Biochimicu et Bioph_vsica Acia, 7 I I Elsevier Biomedical
(I 982) 2 13-223
213
Press
BBA 5 1073
FURTHER CHARACTERIZATION OF THE CHANGES OCCURRING IN THE PLASMA LIPOPROTEIN SPECTRUM IN THE EUROPEAN BADGER (MELES MELES L.) DURING WINTER P. MICHEL
LAPLAUD
a.b, LAURENCE
BEAUBATIE
a and DANIEL
MAUREL
’
* Lahoratoire de Biochimie mkdicale, Faculti de mPdecine et de pharmacre. 2 rue du Dr. Marcland, 87037 Limoges Cedex. h Luhoratoire de Physiologie animale, I/. E.R. des Sciences, 123 rue A. Thomas, 87060 Limoges Cedex and ’ Centre d’ktudes hiologiques des animaux sauvages (CEBAS-CNRS), 79360 Beauvoir-sur-Niort (France) (Received August 3lst, 1981) (Revised manuscript received December
Kqv words: Seasonal vurialion;
21st, 1981)
Lipoprotein
change; (Badger)
The plasma lipoprotein pattern in the European badger has been shown previously to undergo marked and complex quantitative and qualitative seasonal modifications (Laplaud, P.M. et al., 1980, J. Lipid Res., 21, 724-738). However, the conventional ultracentrifugal techniques then in use in our laboratory were of insufficient discriminating power with regard to the numerous lipoprotein fractions whose presence was suggested by our analyses. In the present study, a new density gradient ultracentrifugation procedure was applied to the more detailed determination of the distribution of plasma lipoproteins. The first series of analyses was performed in early December and the second in March, i.e. at the dates when the maximum and minimum, respectively, of lipidemia occur in this species. The fractions thus obtained, each of which corresponded to a narrow density interval, were analyzed subsequently for chemical composition, appearance upon polyacrylamide gel electrophoresis, and for their content of tetramethylurea-soluble apolipoproteins in alkaline-urea gels. Changes occurring from December to March included a large decrease in the plasma concentration of the 1.015-1.065 g/ml lipoproteins, chemical analysis of this material being compatible with the presence of at least two lipoprotein populations. On the other hand, high-density lipoproteins (1.065- 1.162 g/ml) appeared less variable in chemical composition, although the proportion of those with lower density decreased considerably in early spring. Polyacrylamide gel electrophoresis of the native fractions showed multiple bands in most of them; the tetramethylmea-soluble apoprotein profile remained similar at the two dates considered with an apolipoprotein A-I-like component present in large amounts throughout the entire low- and high-density ranges.
Introduction
ent plasma lipoprotein species in this animal undergo seasonal changes in both their concentrations and chemical composition, these being most evident during the cold months. Thus, a maximum of cholesterolemia and phospholipidemia is evident, according to the year, in late November, December or January; this phenomenon occurs concomitantly with the presence of large amounts of LDL (d 1.006- 1.063 g/ml) and with an increase in the relative contribution of lower density
In a previous report [l], we have shown that the European badger exhibits an original and complex plasma lipid transport system. Indeed, the differ-
Abbreviations: VLDL: very low density lipoproteins, JC I.006 g/ml; LDL: low density lipoproteins, density as defined: HDL: high density lipoproteins, density as defined.
OOOS-2760/82/OOOC-0000/$02.75
0 1982 Elsevier Biomedical
Press
214
components in the HDL class (d 1.063- 1.2 1 g/ml). On the other hand, plasma lipids usually attain their minimum level at the beginning of spring; at the same time, the concentration of the low density substances decreases markedly while the lipid composition of this density class is altered, especially with respect to the ratio of the respective amounts of triacylglycerol and cholesteryl esters. During the course of these studies, it became evident that the methodologies then in use in our laboratory, namely conventional preparative ultracentrifugation and analytical ultracentrifugation of components isolated at densities of 1.063 and 1.2 1 g/ml, were of insufficient discriminatory power to resolve further the lipoprotein spectrum in the badger. We have, therefore, taken advantage of a new density gradient ultracentrifugation procedure which allows us to obtain lipoprotein fractions from narrow density intervals. These fractions were used subsequently for chemical analysis, electrophoretic examination, and determination of their tetramethylurea-soluble apolipoprotein content. Materials and Methods Animals and diets The adult male badgers, bred in the Centre d’Etudes Biologiques des Animaux Sauvages, were approximately 2-6 years old. They were kept individually in 6-m2 parks under natural conditions of light, temperature and rainfall. None of the animals exhibited pathological manifestations during the course of the present study. The six badgers used received a diet consisting of a commercial food for dogs (Canina, Duquesne-Purina) and containing the following proportions by weight of the major 20%; animal fat, 6%; constituents: protein, carbohydrate, 5%; vitamin A, 15 000 I.U./kg; vitamin D3, 1500 I.U./kg. Water was provided ad libitum. Collection and treatment of blood For each of the two successive series of manipulations, which began, respectively, on December 3rd, 1980 and March, 25th, 1981, blood samples were taken from six animals which had been fasted overnight for approximately 18 h. Blood was collected on EDTA (final concentration, 1 mM), at
approximately 11.00 a.m., by puncture of the radial vein. Plasma was then separated by low-speed centrifugation and brought to the laboratory on ice. Chemical analysis The techniques used for measurement of plasma total and unesterified cholesterol and triacylglycerol have been described elsewhere [ 11. Phospholipids were determined by means of the ‘Phospholipids B-Test’ (Wako Chemicals, Osaka, choline using phosJapan), which liberates pholipase D; specific measurement of the choline content of all choline-containing phospholipids present is thus obtained. All the techniques in use were checked first for linearity over the entire range of values observed, and were under continuous monitoring by use of several control sera. All the measurements were performed on a semi-automated Clinicon Ultrolab analytical apparatus (LKB Instruments, Bromma, Sweden). Gradient gel electrophoresis of plasma lipoproteins and of lipoprotein fractions Polyacrylamide gel electrophoresis in a threestep gradient was performed according to the method of Fruchart [2]. For the reasons previously discussed [ 11, no attempt was made to quantify the various components. Application of this technique to the electrophoretic examination of the fractions obtained by density gradient ultracentrifugation provided evidence for considerable heterogeneity, with several distinct bands in the LDL region (see Results). In order to ascertain whether this phenomenon was artifactual, we subsequently performed the following manipulations: (1) as heterogeneity was observed first on density gradient fractions originating from pools of plasma from six animals, samples from individual animals were electrophoresed; (2) to eliminate any possible influence of the type of gel used, comparative electrophoreses were conducted on polyacrylamide gradient gels slabs, 216% monomer concentration (PAA 2/16, Pharmacia, Uppsala, Sweden); (3) in some experiments, pre-staining of lipoprotein fractions was replaced by conventional staining, after electrophoretic migration, with 1% Coomassie brilliant blue R (Sigma).
215
In all three types of manipulations, a similar heterogeneity was observed. Furthermore, recombination of the density gradient fractions corresponding to the 1.0155 1.065 g/ml class followed by examination of the resultant mixture under our usual electrophoretic conditions gave a pattern which resembled the more diffuse one obtained with the total 1.006-1.063 g/ml density class isolated by sequential preparative ultracentrifugation (see below) from the same samples. Finally, it is of note that the electrophoretic heterogeneity typical of badger LDL was not observed in several samples from healthy humans examined under identical conditions. Ultracentrifugal methods In each of the manipulations described below, all the NaCl and/or NaBr solutions used for adjustment of densities, as well as for dialysis of the lipoprotein fractions, contained EDTA (0.4 g/l), sodium azide, (0.1 g/l) and merthiolate, (1 mg/l). Monitoring of the actual background densities was performed by means of a DMA 46 calculating precision density meter (Anton Paar KG, Graz, Austria), at 17°C. Sequential preparative ultracentrifugation. Plasma lipoproteins were isolated in the classical density intervals (i.e. d< 1.006 g/ml and d 1,006 1.063 g/ml), according to established procedures [3]. This was performed in a MSE Prepspin 50 ultracentrifuge (MSE, Crawley, U.K.), using an aluminum fixed-angle rotor (capacity 8 X 14 ml), at 17°C. Density gradient ultracentrifugation. Gradients were constructed essentially as quoted by Chapman et al. [4] except that NaCl/NaBr solutions were used in place of NaCl/KBr ones. The gradients were then placed in an MSE 6 X 14 ml titanium swing-out rotor and centrifuged for 48 h at 40000 rev./mm (196000 X g) at 17’C. No braking was employed at the end of the run. The serum-containing gradients were divided into successive 0.5 ml fractions by stepwise aspiration with a micropipette; the background density of each fraction was determined by reference to the density profile obtained form control gradients.
Prior to further chemical or electrophoretical analysis, lipoprotein fractions obtained either by sequential or by density gradient ultracentrifugation were dialyzed according to the conditions previously reported [l]. Chemical analysis of the fractions was performed using the same techniques described above for measurement of the plasma lipids, while the method of Lowry et al. [5] was employed for the assay of protein concentrations. All determinations were made in duplicate, their precision being expressed as the technical error [ 11. This latter parameter, defined for each component assayed, was multiplied by the suitable Student’s t value, thus leading to the determination (in percentage of mass of lipoprotein) of the least significant difference (P-C 0.05): values obtained are quoted in Tables II and III, which summarize the data regarding lipoprotein composition and its seasonal changes. Electrophoresis of apolipoproteins The tetramethylurea-soluble apoprotein components of the lipoprotein fractions isolated by density gradient ultracentrifugation were electrophoresed according to the modification by Kane [6] of the procedure of Davis [7]. For the reasons discussed by Chapman et al. [8], no attempt was made to quantify each band of apolipoprotein. Statistical assay of the quantitative variations of plasma lipids In view of the small number of animals, the statistical significance of variations in plasma lipid levels (total and unesterified cholesterol, triacylglycerol and phospholipids) was assayed by means of the non-parametric test of Mann and Whitney [9]. Results Plasma lipids Table I provides evidence for the large differences noted between the two dates of sampling with regard to all the lipid components assayed. Examination of the plasma levels of the same lipids at several dates before and after those when the present samples were obtained showed that the periods used in this study actually corresponded to the annual maximum and minimum of lipidemia.
216
TABLE I PLASMA
LIPIDS
Values are means? Variable
AND
BODY WEIGHTS
SD. from series of six animals.
assayed
Total cholesterol (mg/lOO Esterified cholesterol
The statistical
AT THE TWO DATES
ml)
December
March
416
169
536
0.682 532 137
0.01 i42 *I2
15.4 2
Observed concentration ratios (December level/ March level) were of the order of 2.50 for total cholesterol, 1.90 for phospholipids and 1.85 for triacylglycerol; at the same time, the mean esterified/total cholesterol concentration ratio decreased by about 0.04. All these seasonal modifications were highly significant (P < 0.01) and consistent with our previous data [ 11.
[9].
Statistical significance of the difference observed between the two dates of sampling *23
0.64-’ 281 74
I.4
OF EXPERIMENT
test used was that of Mann and Whitney
Date
Total cholesterol Phospholipids (mg/ 100 ml) Triacylglycerol (mg/ 100 ml) Body weights (kg)
IN THE BADGERS
P
i-29 il3
13.3 i- 0.5
P
sible only in the December samples. It is of note that authentic VLDL obtained from the same samples by conventional preparative ultracentrifugation were of similar composition. Also in December, most of plasma cholesterol lay in the low density range while in March the situation was reversed, to the benefit of higher density subs-
Eiectrophoresis of plasma lipoproteins The electrophoretic patterns of plasma lipoproteins in our animals were indistinguishable in all badgers examined at the same date and entirely consistent with our published results for corresponding periods of the year [l]. For this reason, these data are not shown.
tances. Thus, the curves representative of the cumulative percentage distribution of cholesterol in the fractions (Fig. 2) show that the mid-point of this distribution was reached at 1.046 g/ml in December, the corresponding value in March being 1.074 g/ml. That plasma cholesterol carried by LDL (thereafter considered as covering the 1.O15- 1.065 g/ml density interval in view of the results of our
Density gradient ultracentrifugation Chemical analysis. The profiles noted when assaying the density distributions of the major components of badger lipoproteins at the two dates of sampling are presented in Fig. 1 and show the redistribution of cholesterol, phospholipids, triacylglycerol and lipoprotein protein occurring during the winter months. At the same time these data allowed measurement (based on the calculated sum of the different components) of the concentrations of lipoprotein material in the successive fractions; using an appropriate factor, these values permitted evaluation of the corresponding plasma concentrations. Lipoproteins with d < 1.015 g/ml were present at low levels both in December (approx. 20 mg/lOO ml) and in March (less than 10 mg/lOO ml); consequently, compositional analysis was pos-
analyses) was predominant in December may be accounted for by two different phenomena, namely the considerable plasma concentration of these lipoproteins (approx. 800 mg/lOO ml) and their concomitant enrichment in cholesteryl esters, when compared with the situation in spring. This second point is particularly evident from the results presented in Table II. However, this table shows equally that, from the viewpoint of lipid composition, the LDL class was heterogeneous at both dates considered. Indeed, the ratio of the relative proportions of triacylglycerol and cholesteryl esters exhibited large variations in successive fractions. In December, the LDL class appeared to include at least two components: firstly, one which was cholesteryl ester-rich and whose maximal concentration lay in the d 1.025 g/ml region, and a second, richer in triacylglycerol and of higher hy-
4
0
,.ol9
December
1.06
,027
1030
1.m
1.0161.m 1130
&
,162 L
Mme
1.2!1
lq
(ii,
from top of tube)
0
3
‘
5
6
7
0
0
10
11
-
_,~_
_L_-_
.,
_.^
--
-
,_
._.,
_-__
___
,
_I
Fig. 1. Badger lipoprotein profiles as evaluated by density gradient ultracenttifugation in samples obtained in December and March. Each of the two diagram refers to a pool made up from equal volumes of plasmas of each of the six badgers examined. See text for experimental conditionso, protein determination, X , cholesterol, A, phospholipids; q, triacylglycerol. The scale inserted in the upper part of the two diagrams refers to the ratio of the concentration of esterified cholesterol (EC) to total cholesterol (TC) in each fraction in which sufficient lipoprotein material was available. With regard to changes in the EC/TC value, calculations made from duplicate determinations of total and unesterified cholesterol (see Materials and Methods) showed that the least significant difference (PcO.05) was less than 0.01 in both the December and March series of measurements.
0
mo-
100
3x-
roc-
N -4
218
TABLE
II
CHEMICAL TION
COMPOSITION
OF PLASMA
LIPOPROTEINS
ISOLATED
BY DENSITY
GRADIENT
ULTRACENTRIFUGA-
Values, expressed as percentages by weight of total lipoprotein, were obtained from the pools of plasma isolated in December (D) and in March (M). See text for experimental conditions. The values obtained for the least significant differences at the PcO.05 level (see Materials and Methods) were, in the December and March series of measurements, respectively, 0.51 and 0.46% for cholesteryl esters, 0.19 and 0.18% for free cholesterol, 0.35 and 0.42% for triacylglycerol, 0.64 and 0.42% for phospholipids, and 1.14 and 1.02% for protein. n.d., Composition not determined, owing to the very low concentrations of lipoproteins present in these fractions. Fraction (ml, from top of
Density
limits
(g/ml)
Date of sampling
Cholesteryl esters
Free cholesterol
Triacylglycerol
Phospholipids
Protein
60.7 n.d.
15.3
10.X
26.4
17.0
tube) 0
-0.5
< 1.013
D M
5.5
7.7
0.5-1.0
1.013-1.015
D M
n.d. n.d.
1.0-1.5
1.015-1.017
D M
n.d. n.d.
1.5-2.0
1.017-1.019
D M
n.d. nd.
2.0-2.5
1.019-1.023
D M
34.0
D M
37.0 13.5
10.8
D M
32.3 17.6
11.4
D M
29. I 16.6
D M
26.3 21.1
D M
28.5 23.5
D M
2.5-3.0
3.0-3.5
3.5-4.0
4.0-4.5
4.5-5.0
5.0-5.5
5.5-6.0
6.0-6.5
6.5-1.0
7.0-7.5
7.5-8.0
8.0-8.5
8.5-9.0
I .023- I.027 I .027- 1.032 1.032-1.039 I .039- 1.046 I .046- I .055 I .055- I.065 1.065- 1.076
1.076-I.087 1.087-1.100
1.100-1.115
1.115-1.130 1.130-1.146
1.146-1.162
12.6
10.1
n.d. 12.5
12.2
7.1 28.4
26.7
16.8
23.0
24.3
9.7 23.3
28.2
17.6
24.3
23.3 18.1
11.0
13.7
II.0
23.9
28.0 24.3
24.3 22.6
9.1
15.8
26.2
II.0
16.9
27.0
24.1
8.4
9.6
21.3
26.2
10.4
9.9
29.0
21.3
31.8 27.0
8.0
4.2
30.3
25.7
9.2
4.6
29.9
29.3
D M
29.8 26. I
6.1
1.8 2.1
29.3
32.4
30.7
32.6
D M
28.7 26.4
5.8
1.5 2.1
31.2
32.9
5.5
31.6
34.5
D M
27.1 24.5
4.2
1.4
30.9
4.5
1.6
31.1
35.x 38.3
D M
24.4 22.8
3.7
1.4
30.6
39.9
3.6
1.7
30.9
40.9
D M
21.8 21.3
D M
22.7 20.9
D M
9.8 10.5
8.0
2.8
1.4
27.5
46.5
3.4
1.5
29.5
44.3
4.7
I.7
20.7
50.3
1.6
1.6
21.9
54.0
15.7 16.8
10.6
2.0
1.9
1.3
1.6
69.1
219
Volume
(ml, II-anlopof
tub2 I
Fig. 2. Cumulative percentage distribution of cholesterol as a function of density, determined in the fractions obtained by density gradient ultracentrifugation, at the two dates of sampling, from the same pools described in the legend to Fig. I. The value of the density at which 50% of total plasma cholesterol was attained is indicated on each curve.
g/ml). In March, the drated density (1.032-1.046 was reduced to trace lower density component amounts. The denser one then became predominant although its concentration was reduced (lipoprotein concentration in the 1.032-1.046 g/ml interval: December, 340 mg/lOO ml; March, 95 mg/lOO ml). At the same time, this later lipopro-
tein material became enriched in triacylglycerol. In addition, LDL of higher density (1.046-1.065 g/ml) exhibited a progressive decrease in triacylglycerol content and thus were intermediate in lipid composition between 1.032- 1.046 g/ml LDL and lower density HDL (1.065 1.076 g/ml). With regard to total HDL, (thereafter denoted as those within the interval 1.0655 1.162 g/ml), only minor variation occurred in the plasma concentration (December, 625 mg/lOO ml; March, 450 mg/lOO ml). However, the ratio of the concentrations of the 1.065-1.100 g/ml vs 1.100-1.162 g/ml components decreased significantly from December (1.57) to March (0.77). Unlike the situation in LDL, no major compositional modification was evident in these high-density substances between the two dates of sampling; however, some enrichment in cholesteryl esters was of note in the lower density HDL in December. Electrophoresis of lipoproteins. Fig. 3 shows the electrophoretic profiles observed in the two series of samples. The December and March patterns shared several features in common: first, no detectable band with VLDL-like migration appeared in the fractions of lowest density; second, heterogeneity was present in each fraction throughout the LDL and HDL regions, and especially in the LDL portion of the pattern where several components migrated as sharp, narrow bands at both dates. As noted previously (see Materials and Methods),
March
Fig. 3. Polyacrylamide gel electrophoresis of the density gradient fractions, from the same pools described in the legend to Fig. I. In each photograph, the direction of migration is from top to bottom; arrows indicate the position of layers; upper and lower dotted lines correspond to limits between 1st and 2nd, and 2nd and 3rd gels, respectively. VLDL, LDL and HDL indicate the typical migration of the corresponding human lipoprotein fractions when examined by this technique.
220
several types of manipulations were performed in order to eliminate an artifactual phenomenon. Major differences between the late fall and early spring patterns can be summarized as follows: 1, in December, large amounts of lipoprotein material migrating in the top of the lower gel were evident in the 1.023-1.027 g/ml fraction, while in March no stainable substances were of note at d < 1.027 g/ml; 2, in the adjacent fractions of higher density (d 1.027- 1.065 g/ml), several components were present at both dates, but heterogeneity was greater in the December sample, with at least four distinct bands; at this same date, large amounts of lipoproteins with mobility intermediate between typical LDL and HDL were present, while they had disappeared in March; on the other hand, some VLDL-like material was of note at this later date in the 1.027- 1.065 g/ml range; 3, apart from a typical HDL band, the 1.065- 1.146 g/ml region presented as a spectrum of slowermigrating lipoproteins; these were stained much more intensely in the late fall samples, where several zone-like components were of note. Electrophoresis of tetramethylurea-soluble apolipoproteins. On both a qualitative and a semiquantitative basis, it was evident that the tetramethylurea - soluble apolipoprotein components
were essentially identical in the two seasons considered. Therefore, only the profile seen in the December samples is presented in Fig. 4. However, in March, the minor amounts of lipoprotein present in the 1.019- 1.033 g/ml interval did not allow us to perform these analyses on the successive fractions corresponding to this region of the density spectrum. Apolipoprotein profile was generally consistent with our previous data [1], and not less than nine distinct components could be observed, numbered I-IX according to their respective migration. As in our preceding report [ 11, results can be interpreted tentatively by comparison of the electrophoreticmobilities of the different peptides with those in man [lO,l 11. Band I, present in trace amounts in the higher density fractions, could be related to human apolipoprotein C-I. Band II, only hetectable at d> 1.100 g/ml, may be a counterpart to apolipoprotein A-IV [12,13] while band III, present at low concentration throughout the density similarly to human spectrum, migrated apolipoprotein E. Band IV, which was previously shown by us [ 1] to possess a molecular weight of 25000-28000, resembles human apolipoprotein AI and is the most prominent tetramethylureasoluble component even in the low-density fractions. Band V may be related to apolipoprotein D, and bands VI-IX to apolipoprotein C-II and different forms of apolipoprotein C-III; these latter peptides appeared distributed relatively evenly across the density range, bands VII and VIII generally being most evident. Discussion
Fume
(ml,fromtopof fube)
Fig. 4. Tetramethylurea-soluble apolipoproteins in the fractions obtained by density gradient ultracentrifugation, from the same pools described in the legend to Fig. 1. For reasons discussed in the text, only the profile observed when examining the December samples is presented in this figure. Electrophoresis is according to the method of Kane [6]; approximately 70 ng total protein was applied to each gel. Upon completion of electrophoresis, gels were fixed in 10% trichloroacetic acid, stained with 0.5% Coomassie brilliant blue and destained for about 48 h in 10% trichloroacetic acid. The arrow indicates the position of the dye front.
The use of density gradient ultracentrifugation has led to results which are in agreement with our previously published data [ 11. At the same time, these results extend considerably our knowledge of the detailed distribution of the plasma lipoprotein components in the European badger, considered at dates when the annual maximum and minimum, respectively, of lipidemia occur. The main interest of this study lies in the discrimination of subspecies of low- and high-density lipoproteins, and in the evaluation of their qualitative and quantitative modification between late fall and early spring. It is thus evident both from
OF THE RELATIVE
III
PROPORTIONS
OF THE VARIOUS
COMPONENTS
OF BADGER
LOW DENSITY
LIPOPROTEINS
1.046- 1.055
4.5-5.0
I .065
D M
I .039- 1.046
4.0-4s
1.055-
D M
1.032-1.039
3.5-4.0
5.0-5.5
D M
I .027- 1.032
3.0-3.5
D M
D M
D M
Date of sampling
1.023-1.027
limits
2.5-3.0
Density
(g/ml)
Fraction (ml, from top of tube)
0.13 (0.12-0.15) 0.17 (0.15-0.19)
0.34 (0.32-0.36) 0.42 (0.40-0.45)
0.60 (0.58-0.63) 0.80 (0.76-0.84)
0.47 (0.45-0.49) I.44 (1.38-1.51)
0.30 (0.28-0.32) 1.32 (1.27-1.38)
0.19 (0.18-0.20) 2.10 (2.00-2.21)
Ratio at each date of sampling
Triacylglycerol/cholesteryl
1.31 (1.00-1.58)
1.24 (1.11-1.41)
1.33 (1.21-1.45)
3.06 (2.82-3.36)
4.40 (3.97-4.93)
11.05 (10.00-12.28)
M/D
esters
1.15 (1.03-1.30)
0.98 (0.86-1.10)
0.96 (0.90- 1.03) 0.94 (0.89-0.99) 0.85 (0.79-0.90) 0.98 (0.93- 1.03)
I .03 (0.90- 1.19)
1.54 (1.34-1.80)
I .55 (1.32-I .79)
1.68 (1.43- 1.96)
M/D
0.86 (0.80-0.93) 0.89 (0.84-0.95)
0.65 (0.59-0.70) 1.OO (0.94- f .06)
0.62 (0.57-0.68) 0.96 (0.90- 1.02)
0.63 (0.57-0.69) 1.06 (0.99-1.12)
Ratio at each date of sampling
Protein/phospholipids
Fractions, obtained by density gradient ultracentrifugation, are from December (D) and March (M) samples. See text for experimental conditions. The values used for calculation of the (t~acylglycerol/choleste~l esters) and (protein/phospho~pids) ratios are taken from Table II. VaIues in parentheses indicate the lower and upper limits of the interval calculated, for each ratio, by application of the values of the leas: significant difference (P-=0.05) obtained for the corresponding chemical components (see Table II and Materials and Methods).
RATIOS
TABLE
M
222
chemical analyses and electrophoretic patterns that LDL (d 1.015- 1.065 g/ml) are largely heterogeneous; furthermore, in some density gradient fractions, the respective proportions of each of the lipid components and of the protein moiety were distinct in the two series of samples. It is customary to discriminate between ‘core’ components’, i.e. triacylglycerol and cholesteryl esters, and ‘surface components’ such as protein and phospholipids. Table III, which summarizes the ratios of the respective percentages of triacylglycerol/cholesteryl esters and protein/phospholipids in successive fractions in the low-density region, thus shows that the lipoproteins that are present in December differ from those that are present in March. With regard to the former ratio, it is evident that its modifications between December and March, expressed as (triacylglycerol/cholester01 esters) M,,,,/(triacylglycerol/cholester01 esters)oecember, become of smaller amplitude with the increasing density of the fraction considered, from about 11 in the d 1.023- 1.027 g/ml fraction to 1.3 in that of d 1.055- 1.065 g/ml. The figures noted in Table III thus suggest a four-compartment model, with successive values of the above-described ratio of I 1.05 (1.023- 1.027 g/ml), 4.40 (1.027- 1.032 g/ml), 3.06 (1.032- 1.039 g/ml) and about 1.25-1.30 (in the three fractions covering the 1.039- 1.065 g/ml interval). For its part, the ratio (protein/phospholipid),,,,/(protein/phospholipid),,,,,,, would rather suggest a two-compartment model (dc 1.039 g/ml, ratio values about 1.50- 1.70, and d > 1.039 g/ml, ratio values about 1.O- 1.15). The marked electrophoretic heterogeneity and the seasonally different appearance of each density gradient fraction clearly substantiate the contention that badger LDL contains a number of discrete molecular components. Among them, those of higher hydrated density may be relatively constant in chemical composition while varying greatly in plasma concentration according to season; in contrast, those of lower density may be modified markedly in composition, especially with respect to their ‘core components’. The constancy of the tetramethylurea-soluble apoprotein profile is, however, of particular note, since it is independent of the marked changes in lipoprotein composition. Furthermore, an apolipoprotein A-I equivalent is
present ubiquitously. This latter phenomenon could be related to the presence, in the low-density range, of fragments resulting from chylomicron metabolism. However, the relative proportion of apolipoprotein B in each fraction, as well as seasonal changes in its contribution to the total protein moiety of individual fractions, remains to be assessed. From a more theoretical point of view, it is of note that lipoprotein material with rather different total apolipoprotein content can possess similar hydrated density. For example, lipoproteins with density 1.023-1.027 g/ml exhibited 16.8% of protein by weight in December, and 24.3% in March. This may be considered in the light of the observations of Mills [14], who found no correlation between the weight percentage of protein in human low-density lipoproteins and their S, rate which, from data compiled by this author, is a linear function of hydrated density at least in the S, 4- 10 range. Contrary to the situation in LDL, the higher density components of the spectrum do not exhibit pronounced seasonal differences in their chemical composition, irrespective of the precise density interval considered. On the other hand, the reduction in the levels of the cholesterol-rich HDL of lower density between December and March most probably accounts for the overall lower cholesterol and greater protein content of total HDL in early spring; this phenomenon is also in accordance with our previous results [ 11. The modifications in the electrophoretic pattern of successive fractions in the 1.065 1.146 g/ml interval primarily consist of a diminution in the slower migrating components in March; this may be related to the concomitant decrease in the concentrations of HDL of lower density. Our electrophoretic procedure is, however, unable to discriminate between human HDL, (d 1.063- 1.125 g/ml) and HDL, (d 1.125-1.21 g/ml), which typically comigrate in the same position, this corresponding to the main, and faster migrating, component of badger HDL. This slower migrating material must consist, therefore, of lipoproteins differing in size and composition. The present study provides evidence that badger plasma lipoproteins are considerably more heterogeneous than was anticipated from our pre-
223
ceding results, and that such heterogeneity varies with the date of sampling. It is important to remember that LDL (d 1.019-1.063 g/ml) from normolipidemic humans, when isolated by our density gradient ultracentrifugation procedure, migrate as a single band under our electrophoretic conditions. At the same time, these LDL are now considered as highly heterogeneous and, in a recent paper, Shen et al. [15] have described the successful separation, by equilibrium density gradient ultracentrifugation, of six subfractions in the density interval 1.019- 1.063 g/ml. Normal human HDL (d 1.063-1.21 g/ml), which equally migrate as a single band (with a faintly stained trail) in our electrophoretic system, are equally heterogeneous both from a physicochemical and a lipid or apolipoprotein composition viewpoint. Each part of the badger plasma lipoprotein spectrum may thus be constituted from multiple discrete components. The analytical challenge now concerns the characterization and isolation of each of these lipoprotein species, as well as the determination of their individual metabolic role. This may be particularly important in view of the most probable interrelationships between at least some of these lipoproteins and the different endocrine factors, especially thyroidal and testicular, which have been shown to undergo non-synchronous annual cycles of considerable amplitude in the badger
WI. Acknowledgments Dr. M.J. Chapman is acknowledged gratefully for helpful criticism, and Dr. J. Maccario for his
valuable aid in the statistical interpretation of the results. The authors wish to thank Mr. J.C. Fage and Mr. J.P. Rambaut for their excellent technical assistance. References 1 Laplaud,
P.M., Beaubatie, L. and Maurel, D. (1980) J. Lipid Res. 21, 724-738 2 Fruchart, J.C. (1976) in Lipides et lipoprotitines (Sezille, G., Fruchart, J.C., Jaillard, J. and Dewailly, P., eds.), pp. 92-95, Crouan et Roques, Lille 3 Havel, R.J., Eder, H.A. and Bragdon, J.H. (1955) J. Clin. Invest. 34, 1345-1353 4 Chapman, M.J., Goldstein, S., Lagrange, D. and Laplaud, P.M. (1981) J. Lipid Res. 22, 339-358 5 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275 6 Kane, J.P. (1973) Anal. B&hem. 53, 350-364 7 Davis, B.J. (1964) Ann. N.Y. Acad. Sci. 121, 404-427 8 Chapman, M.J., Mills, G.L. and Ledford, J.H. (1975) Biothem. J. 149, 423-436 9 Mann, H.B. and Whitney, D.R. (1947) Ann. Math. Stat. 18, 50-60 10 Eisenberg, S. and Levy, R.I. (1975) in Advances in Lipid Research (Paoletti, R. and Kritchevsky, D., eds.), Vol. -13, pp. l-89, Academic Press, New York 11 Bittolo Bon, G., Cazzolato, G. and Avogaro, P. (1981) J. Lipid Res. 22, 998-1002 12 Swaney, J.B., Reese, H. and Eder, H.A. (1974) B&hem. Biophys. Res. Commun. 59, 513-519 13 Weisgraber, K.H., Bersot, T.P. and Mahley, R.W. (1978) B&hem. Biophys. Res. Commun. 85, 287-292 14 Mills, G.L. (1969) B&him. Biophys. Acta 194, 222-226 15 Shen, M.M.S., Krauss, R.M., Lindgren, F.T. and Forte, T.M. (1981) J. Lipid Res. 22, 236-244 16 Maurel, D., Joffre, J. and Boissin, J. (1977) CR. Acad. Sci. Paris 284 D, 1577- 1580