Metabolic and anti-atherogenic effects of long-term benfluorex in dyslipidemic insulin-resistant sand rats (Psammomys obesus)

METABOLIC AND ANTI-ATHEROGENIC EFFECTS OF LONG-TERM BENFLUOREX IN DYSLH’IDEMIC INSULIN-RESISTANT SAND RATS (PSAMMOMYS OBESUS) G. Marquit’,

T. El Madam’, M.L. Solera*, M.T. Pieragp3, P. Hadjiisky4, D. Ravel’, L. Seguin’ and N. Bennani

’ Laboratoire de Recherche des Macrophages, Mediateurs de 1’Inflammation et Interactions Cellulaires; * Laboratoire de Biochimie; 3 Unite INSERM 466, CHU Rangueil, Toulouse, France; 4 Centre de Cardiologie Claude Bernard, Hopital La PitiC-Salpttribre, Paris, France; Institut de Recherches Intemationales Servier, ’ Division Therapeutique Metabolisme, Courbevoie, France; 6 Unite de Nutrition, Universite Mohammed V, Rabat, Morocco.

(Received in final form April l5,19%) Summary Benfluorex is a clinical lipid-lowering agent with antihyperglycemic properties. The effect of long-term oral treatment (10 mg/kg/day for 7.5 months) on carbohydrate and lipid metabolism and aortic morphology was investigated in 24 insulin-resistant sand rats receiving a standard laboratory diet supplemented with cholesterol (2%). Untreated controls (n=34) developed impaired glucose tolerance, hyperinsulinemia, hypertriglyceridemia, hypercholesterolemia and elevated plasma LDL- and VLDL-cholesterol, positively correlated with the proportion of the thoracic aorta displaying oil red O-positive atherosclerosis; ultrastructural examination showed intimal lipid deposits, foam cells, polymorph infiltrates and fibrosis. Benfluorex-treated animals showed decreases in intolerance, hyperinsulinemia, significant glucose hypertriglyceridemia, hypercholesterolemia, and plasma LDL- and VLDLcholesterol, with no evidence of aortic atheroma. The metabolic benefits of benfluorex may protect against the long-term development of atherosclerosis in the insulin-resistant dyslipidemic syndrome. &Y ~onlr: benfluorex, anti-diabetic agent, bypolipidemic agent, sand rat, syndrome X, insulin resistance, dyslipidemia,atherosclerosis The sand rat, Psammomys obesus, when subjected to a standard laboratory diet instead of its native salt bush diet, provides a useful model of the biochemistry and pancreatic/endocrine histomorphology and ultrastructure of maturity-onset (type II) diabetes (l-4). Diabetic sand rats also exhibit degenerative microangiopathy (5). Cholesterol supplementation of the standard diet (2% w/w) induces hypercholesterolemia and cholesterol-overload lesions rather than true atherosclerosis (6); however, addition of an antithyroid agent (0.01% methimazole) to the hypercholesterolemic diet for 45 days induces atherosclerotic lesions (lipid deposits and cell wall changes with interstitial fibrosis, lesions of the elastic lamellae, smooth muscle cell disorientation and proliferation) and complications (fissuring, hemorrhagic infiltration of the vessel wall and occasional fibrinoid necrosis). Additional features are hyperinsulinemia, overweight (not massive obesity), glucose intolerance, insulin resistance, hyperlipidemia, Corresponding author: Pr G. Marquit?, Laboratoire de Recherche des Macrophages, Mediamum de lIr&mmation et Interactions Cellulaires, Batiment Ll CHU rangueil, 1 avenue Jean Poulhes, 31043 Toulouse, France - Tel/Fax : 05.62.14.19.78

66

Anti-athero6derotic Effecta ofBenfluorex

Vol. 63, No. l, l!H?

moderate hyperthyroidism, and hypertension (G. Marquie, M. Sebbar, M.T. Pieraggi, N. Dousset, J. Duhault and N. Bennani, unpublished). All together, these symptoms mimic the syndrome X described in humans, an association of dyslipidemia, hypertension, impaired glucose intolerance, insulin resistance and vascular changes (7). The sand rat is thus an ideal model for testing the effect of new agents on the cardiovascular risk factors associated with diabetes and atherosclerosis. Benfluorex improves insulin resistance in the rat subjected to high fat or fructose diet (8) and, in n5STZinduced diabetic rats, normalizes, hyperglycemia and reverses hepatic insulin resistance (9). It also decreases serum lipids and triacylglycerol synthesis in the adipose tissue of JCR:LA-corpulent (cp) rats (10,ll). Long-term treatment has recently been discovered to have anti-atherogenic effects in insulin-resistant male JCR:LA-cp rats (12). The present study addressed the effect of long-term treatment on insulin resistance, plasma lipids and atherosclerosis in dyslipidemic insulin-resistant sand rats. Methods Animals 36 male and 43 female Psammomys obesus were used in the study. The animals were gathered in the Moroccan desert and transported to the laboratory animal house (Paul Sabatier University, Toulouse, France) where they were kept under constant conditions (25°C 70% hygrometry and a 12 hour light-dark cycle). The study was approved by the Ethics Committee and complied with Principles of Laboratory Animal Care (NM publication no 83-25 revised 1985) and the French law regulating animal experimentation (Decree no 87-848 19 October 1987, and the Ministerial Decrees of 19 April 1988). Study protocol and treatments Male and female sand rats were assigned to three groups: - The vegetable diet (VD) group (n=21) received a diet resembling that in their natural environment. This group controlled for the vascular effects of aging. Three animals were sacrificed at 6 months and the remaining 18 at 9 months. - The high cholesterol diet (HCD) group (n=34) received a standard laboratory diet supplemented with cholesterol (2%), lard (3%) and the antithyroid agent methimazole (0.01%) for the first 45 days to promote atherosclerosis, followed by supplementation with 1% cholesterol and 3% lard alone until the end of the experiment. Three animals were sacrificed at 3,4, 5,6 and 7 months and the 19 remaining at 9 months. - The high cholesterol diet + benfluorex (HCD+B) group (n=24) received the same diet as the HCD group plus oral benfluorex (10 mg/kg by once daily gavage) from day 45 to 9 months (Fig. 1). Three animals were sacrificed at 3 and 6 months and the 18 remaining at 9 months. All animals were weighed weekly. Biochemical parameters Urine and retro-orbital blood samples were taken weekly, and liver fragments after sacrifice at 6 and 9 months, Plasma glucose, total and free cholesterol, triglycerides and free fatty acid fractions were measured in a COBAS Mira S analyser (Roche Diagnostic Systems, Neuillysur-Seine, France) according to the manufacturer’s instructions. Insulin was measured by radioimmunoassay using a SB-INSIkit (CIS Bio international, France) with rat insulin as standard (cross-reactivity with sand rat proinsulin antibodies has not been characterized). HDL, LDL- and VLDL-cholesterol were measured in a combined test (Boehringer). Oxidized lipoprotein was determined using malonedialdehyde (13). Strip tests were used for urinary glucose, ketone and albumin (Miles Laboratories). Hepatic parameters (phospholipids, total, free and esterified cholesterol and triglycerides) were measured after total lipid extraction in chloroform-methanol (2:1, v/v). Glycogen was extracted by boiling with 30% potassium

67

Vol. 63, No. 1,1998

Duration Groups (number of rats)

of diet, in months

O,;,i vegetable

VD (21)

)

2% cholesterol diet + 0.01% antithyroid drug HCD (34)

4

Fig. 1: St&y

(24) design.

1% cholesterol

diet W

)41

2% cholesterol diet + 0.01% antithyroid drug HCD+B

diet

4

VD: vegetable

1% cholesterol diet + benfluorex treatment

4

W

diet (controls). HCD: high cholesterol diet. B:

benfluorex hydroxide followed by measurement of the hydrolyzed glucose. Lecithin acyltransferase (LCAT) activity was determined by radioimmunoassay (14).

cholesterol

Glucose tolerance test A pre and end of study oral glucose test was performed as previously described: briefly, a retro-orbital blood sample was taken after a 12 h overnight fast; glucose solution (2 g/kg body weight) was administered by gavage and blood resampled at 90 min. Plasma glucose 2 9 mmoY1 at 90 min was considered as a state of impaired glucose tolerance (3,14). Morphological parameters After sacrifice by bleeding, specimens of thoracic and abdominal aorta were fixed in Bouin’s solution and processed for histology and histochemistry by staining with Masson’s trichrome and iodine hematoxylin. Some fragments were frozen, fixed in formalin and stained with Sudan black. Atherosclerosis was scored on a 5point scale (16): 0 = normal; 1 = disorganized connective tissue (increased collagen and superficial elastic fiber degeneration); 2 = lipid infiltration (thickening of the subendothelial space with foam cells and numerous collagen fibers); 3 = extensive lipid plaques (fragmented internal elastic lamina; internal zone of the tunica media comprising smooth muscle cells invaded by lipids, scattered elastic elements and disordered collagen fibers; no involvement of the external zone of the tunica media or tunica adventitia); 4 = converging lipid plaques (17). Image analysis topography of extracellular and intracellular lipids was performed from the 3 groups using alcoholic Oil Red 0 (ORO) solution. Samples from both non lipid areas were stained with hematoxylin-eosin-safranin (HES), Gallego fibers), azane, Sudan black (for lipids), and results expressed as an aortic lipidosis area (mm2)/total aortic area (mm2)%].

in 47 aortas positive and (for elastic index [lipid

Statistical analysis (IRIS Biometrics Division, Courbevoie, France) Data were expressed as the mean + SEM and analysed by analysis of variance [plasma insulin, cholesterol and triglycerides by two-way ANOVA (factors: group and time); glucose tolerance test, lipid and hepatic parameters, aortic lipidosis scores by one-way ANOVA (factor: group)] followed by a Newman-Keuls test of significant interaction (~~0.05). Intergroup atherosclerosis scores were compared using Fisher’s exact test.

Vol. 63, No. l, l9W

Anti-atheroscleroticEffects of Benfluorex

Results Since no significant sex differences have been found in any parameter in sand rats receiving a high cholesterol diet with or without benfluotex, results in male and female rats were pooled. Vegetuble diet controls Plasma and tissue biochemistry was similar to that in sand rats living in their biotope, with no hyperglycemia or glycosuria and generally normal glucose tolerance (Fig. 2). Only plasma insulin showed a slight increase from 21.8 + 1.4 to 38.6 + 2.3 VU/ml (Fig. 3a). Lipid (Fig. 3b and c) and lipoprotein (Table 1) profiles were normal. Hepatic tissue biochemistry was normal (Table2). There were no aortic lesions or lipid plaques (Table 3). A few animals showed minor widening of the first media space (Table 4).

VD

HCD

HCD+B

(n=lS)

(n=19)

(n=18)

HDL-cholesterol

(mmol/l)

0.65 I!Z 0.02

1.20 k 0.28”

3.38 zk0.53”

LDL-cholesterol

(mmol/l)

0.67 + 0.05

17.21 f 1.76”

7.65 fi 0.89”.b

0.29 + 0.02

5.33 f 0.55”

1.84 k 0.22”.b

HDLILDL + VLDL-cholesterol: anti-atherogenic index

0.70 + 0.03

0.05 f. 0.01”

0.64 f 0.25

MDA (pmol/l)

55.72 f 3.94

278.68 + 25.36”

118.61 f 7.828.b

LCAT (pmol/h/l)

15.06 + 0.58

51.58 * 3.06”

31.67 I!I2.92”.b

VLDL-cholesterol

(mmol/l)

Table 1: Lipoprotein and LCAT levels at 9 months, in three groups of sand rats receiving a vegetable diet (VD, controls) and a high cholesterol diet (HCD) f benfluorex (B). The number of animals in each group is indicated in parentheses. Values are mean f SEM with” p ~0.001: HCD vs VD and b p
vs

HCD

Standard laboratory diet + cholesterol with or without benjluorex Body weight As previously reported, feeding sand rats with standard laboratory diet supplemented with cholesterol (HCD) resulted in increased body weight gain when compared to control animals (6). This increase could be interpreted as the onset of a moderate obesity, and was prevented by benfluomx administration (Table 2). Glucose tolerance Fasting glucose was not modified by HCD diet. However, almost all tolerance tests, performed in the 9th month, showed glucose intolerance in the HCD group (Fig. 2), with some 90 min values reaching 35 mmol/l. This Cas accompanied by fasting hyperinsulinemia (75.9 f 12.4 vs 38.6 + 2.3 yU/ml in control animals, p&01, Fig. 3a), as well as frequent glucosutia and ketonuria. Benfluorex-treated rats showed normal glucose tolerance with a mean 90 min glucose levels of 8.3 Y!I0.6 vs 16.7 f 1.8 mmol/l in the HCD group (pcO.001; Fig. 2), i.e. glucose intolerance decreased markedly under benfluorex treatment, as did plasma insulin (41.4 * 6.5 vs 75.9 f 12.8 @/ml, respectively; pcO.01; Fig. 3a).

Anti-schcrosclcrotic Effa

Vol. 63, No. l,l!J98

ofBcntluorcx

69

VD

HCD

HCD+B

(n=18)

(n=19)

(n=18)

Body weight(g)

178.9 + 2.0

190.1 f 4.Sb

169.3 + 2.1b.d

Liver weight (g)

6.33 f 0.26

8.73 ?I 0.46b

6.46 k 0.22d

Phospholipids

3.23f0.11

3.94 k 0.25”

3.71 f 0.16

Total cholesterol

0.29 f 0.02

6.27 f 0.55b

3.54 k 0.28d

Esterified cholesterol

0.05 f 0.01

5.02 f 0.53b

2.32 + 0.30d

Free cholesterol

0.24 zk0.01

1.25 rt 0.05b

1.22 + 0.06”

Triglycerides

0.97 zk0.09

3.06 f 0.36b

1.41 + 0.12d

Glycogen

0.78 f 0.09

0.41 * 0.07”

0.95 f 0.10’

Biochemistry (g/100 g fresh tissue)

Table 2: Hepatic biochemistry at 9 months in three groups of sand rats receiving a vegetable diet (VD, controls) and a high cholesterol diet (HCD) * benfluorex (B). Values are mean + SEM and the number of animals in each group is indicated in parentheses. a ~~0.02; b p
vs controls together with hepatic triglycerides, hepatic total cholesterol and, in particular, esterified cholesterol. Hepatic glycogen was reduced vs controls (Table 2). Compared to the HCD group, liver weight was not increased in the benfluorex-treated rats; hepatic triglycerides were lower, as was hepatic cholesterol, particularly the esterified (Table 2). 20-

20-

HCD+B

VD

,J

*MO

r

I

0

*rut0

-e-M9

Minutes

So

-era9

,

0

4M0

I Minutes

”

d-M9

I

0

I Minutes

‘O

Fig. 2: Glucose tolerance tests in three groups of sand rats receiving a vegetable diet (VD, controls) and a high chdesterol diet (HCD) f benfluorex (B), at baseline (MO) and at 9 months (M9). Blood glucose was assayed before (0) and 90 min after an oral glucose bolus (2gkg) and are expressed as mean f SEM. * p
70

Voi.63,No.l,1998

Anti---Of-

Blood lipids In rats fed a HCD feed, plasma phospholipids increased to 9 months together with triglycerides (2.7 f 0.4 vs 1.6 + 0.07 mmol/l in controls; Fig. 3~). However, the increase was greatest in the cholesterol fraction (24.0 & 2.0 vs 1.6 f 0.06 mmol/l, respectively; Fig. 3b), with levels already high at D45 (18.8 ?I 1.4 mmol/l). The VLDL and LDL fractions considerably increased from D45, while HDL decreased from D45 and the anti-atherogenic index fell rapidly to 9 months. Oxidized LDL, already increased at D45, remained high up to 9 months (Table 1). Molar LCAT was greatly increased at 9 months (Table 1). In benfluorex-treated rats, plasma trigl~rides were markedly decreased at 9 months when compared to the HCD group (1.4 f 0.2 vs 2.7f 0.4 mmol/l, p
I

I

Total aorta

HCD (n=19)

I

0.19 f 0.08

8.82 + 1.14”

I

HCD+B (n=18)

I

2.46 +_0.45b

Table 3: Aortic lipid&s index (lipid area (mm’)/total aortic area (mm’)%) at 9 months in three groups of sand rats receiving a vegetable diet (VD, controls) and a high cholesterol diet (HCD) f benfluorex (B). Values are mean + SEM of (n) animals in each group. a pcO.001: HCD vs VD; b p
Advanced widespread lesions developed from 5 months onwards (Fig. 4a), with lipid in the inner media spaces (Fig. 4b) and extracellular lipid composed of cholesterol crystals and intracellular lipid. Lipid deposits were associated with major changes in the vessel wall: interstitial fibrosis, elastic lesions, and smooth muscle cell disorientation and proliferation in the first 3 or 4 layers of the media, i.e. almost the inner half of the wall (Fig. 4c, d and e), with atherosclerosis scores of 3-4 (Table 4). The increased aortic lipidosis index (Table 3) confirmed the histological findings. ORO-positive foci were particularly abundant in the thoracic aorta. The atherosclerosis score increased with the persistent hypercholesterolemia to 6 months, remaining stable thereafter (Table 4). None of the aortic lesions seen in the HCD group were found in the benfluorex-treated group, in particular no fibrous base, parietal fibrosis, aneurysm, hemorragic infiltration, macrophages, or infiltration of the adventitia (Fig. 5a). At 3 and 6 months some splitting and fragmentation appeared, extending to the first spaces. At 9 months, half the aortas (n+ were apparently

Anti-atheroscleroticJZffeusofBentluorex

Vol. 63, No. l, 1998

A = 120E 3 2 5

SO-

HCD

40-

HCD+B VD

: C : ii P

OJ I

0

I

1.5 4

3

6

6

months

) benfluorex

treatment

OI 0

I 1.5

I 3

1

1 6

4

I

I Q

months

) benfluorex

treatment

C 0

e

1

0

I

1.5

6

3

4

9

months

) benfluorex

treatment

Fig. 3: Effects of benfluorex on plasma parameters in three groups of sand rats receiving a vegetable diet (VD, controls; n=Zl; -I%), a high cholesterol diet (HCD; n=34; -O-) and a high cholesterol diet + benfluorex (HCD+B; n=24; -O-). A: Plasma insulin, B: Plasma cholesterol and C: Plasma triglycerides. Each point indicates a mean f SEM value. * ~~0.01 and ** p
71

Anti-atherosclerotic Effects of Benfluorex

Vol. 63, No. l, W98

Fig. 4: Atherosclerosis lesions in sand rats receiving a high cholesterol diet. a) General view of an atherosclerotic aorta with diffuse lipid deposits and variable media penetration (arrow). Masson trichome, x 10. b) Major lipid deposit occupying the internal half of the media. Onset of underlying fibrosis. Sudan black, x 40. c) Detail of atherosclerotic plaque with intra and extracellular lipid deposits. Lesion present in the first two medial layers. x 40. d) Atherosclerotic plaque with lipid deposits, fibrosis and cellular proliferation in the first three layers of the media. x 40. e) Atherosclerotic plaque with a major lipid deposit, fibrosis of the internal media and superficial fibrotic strip (arrow). Masson trichrome, x 40. Atherosclerosis

score

VD (n=18)

0

12

0.5

4

1

2

HCD (n=19)

HCD+B (n=18) 8

3

1.5

5 3

2

6

3

3

4

7

2

Table 4: Distribution of aortic atherosclerosis scores (see text) at 9 months in three groups of sand rats receiving a vegetable diet (VD, controls) and a high cholesterol diet (HCD) * benfluorex (B). a p
73

Fig. 5: Sand rats receiving high cholesterol diet with benfluorex treatment. a) No lesions. Gallego, x 40. b) Variable subendothelial fibrosis. Azane, x 40. c) Minor proliferation of the intima. HES stain, x 40. d) Some foam cells in the intima (above) and lipids in the adventitia. Sudan black, x 40. e) No lipids in the intima, trace amounts in the adventitia. Sudan black, x 40.

normal in structure, while half showed some elastic fiber thickening, muscle cell disorientation and a few intimal foam cells (Fig. 5b, c, d and e). The decrease in the frequency and dimensions of aortic ORO-positive foci resulted in a 4-fold lower lipidosis index than in the HCD group (Table 3). Quantitative morphological outcome confirmed the qualitative data, with low end of study atherosclerosis scores in the benfluorex-treated group (Table 4).

74

Aai-atho

Fxfects of Be&lorex

Vol.63,No.l,1998

Discussion Sand rats fed a standard laboratory diet supplemented with cholesterol and (for a short period) methimazole show: - insulin resistance associated with hyperinsulinemia and glucose intolerance. - dyslipidemia: increased plasma and hepatic cholesterol and triglycerides; markedly increased plasma lipid fractions (LDL and VLDL); and increased oxidized LDL and LCAT activity. - atherosclerotic lesions : lipid deposits, fibrosis, and smooth muscle cell proliferation mimicking those described in humans (18-22). Lesion evolution was monitored by staggered sacrifice from 3 months: months 3-4 were characterized by reversible lesions, equivalent to fatty streaks (22); in months 5-9, the intimal lesions became irreversible, damaging the first spaces of the media, and constituting an atherosclerotic pustule marked by smooth muscle cell proliferation and fibrosis surrounding the lipid core (21-23). These are also features of human syndrome X (10, 12, 24 and G. Marquie, M. Sebbar, M.T. Pieraggi, N. Dousset, J. Duhault and N. Bennani, unpublished) which, as in animal models (25), is associated with dyslipidemia, hypertension, impaired glucose tolerance, and vascular changes. As the sand rat model also comprises several risk factors for atherosclerosis: hyperinsulinemia, overweight, impaired glucose tolerance, hyperlipidemia, moderate hyperthyroidism, and hypertension, it provides a highly instructive model for investigating both syndrome X and its associated atherosclerosis. The study provided further information on the effect of long-term benfluorex, which has established hypolipidemic and antihyperglycemic activity (8-12). Treatment with 10 mg/kg/day for 7.5 months improved the metabolic features of syndrome X displayed by our model, with particular respect to carbohydrates metabolism. Hyperinsulinemia was decreased by 45% vs the HCD group and glucose tolerance significantly improved, indicating enhanced hepatic and muscle tissue sensitivity to insulin. Benfluorex has shown similar effects in other insulin-resistant rat models. Sprague-Dawley rats are insulin-resistant at 52 weeks and have impaired glucose tolerance. Benfluorex 2 x 2.5 mgfkglday p.o for 14 days has been shown to significantly improve the response to an intravenous glucose tolerance test (26). Long-term treatments (50 mg/kg/day) have also been studied in rats receiving a high-lipid or high-fructose diet, both of which induce insulin resistance, primarily in muscle and adipose tissue; benfluorex improved insulin resistance and reduced hyperinsulinemia in the high-fructose animals (8). In addition to its significant antidiabetic effect, benfluorex improves dyslipidemia in sand rats fed a high cholesterol diet. In our study, vs the HCD group, plasma and hepatic total cholesterol levels were halved, and the tissue esterified fraction decreased by 65%. Plasma LDL and VLDL were decreased by 60% and HDL increased by the same proportion. Oxidized LDL was decreased by 57%. These effects confirm recent in vitro and in vivo data, e.g. decrease in significant enhancement of LDL catabolism, together with a dose-dependent radioactive oleic acid incorporation into cholesterol esters, after incubating human pulmonary fibrobiasts and J774 murine cells with the main bioactive metabolite of benfluorex, S422 (27). More recently, Orsibe et al. (28) showed that benfluorex and S422 decrease ACAT activity in rat hepatic microsomes. Decrease in an enzyme which esterifies excess cellular cholesterol decreases the proportion of cholesterol available for oxidation, a step known to be involved in the formation of atherosclerotic lesions. In vivo, the incorporation of benfluorex 30 mg/kg/day into the standard laboratory chow of insulin-resistant JCR:LA-cp rats from weeks 39 to 69 decreases free and ester&d cholesterol levels by 30% and significantly decreases aortic

Vol. 63, No. l, 1998

75

EffectsofB4%fluorex

Anti-e

atherosclerosis (12). This supports the present study in suggesting of benfluorex may prevent the development of atherosclerosis.

that the metabolic

benefits

Benfluorex also decreased plasma and hepatic triglycerides by 47% and 64%, respectively. S422 has been shown to inhibit triglyceride synthesis: it decreases radioactive oleic acid incorporation into triglycerides in human fibroblast cultures (27), while in vim, chronic benfluorex treatment dose-dependently decreases plasma triglycerides in aged Sprague-Dawley rats (27) and JCR:LA-cp rats (12). In summary, long-term benfluorex improves lipid high-cholesterol diet. The decreases in hepatic together with the lower plasma insulin levels contribute to the protective effect on the vascular These data support the use of benfluorex in human

and carbohydrate profiles in sand rats fed a and plasma triglycerides and cholesterol, and improved glucose tolerance, probably wall observed in benfluorex-treated animals. diabetes and related conditions.

Acknowledgements We thank J.C. THIERS for expert technical assistance and J. AUCLAIR for statistical analysis. This investigation was supported by the Group on Multifield Research in Diabetes and Degenerative Complications in the Sand Rat (Coordinator: Prof. G. MARQUIE): Centre de Recherche Cardiologique Cl. Bernard, L. PitiC-SalpCtribre, Paris, France (Prof. P. HADJIISKY); INSERM Unite 63, Lyon, France (Dr. M.C. BOURDILLON); Institut de Recherches Servier, Courbevoie, France (Dr. J. DUHAULT); Laboratoire de Biochimie, CHU Toulouse-Rangueil, France (Dr. M.L. SOLERA) and INSERM Unite 305 (Dr. N. DOUSSET); Laboratoire de Recherche des Macrophages, Mediateurs de 1’Inflammation et Interactions Cellulaires, CHU Rangueil, Toulouse, France (Prof. G. MARQUIE); Unite INSERM 446, CHU Rangueil, Toulouse, France (Drs. M.T. PIERAGGI and M. MAUREL); Department of Anatomy, Academy of Medicine, Sofia, Bulgaria (Prof. P. PETKOV); and Unite de Nutrition, Universitk Mohammed V, Rabat, Morocco (Prof. N. BENNANI).

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.

K. SCHMIDT-NIELSEN, K.B. HAINES and D.B. HACKEL, Science 143 689-690 (1964). G. MARQUIE, P. PETKOV and J. DUHAULT, Diabetologia 19 297 (1980). G. MARQUIE, J. DUHAULT and B. JACOTOT, Diabetes 33 438-443 (1984). P. PETKOV, G. MARQUIE, S. DONEV, Y. DAHMANI and J. DUHAULT, Cell. Mol. Biol. 31 61-74 (1985). G. MARQUIE, J. DUHAULT, P. HADJIISKY, P. PETKOV and H. BOUISSOU, Cell. Mol. Biol. 37 651-667 (1991). G. MARQUIE, P. HADJIISKY, 0. ARNAUD and J. DUHAULT, Am. J. Med. 90 (Supp. 6A) 55S-61s (1991). G.M. REAVEN, Diabetes Care 14 195-202 (1991). L.H. STORLIEN, N.D. OAKES, D.A.B. PAN, M. KUSUNOKI and A.B. JENKINS, Diabetes 42 457-462 (1993). P. SERRADAS, 0. BLONDEL, D. BAILBE and B. PORTHA, Diabetes 42 564-570 (1993). D.N. BRINDLEY, P. HALES, A.I.L. AL-SIENI and J.C. RUSSELL, Biochim. Biophys. Acta 1085 119-125 (1991).

76

Anti-atherosclerotic Effects of Benfluorex

Vol. 63, No. 1, 1998

J.C. RUSSELL, S.E. GRAHAM, P.J. DOLPHIN and D.N. BRINDLEY, Can. J. Physiol. Pharmacol. 74 879-886 (1996). 12. J.C. RUSSELL, S.E. GRAHAM, P.J. DOLPHIN, R.M. ARMY, G.O. WOOD AND D.N. BRINDLEY, Atherosclerosis. In press 13. L.G. ALCINDOR and H. ANTEBI, Molecular Biology of Atherosclerosis, M.J. Halpem (Ed), 383-388, J. Libbey and Company Ltd, (1992). 14. K.T. STOKKE and K.R. NORUM, Scand. J. Clin. Lab. Invest. 17 21-27 (1971). 1.5. J. DUHAULT, F. LACOUR, M. BOULANGER, 0. DELLA-ZUANA, D. RAVEL, M. WIERZBICKI and J. ESPINAL, Diabetologia 37 969-975 (1994). 16. P. GENDRE, Pathol. Eur. 4 235-256 (1969). 17. G. MARQUIE, Diabetologia 14 269-275 (1978). 18. R. ROSS, N. Engl. J. Med. 314 488-500 (1986). 19. R. ROSS, Nature 362 801-808 (1993). 20. C.J. SCHWARTZ, A.J. VALENTE and E.A. SPRAGUE, Am. J. Cardiol. 71 9B-14B (1993). 21. H.C. STARY, Basic Res. Cardiol. 89 (suppl. 1) 17-32 (1994). 22. H.C. STARY, A.B. CHANDLER, S. GLAGOV, J.R. GUYTON, W. INSULL, M.E. ROSENFELD, S.A. SCHAFFER, C.J. SCHWARTZ, W.D. WAGNER and R.W. WISSLER, Atheroscler. Thromb. 14 840-856 (1994). 23. J.H. CAMPBELL and G.R. CAMPBELL, Curr. Opin. Lipid. 5 323-330 (1994) 24. S.M. HAFFNER, L. MYKKANEN, R.A. VALDEZ and M.P. STERN, Arterioscler. Thromb. 14 1430-1437 (1994). 25. E. SHAFRIR, Diabetes Base1 Karger 12 165-191 (1993). 26. F. LACOUR, J. ESPINAL, 0. ARNAUD and J. DUHAULT, Life Sci. 53 1525-1529 (1993). 27. J.C. MAZIERE, C. MAZIERE, M. AUCLAIR, L. MORA and 0. ARNAUD, Eur. J. Clin. Pharmacol. 41 339-344 (1991). 28. T. ORSIERE, M. CHAUVET, M. DELL’AMICO, H. LAFONT and M. BOURDEAUX, Chem. Biol. Interact. 97 297-306 (1995). 11.