: Collagen content (as a % of plaque area), at the time of sacrifice, for control (CTL) and diabetic (DIAB) apo E/ mice with ( þ CN) or without carnosine supplementation. Data are presented as the 95% confidence interval together with the mean (n ¼ 12e18) for both the brachiocephalic and aortic sinus. * Significantly different (p < 0.05) from other groups using GLM ANOVA.
B.E. Brown et al. / Atherosclerosis 232 (2014) 403e409
407
Fig. 3. Panels A and B: extracellular lipid content of atherosclerotic plaques was estimated by measuring the clear areas on a section stained with F4/80 as shown on panel A. The section of interest (bordered by a dashed line) is magnified in panel B with arrows indicating example of lipid pools (false coloured red for image analysis) that were used for estimation of plaque lipid content. Panel C: lipid deposits (as a % of plaque area), at the time of sacrifice, for control (CTL) and diabetic (DIAB) apo E/ mice with ( þ CN) or without carnosine supplementation. Data are presented as box plots bordered by the 1st and 3rd quartile value with the median indicated by a horizontal bar: the range is shown by the vertical lines. # Significantly different (p < 0.05) from other treatments using a KruskaleWallis test.
comparisons amongst the four groups were significant (p ¼ 0.001 in all cases) except between the control and control þ carnosine mice (p ¼ 0.449). SIMPER analysis indicated the data variables which best differentiate between groups and thus, by inference, the strongest effects of diabetes induction and carnosine supplementation. Only data variables that contributed greater than 10% of the total dissimilarity were considered. The data variable that best distinguished between un-supplemented non-diabetic and unsupplemented diabetic mice was blood glucose, followed by glycated haemoglobin and aortic sinus lipid core plaque area (Supplementary Table 1). The variables that contributed most to the differences between carnosine-supplemented non-diabetic and the matched supplemented diabetic mice were blood glucose and glycated haemoglobin (Supplementary Table 1). When comparing the two diabetic groups the variable that best distinguished these un-supplemented vs. the supplemented mice was the greatly reduced aortic sinus lipid core plaque area in the carnosinesupplemented group (Supplementary Table 1). No other data
variable contributed more than 10% to the total dissimilarity between these groups, although there were also large differences between the brachial artery lipid core, as a % of plaque area, between these two groups (8.1%). 4. Discussion Carnosine has been investigated extensively as a potential therapy for diseases in which oxidation and/or glycation is believed to contribute, including atherosclerosis [6,8]. Whilst carnosine administration to humans yields only transient increases in plasma levels, the current (Table 1) and previous studies [9,12] have demonstrated sustained increases in rodents after supplementation via drinking water. The carnosine levels achieved here are higher than those reported previously [12] probably due to the higher dose employed. Studies using Zucker rats have demonstrated significant increases in plasma and kidney carnosine levels with chronic supplementation with D- but not L-carnosine [11],
Fig. 4. Percentage positive staining for F/480 (macrophages) in the brachiocephalic artery and aortic sinus plaques, at the time of sacrifice, for control and diabetic apo E/ mice with or without carnosine supplementation. Data are presented as box plots bordered by the 1st and 3rd quartile value with the median indicated by a horizontal bar: the range is shown by the vertical lines. (n ¼ 11e19). * Significantly different (p < 0.05) from control with the same level of carnosine supplementation using GLM for the square roottransformed data. # Significantly different (p < 0.05) from control with the same level of carnosine supplementation using a KruskaleWallis test.
408
B.E. Brown et al. / Atherosclerosis 232 (2014) 403e409
suggesting that the potential to elevate plasma L-carnosine is species-dependent. The wider range of plasma carnosine levels detected in the diabetic animals when compared to controls is likely to be due to the polydipsia of diabetes, with a three-fold higher level of water consumption observed in the diabetic animals (data not shown). Carnosine levels in other tissues were not analysed as the paraformaldehyde used in tissue fixation, reacts with carnosine. However a previous study has reported increased levels in the hearts and livers of mice supplemented with carnosine using a similar protocol [12]. In the current study, prolonged carnosine administration did not modulate blood glucose (Fig. 1A) or glycated haemoglobin (Fig. 1B) levels, although a previous study [12] has demonstrated effects on blood glucose levels after 4 weeks of supplementation. Transient hyperglycaemia induced by 2-deoxy-D-glucose in rats has also been reported to be inhibited by intravenous and intraperitoneal carnosine administration to give submicromolar plasma concentrations [20]. We conclude that chronic high dose supplementation of carnosine may override the more transient and low dose effects of carnosine upon blood glucose reported previously [12,20]. Whilst carnosine supplementation had no effect upon the hypercholesterolaemia in the diabetic mice, supplementation significantly reduced plasma triglycerides (Table 1); a similar reduction was seen in the control, carnosine-supplemented mice. These data suggest that the effect of carnosine on triglyceride levels is independent of changes induced by diabetes. A differential effect of carnosine on triglycerides and cholesterol has been reported previously in both diabetic and non-diabetic mice [12] with both concentrations of carnosine examined (0.5 or 1.0 g/L in drinking water) blunting the increase in heart and liver triglycerides whereas effects on cholesterol were only seen at the higher dose [12]. Both carnosine and histidine (1 g/L) block increases in plasma triglycerides and cholesterol in fat-fed, non-diabetic, C57BL/6 mice [9] whereas supplementation of fat-fed apo E/ mice with D-carnosine octyl ester for 6 weeks did not reduce plasma cholesterol levels [6]. In contrast, supplementation of hyperlipidaemic (but not hyperglycaemic) Zucker rats with L- or D-carnosine resulted in matched and significant decreases in both plasma cholesterol and triglyceride [11]. Thus while carnosine appears to have a consistent cross-species hypotriglyceridaemic action, this is not true for cholesterol levels. As both diabetes and elevated triglycerides and cholesterol are associated with increased atherosclerosis, we assessed lesion parameters in both the brachiocephalic artery and aortic sinus as these are major sites of development of advanced, rupture-prone, lesions [2]. Diabetes resulted in a significant increase in plaque area at both sites (Table 1) and despite the triglyceride-lowering effects of carnosine; supplementation did not reduce plaque area when compared to the non-supplemented mice matched for glycaemic status. A similar observation has been reported for diabetic apo E/ mice supplemented with conjugated linoleic acid, where no decrease in plaque area was detected despite a substantial lowering of plasma triglycerides [19]. Using this same model, others have reported decreased aortic sinus plaque area on treatment with the PPARg-agonist rosiglitazone, while drug treatment resulted in an elevation of blood triglyceride and cholesterol levels [21]. Thus modulation of hypertriglyceridaemia in this model does not appear to consistently impact on lesion size. These results contrast with recent studies in which D-carnosine octyl ester supplementation of fat-fed apo E/ mice resulted in a reduction in plaque area [6,8]. However, in common with Menini et al [8], carnosine appears to modify plaque composition, and specifically reduce extracellular lipid levels and increase collagen content (Figs. 2 and 3). Extracellular lipid levels within lesions reflect cholesterol precipitation from lipidladen macrophages [22], a process resulting in apoptosis and
necrosis [23]. Thus the reduction in lipid content and higher collagen content (a marker of extracellular matrix levels and hence lesion stability) induced by carnosine appears to reflect a more stable and less advanced plaque phenotype. Consistent with this, the lesion content of macrophages was higher in the carnosine treated group, suggesting that the macrophages within these plaques had not proceeded to apoptosis/necrosis [24]. These findings suggest that carnosine alters plaque composition to a more stable phenotype. The mechanism by which carnosine mediates these changes has not been determined definitively. Carnosine can modulate oxidation and glycation reactions by acting as a scavenger of both radicals and reactive aldehydes (e.g. methylglyoxal and the lipid-derived aldehydes, malondialdehyde and 4-hydroxynonenal) [4]. Thus we and others have shown that carnosine can block (pro-atherosclerotic) aldehyde-mediated modification of low-density lipoproteins as well as copper ion-mediated oxidation [6,7]. As only modest (low mmole/ L) levels plasma levels of carnosine were achieved by supplementation, it is unlikely that radical scavenging is the mode of action, as carnosine is unlikely to compete with other biological targets (e.g. plasma proteins, cf. data in Refs. [25,26]). Detoxification of reactive aldehydes, such as methylgloxal, malondialdehyde and 4-hydroxynonenal, is a more likely mode of action for carnosine, as these aldehydes have longer biological halflives [27] and hence are more likely to encounter carnosine and can be removed by adduction. Reactive aldehydes form adducts with proteins [28], converts LDL into a high-uptake form [29] and have pro-inflammatory activity (e.g. via modulation of the Nrf2 pathway, PPAR-g, haem oxygenase-1 and CD36 expression) [30]. Thus the removal of reactive carbonyls by reaction with carnosine may account for the anti-inflammatory effects of carnosine resulting in a dampening of the low-grade chronic inflammation of the artery wall induced by diabetes and/or a high fat diet [23] and give rise to a more stable lesion phenotype. Carnosine and its analogues have been shown to have antihyperlipidaemic, anti-atherosclerotic, hepato- and renal-protective actions in models of obesity-associated complications [6,8,11]. The current study demonstrates that this well-tolerated dipeptide also modulates diabetes-associated atherosclerosis. The impact of this compound appears to occur via decreases in blood triglyceride levels which are associated with changes in plaque characteristics and greater plaque stability. This data supports further investigation of the use of carnosine (or stable analogues) as an adjunct to current treatments for the vascular complications of diabetes. Conflict of interest The authors have declared no conflict of interest. Funding for this project was provided by an unrestricted grant from Eli Lilly and Company. Acknowledgements The authors thank Herbert Ayala, Kim Berry, Edwin Martin, Gary Martinic and Pam Sheahan for their assistance with animal management and sample collection; Mahmoud Al-hinti and Janice Pamment (Biochemistry Department, Royal Prince Alfred Hospital) for the blood glucose analyses, and Liming Hou and Professor Kerry-Anne Rye for the cholesterol and triglyceride analysis. Christine HJ Kim was a recipient of a University Postgraduate Award from the University of Sydney. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.atherosclerosis.2013.11.068.068.
B.E. Brown et al. / Atherosclerosis 232 (2014) 403e409
References [1] Bianchi C, Miccoli R, Daniele G, Penno G, Del Prato S. Is there evidence that oral hypoglycemic agents reduce cardiovascular morbidity/mortality? Yes. Diabetes Care 2009;32(Suppl. 2):S342e8. [2] Hsueh WA, Abel ED, Breslow JL, et al. Recipes for creating animal models of diabetic cardiovascular disease. Circ Res 2007;100:1415e27. [3] Jandeleit-Dahm K, Cooper ME. The role of AGEs in cardiovascular disease. Curr Pharm Des 2008;14:979e86. [4] Boldyrev AA, Aldini G, Derave W. Physiology and pathophysiology of carnosine. Physiol Rev 2013;93:1803e45. [5] Hipkiss AR. Carnosine and its possible roles in nutrition and health. Adv Food Nutr Res 2009;57:87e154. [6] Barski OA, Xie Z, Baba SP, et al. Dietary carnosine prevents early atherosclerotic lesion formation in apolipoprotein E-null mice. Arterioscler Thromb Vasc Biol 2013;33:1162e70. [7] Rashid I, van Reyk DM, Davies MJ. Carnosine and its constituents inhibit glycation of low-density lipoproteins that promote foam cell formation in vitro. FEBS Lett 2007;581:1067e70. [8] Menini S, Iacobini C, Ricci C, et al. D-carnosine octylester attenuates atherosclerosis and renal disease in ApoE null mice fed a Western diet through reduction of carbonyl stress and inflammation. Br J Pharmacol 2012;166:1344e56. [9] Mong M-C, Chao C-Y, Yin M-C. Histidine and carnosine alleviated hepatic steatosis in mice consumed high saturated fat diet. Eur J Pharmacol 2011;653:82e8. [10] Stvolinsky SL, Bulygina ER, Fedorova TN, et al. Biological activity of novel synthetic derivatives of carnosine. Cell Mol Neurobiol 2010;30:395e404. [11] Aldini G, Orioli M, Rossoni G, et al. The carbonyl scavenger carnosine ameliorates dyslipidemia and renal function in zucker obese rats. J Cell Mol Med 2011;15:1339e56. [12] Lee Y-T, Hsi C-C, Lin M-H, Liu K-S, Yin M-C. Histidine and carnosine delay diabetic deterioration in mice and protect human low density lipoprotein against oxidation and glycation. Eur J Pharmacol 2005;513:145e50. [13] Candido R, Jandeleit-Dahm KA, Cao Z, et al. Prevention of accelerated atherosclerosis by angiotensin-converting enzyme inhibition in diabetic apolipoprotein E-deficient mice. Circulation 2002;106:246e53. [14] McRobb L, Handelsman DJ, Heather AK. Androgen-induced progression of arterial calcification in apolipoprotein E-null mice is uncoupled from plaque growth and lipid levels. Endocrinology 2009;150:841e8. [15] Pattison DI, Hawkins CL, Davies MJ. Hypochlorous acid-mediated protein oxidation: how important are chloramine transfer reactions and protein tertiary structure? Biochemistry 2007;46:9853e64.
409
[16] Ali ZA, Bursill CA, Hu Y, et al. Gene transfer of a broad spectrum CC-chemokine inhibitor reduces vein graft atherosclerosis in apolipoprotein E-knockout mice. Circulation 2005;112:I-235e41. [17] Park L, Raman KG, Lee KJ, et al. Suppression of accelerated diabetic atherosclerosis by the soluble receptor for advanced glycation endproducts. Nat Med 1998;4:1025e31. [18] Candido R, Allen TJ, Lassila M, et al. Irbesartan but not amlidipine suppresses diabetes-associated atherosclerosis. Circulation 2004;109:1536e42. [19] Nestel P, Fujii A, Allen T. The cis-9,trans-11 isomer of conjugated linoleic acid (CLA) lowers plasma triglyceride and raises HDL cholesterol concentrations but does not suppress aortic atherosclerosis in diabetic apoE-deficient mice. Atherosclerosis 2006;189:282e7. [20] Yamano T, Niijima A, Iimori S, Tsuruoka N, Kiso Y, Nagai K. Effect of L-carnosine on the hyperglycemia caused by intracranial injection of 2-deoxy-Dglucose in rats. Neurosci Lett 2001;313:78e82. [21] Levi Z, Shaish A, Yacov N, et al. Rosiglitazone (PPAR-g-agonist) attenuates atherogenesis with no effect on hyperglycaemia in a combined diabetesatherosclerosis mouse model. Diabetes Obes Metab 2003;5:45e50. [22] Small DM, Shipley GG. Physicalechemical basis of lipid deposition in atherosclerosis. Science 1974;185:222e9. [23] Libby P. Inflammation in atherosclerosis. Nature 2002;420:868. [24] Reimers GJ, Jackson CL, Rickards J, et al. Inhibition of rupture of established atherosclerotic plaques by treatment with apolipoprotein A-I. Cardiovasc Res 2011;91:37e44. [25] Davies MJ. The oxidative environment and protein damage. Biochim Biophys Acta 2005;1703:93e109. [26] Gebicki S, Bartosz B, Gebicki JM. Peroxidation of proteins by free radicals and its consequences to living organisms. Curr Topics Biophys 1995;1995: 220e6. [27] Lo TWC, Selwood T, Thornalley PJ. The reaction of methylglyoxal with aminoguanidine under physiological conditions and prevention of methylglyoxal binding to plasma proteins. Biochem Pharmacol 1994;48:1865e70. [28] Requena JR, Fu M-X, Ahmed MU, et al. Quantification of malondialdehyde and 4-hydroxynonenal adducts to lysine residues in native and oxidised human low-density lipoprotein. Biochem J 1997;322:317e25. [29] Hoff HF, O’Neil J, Chisolm GD, et al. Modification of low density lipoprotein with 4-hydroxynonenal induces uptake by macrophages. Arteriosclerosis 1989;9:538e49. [30] Ishii T, Itoh K, Ruiz E, et al. Role of Nrf2 in the regulation of CD36 and stress protein expression in murine macrophages: activation by oxidatively modified LDL and 4-hydroxynonenal. Circ Res 2004;94:609e16.
Contents lists available at ScienceDirect
Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis
Supplementation with carnosine decreases plasma triglycerides and modulates atherosclerotic plaque composition in diabetic apo E/ mice Bronwyn E. Brown a, f,1, Christine H.J. Kim a,1, Fraser R. Torpy b, Christina A. Bursill c, Lucinda S. McRobb d, 2, Alison K. Heather d, e, Michael J. Davies a, f, David M. van Reyk e, * a
Free Radical Group, Heart Research Institute, 7 Eliza Street, Newtown, NSW 2042, Australia School of the Environment, University of Technology, Sydney, PO Box 123, Broadway, NSW 2007, Australia Immunobiology Group, Heart Research Institute, 7 Eliza Street, Newtown, NSW 2042, Australia d Gene Regulation Group, Heart Research Institute, 7 Eliza Street, Newtown, NSW 2042, Australia e School of Medical and Molecular Biosciences, University of Technology, Sydney, PO Box 123, Broadway, NSW 2007, Australia f Faculty of Medicine, University of Sydney, NSW 2006, Australia b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 23 April 2013 Received in revised form 8 November 2013 Accepted 20 November 2013 Available online 18 December 2013
Objective: Carnosine has been shown to modulate triglyceride and glycation levels in cell and animal systems. In this study we investigated whether prolonged supplementation with carnosine inhibits atherosclerosis and markers of lesion stability in hyperglycaemic and hyperlipidaemic mice. Methods: Streptozotocin-induced diabetic apo E/ mice were maintained for 20 weeks, post-induction of diabetes. Half of the animals received carnosine (2 g/L) in their drinking water. Diabetes was confirmed by significant increases in blood glucose and glycated haemoglobin, plasma triglyceride and total cholesterol levels, brachiocephalic artery and aortic sinus plaque area; and lower body mass. Results: Prolonged carnosine supplementation resulted in a significant (w20-fold) increase in plasma carnosine levels, and a significant (w23%) lowering of triglyceride levels in the carnosine-supplemented groups regardless of glycaemic status. Supplementation did not affect glycaemic status, blood cholesterol levels or loss of body mass. In the diabetic mice, carnosine supplementation did not diminish measured plaque area, but reduced the area of plaque occupied by extracellular lipid (w60%) and increased both macrophage numbers (w70%) and plaque collagen content (w50%). The area occupied by a-actin-positive smooth muscle cells was not significantly increased. Conclusions: These data indicate that in a well-established model of diabetes-associated atherosclerosis, prolonged carnosine supplementation enhances plasma levels, and has novel and significant effects on atherosclerotic lesion lipid, collagen and macrophage levels. These data are consistent with greater lesion stability, a key goal in treatment of existing cardiovascular disease. Carnosine supplementation may therefore be of benefit in lowering triglyceride levels and suppressing plaque instability in diabetesassociated atherosclerosis. Ó 2013 Elsevier Ireland Ltd. All rights reserved.
Keywords: Carnosine Diabetes Atherosclerosis Cholesterol Triglycerides
1. Introduction
* Corresponding author. Tel.: þ61 2 95142221; fax: þ61 2 95148206. E-mail addresses: [email protected] (B.E. Brown), [email protected] (C.H.J. Kim), [email protected] (F.R. Torpy), [email protected] (C.A. Bursill), [email protected] (L.S. McRobb), [email protected] (A.K. Heather), [email protected] (M.J. Davies), [email protected] (D.M. van Reyk). 1 These authors contributed equally to the work of the project. 2 Present address: Centre for Vascular Research, University of New South Wales, NSW 2052, Australia. 0021-9150/$ e see front matter Ó 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.atherosclerosis.2013.11.068
The impact of diabetes on cardiovascular health [1] has resulted in significant interest in therapies designed to prevent, inhibit or reverse diabetes-associated atherosclerosis. The apolipoprotein E knock out (apo E/) mouse model [2] has been used extensively to evaluate such approaches and is characterised by hyperglycaemia and dyslipidaemia that contribute to atherosclerosis development. Hyperglycaemia may act through multiple mechanisms including increased production of advanced glycation end products (AGE) mediated by reactive aldehydes and reactive oxidants [3].
404
B.E. Brown et al. / Atherosclerosis 232 (2014) 403e409
Carnosine is an endogenous b-alanyl-L-histidine dipeptide present at high concentrations in several organs including skeletal muscle. Carnosine has been suggested to have multiple functions including as a buffer; as a regulator of Ca2þ sensitivity; as an antioxidant; or by preventing or reversing glycation- and glycoxidation-induced modifications [4,5]. In vitro, carnosine can block both pro-atherosclerotic aldehyde-mediated modifications and copper-mediated oxidation of low-density lipoproteins [6,7]. The inhibitory effect of intact carnosine on aldehyde-mediated LDL modification has been reported as being greater than for its constituents b-alanine and histidine [6,7]. The potential of carnosine and analogues to regulate lipid and lipoprotein metabolism has been investigated, with studies using fat-fed apo E/ mice demonstrating an anti-atherosclerotic activity of D-carnosine octyl ester [6,8]. Carnosine also blocks the increases in plasma triglycerides and cholesterol in fat-fed, nondiabetic, C57BL/6 mice [9]. As an endogenous peptide, carnosine is safe for oral delivery. It is absorbed as the intact dipeptide with hydrolysis in the intestinal mucosa readily saturable [4]. However, human plasma carnosine levels are low due to rapid hydrolysis by carnosinase [4]. Synthetic derivatives are less susceptible to hydrolysis [10], and may have therapeutic potential. In contrast, chronic supplementation of rodents results in persistent and significant elevations of both plasma and tissue (e.g. aorta, heart, kidney, liver and spleen) levels [6,8,9,11,12]; this therefore provides a suitable model to examine the potential of carnosine as an anti-glycation/anti-dyslipidaemic agent. Given the role of enhanced glycation and dyslipidaemia in diabetes-associated atherosclerosis [3], the current study assesses, for the first time, the in vivo anti-atherogenic potential and plaquemodulating potential of chronic supplementation of diabetic animals with carnosine. 2. Methods 2.1. Reagents Chemicals and proteins were purchased from the following sources: aprotinin, butylated hydroxytoluene, EDTA, o-phthaldialdehyde, paraformaldehyde, soybean trypsin inhibitor and streptozotocin (Sigma, St Louis, MO, USA); carnosine (Triton Chemtech Co. Ltd, Shanghai, China); D-phenylalanyl-L-prolyl-L-arginine chloromethyl ketone (PPACK, Calbiochem, KGaA, Darmstadt, Germany); methoxyflurane (Medical Developments International Ltd, Springvale, VIC, Australia); phosphate buffered saline (PBS, Amresco Inc. (Solon, OH, USA)). Solutions were prepared using nanopure water (Milli Q system, Millipore Waters, Billerica, MA, USA) except for injections where pyrogen-free 0.9% w/v saline was used (Baxter Health Care, Old Toongabbie, NSW, Australia). 2.2. Mice Eighty male apo E/ mice (50% C57BL/6, 50% 129SvJ; aged 5e9 weeks) were maintained and monitored in accordance with the Sydney South West Area Health Service Animal Welfare Committee (approval #2007/002). Diabetes was induced in half of the mice via a single intraperitoneal injection of streptozotocin (STZ; 55 mg/kg body mass) on five consecutive days [13]. Control mice received matched volumes of vehicle (sodium citrate buffer, pH 4.5). Post-treatment, half of each group (diabetes and controls) received carnosine (2 g/L) in drinking water to increase plasma carnosine levels. Previous studies have reported increases in plasma carnosine at 1.0 but not 0.5 g carnosine/L drinking water over four weeks [12]. Food (meat-free rat and mice diet, Speciality Feeds, Glen Forrest, WA, Australia: protein (20.0 %),
total fat (4.8%), total carbohydrate (59.4%)) and water were available ad libitum. 2.3. Sample collection and storage At 20 weeks after diabetes induction, and following a 6-h premorning fast, all mice were weighed, anaesthetised using methoxyflurane delivered by nose cone, and euthanised by cardiac exsanguination. Blood was immediately transferred to vials containing a protease inhibitor cocktail (1 mL/mL aprotinin, 0.04 mmoles/L PPACK, 20 mg/mL soybean trypsin inhibitor, 2 mmoles/L EDTA), and used to determine blood glucose and glycated haemoglobin levels (see below). The remaining material was centrifuged (800g, 10 min, 4 C) to obtain plasma that was snapfrozen in liquid N2 and stored at 80 C until analysed. The heart and arterial tree were collected for histological analysis by perfusion through the left ventricle, at physiological pressure, with PBS, containing 2 mmoles/L EDTA and 20 mmoles/L butylated hydroxytoluene, for 4 min, followed with fixation by perfusion with 4 % (w/v) paraformaldehyde (pH 7.5) for 6 min [14]. The tissue was then dissected and stored in paraformaldehyde overnight, then rinsed in PBS and stored in 70% v/v ethanol. 2.4. Determination of blood glucose and glycated haemoglobin Blood glucose was quantified using a True TrackÒ smart system glucometer calibrated using a Roche Modular P800 auto-analyser (the Automated Laboratory, Biochemistry Dept., Royal Prince Alfred Hospital, Sydney). Glycated haemoglobin was determined using A1cNowþ kits (Metrika Inc., Sunnyvale, CA, USA). 2.5. Determination of blood cholesterol and triglycerides Total cholesterol and triglycerides were quantified using commercial kits (Total Cholesterol Kit, Roche Diagnostics, Basel, Switzerland; TG Assay, Novachem, from Merck KGaA, Darmstadt, Germany) in accordance with the manufacturer’s instructions. 2.6. Determination of plasma carnosine levels Plasma-derived low molecular mass filtrates ( 0.05) indicated no linear relationship between plasma carnosine and plaque area for mice of the same glycaemic status. Diabetes was associated with a significant decrease in plaque collagen content and an increase in area occupied by extracellular
Table 1 The measured group parameters at time of sacrifice are presented. For mean body mass, total plasma cholesterol and triglyceride plasma levels of the non-diabetic (control) and diabetic apo E / mice (with or without carnosine (CN) supplementation) groups at the time of sacrifice: results are expressed as mean standard error of the mean (SEM) for 12e19 mice per group. Plasma carnosine levels (mmoles/L), brachiocephalic artery and aortic sinus plaque areas (mm2) are presented as the 95% confidence interval together with the mean (n ¼ 12e18). * Significantly different (p < 0.05) from control group at the same level of carnosine supplementation; and # significantly different (p < 0.05) from the non-supplemented group matched for glycaemic status with data transformed so that it conformed to a normal distribution when necessary.
Mass (g) Total cholesterol (mmoles/L) Triglycerides (mmoles/L) Plasma carnosine (mmoles/L) Brachiocephalic artery plaque area (mm2) Aortic sinus plaque area (mm2)
Control
Control with CN
Diabetic
Diabetic with CN
27.8 0.6 9.6 0.5 1.1 0.1 1.2 0.3 0.045 0.009 0.138 0.013
29.9 0.5 9.0 0.3 0.8 0.1# 8.7 3.4# 0.040 0.009 0.134 0.012
24.6 0.9* 14.6 2.1* 1.7 0.2* 0.7 0.2 0.062 0.010* 0.152 0.016*
24.2 0.5* 15.2 0.7* 1.3 0.1*# 23.6 11.6# 0.059 0.006* 0.150 0.015*
406
B.E. Brown et al. / Atherosclerosis 232 (2014) 403e409
Fig. 1. Blood glucose levels (A) and glycated haemoglobin levels (B), at the time of sacrifice, for control (CTL) and diabetic (DIAB) apo E/ mice with ( þ CN) or without carnosine supplementation. Individual determinations (n ¼ 12e18) are presented, together with the median value (horizontal bar). Out of range results (>33 mmoles/L for blood glucose; and 13% for glycated haemoglobin) are represented as markers placed on the dashed line that delimits the range of detection. A number of animals with diabetes have superimposed data points. * Significantly different (p < 0.05) from control group at the same level of carnosine supplementation using rank-transformed data
lipid in the diabetic animals. Carnosine supplementation blocked these changes. That is, the collagen content and extracellular lipid content of the plaques from the diabetic carnosine-supplemented mice was found to be no different from those of the plaques from their non-diabetic littermates (Figs. 2 and 3). No significant differences were detected in plaque area occupied by a-actin-positive cells at either site (% of total plaque area {median, 1st and 3rd quartile}: CTL {3, 0, 0}; CTL þ CB {2, 1, 4}; DIAB {1, 0, 2}; DIAB þ CN {3, 2, 5}). For the brachiocephalic arteries there was a significant reduction in the area occupied by F4/80-positive cells with diabetic mice but this was unaffected by carnosine
supplementation, while at the aortic sinus there was a significant reduction for non-supplemented diabetic mice only (Fig. 4). 3.4. Multivariate analysis The analysis detected a clear differentiation between the four groups based on simultaneous covariation in the data variables, although there was some overlap between control and control þ carnosine animals. ANOSIM detected a significant difference amongst treatment groups, based on the simultaneous variance in the 16 data variables (Global R ¼ 0.265; p ¼ 0.001). All pairwise
Fig. 2. Panel A: representative images of atherosclerotic plaques stained with picrosirius red in the aortic sinus cross-sections of control (CTL), control þ carnosine (CTL þ CN), diabetes (DIAB) and diabetes þ carnosine (DIAB þ CN) mice. Panel B: Collagen content (as a % of plaque area), at the time of sacrifice, for control (CTL) and diabetic (DIAB) apo E/ mice with ( þ CN) or without carnosine supplementation. Data are presented as the 95% confidence interval together with the mean (n ¼ 12e18) for both the brachiocephalic and aortic sinus. * Significantly different (p < 0.05) from other groups using GLM ANOVA.
B.E. Brown et al. / Atherosclerosis 232 (2014) 403e409
407
Fig. 3. Panels A and B: extracellular lipid content of atherosclerotic plaques was estimated by measuring the clear areas on a section stained with F4/80 as shown on panel A. The section of interest (bordered by a dashed line) is magnified in panel B with arrows indicating example of lipid pools (false coloured red for image analysis) that were used for estimation of plaque lipid content. Panel C: lipid deposits (as a % of plaque area), at the time of sacrifice, for control (CTL) and diabetic (DIAB) apo E/ mice with ( þ CN) or without carnosine supplementation. Data are presented as box plots bordered by the 1st and 3rd quartile value with the median indicated by a horizontal bar: the range is shown by the vertical lines. # Significantly different (p < 0.05) from other treatments using a KruskaleWallis test.
comparisons amongst the four groups were significant (p ¼ 0.001 in all cases) except between the control and control þ carnosine mice (p ¼ 0.449). SIMPER analysis indicated the data variables which best differentiate between groups and thus, by inference, the strongest effects of diabetes induction and carnosine supplementation. Only data variables that contributed greater than 10% of the total dissimilarity were considered. The data variable that best distinguished between un-supplemented non-diabetic and unsupplemented diabetic mice was blood glucose, followed by glycated haemoglobin and aortic sinus lipid core plaque area (Supplementary Table 1). The variables that contributed most to the differences between carnosine-supplemented non-diabetic and the matched supplemented diabetic mice were blood glucose and glycated haemoglobin (Supplementary Table 1). When comparing the two diabetic groups the variable that best distinguished these un-supplemented vs. the supplemented mice was the greatly reduced aortic sinus lipid core plaque area in the carnosinesupplemented group (Supplementary Table 1). No other data
variable contributed more than 10% to the total dissimilarity between these groups, although there were also large differences between the brachial artery lipid core, as a % of plaque area, between these two groups (8.1%). 4. Discussion Carnosine has been investigated extensively as a potential therapy for diseases in which oxidation and/or glycation is believed to contribute, including atherosclerosis [6,8]. Whilst carnosine administration to humans yields only transient increases in plasma levels, the current (Table 1) and previous studies [9,12] have demonstrated sustained increases in rodents after supplementation via drinking water. The carnosine levels achieved here are higher than those reported previously [12] probably due to the higher dose employed. Studies using Zucker rats have demonstrated significant increases in plasma and kidney carnosine levels with chronic supplementation with D- but not L-carnosine [11],
Fig. 4. Percentage positive staining for F/480 (macrophages) in the brachiocephalic artery and aortic sinus plaques, at the time of sacrifice, for control and diabetic apo E/ mice with or without carnosine supplementation. Data are presented as box plots bordered by the 1st and 3rd quartile value with the median indicated by a horizontal bar: the range is shown by the vertical lines. (n ¼ 11e19). * Significantly different (p < 0.05) from control with the same level of carnosine supplementation using GLM for the square roottransformed data. # Significantly different (p < 0.05) from control with the same level of carnosine supplementation using a KruskaleWallis test.
408
B.E. Brown et al. / Atherosclerosis 232 (2014) 403e409
suggesting that the potential to elevate plasma L-carnosine is species-dependent. The wider range of plasma carnosine levels detected in the diabetic animals when compared to controls is likely to be due to the polydipsia of diabetes, with a three-fold higher level of water consumption observed in the diabetic animals (data not shown). Carnosine levels in other tissues were not analysed as the paraformaldehyde used in tissue fixation, reacts with carnosine. However a previous study has reported increased levels in the hearts and livers of mice supplemented with carnosine using a similar protocol [12]. In the current study, prolonged carnosine administration did not modulate blood glucose (Fig. 1A) or glycated haemoglobin (Fig. 1B) levels, although a previous study [12] has demonstrated effects on blood glucose levels after 4 weeks of supplementation. Transient hyperglycaemia induced by 2-deoxy-D-glucose in rats has also been reported to be inhibited by intravenous and intraperitoneal carnosine administration to give submicromolar plasma concentrations [20]. We conclude that chronic high dose supplementation of carnosine may override the more transient and low dose effects of carnosine upon blood glucose reported previously [12,20]. Whilst carnosine supplementation had no effect upon the hypercholesterolaemia in the diabetic mice, supplementation significantly reduced plasma triglycerides (Table 1); a similar reduction was seen in the control, carnosine-supplemented mice. These data suggest that the effect of carnosine on triglyceride levels is independent of changes induced by diabetes. A differential effect of carnosine on triglycerides and cholesterol has been reported previously in both diabetic and non-diabetic mice [12] with both concentrations of carnosine examined (0.5 or 1.0 g/L in drinking water) blunting the increase in heart and liver triglycerides whereas effects on cholesterol were only seen at the higher dose [12]. Both carnosine and histidine (1 g/L) block increases in plasma triglycerides and cholesterol in fat-fed, non-diabetic, C57BL/6 mice [9] whereas supplementation of fat-fed apo E/ mice with D-carnosine octyl ester for 6 weeks did not reduce plasma cholesterol levels [6]. In contrast, supplementation of hyperlipidaemic (but not hyperglycaemic) Zucker rats with L- or D-carnosine resulted in matched and significant decreases in both plasma cholesterol and triglyceride [11]. Thus while carnosine appears to have a consistent cross-species hypotriglyceridaemic action, this is not true for cholesterol levels. As both diabetes and elevated triglycerides and cholesterol are associated with increased atherosclerosis, we assessed lesion parameters in both the brachiocephalic artery and aortic sinus as these are major sites of development of advanced, rupture-prone, lesions [2]. Diabetes resulted in a significant increase in plaque area at both sites (Table 1) and despite the triglyceride-lowering effects of carnosine; supplementation did not reduce plaque area when compared to the non-supplemented mice matched for glycaemic status. A similar observation has been reported for diabetic apo E/ mice supplemented with conjugated linoleic acid, where no decrease in plaque area was detected despite a substantial lowering of plasma triglycerides [19]. Using this same model, others have reported decreased aortic sinus plaque area on treatment with the PPARg-agonist rosiglitazone, while drug treatment resulted in an elevation of blood triglyceride and cholesterol levels [21]. Thus modulation of hypertriglyceridaemia in this model does not appear to consistently impact on lesion size. These results contrast with recent studies in which D-carnosine octyl ester supplementation of fat-fed apo E/ mice resulted in a reduction in plaque area [6,8]. However, in common with Menini et al [8], carnosine appears to modify plaque composition, and specifically reduce extracellular lipid levels and increase collagen content (Figs. 2 and 3). Extracellular lipid levels within lesions reflect cholesterol precipitation from lipidladen macrophages [22], a process resulting in apoptosis and
necrosis [23]. Thus the reduction in lipid content and higher collagen content (a marker of extracellular matrix levels and hence lesion stability) induced by carnosine appears to reflect a more stable and less advanced plaque phenotype. Consistent with this, the lesion content of macrophages was higher in the carnosine treated group, suggesting that the macrophages within these plaques had not proceeded to apoptosis/necrosis [24]. These findings suggest that carnosine alters plaque composition to a more stable phenotype. The mechanism by which carnosine mediates these changes has not been determined definitively. Carnosine can modulate oxidation and glycation reactions by acting as a scavenger of both radicals and reactive aldehydes (e.g. methylglyoxal and the lipid-derived aldehydes, malondialdehyde and 4-hydroxynonenal) [4]. Thus we and others have shown that carnosine can block (pro-atherosclerotic) aldehyde-mediated modification of low-density lipoproteins as well as copper ion-mediated oxidation [6,7]. As only modest (low mmole/ L) levels plasma levels of carnosine were achieved by supplementation, it is unlikely that radical scavenging is the mode of action, as carnosine is unlikely to compete with other biological targets (e.g. plasma proteins, cf. data in Refs. [25,26]). Detoxification of reactive aldehydes, such as methylgloxal, malondialdehyde and 4-hydroxynonenal, is a more likely mode of action for carnosine, as these aldehydes have longer biological halflives [27] and hence are more likely to encounter carnosine and can be removed by adduction. Reactive aldehydes form adducts with proteins [28], converts LDL into a high-uptake form [29] and have pro-inflammatory activity (e.g. via modulation of the Nrf2 pathway, PPAR-g, haem oxygenase-1 and CD36 expression) [30]. Thus the removal of reactive carbonyls by reaction with carnosine may account for the anti-inflammatory effects of carnosine resulting in a dampening of the low-grade chronic inflammation of the artery wall induced by diabetes and/or a high fat diet [23] and give rise to a more stable lesion phenotype. Carnosine and its analogues have been shown to have antihyperlipidaemic, anti-atherosclerotic, hepato- and renal-protective actions in models of obesity-associated complications [6,8,11]. The current study demonstrates that this well-tolerated dipeptide also modulates diabetes-associated atherosclerosis. The impact of this compound appears to occur via decreases in blood triglyceride levels which are associated with changes in plaque characteristics and greater plaque stability. This data supports further investigation of the use of carnosine (or stable analogues) as an adjunct to current treatments for the vascular complications of diabetes. Conflict of interest The authors have declared no conflict of interest. Funding for this project was provided by an unrestricted grant from Eli Lilly and Company. Acknowledgements The authors thank Herbert Ayala, Kim Berry, Edwin Martin, Gary Martinic and Pam Sheahan for their assistance with animal management and sample collection; Mahmoud Al-hinti and Janice Pamment (Biochemistry Department, Royal Prince Alfred Hospital) for the blood glucose analyses, and Liming Hou and Professor Kerry-Anne Rye for the cholesterol and triglyceride analysis. Christine HJ Kim was a recipient of a University Postgraduate Award from the University of Sydney. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.atherosclerosis.2013.11.068.068.
B.E. Brown et al. / Atherosclerosis 232 (2014) 403e409
References [1] Bianchi C, Miccoli R, Daniele G, Penno G, Del Prato S. Is there evidence that oral hypoglycemic agents reduce cardiovascular morbidity/mortality? Yes. Diabetes Care 2009;32(Suppl. 2):S342e8. [2] Hsueh WA, Abel ED, Breslow JL, et al. Recipes for creating animal models of diabetic cardiovascular disease. Circ Res 2007;100:1415e27. [3] Jandeleit-Dahm K, Cooper ME. The role of AGEs in cardiovascular disease. Curr Pharm Des 2008;14:979e86. [4] Boldyrev AA, Aldini G, Derave W. Physiology and pathophysiology of carnosine. Physiol Rev 2013;93:1803e45. [5] Hipkiss AR. Carnosine and its possible roles in nutrition and health. Adv Food Nutr Res 2009;57:87e154. [6] Barski OA, Xie Z, Baba SP, et al. Dietary carnosine prevents early atherosclerotic lesion formation in apolipoprotein E-null mice. Arterioscler Thromb Vasc Biol 2013;33:1162e70. [7] Rashid I, van Reyk DM, Davies MJ. Carnosine and its constituents inhibit glycation of low-density lipoproteins that promote foam cell formation in vitro. FEBS Lett 2007;581:1067e70. [8] Menini S, Iacobini C, Ricci C, et al. D-carnosine octylester attenuates atherosclerosis and renal disease in ApoE null mice fed a Western diet through reduction of carbonyl stress and inflammation. Br J Pharmacol 2012;166:1344e56. [9] Mong M-C, Chao C-Y, Yin M-C. Histidine and carnosine alleviated hepatic steatosis in mice consumed high saturated fat diet. Eur J Pharmacol 2011;653:82e8. [10] Stvolinsky SL, Bulygina ER, Fedorova TN, et al. Biological activity of novel synthetic derivatives of carnosine. Cell Mol Neurobiol 2010;30:395e404. [11] Aldini G, Orioli M, Rossoni G, et al. The carbonyl scavenger carnosine ameliorates dyslipidemia and renal function in zucker obese rats. J Cell Mol Med 2011;15:1339e56. [12] Lee Y-T, Hsi C-C, Lin M-H, Liu K-S, Yin M-C. Histidine and carnosine delay diabetic deterioration in mice and protect human low density lipoprotein against oxidation and glycation. Eur J Pharmacol 2005;513:145e50. [13] Candido R, Jandeleit-Dahm KA, Cao Z, et al. Prevention of accelerated atherosclerosis by angiotensin-converting enzyme inhibition in diabetic apolipoprotein E-deficient mice. Circulation 2002;106:246e53. [14] McRobb L, Handelsman DJ, Heather AK. Androgen-induced progression of arterial calcification in apolipoprotein E-null mice is uncoupled from plaque growth and lipid levels. Endocrinology 2009;150:841e8. [15] Pattison DI, Hawkins CL, Davies MJ. Hypochlorous acid-mediated protein oxidation: how important are chloramine transfer reactions and protein tertiary structure? Biochemistry 2007;46:9853e64.
409
[16] Ali ZA, Bursill CA, Hu Y, et al. Gene transfer of a broad spectrum CC-chemokine inhibitor reduces vein graft atherosclerosis in apolipoprotein E-knockout mice. Circulation 2005;112:I-235e41. [17] Park L, Raman KG, Lee KJ, et al. Suppression of accelerated diabetic atherosclerosis by the soluble receptor for advanced glycation endproducts. Nat Med 1998;4:1025e31. [18] Candido R, Allen TJ, Lassila M, et al. Irbesartan but not amlidipine suppresses diabetes-associated atherosclerosis. Circulation 2004;109:1536e42. [19] Nestel P, Fujii A, Allen T. The cis-9,trans-11 isomer of conjugated linoleic acid (CLA) lowers plasma triglyceride and raises HDL cholesterol concentrations but does not suppress aortic atherosclerosis in diabetic apoE-deficient mice. Atherosclerosis 2006;189:282e7. [20] Yamano T, Niijima A, Iimori S, Tsuruoka N, Kiso Y, Nagai K. Effect of L-carnosine on the hyperglycemia caused by intracranial injection of 2-deoxy-Dglucose in rats. Neurosci Lett 2001;313:78e82. [21] Levi Z, Shaish A, Yacov N, et al. Rosiglitazone (PPAR-g-agonist) attenuates atherogenesis with no effect on hyperglycaemia in a combined diabetesatherosclerosis mouse model. Diabetes Obes Metab 2003;5:45e50. [22] Small DM, Shipley GG. Physicalechemical basis of lipid deposition in atherosclerosis. Science 1974;185:222e9. [23] Libby P. Inflammation in atherosclerosis. Nature 2002;420:868. [24] Reimers GJ, Jackson CL, Rickards J, et al. Inhibition of rupture of established atherosclerotic plaques by treatment with apolipoprotein A-I. Cardiovasc Res 2011;91:37e44. [25] Davies MJ. The oxidative environment and protein damage. Biochim Biophys Acta 2005;1703:93e109. [26] Gebicki S, Bartosz B, Gebicki JM. Peroxidation of proteins by free radicals and its consequences to living organisms. Curr Topics Biophys 1995;1995: 220e6. [27] Lo TWC, Selwood T, Thornalley PJ. The reaction of methylglyoxal with aminoguanidine under physiological conditions and prevention of methylglyoxal binding to plasma proteins. Biochem Pharmacol 1994;48:1865e70. [28] Requena JR, Fu M-X, Ahmed MU, et al. Quantification of malondialdehyde and 4-hydroxynonenal adducts to lysine residues in native and oxidised human low-density lipoprotein. Biochem J 1997;322:317e25. [29] Hoff HF, O’Neil J, Chisolm GD, et al. Modification of low density lipoprotein with 4-hydroxynonenal induces uptake by macrophages. Arteriosclerosis 1989;9:538e49. [30] Ishii T, Itoh K, Ruiz E, et al. Role of Nrf2 in the regulation of CD36 and stress protein expression in murine macrophages: activation by oxidatively modified LDL and 4-hydroxynonenal. Circ Res 2004;94:609e16.
