ORIGINAL RESEARCH

A time course study on prothrombotic parametersand their modulation by anti-platelet drugsin hyperlipidemic hamsters

Vishal Singh & Manish Jain & Prem Prakash & Ankita Misra & Vivek Khanna &

Rajiv Lochan Tiwari & Ravi Shankar Keshari & Shivendra Singh &

Madhu Dikshit & Manoj Kumar Barthwal

Received: 2 July 2010 /Accepted: 9 December 2010 /Published online: 1 February 2011# University of Navarra 2011

Abstract The present study was undertaken to assessthe chronology of major pathological events associatedwith high cholesterol (HC) diet and their modulation byanti-platelet drugs. Male Golden Syrian hamsters werefed HC diet up to 90 days. Plasma lipid, glucose andcoagulation parameters (commercial kits), platelet acti-vation (whole blood aggregation and static adhesion),endothelial dysfunction (aortic ring vasoreactivity),splenocyte TNF-α, IFN-γ and iNOS mRNA transcripts(RT–PCR), and ferric chloride (time to occlusion)induced thrombosis were monitored at 15, 30, 60, and90 days after HC feeding and compared with normolipi-demic hamsters. A significant increase in plasma lipidlevels was observed at 15 days of HC feeding, but otherparameters remain unaltered. Enhanced ADP, collagen,and thrombin-induced platelet aggregation, splenocyteTNF-α expression along with endothelial dysfunctionwere observed from 30 to 90 days of HC feeding.Platelet adhesion on collagen-/fibrinogen-coated surface

and IFN-γ expression were augmented only after60 days, while enhanced iNOS expression, reductionin thrombin time, and potentiation of ferric chloride-induced thrombosis was observed only at 90 days of HCfeeding. Thus, pathological changes induced by HC dietdepend on the duration and extent of hyperlipidemia.Moreover, hamsters treated with anti-platelet drugsaspirin (5 mg/kg) or clopidogrel (10 mg/kg) along withHC feeding exhibited reduction in platelet activation aswell as subsequent changes observed in the above-mentioned parameters following HC feeding. Sincereduction in TNF-α was associated with reversion inendothelial dysfunction and prothrombotic state, the roleof platelets is implicated in the pathological changesassociated with HC feeding.

Keywords Hyperlipidemia . Inflammation . Platelet .

Aspirin . Clopidogrel . Hamsters

Introduction

Thrombotic complications in atherosclerosis are oftenassociated with plaque rupture [12], but importance ofhyperlipidemia in predisposing to thrombotic complica-tions is less clearly defined. Human and experimentalfindings strongly associate hyperlipidemia with endo-thelial dysfunction, platelet activation, and thromboticcomplications [11]. Although high cholesterol (HC) dietcan induce these changes in various animals [34], theirinter-relationship is however not well defined. The

J Physiol Biochem (2011) 67:205–216DOI 10.1007/s13105-010-0065-z

Vishal Singh, Manish Jain and Prem Prakash have contributedequally to this work

CDRI communication number: 7810

V. Singh :M. Jain : P. Prakash :A. Misra :V. Khanna :R. L. Tiwari :R. S. Keshari : S. Singh :M. Dikshit (*) :M. K. Barthwal (*)Division of Pharmacology, Central Drug Research Institute,1. M.G. Marg,Lucknow, UP 226001, Indiae-mail: madhu_dikshit@cdri.res.ine-mail: manojbarthwal@cdri.res.in

mechanism of HC-induced thrombosis can be elucidatedby deciphering the chronology of major pathologicalevents that contribute to this phenomenon. Fromprevious findings, it is difficult to conclude thechronology of thrombosis-associated pathologicalchanges since feeding regimes as well as the amountof cholesterol in HC diet have been variable [3, 8]. Inaddition, associated parameters were also not evaluatedin any single study conducted so far [3]. Spontaneouslyhypertensive rats (SHR) and Wistar–Kyoto rats fed onHC diet up to 12 weeks did not exhibit impaired NO-dependent vasodilation [5]; however, Wistar rats fed onHC diet for 10 weeks exhibited endothelial dysfunction[38]. In yet another study on SHR rats fed on HC dietfor 1 to 6 months, endothelium dysfunction wasobserved after 3 months [8]. However, in these studies,platelet activation and inflammatory status that havebeen associated with the prothrombotic state were notevaluated. Double-homozygous apoE and LDL receptordeficient mice fed on high fat diet showed endothelialactivation at 12 weeks [3], although enhanced plateletreactivity and dynamic coagulation was observed onlyafter 18 weeks [3]. Inflammatory status was notevaluated in this study. Similarly, hypercholesterolemicdiet up to 8 weeks in NZW rabbit induced plateletactivation [27], but endothelial dysfunction and inflam-matory status were not evaluated.

Atherosclerotic patients who undergo balloon angio-plasty or develop in-stent restenosis are often kept ondual anti-platelet therapy to prevent the re-blockage ofthe artery [28]. The role of platelets in the initiation ofatherosclerosis and progression is under intensiveinvestigation [33]. Endothelial dysfunction has beenconsidered as an early event in the progression ofatherosclerosis [9], although others reported thatplatelet activation and other factors are more important[31]. Tumor necrosis factor-alpha (TNF-α), interferon-gamma (IFN-γ), and inducible nitric oxide synthase(iNOS) are important inflammatory mediators that areindividually known to be induced during atherosclero-sis [40]. Although TNF-α and IFN-γ are known toinduce iNOS expression [41], regulation of thispathway in spleens during hyperlipidemia is notknown. TNF-α is also known to induce endothelialdysfunction in pre-diabetic metabolic syndrome [26].Although TNF-α was found to be elevated in the aorticarch of hyperlipidemic hamsters [1], its role in theprogression of athero-thrombosis remains to be ex-plored. Previous studies suggest improvement in

inflammation [43] and endothelial dysfunction [10,14] by anti-platelet drugs aspirin and clopidogrel.However, the mechanism by which these agents exertthis effect is not well defined. Therefore, the presentstudy has been undertaken to systematically evaluatesuch events following HC treatment in hamsters.Hamster is in use for inducing hyperlipidemia andatherosclerosis for quite some time [21]. Hamstershave many advantages as a model for studyingdyslipidemia, and their resemblance to humans on thisaspect [34] makes them a suitable model for screeningtherapeutic candidate drugs. Many candidate drugshave been screened in this model for their hypolipi-demic or anti-atherosclerotic effect, but mostly it hasbeen confined to lesion and lipid analysis [16].Endothelial functionality and thrombotic complicationshave not yet received much attention in this model. Byevaluating the chronology of major pathological eventsin this model, early and late events can be identifiedwhich can explain the intertwining of these processesand might be helpful in evaluating the effect of testdrugs against many associated parameters.

Therefore, to establish the sequence of majorpathological events and suitability of the model forevaluating such changes, we performed a time-dependent study to monitor platelet activation, endothe-lial dysfunction, inflammation, and thrombosis, and alsoevaluated the effect of anti-platelet drugs in this model.

Materials and methods

Materials

Chrono-Par collagen from Chrono-Log Corp. (PA, USA)and coagulation kits from Diagnostica Stago (Vilvoorde,Belgium) were used in the present study. RevertAid™ HMinus first strand cDNA synthesis kit was obtained fromFermentas, USA. All other fine chemicals includingacetylcholine (ACh) chloride, phenylephrine (PE) hydro-chloride, collagen type I (calf-skin type III), fibrinogen(fraction 1, type 1-S), adenosine diphosphate (ADP), andthrombin were obtained from Sigma-Aldrich USA.

Animal diet and treatment

Golden Syrian hamsters (100–120 g) obtained fromthe CDRI National Laboratory Animal Centre wereused in this study as per Institutional Animal Ethics

206 V. Singh et al.

Committee guidelines. HC diet was prepared at thediet production facility of CDRI animal house. Basiccomponents of the diet are listed in Table 1. Animalswere randomized into eight groups receiving chow orHC diet [chow diet supplemented with 3% cholesteroland 15% saturated fat (coconut oil)] for 15, 30, 60,and 90 days. For experiments involving evaluation ofanti-thrombotic drugs, hamsters were divided intofour groups that received chow or HC diet along withvehicle, aspirin 5 mg/kg [36], or clopidogrel 10 mg/kg[42] orally once in a day for 90 days. Each group had aminimum of 16 animals. The entire study includingfeeding and analysis was repeated three times. Bloodwas collected from fasting animals by cardiac puncturein tubes containing 2.5% trisodium citrate. Aggregation

studies were carried out using 1.5 ml of whole blood.For adhesion studies, platelet-rich plasma (PRP) wasisolated from 3 ml of whole blood by centrifugation at1,000 rpm for 10 min. Platelet pellet was obtained bycentrifugation of PRP at 4,000 rpm in the presence of10% acid citrate dextrose (ACD). Blood remainingafter PRP isolation was spun at 4,000 rpm for 10 minto get platelet-poor plasma (PPP) that was used forcoagulation studies.

Plasma biochemistry

Plasma total cholesterol (TC), low density lipoproteincholesterol (LDLC), high density lipoprotein choles-terol (HDLC), triglycerides (TGs), and glucose were

Sample no. Composition Quantity in grams/100 kg feed

1 Choline chloride 100

2 PABA 10

3 Alpha tocopherol 12

4 Inositol 10

5 Pantothenic acid 1.2

6 Thiamine 1.2

7 Niacin 1

8 Riboflavin 0.5

9 Pyridoxine 0.6

10 Folic acid 0.1

11 Biotin 0.04

12 Vitamin K 0.15

13 Vitamin D 0.21

14 Cynocobalamin 0.005

15 Dicalcium phosphate 1,250

16 Calcium carbonate 550

17 Sodium chloride 300

18 Magnesium sulfate 229.2

19 Ferrous sulfate 16.4

20 Potassium iodide 1,000

21 Zinc sulfate 2,192

22 Copper sulfate 2,192

23 Cobalt chloride 12

24 Starch 363.21

25 Wheat 22,500

26 Roaster Bengal gram dal 60,000

27 Skim milk powder 5,000

28 Casein 4,000

29 Vegetable oil 4,000

Table 1 Composition ofchow diet (quantity ingrams/100 kg)

High cholesterol diet wasprepared by supplementingthe chow diet with 3% cho-lesterol and 15% saturatedfat

Development of prothrombotic state in hyperlipidemic hamsters 207

measured enzymatically using commercial kits fromPointe Scientific, USA and RFCL Limited, India. Inbrief, fresh plasma was isolated from citrated bloodand treated with kit reagents as per manufacturer’sprotocol and color produced was measured by ELISAplate reader at parameter-specific wavelengths(Powerware XS; Biotek, USA).

Vascular reactivity

Transverse 4-mm-wide rings of thoracic aorta werecut and mounted in organ baths containing Krebssolution.

Changes in isometric force was amplified viaCompact Research System Amplifiers and wererecorded with force transducers (FSG-01; Budapest,Hungary) using SPEL Solution Pack for ExperimentalLaboratories ADVANCE ISOSYS data acquisition andanalysis software. After equilibration, the aortic ringswere exposed to KCl Krebs buffer (80 mM) in order toassess the maximum tissue contractility. Cumulativeconcentration-dependent contraction responses to PE(10 pM–100 μM) were assessed. The presence of afunctional endothelium was then assessed by measuringrelaxation to ACh (3 nM to 30 mM) in PE (1 μM) pre-contracted rings. Finally, tissue contractility and viabilitywere assessed by exposing the rings to KCl Krebs buffer(80 mM) in all the groups [30].

Assessment of inflammatory cytokines

Spleens were aseptically removed from the hamstersand placed in RPMI medium. Single cell suspensionsof splenocytes (spleen lymphocytes) were prepared asdescribed previously [22]. RNA was isolated usingTRIZOL as per manufacturer’s protocol. cDNA wassynthesized using RevertAid™ H Minus first strandcDNA synthesis kit (Fermentas). Gene amplificationof TNF-α, IFN-γ, iNOS, and hypoxanthine–guaninephosphoribosyltransferase was achieved using specificprimers and amplification conditions as describedpreviously [6]. The PCR products were analyzed onagarose gel and quantified using a calibrated densi-tometer (BioRad Inc. USA).

Whole blood aggregation

Whole blood aggregation studies were performed on adual channel aggregometer (Chrono-Log) using the

impedance method [17]. Aggregation was induced byADP (10 μM), thrombin (1 U/ml), and collagen(2.5 μg/ml). Impedance was measured over a timeinterval of 6 min and aggregation was calculated bythe method as described earlier [17].

Static platelet adhesion assay

Platelet adhesion to collagen and fibrinogen wasmeasured in polystyrene 96-well microtiter plates asdescribed earlier [37]. Briefly, the plates were coatedwith fibrillar collagen (2 μg/well) or fibrinogen(200 μg/well). Washed platelets (2×107 cells/well)were added to the coated plates, and after threewashings 5 mM p-nitrophenyl phosphate was addedto measure adhesion [25]. The relative OD units wereused to calculate the number of adhered platelets [25].

Coagulation parameters

Coagulation parameters, namely prothrombin time(PT), activated partial thromboplastin time (aPTT),thrombin time (TT), and fibrinogen time (FT), wereassayed in plasma samples of all the groups within2 h of sample collection. All assays were performedusing commercial kits as per manufacturer’s instruc-tions and measured by using a coagulometer (Start4semi-automated; Young Instruments, Stago, France).

Ferric chloride-induced thrombosis

Hamsters were anesthetized by urethane (1.25 g/kg, i.p.).The carotid artery was carefully dissected and a pulsedDoppler flow probe (DBF-120A-CPx) was placedaround it to record the blood flow using Biopac DataAcquisition System (CBI-8000; Crystal Biotech, USA).Thrombosis was monitored as the time taken for thecessation of carotid artery blood flow, time to occlusion(TTO) after the application of 20% ferric chloride-soaked filter paper as described earlier [35].

Statistical analysis

Data have been represented as mean ± SEM.Significance of difference between groups was ana-lyzed by performing one-way analysis of variancefollowed by Bonferroni’s multiple comparison test forvarious parameters between control and treatedgroups; a p value less than 0.05 was considered

208 V. Singh et al.

statistically significant. GraphPad Prism Software5.00 was employed for statistical analyses.

Results

Body weight, plasma biochemistry, and endothelialdysfunction

A time-dependent increase in the body weight ofhamsters fed HC diet was observed. Weights of hamsterson HC diet at 30 (159±4 g), 60 (174±5.4 g), and 90(190±6.7 g) days, and HC+aspirin (194±4.61 g) andHC+clopidogrel (186±5.05 g) were significantly morethan the corresponding control animals at 30 (144±2.1 g), 60 (154.5±3.1 g), and 90 (165.3±4.9 g)days. At15 days, weights of chow- and HC-fed animals weresimilar. Lipid levels in hamsters analyzed at differentdays of high cholesterol feeding were significantly morethan their matched chow-fed controls (Table 2). Inaddition, chow-fed animals analyzed at different dayshad similar lipid levels (Table 2). Plasma TC, TGs,LDLC, and HDLC were increased significantly after15 days of HC treatment, indicating an early develop-ment of hyperlipidemia in these animals (Table 2). Thiswas further increased at 30 days (Table 2). However, nofurther increase was observed after 30 days, and lipidlevels at 60 and 90 days were almost similar to 30 days(Table 2). HC diet did not influence glucose levels inthese hamsters up to 90 days. In the present study, HC

diet and a lower dose of aspirin 5 mg/kg resulted in asignificant decrease in TC (p<0.05) (Table 2) and TGs(p<0.05) (Table 2) levels, but no change in animalweight was observed (data not shown). However,clopidogrel did not alter the circulating lipid levels(Table 2). To assess endothelial dysfunction in theseanimals, dose-dependent PE-induced contractions andACh-mediated relaxations were monitored. In chow-and HC-fed animals, PE-induced (10 pM to 100 μM)contractions were similar and there was no effect ofduration of HC feeding on this parameter (data notshown). Similarly, ACh (3 nM to 30 nM) inducedrelaxation (data not shown) in PE pre-contracted ringsfrom chow-fed animals of different time points were notsignificantly different from each other, and thereforetheir data were pooled. ACh-induced relaxations weresignificantly reduced in the aortic rings obtained fromhamsters after 30, 60, and 90 days on HC diet whencompared to the control values (Fig. 1a). Significantrecovery in endothelial function was observed inaspirin- and clopidogrel-treated, HC-fed hamsters(Fig. 1b).

Expression of inflammatory mediators

TNF-α, iNOS, and IFN-γ expression was measuredin the splenocytes. Around 2-fold induction in TNF-αwas observed at 30 days of HC feeding which remainaugmented up to 90 days (Fig. 2a, d). Similarly, a 2-fold induction in IFN-γ was observed at 60 days that

Table 2 Time-dependent hyperlipidemia

Days Animal groups TC (mg/dl) TG (mg/dl) LDLC (mg/dl) HDLC (mg/dl)

15 Chow 98.2±8.3 132.1±13.8 28.7±2.1 26.7±2.1

HC 283.6±16.1** 318±37.8* 146±20.1** 46.48±1.7*

30 Chow 94.2±6.7 140±11.2 31.6±1.3 29.3±1.7

HC 749.3±44*** ααα 753.2±107.8*** ααα 441±52.1*** ααα 64.8±2.6*** αα

60 Chow 98.3±4.9 141±17.8 34.7±2 31.1±2.3

HC 692±85.2*** 708.1±147.5*** 404±36.7*** 68.5±2.6***

90 Chow 97.2±13.1 152.1±8.8 38.13±2.37 33.34±2.2

HC 738.1±113*** 687±91.4*** 360±24.1*** 67.74±3.3***

HC+aspirin 385±80*** # 359.3±41.9*** # 347±36.1*** 68.37±2.3***

HC+clopidogrel 703±61*** 692±85.2*** 355±28.5*** 64.72±1.9***

Results are mean ± SEM from at least ten animals in each group

TC total cholesterol, TGs triglycerides, LDLC low density lipoprotein cholesterol, HDLC high density lipoprotein cholesterol

*p<0.05, **p<0.01, ***p<0.001 vs. chow fed, # p<0.01 vs. high cholesterol (HC) 90 days, ααα p<0.0001, αα p<0.01 vs. HC 15 days

Development of prothrombotic state in hyperlipidemic hamsters 209

was further enhanced to 4.5-fold at 90 days (Fig. 2b,d). A 3-fold induction in iNOS transcript wasobserved only at 90 days (Fig. 2c, d). Although atendency of increase in iNOS transcript was evident at60 days, this was not statistically significant (Fig. 2c,d). Both aspirin and clopidogrel inhibited HC-inducedTNF-α expression (Fig. 2a, d). Although aspirin alsoinhibited HC-induced increase in iNOS and IFN-γlevels, clopidogrel had no effect (Fig. 2b–d).

Platelet activation in hyperlipidemic hamsters

To determine the time-dependent influence of lipidson platelet reactivity, platelet adhesion and aggrega-tion studies were performed on hamsters with andwithout HC dietary manipulation at various timepoints. Platelet counts were similar in all the groupswith a mean count of 1×108 platelets per milliliter.No change in platelet aggregation or adhesion couldbe observed at 15 days of HC feeding when the lipidlevels were significantly high. Significant increase inplatelet aggregation could be observed at 30 days ofHC feeding (Fig. 3a). At subsequent times of analysis,a tendency of increase in platelet aggregation wasobserved but this was not statistically different fromthe changes observed at 30 days (Fig. 3a). Meanmaximal platelet aggregation in response to ADP

(10 μM) was significantly more in hamsters after 30(40±3%), 60 (43±3%), and 90 (60±7%) days of HCdiet feeding than their respective controls at 30 (30±1.8%), 60 (28±3.4%), and 90 days (29±1.3%),respectively (Fig. 3a). Similarly, whole blood aggre-gation induced by collagen (2.5 μg/ml) was alsohigher in 30 (51.0±4%), 60 (53±3%), and 90 (62±4%) days of HC-fed animals as compared to theirchow-fed controls at 30 (37±2%), 60 (31±1%), and90 (32±1.1%) days, respectively (Fig. 3a). Thrombin(1 U/ml) induced aggregation was also higher in 30(46±2%), 60 (45±3%), and 90 (51±3%) days HC-fedgroup as compared to their chow-fed controls at 30(33%±2), 60 (30%±2), and 90 (33±1.1%) days,respectively (Fig. 3a). Hamsters on HC diet random-ized to receive aspirin and clopidogrel displayedreduced aggregation as compared to HC-fed groups(Fig. 3a). Significant decrease in ADP (38±5%) andcollagen (37±4%) induced platelet aggregation inaspirin-treated HC group was observed as comparedto HC-fed group alone (Fig. 3a). Similarly, significantdecrease in ADP (p<0.001) and collagen (p<0.05)induced platelet aggregation in clopidogrel-treatedHC group was also observed as compared to HC-fedgroup alone (Fig. 3a). Platelets isolated from HC-fedhamsters showed enhanced adhesion to collagen-(Fig. 3b) and fibrinogen-coated surfaces (Fig. 3c).

Fig. 1 Effect of HC diet on the vascular reactivity at variousduration of feeding. a, b Concentration-dependent response ofACh (3 nM to 30 mM) in the PE (1 μM) pre-contracted aorticrings of hamsters in different groups. Results are mean ± SEMfrom at least ten animals from each group. *p<0.001 HC

90 days vs. pooled control group, +p<0.05, +++p<0.001 HC60 days vs. pooled control group, δδp<0.01, δδδp<0.001 HC30 days vs. pooled control group, #p<0.001 HC+aspirin vs.HC 90 days, σp<0.05 HC+clopidogrel vs. HC 90 days

210 V. Singh et al.

Significantly enhanced platelet adhesion was ob-served on collagen-coated plates in 60 (p<0.01) and90 days (p<0.001) HC-fed groups in comparison tochow-fed control (Fig. 3b). However, in case offibrinogen, a significant increase was observed onlyat 90 days (p<0.001) (Fig. 3c). Aspirin had no effecton adhesion as compared to HC-treated group(Fig. 3b, c). However, clopidogrel significantlyinhibited the HC-induced enhanced platelet adhesionon fibrinogen (p<0.001) and collagen (p<0.001)coated plates as compared to HC-treated groups(Fig. 3b, c).

Coagulation parameters and ferric chloride-inducedthrombosis

To assess the coagulability of blood, TT, PT, aPTT,and FT were assessed in the plasma of all the groups(Table 3). A significant decrease in the TT (p<0.05)was observed only at 90 days in HC-fed animals (25±1 min) when compared to their chow-fed controls(29±1 min) (Table 3). No differences were observedin aPTT, FT, and PT at all the time points of HC diettreatment in comparison to respective chow-fed

controls (Table 3). Aspirin treatment showed signif-icant reversion in the TT (29±0.7) with no change inother parameters. Similarly, clopidogrel also showedsignificant reversion in TT (31±2) with no effect onother parameters (Table 3). Prothrombotic potential inHC-fed animals was assessed by measuring TTO afterthe application of ferric chloride-soaked filter paper(Fig. 4). A significant reduction in the ferric chloride-induced TTO was observed in 90 days HC-fedanimals in comparison to their respective control(p<0.05) (Fig. 4). However, there was no change atall the other time points analyzed. Remarkably,aspirin and clopidogrel showed significant protectionagainst ferric chloride-induced thrombosis in HC-treated group (p<0.001) (Fig. 4).

Discussion

In the present study, the chronology of major patholog-ical events in the hamsters following HC diet has beensystematically monitored. Cholesterol from 0.1% to 3%has been used to induce atherosclerosis in the hamsters[21]. The modified diet used in this study has also been

Fig. 2 Time-dependent inflammatory response in hyperlipidemichamsters. Bar diagram representing fold change in the transcriptlevels of a TNF-α, b IFN-γ, and c iNOS. d Expression profileafter RT–PCR in chow-fed, HC, HC+aspirin, and HC+clopi-

dogrel-fed groups. Results are mean ± SEM from at least fiveanimals from each group. *p<0.05, **p<0.01, ***p<0.001 HCvs. chow fed, #p<0.05, ###p<0.001 vs. HC 90 days

Development of prothrombotic state in hyperlipidemic hamsters 211

used earlier to induce accelerated atherosclerotic load[21]. Despite being on the higher side, it induced atime-dependent increase in the circulating plasmalipids. Earlier reports on cholesterol (0.05%) fed apoEand LDL receptor double knockout mice revealed thatendothelial dysfunction (at 3 months) preceded plateletactivation (4.5 months) [3]. The role of inflammationin this model was, however, not explored. In thepresent study, TNF-α induction, endothelial dysfunc-tion, and platelet activation were observed at 4 weeks.A higher concentration of cholesterol (3%) was used inthis study, which explains the early and simultaneousonset of major pathological changes. Since a signifi-cant increase in the lipid levels was observed from 15to 30 days of HC diet, it can be concluded that athreshold level of circulating lipids was required toinduce major pathological changes.

Local inflammatory process precedes atherosclero-sis; expression of cytokines in the plaque is usually inparallel with spleen cells [39]. A T helper type 1(Th1)/Th2 switch was found to be induced byhypercholesterolemia in apoE-deficient mice [44]. Inthis study, we found that the splenocyte inflammatoryload was augmented under atherosclerotic conditions,implicating the role of systemic inflammation in thisprocess. Response of various inflammatory cellspresent in the spleen should therefore be investigatedfor further elucidating the mechanisms involved.Since a significant increase in iNOS induction wasseen at 90 days, it is possible that early induction ofTNF-α and IFN-γ was required for this induction,which needs experimental verification. Previousreports have also suggested TNF-α-independentiNOS induction in TNF-α knockout cultured macro-

Fig. 3 Effect of HC feeding on platelet activation. a Bar diagramrepresenting whole blood aggregation responses of chow-fed andHC-fed hamsters represented as percent aggregation. Data areexpressed as mean ± SEM from at least ten animals in each group.*p<0.05, **p<0.01, ***p<0.001 vs. chow fed, #p<0.05, ##p<0.001 vs. HC 90 days. b Elevated platelet adhesion on collagen-coated surface after HC diet treatment. Data are expressed as

mean ± SEM from at least ten animals from each group. cEnhanced platelet adhesion on fibrinogen-coated surface after HCdiet treatment. Data are expressed as mean ± SEM from at leastten animals from each group. **p<0.01, ***p<0.001 vs. chowfed; @p<0.05, @@@p<0.001 HC 30 days, #p<0.001 vs. HC90 days

212 V. Singh et al.

phages [7]. Induction of iNOS has been considered asa marker of vascular inflammation, which mightpromote thrombosis and atherosclerosis [20]. Endo-thelial dysfunction and augmentation in TNF-αexpression happened simultaneously; the role ofTNF-α in endothelial dysfunction is thus quite likelyas earlier reports have demonstrated improvement ofaortic stiffness in the patients with inflammatoryarthropathies [2] by using TNF-α antagonist [26]. Asignificant reduction in TT was observed after HCtreatment, indicating enhanced thrombin activity andactivation of the coagulation cascade under hyper-lipidemic conditions. This is in agreement to theplatelet-dependent thrombin generation and activationof coagulation cascade [23]. TTO was significantlyreduced in 90 days HC-fed animals as compared tothe normolipidemic controls, indicating that these

animals were more prone to thrombosis. Since TTOremained unaltered at all the earlier time points, it canbe suggested that sustained hyperlipidemia for3 months was required to induce the prothromboticstate in hamsters.

Earliest change was platelet activation, induction inTNF-α, and endothelial dysfunction. We thereforeexplored the effect of standard anti-platelet drugsaspirin and clopidogrel on changes observed subse-quent to HC diet. Augmentation in TC and TGs levelwas reverted by aspirin treatment, but not withclopidogrel. The lipid-lowering effect of aspirin hasbeen attributed to antioxidant activity [19] of gentisicacid, a metabolic product of salicylate [19]. Anti-lipolytic effect of aspirin due to enhanced insulinsecretion leading to reduced lipolysis and decrease inserum cholesterol and TG has also been demonstrated

Table 3 Coagulation parameters

Days Animal groups TT (s) PT (s) aPTT (s) FT (s)

15 Chow 27±0.9 13±0.8 16±0.9 15±0.7

HC 29±1.3 11±0.3 16±0.8 14±0.9

30 Chow 28±0.7 14±1.2 16±0.9 14±1

HC 28±0.4 12±0.2 15±0.9 15±0.6

60 Chow 28±0.9 13±0.8 15±0.9 15±0.9

HC 29±1.3 11±0.8 14±0.6 16±1

90 Chow 29±0.8 12±0.3 15±0.4 15±0.6

HC 25±0.7* 11±0.2 15±0.5 15±1

HC+aspirin 29±0.7# 12±0.3 18±1.6 16±1

HC+clopidogrel 31±1# 11±0.6 14±0.6 15±0.9

Table representing mean thrombin time (TT), prothrombin time (PT), activated partial thromboplastin time (aPTT), and fibrinogentime (FT) recorded after monitoring the time taken by plasma to form clot after the addition of parameter specific reagents. Results aremean ± SEM from at least ten animals in each group

*p<0.05 vs. chow fed, # p<0.05 vs. HC 90 days

Fig. 4 Thrombotic potential of HC-fed hamsters. Bar diagramrepresenting fold change in the time taken to occlusion after theapplication of 20% ferric chloride to the carotid artery in

various groups. Results are mean ± SEM from at least tenanimals from each group. *p<0.05 vs. chow fed, ###p<0.001vs. HC 90 days

Development of prothrombotic state in hyperlipidemic hamsters 213

[19]. However, LDLC and HDLC levels were notaltered following aspirin treatment. The presence of ahigh level of very low density lipoprotein cholesterolalong with intermediate density lipoprotein cholester-ol and LDLC in hamsters as compared to other rodentspecies conceivably affects the aspirin response. Thispresent finding corroborate with other reports withhigh doses of aspirin which reduced TC withoutaffecting LDLC [15]. Moreover, glucagon-likepeptide-1 receptor agonist exendin-4 reduced plasmatotal cholesterol without affecting LDLC and HDLCcholesterol levels [4]. Although aspirin did notinfluence enhanced adhesion observed after HC diettreatment, clopidogrel could significantly inhibitadhesion on both collagen- and fibrinogen-coatedsurfaces [32]. TXA2 and ADP, by potentiating plateletactivation, led to thrombus formation. However,earlier studies have illustrated that COX inactivationby aspirin did not hamper platelet adhesion overeither collagen or fibrinogen, indicating that TxA2-mediated activation of the GP IIb–IIIa complex mightonly be required for platelet spreading [24]. On thecontrary, P2Y12 receptor antagonist-mediated reduc-tion in the surface coverage was not due to change inthe morphology of the adhered platelets but was dueto a decreased number of platelets adhering to eitherimmobilized fibrinogen or collagen, suggesting thatplatelet adhesion was ADP dependent [29].

Interestingly, by inhibiting platelet activation,aspirin and clopidogrel also reduced the generationof inflammatory response and endothelial dysfunctionsuggesting that platelets could be central to endothe-lial dysfunction and initiation of inflammatory re-sponse. We observed reversion in TNF-α transcriptby aspirin and clopidogrel. At the same time, recoveryin endothelium functionality by aspirin and clopidog-rel was also observed; it can be speculated that theprotective effect of aspirin and clopidogrel onendothelial function and prothrombotic state couldbe associated with the reduction in TNF-α expression.However, this needs further confirmation. Bothaspirin and clopidogrel can modulate TNF-α expres-sion [18]. Moreover, clopidogrel treatment did notshow a change in the expression pattern of IFN-γ andiNOS. Since clopidogrel treatment is well known toreduce the circulating levels of TNF-α [13], thismight also be associated with reduced expression ofsplenocyte TNF (as observed in the present study). Itis possible that certain mediators (not analyzed in the

present investigation) might be involved in iNOS andIFN-γ induction, which were not affected by clopi-dogrel but were reduced by aspirin. The detailedmechanism involved, however, needs further experi-mentation and validation.

Restoration of endothelial function and TNF-αlevel by both clopidogrel and aspirin thereforesuggested a potential role of platelets in thehyperlipidemia-associated pathology. Alterations inTTO observed after 90 days of HC diet treatmentwere significantly reversed by both aspirin andclopidogrel.

Besides suggesting that platelets subsequent tohyperlipidemia could be important to promote endo-thelial dysfunction and inflammation, the presentstudy also highlighted importance of the durationand extent of hyperlipidemia in the induction ofprothrombotic state. Moreover, enhanced splenocyteinflammatory load indicated the role of systemicinflammation in hyperlipidemia associated with theprothrombotic state. Hamsters as an experimentalmodel in this study thus closely mimicked the clinicalatherothrombotic state, suggesting their suitability forthe evaluation of new anti-platelet molecules onhyperlipidemia-associated prothrombotic state.

Acknowledgments Financial support to Manoj KumarBarthwal and Madhu Dikshit from Central Drug ResearchInstitute and NWP0034 is acknowledged. Financial supportfrom the Department of Science and Technology, India toManoj Kumar Barthwal is also acknowledged. Fellowshipmoney from Indian Council of Medical Research, Council ofScientific and Industrial Research, India to the other teammembers is acknowledged. The authors are also thankful to Mr.MPS Negi from the Biometry and Statistics Division, CentralDrug Research Institute, Lucknow for helping in the statisticalanalysis of the data. Both Manoj Kumar Barthwal and MadhuDikshit were equally responsible for the present work.

References

1. Alexaki A, Wilson TA, Atallah MT, Handelman G,Nicolosi RJ (2004) Hamsters fed diets high in saturatedfat have increased cholesterol accumulation and cytokineproduction in the aortic arch compared with cholesterol-fedhamsters with moderately elevated plasma non-HDLcholesterol concentrations. J Nutr 134:410–415

2. Angel K, Provan SA, Gulseth HL, Mowinckel P, KvienTK, Atar D (2010) Tumor necrosis factor-alpha antagonistsimprove aortic stiffness in patients with inflammatoryarthropathies: a controlled study. Hypertension 55:333–338

214 V. Singh et al.

3. Aoki R, Ikarugi H, Naemura A, Ijiri Y, Yamashita T,Yamamoto J (2006) Endothelial dysfunction precedesatherosclerotic lesions and platelet activation in high fatdiet-induced prothrombotic state. Thromb Res 117:529–535

4. Arakawa M, Mita T, Azuma K, Ebato C, Goto H,Nomiyama T, Fujitani Y, Hirose T, Kawamori R, WatadaH (2010) Inhibition of monocyte adhesion to endothelialcells and attenuation of atherosclerotic lesion by aglucagon-like peptide-1 receptor agonist, exendin-4. Diabetes59:1030–1037

5. Bartus M, Lomnicka M, Lorkowska B, Franczyk M,Kostogrys RB, Pisulewski PM, Chlopicki S (2005) Hyper-triglyceridemia but not hypercholesterolemia induces en-dothelial dysfunction in the rat. Pharmacol Rep 57Suppl:127–137

6. Basu R, Bhaumik S, Basu JM, Naskar K, De T, Roy S(2005) Kinetoplastid membrane protein-11 DNA vaccina-tion induces complete protection against both pentavalentantimonial-sensitive and -resistant strains of Leishmaniadonovani that correlates with inducible nitric oxide syn-thase activity and IL-4 generation: evidence for mixed Th1-and Th2-like responses in visceral leishmaniasis. J Immunol174:7160–7171

7. Bekker LG, Freeman S, Murray PJ, Ryffel B, Kaplan G(2001) TNF-alpha controls intracellular mycobacterialgrowth by both inducible nitric oxide synthase-dependentand inducible nitric oxide synthase-independent pathways.J Immunol 166:6728–6734

8. Cappelli-Bigazzi M, Rubattu S, Battaglia C, Russo R, EneaI, Ambrosio G, Chiariello M, Volpe M (1997) Effects ofhigh-cholesterol and atherogenic diets on vascular relaxationin spontaneously hypertensive rats. Am J Physiol 273:H647–H654

9. Davignon J, Ganz P (2004) Role of endothelial dysfunctionin atherosclerosis. Circulation 109:III27–III32

10. Felmeden D, Nadar SK, Lip GY (2005) Aspirin andendothelial function in hypertension. J Hum Hypertens19:663–665

11. Ferroni P, Basili S, Davi G (2003) Platelet activation,inflammatory mediators and hypercholesterolemia. CurrVasc Pharmacol 1:157–169

12. Glass CK, Witztum JL (2001) Atherosclerosis. The roadahead. Cell 104:503–516

13. Gurbel PA, Bliden KP, Tantry US (2006) Effect ofclopidogrel with and without eptifibatide on tumor necrosisfactor-alpha and C-reactive protein release after electivestenting: results from the CLEAR PLATELETS 1b study. JAm Coll Cardiol 48:2186–2191

14. Heitzer T, Rudolph V, Schwedhelm E, Karstens M, SydowK, Ortak M, Tschentscher P, Meinertz T, Boger R, Baldus S(2006) Clopidogrel improves systemic endothelial nitricoxide bioavailability in patients with coronary arterydisease: evidence for antioxidant and antiinflammatoryeffects. Arterioscler Thromb Vasc Biol 26:1648–1652

15. Hundal RS, Petersen KF, Mayerson AB, Randhawa PS,Inzucchi S, Shoelson SE, Shulman GI (2002) Mechanismby which high-dose aspirin improves glucose metabolismin type 2 diabetes. J Clin Invest 109:1321–1326

16. Ikenoya M, Yoshinaka Y, Kobayashi H, Kawamine K,Shibuya K, Sato F, Sawanobori K, Watanabe T, Miyazaki

A (2007) A selective ACAT-1 inhibitor, K-604, suppressesfatty streak lesions in fat-fed hamsters without affectingplasma cholesterol levels. Atherosclerosis 191:290–297

17. Ivandic BT, Schlick P, Staritz P, Kurz K, Katus HA,Giannitsis E (2006) Determination of clopidogrel resistanceby whole blood platelet aggregometry and inhibitors of theP2Y12 receptor. Clin Chem 52:383–388

18. Joussen AM, Poulaki V, Mitsiades N, Kirchhof B, KoizumiK, Dohmen S, Adamis AP (2002) Nonsteroidal anti-inflammatory drugs prevent early diabetic retinopathy viaTNF-alpha suppression. FASEB J 16:438–440

19. Kouraklis G, Patapis P, Misiakos E, Glinavou A, Sioka C,Karayiannakos PE (2004) Effects of acetylsalicylic acid onexperimental atherogenesis induced in rabbits. Int Angiol23:139–143

20. Lazar HL, Joseph L, San Mateo C, Frame J, Cabral HJ,McDonnell M, Chipkin S (2010) Expression of induciblenitric oxide synthase in conduits used in patients withdiabetes mellitus undergoing coronary revascularization. JCard Surg 25:120–126

21. Mangiapane EH, McAteer MA, Benson GM, White DA,Salter AM (1999) Modulation of the regression ofatherosclerosis in the hamster by dietary lipids: comparisonof coconut oil and olive oil. Br J Nutr 82:401–409

22. Mito N, Yoshino H, Hosoda T, Sato K (2004) Analysis ofthe effect of leptin on immune function in vivo using diet-induced obese mice. J Endocrinol 180:167–173

23. Monroe DM, Hoffman M, Roberts HR (2002) Platelets andthrombin generation. Arterioscler Thromb Vasc Biol22:1381–1389

24. Nakamura T, Jamieson GA, Okuma M, Kambayashi J,Tandon NN (1998) Platelet adhesion to native type Icollagen fibrils. Role of GPVI in divalent cation-dependentand -independent adhesion and thromboxane A2 generation. JBiol Chem 273:4338–4344

25. Onley DJ, Knight CG, Tuckwell DS, Barnes MJ, FarndaleRW (2000) Micromolar Ca2+ concentrations are essentialfor Mg2+-dependent binding of collagen by the integrinalpha 2beta 1 in human platelets. J Biol Chem 275:24560–24564

26. Picchi A, Gao X, Belmadani S, Potter BJ, Focardi M,Chilian WM, Zhang C (2006) Tumor necrosis factor-alphainduces endothelial dysfunction in the prediabetic metabolicsyndrome. Circ Res 99:69–77

27. Prociuk MA, Edel AL, Richard MN, Gavel NT, Ander BP,Dupasquier CM, Pierce GN (2008) Cholesterol-inducedstimulation of platelet aggregation is prevented by ahempseed-enriched diet. Can J Physiol Pharmacol86:153–159

28. Raymond C, Menon V (2009) Dual antiplatelet therapy incoronary artery disease: a case-based approach. Cleve ClinJ Med 76:663–670

29. Remijn JA, Wu YP, Jeninga EH, IJ MJ, van Willigen G, deGroot PG, Sixma JJ, Nurden AT, Nurden P (2002) Role ofADP receptor P2Y(12) in platelet adhesion and thrombusformation in flowing blood. Arterioscler Thromb Vasc Biol22:686–691

30. Santhanam AV, Viswanathan S, Dikshit M (2007) Activationof protein kinase B/Akt and endothelial nitric oxide synthasemediates agmatine-induced endothelium-dependent relaxation.Eur J Pharmacol 572:189–196

Development of prothrombotic state in hyperlipidemic hamsters 215

31. Schafer A, Bauersachs J (2008) Endothelial dysfunction,impaired endogenous platelet inhibition and platelet activationin diabetes and atherosclerosis. Curr Vasc Pharmacol 6:52–60

32. Shenkman B, Matetzky S, Fefer P, Hod H, Einav Y,Lubetsky A, Varon D, Savion N (2008) Variable respon-siveness to clopidogrel and aspirin among patients withacute coronary syndrome as assessed by platelet functiontests. Thromb Res 122:336–345

33. Siegel-Axel D, Daub K, Seizer P, Lindemann S, Gawaz M(2008) Platelet lipoprotein interplay: trigger of foam cellformation and driver of atherosclerosis. Cardiovasc Res78:8–17

34. Singh V, Tiwari RL, Dikshit M, Barthwal MK (2009)Models to study atherosclerosis: a mechanistic insight. CurrVasc Pharmacol 7:75–109

35. Surin WR, Prakash P, Barthwal MK, Dikshit M (2010)Optimization of ferric chloride induced thrombosis modelin rats: effect of anti-platelet and anti-coagulant drugs. JPharmacol Toxicol Methods 61:287–291

36. Tancevski I, Wehinger A, Schgoer W, Eller P, Cuzzocrea S,Foeger B, Patsch JR, Ritsch A (2006) Aspirin regulatesexpression and function of scavenger receptor-BI in macro-phages: studies in primary human macrophages and inmice. FASEB J 20:1328–1335

37. Tandon NN, Kralisz U, Jamieson GA (1989) Identificationof glycoprotein IV (CD36) as a primary receptor forplatelet–collagen adhesion. J Biol Chem 264:7576–7583

38. Tauseef M, Shahid M, Sharma KK, Fahim M (2008)Antioxidative action of aspirin on endothelial function in

hypercholesterolaemic rats. Basic Clin Pharmacol Toxicol103:314–321

39. Tedgui A, Mallat Z (2006) Cytokines in atherosclerosis:pathogenic and regulatory pathways. Physiol Rev 86:515–581

40. Tiwari RL, Singh V, Barthwal MK (2008) Macrophages: anelusive yet emerging therapeutic target of atherosclerosis.Med Res Rev 28:483–544

41. Venter JC, Adams MD, Myers EW, Li PW, Mural RJ,Sutton GG, Smith HO, Yandell M, Evans CA, Holt RA etal (2001) The sequence of the human genome. Science291:1304–1351

42. Waksman R, Ajani AE, White RL, Pinnow E, Dieble R,Bui AB, Taaffe M, Gruberg L, Mintz GS, Satler LF et al(2001) Prolonged antiplatelet therapy to prevent latethrombosis after intracoronary gamma-radiation in patientswith in-stent restenosis: Washington radiation for in-stentrestenosis trial plus 6 months of clopidogrel (WRISTPLUS). Circulation 103:2332–2335

43. Waksman R, Pakala R, Roy P, Baffour R, Hellinga D,Seabron R, Chan R, Scheinowitz M, Kolodgie F, VirmaniR (2008) Effect of clopidogrel on neointimal formation andinflammation in balloon-denuded and radiated hypercho-lesterolemic rabbit iliac arteries. J Interv Cardiol 21:122–128

44. Zhou X, Paulsson G, Stemme S, Hansson GK (1998)Hypercholesterolemia is associated with a T helper (Th) 1/Th2 switch of the autoimmune response in atheroscleroticapo E-knockout mice. J Clin Invest 101:1717–1725

216 V. Singh et al.