Effects of inhibiting cholesterol absorption and synthesis on cholesterol and lipoprotein metabolism in hypercholesterolemic non-insulin-dependent diabetic men Helena Gylling and Tatu A. Miettinen' Division of Internal Medicine, Department of Medicine, University of Helsinki, Haartmaninkatu 4, FIN-00290 Helsinki, Finland

Abstract Effectiveness of a simultaneous inhibition of cholesterol absorption and synthesis, caused by sitostanol ester margarine and pravastatin, was studied to control mild hypercholesterolemia in men with non-insulin-dependent diabetes mellitus (NIDDM) (n = 8). Margarine, 24 g daily, was a basal dietary treatment. Four 7-week intervention periods included margarine, sitostanol (3 g/day) ester margarine, pravastatin (40 mg/day), and sitostanol ester margarine plus pravastatin in a random order. Pravastatin lowered serum total (-32%) and LDL cholesterol (-38%)and apolipoprotein B (-39%) because of enhanced removal (+20%) and decreased production (-26%) of LDL apolipoprotein B, and reduced synthesis (-9%) and turnover (-8%)of cholesterol, which resulted in reduced biliary cholesterol secretion (-18%). Even though serum triglycerides were lowered by 28%, VLDL, IDL, and light and dense LDL became triglycerideenriched. Despite increasing cholesterol synthesis, sitostanol lowered LDL cholesterol (-14%) by inhibiting cholesterol absorption (-68%) and LDL apolipoprotein B production rate (-20%).Combination of pravastatin and sitostanol ester lowered serum total, VLDL, IDL, and LDL cholesterol and LDL apolipoprotein B by the highest rate, 35%, 50%, 35%, 44%, and 45% from the control margarine period, respectively, because of reduced apolipoprotein B transport rate (but unchanged removal), in both the total and dense LDL subfractions. HDL cholesterol and apolipoprotein A-I kinetics were unchanged. In spite of decreased absorption, cholesterol synthesis was not compensatorily increased. In conclusion, simultaneous inhibition of cholesterol absorption and synthesis lowers LDL cholesterol and apolipoprotein B by 44-45% solely through inhibition of LDL apolipoprotein B production rate in hypercholesterolemic NIDDM patients. A combination of statin to sitostanol ester margarine-resistant patients offers a safe and effective measure to normalize abnormally high cholesterol values, probably with a lowered statin dose.-Gylling, H., and T. A. Miettinen. Effects of inhibiting cholesterol absorption and synthesis on cholesterol and lipoprotein metabolism in hypercholesterolemic non-insulin-dependent diabetic men. J. Lipid Res. 1996.37: 1776-1785.

Supplementary key words pravastatin plus sitostanol treatment NIDDM apolipoprotein B kinetics apolipoprotein A-I kinetics

cholesterol precursors plant sterols

1776 Joumd of Lipid Research Volume 37, 1996

In general, inhibition of cholesterol synthesis by hydroxymethyl-glutaryl-CoA (HMG-CoA) reductase inhibitors lowers serum total and LDL cholesterol level by 25-30% (1). On the other hand, the inhibition of cholesterol absorption with sitostanol ester lowered LDL cholesterol level by 14% in a mildly hypercholesterolemic non-diabetic population (2), and by 9% in mildly hypercholesterolemic non-insulin-dependent diabetics (NIDDM) (3), whereas the combination of two inhibitors of cholesterol absorption, neomycin and sitostanol ester, lowered LDL cholesterol by 36% in the NIDDM subjects (4). Although NIDDM is frequently associated with hypertriglyceridemia, high serum low density lipoprotein (LDL) cholesterol levels were present in two-thirds of dyslipidemic NIDDM subjects (5), which is the major atherogenic lipid profile in diabetics also (6). HMG-CoA reductase inhibitors are effective cholesterol-loweringagents also in diabetics (7-1 l),but the detailed lipid-lowering mechanisms are not completely evaluated in them. The question now arises whether a more potent hypocholesterolemic effect could be achieved by a simultaneous inhibition of cholesterol absorption by sitostanol ester and synthesis by an HMG-CoA reductase inhibitor, and what are the detailed metabolic consequencies of the combination therapy. Thus, the aim of this study was to evaluate the effects of pravastatin alone and in combination with

Abbreviations: VLDL, very low density lipoprotein; IDL, intermediatedensity lipoprotein; LDL, low density lipoprotein; HDL, high density lipoprotein; NIDDM, non-insulindependent diabetes mellitus; HMG-CoA, 3-hydroxy-3-methylgluta1ylcoenzyme A, FCR, fractional catabolic rate; TR, transport rate. 'To whom correspondence should be addressed.

sitostanol ester on serum lipids and cholesterol and lipoprotein metabolism in hypercholesterolemic NIDDM men. Special attention was paid to the composition of apoB-containing lipoproteins, LDL apoB and high density lipoprotein (HDL) kinetics and cholesterol absorption and synthesis.

SUBJECTS AND METHODS Subjects

The study group consisted of eight NIDDM men with a mean age of 60.2 f 1.6 (SE) years and body mass index 26.6 f 1.1kg/m2. The primary selection criteria were as follows: serum cholesterol concentration 2 6.0 mmol/l; serum triglyceride concentration I 2.5 mmol/l; body mass index I 28.0 kg/m2; and no present insulin or hypolipidemic therapy. As cholesterol malabsorption is effective only for cholesterol, not triglycerides, the subjects were selected to have a lipid profile of primary moderate hypercholesterolemia and they did not represent dyslipidemic diabetics in general. The glycemic control was good to moderate during the metabolic studies (Table 1). The daily intakes of cholesterol and fat were 233 f 30 mg/day and 79 f 7 g/day, and they were practically unchanged during the interventions. None of the subjects had microalbuminuria, retinoor neuropathia or hepatic, thyroid, or gastrointestinal disease. Five subjects had coronary artery disease. Two patients were treated with beta-blocking agents, three with calcium channel blockers, and one patient received diuretics. Four patients were on glibenclamide and one patient was on biguanidine therapy. Neither weight, serum lipids, nor the metabolic parameters differed among the subjects with or without beta-blocking therapy or with or without glibenclamide. All subjects volunteered for the study, and the study protocol was accepted by the Ethics Committee of the Second Department of Medicine, University of Helsinki. After the run-in period of 4 weeks on baseline ad libitum diet, the four 7-week intervention periods (rapeseed oil control margarine, sitostanol ester margarine, pravastatin plus control margarine, and pravastatin plus sitostanol ester margarine) were randomly selected and blinded for the type of the margarine. Seven weeks, in several cases up to 21 weeks (e.g., sitostanol followed by pravastatin and pravastatin plus sitostanol) on reduced cholesterol levels, were considered a long enough time because even shorter periods have been considered to reach a metabolic steady state of sterol metabolism (12). The subjects were advised to replace 24 g daily of their normal dietary fat with the rapeseed oil margarine according to detailed instructions of a dietician. The margarine was distributed in 8-g buttons, and the par-

ticipants used one button on a slice ofbread at breakfast, lunch, and dinner. The sitostanol ester margarine contained 1 g of sitostanol dissolved in the 8 g of rapeseed oil margarine. Pravastatin, 40 mg daily, was taken as a single dose. The rapeseed oil margarine contained campesterol and sitosterol, 209 and 288 mg/100 g of margarine, respectively. Sitostanol ester margarine contained campesterol, campestanol, sitosterol, and sitostanol 288, 921, 1,138,and 11,400 m u 1 0 0 g of margarine, respectively. Sitosterol was hydrogenated to sitostanol (Kaukas Inc., Lappeenranta, Finland) and transesterified with rapeseed oil fatty acids and dissolved in the margarine (Raisio Inc., Raisio, Finland). At the end of the treatment periods, metabolic and kinetic studies were performed. The subjects kept a food record for 7 days, from which the dietary constituents were calculated (13). Also, they were given a capsule containing [4'4C]cholesterol, [22,23-3H]~sitosterol, and 200 mg CrzOs three times a day with their regular meals during the 7-day period. Threeday stool collections were performed at the end of the 7-day periods. LDL and HDL turn-over studies were performed at the end of each period, during which serum lipids, lipoproteins and apolipoproteins, and serum noncholesterol sterols were analyzed four times from serum samples after a 12-h fast, and the mean values of these specimen are given. Liver enzymes, creatinine phosphokinase, and parameters for the glycemic control were monitored during each intervention. Methods Serum total and free cholesterol, triglycerides, phospholipids, and apolipoproteins A-I, A-11, and B were measured automatically with the use of commercial kits (Boehringer Diagnostica, Germany, Wako Chemicals, Germany, and Orion Diagnostica, Finland). Serum lipoproteins were separated by ultracentrifugation into density classes as described in Manual of Laboratory Operations of Lipid Research Clinics Program to very low (VLDL), intermediate (IDL), LDL, and HDL (14). LDL was fractionated into subfractions by density gradient ultracentrifugation so that the light fraction consisted of densities from 1.019 to 1.036 g/ml, the dense fraction from 1.037 to 1.055 g/ml, and the very dense fraction from 1.056 to 1.063 g/ml, respectively (15). Serum cholesterol precursors, squalene, A8cholestenol, desmosterol, and lathosterol, and serum plant sterols, campesterol and sitosterol, and cholestanol, were quantified with gas-liquid chromatography on a 50-m long SE-30 capillary column (2, 16). Cholesterol absorption was measured from the 3day stool collections with the peroral doubleisotope continuous feeding method validated by Crouse and

Gylling and Miettina Inhibition of cholesterol absorption and synthesis 1777

Grundy (17). Chromic oxide was analyzed from the 3-day fecal specimen (18),and fecal sterols were determined with gas-liquid chromatography using the 50-m long SE-30 capillary column (16, 19, 20). Fecal analyses were performed from the 3-day stool collections, which is a long enough period to give reliable information on fecal elimination of cholesterol (12, 19-21). A picture of daily variation is obtained from our separate series of 21 subjects; the mean of three successive daily sterol analyses varied by 7.3 k 1.5% from the upper and lower values. Cholesterol synthesis was calculated with the sterol balance technique (19, 20). For the kinetic studies, 50 ml plasma was drawn from fasted subjects and treated with EDTA; autologous HDL and total and dense LDL were separated by serial preparative ultracentrifugations. ApoA-I was isolated from HDL as described previously (22). Dense LDL apoB and apoA-I were iodinated with lZ51,and total LDL apoB with I3'I by a modification of the iodine-monochloride method (23, 24). From 95 to 99% of the label was recovered from the apolipoproteins by 15% trichloracetic acid precipitation. Three days before injection the subjects started to take peroral potassium iodide. Approximately 1 mg of the labeled autologous total LDL apoB and apoA-I and 0.5-1 mg of dense LDL were mixed with 5% human serum albumin, filtered, and injected simultaneously. The total amount of radioactivity did not exceed 60 pCi. After the injection, blood samples of 10 ml were collected and counted for 14 days. The die-away curves were constructed in whole plasma for 1311-labeledLDL and after ultracentrifugation for 1251-labeled dense LDL and 1251-labeled HDL. Fractional catabolic rates (FCR) for total and dense LDL and apoA-I were determined using a two-pool model (25). Transport rate (TR) was calculated by multiplying FCR by the pool size. Pool size was the apolipoprotein plasma concentration multiplied by plasma volume, which was calculated to be 4.5% of body weight. ApoB from total and dense LDL and apoA-I in serum were measured twice from postinjection samples, and their mean values were used in the calculations of TRs.

Calculations Cholesterol synthesis = difference between the fecal steroids (neutral and acidic) of cholesterol origin and dietary cholesterol. Total intestinal cholesterol flux = fecal neutral sterols divided by (1 - fractional cholesterol absorption). Cholesterol transport or turnover = sum of fecal bile acids and fecal endogenous neutral sterols calculated by subtracting ( 1 - cholesterol absorption efficiency)multiplied by dietary cholesterol from fecal neutral sterols. Biliary cholesterol secretion = total intestinal flux minus dietary cholesterol. Lipid to apoB ratios were calculated from the uncorrected values obtained from every fraction, while the concentrations are corrected for volume and recovery. Serum noncholesterol sterol concentrations are standardized to serum cholesterol value and are given as lo2 x mmol/mol cholesterol and called proportions in the following. The hypothesis testing was performed with analysis of variance, two-sided Student's t-test, paired t-test, and Pearson's product-moment correlation. Logarithmic transformations were used when appropriate. A P value < 0.05 was considered statistically significant.

RESULTS Seven subjects completed the whole study; one patient dropped out in the middle of his last period because of a personal reason. Weight of the patients and glycemic control were unchanged during the interventions (Table 1). Pravastatin and sitostanol ester margarine were well tolerated without any side effects. The patients could not distinguish between the two types of margarine so the study could be completed double blinded for the margarine. The dietary intakes of campesterol and sitosterol were calculated to be increased by 20 mg/day and 220 m u d a y when switched from margarine without to that with sitostanol ester, while the

TABLE 1. Clinical characteristics and serum lipids (n = 8) Pravasmtin

Margarine Variables

Weight, kg Fasting blood glucose, mmol/l Glycated hemoglobin, % Serum cholesterol, mmol/l Serum triglycerides, mmol/l Serum phospholipids, mmol/l ~

Sitostanol Estcr

(M)

+M

75.8 f 2.7 7.7 0.7 7.0 f 0.5 6.64 0.12 2.43 f 0.17 3.56 f 0.10

81.1 5.3 8.1 0.8 7.3 It 0.6 5.94 2 0.2w 2.38 10.29 3.30 I 0 . 1 P

*

*

* *

Pravastatin +

M

80.5 f 4.8 8.2 f 0.5 8.1 f 0.3 4.53 k 0.17'." 1.74 f 0.1VJ 2.64 f O.IW,b

+M 80.5 f 4.8 8.1 f 0.6 7.1 f 0.5 4.29 f 0.16'-' 1.70 f 0.09.' 2.64 f 0.09."

~~

Mean f SE. For conversion to mg/dl, multiply cholesterol values by 38.7, triglycerides by 88.2, and phospholipids by 75.0 "Significantly different from M. 'Significantly different frnm sitostanol esler + M; ANOVA.

1778 Journal of Lipid Research Volume 37, 1996

+

Sitostanol Ester

respective calculated intake values of campestanol and sitostanol were 240 and 3000 mg/day. In the LDL subfraction kinetics there was a back flux from the dense to the light fraction varying from 5 to 18% in the different individuals but remaining very constant in one person throughout the study. The shape of the decay curves was bimodal. In addition, in one patient the die-away curves of the light and very dense fractions could be constructed at baseline because the fractions were separated and counted from every postinjection sample, The FCR for light LDL apoB was faster than for dense LDL apoB, and slowest for the very dense fraction. Less than 5% of the labeled dense apoB was recovered in the very dense subfraction.

7.0

-

6.0

--

5.0

--

-4.0

Sitostanol ester Sitostanol ester margarine lowered serum total, VLDL, and LDL cholesterol by 11%, IO%, and 14%, respectively, with no effect on HDL cholesterol and serum triglycerides (Table 1, Fig. 1). The treatment reduced total cholesterol below 5.2 mmol/l in three out of eight patients, and LDL cholesterol below 3.5 mmol/l in four subjects. In VLDL and IDL, the triglyceride/cholesterol ratios were significantly increased (Table 2). LDL apoB was reduced only by 9% due to decreased TR for LDL apoB (Table 3), so that the cholesterol/apoB ratio was significantly decreased (Table 2). Cholesterol and apoB concentrations were diminished mainly in the dense subfraction so that the relative proportion of

CHOLESTEROL

--

\

E '3.0

--

2.0

--

1.0

--

-

i

T

L

HDL-C

*+

-.+

VLDL-C

.+

W

;+ IDL-C

TRIGLYCERIDES

L-----L *+ TOTALTG

*+

z

VLDL-TG

Fig. 1. Serum total and lipoprotein cholesterol and serum total and VLDL triglyceride concentrations during treatment with rapeseed oil margarine (M) without and with sitostanol ester and pravastatin plus M without and with sitostanol ester in noninsulin-dependent diabetic men (n = 8). Values are mean f SE. For conversion to mg/dl, multiply cholesterol values by 38.7 and triglycerides by 88.2. * P < 0.05 from M; +P< 0.05 from sitostanol ester + M; "P < 0.05 from pravastatin; ANOVA.

Cylling and Miettina Inhibition of cholesterol absorption and synthesis 1779

TABLE 2. The composition of VLDL, IDL, and LDL Variables

VLDL Cholesterol, % Triglycerides, % Phospholipids, % Triglyceride/cholesterol IDL Cholesterol, % Triglycerides, % Phospholipids, % Triglyceride/cholesterol LDL Cholesterol, % Triglycerides, % Phospholipids, % Triglyceride/cholesterol Cholesterol/apoB Triglyceride/apoB Light LDL Triglyceride/cholesterol Cholesterol/apoB Triglyceride/apoB Dense LDL Triglyceride/cholesterol Cholesterol/apoB Triglyceride/apoB Very dense LDL Triglyceride/cholesterol Cholesterol/apoB Triglyceride/apoB

Margarine (M) -

.

Sitostanol Ester + M

Pravastatin +M .__

27.3 f 0.7 53.0 f 0.5 19.7 f 0.2 1.95 f 0.06

26.1 f 1.1 54.2 f 1.4 19.6 f 0.4 2.11 f 0.15a

22.8 f 0.6a,b 57.7 f 0.6n,b 19.5 f 0.3 2.54 f 0.09a,6

23.5 0.8' 56.4 0 . e 20.2 f 0.3 2.42 k 0 . 1

54.5 f 3.0 19.2 f 1.5 26.3 f 4.1 0.35 f 0.02

52.9 f 0.4 23.6 f 0.50 23.5 f 0.3 0.45 f 0.014

49.5 f 0.7b 27.4 f 0.9a,b 23.1 f 0.5 0.56 f 0.0246

49.1 f 1 . 3 1 28.3 f I.?'.* 22.6 f 0.6 0.58 f 0.0@,*

70.7 f 0.6 5.4 k 0.4 23.8 f 0.4 0.08 f 0.01 1.94 f 0.05 0.34 k 0.02

69.9 f 0.5 6.0 f 0.5 24.1 f 0.1 0.09 f 0.01 1.86 f 0.020 0.36 f 0.03

69.6 k 0.5 6.5 f 0.4 23.9 f 0.2 0.09 f 0.01 1.99 f 0.02* 0.43 f 0.03'

69.1 f 0.6 6.9 f 0.fP 24.0 f 0.2 0.10 f 0.01 1.99 k 0.05b 0.45 k O.O@

0.14 f 0.01 2.08 f 0.19 0.73 f 0.05

0.44 f 0.13" 2.13f0.10 1.50 f 0.460

0.22 f 0.03' 2.25 f 0.10 1.10 f 0.110

0.25 f 0.04 2.12 f 0.06 3.09 f 1.74

0.09 f 0.01 2.25 f 0.09 0.45 f 0.04

0.17 f 0.05 2.26 f 0.07 0.60 f 0.09

0.13 f 0.03 2.36 f 0.07 0.69 f 0.16

0.15 f 0.03 2.09 f 0.09,b.' 0.72 f 0.lP

0.35 f 0.1 I d 2.75 f 0.27 1.74 f 0.47d

0.54 f 0.22 2.27 f 0.10 1.92 f 0.3sd

0.41 f 0.14 2.18 f 0.07 2.03 f 0.64

0.41 f 0.06d 2.27 f 0.09 2.12 f 0.30"

* *

Mean k SE. "Significantly different from M. *Significantlydifferent from sitostanol ester + M "Significantly different from pravastatin + M. dSignificantlydifferent from dense fraction.

triglyceride to apoB was increased. The compositions of the light and very dense subfractions were practically unchanged. FCR for dense LDL apoB tended to be higher than that for total LDL apoB suggesting that the removal of the light LDL apoB was the lowest. Sitostanol ester had no effect on the metabolic parameters related to apoA-I concentration. Serum plant sterol proportions campesterol and sitosterol, indicators of cholesterol absorption (26), were diminished by 46% and 43% (Table 4),and cholesterol absorption efficiency was diminished by 68% (Table 5). Of the cholesterol precursor sterols, indicators of cholesterol synthesis (26), only desmosterol increased significantly by 11%. Bile acid synthesis was unchanged, while fecal neutral sterol excretion, cholesterol synthesis, and cholesterol turnover were significantly increased by 29-39% (Table 5 ) . Biliary secretion of cholesterol was significantly increased by 11%.

Pravastatin Comparison to margarine period. Pravastatin significantly reduced serum total, VLDL, IDL, and LDL cho-

1780 Journal of Lipid Research Volume 37, 1996

lesterol by 34-44% compared with the control margarine period (Table 1, Fig. l), and serum total and lipoprotein triglycerides by 8 (1DL)-31% (VLDL and LDL). In VLDL and IDL, the relative proportions of cholesterol were decreased and those of triglycerides increased so that the triglyceride/cholesterol ratios were significantly increased (Table 2). LDL cholesterol/apoB ratio was unchanged from the control margarine period. Cholesterol and apoB contents were reduced in both the light and dense LDL fractions, while the relative triglyceride content was significantly increased especially in the light LDL fraction. FCR for total LDL apoB was increased and that of TR decreased (Table 3). However, in the dense fraction, both FCR and especially TR for LDL apoB were significantlydiminished, suggesting that the catabolism of LDL apoB was increased in the light fraction. The composition of the very dense fraction was practically unaffected. The composition and kinetics of HDL apoA-I were unchanged (Table 6). Pravastatin significantly reduced the precursor sterol proportions by 27-53% (Table 4), fecal neutral sterol excretion by 20%, and cholesterol synthesis and turn-

TABLE 3. LDL apoB kinetics Pravastatin + Variables

Total LDL Cholesterol, mg/dl ApoB, m u d l FCR, pools/d TR,mg/kg/ Light LDL Cholesterol, mg/dl ApoB, mg/dl Dense LDL Cholesterol, mg/dl ApoB, mg/dl FCR, pools/d TR, mg/kg/d Very dense LDL Cholesterol, m u d l ApoB, m d d l

Margarine (M)

Sitostanol Ester + M

Pravastatin + M

100.1 f 7 . 1 ~ ~ 50.4 f 3.94 0.385 f 0 . 0 2 7 ~ ~ 8.3 f 0 . 9

Sitostanol Ester + M

161.3 f 4.2 83.1 f 0.9 0.321 f 0.020 11.2 f 1.0

138.1 f 5.40 74.2 f 2,4a 0.305 f 0.013 9.0 f 0.5a

90.4 f 5.9.4' 45.3 f 2.346 0.320 f 0.034 6.6 f O.@br

54.6 f 7.8 27.6 f 4.1

51.6 rt 9.6 28.2 f 4.7

33.3 f 5.946 18.2 f 2.646

29.3 f 7.346 15.1 f 3 . W

98.3 f 6.1 52.8 f 4.1 0.365 f 0.029 7.9 f 1.2

77.2 f 8.W 41.2 f 4.3 0.386 f 0.042 6.8 f 1.1

61.3 f 9 . P 27.8 f 4.446 0.333 f 0.029 4.2 f 0.8R6

55.1 f 7.44b 27.5 f 3.9.b 0.367 f 0.030 4.3 f 0.8q6

0.21 f 0.07 2.7 f 0.4

0.24 f 0.07 4.9 f 1.6

0.21 f 0.09 4.4 f 1.9

0.18 f 0.05 3.6 f 1.0

Mean f SE. "Significantlydifferent from M. %Significantlydifferent from sitostanol ester M. 'Significantly different from pravastatin M; ANOVA.

+

+

over by 8% (Table 5). The decrease of the latter was related to the decrease (-18%)in biliary secretion of cholesterol and the increase in the s e x " sitosterol proportion ( T = 0.661). The drug increased the cholestanol and plant sterol proportions (Table 4), but did not affect cholesterol absorption (Table 5). Comparison to sitostanol ester period. With the exception of HDL cholesterol, the reductions of all lipid fractions were higher than those caused by sitostanol ester (Tables 2 and 3, Fig. 1). Triglyceride-richlipoproteins became more enriched in triglycerides with decreased cholesterol proportions. The higher decrease of LDL apoB was due to a marked decrease in both the light and dense LDL apoB, decreased TR of the dense LDL apoB, and an obvious increase in FCR of the light LDL apoB. These findings were associated with decreased biliary secretion, synthesis and turnover of cholesterol, and increased absorption of cholesterol (Tables 4 and 5).

Pravastatin + sitostanol ester Inhibition of both synthesis and absorption of cholesterol (Tables 4 and 5) further lowered total, VLDL, IDL, and LDL cholesterol, the respective serum lipid reductions ranging from 8 to 10%from the pravastatin period, and from 35 to 44% from the control margarine period (Fig. 1, Table 3). The changes for LDL cholesterol were solely due to decreased TR, detectable in the dense LDL fraction, while the removal of LDL apoB was unchanged (Table 3). The changes were associated with increased

triglycerideand decreased cholesterolproportions in the lipoproteins(Table2). Cholesterolabsorption and synthesiswere expectedto beroughlyinbetween thevaluesduringthesitostanoland pravastatin periods. As shown in Tables 4 and 5, this was true for absolute and relative absorption and biliary secretion of cholesteroland the plant sterol proportions, so that all these values were below those of basal margarine periods (respective reductions -34%,-29%, -15%, -26%,and -9%).Cholesterol synthesis and turnover had virtuallyreturned to the basal level (respectivereductions only -2%and -3%),yet the precursor sterol proportions were similar to those during the pravastatin period, and theywere23-50%below thebasalvalues. DISCUSSION Pravastatin significantly reduced the cholesterol and triglyceride contents of all apoB-containinglipoproteins by enhancing FCR and decreasing TR for LDL apoB, and synthesis and turnover of cholesterol. Reduced cholesterol synthesis was followed by lowered biliary secretion of cholesterol. Additional inhibition of cholesterol absorption by the combination of sitostanol ester with pravastatin further decreased LDL cholesterol and apoB by decreasing still more the TR for LDL apoB. Although the cholesterol/apoB ratio was decreased only in dense LDL, VLDL and IDL and the light and dense LDL became enriched in triglyceride. Pravastatin alone or with sitostanol had no effect on HDL cholesterol level and apoA-I metabolism.

Gylling and Miettinen Inhibition of cholesterol absorption and synthesis 1781

TABLE 4. Serum cholesterol Drecursor and Dlant sterol DroDortions Pravastatin + Sitostanol Variables

Margarine (M)

Sitostanol Ester + M

Pravastatin + M

Ester + M

102mmol/mol cholesterol

Squalene A8 lathostenol Desmosterol Lathosterol Campesterol Sitosterol Cholestanol

35 f 5 30 f 3 103 It 7 215 f 23 250 29 112 f 11 92 f 6

38 f 6 35 f 3 114 f F 236 f 29 136 f 1G‘ 64 f 9 80 7.

*

40 f 3 14 f 10,’ 67 f 54b 156 f 8~~ 282 24b 147 i 12a.’ 99 f 56

*

* -

41 f 4 15 f I,’.’ 70 f 5QbJ 165 f 14 184 f 10%’~ 102 64c 98 f 5)

*

Mean f SE. “P 0.05 from M. bP C 0.05 from sitostanol ester M. ‘P < 0.05 from pravastatin M.

+

+

Pravastatin lowered LDL cholesterol in the present NIDDM subjects similarly to results of previous studies in non-diabetics (27-31)or in diabetics (3). However, the cholesterol synthesis-loweringeffect of pravastatin, assessed either with the sterol balance data or the precursor sterol proportions, was only modest when compared with results obtained in familial hypercholesterolemia (FH) (27), primary moderate hypercholesterolemic (30), or gallstone patients (31). Interestingly, in the present study, the inhibition of cholesterol synthesis by 9% caused the more potent hypocholesterolemic effect (32%) than the respective 11% decrease caused by 68% inhibition of cholesterol absorption efficiency by sitostanol ester. The combination treatment of pravastatin and sitostanol ester revealed that the most efficient (44%) LDL cholesterol lowering was associated with cholesterol malabsorption and apparently slightly depressed cholesterol synthesis. Serum cholesterol precursor proportions of A* lathostenol, desmosterol, and lathosterol, but not squalene,

reflected cholesterol synthesis in the present series also so that the values were increased by sitostanol and decreased by pravastatin, but were unchanged by the addition of cholesterol malabsorption with sitostanol to pravastatin-induced synthesis lowering. Despite increased fecal neutral sterol output by the sitostanol addition, sterol balance values were not changed significantly from the pravastatin period; cholesterol synthesis was, in fact, virtually the same as during the control margarine period, yet the precursor sterol values were 25-50% lower during the pravastatin-sitostanol period than the basal margarine period. We speculate that the enzyme activities converting the precursor sterols to cholesterol are down-regulated less than HMG-CoA reductase resulting in decreased precursor sterol proportions. We have earlier observed a similar dissociation between sterol balance data and cholesterol precursors when pravastatin was associated with gemfibrozil treatment (32).The latter increased cholesterol synthesis according to both fecal and precursor analysis. During

TABLE 5. Cholesterol absorption and metabolism Pravastatin 4 Margarine (M)

Variables

Sitostanol Ester + hl

Pravastatin + M

Sitostanol Ester + M

30.3 f 1.7b

17.9 f 3.06,r

0.28 f 0.10 1.17 f 0.350 1.45 f 0.440 14.4 f 0.F

0.86 f 0.07b 3.23 f 0.25b 4.08 f 0.29 10.6 f 0 . e b

0.62 f 0.17b 2.03 f 0.42ab.‘ 2.65 f 0.55a”‘ 1 1 .o f 0.94b

5.2 f 0.4 11.7 f 0.7 1.29 f 0.07

5.3 f 0.3 16.3 f 0 . 9 1.68 f 0.17

6.3 f 1.5 9.4 0 . P 1.51 f 0.11

*

5.5 f 0.3 11.7 f 0.6”‘ 1.41 f 0.12

14.2 f 0.6 14.9 f 0.5

18.3 k O.Sa 18.6 f O B a

12.9 f 1.74 13.7 f 1.74)

13.9 f 0.86 14.5 f O B b

Cholesterol absorption, %

25.3 f 2.8

8.1 f 2.4a

Dietary cholesterol absorbed Biliary cholesterol absorbed Total cholesterol absorbed Biliary cholesterol secretion

0.72 f 0.13 3.28 f 0.38 4.00 f 0.51 12.9 f 0.4

Fecal bile acids Fecal neutral sterols Fecal campesterol Cholesterol synthesis Cholesterol turnover

mdWhY

Mean f SE. “ P C 0.05 from M. ’f‘ < 0.05 from sitostanol ester M . ‘P < 0.05 froni pravastatin M.

+

+

1782 Journal of Lipid Research Volume 37, 1996

TABLE 6. HDL apoA-I kinetics Pravastatin + Sitostanol Variables

Margarine (M)

Sitostanol Ester + M

Pravastatin + M

Ester + M

46.4 f 3.2

45.9 f 2.5

46.8 f 4.4

46.6 f 4.2

ApoA-I, m g / d

134.9 f 4.7

132.8 f 3.2

125.4 f 5.0

125.7 f 5.5

ApoA-11, m u d l

FCR apoA-I, pools/d

29.1 f 0.7 0.208 f 0.014

31.0 f 1.1 0.218 k 0.018

28.0 k 0.8 0.226 f 0.014

28.5 f 1.0 0.213 f 0.010

TR apoA-I, mg/kg/d

12.3 f 0.8

12.6 k i.0

12.2 k 0.7

11.6f0.5

HDL cholesterol, m u d l

Mean f SE. FCR, fractional catabolic rate; TR, transport rate.

the combination of pravastatin and gemfibrozil, fecal data showed virtually no change in synthesis, while the precursor sterol proportions were decreased by up to 36%(32). On the other hand, cholesterol absorption was unchanged by pravastatin in the diabetics, and the elevated serum plant sterol and cholestanol proportions after pravastatin treatment result from their accumulation in serum sterol mixture due to the reduced biliary sterol secretion. In fact, the higher the decrease in cholesterol turnover or biliary secretion, the higher was the increase in the serum sitosterol proportions. Thus, the statin-induced decrease in cholesterol turnover, not discovered by cholesterol kinetics (33), is actually shown for the first time in the present study, and it explains our frequent finding of increased plant sterol proportions during statin treatments (29, 32, 34). The present study is the first to evaluate the effects of pravastatin alone and in combination with cholesterol malabsorption on LDL composition and kinetics in NIDDM. Pravastatin lowered the number of LDL particles through enhanced FCR and decreased TR for LDL apoB, resulting in increased triglyceride proportion of most remaining apoB-containing lipoproteins. The kinetic findings are, in general, consistent with the previous observations in animal models (35), in FH (8), in non-FH (36), and in combined hyperlipidemia (37), whose lipid profile resembles that of the present NIDDM series. However, in the dense LDL fraction both FCR and TR for apoB were significantly diminished from the control period (Table 3) suggesting that removal of the light particle was accelerated by pravastatin, leaving less substrate to enter into the dense compartment. It has been shown previously in some studies (38-40) that statins, instead of up-regulating the catabolism of LDL, reduce the TR for LDL apoB. Sitostanol ester, too, reduced serum cholesterol by inhibiting only the TR for LDL apoB, when VLDL and IDL also tended to be decreased. In fact, in hamsters, lovastatin and cholesterol malabsorption, caused by stigmastanyl-phosphocholine, lowered LDL cholesterol without any effect on the removal rate of LDL, but enhanced hepatic uptake of VLDL and decreased, accordingly, synthesis

of LDL (41). The authors considered that cholesterol malabsorption contributed to cholesterol lowering by a reduced flux of intestinal cholesterol to the liver, resulting in lowered VLDL production and enhanced VLDL removal due to a mechanism perhaps different from LDL receptor activity. The present results suggest that the combination treatment reduced intestinal flux of cholesterol to the liver without compensatory increase in synthesis, resulting in hepatic lack of cholesterol. The latter is reflected by reduced efflux of hepatic cholesterol as decreased biliary and VLDL secretion so that serum VLDL and IDL cholesterol levels are lowered through their reduced formation and reduced conversion to LDL. Consequences of these changes apparently are that in our diabetic patients the LDL particle and also its precursors, VLDL and IDL, became proportionally triglyceride-enriched by the treatments, and the dense LDL became poor in cholesterol during the combined treatment. This altered lipoprotein composition may be associated with NIDDM, because in non-NIDDM subjects with combined hyperlypidemia lovastatin did not change the triglyceride to apoB ratio (42).Neomycin also reduces serum cholesterol by inhibition of TR for LDL apoB with no consistent effect on removal (43). Thus, cholesterol malabsorption, causing a reduced flux of intestinal cholesterol to the liver, may result, depending on compensatory increase in synthesis of cholesterol, in a decreased synthesis of VLDL (probably rich in triglycerides and low in cholesterol), and, accordingly, in a reduced transport of VLDL to LDL and in lowering of cholesterol in serum. Up-regulation of hepatic receptor activity may occur picking up mainly remnants and light LDL. The present study showed that sitostanol ester margarine combined as a normal dietary ingredient to HMGCoA reductase inhibitors lowered serum cholesterol by about 35% and LDL cholesterol by 45%. It is clear that dietary sitostanol ester margarine normalizes serum cholesterol of many patients with increased values, while patients resistant to this dietary measure can be treated with addition of a statin, most likely frequently at a reduced dose.

Gylling and Miettinen

Inhibition of cholesterol absorption and synthesis 1783

This study was supported by grants from the Finnish Academy of Medical Sciences, the Finnish Heart Research Foundation, Helsinki, the Juho Vainio Foundation, Helsinki, and University of Helsinki. Pravastatin was received from Bristol-MeyerSquibb, Helsinki, Finland. The expert technical assistance of Leena Kaipiainen, Orvokki Ahlroos, Pia Hoffstrom, Elli Kempas, Ritva Nissila, Leena Saikko, Anja Salolainen, and Antti b i n e is greatly acknowledged. Manuscrip received 28 December 1996 and in rwisedform 13 May 1996.

REFERENCES 1. Grundy, S. M. 1988. HMG-CoA reductase inhibitors for treatment of hypercholesterolemia. N. Engl. J. Med. 319: 24-33. 2. Miettinen, T. A., P. Puska, H. Gylling, H. Vanhanen, and E. Vartiainen. 1995. Reduction of serum cholesterol with sitostanol-ester margarine in a mildly hypercholesterolemic population. N. Engl. J. Med. 333: 1308-1312. 3. Gylling, H., and T. A. Miettinen. 1994. Serum cholesterol and cholesterol and lipoprotein metabolism in hypercholesterolemic NIDDM patients before and during sitostanol ester-margarine treatment. Diabetologia. 37: 773-780. 4. Gylling, H., and T. A. Miettinen. 1995. The effect of cholesterol absorption inhibition on low density lipoprotein cholesterol level. Atherosclerosis. 117: 305-308. 5 . Harris, M. I. 1991. Hypercholesterolemia in diabetes and glucose intolerance in the U.S. population. Diabetes Care. 1 4 3&3?4. 6. Ginsberg, H. N. 1991. Lipoprotein physiology in nondiabetic and diabetic states. Relationship to atherogenesis. Diabetes Care. 1 4 839-855. 7. Yoshino, G., T. Kazumi, M. Iwai, M. Matsushita, K. Matsuba, R. Uenoyama, I. Iwatani, and S. Baba. 1989. Longterm treatment of hypercholesterolemic non-insulin-dependent diabetics (NIDDM) with pravastatin (CS-514). Atherosclerosis. 75: 67-72. 8. Garg, A., and S. M. Grundy. 1988. Lovastatin for lowering cholesterol levels in non-insulin-dependent diabetes mellitus. N. Engl..J. Med. 318: 81-86. 9. Miccoli, R., A. Bertolotto, M. G. Giovannitti, I. Tedesco, S. Manfredi, G. Galantini, and R. Navalesi. 1992. Simvastatin for lowering cholesterol levels in non-insulin-dependent diabetes mellitus and in primary hypercholesterolemia. Curr. Ther. Res. 51: 66-74. 10. Zambon, S., A. Lapolla, G. Sartore, C. Gherardingher, A. Cortella, E. Manzato, G. Crepaldi, and D. Fedele. 1992. Long-term treatment with simvastatin in hypercholesterolemic non-insulindependent diabetic patients. Curr. Ther. Res. 5 2 221-229. 11. Cassader, M., G. Ruiu, R. Gambino, N. Alemanno, F. Veglia, and G. Pagano. 1993. Hypercholesterolemia in non-insulin-dependent diabetes mellitus: different effect of simvastatin on VLDL and LDL cholesterol levels. Atherosclerosis. 99: 4 7-53. 12. Miettinen, T. 1979. Effects of neomycin alone and in combination with cholestyramine on serum cholesterol and fecal steroids in hypercholesterolemic subjects. J. Clin. Invest. 64: 1485-1493. 13. Knuts, L-R., M. Rastas, and P. Haapala. 1991. Micro-Nutrica. Version 1.O. Helsinki: Kansaneliikelaitos.

1784 Journal of Lipid Research Volume 37, 1996

14. Lipid Research Clinics Program. 1974. Manual of Laboratory Operations. Lipid and Lipoprotein Analysis. National Institutes of Health. DHEW Publication No. (NIH) 75-628. Bethesda, MD. 1:51-59. 15. Gylling, H., H. Vanhanen, and T. A. Miettinen. 1993. Effects of ketoconazole on cholesterol precursors and low density lipoprotein kinetics in hypercholesterolemia. ]. Lipid Res. 34: 5 9 4 7 . 16. Miettinen, T. A. 1982. Gas-liquid chromatographicdetermination of fecal neutral sterols using a capillary column. Clin. Chim. Acta. 124: 245-248. 17. Crouse, J. R., and S. M. Grundy. 1978. Evaluation o f ;I continuous isotope feeding method for measurement of cholesterol absorption in man.J Lzpzd Res. 19: 967-971. 18. Bolin, D. W., R. P. King, and E. W. Klosterman. 1952. A simplified method for the determination of chromic oxide (Cr203) when used as an index substance. Science. 116: 634-4535. 19. Miettinen, T. A., E. H. Ahrens, Jr., and S. M. Grundy. 1965. Quantitative isolation and gas-liquid chromatographic analysis of total dietary and fecal neutral steroids.J. Lipid Res. 6: 41 1-424. 20. Grundy, S. M., E. H. Ahrens, Jr., and T. A. Miettinen. 1965. Quantitative isolation and gas-liquid chromatographic analysis of totd fecal bile acids.]. Lipid Res. 6: 397-410. 21. Czubayko, F., B. Beumers, S. Lammsfuss, D. Lutjohann, and K. von Bergmann. 1991. A simplified micremethod for quantification of fecal excretion of neutral and acidic sterols for outpatient studies in humans.J. LipidRes. 32: 1861-1867. 22. Vega, G. L., H. Gylling, A. V. Nichols, and S. M. Grundy. 1991. Evaluation of a method for study of kinetics of autologous apolipoprotein A-I. J. Lipid Res. 32: 867-875. 23. McFarlane, A. S. 1958. Efficient trace-labeling of proteins with iodine. Natum. 182: 53-57. 24. Bilheimer, D. W., S. Eisenberg, and R. I. Levy. 1972. The metabolism of very low density lipoproteins. I. Preliniinary in vitro and in vivo observations. Biochim. Biophys. Acta. 260: 212-221. 25. Matthews, C. M. E. 1957. The theory of tracer experiments with 1311-labeled plasina proteins. Phys. Med. Biol. 2: 36-53. 26. Miettinen, T. A,, R. S. Tilvis, and Y. A. Kesaniemi. 1990. Serum plant sterols and cholesterol precursors reflect cholesterol absorption and synthesis in volunteers of a randomly selected male population. Am.J. Epihniol. 131: 20-31. 27. Vanhanen, H., Y. A. Kesaniemi, and T. A. Miettinen. 1992. Pravastatin lowers serum cholesterol, cholesterolprecursor sterols, fecal steroids, and cholesterol absorption in man. Metabolism. 41: 588-595. 28. Vega, G. L., R. M. Krauss, and S. M. Grundy. 1990. Pravastatin therapy in primary moderate hypercholesterolemia: changes in metabolism of apolipoprotein B-containing lipoproteins. J. Intem. Med. 227: 81-94. 29. Miettinen, T. A., H. Vanhanen, J-P. Ojala, and M. J. Tikkanen. 1992. Noncholesterol sterols and fecal elimination of cholesterol during statin and fibrate treatment. Atherosclerosis. 97: S73-90. 30. Vanhanen, H. 1994. Cholesterol malabsorption caused by sitostanol ester feeding and neomycin in pravastatintreated hypercholesterolemic patients. Eur. J. Clin. Pharmacol. 47: 169-176. 31. Reihner, E., M. Rudling, D. Stahlberg, L. Berglund, S. Ewerth, 1. Bjorkhem, K. Einarsson, and B. Angelin. 1990. Influence of pravastatin, a specific inhibitor of HMG-CoA

reductase, on hepatic metabolism of cholesterol. N. Engl. J. Med. 323: 224-228. 32. Vanhanen, H. T., and T. A. Miettinen. 1995. Cholesterol absorption and synthesis during pravastatin, gemfibrozil and their combination. Atherosclerosis. 115: 135-146. 33. Goldberg, I. J., S. Holleran, R. Ramakrishnan, M. Adams, R. H. Palmer, R. B. Dell, and D. S. Goodman. 1990. Lack of effect of lovastatin therapy on the parameters of whole-body cholesterol metabo1ism.J. Clin. Invest. 86: 801-808. 34. Uusitupa, M. I.J., T. A. Miettinen, P. Happonen, T. Ebehg, H. Twola, E. Voutilainen, and K. Pyodii. 1992. Lathosterol and other noncholesterol sterols during treatment of hypercholesterolemiawith lovastatin alone and with cholestyramine or guar gum. A T t " c h . T h d . 12 807-813. 35. Ma, P. T. S., G. Gil, T. C. Sudhof, D. W. Bilheimer,J. L. Goldstein, and M. S. Brown. 1986. Mevinolin, an inhibitor of cholesterol synthesis, induces mRNA for low density lipoprotein receptor in livers of hamsters and rabbits. Roc. Natl. Acad. Sci. USA. 83: 83-70-8374. 36. Kentinen, K., M. J. Savolainen, J. I. Heikkila, and Y. A. Kesaniemi. 1993. Lovastatin enhanceshepatic uptake oflow density lipoprotein in humans.J. Lipid &. 3 4 1975-1982. 37. Parhofer, K. G., P. H. R. Barrett, J. Dunn, and G. Schonfeld. 1993. Effect of pravastatin on metabolic parameters of apolipoprotein B in patients with mixed hyperlipoproteinemia. Clin. Invest. 71: 939-946.

38. Grundy, S. M., and G. L. Vega. 1985. Influence of mevinolin on metabolism of low density lipoproteins in primary moderate hypercholesterolemia. ]. Lipid Res. 2 6 1464-1475. 39. Gaw, A., C.J. Packard, E. F. Murray, G. M. Lindsay, B. A. Griffin, M. J. Caslake, B. D. Vallance, A. R. Lorimer, and J. Shepherd. 1993. Effects of simvastatin on apoB metabolism and LDL subfraction distribution. Arterioscler. Thromb. 13: 170-189. 40. Arad, Y., R. Ramakrishnan, and H. N. Ginsberg. 1990. Lovastatin therapy reduces low density lipoprotein apoB levels in subjects with combined hyperlipidemia by reducing the production of apokontaining lipoproteins: implications for the pathophysiology of apoB production.]. Lipid Res. 31: 567-582. 41. Himber, J., B. Missano, M. Rudling, U. Hennes, and H. J. Kempen. 1995. Effects of stigmastanyl-phosphocholine (Ro 16-6532)and lovastatin on lipid and lipoprotein levels and lipoprotein metabolism in the hamster on different diets.J. Lipid Res. 3 6 1567-1585. 42. Tilly-Kiesi, M. 1991. The effect of lovastatin treatment on lowdensity lipoprotein hydrated density distribution and composition in patients with intermittent claudication and primary hypercholesterolemia.Metabolism. 4 0 623-628. 43. Kesaniemi, Y. A., and S. M. Grundy. 1984. Turnover of low density lipoproteins during inhibition of cholesterol absorption by neomycin. Arteriosclerosis. 4 4 1 4 .

Gylling and Miettina Inhibition of cholesterol absorption and synthesis 1785