Catabolism of very low density lipoproteins in the rabbit effect of changing composition and pool size

176

Bioehimica et Biophysics Acta, 528 (1978) 176-189 @ Elsev~er/North-Holland Biomedical Press

BBA 57127

CATABOLISM

OF VERY LOW DENSITY LIPOPROTEINS IN THE RABBIT

EFFECT OF CHANGING

RAMPRATAP

COMPOSITION AND POOL SIZE

S. KUSHWAHA and WILLIAM R. HAZZARD

northwest Lipid Research Clinic, ~ar~or~iew Seattle, Wash. 98104 (U.S.A.)

Medical Center,

Uni~rs~ty

of Washington,

(Received August 4th, 1977)

Summary To determine the metabolic mechanism of hypercholesterolemia in rabbits produced by feeding cholesterol-rich diets, control and hypercholesterolemic rabbits were injected with I-labelled very low density lipoproteins (VLDL, a! 1.006 g/ml) from control and/or hypercholesterolemic donors. Apolipoprotein B in VLDL decayed biphasically. The first phase occurred much more rapid than the second. 95% of the VLDL ap~lipopro~in B was catabolized via the first phase (t,,z = 0.55 + 0.19 h) in normal rabbit with the immediate appearance of this radioactivity in intermediate density lipoproteins (IDL, d 1.0061.025 g/ml) and low density lipoproteins (LDL, d 1.025-1.063 g/ml). The apolipoproteins C and E at the same time were transferred to high density lipoproteins where they decayed biphasically. The apolipoprotein B from hypercholesterolemic VLDL in the normal recipient disappeared at a similar rate as from normal VLDL via phase I; however, it was incompletely converted to IDL and LDL. Apolipoprotein B from normal VLDL in cholesterol-fed rabbits disappeared at a normal rate via phase I, but only 82% was catabolized by this phase. Hypercholesterolemic VLDL injected into the hypercholesterolemic recipient was less rapidly ca~bolized via phase f (t,,, = 2.5 ?I 0.89 h) and only a small fraction was converted to IDL and LDL. -_ Introduction A short time after beginning a cholesterol-enriched an increased level of very low density lipoprotein

diet, the rabbit develops (VLDL, d 1.006 g/ml)

Abbreviations: VLDL, very low density lipoproteins (d 1.006 g/ml): IDL. intermediate density lipoproteins (d 1.006-1.025 g/ml): LDL, low density lipoproteins (d 1.026-1.063 B/ml); HDL. high density lipoproteins (d 1.063-1.21 g/ml); SDS, sodium dodecyl sulfate.

177

cholesterol [l-3]. With continued cholesterol ingestion, VLDL and intermediate density lipoproteins (IDL, d 1.006-1.019 g/ml) rise precipitously, low density lipoproteins (LDL, d 1.019-1.063 g/ml) increase moderately, but high density lipoprotein (HDL, d 1.063-l .21 g/ml) levels remain relatively constant. Moreover, the composition of these lipoproteins changes [ 4-61: both VLDL and IDL becoming very rich in cholesterol, while all classes become enriched in a peptide homologous to the arginine-rich peptide of the human (apolipoprotein E). With the persistence of these changes during continued cholesterol feeding, atherosclerosis rapidly supervenes, and thus, the rabbit is used as a model of diet-induced atherosclerosis. However, the metabolic mechanism of hypercholesterolemia in the rabbit remains uncertain. Recent studies have suggested that VLDL catabolism is retarded in the cholesterol-fed rabbit [7]. These studies failed to specify the respective roles of the induced changes in VLDL composition and pool size in contributing to this delay. We have studied VLDL turnover in normal a&cholesterol-fed rabbits in an effort to answer the following questions: (1) What is the effect of changing VLDL pool size and composition upon VLDL catabolism? (2) What is the relationship among apolipoprotein B, apolipoprotein E, and the C apolipoproteins during VLDL catabolism? (3) What are the probable metabolic relationships between VLDL, IDL, LDL, and HDL in the normal and cholesterol-fed rabbit? Materials and Methods

Animals and diets. Female New Zealand white rabbits (3.5-4.0 kg, 12-15 weeks old) were used and the control animals were fed a rabbit chow diet (Carnation Co.). Rabbits in the hypercholesterolemic group received the same diet with the addition of 1% cholesterol (w/v) dissolved in 10% sesame oil [2]. Both groups fed ad libitum (about 120-150 g/day). The cholesterol-fed rabbits were maintained on this diet for 8 weeks, with the exception of two animals studied after 2 days on the diet. Isolation and lubelling of VLDL. To isolate VLDL for labelling, non-fasting rabbits were bled from the ear. The blood was collected in tubes containing EDTA (1 mg/ml) and pooled; plasma was separated by low speed centrifugation and then centrifuged in a Beckman model L2-65B ultracentrafuge using a 50 Ti rotor at a speed of 40 000 rev./min at 60°C for 20 h. The tubes were then sliced and the 5 ml top layer containing VLDL was aspirated. The VLDL thus obtained was diluted with one volume saline (d = 1.006 g/ml) and centrifuged again under similar conditions. At the end of this centrifugation, the top 5 ml were aspirated and this washed VLDL used for labelling experiments. After determination of protein content [ 81, this VLDL was labelled by the ICl procedure of McFarlane [9], as modified by Bilheimer et al. [lo]. Free “‘1 or “‘1 was removed by dialysis against 0.15 M NaCl/O.Ol M EDTA (pH 7.4, 6°C) with 6-8 changes. After dialysis, 10 ~1 of the labelled VLDL (added to non-radioactive VLDL) was used for determining the trichloroacetic acidprecipitable radioactivity (12.5% trichloroacetic acid) and free iodide. Iodide was converted to I2 using 40% KI and 30% H202 and extracted with chloroform. The lipid radioactivity was determined with 20 volumes of chloroform/methanol (2 : 1, v/v), washing the extract with 0.7% KC1 in 0.02 M HCl and counting

178

after evaporation to dryness. The radioactivity in apolipoprotein B and other smaller peptides of VLDL was determined by taking a small aliquot of labelled VLDL (diluted with 10 volumes of non-labelled VLDL) and adding the same volume of tetramethylurea [ 11,123. Radioactivity in tetramethylurea-soluble and -insoluble (apolipoprotein B) fractions was then determined (Table I) separately using a gamma spectrometer (Nuclear Chicago). A small aliquot was used to isolate the ~tr~ethylurea-soluble apolipoproteins on polyacrylamide gel electrophoresis and the radioactivity associated in major bands was determined (Table I). VLDL turnover. Non-fasted control and cholesterol-fed rabbits were injected with labelled VLDL in the marginal ear vein. Two turnover experiments were normally conducted concurrently. Each animal received l-2 mg I-labelled VLDL protein (12-25 FCi, 1-Z ml). Blood samples were drawn at 5 min, 1, 2, 4, 6, 8, 12, 24, 32, and 48 h from the contralateral ear artery, In later experiments, the samples were only drawn for a period of 24 h, since most of the radioactivity has disappeared from VLDL in this interval. Blood was collected in tubes containing EDTA (1 mg/ml) and centrifuged to obtain the plasma. A small aliquot of whole plasma was counted in a gamma counter to determine recovery at each time point. 3 ml plasma was used to isolate individual lipoprotein fractions as follows: The plasma was overlaid with saline (d 1.006 g/ ml) and centrifuged in the 40.3 rotor in the Beckman model L5-40 ultracentrifuge (40 000 rev.fmin, 20 h, 6°C). VLDL was obtained by pipetting the top layer after slicing the tube. Similarly IDL (d 1.063-1.025), LDL (d 1.0251.063 g/ml) and HDL (d 1.063-1.21 g/ml) were obtained by sequential ultracentrifugation by adjusting the densities of the material to corresponding lipoprotein fraction. The lipoproteins thus separated gave a single band on agarose gel electrophoresis [13]. The radioactivity in a small aliquot of each fraction was counted to allow plotting of the decay curves. The same aliquot was then treated with one volume of tetramethylurea and both tetramethylurea-soluble (C and E peptides) and -insoluble (apolipoprotein B and lipid) fractions were counted. Another aliquot of each fraction was used to extract the lipids as described above, and both protein and lipid fractions (after evaoorating the solvent) were counted. A small aliquot was used to quantify the total and tetramethylurea-soluble protein. A portion of each fraction was used to determine the radioactivity associated with individual apolipoproteins soluble in tetramethylurea. Lipid and protein analysis. Cholesterol and triglycerides in plasma and all other fractions were measured by autoanalyzer II techniques according to standard procedures [ 141. The phospholipids were estimated by determining phosphorus contents of plasma lipid extracts [ 151. The apoprotein B was determined as the difference between total and tetramethylurea-soluble protein as determined by the Lowry procedure [8] using human serum albumin as standard. In man [ 111, rat [12], and also in the rabbit, only apolipoprotein B proved to be a tetramethylurea-insoluble protein, while the radioactivity in the total tetramethylurea-soluble fraction was found to equal the total of that in all VLDL apolipoprotein bands on polyacrylamide gel. Tetramethylurea-insoluble protein was subjected to electrophoresis (SDS and urea gels), but it did not enter the gel for any appreciable distance. VLDL apolipoprotein B was also

I

OF 131I-LABELED

NORMAL

AND

* Mean + S.D. for four experiments.

Trichloroacetic acid precipitable Free iodine Lipids Apolipoprotein B (tetramethylurea insoluble minus lipids) Tetnunethyluxca-soluble proteins Zone 2 (apollpoprotein C-I) Zone 3 (apolipoprotein E) Zone 4 (apoltpoprotein C-II and C-III)

Fm&kln

CHARACTERIZATION

TABLE HYPERCHOLESTEROLEMIC

91.9 2.2 17.2 53.4 29.4 3.0 7.7 18.7

f f 2 f f * + f

2.3 * 0.8 1.9 9.4 7.4 0.5 0.7 1.1

radioactivitY

Percent of total

11.0 36.5 53.5

-

-

Densitometric area of polyacrylamide gel electrophoresis (percent of total tetramethylurea soluble)

13 1I-labelednormal VLDL

125 I-LABELED VERY

LIPOPROTEINS

93.7 2.9 25.6 47.0 27.3 3.2 13.3 9.9

f * f f f f * f

1.1 1.1 2.5 3.6 5.8 0.6 1.0 1.4

Percent of total radioactivity

VLDL

7.5 77.0 15.6

-

Densltometic area of polyacrylamide gel electrophoresls (percent of total tetramethyl-a soluble)

1251-labeled hypercholesterolemlc

LOW DENSITY

180

determined by delipidating the VLDL with chloroform/methanol (2 : 1, v/v) at 4°C and washing the C and E peptides with 8 M urea. Apolipoprotein B was dissolved in 0.1 M SDS and the protein determined [8].

Pulyac~lamide gel electrophoresis and determination of ~di~activity tetramethylurea-soluble apolipoproteins. The apolipoproteins were separated

in

in 10% polyacrylamide gels containing 8 M urea. The delipidation was carried out by tetramethylurea in the gel tubes as described by Kane [ll]. The gels were stained with Coomassie Brilliant Blue after fixing with 10% trichloroacetic acid for 15 min and the distribution of protein among the respective bands estimated by densitometric scanning. The gels were sliced into four parts as shown in Fig. 1 and the radioactivity determined. The bands obtained in VLDL and HDL were similar to those described by Shore et al. [4,6]. Zone 2 contained apolipoprotein CI (a minor band), Zone 3 contained apolipoprotein E

A

Fig. 1. Stained 10% polyacrylamide gel of tetramethylurea-soluble apolipoproteins of rabbit VLDL. A. normolipidemic and B. hypercholesterolemic. To determine the distribution of radioactivity among aPolipoproteins, the gels were sliced into zones aa shown by parallel lines and counted. Zone I contains the proteins not entering the gel; Zone 2 contains apolipoprotein CI; Zone 3 contains apolipoproteln E and; Zone 4 contains apoii~oprotein CII and CIII’s.

181

(two major bands, similar to RI1 and R3 of Shore et al. 141) and Zone 4 contained faster-moving CII and CIII peptides. Calculations. The decay curves for apolipoprotein B in VLDL and other fractions were plotted by considering the total radioactivity in apolipoprotein B of these fractions, at 5 min after injection, as the 100% injected dose. The radioactivity, at each interval in apolipoprotein B in each lipoprotein fraction, was expressed as the percentage of this initial value. The half-lives for the first and second phases were calculated by the least square method. For the first phase, points from 5 min to 2 or 4 h were used, while for the second phase, points from 6 h and later were included. The fractional catabolic rates for apolipoprotein B of VLDL (within intrav~cular pool) were calculated [7,16] as follows: Fractional Catabolic Rate =

1 A’/a +I?//3

where A’ = A/P,, B’ = B/P,, and PO= 100 (see Fig. 3). Turnover rates for the apolipoprotein B in VLDL was calculated as fractional catabolic rate X pool size and were expressed as mg/h per 100 ml plasma. Results and Discussion Plasma and lipopr~te~n lipids

Results summarized in Table II reveal that feeding a diet rich in cholesterol to rabbits resulted in a 22-fold increase in plasma cholesterol, a 4-fold increase in phospholipid and a 1.5fold increase of triglyceride levels. Among the lipoprotein fractions, cholesterol was greatly increased in VLDL (155fold) and IDL (48-fold), as reported earlier [4,5], while LDL cholesterol increased only 3-fold and HDL increased 1.6-fold. At the same time, triglyceride in VLDL increased only 5-fold; in LDL, increased 3-fold; and, in both LDL and HDL, decreased to 20% of pre-treatment values. Apolipoprotein

responses to cholesterol feeding

The changes in lipid levels and dis~ibution were paralleled by major changes in the apolipoproteins within each lipoprotein fraction (Table If). Tet~ethylurea-msoluble protein increased by 17-fold in VLDL and 15-fold in IDL, -but only a-fold in LDL and 2.5-fold in HDL. This was associated with a reduction in the proportion of the total protein represented by the tetramethylureasoluble peptides within both VLDL and IDL. However, the distribution between tetramethylurea-insoluble and soluble proteins within LDL did not change and the soluble fraction in HDL decreased only slightly. Moreover, additional changes occurred within the tetramethylurea-soluble apolipoproteins of all fractions (Fig. 2), which were richer in a protein, which migrates in two bands on polyacrylamide gel electrophoresis (Zone 3 in Fig. 1). This protein has been shown to be homologous to the arginine-rich peptide (apolipopro~in E) in man [4 1. The fraction of VLDL proteins represented by these bands (Table I) increased by Z-fold in the choles~rol-fed animal; the proportional reduction in the apolipoprotein C peptides being 30% for CI and 70% for CII-CIII. These changes translate into an absolute rise in apolipopro-

II

Whole plasma VLDL IDL LDL HDL Recovery

Whole plasma VLDL IDL LDL HDL Recovery

(A) Normal rabbits

(B) Cholesterol-fed

* Mean + S.D. for four animals.

Fraction

1515.0 790.0 608.0 89.0 25.0

69.8 5.1 12.7 31.0 14.8

f 176.7 f 62.3 + 68.0 + 5.8 i: 2.8

+- 9.2 * f 3.6 f. 18.2 ir 1.2 r 5.8

100.00 52.1 40.1 5.9 1.7 99.8

100.00 7.3 18.2 44.4 21.1 91.0 117.0 75.0 33.0 7.0 3.0

76.5 16.9 11.1 32.6 15.1

f 322 f 11.1 k 4.2 r 2.1 k 0.6

f 11.5 + 7.6 + 1.3 f 8.9 f 2.1

mg/lOO ml

Percent of total

IN NORMAL

mg/lOO ml

FRACTIONS Triglyceride

OF LIPOPROTEIN

Cholesterol

COMPOSITION

Dietary status

LIPID AND APOLIPOPROTEIN

TABLE

100.00 64.1 28.2 5.9 2.5 100.7

100.00 20.8 14.5 42.6 19.7 97.6

Percent of total

AND

9.9

ml)

579.0 f 28.1 -

146.3 + -

(w/l00

Phospholipid

CHOLESTEROL-FED

127.1 123.2 16.2 139.5

7.6 8.3 8.8 142.9

k 18.2 * 15.1 t 4.3 + 24.2

t 2.0 c 1.7 * 1.1 f 14.4

Total protein (mg/lOO ml)

RABBITS

72.2 82.1 86.2 19.25

7.9

39.5 39.8 86.8

Tetramethylureainsoluble protein (percent of total)

Fig. 2. 10% polyacrylamide gel patterns of tetramethylurea-soluble apolipoproteins of rabbit plasma lipoproteins, A, normal VLDL; B, hypercholesterolemic VLDL: C, normal IDL; D, hypercholesterolemic IDL, E, normal LDL; F, hypercholesterolemic LDL: G. normal HDL; and H. hypercholesterolemic HDL. The gels were stained with Coomassie Brilliant Blue. G-260 dye, The lipoprotein fractions were isolated as described in Materials and Methods.

tein E concen~ation during cholesterol feeding of 17-fold in VLDL, while apolipoprotein CI rose only &fold and apolipoprotein CII-CIII only 2-fold. Mahley and Holcombe [ 171 have recently demonstrated similar changes in cholesterolfed rats. The enrichment of LDL with apolipoprotein E may be due to the presence of HDL, which has been reported to be formed in cholesterol-fed animals [17-191 and has the mobility similar to LDL. These changes are induced in rabbit lipoproteins in a very short time, which make this species a unique model for the study of hyperlipopro~inemia and atherosclerosis. The changes induced in rabbit lipoproteins could be compared with lipoproteins in human type III hyperlipoproteinemia or Broad-0 disease and hypothyroidism, where VLDL and IDL are rich in cholesterol and apolipoprotein E [20,21] with the predominant occurrence of /I-VLDL. Lipoprotein turnover studies The biphasic nature of VLDL apolipopro~~ B disappearance (Fig. 3) was similar to the whole VLDL decay reported by Rodriquez et al. [7]. The first phase was much shorter than the second phase, in both normal and cholesterolfed rabbits (Table III). In the case of normal animals when their own VLDL was injected (Fig. 3 and Table III), VLDL apolipoprotein B disappearance via the first phase (95%) greatly exceeded that via the second (5%). IDL and LDL

I&

24 32 Hours

40

48

Fig. 3. Decay of radioactivity in apolipoprotein B of plasma lipoprotein fractions (VLDL. IDL, and LDL) from normal rabbits when their own *2sI-labeled VLDL was injected. Lipoproteins were isolated as described in Materials and Methods and the radioactivity in apolipoprotein B was determined as the difference between total radioactivity in the lipoprotein fraction and the lipid plus tetramethylurea-soluble radioactivity. Each point is a mean of four experiments.

radioactivity at 5 min already represented 20 and 17.5%, respectively, of the apolipoprotein B radioactivity circulating at that time, suggesting a very rapid conversion from VLDL during those 5 min. Since the disappearance curve for VLDL apolipoprotein B intersected that for IDL at the peak of its radioactivity (assuming constant apolipoprotein B pool size), this was consistent with a precursor vs. product relationship between VLDL and IDL apolipoprotein B in these animals. This suggests that all IDL was derived from VLDL and the first phase mainly represented a lipoprotein lipase-mediated conversion of VLDL to IDL and LDL. The decay curves for tetramethylurea-soluble apolipoproteins in case of normal VLDL injected into the normal rabbit are given in Fig. 4. Radioactivity in all zones (Fig. 1) of polyacrylamide gels from VLDL decayed biphasically. The first phase was very rapid; most of the radioactivity was transferred to HDL (where the radioactivity peaked in 1 h) and to a lesser extent to IDL and LDL, where it also decayed biphasically. Eisenberg and Rachmilewitz [22,23] have similarly reported in rats that apolipoproteins in VLDL and HDL represent one common pool and the transfer of these apolipoproteins from VLDL to HDL depends upon the extent to which VLDL triglycerides have been hydrolyzed. In the case of normolipidemic and hypercholesterolemic VLDL injected into the normolipidemic recipients, the fractional catabolic rates and the half-lives for the first phase in both cases were similar (Table III). But, the conversion of normal VLDL apolipoprotein B to IDL (20% at 2 h) and LDL (17.5% at 2 h) was much more than that from hypercholesterolemic VLDL apolipoprotein B (IDL 10.5% and LDL 8.5% at 2 h). This suggests that, in addition to conversion

OF VLDL

RATES

22.7 it 5.5

0.90 * 0.20 18.2 r 8.1

21.0

0.64

+ 0.89

16.0

0.52 t

2.5

24.1 f 8.0

9.55 f 0.19 **

Values obtaked by plotting the mean values of each time interval. Mean f S.D. for four animals. Hypercholesterolemic. Mean values for two animals.

1515.0 k 176.7

hc VLDL/ hc recipient (after awd240n thediet)

* ** *** t

1516.0 f 176.7

12.9

Normal VLDL/ ha re&ient

202.4

79.4

hc *** VLDLj normal recipient

he VLDL/hc recipient (after 3 days on the diet)

57.3 f

(h)

00

9.06 t 0.01

0.12 * 0.05

0.28

0.37

0.37 t 0.12

Fractional catabolic rate (h-l)

AND CHOLESTEROL-FED

fli2 (second phase)

IN NORMAL

G/2 (first phase)

B METABOLISM

ChoIesterol pools&e of recipient snimals (mg/lOO ml plapma)

APOLIPOPROTEIN

Normal VLDL/ normal recipient

Donor/recipient

III

TABLE

8.5 10.5

X3.6

10.5 14.2

23.3

2.20

5.6 t 1.2

5.8

17.5

20.0

0.90 f 0.24

9.4

MaximUUl radioactivity * appearing in LDL (percent of dose)

(mrrlh per 100 ml plasma)

MaximUm radioactivity * appearing in IDL (percent of dose)

Turnover rate

RABBITS

HVLDL -1DL

VLDL WIDL

-

-LDL

LDL

wHDL

16 5 minutes

24

32

5 mfnutes HOUfS

O-OHDL

40

48

8 5 mlnules

16

24

32

40

48

F&. 4. Decay of radioactivity in tetramethyluxea-soluble apolipoproteins of plasma lipoprotein fractions from normal rabbits when their own 1 z 5 I-labeled VLDL was injected. Tetramethylurea-soluble apolipoProteins Were separated on 10% polyacrylamide gels and stained with Coomassie Brilliant Blue G-250 dye and sliced as described in Fig. 1. Each point is a mean of four experiments.

of VLDL apolipoprotein B to IDL and LDL, there was another pathway for the removal of apolipoprotein B within the first phase and that this pathway removed larger proportions of hypercholesterolemic VLDL apolipoprotein B. This pathway may involve a direct removal of apolipoprotein by the liver in the form of remnants as observed in case of rats [24,25]. Since both the processes were rapid, it was difficult to distinguish them. However, in the case of hypercholesterolemic animals, the radioactivity from hypercholesterolemic VLDL apolipoprotein B decayed much slower than that from normal VLDL apolipoprotein B. Half-life for the first phase of normal VLDL apolipoprotein B catabolism (Table III) (0.9 h) was much shorter than that of hypercholesterolemic VLDL apolipop~tein B (2.5 h). The catabolism of normal VLDL proceeded to a large extent (82%) via the first phase with a rapid appearance of radioactivity in apolipoprotein B of IDL and LDL (Fig. 6). However, in the case of hypercholesterolemic VLDL less than 60% of VLDL apolipoprotein B was catabolized via the first phase. This resulted in much lower fractional catabolic rate for hypercholesterolemic VLDL apolipoprotein B (Table III), like that of LDL in case of cholesterol-fed guinea pigs [26]. Similarly, the fractional catabolic rate of normal VLDL apolipoprotein B injected into the hypercholesterolemic animals (0.12 h-l) was lower than that from normal VLDL injected into the normal recipients (0.37 rt 0.12 h-l) even though the first and second phases had similar half-lives. The only difference was that normal VLDL apolipoprotein B in normal animals (Fig. 3) via the first phase was ca~bol~zed to a greater extent (95%) than that into the hypercholesterolemic animals (82%). Thus, the normal VLDL apolipoprotein B in hypercholesterolemic animals was not completely catabolized via first phase even

187 Hypercholesterolem~c donor

tNormollpidemlc donor

16

I

1

24

C/

-

VLDL

-

IDL LDL

I

8

16

f

24

I

Sminutes

5 m tnutes Hoers

Fig. 6. Decay of radioactivity in apolipoproteln B of plesma lipoprotein iraction~ (VLDL. IDL and LDL) from normal rabbits: A. with own 1311-labeled VLDL, and B. with 1251-&beled VLDL isolated from cholesterol-fed rabbits were injected. The lipoproteins were isolated BP described in Materi& and Methods end the radioactivity in apolipoprotein B wm determined as the difference between total radioactivity in the lipoprotein traction and the lipid plus tetramethyluxeaaoluble radioactivitv. Each point is a mean of two experiments.

though the formation of IDL and LDL was like the normal VLDL into the normal animal. This suggests that the VLDL apolipoprotein B which would have been removed directly as a remnant probably by the liver [24,25] has not been removed due to saturation of this mechanism in hypercholesterolemic rabbits. Therefore more VLDL apolipoprotein B (both normal and hypercholesterolemic) would be catabolized via the slower second phase in these animals. Thus, in hypercholesterolemic rabbits these remnants would accumulate due to the saturation of this pathway. This saturation started to show as early as after 2 days of cholesterol feeding. The first phase of VLDL apolipoprotein B catabolism in case of animals fed cholesterol-rich diet for 2 days was slightly slower (0.64 h) than normal VLDL injected into the normal animals. At the same time slightly lesser VLDL was catabolized via first phase (Table III), with a lesser conversion of VLDL to IDL and LDL resulting in slightly lower fractional catabolic rate. The second phase of VLDL apolipoprotein B catabolism was much slower in all cases and the half-lives for this phase in VLDL were similar to that in IDL

li_

B-~Oynp~tcholesteroiemlc

0-0

VLDL IDL

-

CI

LDL

-LDL

e-0

.,I 0

8

5 minutes

16

24

d

VLDL IDL

t

, 8

16

24

5 mtnutes

Hours Fig. 6. Decay of radioactivity in apolipoprotein B of Plasma lipoprotein fractions (VLDL, IDL, and LDL) from rabbits fed a cholesterol-rich diet for 8 weeks; A. with own 12sI-labeled VLDL and B. when 1311. labeled VLDL isolated from normal rabbits were injected. The lipoproteins were isolated as described in Materials and Methods and the radioactivity ln apollpoprotein B was determined as the difference between total radioactivity in the lipoprotein fraction and tbe lipids plus tetramethylurea-soluble radioactivity. Each point is a mean of four experiments.

and LDL. Thus the catabolism of VLDL apolipoprotein B via this phase may be similar to LDL catabolism mediated by the membrane receptors [Z?]. Weasel et al. [ZS] has calcula~d that LDL degraded by peripheral cells could qu~ti~tively account for all in vivo LDL turnover in man. Thus, VLDL apolipoprotein B, catabolism via the second phase may represent VLDL degradation by the peripheral cells. In the case of hypercholesterolemic rabbits a large proportion of the VLDL apolipoprotein B was catabolized via the second phase. Therefore, the peripheral cells would metabolize large amounts of VLDL which is rich in apolipoprotein E and has been shown to be more atherorgenic [4,18]. This may be the reason why cholesterol-fed rabbits develop atherosclerosis faster. Thus, there may be a relationship between VLDL catabolized via second phase and atherogenesis.

This research was supported by N.H.L.B.I. contract HV-l-2157L and N.I.H. grant HL 18645. W.R.H. is an Investigator of the Howard Hughes Medical

189

Institute. We thank Rita Amalfitano for her skillful technical help, and Laurie Tilden for her secretarial help. References

7 8 0 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 26 26

Garlick. D.G. and Courtice. F.D. (1962) J. EXR. Physiol. 47,211-219 Camejo. G.. Bosch, V., Arreaao, C. and Mender, M.D. (1973) J. Lipid Res. 14,61-68 ROB, A.C. and Zilyemmit, D.B. (1977) J. Lipid Res. 18.169-181 Shore, V-G., Shore, B. and Hart, R.G. (1974) Biochemistry 13.1579-1684 Rodriguez, J.L., GhiseRi, G.D., Torreggiani and Sirtori. C.R. (1976) Atherosclerosis 23.73-83 Shore. B. and Shore, V. (1976) in A~e~~e~~s Drug Discovery (Day, C.E.. ed.), PP. 123-141, Plenum Press. New York Rodriquez, J.L., Catapano. A., GhiseIli, G.C. and Sirtori. C.R. (1976) Atherosclerosis 23.85-Q6 Lowry. OH.. Rosebrougb, N.J.. Fair. A.L. and Randall. R.J. (1961) J. Biol. Chem. 193.265-276 McFarlsne. AS. (1958) Nature 182. 63-64 Bilheimer, D.W., Eisenberg, S. and Lew, R.I. (lQ72) Biochim. Biopbys. Acta 260, 212-224 Kane, J.P. (1973) Anal. Biochem. 63,497-606 Faergeman. 0.. Sata. T.. Kane, J.P. and Havel, R.J. (1976) J. Clin. Invest. 56, 1396-1403 Noble, R.P. (1968) J. Lipid Res, 9.693-700 Lipid Research Clinics Progmm (1974) Manual of Laboratory Operations, Vol, 1. D.H.E.W., Pub&ation No. (N.I.H.) 75628 Bartlett, G.R. (1969) J. Biol. Chem. 234.466468 Matthews, C.M.E. (1967) Phys. Med. Biol. 2.36-63 Mahley, R.W. and Holcombe. K.S. (1977) J. Lipid Res. 12.314-324 MaHey, R.W., Weisgraber, K.H. and Innerarity, T. (1974) Circ. Res. 36,722-733 Mahiey, R.W., Weisgraber. K.H., Inneraritv, T., Brewer, Jr.. H.B. and Amman, G. (1976) Biochemistry 14.2817-2823 HaveI, R.J. and Kane, 3-P. (1973) Froc. NatI. Acad. Sci. U.S. 70,2015-2019 Kushwaha, R.S., Hazzar, W.R.. WahI, P.W. and Hoover, J.J. (1977) Anal. Int. Med. 87. 509-616 Eisenbenc. S. and RacbmiIewitz, D. (1973jBiochim. Biophyr Acfa 326.378-396 Eisenberg. S. and RachmiIewits. D. (1976) J. Lipid Res. 16,341-361 Faergeman, 0. and Havel, R.J. (1976) J. Chn. Invest. 66,121+1218 N&on. R. (1976) Bioehem. Biophys. Res. Commun. 66.6Q-66

Mills. G.L. and McTaggart. F. (1974) in Atherosclerosis III (Schettler. G. and Weizel, A., eds.), pp. 398401, Springer-Verlapl Press. Berlin 27 Brown, M.S. and Goldstein, J.L. (1976) Science 191,16Q-154 28 Weinstein. D.B.. Camw, T.E. and Steinberg, D. (1976) Biochim. Biopbys. Acta 424.404-421