Evidence that two synthetic pathways contribute to the apolipoprotein B pool of the low density lipoprotein fraction of rabbit plasma

311

Biochimica et Biophysics Acta, 7 11 ( 1982) 3 11-3 15 Elsevier Biomedical Press

BBA 5 1095

EVIDENCE THAT TWO SYNTHETIC PATHWAYS CONTRIBUTE TO THE APOLIPOPROTEIN POOL OF THE LOW DENSITY LIPOPROTEIN FRACTION OF RABBIT PLASMA GIANCARLO

GHISELLI

*

Istituto di Farmacologia e Farmacognosia, (Received

October

B

Universita’ degli Studi di Milano, Milan (lta!v)

Sth, 198 I)

Kqv words: Apolipoprolein

B; LDL; Lipoprotein synthesis; (Rabbit plasma)

The plasma specific activity of apolipoprotein B was determined in normal rabbit injected with homologous radioiodinated very low and intermediate density lipoproteins (VLDL and IDL). The specific activity was followed in the same density range of the injected lipoproteins and in two low density lipoprotein subfractions (LDL, and LDL,). VLDL B (d< 1.006 g/ m I) is cleared from plasma mainly in IDL (d= 1.006-1.019 g/ml), with minor radioactivity recovery in LDL, (d= 1.019-1.040 g/ml) and LDL, (d= 1.040-1.063 g/ml) fractions. The main catabolic product of IDL B is, on the other hand, LDL, B. LDL, B, which represent more than 20% of the whole LDL B plasma pool in this animal species, is not derived to a significant extent from either VLDL or IDL and, conceivably, is synthesized independently.

Introduction

only part of the LDL B pool, namely LDL,, is derived from VLDL B; in contrast, a negligible amount of LDL, B comes from this source [6]. The present study was undertaken to determine if an independent LDL B synthesis is also operative in rabbit. This could be of particular significance in evaluating the general control mechanisms of LDL plasma level and in the identification of the apolipoprotein B synthetic pathways in animals.

Apolipoprotein B is the major protein component of very low and low density lipoproteins in all the animal species so far studied [l-3]. It is now firmly established that VLDL B is transferred to the LDL density range in humans [4], and the same mechanism has been described for a number of animals [5-lo]. The fraction of LDL B, which is derived from the VLDL B and from other sources, is, however, variable in different clinical and experimental conditions. Whereas, LDL B apparently is derived totally from VLDL B in normal individuals [ 1 l- 141, a significant fraction of LDL B is synthetized independently in patients with familial hypercholesterolemia [ 14- 161. In rats,

Materials and Methods Normal New Zealand male rabbits (weight 2.83.2 kg) obtained from Charles River Laboratories (Calco, Italy) were used throughout the studies. They were individually caged and allowed free access to tap water; 1OOg of food (Charles River diet for rabbits) were given daily. For all turnover experiments the rabbits were fasted for 12 h prior to blood drawing or injection of labelled lipoproteins. Rabbits were never fasted longer than 24 h in the course of the turnover experiments, as detailed below.

* Present

address: National Institutes of Health, Molecular Disease Branch, Bethesda, MD 20205, U.S.A. Abbreviations: VLDL, very low density lipoproteins; LDL, low density lipoproteins; IDL, intermediate density lipoproteins; HDL, high density lipoproteins; VLDL B, IDL B and LDL B are, respectively, abbreviations for VLDL, IDL and LDL apolipoprotein B.

ooO5-2760/OOOWXWO/82/$02.75

0 1982 Elsevier Biomedical

Press

312

Blood was withdrawn from fasted donors by cardiac puncture. Blood pooled from four animals was collected in iced tubes containing 50 mM Na,EDTA/3 mM NaN, (final concentrations) and centrifuged at 4°C 3000 rpm for 15 min. The VLDL isolations were carried out by ultracentrifugation, in a 50 Ti rotor (Beckman Instrument, Palo Alto, CA), of the pooled plasma, spinning at 40000 rpm for 24 h at 15°C according to the suggested guidelines [ 171. Preliminary experiments have shown that fasted rabbits have a negligible amount of lipoproteins with S, < 400. The IDL were prepared by centrifuging the pooled plasma twice at d = 1.006 g/ml, both times discarding the floating lipoproteins. The density of the 1.006 g/ml infranate was then adjusted to 1.019 g/ml adding calculated volumes of 1.35 g/ml KBr solution. The material floating after a third run at d = 1.O19 g/ml was used for the experiments. The separated lipoprotein fractions were concentrated by ultracentrifugation at their nominal flotation density. The lipoprotein iodination was performed as previously described [ 181. Briefly, the pH of the lipoprotein preparations was adjusted by dialysis against 0.4 M glycine/NaOH buffer at pH 10 for 30 min. To 1 ml of lipoprotein solution, containing 4-10 mg of proteins, was added first 2 mCi of carrier-free I ‘*‘Na (Sorin, Saluggia, Italy) and then an aliquot of 3.3 mM ICI in 2 M NaCl to give a final protein/iodine molar ratio of 0.1. 1 min after the start of the reaction, the iodination mixture was passed over a small column (0.8 X 15 cm) packed with Sephadex G-50 (Pharmacia, Uppsala, Sweden) and eluted with 50 mM NaCl/ Na,EDTA buffer at pH 7.4, at the same salt density of the separated lipoproteins. Iodinated samples were dialyzed against NaCl, 50 mM Na,EDTA and 3 mM NaN, (pH 7.4), adjusted to the desired density for 6 h with four changes of 1 1 of the same dialyzing solution. The labelled lipoproteins were ultracentrifuged once at the appropriate density and finally dialyzed against Krebs-Ringer buffer before injection. 17 and 28% of total apolipoprotein-bound radioactivity was found to be associated with the apolipoprotein B of VLDL and IDL, respectively as determined by SDS-polyacrylamide gel electrophoresis of the delipidated radioiodinated lipoproteins [ 18,191. Each recipient animal received 0.5-2 mg of the labelled

lipoproteins intravenously as a bolus in a 0.5 ml volume. Injected radioactivity was in any case greater than 5 . lo6 cpm per animal. At selected time intervals, blood was collected from the central ear artery of the animals. The subsequent plasma lipoprotein fractionation was carried out on 5-ml samples Na ZEDTA/NaN, added plasma, pooled from two recipient rabbits. VLDL, IDL, LDL, and LDL, were then separated in a 40.3 Beckman rotor spinning each time at 40000 rpm for 22 h at 15°C and adjusting sequentially the working density as previously described. After exhaustive dialysis against buffered saline, the isolated lipoprotein were processed for the apolipoprotein B specific activity by tetramethylurea (Carlo Erba, Milano, Italy) apolipoprotein B precipitation [20]. Tetramethylurea was added (1: 1, v/v) to the samples, and the tube were incubated at 37°C for 2 h. Precipitate was filtered on a layer of celite held in the conical portion of Pasteur pipettes, followed by thorough washing with chloroform/methanol (2: 1). The celite layer was then removed and placed in small test tubes containing 0.3 ml of 1 M NaOH overnight at room temperature. After centrifugation, protein were determinated in duplicate on the supernatants by the method of Lowry et al. [21] against bovine serum albumin standards, followed by radioactivity counting of the reacted protein solutions. Lipoprotein chemical composition was determined as previously described [22] on the ultracentrifugally isolated lipoprotein fractions. Lipoprotein polyacrylamide gel electrophoresis was performed on the same isolated lipoproteins as described by Naito et al. [23]. Results and Discussion The specificity of apolipoprotein B precipitation was tested by polyacrylamide gel electrophoresis performed on the insoluble tetramethylurea precipitated material. No SDS-soluble apolipoproteins other than apolipoprotein B were detected in the lipoprotein precipitate after 2 h of incubation at 37’C with redistilled tetramethylurea. The validity of the lipoprotein separation was determined by polyacrylamide gel electrophoresis on the ultracentrifugally isolated lipoprotein fractions (see Fig. 1). By this procedure, IDL appears as a discrete lipoprotein family in the

313

VLDL

IDL

LDL,

HDL

Fig. I. Polyacrylamide gel electrophoresis of the plasma lipoproteins of normal rabbit. Lipoproteins were isolated by sequential ultracentrifugation, as described in Materials and Methods, from plasma pooled from four rabbits, The dialyzed fractions were stained before running with Sudan black B, as described by Naito et al. [23].

rabbit plasma. No overlap between the LDL, and the HDL fractions was detectable. The cholesterol, triacylglycerols, proteins and apolipoprotein B levels in the different lipoprotein fractions are reported in Table I. Most of the plasma cholesterol in rabbit is carried by HDL: a comparison of the two LDL subfractions indicates that approximately one-fifth of the LDL cholesterol is associated with LDL,. During the kinetic studies the steady-state conditions were verified by checking the mass of

TABLE LIPIDS,

I PROTEIN

Values are expressed Density

apolipoprotein B in the different lipoprotein fractions. The variations were not greater than 10% in any case. Upon VLDL injection, the apolipoprotein B specific activity was followed in the same density range as well as in the IDL, LDL, and LDL, ranges. Examination of the pattern of the apolipoprotein B specific activity in the different isolated lipoproteins (Fig. 2, panels A-D) indicates that by application of the precursor-product relationship criteria [24] VLDL B is a major precursor of IDL B. The apolipoprotein B specific activity of LDL, and LDL, are below the time corresponding points of VLDL and IDL, suggesting that neither LDL, B nor LDL, B are derived significantly from VLDL B. Following the injection of iodinated IDL, the apolipoprotein B specific activity curves of LDL, and LDL, on the other hand, (see Fig2, panels E-H) behave differently. IDL B is, in fact, converted rapidly into LDL, B. The peak of LDL, B specific activity does not, however, occur at the same time it crosses the IDL B specific activity curve, i.e. it does not obey strictly the Zilversmit criteria for a complete precursor-product relationship [24]. This discrepancy, also noted by other authors performing apolipoprotein B kinetics in normal rats [6], may be due to the heterogeneity of the circulating IDL and LDL, or to the lipoprotein conversion delay related to the lipolytic mechanism [14]. The data of these experiments, nevertheless, strongly indicate that neither IDL B nor VLDL B are significant precursors of LDL, B. The catabolism of VLDL to LDL is a complex phenomenon involving several enzyme-controlled

range

AND

APOLIPOPROTEIN

as mg/dl,

mean*

B CONCENTRATION

IN PLASMA

LIPOPROTEINS

OF NORMAL

S.E.; n =4.

Triacylglycerol

Cholesterol

Protein

I& Tetramethylurea-insoluble

17.1*2.8 11.7* 1.1 4.OkO.9 0.9kO.2 5.1kl.O

3.1*0.7 7.4* 1.2 10.3* 1.2 2.5*0.4 23.2* 1.8

3.9kO.6 4.7*0.8 9.4* 1.4 2.5 -to.4 39.9k2.1

27.6 = 47.7 68.7 65.4 n.d.

(g/ml)
RABBITS

a Mean of two determinations.

314

D

2ooI\

I

'OOW

120

1

I

I

4

8

I

I

12 0

I

I

I

8

4

I

I

12

TIME (hours) Fig. 2 Apolipoprotein B specific activity curves of VLDL (A), IDL (O), LDL, (W) and LDL, (0) after injection of ‘251-labelled VLDL (panels A-D) or ‘ZSI-labelled IDL (panels E-H). Each point represents the apolipoprotein B specific activity mean of duplicate determinations. The lipoproteins were isolated from plasma pooled from two rabbits. For the experiments shown in panels was injected into the A and B the same ‘251-labelled VLDL preparation has been utilized. The same ‘251-labelled IDL preparation rabbits for the experiments presented in panels E and F

steps [4].The VLDL heterogeneity has been suggested either from compositional studies [ 17,25,26] or from the catabolic models proposed by various authors [4,14,27]. In normal rabbits a catabolic heterogeneity of the triacylglycerol-rich lipoproteins (VLDL and IDL) is demonstrated here. Based on the analysis of their catabolic products, VLDL and IDL present marked differences. Whereas, in fact, LDL, B is derived significantly from IDL B, the catabolism of VLDL B apparently does not lead efficiently to the formation of LDL B. Different interpretations may be considered to explain this catabolic heterogeneity. A different site of origin may affect the catabolic fate of particles floating at density less than 1.019 g/ml. There are indications, in fact, that the intestinal lipoprotein B pool does not contribute significantly to the

mass of the plasma LDL B pool either directly or after catabolism [28-301. On the other hand, liver-secreted VLDL B may be catabolized totally to LDL B in man, as shown by various authors after injection of protein radioiodinated VLDL drawn from fasted subjects [ 1 l- 141. An alternative interpretation may take into account the large spectrum of lipoprotein remnant particle produced by the activity of the lipoprotein-lipase on the triacylglycerol-rich lipoproteins [31]. In this case the density of the post-lipolytic particles depends on the flotation characteristics of the parent lipoproteins and on the extent of the delipidization in plasma [32]. These studies also indicate that the LDL spectrum is composed of subfractions which derive to a different extent from VLDL and IDL. A minimal

315

portion of LDL, B originates, in fact, from these two lipoproteins. Most of the LDL, B, in contrast, derives from IDL B, a minor fraction coming from VLDL. These findings suggest that in normal rabbit part of the LDL B pool is derived from sources independent of the catabolic pathway of VLDL B. Individual or species differences in the synthesis and the catabolism of the LDL subfractions might have important consequences, since the relative proportion of the LDL B derived from VLDL and from independent synthesis make a significant contribution in determining the level of LDL in plasma [33]. Independent LDL B synthesis has been shown for several animal species [5- 10,33-361 as well as for man [ 14-161. In the rat [6], the independently produced fraction can be recovered in the LDL, density range which is predominant quantitatively in that animal. In rabbit, the independently produced LDL B is also recovered in the LDL, density range, which is, however, less significant. As a general mechanism, it may be hypothesized that a low production of LDL B from VLDL is associated with a higher percentage of LDL, subfraction. In familial hypercholesterolemia, characterized by an absence of an adequate number of functional peripheral LDL receptors, the increased apolipoprotein B-independent synthesis may be regarded as a compensatory physiological production [37]. Consequently, the dissimilarity in the relative proportion of the independent LDL B inputs in human and animals may be a reflection of the individual and species mechanisms for lipoprotein cholesterol transport. References I Goldstein, S., Chapman, M.J. and Mills, G.L. (1977) Atherosclerosis 28, 93-100 2 Chapman, M.J. (1980) J. Lipid Res. 21, 789-853 3 Fredrickson, D.S., Lux, S.E. and Herbert, P.N. (1972) Adv. Exp. Med. Biol. 26, 25-46 4 Eisenberg, S. (1975) Prog. B&hem. Pharmacol. 15, 139- 165 0.. Sata, T., Kane, J.P. and Havel, R.J. (1975) 5 Faergeman, J. Clin. Invest. 56, 1396-1403 6 Fidge, N.H. and Poulis, P. (1978) J. Lipid Res. 19, 342-349 7 Suri, B.S., Targ, M.E. and Robinson, D.S. (1979) Biochem. J. 178, 455-466 8 Weech, P.K. and Mills, G.L. (1978) Biochem. J. 175, 413419 9 Marcel, Y.L., Nestruk, A.C., Bergseth, M., Bidallier, M., Robinson, W.T. and Jeffries, D. (1978) Can. J. B&hem. 56, 963-967 D.R. (1975)Biochim. Biophys. Acta 388, 38-5 1 10 llhngworth,

1 I Sigurdsson,

G., Nicoll,

A. and

Lewis,

B. (1975) J. Chn.

Invest. 56. 1481-1490 12 Reardon, M.F., Fidge, N.H. and Nestel, P.J. (1978) J. &in. Invest. 61, 850-860 H.N. and 13 Le, N.A., Melish, J.S., Roach, B.C., Ginsberg, Brown, W.V. (1978) J. Lipid Res. 19, 578-584 14 Berman, M., Hall, M. Ill, Levy, R.I., Eisenberg, S., BiIheimer, D.W., Phair, R.D. and Goebel, R.H. (1978) J. Lipid Res. 19, 38-56 G.R. (1979) 15 Soutar, A.K., Myant, N.B. and Thompson, Atherosclerosis 32, 315-325 16 Janus, E.D., Nicoll, A., Wootton, R., Turner, P.R., Magill, P.J. and Lewis, B. (1980) Eur. J. Clin. Invest. 10, 149-159 F.T., Jensen, L.C. and Hatch, F.T. (1972) in 17 Lindgren, Blood, Lipids and Lipoproteins (Nelson, G.T., ed.), pp. 18 l-274, Wiley and Sons, New York 18 Rodriguez, J.L., Catapano, A., Ghiselh, G. and Sirtori, C.R. (1976) Atherosclerosis 23, 85-96 19 Laemmh, U.K. (1970) Nature 227, 680-685 20 Kane, J.P. (1973) Anal. B&hem. 53, 350-364 21 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275 22 Rodriguez, J.L., Ghiselli, G., Torreggiani, D. and Sirtori, CR. (1976) Atherosclerosis 23, 73-83 23 Naito, H.K., Wada, M., Ehrhart, L.A. and Lewis, L.A. (1973) Clin. Chem. 19, 228-234 24 Shipley, R.H. and Clark, R.E. (1972) Tracer Methods for In Vivo Kinetics, pp. 93- 109, Academic Press, New York S.H., Nathans, A., Dowdee, M. and Hilderman, 25 Quarfordt, H.L. (1972) J. Lipid Res. 13, 435-444 26 Sata, T., Havel, R.J. and Jones, A.L. (1972)J. Lipid Res. 13, 757-768 27 Fisher, W.R., Zech, L.A., Bardalaye, P., Warmke, G. and Berman, M. (1980) J. Lipid Res. 21, 760-774 28 Sparks, C.E. and Marsh, J.B. (1981) J. Lipid Res. 22, 5 19-527 29 Schaefer, E.J., Jenkins, L.L. and Brewer, H.B., Jr. (1978) B&hem. Biophys. Res. Comm. 80, 405-412 30 Ross, A.C. and Zilversmit, D.B. (1977) J. Lipid Res. 18, 169-181 31 Higgins, J.M. and Fielding, C.J. (1975) Biochemistry 14, 2288-2293 32 Catapano, A.L., Kinnunen, P.K.J., Packard, CL., Gotto, A.M., Jr. and Smith, L.C. (1978) in International Conference on Atherosclerosis (Carlson, L.A., Paoletti, R., Sirtori, C.R. and Weber, G., eds.), pp. 315-318, Raven Press, New York 33 Ghiselli, G. and Sirtori, C.R. (1980) in Drugs Affecting Lipid Metabolism (Fumagalh, R., Kritchevsky, D. and Paoletti, R., eds.), pp. 243-250, Elsevier/North-Holland Biochemical Press, New York 34 Nakaya, N., Chung, B.H. and Taunton, O.D. (1977) J. Biol. Chem. 252, 5258-5261 35 Heimberg, D.M., Weinstein, I., Dishmon, G. and Fried, M. (1967) Am. J. Physiol. 209, 1053-1060 M.J., Mills, G.L. and Taylaur, C.E. (1973) Bio36 Chapman, them. J. 131, 177-185 37 Brown, M.S. and Goldstein, J.L. (1976) New Engl. J. Med. 294, 138661390