- Email: [email protected]
Effects of cholesterol feeding on primate serum lipoproteins
BIOCHEMICAL
MEDICINE
29,
2 14-226
( 1!%3)
Effects of Cholesterol Feeding on Primate Serum Lipoproteins III. The Change in High Density Lipoprotein CHARLES A. NELSON,’
MANFORD D. MORRIS. AND WILLIAM
of Biochemistry
Departments Sciences.
4301 South
West
Components’
and
Murkham
Rrseurch
Pediutrics. Street. Little
Institrcte.
Nen,
Unil,ersity ofArkun.ws Rock. Arkunsus 72205, Iheriu.
Louisianrr
E. GREER for Mrdictrl und Gulf
70560
Received January 14. 1981
A variety of detailed studies of nutrition and atherosclerosis have been conducted on subhuman primates, but only a limited number of species have been utilized. Addition of relatively small amounts of cholesterol to the subhuman primate diet leads to a variable response in blood lipids and lipoproteins, but most animals develop hypercholesterolemia and hyperbetalipoproteinemia. In animals whose serum cholesterol is increased, the LDL3 has an abnormal structure when compared to those of chowfed animals (1). In addition, shortly after starting animals on a cholesterol containing diet, we observed, on numerous occasions, that the HDL cholesterol was elevated. However, with increased time on the cholesterol diet, this fraction was reduced below the control diet concentration. Almost nothing is known about this biphasic response to cholesterol feeding. In this paper, we have extended cholesterol feeding studies to an additional subhuman primate, the sooty mangabey (Cercocebus UQS). an Old World monkey found in Africa. This species is of interest because the sooty mangabey has previously been reported to develop coronary atherosclerosis (2) and because our preliminary studies showed that juvenile ’ A preliminary report of this work was presented at the American Society for Biological Chemists meeting in June 1978. ’ To whom requests for reprints should be addressed: Department of Biochemistry, University of Arkansas for Medical Sciences, 4301 W. Markham St., Little Rock, Arkansas 72205.
’ Abbreviations used are: VLDL, very low density lipoproteins; lipoproteins: and HDL. high density lipoproteins. 214
0006-2944/83 Copyright All rights
$3.00 0 1983 by Academic Prey. Inc. of reproductwn in any form rewrved
LDL.
low density
CHOLESTEROL-FED
MONKEY
215
HDL
and adult sooty mangabeys had a low percentage of total plasma cholesterol transported as HDL cholesterol. We report here, the response of male and female sooty mangabeys to cholesterol feeding, and the relation of HDL*/HDL:, in response to the cholesterol-containing diet. Additionally, we have found a higher proportion of total serum cholesterol is transported in a pre-/3 lipoprotein fraction than previously observed in subhuman primates which corresponds to the lipoprotein we previously described in rhesus monkeys (3); it is similar to the Lp(a) of the human. METHODS AND MATERIALS
The sooty mangabey serum lipid and lipoprotein data cited in Table 1, were obtained from blood samples provided through the Yerkes Regional Primate Research Center in Atlanta, Georgia. Otherwise, all detailed lipoprotein studies were carried out with I1 female and 3 male sooty mangabey monkeys housed at the Gulf South Research Institute, New Iberia, Louisiana. The latter consisted of five monkeys of about 5.3 kg body wt and nine monkeys of 7.9 kg weight. No difference in lipoprotein properties was found between the two groups. The serum was obtained as previously described (1). Two baseline samples of serum were obtained while the monkeys were fed Purina monkey chow. The animals were then fed for 5 weeks a diet which contained 45% of the calories as lard. This was followed by a diet containing l.Omg/kcalcholesterol and 45% of the calories as lard fed for 5 weeks. Finally, the monkeys were returned to the Purina chow diet for 5 weeks. Serum cholesterol concentrations were measured at least two times during TABLE PLASMA
TOTAL
AND
HDL
CHOLESTEROL
AND
1
TRIGLYCERIDES
OF YERKES
SOOTY
MANGABEYS”
-
-
Cholesterol Age/sex
Number
Juvenile male
5
Juvenile female
4
Adult male
5
Adult female
6
All
20
Total 0wh-W
HDL (mg/dl)
HDL (96)
217 f 39 167-266 213 + 32 178-256 152 * 14 132-168 173 -+ 35 136-217 187 k 40
97 2 2s 59-122 84 ” 10 72-94 73 2 10 60-84 71 t 16 54-98 81 -r 9
45 t 7 35-54 39 k 3 37-44 48 -+ 8 42-61 41 k 6 32-48 43.5 r+ 7
Triacylglycerol
(m&II) 41 k 5 34-44 33 + 3 30-37 27 t 7 19-36 43 2 12 31-55 36 k 10
Note. The mean 2 SD and range are given. y Serum was provided through the courtesy of Drs. Irwin S. Bernstein and Brent Swenson, Yerkes Regional Primate Research Center, Atlanta, Georgia.
216
NELSON.
MORRIS.
AND
GREER
each diet period; at the end of the final chow feeding period, serum cholesterol concentrations had returned to the prediet concentrations. The lipoproteins were isolated by two methods: either the conventional preparative ultracentrifugal flotation method between densities 1.063 and 1.22 g/ml (HDL) or by the method of Rude1 ef uf. (4) in which ultracentrifugal isolation of total serum lipoproteins is carried out at density 1.25 g/ml through a layer of 1.22 g/ml KBr in a Beckman SW40 swinging bucket rotor at 40,000 t-pm for 40 hr followed by 6% agarose gel filtration chromatography. The recovery was 97-100% of cholesterol in the density 1.22 g/ml supernatant. VLDL, an Lp(a)-like lipoprotein, LDL,, LDL?, and HDL are separated in order as described (3) from 2.4 x 85cm columns of 6% agarose (Bio-Rad Laboratories, Richmond, Calif.) eluting at a flow rate of 10 ml/hr in the cold. The purity of the isolated lipoproteins was established by agarose electrophoresis (5) and immunoelectrophoresis (6). No albumin or other serum lipoproteins were present. Furthermore. their presence was not seen when the samples were subjected to analytical ultracentrifugation. Flotation velocity in the analytical ultracentrifuge was carried out at density 1.2 g/ml at 47,660 or 48,000 rpm using a double sector cell to include a baseline for the floating lipoprotein analytical ultracentrifugal schlieren peak. The HDL2 and HDL3 areas under the often asymmetric peak were estimated by first normalizing to a linear baseline, then drawing each HDL component symmetrically with the leading edge of HDLz and/ or the trailing edge of HDL,. The complete HDLz and HDL3 components drawn in under the schlieren peak were symmetrical and left no area unaccounted for. These areas were corrected for the increased concentration of the faster floating component in the narrower upper part of the sector shaped cell. No correction was made for the Johnston-Ogston effect (7) since Anderson ef al. (8) reported it to be small, and we have found that mixtures of purified HDLz and HDL, yielded areas which were similar to those obtained for each fraction alone. Dilution of mixtures of HDLz and HDL3 yielded identical percentage areas for all concentrations. The Sf measured from the apparent peak of the major component was corrected to the true Sf by comparison to the actual peak observed when the baseline was normalized and the HDLz and HDL3 peaks were drawn in. In a similar manner, the Sr of a minor component could be calculated from the peak of that component if it were greater than 20% of the total. The actual densities of the solvent used to determine Sf were measured in a pycnometer and a slight correction to Sr, 1.2000, was made to the measured Sr based on partial specific volumes of 0.90 ml/g for HDLz and 0.87 ml/g for HDLj (9). Other analytical methods have been reported previously (1). All statistical notations are standard deviations. Since measurements were obtained from chow-fed animals both before
CHOLESTEROL-FED
MONKEY
217
HDL
and 5 weeks after cholesterol feeding, it was essential to determine the completeness of the return of HDL structure to prediet conditions. The sera of some chow-fed monkeys (5 weeks after discontinuing cholesterolfat feeding) were spun at density 1.22 g/ml as described. The supematant solution was then dialyzed directly against solvent of density 1.2 g/ml and studied in the analytical ultracentrifuge. Of five monkeys which had been studied before cholesterol feeding, four were within 9% of the percentage HDL,? (two higher and two lower) found originally, while they averaged 4 mg/dl differences in the HDL cholesterol in these determinations which were 5 months apart. The HDL cholesterol of the remaining monkey was increased 50% while the percentage HDLz rose 27%. RESULTS Serum and HDL cholesterol by sex and age. We initially surveyed a colony of 20 free-ranging sooty mangabeys and observed increased total serum cholesterol and HDL cholesterol in juvenile male and female animals when compared to adults of both sexes (Table I). Although the mean total cholesterol concentration in the juvenile animals was greater than 200 mg/dl, there was no increased triacylglycerol when compared to the adult animals. In adult females, the mean value for total serum cholesterol was about 20 mg/dl greater than in males although the HDL cholesterol of males and females was similar. Thus, only 41% of the total serum cholesterol was found in HDL in adult females, whereas 48% was present HDL in adult males. Lipoprotein in response to fat and cholesterol. The mean and SD of total and HDL serum cholesterol concentrations for the 14 monkeys which were included in the cholesterol-fat diet studies are shown in Table 2. There was a 23% mean rise in total serum and 17% increase in HDL cholesterol on the fat diet. After the animals had been placed TABLE 2 CHANGESIN TOTAL AND HDL CHOLESEROL
ON VARIOUS
DIETS
Serum mg/dl Diet (N = 14)
Time on diet
Chow Fat Cholesterol-fat Cholesterol-fat Chow
1 month l-3 weeks” 5 weeks 5 weeks
Total cholesterol 14s 179 328 398 148
” it +-e
19 26 53 96 12
HDL cholesterol 52 61 80 45 57
If: k k 2 k
7 13 10 17 II
” Each monkey (except No. 6005) was taken at the time (1-3 weeks) of its highest HDL cholesterol.
218
NELSON.
MORRIS.
AND
GREEK
on the cholesterol-fat diet for l-3 weeks. all monkeys except one male (No. 6005) exhibited an absolute increase in HDL cholesterol. All increased in total cholesterol. The HDL cholesterol increase constituted only about 15% of the observed increase in total cholesterol. After 5 weeks of cholesterol-fat feeding, the mean HDL cholesterol concentration decreased 13% below the chow-diet concentration, while the total cholesterol continued to increase in all monkeys. The distribution of serum cholesterol among the various lipoprotein fractions from some mangabey monkeys was determined on the chow diet and at two time intervals on the cholesterol-fat diet as shown in Table 3. After l-3 weeks on the cholesterol-fat diet, in these representative animals the total serum cholesterol doubled, the LDL cholesterol rose l55%, and the HDL cholesterol rose 50%. All of the HDL rise was confined to HDL?, while HDLi did not change. When the serum cholesterol of five monkeys had increased to 464 mg/dl and the LDL was about six times greater than baseline, the HDL decreased to only 46% of the chowfed mean concentration. HDL, subsequently decreased greatly to just below its original concentration, while HDL3 also decreased lower than originally found. VLDL also increased about seven times on the cholesterol-fat diet, constituting about 2% of the total serum cholesterol. HDL component ~han~rs. On the Purina chow diet, HDLz and HDL-( were generally poorly separated. even at high speed in the analytical ultracentrifuge (Fig. I, bottom). After 1-3 weeks of cholesterol feeding, when the total HDL concentration increased, the HDLz and HDL7 were better separated when floating at high speed (Fig. I, middle). After further cholesterol feeding, when the HDL concentration had decreased much lower than in the chow-fed state, in some monkeys only HDL3 was visible (Fig. I, top). The same eight mangabey monkeys were compared in percentage of HDLl and HDL, before and after l-3 weeks of cholesterol feeding to avoid the variability of the original chow-fed HDL?IHDL, ratios among monkeys (Table 4). All but one monkey rose in percentage HDL?, the mean increase in HDLz being 28 i 16%. HDL,, at this time, represented over 50% of the total HDL. Later, after 5 weeks of cholesterol feeding, the percentage of HDL: fell 25%, near the original HDL? amount on the chow diet. Two monkeys whose serum cholesterol was 415 and 581 mg/dl, exhibited no HDL, (see Fig. I, top). The percentage of HDLz in all monkeys was highly correlated with the concentration of HDL cholesterol. Figure 2 is such a plot of all of the mangabey data. The correlation coefficient of percentage of HDLz with HDL cholesterol concentration was 0.77, P < 0.001. Molecular rxclusion projiles. Elution profiles of total HDL from 6% agarose columns were symmetrical while the monkeys ingested Purina chow diets (Fig. 3, top). From every mangabey monkey fed the cholesterol-
2% 2 23
464 2 80
8
5
5
Chow Cholesterol-fat (1-3 weeks) Cholesterol-fat (5 weeks)
923
1.4 k 1.4
1.2 2 0.7
VLDL
(MGJDL
19 2 8
17 2 5
I-p(a)
DISTRIBUTION
3
420 5 89
1.3 2 1.5
1.8 _’ 1.4
LDL,
OF CHOLESTEROL)
TABLE
184 2 22
72 k 22
LDLz
IN MANCABEY
28 t 13
90 2 12
61 +- 11
HDL
SERUMS
37 2 11
93 2 10
54 2 to
HDL (Hp-Mn)”
and HDL were
36 2 8
35 k 12
19 I!z 15 57 ? 10
HDL,’
HDLzb
” Hp-Mn is the heparin-manganese precipitation method of determining HDL cholesterol (1 I) VLDL, Lp(a), LDL,, LDL,, separated as described in Methods. ’ Calculated from weight percentage assuming 20% cholesterol in HDLz and 17% in HDL, (10). ’ includes Lp(a), LDL,, and LDL,. The increased concentration of the LDL fraction overlaps the other two fractions.
153 -c 32
N
Diet
Serum cholesterol
LIFVPROTEIN
s F
2
2 E p
in
FIG.
I.
(Top)
Monkey
No.
6005
HDL
tS,
=
3.0)
on
a cholesterol-fat
diet
having
an
HDL cholesterol of 32 mgidl. (Middle) Monkey No. 6007 HDL (S, = 7.3) on a cholesterolfat diet having an HDL cholesterol of 80 mgidl. (Bottom) Monkey No. 6007 HDL (S, of fast component = 7.3) on a chow diet having an HDL cholesterol of 45 mg/dl. The pictures were taken at 64 min after reaching speed. Flotation is from right to left in I.7 giml KBr at 48,000 molecular various
rpm. 25”. at a schlieren weight moves faster. diets.
bar angle Note the
of 60”. relative 220
HDL, being positions
of lower of HDL,
density and and HDI.,
higher on the
I
8
10
24
24 2 22
56 5
28 f
HDL,
76 *
22
44 ‘- 10
72 2 24
HDL,
WITH
TABLE
f
16
- 25b k 25
+28
4
474
k 83
306 k 59
143 ‘- 20
Serum
Cholesterol
TIME ON CHOLESTEROL ~ ~-___--~
Change in HDLz
CHANGES
7
II
2 II
34 2
82
522
HDL
(mg/dl)
DIET
a The same eight monkeys were used in both chow and cholesterol-fat (l-3 weeks) diets. b Decrease from the HDL, concentration at 5 weeks from that on the chow diet of each of four monkeys. ’ The concentration of HDLz was too low to measure S,.
Chow Cholesterol-fat (l-3 weeks) Cholesterol-fat (5 weeks)
N
COMPONENT
Weight percentage
-.~
HDL
7.0(J)
J.2(6)
0.3
.__
k 0.2
f
HDLz
(7)
2.8 (7)
_’ 0.2
2 0.6
+ 0.4
HDL,
3.15(3)
4.2
-
Mean S, (N) z
2
s
8
P
‘22
NELSON.
MORRIS. AND GREEK
60
20 SERUM
40 HDL
60 80 CONCENTRATION
100 mg/dl
FIG. 2. The weight percentage of HDLz of total HDL as a function of HDL cholesterol concentration for all the data obtained for the 13 sooty mangabey monkeys at various times on various diets is shown. Open circles (0). chow diet: half-closed circles Cc)), the l-3 week increase in HDL cholesterol; closed circles CO), 5 weeks on the cholesterol diet. The correlation coefficient was 0.77 and was significant at P < 0.001.
fat diet, the HDL eluted asymmetrically, and appeared to have two components (Fig. 3, middle). A typical pattern where similar amounts of the two components were very clearly separated is shown in Fig. 3. bottom. Calculation of the percentage of each component from the uv absorbancy while correcting for the different percentage protein in HDLz 0.2
0
300
250 ELUTION
350 VOLUME
ml
FIG. 3. Six-percent agarose gel filtration patterns of HDL from mangabey No. 6007 on: (TOP) a chow diet, closed circles (0); (MIDDLE) from the cholesterol-fat diet at I week, half-closed circles (8); and (BOTTOM) at 5 weeks, open circles (0). The elution volumes are not strictly comparable since three different columns were used. The flotation velocity patterns of the chow-fed and I week cholesterol-fed HDL are given in Fig. I.
CHOLESTEROL-FED
MONKEY
HDL
223
and HDL, agreed very well with that determined by the analytical ultracentrifugal method. The improved separation of HDLz and HDL3 from cholesterol-fat-fed monkeys by molecular exclusion column chromatography suggested that on the cholesterol-fat diet, HDLz and HDL, increased their size difference. Flotation rates. The relation among percentage HDL2 and HDL3, total HDL cholesterol concentration, and the Sf of the major HDL components during the chow and cholesterol-fat diet periods are shown in Table 4. The HDS Sf rates were considerably smaller (2.8 vs 4.2) after 5 weeks on the cholesterol-fat diet when compared to the chow diet while the HDLz Sf rates were almost identical (7.0 vs 7.2) during the l-3 week cholesterol-fat time period. (Five week HDL, concentrations were too low to measure S,.) This increased difference between the Sf of HDL> agrees with the previously discussed increased separation on molecular sizing gel filtration. Typical schlieren photographs of HDL from monkeys on the indicated diets taken at identical times are shown in Fig. 1. It can be observed that the HDL3 (the peak on the right in all photographs) floated slower when the cholesterol-fat diet was fed. In Fig. I, top, after 5 weeks on the cholesterol-fat diet, HDL3 remained very near the cell bottom (right edge) while in Fig. 1, bottom, on the chow diet, HDL, had moved farther away from the right-hand edge. DISCUSSION The present studies are significant in that they add another subhuman primate to the list of those that are highly responsive to cholesterol feeding by increasing both plasma cholesterol and LDL. A small but significant increase in total cholesterol and HDL cholesterol was observed when the animals ingested a high fat diet as compared to pelleted monkey chow. In addition, for the first time, studies specifically designed to study such changes in HDL subfractions during cholesterol feeding are reported. Addition of cholesterol to the high fat diet resulted in a prompt and sustained increase in plasma cholesterol and substantial alteration in its distribution among lipoproteins. Prior to fat or cholesterol feeding only 36% of plasma cholesterol was transported as HDL cholesterol. This percentage of HDL cholesterol was low for an Old World monkey (12). but not as low as those of some New World monkeys: spider 9% and woolly 8% (of the total cholesterol as HDL) being the lowest (12,13). The major change in the plasma lipoproteins was a marked increase in LDL during cholesterol-fat feeding. After 5 weeks on the cholesterolfat diet, LDL cholesterol accounted for greater than 85% of the total cholesterol. Five weeks after return to a chow diet, the total and HDL cholesterol were at the prediet concentration. In several primate species, when total serum cholesterol is increased
724
NELSON.
MORRIS.
AND
GREEK
to a concentration of 200-450 mg/dl. it has generally been found that an increase in serum HDL also occurs (1 l-1.5). A further elevation in total serum cholesterol values is associated with a serum HDL concentration which is lower than controls (16-19). In rhesus monkeys, a small increase in HDL during cholesterol feeding resulted in a slightly increased ratio of 1.125 g/ml floating HDL to sedimenting HDL (I I ). while the decrease in HDL (macaque monkeys) resulted in a decreased ratio of I. I25 g/ml floating HDL to sedimenting HDL C16). Unexpectedly. the changes in HDL in response to cholesterol-fat feeding were associated with several new observations. First, during the early period of induction of hypercholesterolemia, the increase in HDL cholesterol was found in HDL? only. Second, during the latter part of the cholesterol-fat diet period, as total plasma cholesterol increased further, both HDLz and HDL, were decreased to levels below that found on the chow diet. In several monkeys, HDL, was undetectable during the latter stages of cholesterol-fat feeding regimen. Finally, an additional and unique observation was that HDL isolated from cholesterol-fat-fed sooty mangabeys separated into two components upon agarose column chromatography gel filtration where components are separated according to molecular size. During the early period of cholesterol-fat feeding, when plasma cholesterol had increased about two times and HDL cholesterol was at its highest serum concentration, the HDL profile from 6% agarose gel chromatography revealed a skew to the right indicative of increased heterogeneity in the HDL fraction (Fig. 3, middle). At the end of the cholesterol-fat feeding period (5 weeks) the agarose column pattern clearly revealed two peaks in the purified HDL of many monkeys. Thus. HDL? and HDL, increased in size difference on cholesterol-fat diets. During cholesterol-fat feeding, it was the HDL3 that became progressively altered structurally so that by the end of 5 weeks on this diet, the Sf was reduced by 33%. from S, 4.2 to 2.8. These data indicate that the HDL? was smaller and/or denser as a result of cholesterol-fat feeding. Such alterations in the metabolism of HDL subfractions, the specifically altered concentration of circulating HDLl during the early period of cholesterol-fat feeding, and the apparent selective change in the physical properties of HDL, after cholesterol ingestion, have not been reported previously. We have previously reported that the HDL of chow-fed rhesus monkeys is eluted from 6% agarose columns as a single symmetrical peak Cl). Cholesterol-fed rhesus and grivet monkeys have likewise not been reported to show HDL heterogeneity when characterized by gel filtration chromatography (20,211. In both species, the HDL may have been primarily HDL, judging from the decreased serum HDL concentration. Other in-
CHOLESTEROL-FED
MONKEY
HDL
225
vestigators have not looked for heterogeneity, but have arbitrarily divided subhuman primate HDL into two density fractions and found differences in chemical composition (10,ll). Our attempts to study the metabolismof HDL in rhesus monkeys have been complicated by the failure to separate HDL, and HDL, by gel filtration chromatography and the lack of obvious heterogeneity of the chow-fed rhesus monkey HDL. It is apparent that the existing analytical separation techniques are still too limited to permit a precise definition of all HDL subfractions. However, based on our present data in the HDL of the sooty mangabey, this animal appears to be the preferred one to study changes in HDL metabolism during cholesterol feeding. When the mean flotation rates of chow-fed sooty mangabey HDLz and HDL, are compared to their corresponding human fractions, the rates are somewhat faster. The mangabey Sr for HDLz was 7.1 + 0.3 compared to human 6.4 + 0.4 (N = 8) (our unpublished data), whereas HDL, was Sf 4.2 ? 0.4 in the mangabeys and 3.35 + 0.3 (N = 33) in the humans. After the mangabeys had ingested the cholesterol-fat diet for 5 weeks, the HDL, Sf was quite similar to that observed in human HDL+ We are unaware of reports in humans comparing the properties of HDL subfractions during diet-induced alterations of cholesterol and/or fat intake. In addition to the observations on HDL, the sooty mangabey plasma lipoproteins were characterized by a higher concentration of the pre+ non-VLDL lipoprotein, Lp(a), in every mangabey when compared to other monkey species. During cholesterol feeding this peak becomes obscured when lipoproteins were isolated column chromatographically so we were unable to judge changes in this fraction. SUMMARY
Sooty mangabey (Cercocebus atys) monkeys had a lower serum HDL cholesterol concentration than any other Old World monkey species reported. In addition, they had a higher serum Lp(a) concentration than other species. The mangabeys were fed a cholesterol-fat diet for 5 weeks. HDLz and HDL3 amounts were determined from the two peaks apparent upon analytical ultracentrifugation. In the first l-3 weeks, 13 of the 14 mangabeys increased 30% (mean) in total HDL, this increase occurring only in the HDL, fraction. After 5 weeks, HDL and HDLz decreased markedly. During the cholesterol feeding, HDL3 continually decreased in flotation rate, indicating it was either smaller and/or denser. HDL2 and HDL3 separated well on molecular sieving agarose columns during the diet period, whereas a single symmetrical elution peak was found for chow-fed HDL. Thus on a cholesterol-fat diet, HDLz and HDLl increased in difference in molecular size.
226
NELSON,
MORRIS. AND CREER
ACKNOWLEDGMENTS We would like to acknowledge the expert technicaf assistance of Rosa Haidar. We also appreciate and acknowledge Dr. Irwin S. Bernstein and Dr. Brent Swenson of the Yerkes Regional Primate Research Center. Emory University, Atlanta, Geargia for obtaining the serum sampies from the free ranging mangabeys. They were supported by NIH Core Research Grant RR00165 to the Yerkes Regional Primate Center and NIH Grant 13864.
REFERENCES I. 2. 3. 4. 5. 6. 7. 8.
Nelson, C. A., and Morris, M. D., &&em. &fed. $7, 3tQ (t9771. Hamerton, A. E.. Pruc. Zuol. Sot. Lnndor? 111, IS1 f 1941). Nelson, C. A., and Morris, M. IX. Biuckem. B&&s. Res. Cnmmun. 71, 438 (1976). Rude], L. L., Lee, 1. A., Morris, M. D., and Felts, J. M., ~~~~h~rn. f. 139, 89 (1976). Noble, R. P., J. Lipid Res. 9, 693 (1468). Scheidegger, J. J., Int. Arch. Allergy Appl. Immunol. 7, 103 (1955). Johnston, J. P., and Ogston, A. G.. Tucms. Faroduy SW. 42, 789 fi946). Anderson, D. W. ~Nichols, A. V ., Forte. T. M.. and Lindgren, F. T., ~~u~k~rn. Biopkys. Acta 493, 55 (1977). 9, Ha&wood, R. N., I. Amer. Ckmm. Sue. 80, 2152 (19571. IO, Scanu, A. M., Edelstein, C., Vitello. L., Jones, R.. and Wissler, R,, J. Biol. Ckem. 248, 7648 (1973). Il. Burstein, M., and Samille, J., C/in. Ckim. Acta 5, 609 f1960). 12. Rudei, L. L., and Lofland, W. B.. Jr.. in “Primates in Medicine,” (J. P. Strong. Ed.), 9th ed., p. 224. Karger, Base!, 1976. 13. Lofland, H. B., St. Cfair, R. W.. MacNintch, J. E., and Prichard, R. W,, Arch. ~~t~~~~, 83, 211 (1967). 14. Portman, 0. W., Alexander, M.. Neuringer, M., fllingworth, D. R., and Alam, S. S,, f. Nutr. 11X, 733 (1981). IS. Kushwaha, R. S,, Hazzard, W. R., Harker, L. A., and Engblom, J., ~~~?~~~s~~e~~.~j~~ 31, 65 f197fQ. 16. &show, A. C., Nicolosi, R. J., and Hayes, K. C.. Amer. J. Cfin. Nutr. 34,830 (19811. 17. Srinivasan, S. R.. Rad~akrisbnamu~by, 3.. Dalferes, E. R., Jr., and Berenson. G. S., Lipids
14, 559 (1X79),
18. Srinivasan, S. R., Radhakrishnamurthy, B., Smith, C. C.. Wolf, R. H., and Berenson. Cl. S.. J. Nutr. 106, 1757 (1976). 19. Nelson, C. A.. Casaii. R. E., and Morris, M. D., Biackem. Med. 22, 110 f 1979). 20. Rudel, L. L., Pit&, L. L., II, and Nelson, C. A.. 1. Lipid Res. 18, 21t (19771. 21. Rude!, L. L., Shah. R., and Greene, D. C., J. Lipid Res. 20, 55 (197%.
MEDICINE
29,
2 14-226
( 1!%3)
Effects of Cholesterol Feeding on Primate Serum Lipoproteins III. The Change in High Density Lipoprotein CHARLES A. NELSON,’
MANFORD D. MORRIS. AND WILLIAM
of Biochemistry
Departments Sciences.
4301 South
West
Components’
and
Murkham
Rrseurch
Pediutrics. Street. Little
Institrcte.
Nen,
Unil,ersity ofArkun.ws Rock. Arkunsus 72205, Iheriu.
Louisianrr
E. GREER for Mrdictrl und Gulf
70560
Received January 14. 1981
A variety of detailed studies of nutrition and atherosclerosis have been conducted on subhuman primates, but only a limited number of species have been utilized. Addition of relatively small amounts of cholesterol to the subhuman primate diet leads to a variable response in blood lipids and lipoproteins, but most animals develop hypercholesterolemia and hyperbetalipoproteinemia. In animals whose serum cholesterol is increased, the LDL3 has an abnormal structure when compared to those of chowfed animals (1). In addition, shortly after starting animals on a cholesterol containing diet, we observed, on numerous occasions, that the HDL cholesterol was elevated. However, with increased time on the cholesterol diet, this fraction was reduced below the control diet concentration. Almost nothing is known about this biphasic response to cholesterol feeding. In this paper, we have extended cholesterol feeding studies to an additional subhuman primate, the sooty mangabey (Cercocebus UQS). an Old World monkey found in Africa. This species is of interest because the sooty mangabey has previously been reported to develop coronary atherosclerosis (2) and because our preliminary studies showed that juvenile ’ A preliminary report of this work was presented at the American Society for Biological Chemists meeting in June 1978. ’ To whom requests for reprints should be addressed: Department of Biochemistry, University of Arkansas for Medical Sciences, 4301 W. Markham St., Little Rock, Arkansas 72205.
’ Abbreviations used are: VLDL, very low density lipoproteins; lipoproteins: and HDL. high density lipoproteins. 214
0006-2944/83 Copyright All rights
$3.00 0 1983 by Academic Prey. Inc. of reproductwn in any form rewrved
LDL.
low density
CHOLESTEROL-FED
MONKEY
215
HDL
and adult sooty mangabeys had a low percentage of total plasma cholesterol transported as HDL cholesterol. We report here, the response of male and female sooty mangabeys to cholesterol feeding, and the relation of HDL*/HDL:, in response to the cholesterol-containing diet. Additionally, we have found a higher proportion of total serum cholesterol is transported in a pre-/3 lipoprotein fraction than previously observed in subhuman primates which corresponds to the lipoprotein we previously described in rhesus monkeys (3); it is similar to the Lp(a) of the human. METHODS AND MATERIALS
The sooty mangabey serum lipid and lipoprotein data cited in Table 1, were obtained from blood samples provided through the Yerkes Regional Primate Research Center in Atlanta, Georgia. Otherwise, all detailed lipoprotein studies were carried out with I1 female and 3 male sooty mangabey monkeys housed at the Gulf South Research Institute, New Iberia, Louisiana. The latter consisted of five monkeys of about 5.3 kg body wt and nine monkeys of 7.9 kg weight. No difference in lipoprotein properties was found between the two groups. The serum was obtained as previously described (1). Two baseline samples of serum were obtained while the monkeys were fed Purina monkey chow. The animals were then fed for 5 weeks a diet which contained 45% of the calories as lard. This was followed by a diet containing l.Omg/kcalcholesterol and 45% of the calories as lard fed for 5 weeks. Finally, the monkeys were returned to the Purina chow diet for 5 weeks. Serum cholesterol concentrations were measured at least two times during TABLE PLASMA
TOTAL
AND
HDL
CHOLESTEROL
AND
1
TRIGLYCERIDES
OF YERKES
SOOTY
MANGABEYS”
-
-
Cholesterol Age/sex
Number
Juvenile male
5
Juvenile female
4
Adult male
5
Adult female
6
All
20
Total 0wh-W
HDL (mg/dl)
HDL (96)
217 f 39 167-266 213 + 32 178-256 152 * 14 132-168 173 -+ 35 136-217 187 k 40
97 2 2s 59-122 84 ” 10 72-94 73 2 10 60-84 71 t 16 54-98 81 -r 9
45 t 7 35-54 39 k 3 37-44 48 -+ 8 42-61 41 k 6 32-48 43.5 r+ 7
Triacylglycerol
(m&II) 41 k 5 34-44 33 + 3 30-37 27 t 7 19-36 43 2 12 31-55 36 k 10
Note. The mean 2 SD and range are given. y Serum was provided through the courtesy of Drs. Irwin S. Bernstein and Brent Swenson, Yerkes Regional Primate Research Center, Atlanta, Georgia.
216
NELSON.
MORRIS.
AND
GREER
each diet period; at the end of the final chow feeding period, serum cholesterol concentrations had returned to the prediet concentrations. The lipoproteins were isolated by two methods: either the conventional preparative ultracentrifugal flotation method between densities 1.063 and 1.22 g/ml (HDL) or by the method of Rude1 ef uf. (4) in which ultracentrifugal isolation of total serum lipoproteins is carried out at density 1.25 g/ml through a layer of 1.22 g/ml KBr in a Beckman SW40 swinging bucket rotor at 40,000 t-pm for 40 hr followed by 6% agarose gel filtration chromatography. The recovery was 97-100% of cholesterol in the density 1.22 g/ml supernatant. VLDL, an Lp(a)-like lipoprotein, LDL,, LDL?, and HDL are separated in order as described (3) from 2.4 x 85cm columns of 6% agarose (Bio-Rad Laboratories, Richmond, Calif.) eluting at a flow rate of 10 ml/hr in the cold. The purity of the isolated lipoproteins was established by agarose electrophoresis (5) and immunoelectrophoresis (6). No albumin or other serum lipoproteins were present. Furthermore. their presence was not seen when the samples were subjected to analytical ultracentrifugation. Flotation velocity in the analytical ultracentrifuge was carried out at density 1.2 g/ml at 47,660 or 48,000 rpm using a double sector cell to include a baseline for the floating lipoprotein analytical ultracentrifugal schlieren peak. The HDL2 and HDL3 areas under the often asymmetric peak were estimated by first normalizing to a linear baseline, then drawing each HDL component symmetrically with the leading edge of HDLz and/ or the trailing edge of HDL,. The complete HDLz and HDL3 components drawn in under the schlieren peak were symmetrical and left no area unaccounted for. These areas were corrected for the increased concentration of the faster floating component in the narrower upper part of the sector shaped cell. No correction was made for the Johnston-Ogston effect (7) since Anderson ef al. (8) reported it to be small, and we have found that mixtures of purified HDLz and HDL, yielded areas which were similar to those obtained for each fraction alone. Dilution of mixtures of HDLz and HDL3 yielded identical percentage areas for all concentrations. The Sf measured from the apparent peak of the major component was corrected to the true Sf by comparison to the actual peak observed when the baseline was normalized and the HDLz and HDL3 peaks were drawn in. In a similar manner, the Sr of a minor component could be calculated from the peak of that component if it were greater than 20% of the total. The actual densities of the solvent used to determine Sf were measured in a pycnometer and a slight correction to Sr, 1.2000, was made to the measured Sr based on partial specific volumes of 0.90 ml/g for HDLz and 0.87 ml/g for HDLj (9). Other analytical methods have been reported previously (1). All statistical notations are standard deviations. Since measurements were obtained from chow-fed animals both before
CHOLESTEROL-FED
MONKEY
217
HDL
and 5 weeks after cholesterol feeding, it was essential to determine the completeness of the return of HDL structure to prediet conditions. The sera of some chow-fed monkeys (5 weeks after discontinuing cholesterolfat feeding) were spun at density 1.22 g/ml as described. The supematant solution was then dialyzed directly against solvent of density 1.2 g/ml and studied in the analytical ultracentrifuge. Of five monkeys which had been studied before cholesterol feeding, four were within 9% of the percentage HDL,? (two higher and two lower) found originally, while they averaged 4 mg/dl differences in the HDL cholesterol in these determinations which were 5 months apart. The HDL cholesterol of the remaining monkey was increased 50% while the percentage HDLz rose 27%. RESULTS Serum and HDL cholesterol by sex and age. We initially surveyed a colony of 20 free-ranging sooty mangabeys and observed increased total serum cholesterol and HDL cholesterol in juvenile male and female animals when compared to adults of both sexes (Table I). Although the mean total cholesterol concentration in the juvenile animals was greater than 200 mg/dl, there was no increased triacylglycerol when compared to the adult animals. In adult females, the mean value for total serum cholesterol was about 20 mg/dl greater than in males although the HDL cholesterol of males and females was similar. Thus, only 41% of the total serum cholesterol was found in HDL in adult females, whereas 48% was present HDL in adult males. Lipoprotein in response to fat and cholesterol. The mean and SD of total and HDL serum cholesterol concentrations for the 14 monkeys which were included in the cholesterol-fat diet studies are shown in Table 2. There was a 23% mean rise in total serum and 17% increase in HDL cholesterol on the fat diet. After the animals had been placed TABLE 2 CHANGESIN TOTAL AND HDL CHOLESEROL
ON VARIOUS
DIETS
Serum mg/dl Diet (N = 14)
Time on diet
Chow Fat Cholesterol-fat Cholesterol-fat Chow
1 month l-3 weeks” 5 weeks 5 weeks
Total cholesterol 14s 179 328 398 148
” it +-e
19 26 53 96 12
HDL cholesterol 52 61 80 45 57
If: k k 2 k
7 13 10 17 II
” Each monkey (except No. 6005) was taken at the time (1-3 weeks) of its highest HDL cholesterol.
218
NELSON.
MORRIS.
AND
GREEK
on the cholesterol-fat diet for l-3 weeks. all monkeys except one male (No. 6005) exhibited an absolute increase in HDL cholesterol. All increased in total cholesterol. The HDL cholesterol increase constituted only about 15% of the observed increase in total cholesterol. After 5 weeks of cholesterol-fat feeding, the mean HDL cholesterol concentration decreased 13% below the chow-diet concentration, while the total cholesterol continued to increase in all monkeys. The distribution of serum cholesterol among the various lipoprotein fractions from some mangabey monkeys was determined on the chow diet and at two time intervals on the cholesterol-fat diet as shown in Table 3. After l-3 weeks on the cholesterol-fat diet, in these representative animals the total serum cholesterol doubled, the LDL cholesterol rose l55%, and the HDL cholesterol rose 50%. All of the HDL rise was confined to HDL?, while HDLi did not change. When the serum cholesterol of five monkeys had increased to 464 mg/dl and the LDL was about six times greater than baseline, the HDL decreased to only 46% of the chowfed mean concentration. HDL, subsequently decreased greatly to just below its original concentration, while HDL3 also decreased lower than originally found. VLDL also increased about seven times on the cholesterol-fat diet, constituting about 2% of the total serum cholesterol. HDL component ~han~rs. On the Purina chow diet, HDLz and HDL-( were generally poorly separated. even at high speed in the analytical ultracentrifuge (Fig. I, bottom). After 1-3 weeks of cholesterol feeding, when the total HDL concentration increased, the HDLz and HDL7 were better separated when floating at high speed (Fig. I, middle). After further cholesterol feeding, when the HDL concentration had decreased much lower than in the chow-fed state, in some monkeys only HDL3 was visible (Fig. I, top). The same eight mangabey monkeys were compared in percentage of HDLl and HDL, before and after l-3 weeks of cholesterol feeding to avoid the variability of the original chow-fed HDL?IHDL, ratios among monkeys (Table 4). All but one monkey rose in percentage HDL?, the mean increase in HDLz being 28 i 16%. HDL,, at this time, represented over 50% of the total HDL. Later, after 5 weeks of cholesterol feeding, the percentage of HDL: fell 25%, near the original HDL? amount on the chow diet. Two monkeys whose serum cholesterol was 415 and 581 mg/dl, exhibited no HDL, (see Fig. I, top). The percentage of HDLz in all monkeys was highly correlated with the concentration of HDL cholesterol. Figure 2 is such a plot of all of the mangabey data. The correlation coefficient of percentage of HDLz with HDL cholesterol concentration was 0.77, P < 0.001. Molecular rxclusion projiles. Elution profiles of total HDL from 6% agarose columns were symmetrical while the monkeys ingested Purina chow diets (Fig. 3, top). From every mangabey monkey fed the cholesterol-
2% 2 23
464 2 80
8
5
5
Chow Cholesterol-fat (1-3 weeks) Cholesterol-fat (5 weeks)
923
1.4 k 1.4
1.2 2 0.7
VLDL
(MGJDL
19 2 8
17 2 5
I-p(a)
DISTRIBUTION
3
420 5 89
1.3 2 1.5
1.8 _’ 1.4
LDL,
OF CHOLESTEROL)
TABLE
184 2 22
72 k 22
LDLz
IN MANCABEY
28 t 13
90 2 12
61 +- 11
HDL
SERUMS
37 2 11
93 2 10
54 2 to
HDL (Hp-Mn)”
and HDL were
36 2 8
35 k 12
19 I!z 15 57 ? 10
HDL,’
HDLzb
” Hp-Mn is the heparin-manganese precipitation method of determining HDL cholesterol (1 I) VLDL, Lp(a), LDL,, LDL,, separated as described in Methods. ’ Calculated from weight percentage assuming 20% cholesterol in HDLz and 17% in HDL, (10). ’ includes Lp(a), LDL,, and LDL,. The increased concentration of the LDL fraction overlaps the other two fractions.
153 -c 32
N
Diet
Serum cholesterol
LIFVPROTEIN
s F
2
2 E p
in
FIG.
I.
(Top)
Monkey
No.
6005
HDL
tS,
=
3.0)
on
a cholesterol-fat
diet
having
an
HDL cholesterol of 32 mgidl. (Middle) Monkey No. 6007 HDL (S, = 7.3) on a cholesterolfat diet having an HDL cholesterol of 80 mgidl. (Bottom) Monkey No. 6007 HDL (S, of fast component = 7.3) on a chow diet having an HDL cholesterol of 45 mg/dl. The pictures were taken at 64 min after reaching speed. Flotation is from right to left in I.7 giml KBr at 48,000 molecular various
rpm. 25”. at a schlieren weight moves faster. diets.
bar angle Note the
of 60”. relative 220
HDL, being positions
of lower of HDL,
density and and HDI.,
higher on the
I
8
10
24
24 2 22
56 5
28 f
HDL,
76 *
22
44 ‘- 10
72 2 24
HDL,
WITH
TABLE
f
16
- 25b k 25
+28
4
474
k 83
306 k 59
143 ‘- 20
Serum
Cholesterol
TIME ON CHOLESTEROL ~ ~-___--~
Change in HDLz
CHANGES
7
II
2 II
34 2
82
522
HDL
(mg/dl)
DIET
a The same eight monkeys were used in both chow and cholesterol-fat (l-3 weeks) diets. b Decrease from the HDL, concentration at 5 weeks from that on the chow diet of each of four monkeys. ’ The concentration of HDLz was too low to measure S,.
Chow Cholesterol-fat (l-3 weeks) Cholesterol-fat (5 weeks)
N
COMPONENT
Weight percentage
-.~
HDL
7.0(J)
J.2(6)
0.3
.__
k 0.2
f
HDLz
(7)
2.8 (7)
_’ 0.2
2 0.6
+ 0.4
HDL,
3.15(3)
4.2
-
Mean S, (N) z
2
s
8
P
‘22
NELSON.
MORRIS. AND GREEK
60
20 SERUM
40 HDL
60 80 CONCENTRATION
100 mg/dl
FIG. 2. The weight percentage of HDLz of total HDL as a function of HDL cholesterol concentration for all the data obtained for the 13 sooty mangabey monkeys at various times on various diets is shown. Open circles (0). chow diet: half-closed circles Cc)), the l-3 week increase in HDL cholesterol; closed circles CO), 5 weeks on the cholesterol diet. The correlation coefficient was 0.77 and was significant at P < 0.001.
fat diet, the HDL eluted asymmetrically, and appeared to have two components (Fig. 3, middle). A typical pattern where similar amounts of the two components were very clearly separated is shown in Fig. 3. bottom. Calculation of the percentage of each component from the uv absorbancy while correcting for the different percentage protein in HDLz 0.2
0
300
250 ELUTION
350 VOLUME
ml
FIG. 3. Six-percent agarose gel filtration patterns of HDL from mangabey No. 6007 on: (TOP) a chow diet, closed circles (0); (MIDDLE) from the cholesterol-fat diet at I week, half-closed circles (8); and (BOTTOM) at 5 weeks, open circles (0). The elution volumes are not strictly comparable since three different columns were used. The flotation velocity patterns of the chow-fed and I week cholesterol-fed HDL are given in Fig. I.
CHOLESTEROL-FED
MONKEY
HDL
223
and HDL, agreed very well with that determined by the analytical ultracentrifugal method. The improved separation of HDLz and HDL3 from cholesterol-fat-fed monkeys by molecular exclusion column chromatography suggested that on the cholesterol-fat diet, HDLz and HDL, increased their size difference. Flotation rates. The relation among percentage HDL2 and HDL3, total HDL cholesterol concentration, and the Sf of the major HDL components during the chow and cholesterol-fat diet periods are shown in Table 4. The HDS Sf rates were considerably smaller (2.8 vs 4.2) after 5 weeks on the cholesterol-fat diet when compared to the chow diet while the HDLz Sf rates were almost identical (7.0 vs 7.2) during the l-3 week cholesterol-fat time period. (Five week HDL, concentrations were too low to measure S,.) This increased difference between the Sf of HDL> agrees with the previously discussed increased separation on molecular sizing gel filtration. Typical schlieren photographs of HDL from monkeys on the indicated diets taken at identical times are shown in Fig. 1. It can be observed that the HDL3 (the peak on the right in all photographs) floated slower when the cholesterol-fat diet was fed. In Fig. I, top, after 5 weeks on the cholesterol-fat diet, HDL3 remained very near the cell bottom (right edge) while in Fig. 1, bottom, on the chow diet, HDL, had moved farther away from the right-hand edge. DISCUSSION The present studies are significant in that they add another subhuman primate to the list of those that are highly responsive to cholesterol feeding by increasing both plasma cholesterol and LDL. A small but significant increase in total cholesterol and HDL cholesterol was observed when the animals ingested a high fat diet as compared to pelleted monkey chow. In addition, for the first time, studies specifically designed to study such changes in HDL subfractions during cholesterol feeding are reported. Addition of cholesterol to the high fat diet resulted in a prompt and sustained increase in plasma cholesterol and substantial alteration in its distribution among lipoproteins. Prior to fat or cholesterol feeding only 36% of plasma cholesterol was transported as HDL cholesterol. This percentage of HDL cholesterol was low for an Old World monkey (12). but not as low as those of some New World monkeys: spider 9% and woolly 8% (of the total cholesterol as HDL) being the lowest (12,13). The major change in the plasma lipoproteins was a marked increase in LDL during cholesterol-fat feeding. After 5 weeks on the cholesterolfat diet, LDL cholesterol accounted for greater than 85% of the total cholesterol. Five weeks after return to a chow diet, the total and HDL cholesterol were at the prediet concentration. In several primate species, when total serum cholesterol is increased
724
NELSON.
MORRIS.
AND
GREEK
to a concentration of 200-450 mg/dl. it has generally been found that an increase in serum HDL also occurs (1 l-1.5). A further elevation in total serum cholesterol values is associated with a serum HDL concentration which is lower than controls (16-19). In rhesus monkeys, a small increase in HDL during cholesterol feeding resulted in a slightly increased ratio of 1.125 g/ml floating HDL to sedimenting HDL (I I ). while the decrease in HDL (macaque monkeys) resulted in a decreased ratio of I. I25 g/ml floating HDL to sedimenting HDL C16). Unexpectedly. the changes in HDL in response to cholesterol-fat feeding were associated with several new observations. First, during the early period of induction of hypercholesterolemia, the increase in HDL cholesterol was found in HDL? only. Second, during the latter part of the cholesterol-fat diet period, as total plasma cholesterol increased further, both HDLz and HDL, were decreased to levels below that found on the chow diet. In several monkeys, HDL, was undetectable during the latter stages of cholesterol-fat feeding regimen. Finally, an additional and unique observation was that HDL isolated from cholesterol-fat-fed sooty mangabeys separated into two components upon agarose column chromatography gel filtration where components are separated according to molecular size. During the early period of cholesterol-fat feeding, when plasma cholesterol had increased about two times and HDL cholesterol was at its highest serum concentration, the HDL profile from 6% agarose gel chromatography revealed a skew to the right indicative of increased heterogeneity in the HDL fraction (Fig. 3, middle). At the end of the cholesterol-fat feeding period (5 weeks) the agarose column pattern clearly revealed two peaks in the purified HDL of many monkeys. Thus. HDL? and HDL, increased in size difference on cholesterol-fat diets. During cholesterol-fat feeding, it was the HDL3 that became progressively altered structurally so that by the end of 5 weeks on this diet, the Sf was reduced by 33%. from S, 4.2 to 2.8. These data indicate that the HDL? was smaller and/or denser as a result of cholesterol-fat feeding. Such alterations in the metabolism of HDL subfractions, the specifically altered concentration of circulating HDLl during the early period of cholesterol-fat feeding, and the apparent selective change in the physical properties of HDL, after cholesterol ingestion, have not been reported previously. We have previously reported that the HDL of chow-fed rhesus monkeys is eluted from 6% agarose columns as a single symmetrical peak Cl). Cholesterol-fed rhesus and grivet monkeys have likewise not been reported to show HDL heterogeneity when characterized by gel filtration chromatography (20,211. In both species, the HDL may have been primarily HDL, judging from the decreased serum HDL concentration. Other in-
CHOLESTEROL-FED
MONKEY
HDL
225
vestigators have not looked for heterogeneity, but have arbitrarily divided subhuman primate HDL into two density fractions and found differences in chemical composition (10,ll). Our attempts to study the metabolismof HDL in rhesus monkeys have been complicated by the failure to separate HDL, and HDL, by gel filtration chromatography and the lack of obvious heterogeneity of the chow-fed rhesus monkey HDL. It is apparent that the existing analytical separation techniques are still too limited to permit a precise definition of all HDL subfractions. However, based on our present data in the HDL of the sooty mangabey, this animal appears to be the preferred one to study changes in HDL metabolism during cholesterol feeding. When the mean flotation rates of chow-fed sooty mangabey HDLz and HDL, are compared to their corresponding human fractions, the rates are somewhat faster. The mangabey Sr for HDLz was 7.1 + 0.3 compared to human 6.4 + 0.4 (N = 8) (our unpublished data), whereas HDL, was Sf 4.2 ? 0.4 in the mangabeys and 3.35 + 0.3 (N = 33) in the humans. After the mangabeys had ingested the cholesterol-fat diet for 5 weeks, the HDL, Sf was quite similar to that observed in human HDL+ We are unaware of reports in humans comparing the properties of HDL subfractions during diet-induced alterations of cholesterol and/or fat intake. In addition to the observations on HDL, the sooty mangabey plasma lipoproteins were characterized by a higher concentration of the pre+ non-VLDL lipoprotein, Lp(a), in every mangabey when compared to other monkey species. During cholesterol feeding this peak becomes obscured when lipoproteins were isolated column chromatographically so we were unable to judge changes in this fraction. SUMMARY
Sooty mangabey (Cercocebus atys) monkeys had a lower serum HDL cholesterol concentration than any other Old World monkey species reported. In addition, they had a higher serum Lp(a) concentration than other species. The mangabeys were fed a cholesterol-fat diet for 5 weeks. HDLz and HDL3 amounts were determined from the two peaks apparent upon analytical ultracentrifugation. In the first l-3 weeks, 13 of the 14 mangabeys increased 30% (mean) in total HDL, this increase occurring only in the HDL, fraction. After 5 weeks, HDL and HDLz decreased markedly. During the cholesterol feeding, HDL3 continually decreased in flotation rate, indicating it was either smaller and/or denser. HDL2 and HDL3 separated well on molecular sieving agarose columns during the diet period, whereas a single symmetrical elution peak was found for chow-fed HDL. Thus on a cholesterol-fat diet, HDLz and HDLl increased in difference in molecular size.
226
NELSON,
MORRIS. AND CREER
ACKNOWLEDGMENTS We would like to acknowledge the expert technicaf assistance of Rosa Haidar. We also appreciate and acknowledge Dr. Irwin S. Bernstein and Dr. Brent Swenson of the Yerkes Regional Primate Research Center. Emory University, Atlanta, Geargia for obtaining the serum sampies from the free ranging mangabeys. They were supported by NIH Core Research Grant RR00165 to the Yerkes Regional Primate Center and NIH Grant 13864.
REFERENCES I. 2. 3. 4. 5. 6. 7. 8.
Nelson, C. A., and Morris, M. D., &&em. &fed. $7, 3tQ (t9771. Hamerton, A. E.. Pruc. Zuol. Sot. Lnndor? 111, IS1 f 1941). Nelson, C. A., and Morris, M. IX. Biuckem. B&&s. Res. Cnmmun. 71, 438 (1976). Rude], L. L., Lee, 1. A., Morris, M. D., and Felts, J. M., ~~~~h~rn. f. 139, 89 (1976). Noble, R. P., J. Lipid Res. 9, 693 (1468). Scheidegger, J. J., Int. Arch. Allergy Appl. Immunol. 7, 103 (1955). Johnston, J. P., and Ogston, A. G.. Tucms. Faroduy SW. 42, 789 fi946). Anderson, D. W. ~Nichols, A. V ., Forte. T. M.. and Lindgren, F. T., ~~u~k~rn. Biopkys. Acta 493, 55 (1977). 9, Ha&wood, R. N., I. Amer. Ckmm. Sue. 80, 2152 (19571. IO, Scanu, A. M., Edelstein, C., Vitello. L., Jones, R.. and Wissler, R,, J. Biol. Ckem. 248, 7648 (1973). Il. Burstein, M., and Samille, J., C/in. Ckim. Acta 5, 609 f1960). 12. Rudei, L. L., and Lofland, W. B.. Jr.. in “Primates in Medicine,” (J. P. Strong. Ed.), 9th ed., p. 224. Karger, Base!, 1976. 13. Lofland, H. B., St. Cfair, R. W.. MacNintch, J. E., and Prichard, R. W,, Arch. ~~t~~~~, 83, 211 (1967). 14. Portman, 0. W., Alexander, M.. Neuringer, M., fllingworth, D. R., and Alam, S. S,, f. Nutr. 11X, 733 (1981). IS. Kushwaha, R. S,, Hazzard, W. R., Harker, L. A., and Engblom, J., ~~~?~~~s~~e~~.~j~~ 31, 65 f197fQ. 16. &show, A. C., Nicolosi, R. J., and Hayes, K. C.. Amer. J. Cfin. Nutr. 34,830 (19811. 17. Srinivasan, S. R.. Rad~akrisbnamu~by, 3.. Dalferes, E. R., Jr., and Berenson. G. S., Lipids
14, 559 (1X79),
18. Srinivasan, S. R., Radhakrishnamurthy, B., Smith, C. C.. Wolf, R. H., and Berenson. Cl. S.. J. Nutr. 106, 1757 (1976). 19. Nelson, C. A.. Casaii. R. E., and Morris, M. D., Biackem. Med. 22, 110 f 1979). 20. Rudel, L. L., Pit&, L. L., II, and Nelson, C. A.. 1. Lipid Res. 18, 21t (19771. 21. Rude!, L. L., Shah. R., and Greene, D. C., J. Lipid Res. 20, 55 (197%.