The plasma and tissue turnover and distribution of two radio-iodine-labelled pig plasma low density lipoproteins

Atherosclerosis,

601

22 (1975) 601-628

0 Elsevier Scientific Publishing Company, Amsterdam - Printed in The Netherlands

THE

PLASMA

AND

TISSUE

RADIO-IODINE-LABELLED

G. D. CALVERT, P. J. SCOTT

TURNOVER PIG PLASMA

AND

AND

DISTRIBUTION

LOW DENSITY

OF TWO

LIPOPROTEINS

D. N. SHARPE

Department of Medicine (P.J.S. and D.N.S.), University of Auckland School of Medicine, Auckland (New Zealand), and Department of Clinical Chemistry (C.D.C.), Queen Elizabeth Medical Centre, Edgbaston, Birmingham B15 2TH (Great Britain)

(Revised, received March 14th, 1975) (Accepted April 2nd, 1975)

SUMMARY

Two classes of pig plasma low density lipoprotein (LDh and LDL2) different densities and molecular sizes were isolated by zonal ultracentrifugation

with and

were further purified by flotation. The peptide component was iodinated with 1251, and the labelled lipoprotein was injected intravenously. 1251-LDLl turnover studies were performed on 22 3-4 month old female Large White pigs, and 1251-LDLz turnover studies on 4 similar pigs. A biological screening experiment confirmed that the shape of the plasma activity curve was not a function of protein denaturation. The pattern of radioactivity decline in plasma was not affected by the degree of LDL iodination.

1251-LDL1 turnover The curve of plasma radioactivity plotted against time over the first 5 days after injection could be resolved into two exponentials. The plasma biological half-life (T+) was calculated from the slower exponential predominant from the second day. The mean Td over 2-5 days was 22.9 hr (range 17.2-28.5 hr). Multicompartmental analysis of the plasma decay curve using an open mammillary model gave a mean fractional catabolic rate per day for LDL1 of 1.4 (range 0.9-1.9). The mean T, was 0.26-0.31 times and the fractional catabolic rate 3.0-3.9 times those values found in two studies on adult humans. The tissue distribution of 1251 was analysed in a series of 20 animals killed from 1.0 to 33.8 days after 1251-LDL1 injection. Most tissue 1251 (86-89%) was protein

This work was supported by grants from the New Zealand Medical Research Council, the National Heart Foundation of New Zealand, the Life Insurance New Zealand and the Birmingham Regional Hospital Board.

Medical Fund of Australia

and

602

G. D. CALVERT,

P. J. SCOTT

AND D. N. SHARPE

bound. An appropriate correction was made to the 1251 counts for retained plasma in liver and spleen (using tslI-albumin); retained plasma in other tissues was negligible. Highest lz51 tissue levels were found in the liver, supporting other evidence that the liver may be the major site of LDLl catabolism. After 2.06 and 4.06 days the livers in two animals contained I .6”/, and 0.7 y0 respectively of the total injected i251, equal to 33 ‘A and 54 y0 of the total plasma lz51 at those times. The skin contained about one-third to one-ninth the 1251 in the liver at various times. Distribution in other organs was quantitatively minimal. Higher levels of radioactivity were found in the intima and inner media of the aorta than in the outer media. These results suggest that plasma LDL in the pig diffuses

through

confirmed

the endothelial

surface

into

the arterial

wall. These

findings

are

by autoradiography.

1251-LDLa turnover Parallel studies of plasma i2jI-LDLs turnover and tissue distribution were performed. The plasma biological decay curve was multi-exponential, suggesting that LDLa metabolism is complex, and possibly more rapid than that of LDLl (LDLe is smaller and denser than LDLr). The tissue distribution of 1251-LDLz in these pigs was very similar to that of 1251-LDLr. As LDLl and LDLs differ in the amount of lipid they contain, they may have different roles to play in lipid transport, and there may be interconversion of one into the other at different

Key words:

sites. This hypothesis

remains

conjectural.

Iodine labelling - Low density lipoprotein - Plasma catabolism - Swine - Tissue protein distribution

turnover - Protein

INTRODUCTION

Cholesterol-rich

low density

the pathogenesis of atherosclerosis protein is labelled with radio-iodine

lipoprotein (LDL) is held to be a major factor in 1J. Experiments in man using LDL in which the are limited by several factors.

The radiation

dose

must be small, the analysis must generally be confined to the distribution of labelled LDL in plasma and urine with tissue biopsy excluded, and the manoeuvres that might affect LDL metabolism are limited to those risk-free procedures which are acceptable to the subject. The pig presents a number of advantages in atherosclerosis research3. In particular, experiments using 1251-labelled LDL may be extended by increasing the size of the radiation dose, analysis of tissue radioactivity, and by dietary and drug manipulations. The porcine serum lipoproteins in the density class 1.006-l .063 g/ml (LDL) have been describedd-7. This fraction includes two lipoproteins of different densities,

PIG PLASMA

migrating

LIPOPROTEIN

TURNOVER

IN PLASMA

AND

at Sf 4.9 and Sf 1.8 in the analytical

TISSUES

603

ultracentrifuge4s5,

named

LDLr and

LDLz respectively. Previous studies5 suggest that the apoprotein moieties of LDLl and LDLz are identical, but the proportions of lipid in these molecule&s and their densities

and sizes@ differ. The modal

densities

of LDLr and LDLs are 1.035 g/ml

and 1.050 g/ml, and the molecular diameters 217 A and 195 A respectivelys. Use of the zonal ultracentrifuge allows preparation of large amounts of LDLl and LDL$. This purified

paper

describes

the serum

and

LDLI and LDLz after intravascular

tissue

distributions

injections

of radio-iodinated

in pigs.

METHODS

LDL preparation All ultracentrifugation and LDLs using a Beckman

was performed at 4°C. Methods of preparation of LDLl Spinco Ti15 zonal rotor in a Spinco L2-65B ultracentri-

fuge have already been describeds. After preparation with the zonal rotor the lipoprotein solution was layered carefully under at least an equal volume of NaCl solution (density 1.063 g/ml) and spun at 40,000 rpm for 18-24 hr in a Spinco type 40 rotor or 16 hr at 50,000 rpm in a Spinco Ti60 rotor. The lipoprotein layer at the top of the tube was carefully aspirated using a Pasteur pipette, and found to be free of albumin by immunoelectrophoresis. For one turnover study (on pig 38) LDLl was prepared in a fixed angle rotor. Plasma was layered under an equal volume of NaCl, density 1.006 g/ml, and spun in a Spinco type 30 rotor at 29,000 rpm at 4°C for 12 hr. The very low density lipoprotein (VLDL), clearly visible at the top of the tube, was carefully aspirated with a Pasteur pipette, together with most of the underlying NaCl solution. The smallmolecule density of the remaining plasma was adjusted to 1.05 g/ml with a solution of NaCl-KBr. This was overlaid with two volumes of NaCl-KBr solution, density 1.05 g/ml, and spun at 45,000 rpm in a Spinco

Ti60 rotor

at 4°C for 24 hr. The top

1 cm layer of LDLl was carefully aspirated with a Pasteur pipette, and the small molecule density raised to 1.063 g/ml with solid NaCl. This solution was overlaid with an equal volume of NaCl solution, density 1.063 g/ml, and spun in a Spinco Ti60 rotor at 4°C and 55,000 rpm for 10 hr. The top 0.5 cm layer of lipoprotein then aspirated and labelled with 1251.

was

Chemical estimations of lipoprotein fractions Protein content was estimated by the method of Lowry et aLa, using bovine serum albumin as a standard. Because of the low LDL concentrations it was not necessary to remove turbidity due to lipid during this estimation. Cholesterol9 and triglyceride10 estimations were carried out by standard techniques after isopropanol extraction. Electrophoresis was performed on papertl, polyacrylamide gelz4 or cellulose acetate12 and immunoelectrophoresis on cellulose acetate (Cellogel)12, using serum from New Zealand White rabbits immunised against pig albumin or pig serum as the source of antibody. All density measurements were made with a Westphal

604

G.

specific

gravity

balance

(Griffin

and

D. CALVERT,

George,

London)

P. J. SCOTT AND

at 20°C

D. N. SHARPE

or, on pure

salt

solutions, with an Abbe high performance refractometer, interpolating graphically between values of density and refractive index given in standard tablesls. Labeling This was carried out within

1-4 hr of the period of ultracentrifugation

described

above, using a modification of an ICl method of McFarlane14. Two to four 6 ml samples of washed lipoprotein solutions were each mixed with 0.5 ml of NH40H NH&l

buffer (pH 9.0), and labelled

separately.

Approximately

2-3 mCi of carrier-

free rz51 (prepared for protein iodination, The Radiochemical Centre, Amersham, England) per 6 ml lipoprotein sample was used for nearly all labellings, though up to 6.5 mCi per labelling was used in three animals (pigs 17, 19 and 20, all 24 hr studies)

to provide

plasma others.

radioactivity

high grain counts

for autoradiographic

in the first 24 hr was no greater

purposes.

The decline in

in these three animals

than in

The 12sI was added to a calculated amount of 6.2 x 10-s M ICI solution (for calculation see below). This solution was then mixed with 0.5 ml NHdOH - NH&l buffer and immediately jetted into the buffered lipoprotein mixture, using a disposable syringe with a wide bore needle, with constant manual agitation of the LDL solution. The resulting mixture was allowed to stand for 8 min, after which 0.1 ml 1.66 M KI solution was added. The mixture was shaken and left a further 15 min, after which it was passed slowly (4-5 min) down a 229 x 13 mm column containing chloridecharged solution aliquots paration

‘Amberlite’ IRA-400 resin to absorb “free was then passed down the column to elute were taken the collected rz51-LDL solutions took place under sterile conditions, and

iodide”.

Seven ml 0.15 M NaCl

any remaining

LDL and after

were usually pooled. This presolutions were maintained at

45°C except for the passage down the ion exchange column at room temperature, process taking 10 to 15 min. Aliquots were taken for estimation of labelling efficiency (the proportion

a of

lz51 bound to LDL), and for precipitation with an equal volume of 20% trichloroacetic acid at 4°C to calculate the non-protein bound “free iodide”. The proportion of lz51 attached to lipid was estimated after lipid in an 125I-LDL aliquot had been exextract was then washed tracted by the method of Folch et al. 15. The chloroform four times with 0.1 A4 NaCl, and a portion counted. The nature of the lipid-lz51 complex was not investigated; Langer et aZ.16, using a similar method to label human LDL, reported that most lipid-bound radio-iodine was attached to phospholipid. Basic data for the LDLr and LDLs turnover studies are shown in Tables 1 and 7 respectively. LDL iodination was performed with labelling efficiencies from 12.1 % to 78.9 % (mean 41%). Two per cent or less 1251 was bound to lipid. “Free iodide”, 1251, averaged 6.8%. Analysis of results i.e. non-trichloroacetic acid-precipitated suggested that this was a rather unsatisfactory method of calculating “free iodide”, giving variable results probably due to 1251 release from precipitated LDL. Others have found that trichloroacetic acid precipitation released a variable amount of

605

PIG PLASMA LIPOPROTEIN TURNOVER IN PLASMA AND TISSUES TABLE BASIC

Animal

1

DATA

No.

LDLl

FOR

Labelling efficiency

TURNOVER

( %)a

STUDIES

lz51 ( % bound to lipid)”

Free iodide”

Duration of study (days)

Weight at death (kg)

17

67

1.0

20

19

50

1.0

20

20

43

-

1.0

20

21 22

36

5.8

33.8 33.8

20 21

23 24

49

-

11.5

27.9 25.9

21 14

25

57

0.8

4.8

4.0

25

26

79

1.2

4.6

4.8

18

29 30

46

0.3

2.7

18.8 18.8

11.5 10.5

31 32 33 34

42

0.7

5.6

9.9 12.9 12.9 3.8

12.5 11.5 11.5 14.5

35 36

31

2.0

8.4

13.1 13.1

11

38

28

-

5.3

7.9

8

39

12

0.3

10.9

4.06

17

40

19

0.2

12.8

2.06

20.5

15.5

a In some studies mean values for labelling efficiency, free iodide and i251 bound to lipid are given. In these studies separately labelled samples of LDLl were pooled and mixed before being divided into equal portions for the individual turnover studies. A dash indicates that data were not available.

radio-iodine

from labelled

fibrinogen17

and albuminls.

Some of our samples were left

in contact with trichloroacetic acid for over an hour before “free iodide” was measured, in retrospect a procedural mistake. Immediately after labelling porcine plasma was added to the lz51-LDL solutions (sometimes with additional albumin) to give a protein concentration of about 4 g/100 ml. In this way radiation damage to the labelled protein was minimised. The lipoprotein solution in a sterile glass container was kept in ice and taken as quickly as possible to be injected into the test pigs. Injection took place within an hour after labelling. Calculations Calculation

of the degree of substitution

of iodine atoms per molecule

of lipo-

606

G. D.

CALVERT,

I’. J. SCOTT AND D. N. SHARPE

protein was hampered by a lack of knowledge of the lipoproteinsubunit structure. Human LDL may be composed of 20 subunits lg. As the molecular weight of human LDL is about 2.2-2.3 x lo6 daltonGO, and the protein content 22-24x the “molecular weight” of the protein component of each subunit may be about daltons. Apo-LDL in 8 M urea has been considered to have a molecular 2.75 x 10” daltonslQ, an even lower subunit

though

this figure may represent

molecular

a minor

2.6 x lo* weight of

component21,

and

weight of 8,000 daltons

has been suggesteds2. Pig LDLr and LDL2 have molecular weights of about 2.7 x 106 and 2.0 x lo6 daltons respectively4. For the purpose of rough calculation of the iodine substitution ratio we have taken the molecular weight of a porcine LDL protein subunit to be 2.75 x lo4 daltons. The data on labelling for LDLr and LDLQ studies are shown in Tables 1 and 7 respectively. Checks were made that the radioactivity

was in the appropriate

LDL fraction

by using paper and polyacrylamide gel electrophoresis and gradient ultracentrifugation. In both LDLr and LDLQ turnover studies radioactivity was counted in 1 cm bands cut from paper almost all radioactivity see Fig.

electrophoretic strips. During migrated with beta-lipoprotein

1). In the longer term studies it was apparent

plasma biological decay curve after radioactivity to non beta-migrating

8 days coincided protein fractions,

the first 8 days of the studies (for three typical experiments that a change

in slope in the

with a relative shift of the and data from this latter

Fig. 1. Radioactivity in 1 cm bands cut from paper electrophoretic strips. The strips were stained with Oil Red 0. a: pig 39, showing radioactivity in /? band during the first 2 days of an 1251-LDL1 turnover study.

PIG PLASMA LIPOPROTEIN TURNOVER IN PLASMA AND TISSUES

607

PIG 25

~ ol,’

* 33.8 DAYS PIG 22

Fig. lb: pig 25 (upper), showing radioactivity in the /3 band during the first 4 days of an lz51-LDL1 turnover study. The sample from pig 22, at 33.8 days, confirms that the radioactivity is not predominantly in the /3 band after the first 9-10 days of an 12jI-LDLl study (see Fig. 3).

period were not used to calculate turnover parameters. The radioactivity in serum after 8 days was not predominantly associated with thyroid hormone+. Its nature was not fully characterised, as the plasma specific activity after 10 days was about onethousandth that at zero time. Two LDLl turnover studies (pigs 81 and 82) were further analysed in order to show that the lz51 remained in the LDLl fraction. In each of these (not shown else-

608

G.

% ‘s’^”

804

4

D. CALVERT,

P. J. SCOTT AND D. N. SHARPE

ACTIVITY 0 DAYS

80

0.88 DAYS

l-77 DAYS

8d

4.00 DAYS

60 40

~0 B a( Fig. lc: pig 27, showing radioactivity in the /!Iband during the first 4 days in an 1251-LDL~ turnover study. Radioactivity is not as sharply localised to the p band as in LDLl studies.

where in results) approximately 650 ,uCi 1251-LDLi was injected intravenously. The labelling and turnover characteristics were similar to those of our other studies. Serum was taken on each of the first 8 days and polyacrylamide gel electrophoresis performed after the lipoprotein had been stained with acetylated Sudan Black (George T. Gurr)24. LDLi and LDLs were clearly separated (see ref. 5) in this case by 2-2.5 mm. Gels were carefully washed with 7 % acetic acid and sliced, and the 1251activity in slices, with and without lipoprotein fractions, counted. The injected “1251-LDLi” contained 12.8 ‘A 1251-LDLs estimated by this method. The proportions of radioactivity in the LDLs fraction in plasma from the two pigs are set out in Table 2. It can be seen that the proportion of radioactivity in the LDLs fraction initially declined, though later in the studies it increased. This may be a manifestation of the complex 1251-LDL~ plasma decay curve obtained from four different animals (see Fig. 5), with a rapid initial fall in activity and a slower fall later in the studies. The serum samples analysed in Table 2 were also subjected to density gradient ultracentrifugation. One ml samples were brought to small-molecule density 1.23 g/ml with solid KBr25, then layered under solutions of KBr d 1.21 g/ml (3 ml) and

PIG PLASMA LIPOPROTEIN TURNOVER IN PLASMA AND TISSUES TABLE

609

2

THEPROPORTIONSOFPLASMARADI~ACTIVIT~INTHE GEL ELECTROPHORESIS)IN

TWO

“LDLl”

TURNOVER

LDL~ FRACTION(SEPARATEDBYPOL~ACRYLAMIDE STUDIES

(Expressed as % total LDL radioactivity) Time (hr) from zero

Pig 81

Pig 82

0 0.25 1.0 24 48 69 95 119 166

12.8 12.1 9.2 4.9 3.3 2.4 6.4 11.3 13.9

12.8 12.0 8.1 4.1 2.0 1.7 4.4 12.7

NaCl d 1.063 g/ml (2 ml), 1.045 g/ml (2 ml), 1.019 g/ml (1.5 ml) and 1.006 g/ml (2.5-3.0 ml) sequentially in 13 ml cellulose nitrate centrifuge tubes. These were placed in a Beckman SW 41 Ti rotor and spun at 40,000 rpm in a Beckman L5 65 ultracentrifuge for 44 hr at 20°C. The tubes were pierced at the bottom end and the contents pumped out upwards through a Gilford density gradient scanner, absorbance at 280 nm was recorded, and 0.4 ml fractions collected. VLDL, HDL, and the KBr containing dense plasma proteins

emerged as separate peaks. LDLr was a clear peak between

VLDL and HDL, with LDLz emerging either as a denser separate peak or as a shoulder on the LDLr peak. Radioactivity was mostly in the LDLr peak, but this trailed off through the denser fractions distribution of radioactivity

into the LDLa peak in a skewed distribution. The skew was such that a definite fractionation into LDLr- and

LDLz-associated activity could not be made, though those obtained with polyacrylamide gel electrophoresis. Experimental Female

animals Large White

pigs, aged 3-4 months

results

were consistent

were used. The animals

with

weighed

between 8-12.5 kg when purchased and between 8-21 kg at death (see Tables 1 and 7). All pigs were fed daily; sufficient food was given to ensure total or almost total consumption, and water was freely available. A standard mash diet (Northern Roller Mills, Auckland, New Zealand, weaner-grower mash) containing 4.5-4.9 % animal tallow, 24.5 % protein and 71% vegetable products was given, supplemented at intervals with raw cabbage or carrot. On this regimen the mean fasting serum cholesterol concentration was 63 i 17 (Jo SD) mg/lOO ml; triglyceride concentrations were of the same order, 30-100 mg/ 100 ml but whereas the serum cholesterol altered very little in the individual animal, triglyceride concentrations fluctuated considerably. No pre-B band was visible on regular lipoprotein paper electrophoresis. No significant trend occurred in serum

610

G. D. CALVERT,

cholesterol

or triglyceride

during

the turnover

P. J. SCOTT AND D. N. SHARPE

experiments.

Body weight

increased

steadily about 2 kg per week in most pigs, though two (pigs 24 and 38) failed to grow more than 0.5 kg per week during the experiments. Though we recognised that these studies were carried out on animals with an expanding plasma volume, as the weight increase

was only about

3 ‘A of body weight per T+ of plasma

we did not apply any correction for growth to our data. Thyroid uptake of radio-iodine was blocked by intramuscular

radioactivity,

injection

of 50 mg

NaI in 1% solution 24 hr before a study began, again immediately before injection of labelled lipoprotein and daily thereafter. This was effective, as we were unable to demonstrate radio-labelled circulating thyroid hormones with an ion exchange resin method23

during

pig LDL turnover

studies.

Nevertheless

there was some 1251

in the thyroid of pigs at post mortem, which on autoradiography seemed to be around the follicular walls but not in the colloid. The pigs were restrained supine on a trestle and labelled lipoprotein was injected into the anterior vena cava through a 64 mm 18-20 gauge hypodermic needle. The needle was inserted at an angle of approximately 45” to the skin, in the midline just rostra1 to the sternum, and venous blood from the anterior vena cava was drawn back into the syringe before and after injection to check that it was intravascular. The shape of the plasma

decay curve over the 24 hr following

further

or not it was intravascular. was generally withdrawn

check on whether Blood for counting

from

injection

provided

a

ear veins or, when these

proved too difficult, from the left ventricle by percutaneous puncture. Blood was taken into test tubes containing disodium EDTA (1 mg/ml blood), and plasma was separated and stored at 4°C until radioactivity was counted. Animals were given heparin just before death and killed with an intra-cardiac injection of pentobarbitone and succinylcholine sufficient to cause immediate and complete loss of consciousness. After cardiac asystole they were suspended in a headdown position and drained of as much blood as possible. In order to check that the turnover data were not influenced by iz51-LDL denaturation a biological screening experiment was performed. 350 ml heparinised blood was aspirated under anaesthesia from a donor animal 4 days after injection of 1251-LDLi. 200 ml plasma was separated and injected intravenously into a recipient animal (pig 44), and plasma radioactivity followed over the next few days. The biological decay curve was identical to that in the donor animal and the curves were within the normal range observed in the study. In pigs 39 and 40 an estimate of the total plasma volume and the volume remaining in the tissue after washing was made by intravenous injection 15 min before death of an accurately weighed amount of iarI-human serum albumin. Analysis of tissues and plasma Post mortem examination was begun within an hour after death. The pigs were weighed at death and in all later studies the major internal organs were weighed. Specimens (Table 3) were taken for radioactivity counting and histological exa-

PIG PLASMA LIPOPROTEIN TURNOVER IN PLASMA AND TISSUES

611

TABLE 3 TISSUE SPECIMENS

TAKEN

Skin Striated muscle Fat Liver Spleen Kidney Lung Lymph nodes Thyroid Salivary glands Nerve Brain (1 animal only) Heart Aorta Pulmonary artery Inferior vena cava

AT POST-MORTEM

EXAMINATION

Upper abdomen, adjacent to midline Upper rectus abdominis Overlying upper rectus abdominis Anterior border right lobe (in addition in some animals middle of left lobe) Tail Either pole Periphery Para-aortic and supraclavicular (sometimes also hilar) In tot0 Submaxillary Brachial plexus Mid-temporal cerebral cortex, stripped of leptomeninges, and pons Apex of left ventricle In toto, from aortic valves to just distal to the bifurcation Right ventricle to bifurcation Rostra1 portion

mination (including autoradiography). Specimens were washed under running tap water to remove as much blood as possible and dabbed dry without excessive rubbing with a clean paper towel. Excess fat, connective tissue and blood vessels were removed, as was loose adventitia around the aorta and pulmonary artery. Tissue specimens were fixed in CPC formalin. divided at the origin of the fourth intercostal artery

The aorta, after fixation, was and just above the coeliac axis

artery, to provide arch, descending thoracic and abdominal segments. Sections of aorta for histology, autoradiography and radioactive counting were taken, and the section for counting was cut, with clean scissors, into small pieces. The intima and inner third of the media were stripped

from the outer two-thirds

of the media, using

magnifying lenses for accurate control, at a natural plane of cleavage. The method was identical to that used in human studies 26, though the inner layer of the aorta in the pig is a little thinner than the corresponding layer in the human. Plasma and tissue radioactivity counting was done on a Packard Selektronic Gamma Counter. Counting times were such that the potential errors from counting did not exceed 2 % (in most cases radioactivity was such that errors were much less). In the two animals injected with both 125I-LDL and 1311-serum albumin (pigs 39 and 40), gate settings used ensured that there was no contamination of the lstI-radioactivity spectrum by 1251; the minor degree of 1251 spectrum contamination by Is11 was measured, using standard solutions, and an appropriate correction made. Counts were performed on exactly 1.O ml plasma, in plastic disposable counting tubes. Tissue for counting was adjusted so that the volume of all samples wa sapproximately 1.0 ml. Preliminary experiments showed that the effect of increases in volume of equally packed tissue samples in the range 1-2 ml on the count per gram tissue was under 5 7:.

612

G. D. CALVERT,

Aortic wall was homogenised

and precipitated

P. J. SCOTT AND D. N. SHARPE

with trichloroacetic

acid 24 or 48

hr after 1251-LDLr injection. 89 % (1 animal, 48 hr after injection) or 88 % (4 animals, 24 hr after injection) of lz51 was thus precipitable. 86% of the radioactivity in an aortic homogenate was not removed by 24 hr dialysis in 3 changes of 0.15 A4 NaCl solution containing 0.1 % disodium EDTA and 0.05 % NaI. The greater part of tissue radioactivity therefore was attached to protein, and did not simply reflect an influx of free lz51 into tissues. Tissue counts (1251-LDL) per gram tissue were calculated as a proportion of plasma counts per ml plasma at death, and also as a proportion of plasma counts 15 min after intravascular injection. Counts for liver and spleen were adjusted to allow for the effect of contained

plasma.

The magnitude

of this adjustment

was small;

washed liver and spleen both contained on average under 0.055 ml plasma (0.0520.062 ml and 0.052-0.059 ml respectively) per gram tissue. No adjustment was made for other tissues as the amounts involved were so small (e.g. washed skin, fat, muscle and nerve contained

0.007 ml, 0.003 ml, 0.005 ml and 0.008 ml plasma

respectively

per gram tissue). We attempted at some length to demonstrate the immunological identity of LDL-apoprotein with the 125I-labelled protein present in post mortem tissue specimens, using the tissue immunoelectrophoresis method of Smith and Slaterz7. We failed to demonstrate this identity conclusively. This may have been due to several factors; for instance, tissue apo-LDL may have become immunologically unreactive, or it may have become fixed in the tissue, e.g. bound to mucopolysaccharide ground substance. No free tissue lipid was visible in frozen sections of vessel walls stained with Oil Red 0. As soon as lipid is removed, apo-LDL (apoprotein B) becomes insoluble in aqueous buffer+, and (cJ the atherosclerotic human aortae studied by Dr. Smith) this lack of tissue lipid may explain our negative findings. Professor K. W. Walton (personal communication, 1975) has work suggesting that only a labile portion of artery wall LDL can be removed by electrophoresis, and further work is under way with him to clarify this point. Other workers2s,28-32 have doculipoproteins as tissue r2sI, and, while they mented the tissue uptake of rssI-labelled have likewise not conclusively demonstrated identity, the weight of published evidence (using radio-iodinated lipoproteins, lipoproteins labelled in the lipid moiety30,33,31, and immunofluorescentas@ work) suggests that uptake of and immunological27 plasma lipoproteins into tissues occurs. Free iodide distribution

was deterThe plasma and tissue distribution of free rssI- injected intravenously mined in two pigs, one killed I hr after injection and the other after 24 hr. RESULTS

LDLl plasma turnover

The changes

in plasma

activity

with time were plotted

on a semi-logarithmic

PIG PLASMA LIPOPROTEIN TURNOVER IN PLASMA AND TISSUES

613

Fig. 2. The changes in plasma activity in iz51-LDLr studies. The mean & 1 SD of the plasma activity as a % of activity at zero time for all animals each day is plotted. The period after 5 days was not used for analysis. 100

. PIG 39 *PIG 40

Fig. 3. lz51-LDLr turnover in two pigs showing that close analysis of serum decay curves reveais their multi-exponential nature. The ordinate is as for Fig. 2.

614

G. D. CALVERT, P. J. SCOTT AND D. N. SHARPE

TABLE 4 CHANGES

IN PLASMA1251-LDL1

IN THE FIRST

1%)

MIN AFTER

INTRAVENOUS

INJECTION

IN TWO

PIGS

An equal volume of the same labelled preparation was injected into each pig. Pig 39 weighed 16 kg and pig 40 19.5 kg. All blood specimens were taken by cardiac puncture. Time after injection (min)

5 10 15 20 40 50 60 90 120 150

% Zero counts pig 39

pig 40

100 88.1 85.1 84.4 83.7 73.5 71.7 67.1 61.9 59.5

100 88.7 85.5 85.6 83.4 70.5 63.8 55.2 54.3 52.8

scale. The initial steep drop in plasma specific activity in the first 24 hr was a constant (as demonstrated feature of all studies. This was not due to 1251-LDL denaturation in the biological screening experiment), and was presumably due to lz51-LDL equilibration in tissues. The decay curves are combined in Fig. 2. Although in studies when daily blood specimens were taken a mono-exponential decline in plasma specific activity seemed to begin after the first day, closer analysis of the first hours of two LDLi turnover studies (Fig. 3) suggested that a monoexponential decline was not apparent until 2 days had passed. The results from the first 150 min of these turnover studies are shown in Table 4. The T, was calculated from days 2-5 inclusive. The decay curves were monoexponential during this period, and were clearly shown to be due largely to lz51-LDLt turnover (Fig. 1, and see validation in METHODS section). The mean T* of the main exponential (2-5 days) was 22.9 hr (range 17.2-28.5 hr). The biological half-life (T+) was unaffected by the degree of protein iodination used, as shown in Table 5. The biological decay curve was similar when LDLl was prepared by standard preparative ultracentrifugal techniques, and when LDLl had been “biologically screened” by prior injection into another animal. One ml samples of plasma from several experimental animals were put down a Sephadex G-100 column (e.g. Fig. 4). There was good separation between the main protein peak containing most radioactivity and the “free iodide” peak and the latter was less than 10 % (usually about 9 %) of the total plasma radioactivity in the 2-5 day period of the study. The limitations of the turnover data restricted the choice of available analyses37. No reliable urine collections were possible (though in future the use of metabolic

615

PIG PLASMA LIPOPROTEIN TURNOVER IN PLASMA AND TISSUES

TABLE LACK

5

OF CORRELATION

BETWEEN

DEGREE

OF

LDLl

IODINATION

AND BIOLOGICAL

HALF-LIFE

In some experiments two or more samples of iodinated LDL were pooled; the mean figures for the iodine/protein ratio are given in these cases. Pig No.

I/protein

39 38 35 36 31 32 34 21 22 29 30 23 24 25 26

0.21 0.6 ;

ratio

23.9 21.0 17.7 24.6 17.2 17.2 22.4 26.9 27.0 20.6 22.3 28.5 21.4 24.0 22.8

0.82

1 ‘ 1.15 \ ( 2.2

1 i ,

Tb (hr)

3.15 6.7 8.5 9.5

Mean 22.9

PIG 22

SERUM

4.0 DAYS : SEPHADEX RADlOACTlVlTY:

0.3-

IIq

ABSORBANCE

-

G-100.

- - -

: -

-3000

,‘\

L~~ll.l111111111 0 10 20 30

0 40

FRACTION

50

60

70

60

NO.

Fig. 4. Separation of ‘aSI-LDL1 and “free iodide” from plasma taken during a turnover study. Absorbance at 280 nm and radioactivity in 3.4 ml fractions obtained from a 450 x 15 mm column of Sephadex G-100. A 1 ml sample of plasma from Pig 22 on the 4th day of the turnover study was eluted with Tris-HCl buffer containing 0.002 M sodium dodecyl sulphate at pH 8.2. The total radioactivity of the “free iodide” peak (in the salt volume, using CuSO4) is approximately 9.2% that of the plasma.

616

G. D. CALVERT, P. J. SCOTT AND D. N. SHARPE

TABLE 6 MATTHEWSS8

MULTICOMPARTMENTAL

ANALYSIS

OF

LDLl

TURNOVER

DATA

Derived from changes in plasma specific activity, assuming one intravascular compartment. Pig No.

Tt main exponential (hr)

I&Z (fractional catabolic rate per day)

21 22 23 24 25 29 30 32 34 36 38 39

26.9 27.0 28.5 27.4 24.0 20.6 22.3 17.2 22.4 24.6 21.0 23.9

1.5 1.9 1.3 1.2 0.9

Mean

23.8

1.4

Range

17.2-28.5

0.9-I .9

and one extravascular

1.1 1.3 1.8 1.1 1.7 1.3 1.9

cages may allow some data on urine isotope excretion), nor available. Multicompartmental analysis38 seemed the most though the rapidity of LDL turnover and the consequent and relative delay in iodide excretion must lead to some

was a whole-body counter suitable available system, high plasma “free iodide” error in the results. Other

methods of analysis taking into account the iodide space seemed complex3g~40 and inaccurate in our experimental circumstances41*42. The results of multicompartmental analysis of the data are shown in Table 6. All curves were regarded as the sum of two exponentials. The mean calculated fractional

catabolic

LDLz plasma

rate was 1.4 per day (range 0.9-1.9). turnover

Basic data for the LDLz turnover

studies

are shown

in Table

7. The plasma

TABLE I BASIC

DATA

Animal No.

FROM

l=I-LDLz

Label&g efficiency

PLASMA

I/protein ratio

TURNOVER

STUDIES

l=r ( % bound to lipid)

( %) (

Free iodide

Duration of study (days)

Th (hr)

(%I

Weight at death (kg)

27 28

43.7

8.6

1.4

5.3

7.0 7.0

23.1 23.2

12.25 10.25

41 42

12.2 13.8

0.16 0.18

0.22 0.21

14.5 7.9

3.0 6.9

22.2

20 17.5

/

1

PIG PLASMA LIPOPROTEIN TURNOVER IN PLASMA

617

AND TISSUES

z 10: 5

6 a 5%

-

3

-

8 ‘z E 1 * 0.5-

0.1 ;

0

I

1

I

2

I

3

I

4

I

5

I

6

DAYS

Fig. 5. lZ51-LDLz plasma turnover curves. As with Figs. 2 and 3, the ordinate is the plasma activity as a ‘A of that at zero time. The curves are plainly multi-exponential.

turnover reflected

curves are summarised in Fig. 5. Although the turnover data predominantly catabolism of a ,&lipoprotein (Fig. lc) the plasma radioactivity was a little

more widely distributed

in the electrophoretic

fractions

than in the LDLl

turnover

studies. LDLI

tissue distribution

The 1251 distribution

in tissues at death, expressed

as a percentage

of the plasma

activity at death, is shown in Table 8. After 4 days liver radioactivity per gram is a little higher than that of plasma, and after 10 days (when plasma specific activity is per gram is 1000 or more about 3 x IO-5 times that at zero time) liver radioactivity times that of plasma. A possible explanation is that at least part of the liver is in an extravascular compartment. The I251 tissue distribution at death was also expressed as a percentage of the W-LDLr injection) in Table 9. serum counts at zero time (15 min after intravascular It can be seen that the liver radioactivity in fact is about l/30 the zero serum activity per gram at the end of day one, suggesting that in absolute terms the amount of radioactivity retained in the liver is very small. The liver at all times retains a greater proportion of injected 1251 per gram of tissue than other tissues (with the exception of the thyroid gland). The proportion of injected 1251 in the whole liver and skin was quantitated in several animals at various intervals after 125I-LDLr injection. The

rz51-LDLr

IN TISSUES

AT DEATH

-

-

B b C d

5.0 3.5 3.9 2.6 2.5 5.8

5.1 7.3 6.6 0.5 3.7 2.5 1.1 4.2 _ -

-

6.7 1.4 0.4 4.6 3.5& 1.5”

18 12 12 12 27 15 9.4 2.3 1.8 5.4 -

17 7.1 12 11 25 17

24 11 15 16

51 19

18 16 28 23 16 15

18 25 -

80 18

4.0

2.5

4.9 13 2.8 2.2 3.4 2.8 9.9 7.5 -

21 20 18 18 17 15

15 14 10

30 10

3.8

34 26

25 49 19 34 22 28

50 32 24 15 51 51

19 20 21 45

115 56

7.9

38

10 24 1.4 6.2 2.6 1.6 7.6 6.0 4.9 15 16 ---_

30 19 18 17 47 26

46 199 161732 42

106 183 29 -

4.06 4.8

39

55 7.9 14 45

66 61 115 192 945c 272

1708 507 180 384

708 270

9.9

31

1350 492

12.9

33

858 377

13.1

35

-

88 18 21 47

111 80 95 66 45 97

_

61 18 23 94

87 78 111 64 175 100

24 36 52 31 160 55

110 103 49 87

305c 184

13.1

36 19.8

30

105 47 28 70 ~

184 119 163 102 112 95

121 166 154 43

1288 446

27.9

23

252 69 67 209 _

149 131 138 106 120 80

_

62 6.7 27 63

171 108 133 90 154 62

157 350 199122 53 -

1632 2220 150 320

19.8

29

-

70 17 41 63 __

168 131 154 114 99 86

196 366

370

2056 1159

33.8

21

-

77 12 5.0 74

141 105 107 92 99 82

145 167

405

2105 568

33.8

22

nodes from pigs 32 and 33 respectively

60 34 5.2 5.9 32 8.2 44 27 _ _

108 64 118 95 -

556” 1147d 2769 340 244 270 156 135 117 268 241 201

1441 489

12.9

32

Cerebral hemisphere, white and gray matter. Medulla oblongata. Possible contamination or sample error. Haemorrhagic lymph nodes (as were many nodes from other animals). Figures for pale (non-haemorrhagic) are 304 and 574.

Skin Striated muscle Fat Nerve Brain

inner l/3 outer 2/3 thoracic, inner i/a outer 2/3 abdomen, inner r/a outer 2/3

Aorta arch,

7.4 5.3 2.9 3.6 3.3 8.8

25 8.6 2.7 23

-

-

Lymph nodes Salivary glands Kidney Lung

5.8 4.2 5.8 5.2 4.2 9.4

55 11

53 44 7.6 11

38 11

Tissue

48

2.06 3.7

Liver Spleen

I.0

1.0

40

1.0

(days)

20

Duration ofexpt.

19

17

activity per g tissue x 100 corrected in the case of liver and spleen for contained plasma. activity per g serum at death’

OF

Animal No.

Expressed as

THE DISTRIBUTION

TABLE 8

r=l-LDLr

17

1.0

Duration of expt.

1.2

0.8 1.2 1.O 0.8 1.9

l/3

2/3 i/a 2/3 l/a

Aorta arch,

AFTER

DEATH

1.3 0.50 0.84 _ -

1.2

1.8

(0.6)

(0.2) (0.4) (0.4) (0.8)

(0.6)

(0.8) (0.4) (0.5) (0.5)

(1.7) (0.7)

3.7

48c

0.3 (0.3) 0.07 (0.08) 0.02 (0.06) 0.22 (0.18) 0.17a0.07b

0.7

0.6 0.6 0.6 1.3

0.9

1.2 0.4 0.1 1.1

2.7 0.5

2.06

40

0.20 0.11 0.13 0.38 -

0.6

0.8 0.7 0.7 0.7

0.8

0.6 0.5 0.4

1.2 0.4

3.8

34

0.16 0.10 0.13 0.35 -

0.7

0.7 1.3 1.1 0.7

0.8

0.8 1.2 -

3.7 0.8

4.0

25

0.13 0.02 0.02 0.06 -

0.3

0.2 0.2 0.2 0.6

0.4

0.6 0.2 0.2 0.4

1.3 0.4

4.06

39

0.36 0.09 0.11 0.23 _

0.4

0.7 0.3 0.5 0.3

0.4

3.0 0.6

2.8 -

4.8

26

0.003 0.007 0.02 -

0.06

0.04 0.03 0.02 0.06

0.06

0.02 0.03 0.03 0.06

0.1 0.1

7.9

38

0.02 0.003 0.006 0.02 _

0.02 0.05 0.08 0.4 0.1

0.03

0.7 0.2 0.07 0.2

0.3 0.1

9.9

31

0.02 0.005 0.007 0.028 -

0.03

0.02 0.03 0.02 0.05

0.03

0.3 0.1 0.04 0.07

0.4 0.1

12.9

33

0.01 0.001 0.006 0,008 _

-

0.01 0.02 0.02 -

0.02

0.5 0.05 0.02 0.04

0.2 0.1

13.1

35

an approximation,

0.02 0.004 0.004 0.009 -

0.02

0.02 0.02 0.01 0.01

0.02

0.1 0.1 0.03 0.05

0.3 0.1

12.9

32

0.05 0.13 0.12 0.039 _

0.02

0.02 0.03 0.02 0.02

0.03

0.03 0.04 0.02 0.01

0.1 0.06

19.8

30

0.01 0.001 0.002 0.006 -

0.01 0.01 0.01 0.01 0.01

0.02

0.03 -

0.04 0.04

27.9

23

by extrapolating

0.02 0.11 0.006 0.16 -

0.02

0.03 0.04 0.02 0.03

0.04

0.03 0.04 0.04 0.01

0.4 0.03

19.8

29

obtained

0.03 0.005 0.006 0.021 _

0.04

0.02 0.04 0.02 0.1

0.02

0.1 0.08 0.04 0.07

0.2 0.1

13.1

36

corrected in the case of liver and spleen for contained plasma.

’

& Cerebral hemisphere, white and gray matter. b Medulla oblongata. C Some iz51-LDL extravasated at injection. These figures for this animal are therefore plasma decay curve between days 2 and 4 to zero time.

Skin Striated muscle Fat Nerve Brain

1.0 0.7 0.8 0.5 0.5

_

1.1 0.6 0.7 0.7

1.5

1.5 0.74 -

-

1.0 0.10 0.22 -

-

-

Lymph nodes Salivary glands Kidney Lung

2/3

-

7.6 2.2

inner outer thoracic, inner outer abdomen, inner outer

1.0

20

10.6 8.8 1.5 2.2

1.0

19

Liver Spleen

Tissue

(days)

IN TISSUES

activity per g tissue x 100 activity per g serum at zero time (15 min after injection)

OF

Animal No.

Expressed as

THEDISTRIBUTION

TABLE 9

0.01 0.002 0.001 0.010 _

0.01 0.02 0.01 0.01 0.01

0.02

0.06 0.02 0.02

0.3 0.08

33.8

22

the observed

0.01 0.002 0.005 0.007 _

0.02 0.02 0.01 0.01 0.01

0.02

0.04 0.02 0.04

0.2 0.1

33.8

21

620

G. D. CALVERT, P. J. SCOTT AND D. N. SHARPE

TABLE THE

10

PROPORTION

OF THE INJECTED

‘=I-LDLl

DOSE IN THE LIVER

AT

2

AND

Time between lz51-LDL1 injection and death (days) Plasma volume (ml) Plasma volume (ml/kg) Total 1251-LDLl in the liver at death as a proportion of total plasma 1251-LDL1 at death (corrected for contained plasma in liver) (%) Specific activity of plasma lz51-LDLl at death as a proportion of that 15 min after injection (%) Proportion total injected l25I-LDLl in the liver at death (%)

4

DAYS

Pig 40

Pig 39

2.06 1078 52.6

4.06 879 51.7

33

54

4.8 1.6

1.3 0.7

results (Tables 10 and 11) show that, for instance, after 2.06 days the 1251in the liver (not including contained plasma) is about 33 % that of total plasma 1251but the proportion of total injected l25I in the liver is only 1.6%. The figures after 4.06 days are 54% and 0.7% respectively, and measurements after longer intervals continue these trends. Thus, although the proportion of 1251in the liver after the first few days may seem large compared with that at the same time in the plasma, the proportion of the injected l251 in the liver is in fact quite small. The data also suggest that skin is a not unimportant site of 1251-LDh turnover (Table 11). This is apparent when one considers the total organ weight. In man this organ is about 7 % of body weight and is, apart from striated muscle, the largest soft TABLE 11 TOTAL

‘=I-LDL1

IN LIVER

AND SKIN

AS A PROPORTION

OF THAT

IN PLASMA

AT DEATH

The figure for lz51-LDL1 in organs is corrected for contained plasma. Pig No.

Duration

of turnover expt.

(days)

Total organ ‘251-LDL~ at death Total plasma liver”

x 100

‘251-LDL1 at death skin b

liver skin

40 39 31 32 33 35 29 30

2.06 4.06 9.9 12.9 12.9 13.1 18.8

33 54 453 845 806 614 787

18.8

1011

10 15 83 133 92 91 159 382

3.3 3.6 5.5 6.4 8.8 6.7 4.9 2.6

z Liver weight measured directly (mean liver weight 52 g/kg body weight, i.e. approx. 5% body weight). b Skin weight estimated as 8 ‘A body weight (cJ estimated 7 % in rnarF).

PIG

PLASMA

TABLE

LIPOPROTEIN

IN PLASMA

AND

TISSUES

Expressed as

OF ‘251-LDL2

IN TISSUES

AT DEATH

activity per g tissue X 100 corrected in the case of liver and spleen for contained activity per g serum at death ’

41

42

27

28

3.0

6.2

7.0

7.0

Liver Spleen

68 38

255 137

112 50

202 73

Lymph nodes Salivary glands Kidney Lung

30 17 22 45

86 36 60 97

21 20 15 16 6.3 2.3

21 27 44 34

Duration

621

12

THE DISTRIBUTION

Animal

TURNOVER

No. of expt.

(days)

Tissue

Aorta arch,

inner l/3 outer 2/3 thoracic, inner l/3 outer 2/3 abdomen, inner l/a outer 2/3

Skin Striated muscle Fat Nerve

9.1 2.2 1.3 16

9.6 19 4.4 14

23 -

55 -

24 38

22 67

16 II 6.8 10 22 17

8.9 3.6 15 10

10 1.9 2.2 JO

22 17 1.5 3.6 II

tissue organ in the bodyds; in the smaller pig, with thicker skin, it is probably more, e.g. 8% of total body weight. Data in Table 11 suggest (ride the liver: skin ratios in the last column of the table) that turnover in skin is considerably slower than that in liver. The spleen retains less 1251 than the liver, and its size (20-36 g) is insignificant compared to the liver (255-655 g) in the animals studied. Striated muscle, fat, nervous tissue and most viscera do not retain any great proportion of LDLr (Tables 8 and 9). The data in Tables 8 and 9 did not suggest different rates of entry of LDLl into the arch, thoracic and abdominal segments of the aortic wall. Overall, there is significantly more 12sI in the inner one-third of the aorta than in the outer twothirds, weight for weight (using Wilcoxon’s signed ranks test44). LDLz tissue distribution The I251 distribution in tissues at death, corrected for contained plasma and expressed as a percentage of the plasma radioactivity at death, is shown in Table 12 (cf: Table 8). In general the tissue distribution seems similar to that obtained with LDLi. The differences (e.g. in 1251 in the aortic wall at 7 days) invite speculation, but the

622

G. D. CALVERT, P. J. SCOTT AND D. N. SHARPE

TABLE THE

13

TISSUE

AFTER

DISTRIBUTION

INJECTION

The radioactivity

OF FREE

AND THE OTHER

lz51AFTER

INJECTED

24

INTRAVENOUSLY

IN TWO

PIGS,

ONE

KILLED

1 HOUR

HOURS

per g tissue is expressed as percent of activity per g serum.

Duration of experiment

Pig 71 1 hr

Pig 72 24 hr

Liver& Spleen& Lymph nodes Salivary glands Aorta thoracic, inner l/s thoracic, outer 2/~ Skin Striated muscle Fat

8.4 7.6 5.2 3.6

23 8.3 7.1 3.0

19 14 1.3 3.3 1.6

16 11 7.1 5.6 1.9

&Corrected for contained plasma.

number of experimental animals is far too few to draw any conclusions, results must be looked upon as preliminary. As with LDL1, these data display a significant difference activity in the inner one-third and outer two-thirds of the aorta, are considered

together

and these

between the radiowhen all specimens

(2P < 0.05).

Free iodide distribution The plasma T6 for 1251- injected intravenously in two pigs was approximately 1.6 hr over the first 5 hr, though between 5 and 24 hr the decline in plasma activity was much less precipitous, presumably reflecting an expanded body pool. The tissue counts expressed as a proportion of the plasma counts at death are shown in Table 13. The values for tissue radioactivity thus expressed at 1 and 24 hr are similar, suggesting that the tissue radioactivity in these two animals was a function of the rapidly excreted plasma 1251-. As 10% or less of plasma radioactivity during an 12jI-LDL turnover study is due to free ra51-, this is further evidence that tissue radioactivity is largely due to protein-bound 1251. DISCUSSION

Because the serum lipoprotein+7145,46, cardiovascular systemaT-50 and atherosclerotic lesions51-5s in pigs and men are rather similar, pigs are an attractive model for the study of experimentally produced or naturally occurring atherosclerosis3. We have sought in this study of the dynamic aspects of lipoprotein metabolism to extend the comparison. The results indicate that the catabolism of low density lipoproteins differs considerably in the two species.

PIG PLASMA LIPOPROTEIN

In young

TURNOVER

pigs LDLl

(which

623

IN PLASMA AND TISSUES

is similar

to human

LDLa,

density

1.019-1.063

g/ml) is metabolised much faster than in humans. The mono-exponential phase of plasma decay has a Tt of about 23 hr in the pig, and about 34 days in man16*54. Assuming that analysis of plasma decay curves is valid, the mean fractional catabolic rate of LDLl in young pigs is also 3-4 times the values found in man. The rate of turnover partments

was such that equilibrium was not attained

before

between

intravascular

the 8th day (as judged

the mono-exponential nature of the plasma approximation allowing curve analysis.

and extravascular by tissue

com-

radioactivity);

decay curve seems therefore

a fortunate

None of the studies with rssI-LDLa demonstrates a mono-exponential rate of plasma biological decay. If a mean 2-5 day T+is calculated the value obtained is similar to that for LDLl (Table 7), though the initial rate of decay may be much faster. Perhaps the decay curve reflects equilibration in multiple metabolic pools, rather than the two or three pools commonly assumed. Alternatively the turnover rate or the tissue distribution may have been such that equilibrium was never attained. It seems unlikely that LDL size heterogeneity contributed to the multi-exponential nature of the plasma decay curve (though size heterogeneity can produce such a islI-polyvinylpyrrolidone). curve, as Regoeczii7 has demonstrated with intravascular We hope that further work may provide an answer. The problem

recalls the work of Hay and others55, who found that 1251-labelled

serum LDL in normal male rats had a plasma decay curve over 50 hr in which, after an initial steep drop in plasma radioactivity after injection, two exponentials were demonstrated. In contrast, in oestrogen-treated male rats the plasma decay curve was multi-exponential.

They found

no obvious

explanation

for this difference.

There may in fact be no fundamental difference between the metabolism of LDLi and LDLs other than a chance distribution of metabolic pools in LDLl metabolism such that a two-pool analysis is possible, whereas the metabolic pools (as measured

by their effect on plasma

tidily into two groups. fundamental difference

specific activity)

in LDLs metabolism

do not fall

Alternatively the data in Table 2 and Fig. 5 may reflect a in the functions of LDLi and LDLs in lipid transport, as

discussed below. The equilibrium attained after 8-10 days is probably a manifestation of an expanded free iodide pool rather than a function of LDLl metabolism. The information that we could obtain from plasma decay curves in pigs was limited and should be interpreted with caution. Neither whole body counting nor complete urine collections were feasible, and we had therefore to base our analysis solely on changes in plasma is5I-LDL. An assumption on the site of catabolism was necessary3’, and we assumed that this is either in the intravascular compartment or in a site in rapid equilibrium with it. This may be valid for studies on labelled albumi+, (though some workers question this 57959, but there is some evidence that gamma globulins in humans59 and fibrinogen60961 and albumins0 in rabbits may be catabolised partly in an extravascular compartment. Our studies on the tissue stribution of 1251-LDLi suggest that some extravascular catabolism of LDL may OCCUT in the pig, as the radioactivity of liver relative to that of plasma rises throughout

624

G. D. CALVERT,

P. J. SCOTT AND D. N. SHARPE

the study. In addition, the work of Eisenberg et ~1.6~suggests that some free iodide (liberated by the liver as a result of lipoprotein catabolism) is excreted with the bile into the intestine, in effect extravascular catabolism. The effect of the free iodide spaces3 may be significant in studies on a protein with rapid turnover. only about exponential released

In these studies

on labelled

LDLi

we have good evidence

that

10% or less of the 1251 in the plasma is non-protein bound. If the multidecay curve for 1251-LDLa is due to a high level of plasma free iodide

by catabolism,

LDLs may have a much more rapid turnover

than LDLr.

It

will be necessary to investigate LDLs metabolism further, perhaps using a method of analysis that takes into account the free iodide space37l5s (although the size of the free iodide space60,63 and iodide excretion rate41 are probably unstable in small pigs, accentuating the possible error in an unstable analysis method4a). The very low level of circulating radioactivity after 10 days may represent a greatly expanded iodide pool. Iodine may re-cycle through the gastrointestinal tract (some iodide freed from LDL apoprotein

by catabolism

in the liver is excreted in the biless, partly reabsorbed

in the

colon, and perhaps with plasma free iodide excreted into the stomach). Apparent expansion of the iodide pool becomes more important in the later stages of turnover experiments, and may, for instance, be the explanation for the appearance of a late third exponential in human LDL turnover studiess*. It seemed likely to us that studies on tissue radioactivity might provide more information about LDL catabolism. The liver consistently had the highest tissue radioactivity (except for the thyroid, in which 125I accumulation is not a primary function of LDL metabolism). After about 2 days the liver contained about 33 % of the activity present in the plasma, and after 4 days about 54%. This radioactivity is slow to leave the liver, so that by 9.9 days the liver contained about 45 y0 of the radioactivity in the plasma compartment. These figures relate to. plasma; the fraction of the injected dose of I251 retained in the liver is small, 1.6 y0 at about 2 days, 0.7 y0 at about 4 days, and less than 0.2% by 9.9 days. As we have shown, most of this tissue radioactivity was precipitable with trichloroacetic acid, suggesting that it represented rssI-LDL rather than free iodide. These findings on plasma T, and 1251-LDL in the liver are very similar to those reported in an abstract of a series of four animals by Sniderman et a1.e4. There seems to be no other tissue containing a pool of LDL of comparable size,The 1251 in liver appears on autoradiography to be intracellular (unpublished findings). As LDL is responsible for carrying most of the plasma cholesterol, and as cholesterol is catabolised in the liver, it seems likely that the liver is the site of LDL catabolism. Against this we have the findings of Sniderman et cd.65 (reviewed by SteinbergsG), who performed LDL turnover studies before and immediately after hepatectomy in dogs and pigs. They found that the T+ of LDL dropped significantly after hepatectomy (20 hr and 27.4 hr before hepatectomy, and 8.8 hr and Il.3 hr after hepatectomy, for pigs and dogs respectively), and concluded that irreversible LDL removal by the liver seemed quantitatively minor. It is difficult, however, to find any direct evidence to support a mechanism of peripheral catabolism of LDL. Unlike

PIG PLASMA

albumin,

LIPOPROTEIN

TURNOVER

very little ia5I-LDL

IN PLASMA

was found

AND

in striated

TISSUES

625

muscle (if we take 40% of body

weight to be striated muscle, it contains less than one-third as much radioactivity at any one time as does the liver). Likewise, the radioactivity in fat was very low. Nevertheless,

as Steinberg

has pointed

can be invoked to explain their findings on this subject. Perhaps The lipid-poor

LDLi

is converted

fragment,

out66 quite plausible and no doubt

peripherally

and testable

they will publish

hypotheses further

into LDLa (by the removal

LDLa, may recirculate

work

of lipid).

to receive more lipid at the liver.

Our work suggests that LDLs may be metabolised more rapidly than LDLI. If so, this may explain the “LDL-stabilising effect” of the liver postulated by Steinberg66. Evidence that the liver may be a major site of LDL catabolism comes from two other groups of workers who injected radio-iodinated human HDL, LDL and VLDL62 and rat HDLa2967~68, LDL55 and VLDLss,70 intravenously into rats. In all cases the liver proved to be the major site of 1251 uptake, and the authors concluded that it was therefore the main catabolic site. Skin, and artery wall, have a level of radioactivity intermediate between liver and tissues

such as striated

muscle and fat. Skin is probably

less than one-third

the

weight of liver43, and contains about one-tenth the radioactivity per gram tissue. The evidence suggests that skin and artery wall (in particular the inner third of the artery wall) are sites of significant LDL exchange with plasma. This accords with evidence obtained by other workers using immunofluorescencea techniques for localising B apoprotein in LDL. The skin in experimental hypercholesterolaemia is a major cholesterol pool7l, and in hypercholesterolaemic patients skin and artery wall are major sites of LDL and cholesterol accumulation as xanthomas or atheromatous plaques. Xanthomas may regress with prolonged treatment on hypolipidaemic regimens, suggesting that skin cholesterol and perhaps LDL is in slow equilibrium with the plasma compartment. Our autoradiographic findings (to be published) suggest that LDL enters the aortic wall at the intimal surface. Our finding that there is a significantly higher level of radioactivity in the inner one-third of the aortic wall than in the outer two-thirds supports but does not prove this concept, that LDL enters the artery wall at the intimal surface. Several other tissues, in particular striated muscle and fat, account for a large part of body weight, yet have a very low content of 1251, and presumably therefore play a minor part in LDL catabolism. The proportionate weight and water content of organs in the young pig of this age are not markedly different from those in the adult pigds, and these observations on tissue distribution may well be valid also for the adult pig. Our findings on LDLs, a smaller, denser form of pig serum LDL containing apoprotein B, are preliminary and difficult to interpret. They suggest, though, that LDLz may have a more rapid turnover than LDLi. We have mentioned the possibility that in viva there may be some equilibration between LDLl and LDLa; this might contribute to the non-exponential serum decay curve of LDLs.

626

G. D. CALVERT, P. J. SCOTT AND D. N. SHARPE

REFERENCES 1 Intersociety Commission, Primary prevention of the atherosclerotic diseases. Atherosclerosis Study Group and Epidemiology Study Group, Circulation, 42 (1970) A55. 2 WALTON, K. W., The biology of atherosclerosis. In: E. E. BITTAR AND N. BITTAR(Eds.), The York, 1969, Chap. 7, pp. 193-233. Biological Basis of Medicine, Academic Press, London-New 3 RATCLIFFE,H. L. AND LUGINB~~HL,H., The domestic pig - A model for experimental athero13 (1971) 133. sclerosis, Atherosclerosis, 4 JANADO, M., MARTIN, W. G. AND COOK, W. H., Separation and properties of pig-serum lipoproteins, Can. J. Biochem., 44 (1966) 1201. 5 CALVERT,G. D. AND SCOTT, P. J., Properties of two pig low density lipoproteins prepared by zonal ultracentrifugation, Atherosclerosis, 22 (1975) 583. 6 FIDGE, N., The isolation and properties of pig plasma lipoproteins and partial characterisation of their apoproteins, Biochim. Biophys. Acta, 295 (1973) 258. 7 KALAB, M. AND MARTIN, W. E., Gel filtration of native and modified pig serum lipoproteins, J. Chromafogr., 35 (1968) 230. N. J., FARR, A. L. AND RANDALL,R. J., Protein measurement with 8 LOWRY,0. H., ROSEBROUGH, the Folin phenol reagent, J. Biol. Chem., 193 (1951) 263. 9 AutoAnalyzer method 24a. In: AufoAnalyzer Manual, Technicon Instruments Corp., Tarrytown, N.Y. Manual, Technicon Instruments Corp., Tarrytown, 10 AutoAnalyzer method N78. In: AutoAnalyzer N.Y. Based on KESSLER,G. AND LEDERER,H., Fluorometric measurement of triglycerides. In: L. T. SKEGGS,JR. (Ed.), Automation in Analytical Chemistry, Technicon Symposium, 1965, p. 341. 11 LEES,R. S. AND HATCH, F. T., Sharper separation of lipoprotein species by paper electrophoresis in albumin-containing buffer, J. Lab. Clin. Med., 61 (1963) 518. 12 Microzone electrophoresis or immunoelectrophoresis using model R-101 and manufacturer’s method, Beckman Instruments Inc., Fullerton, Calif. 13 Handbook of Chemistry and Physics, 50th edition, edited by R. C. WEAST,The Chemical Rubber Co., Clevedon, Ohio, 1969. 14 MCFARLANE,A. S., Efficient trace-labelling of proteins with iodine, Nature (London), 182 (1958) 53.

15 FOLCH, J., LEES, M. AND SLOANESTANLEY,G. H., A simple method for the isolation and purification of total lipids from animal tissues, /. Biol. Chem., 226 (1957) 497. 16 LANCER,T., STROBER,W. AND LEVY, R. I., The metabolism of low density lipoprotein in familial Type II hyperlipoproteinemia, J. C/k. Invest., 51 (1972) 1528. 17 REGOECZI,E., Iodine-labelled fibrinogen - A review, Brif. J. Haemafol., 20 (1971) 649. 18 MCFARLANE,A. S., Sites of protein catabolism. In: G. BIRKE, R. NORBERGAND L. 0. PLANTIN (Eds.), Physiology and Pathophysiology of Plasma Protein Metabolism, Pergamon, Oxford, 1969, pp. 87-93. 19 POLLARD,H., SCANU, A. M. AND TAYLOR,E. W., On the geometrical arrangement of the protein subunits of human serum low-density lipoprotein: evidence for a dodecahedral model, Proc. Nat. Acad. Sci. U.S., 64 (1969) 304.

20 SCANU, A. M. AND WISDOM,C., Serum lipoproteins

structure and function, Ann. Rev. Biochem.,

41 (1972) 703.

21 KANE, J. P., RICHARDS,E. G. AND HAVEL, R. J., Subunit heterogeneity in human serum beta lipoprotein, Proc. Nat. Acad. Sci. U.S., 66 (1970) 1075. 22 MATEU, L., TARDIEU,A., LUZZATI, V., AGGERBECK,L. AND SCANU, A. M., On the structure of human serum low density lipoprotein, J. Mol. Biol., 70 (1972) 105. 23 GALTON, V. A. AND PITT-RIVERS,R., A quantitative method for the separation of thyroid hormones and related compounds from serum and tissues with an anion-exchange resin, Biochem. J., 72 (1959) 310. 24 FRINGS, C. S., FOSTER,L. B. AND COHEN, P. S., Electrophoretic separation of serum lipoproteins in polyacrylamide gel, C/in. Chem., 17 (1971) 111. 25 HATCH, F. T. AND LEES, R. S., Practical methods for plasma lipoprotein analysis, Advan. Lipid Res., 6 (1968) 1.

26 SCOTT, P. J. AND HURLEY, P. J., The distribution of radio-iodinated serum albumin and lowdensity lipoprotein in tissues and the arterial wall, Atherosclerosis, 11 (1970) 77.

PIG PLASMA

LIPOPROTEIN

TURNOVER

IN PLASMA

AND

TISSUES

627

27 SMITH,E. B. AND SLATER,R. S., Relationship between low density lipoprotein in aortic intima and serum-lipid levels, Lance?, i (1972) 463. 28 OKISHIO,T., Studies on the transfer of I 13t labelled serum lipoproteins into the aorta of rabbits with experimental atherosclerosis, Med. J. Osaka Univ., 11 (1961) 367. 29 DUNCAN, JR., L. E., BUCK, K. AND LYNCH, A., Lipoprotein movement through canine aortic wall, Science, 142 (I 963) 972. 30 ADAMS, C. W. M., VIRAG, S., MORGAN,R. S. AND ORTON, C. C., Dissociation of [3H]cholesterol and [1251]labelled plasma protein influx in normal and atheromatous rabbit aorta, J. Atheroscler. Res., 8 (1968) 679. 31 SCOTT, P. J. AND HURLEY, P. J., Low-density lipoprotein accumulation in aortic and coronary artery walls, Israel J. Med. Sci., 5 (1969) 631. 32 STEIN,O., STEIN,Y. AND EISENBERG, S., A radioautographic study of the transport of 1251-labelled serum lipoproteins in rat aorta, Z. Zellforsch., I38 (1973) 223. 33 CHARMAN,R. AND LIPSKY, S. R., Study of the movement in vivo and in vitro of labelled cholesterol

across the aorta of the normal rat, J. Atheroscler. Res., 7 (1967) 143. 34 KRAMSCH,D. M., GORE, I. AND HOLLANDER,W., The distribution of intravenously administered aH cholesterol in the arteries and other tissues, Part 2 (Radioautographic findings), J. Atheroscler. Res., 7 (1967) 501. localization of serum 35 KAO, V. C. Y. AND WISSLER,R. W., A study of the immunohistochemical lipoproteins and other plasma proteins in human atherosclerotic lesions, Exp. Molec. Puthol., 4 (1965) 465. 36 WALTON, K. W. AND WILLIAMSON,N., Histological and immunofluorescent studies on the evolution of the human atherosclerotic plaque, J. Atheroscler. Res., 8 (1968) 599. 37 NOSSLIN,B., Analysis of disappearance time-curves after single injection of labelled proteins. In: Protein Turnover, Ciba Foundation Symposium 9 (New Series), Associated Scientific Publishers, Amsterdam, 1973, pp. 113-l 30. 38 MATTHEWS,C. M. E., The theory of tracer experiments with 1311-1abelledplasma protein, Physics Biol. Med., 2 (1957) 36. 39 BIANCHI, R., MARIANI, G., PILO, A., TONI, M. G. AND DONATO, L., Short-term determination of plasma protein turnover by a two-tracer technique using plasma only or plasma and urine data. In: Protein Turnover, Ciba Foundation Symposium 9 (New Series), Associated Scientific Publishers, Amsterdam, 1973, pp. 47-65. 40 VITEK, F., BIANCHI, R. AND DONATO, L., The study of distribution and catabolism of labelled serum albumin by means of an analog computer technique, J. Nucl. Biol. Med., 10 (1966) 121. 41 HOFFENBERG,R., ZALIN, A., MCFARLANE, A. S. AND BLACK, E. G., A consideration of short-term measurement of plasma protein catabolic rates. In: H. PEETERS,(Ed.), Protides of the Biological Fluids (19th colloq.), Pergamon, Oxford, 1972, pp. 467471. 42 ZALIN, A. M., CARSON,E., BLACK, E. G. AND HOFFENBERG,R., The short-term measurement of plasma protein catabolic rates, J. Nucl. Biol. Med., 16 (1972) 73. 43 WIDDOWSON,E. M. AND DICKERSON,J. W. T., The effect of growth and function on the chemical composition of soft tissues, Biochem. J., 77 (1960) 30. 44 LANGLEY, R., In: Practical Statistics, Pan Books Ltd., London, 1968, pp. 179-189, based on F. WILCOXON, Biometrics, 1 (1945) 80. 45 MILLS, G. L. AND TAYLAUR, C. E., The distribution and composition of serum lipoproteins in eighteen animals, Comp. Biochem. Physiol., 40B (1971) 489. 46 HAVEL, R. J., EDER, H. A. AND BRAGDON, J. H., The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum, J. C/in. Invest., 34 (1955) 1345. 47 ENGELHARDT,W. V., Swine cardiovascular physiology - A review. In: L. K. BUSTAD AND R. 0. MCCLELLAN (Eds.), Swine in Biomedical Research, U.S. Atomic Energy Commission, Division of Technical Information, and Battelle Memorial Inst., Frayn Printing Co., Seattle, Wash., 1965, pp. 307-329. 48 LUMB, G. AND HARDY, L., Collateral circulation and survival related to gradual occlusion of the right coronary artery in the pig, Circulation, 27 (1963) 717. 49 PENSINGER,R. R., Comparative coronary artery studies, Circulation, Suppl. II to Vols. 35 and 36 (1967) 11-209. 50 FRENCH, J. E., The structure of the tunica intima of large arteries. In: D. G. CHALMERSAND

628

51

52 53

54 55

56 57 58 59 60 61

62 63 64 65 66

67

68

69 70 71

G. D. CALVERT, P. J. SCOTT AND D. N. SHARPE

G. A. GRESHAM,(Eds.),Symposium on the Biological Aspects of Occlusive Vascular Disease, Cambridge University Press, London, 1964, pp. 24-30. LUGINB~~HL, H., Spontaneous atherosclerosis in swine. In: L. K. BUSTADAND R. 0. MCCLELLAN (Eds.), Swine in Biomedical Research, U.S. Atomic Energy Commission, Division of Technical Information, and Battelle Memorial Inst., Frayn Printing Co., Seattle, Wash., 1965, pp. 347-363. FRENCH, J. E., JENNINGS,M. A., POOLE, J. C. F., ROBERTSON,D. S. AND FLOREY, H., Intimal changes in the arteries of ageing pigs, Proc. Roy. Sot. (London), B 158 (1963) 24. GETTY, R., The gross and microscopic occurrence and distribution of spontaneous atherosclerosis in the arteries of swine. In: J. C. ROBERTSAND R. STRAUSS(Eds.), Comparative Atherosclerosis, Hoeber Medical Div., Harper and Row, New York, 1965, pp. 1 l-20 and Atlas pp. 6-11. HURLEY, P. J. AND SCOTT, P. J., Plasma turnover of $0-9 low-density lipoprotein in normal men and women, Atherosclerosis, 11 (1970) 51. HAY, R. V., POTTENGER,L. A., REINGOLD,A. L., GETZ, G. S. AND WISSLER,R. W., Degradation of 1125-Iabelled serum low density lipoprotein in normal and estrogen-treated male rats, Biochem. Biophys. Res. Commrm., 44 (1971) 1471. CAMPBELL,R. M., CUTHBERTSON, D. P., MATTHEWS,C. M. AND MCFARLANE, A. S., Behaviour of 14C- and 1311-labeIled plasma proteins in the rat, Znt. J. Appl. Radiation Isotopes, 1 (1956) 66. KATZ, J., BONORRIS,G., GOLDEN, S. AND SELLERS,A. L., Extravascular albumin mass and exchange in rat tissues, C/in. Sci., 39 (1970) 705. KATZ, J., BONORRIS,G. AND SELLERS,A. L., Extravascular albumin in human tissues, Clin. Sci., 39 (1970) 725. ALPER, C. A., FREEMAN,T. AND WALDENSTR~M,J., The metabolism of gamma globulins in myeloma and allied conditions, J. C/in. Invest., 42 (1963) 1858. MCFARLANE, A. S. AND KOJ, A., Short-term measurement of catabolic rates using iodine-labelled plasma proteins, J. Clin. Invest., 49 (1970) 1903. REGOECZI,E. AND ALLISON,A. C., Unpublished findings, 1969. In: Discussion following reference 26, Protein Turnover, Ciba Foundation Symposium 9 (New Series), Associated Scientific Publishers, Amsterdam, 1973, p. 68. EISENBERG,S., WINDMUELLER,H. G. AND LEVY, R. I., Metabolic fate of rat and human lipoprotein apoproteins in the rat, J. Lipid Res., 14 (1973) 446. ZIZZA, F., CAMPBELL,T. J. AND REEVE, E. B., The nature and rates of excretion of radioactive breakdown products of 1131-aIbumin in rats, J. Gen. Physiol., 43 (1959) 397. SNIDERMAN,A. D., CAREW, T. E. AND STEINBERG,D., Metabolism of low density lipoprotein apoprotein in the pig, Circulation, Suppl. II to Vols. 45 and 46 (1972) 11-249. SNIDERMAN,A. D., CAREW, T. E., CHANDLER,J. G., HAYES,S. AND STEINBERG,D., The role of the liver in metabolism of low density lipoproteins (LDL), J. C/in. Invest., 52 (1973) 79a. STEINBERG,D., Lipoprotein catabolism. In: G. SCHETTLERAND A. WEIZEL (Ed%), Atherosclerosis ZZZ,Proceedings of the Third International Symposium, Springer-Verlag, Berlin, Heidelberg, New York, 1974, p. 658. ROHEIM,P. S., RACHMILEWITZ,D., STEIN, 0. AND STEIN, Y., Metabolism of iodinated high density lipoproteins in the rat, Part 1 (Half-life in the circulation and uptake by organs), Biochim. Biophys. Acta, 248 (1971) 315. RACHMILEWITZ,D., STEIN,O., ROHEIM,P. S. AND STEIN,Y., Metabolism of iodinated high density lipoproteins in the rat, Part 2 (Autoradiographic localization in the liver), Biochim. Biophys. Acta, 270 (1972) 414. EISENBERG,S. AND RACHMILEWITZ,D., Metabolism of rat very low density lipoprotein, Part I (Fate in circulation of the whole lipoprotein), Biochim. Biophys. Acta, 326 (1973) 378. EISENBERG,S. AND RACHMILEWITZ,D., Metabolism of rat very low density lipoprotein, Part 2 (Fate in circulation of apoprotein subunits), Biochim. Biophys. Acta, 326 (1973) 391. ARMSTRONG,M. L., CONNOR, W. E. AND WARNER, E. D., Tissue cholesterol concentrations in the hypercholesteremic rhesus monkey, Circulation, Suppl. II to Vols. 35 and 36 (1967) 11-3.