Inhibition of cholesteryl ester transfer protein activity in hamsters alters HDL lipid composition

ATHEROSCLEROSIS

Atherosclerosis 110 (1994) 101-109

Inhibition of cholesteryl ester transfer protein activity in hamsters alters HDL lipid composition B.J. Gaynor, Department

Tom Sand, Ronald

of Cardiovascular

W. Clark, Robert James B. Moberly*

and Metabolic Diseases,

J. Aiello, Mark J. Bamberger,

Central Research Division, Pfizer, Inc., Groton, CT 06340, USA

(Received 8 March 1994; revision received 22 April 1994; accepted 25 April 1994)

Abstract

We investigated the role of cholesteryl ester transfer protein (CETP) in hamsters by using a monoclonal antibody (MAb) that inhibited hamster CETP activity. MAbs were prepared against partially purified human CETP and screened for inhibition of ‘H-cholesteryl oleate (CE) transfer from LDL to HDL in the presence of human plasma bottom fraction (d > 1.21 g/ml). Antibody lC4 inhibited CE transfer activity in both human plasma bottom fraction (IC,, = -4 pg/ml) and in whole plasma from male Golden Syrian hamsters (IC,, = -30 &ml). Purified MA\, lC4 was injected into chow- and cholesterol-fed hamsters, and blood was collected for analysis of plasma CETP activity and HDL lipid composition. Plasma CETP activity was inhibited by 70%-80% at all times up to 24 h following injection of 500 pg MAb lC4 (- 3.7 mg/kg). The amount of antibody required for 50% inhibition at 24 h post-injection was 200 pg (- 1.5 mg/kg). Inhibition of hamster CETP activity in vivo increased hamster HDL cholesterol by 33% (P < O.OOOl),increased HDL-CE by 31% (P < 0.0001) and decreased HDL-triglyceride by 42% (P < 0.0001) (n = 36) as determined following isolation of HDL by ultracentrifugation. An increase in HDL cholesterol and a redistribution of cholesterol to a larger HDL particle were also observed following fast protein liquid chromatography (FPLC) gel filtration of plasma lipoproteins. These results demonstrate that inhibition of CETP activity in hamsters alters HDL lipid composition and particle size and further demonstrate the utility of hamsters as a model for CETP inhibition. Keywords:

HDL cholesterol; Cholesteryl ester transfer; CETP; Lipoprotein; Atherosclerosis; Golden Syrian hamster

1. Introduction Cholesteryl ester transfer protein facilitates the transfer of neutral lipids

(CETP) between

* Corresponding author, Baxter Healthcare Corporation, Renal Division Research MPR-DI, 1620 Waukegan Road, McGaw Park, IL, 60085-6730, USA. Tel.: 708 473 6457; Fax: 708 473 6923. 0021-9150/94/$07.00 0 1994 Elsevier Science Ireland SSDI 002 I-91 50(94)05276-O

plasma lipoproteins. When incubated with plasma lipoproteins in vitro, CETP mediates the transfer of cholesteryl ester (CE) from HDL to triglyceriderich lipoproteins such as VLDL, and the reciprocal exchange of triglyceride (TG) from VLDL into HDL [1,2]. Studies of CETP deficiency in man [3-51, inhibition of CETP activity in rabbits [6,7], and introduction of human CETP into rats [8,9]

Ltd. All rights reserved

102

B.J. Gaynor et al. /Atherosclerosis

and mice [ 10,l l] suggest that CETP plays a major role in determining plasma lipoprotein composition, particularly the CE and TG composition of HDL. Hamsters are a valuable model for human lipid metabolism due to appreciable LDL cholesterol and similar rates of sterol synthesis [ 121. Hamsters also possess measurable CETP activity, which increases in response to dietary cholesterol [13,14]. We have isolated a monoclonal antibody, MAb lC4, that inhibits hamster CETP activity in vitro. When administered intravenously to hamsters, MAb lC4 inhibits plasma CETP activity by up to 80%. We investigated the role of CETP inhibition on hamster lipoprotein composition, particularly that of HDL. Our studies indicate that inhibition of CETP activity in hamsters alters HDL lipid composition and increases HDL particle size. These studies support the use of hamsters as an animal model for pharmacological inhibition of plasma CETP activity. 2. Materials and methods 2.1. Hamsters and diets Male Golden Syrian hamsters (Charles River Laboratories) were maintained in colony cages (41cage) on a 12 h light/dark cycle. Hamsters were allowed free access to water and powdered Agway Prolab chow or chow mixed with 1.5% cholesterol and 0.25% cholic acid (w/w). Cholesterol feeding was started 5 days prior to antibody injections and continued until sacrifice. For studies to determine dietary effects on plasma CETP activity and lipoproteins, hamsters were anesthetized by i.p. injection of pentobarbitol (- 110 mg/kg) and exsanguinated from the abdominal aorta. For antibody injections, hamsters were anesthetized with carbon dioxide, and MAb lC4 (lo-500 pg in 0.2 ml saline) or saline (controls) was administered intravenously. Injections were made into an exposed jugular vein using a syringe fitted with a 28 g needle to minimize blood loss. The incision was closed with wound clips, and animals were permitted to recover. After the specified duration, hamsters were anesthetized with pentobarbitol and exsanguinated. Hamsters weighed between 1lo-160 g. All animal experiments were perform-

110 (1994) 101-109

ed according to standards established by the Animal Welfare Act. 2.2. Sample collection and analysis Blood from anesthetized hamsters was collected into tubes containing 40 ~1 of 0.4 M EDTA/4% NaN3 and held on ice until separation. Plasma was obtained by low-speed centrifugation at 4°C. HDL (d = 1.055 to 1.21 g/ml) was isolated at 4°C by preparative ultracentrifugation at 40 000 rev./min for 36 h in a Beckman 50.4 Ti rotor. Solutions of NaBr were used for density adjustments, and densities were checked by refractometry. Total plasma and lipoprotein cholesterol (total and free) and triglyceride concentrations were determined by enzymatic assay (Wako Biochemicals, Dallas, TX; Boehringer-Mannheim, Indianapolis, IN). Esteritied cholesterol was calculated as the difference between free and total cholesterol mass. Assays were standardized with normal control serum (Boehringer-Mannheim); standards were 102% * 2% of the reported values for cholesterol and 97% f 1% for triglyceride. 2.3. Separation of plasma lipoproteins by FPLC Separation of plasma lipoproteins was performed using fast protein liquid chromatography (FPLC, Pharmacia) gel filtration as previously described [15]. Plasma (0.5 ml) was injected onto two Superose 6 columns connected in series. Lipoprotein fractions (0.5 ml) were eluted with saline (pH 8.2) containing 1 mM EDTA and stored at 4°C until assayed for cholesterol content as described above. Cholesterol contents of VLDL, LDL and HDL were calculated by summation of cholesterol values in their respective peaks. Recovery of cholesterol following FPLC gel filtration was 95%-100% that of total cholesterol measured in aliquots before separation. 2.4. Purification of CETP CETP was partially purified from human plasma using a three-step procedure employing hydrophobic interaction and both cation and anion exchange. Human plasma bottom fraction (d > 1.21 g/ml) was loaded onto a butyl Toyopearl650 M column in 10 mM Tris plus 500 mM NaCl, pH 7.4, and CETP was eluted with a linear gradient

B.J. Gaynoret al. /Atherosclerosis110 (1994) 101-109

beginning with 50 mM Tris and ending with H20 [ 161. Fractions containing CETP activity, determined as 3H-CE transfer from LDL to HDL, were pooled, dialyzed against 50 mM acetate, pH 4.5, and loaded onto a CM-Sepharose column. CETP was then eluted using a O-l .O M linear NaCl gradient in acetate buffer. The pooled CETP fraction was next dialyzed against 10 mM Tris, pH 7.4, applied to a Mono-Q column and eluted with 0- 1.0 M NaCl. The three purification steps successively increased the specific activity x 175, x 10 and x 12, respectively, yielding a predicted total -21 OOO-foldincrease in specific activity and a purity of 60%-70%. 2.5. Preparation of labeled and unlabeled lipoproteins Lipoprotein substrates were prepared according to the method described by Morton and Zilversmit [ 171.Blood was collected from fasted donors using EDTA as anticoagulant. Plasma was adjusted to 3.5 mM with N-ethylmaleimide, an inhibitor of lecithin cholesterol acyltransferase. Radiolabeled lipoproteins were produced by incubating (37°C 18 h) 150 ml plasma with 3.0 ml of a liposome solution consisting of egg phosphatidyl choline, cholesteryl oleate (CE), triolein and BHT in a ratio of 82:12:5.6:0.4 mol %, respectively, containing 4.0 mCi of [3H]-CE (250 nmol). Lipoprotein subclasses from both normal and radiolabeled human plasma were obtained by sequential ultracentrifugation following density adjustments with potassium bromide. LDL and HDL were isolated from the density ranges 1.019-1.063 g/ml and 1. lo- 1.21 g/ml, respectively. Isolated lipoproteins were dialyzed against 100 mM sodium phosphate (pH 7.4) containing 1 mM EDTA and 0.02% sodium azide. 2.6. Preparation of monoclonal antibodies Balb/c mice were immunized with - 10 pg partially purified human CETP in complete Freund’s adjuvant and boosted after 1 month using CETP in incomplete adjuvant. Spleens were removed and spleen lymphocytes were fused with mouse myeloma cells (SP2/0) using PEG 1000 as previously described [ 181. Hybridoma supematants (50 ~1 aliquots) were screened after 2 weeks

103

for inhibition of [ 3H]-cholesteryl oleate transfer from LDL to HDL using bottom fraction (d > 1.21 g/ml) from human plasma as the source of transfer activity. Hybridoma supematant (50 ~1) was combined with [3H]-CE labeled LDL (3.0 nmol CE), unlabeled HDL (1.9 nmol CE) and 150 rg bottom fraction in 100 mM sodium phosphate (pH 7.4) (final volume 0.15 ml) and incubated for 18 h at 37°C. Lipoproteins were precipitated by addition of 160 mM phosphate buffer (125 ~1) and 120 mM manganese chloride (25 ~1) followed by centrifugation for 20 min. An aliquot of the supernatant (200 ~1) was added to scintillation fluid and counted. Inhibition of transfer was detected by a decrease in counts compared to control tubes containing unused hybridoma medium. Confirmed positive hybridomas were cloned by limiting dilution and stored under liquid Nz or injected i.p. into pristane-primed Balb/c mice for ascites production. MAb lC4 was purified by Protein A Superose (Pharmacia) affinity chromatography from delipidated mouse ascites fluid. Purified 1C4 was dialyzed against borate buffer (50 mM boric acid, 150 mM NaCl, pH 7.4) and stored at 4°C. Prior to injection into hamsters, MAb lC4 was dialyzed against PBS (pH 7.4). 2.7. Determination of CETP activity in human and hamster plasma

In vitro plasma CETP activity was determined by incubating human LDL containing 3Hcholesteryl ester (4 nmol CE containing 40 000 dpm) for 18 h with 2 nmol unlabeled human HDL and either human or hamster lipoprotein-free bottom fraction (d > 1.21 g/ml) in a total volume of 170 ~1. Radiolabeled LDL was precipitated by the addition of 136 mM sodium phosphate buffer, pH 7.4 (125 ~1) and 85 mM manganese chloride (25 ~1). Percent transfer of counts to the supernatant was determined by liquid scintillation. In vivo CETP activity in whole plasma was measured by incubating (4 h at 37°C) 150 ~1 plasma from control or MAb lC4-treated hamsters in the presence of 3H-CE-labeled HDL (3 nmol CE) and 2 mM 5,5’-dithio-bis(2-nitrobenzoic acid) (DTNB). After precipitation of the VLDL and LDL by addition of 248 mM phosphate buffer (125 ~1) and 310 mM manganese chloride (25 pl), aliquots of

lo4

B. J. Gaynor et al. /Atherosclerosis

101-109

3. Results

supernatant were counted as above. The effect of dietary modification on hamster CETP activity was assessed with 2-4 ~1 hamster plasma (CETP source) incubated for 18 h at 37°C with 3H-CElabeled HDL (10 fig cholesterol) and an excess of unlabeled LDL (250 pg cholesterol). After precipitation of the VLDL and LDL by addition of 200 mM phosphate buffer (125 ~1) and 320 mM manganese chloride (25 kl), supernatant aliquots were counted as above.

3.1. Cholesterol feeding stimulates hamster CETP activity Since endogenous hamster CETP activity is low relative to that of rabbits and man, cholesterol feeding was undertaken to stimulate hamster CETP activity. Addition of 1.5% cholesterol to the diet for 5 days increased plasma CETP activity by -50% (32% f 2% vs. 21% f 1% transfer, P < 0.01, n = 4) (Fig. 1). Cholesterol feeding also resulted in an increase in cholesterol mass in LDL (68.2 f 7.7 vs. 33.4 f 2.8 mg/dl, P < 0.01) and VLDL (59.9 f 16.2 vs. 9.9 f 2.3 mg/dl, P < 0.01). HDL cholesterol was not affected by cholesterol feeding (60.2 f 5.3 mg/dl for cholesterol-fed vs. 65.8 f 8.6 mg/dl for chow-fed). CETP activity

2.8. Statistics Statistical analysis of the data was performed using an RS/l (BBN, Cambridge, MA) procedure based on Student’s t-test for unpaired samples. Results are reported as means f 1 standard deviation, except as noted in the Fig. 4 legend.

80

110 (1994)

I

I ClFrP Activity: % Transfer = 2 1 +l Cholesterol (mg/dl) 9.9 + 2.3 VLDL 33.4 f 2.8 LDL 65.8 k 8.6 HDL 109.0 +l 1.6 Total

CETP Activity: % Transfer = 32f2 Cholesterol (mg/dl) 59.9 +16.2 VLDL 68.2 f 7.7 LDL 60.2 + 5.3 HDL 188.0 k23.0 10

20

30

40

50

Fraction Number Fig. 1. Effect of dietary cholesterol on hamster lipoprotein cholesterol distribution and CETP activity. Hamsters were fed a chow diet, with or without the addition of 1.5% cholesterol, for 5 days. Lipoprotein cholesterol was determined in fractions separated by FPLC gel filtration. VLDL, LDL and HDL cholesterol were determined by summation of the cholesterol content in fractions l-14 (VLDL), 15-30 (LDL) and 31-42 (HDL). Results are plotted for 4 individual hamsters to portray the relative consistency of lipoprotein profiles obtained on a given diet. CETP activity was determined as described in Materials and methods. A line has been drawn at the abscissa for 50 cg cholesterol to denote the difference in ordinate scales for the two treatments. Results are provided as mean f SD. (n = 4). All values of the two treatments were significantly different (P < 0.01) except for HDL cholesterol (NS). Chow, Agway Prolab chow; 1.5% Chol, Agway chow mixed with 1.5% cholesterol and 0.25% cholic acid (w/w).

B. J. Gaynor et al. /Atherosclerosis

1 IO (1994)

105

101-109

was not further stimulated by the addition of saturated fat (15% coconut oil) or extended feeding for 4 weeks, although these dietary manipulations resulted in increased hypercholesterolemia (data not shown). 3.2. MA6 1 C4 inhibits CETP activity in human and 1

hamster plasma

MAb lC4 in hybridoma supernatant or purified from mouse ascites fluid inhibited cholesteryl ester transfer activity in both human and hamster plasma (Fig. 2). MAb lC4 inhibited human CETP activity with an ICsO of -4 &ml (- 2 nM) in an assay using lipoprotein-free human plasma (density > 1.21 g/ml) as the CETP source. Endogenous CETP activity in whole hamster plasma was inhibited with an I& of -20-30 &ml. Control transfer in the two assays was comparable. High

0

ll,‘).I.I.I.I. 2 4

6

8

0 ‘I’I.‘I.I“,0

10

Antibody

20

30

10

40

12

50

0

.

I

.

100

I

200

Antibody

I

300

.

I

1

400

I

.

500

I

600

1C4 (pg)

Fig. 3. Dose-dependent inhibition of CETP activity in vivo by MAb 1C4. Cholesterol-fed hamsters were injected with purified MAb lC4 at doses of 20, 100, 200, 300, 400 and 500 @g/hamster. CETP activity was determined in individual plasma samples obtained 24 h after injection. Control 3Hcholesteryl oleate transfer from HDL was 41% f 8%. Hamsters weighed 125- 146 g (mean 134 f 8 g). Values are expressed as mean f S.D., n = 4.

concentrations of MAb lC4 were unable to completely inhibit CETP activity (maximum inhibition - 80%) in either human plasma bottom fraction or hamster plasma. Maximal CETP inhibition in human bottom fraction and hamster plasma was obtained with - 20 pg/rnl and - 100 &ml MAb lC4, respectively.

a 0

o!

14

60

1 C4 (pg/ml)

Fig. 2. Inhibition of human (a) and hamster (b) CETP activity by monoclonal antibody lC4. MAb 1C4 was added to tubes containing isolated human HDL, radiolabeled LDL and either human or hamster bottom fraction (d > I .21 @ml). CETP activity was determined as described in Materials and methods. Error bars are f SD. and, when absent, are smaller than the data point.

3.3. MAb lC4 inhibits CETP activity hamster HDL composition in vivo

and alters

Intravenous injection of a single bolus of MAb lC4 (3.7 f 0.2 mg/kg) into hamsters caused a rapid and prolonged inhibition of plasma CETP activity. The dose of MAb lC4 required for 50% inhibition of CETP activity in cholesterol-fed hamsters (ED,,) was - 200 pg, or 1.5 mg/kg (Fig. 3). Maximal inhibition (-70%) was achieved following injection of -300 pg of MAb lC4. CETP inhibition was prolonged, continuing for at least 24 h after administration of MAb lC4 (Table 1). CETP activities at 0.5, 3, 6, and 24 h post-dose were inhibited by 79%, 83%, 71%, and 76% respectively, compared to control transfer in saline-injected hamsters. Injection of MAb lC4 into hamsters altered HDL lipid composition determined following isolation of HDL by ultracentrifugation. The effects

B.J. Gaynor et al. /Atherosclerosis

106

Table I Time course of CETP inhibition in hamsters Time (h)

0.5 3 6 24

CETP Activity (% transfer) Control

Antibody lC4

33.5 (14.6) 43.4 (13.8) 40.0 (14.3) 35.6 (5.4)

6.9 (1.2)’ 7.2 (3.6)* 11.7 (2.9)* 8.7 (2.8)*

101-109

Table 3 CETP inhibition by MAb 1C4 alters HDL composition in chow-fed hamsters n

9 II 12 12

Cholesterol-fed hamsters were injected with 500 pg CETP (- 3.7 mg/kg) or saline. Animals were sacrificed and plasma CETP activity was determined at 0.5, 3, 6, and 24 h after injection. The results are expressed as mean % transfer f SD. (in parentheses). *P < 0.001.

in of MAb lC4 on HDL lipid composition cholesterol- and chow-fed hamsters are presented in Tables 2 and 3, respectively. In cholesterol-fed hamsters, HDL cholesterol increased by 44% (P < O.OOS),HDL-CE increased by 38% (P < 0.005) and HDL-triglyceride decreased by 32% (P C 0.05). As a result, the ratio of HDL cholesterol to HDL triglyceride more than doubled, from 7.2 to 14.9 (Table 2). Similar results were obtained in three separate experiments in cholesterol-fed hamsters, for which the mean inhibition of CETP

Table 2 CETP inhibition by MAb IC4 alters HDL composition in cholesterol-fed hamsters

Control (n = 12) MAb 1C4 (n = 12)

II0 (1994)

HDL-C (mgidt)

HDL-CE (mg/dl)

HDL-TG (mg/dU

Ratio CE/TG

57.1 (14.9) 82.1** (9.7)

49.3 (11.3) 68.0** (8.5)

7.6 (2.7) 5.2* (2.2)

7.2 (2.5) 14.9** (6.1)

Cholesterol-fed hamsters were injected with 500 pg MAb lC4 (3.8 f 0.04 mg/kg) or saline. Animals were sacrificed after 24 h and HDL (d > 1.21 g/ml) was isolated by ultracentrifugation as described in Materials and methods. HDL total cholesterol (HDL-C), unesterified cholesterol and triglyceride (HDL-TG) were determined by enzymatic assay. HDL cholesteryl ester (HDL-CE) and the ratio of HDL-CE to HDL-TG were calculated. CETP activity was inhibited 79% (8% * 2%** vs. 40% * 9% transfer). Results are expressed as mean & SD. *P < 0.02, ?? *P < 0.005.

Control (n= 11) MAb 1C4 (n = 12)

HDL-C (mg/dl)

HDL-CE (mg/dt)

HDL-TG (mg/df)

Ratio CE/TG

71.2 (11.0) 84.0* (12.7)

58.2 (10.5) 69.8* (10.7)

;;:6) 1.9** (0.3)

22.0 (6.1) 38.6** (9.7)

Chow-fed hamsters were injected with 500 pg 1C4 (3.9 f 0.2

mgkg)and HDL lipid composition determined as described in the legend for Table 2. CETP activity was inhibited 68% (6% f 3%” vs 20% f 3% transfer).

activity was 76% i 3% (P < O.OOOl),mean HDL cholesterol increased by 33% f 9% (P < O.OOOl), mean HDL-CE increased by 31% f 8% (P c O.OOOl),and mean HDL triglyceride decreased by 42% f 9% (P c 0.0001) (n = 36 hamsters/group). In these experiments, total plasma cholesterol also increased significantly by 18% f 8% (P < 0.05) following injection of MAb lC4, whereas plasma triglyceride levels were unchanged from controls. Qualitatively similar changes were achieved when MA\, lC4 was injected into chow-fed hamsters (Table 3). HDL cholesterol and HDL-CE increased by 18% and 20% (P < O.OS), respectively, whereas HDL triglyceride decreased by 32% (P < O.OOS),resulting in a 75% increase in the HDL cholesterol/triglyceride ratio.

Table 4 The effect of MAb 1C4 on hamster plasma lipoproteins Lipoprotein cholesterol (mg/dl)

Control (n = 14) MAb lC4 (n = 15)

VLDL

LDL

HDL

TOTAL

163.8 (70.3) 159.2 (72.8)

106.4 (32.2) 124.2 (32.5)

76.2 (9.7) 104.0* (17.8)

346.4 (89.8) 387.2 (81.3)

Cholesterol-fed hamsters were injected with MAb lC4 or saline and lipoproteins were separated by FPLC as described in the legend for Fig. 4. The total cholesterol mass in each lipoprotein peak was determined by summation of the fractions in the respective peak. Values are expressed as mean f S.D. of the samples profiled in Fig. 4.

B.J. Gaynor et al. /Atherosclerosis

107

110 (1994) 101-109

300 250 200 150loosoni-0

10

20

30

40

Fraction Number Fig. 4. The effect of MAb 1C4 on the distribution of cholesterol in lipoproteins from hamster plasma. Cholesterol-fed hamsters were injected with 500 cg MAb 1C4 or saline and sacrificed after 24 h. Plasma lipoproteins were separated by FPLC as described in Materials and methods. Cholesterol values are plotted as mean f standard error of the mean (n = 15 for MAb 1C4 and n = 14 for saline injected). Total cholesterol mass in each lipoprotein peak are presented in Table 4. Plasma CETP activity was inhibited 76% (P < 0.02) in this experiment.

Changes in HDL cholesterol concentration were also observed when hamster lipoproteins were separated by gel filtration (Fig. 4 and Table 4). Total HDL cholesterol mass in the lC4-treated group was 36% higher than in controls (P < 0.01). The HDL cholesterol concentration expressed as a percentage of total plasma cholesterol was significantly greater for lC4 treated hamsters than controls (28% f 6% vs. 22% f 5%, P c 0.02), but LDL, VLDL and total cholesterol were unaffected. A shift toward a larger HDL particle was also observed when hamster plasma lipoproteins were separated by FPLC (Fig. 4), consistent with the formation of larger, cholesterol-enriched HDL following CETP inhibition. 4. Discussion

MAb lC4 effectively inhibits CE transfer in both human and hamster plasma. When administered intravenously to hamsters, MAb lC4 raises HDL cholesterol, increases HDL cholesteryl ester content and decreases HDL triglyceride. As a result, the HDL CE/TG ratio increases dramatically, and the HDL particle shifts to a larger size as shown by gel filtration of plasma lipoproteins. These changes in HDL composition and HDL particle size are consistent with altera-

tions in HDL of human subjects with a deficiency of plasma CETP and with changes induced by infusion of CETP-inhibitory antibodies in rabbits. In homozygous CETP-deficient humans, HDL cholesterol and HDL-CE are significantly elevated [4], and HDL particle size is larger than in control individuals [5]. In rabbits, inhibition of CETP activity by intravenous infusion of CETPneutralizing antibodies doubles HDL-CE, decreases HDL-TG, and increases HDL size from a radius of 4.7 to 5.4 nm [6,7]. Infusion of partially purified CETP into rats, a species that lacks endogenous CETP activity, results in the loss of large, cholesteryl ester-rich HDLi and the appearance of triglyceride in HDL [8,9]. Similarly, expression of human CETP in transgenic mice significantly reduces HDL cholesterol [lo], especially when coexpressed with human apolipoprotein A-I [l 11. Our results of CETP inhibition in hamsters are consistent with a major role for CETP in determining HDL lipid composition and HDL particle size in all species expressing CETP. The changes in HDL composition induced by MAb lC4 are independent of diet, but are enhanced by exogenous cholesterol. Hamsters respond rapidly to dietary cholesterol with significant increases in plasma CETP activity (50%) and LDL cholesterol (200%) within 5 days, while HDL cho-

108

B.J. Gaynor et al. /Atherosclerosis

lesterol remains unchanged. In both chow- and cholesterol-fed hamsters, MAb-IC4 comparably inhibits CETP activity (68% and 76%, respectively) and elevates total plasma cholesterol (14% and 18%, respectively), but unequally increases the ratio of HDL-CE to HDL-TG (75% and 230%, respectively). The cholesterol-induced stimulation of CETP activity we observe is comparable to that observed in hamsters fed less cholesterol for a longer period [ 13,191,but less than the 2-fold stimulation seen by others [14]. It is unlikely that inclusion of dietary cholic acid, which enhances cholesterol absorption from the intestine and indirectly increases plasma LDL cholesterol levels [20], would diminish the response of CETP to cholesterol in our study. Rather, CETP activity differences may reflect variability among different strains of hamsters, dietary responsiveness, or the amount of endogenous CETP inhibitor protein present. Cholesterolinduced changes in plasma lipoproteins and CETP activity may be independently regulated in hamsters, as has been demonstrated in transgenic mice [21]. The reason for the failure to achieve complete inhibition of CETP using MAb lC4 is unclear. We rarely observed inhibition of cholesteryl ester transfer greater than 80% in assays using whole hamster plasma or human bottom fraction. A possible explanation for this lack of complete inhibition is that a portion of radiolabeled CE in the donor particles may exist in a readily exchangeable pool, perhaps on the surface of the lipoprotein. Alternatively, a fraction of CETP in hamster plasma may be inaccessible to MAb 1C4, even when the antibody is present at high concentrations. Whatever the reason, the failure to achieve complete inhibition of CETP does not detract from the conclusions drawn from in vivo studies, in which the majority of CETP activity was inhibited. The inverse correlation between plasma CETP activity and HDL cholesterol and the positive correlation between CETP expression and atherosclerosis susceptibility among animal species suggest that elevated levels of CETP may be atherogenic. In fact, recent studies demonstrate a dramatic increase in atherosclerosis in transgenic mice expressing simian CETP [22]. The demon-

II0

(1994)

101-109

stration that suppression of CETP activity in hamsters produces an increase in HDL cholesterol supports the hypothesis that inhibition of CETP may offer a therapeutic benefit. The hamster, which mimics the response of humans to diet and develops atherosclerotic lesions, provides an attractive model in which to further study this hypothesis. References 1 Hesler, C.B., Swenson, T.L. and Tall, A.R., Purification and characterization of a human plasma cholesteryl ester transfer protein, J. Biol. Chem., 262 (1987) 2275. 2 Quig, D.W. and Zilversmit, D.B., Plasma lipid transfer activities, Ann. Rev. Nutr., IO (1990) 169. 3 Inazu, A., Brown, M.L., Hesler, C.B. et al., Increased high-density lipoprotein levels caused by a common cholesteryl-ester transfer protein gene mutation, N. Engl. J. Med., 323 (1990) 1234. 4 Koizumi, J., Inazu, A., Yagi, K. et al., Serum lipoprotein lipid concentration and composition in homozygous and heterozygous patients with cholesteryl ester transfer protein deficiency, Atherosclerosis, 90 (1991) 189. 5 Yamashita, S., Hui, D.Y., Wetterau, J.R. et al., Characterization of plasma lipoproteins in patients heterozygous for human plasma cholesteryl ester transfer protein (CETP) deficiency: plasma CETP regulates highdensity lipoprotein concentration and composition, Metabolism, 40 (1991) 756. 6 Whitlock, M.E., Swenson, T.L., Ramakrishnan, R., Leonard, M.T., Marcel, Y.L., Milne, R.W. and Tall, A.R., Monoclonal antibody inhibition of cholesteryl ester transfer protein activity in the rabbit, J. Clin. Invest., 84 (1989) 129. 7 Abbey, M. and Calvert, G.D., Effects of blocking plasma lipid transfer protein activity in the rabbit, B&him. Biophys. Acta, 1003 (1989) 20. 8 Quig, D.W. and Zilversmit, D.B., Disappearance and effects of exogenous lipid transfer activity in rats, B&him. Biophys. Acta, 879 (1986) 171. Ha, Y.C., Chang, L.B.F. and Barter, P.J., Effects of injecting exogenous lipid transfer protein into rats, Biochim. Biophys. Acta, 833 (1985) 203. Agellon, L.B., Walsh, A., Hayek, T. et al., Reduced high density lipoprotein cholesterol in human cholesteryl ester transfer protein transgenic mice, J. Biol. Chem., 266 (1991) 10796. Hayek, T., Chajek-Shaul, T., Walsh, A., Agellon, L.B., Moulin, P., Tall, A.R. and Breslow, J.L., An interaction between the human cholesteryl ester transfer protein (CETP) and apolipoprotein A-I genes in transgenic mice results in a profound CETP-mediated depression of high density lipoprotein cholesterol levels, J. Chn. Invest., 90 (1992) 505.

B.J. Gaynor et al. iAtherosclerosis

12

13

14

15

16

17

Spady, D.K. and Dietschy, J.M., Interaction of dietary cholesterol and triglyceride in the regulation of hepatic low density lipoprotein transport in the hamster, J. Clin. Invest., 81 (1988) 300. Quig, D.W., Arbeeny, C.M. and Zilversmit, D.B., Effects of hyperlipidemias in hamsters on lipid transfer protein activity and unidirectional cholesteryl ester transfer in plasma, Biochim. Biophys. Acta, 1083 (1991) 257. Stein, Y., Dabach, Y., Hollander G. and Stein, O., Cholesteryl ester transfer activity in hamster plasma: increase by fat and cholesterol rich diets, Biochim. Biophys. Acta, 1042 (1990) 138. Cole, T.G., Kitchens, R., Daugherty, A. and Schonfeld, G., An improved method for separation of triglyceriderich lipoproteins by FPLC, Pharmacia FPLC Biocommunique, 4 (I 988) 4. Ohnishi, T., Yokoyama, S. and Yamamoto, A., Rapid purification of human plasma lipid transfer proteins, J. Lipid Res., 31 (1990) 397. Morton, R.E. and Zilversmit, D.B., Inter-relationship of lipids transferred by the lipid-transfer protein isolated

18

19

20

21

22

110 (19941

101-109

109

from human lipoprotein deficient plasma, J. Biol. Chem.. 258 (1983) 11751. Kiihler, G. and Milstein, C., Continuous cultures of fused cells secreting antibody of predetermined specificity, Nature, 256 (1975) 495. Jiang, X.C., Moulin, P., Quinet, E. et al.. Mammalian adipose tissue and muscle are major sources of lipid transfer protein mRNA, J. Biol. Chem., 266 (1991) 4631. Spady, D.K., Stange, E.F., Bilhartz, L.E. and Dietschy. J.M., Bile acids regulate hepatic low density lipoprotein receptor activity in the hamster by altering cholesterol flux across the liver, Proc. Natl. Acad. Sci. 83 (1986) 1916. Jiang, X.C., Agellon, L.B., Walsh, A., Breslow, J.L. and Tall, A.R., Dietary cholesterol increases transcription of the human cholesteryl ester transfer protein gene in transgenie mice, J. Clin. Invest., 90 (1992) 1290. Marotti, K.R., Castle, C.K., Boyle, T.P., Lin, A.H., Murray, R.W. and Melchior, G.W., Severe atherosclerosis in transgenic mice expressing simian cholesteryl ester transfer protein, Nature, 364 (1993) 73.