Journal of Clinical Investigation Vol. 43, No. 11, 1964 The In Vivo Turnover of Individual Cholesterol Esters in Human Plasma Lipoproteins* DEWITT S. GOODMAN t (From the Department of Medicine, College of Physicians & Sutrgeonis, Columbia University, New York, N. Y.) It has long been known that plasma cholesterol consists largely of a mixture of several different fatty acid esters of cholesterol. This was first in- dicated in 1896, when Hiirthle isolated cholesteryl palmitate, oleate, and stearate from an alcoholic extract of serum (1). Subsequent early studies on the ratio of free to ester cholesterol in plasma demonstrated that most of the plasma cholesterol normally exists in ester form (2). More recent studies, using gas-liquid chromatography for the analysis of the cholesterol ester fatty acids, have defined the composition of the plasma cholesterol esters in a variety of species (3-8). In addition, plasma cholesterol normally is distributed among several different plasma lipo- protein fractions. Most of the plasma cholesterol of man is typically present in the low density, beta-lipoprotein fraction, with smaller amounts present in the high density and the very low den- sity lipoproteins (9-12). These lipoproteins show small but consistent differences in their ratio of free to ester cholesterol (11, 13). Recent analyses of the cholesterol ester composition of each human plasma lipoprotein have demonstrated strong simi- larities in the cholesterol ester composition of the different lipoproteins (13). Small but statistically significant differences do, however, exist between the different lipoproteins. Larger differences be- tween lipoprotein cholesterol ester patterns have been reported by Lindgren and Nichols (14). Previous studies on the turnover of cholesterol in human plasma have treated plasma cholesterol ester as a single homogeneous entity (15-17). Since plasma cholesterol ester exists as a mixture * Submitted for publication April 2, 1964; accepted July 2, 1964. Supported by research grants AM-05986 and HE-05741 from the National Institutes of Health. t Recipient of Career Scientist Award of the Health Research Council of the City of New York under con- tract I-399. of different esters distributed among several dif- ferent lipoproteins, however, multiple possibilities exist for the heterogeneous metabolism of plasma cholesterol esters. Thus, the turnover rates of different cholesterol esters, or of the same or dif- ferent esters within different lipoproteins, may not be the same. Recent studies on the turnover of individual cholesterol esters in rat plasma have, in fact, demonstrated substantial differences in the turnover rates of the different esters (18, 19). The present study was undertaken to examine the turnover of each different cholesterol ester within each of the three major plasma lipoprotein fractions in man. A preliminary report of some of these results has been published (19). Methods Forty-four microcuries of DL-2-C'4-mevalonic acid, in solution in isotonic saline, was injected intravenously into each of two fasting normal men. The C'4-mevalonic acid 1 was prepared for injection as described earlier (18). The subjects, a 39-year-old Negro (subject WB) and an 18-year-old Caucasian (subject EH), ate their usual diets except for a slight reduction in fat intake for one week preceding, and during, the period of the study. Serial blood samples were collected in syringes moistened with a solution of heparin, and plasma samples were col- lected and processed as described in detail elsewhere (13). In brief, small samples of plasma were extracted immediately, and larger volumes of plasma were used for the serial separation of lipoproteins as discussed by Havel, Eder, and Bragdon (10). Three lipoprotein frac- tions were collected from each large sample by ultracen- trifugation at densities (before centrifugation) of 1.019, 1.063, and 1.21. The lipoprotein and whole plasma sam- ples were extracted with CHCl3: MeOH, 2: 1 (vol/vol), and the total lipid so obtained was separated into hydro- carbon, cholesterol ester, free cholesterol + triglyceride, and phospholipid fractions by silicic acid column chroma- tography as previously described (13). The separations were checked repeatedly to ensure a quantitative re- covery of cholesterol esters uncontaminated by triglycer- ides in the second column fraction. Portions of each 1 New England Nuclear Corp., Boston, Mass. 2026
Transcript
Journal of Clinical InvestigationVol. 43, No. 11, 1964
The In Vivo Turnover of Individual Cholesterol Esters inHuman Plasma Lipoproteins*
DEWITT S. GOODMANt(From the Department of Medicine, College of Physicians & Sutrgeonis, Columbia University,
New York, N. Y.)
It has long been known that plasma cholesterolconsists largely of a mixture of several differentfatty acid esters of cholesterol. This was first in-dicated in 1896, when Hiirthle isolated cholesterylpalmitate, oleate, and stearate from an alcoholicextract of serum (1). Subsequent early studieson the ratio of free to ester cholesterol in plasmademonstrated that most of the plasma cholesterolnormally exists in ester form (2). More recentstudies, using gas-liquid chromatography for theanalysis of the cholesterol ester fatty acids, havedefined the composition of the plasma cholesterolesters in a variety of species (3-8).
In addition, plasma cholesterol normally isdistributed among several different plasma lipo-protein fractions. Most of the plasma cholesterolof man is typically present in the low density,beta-lipoprotein fraction, with smaller amountspresent in the high density and the very low den-sity lipoproteins (9-12). These lipoproteins showsmall but consistent differences in their ratio offree to ester cholesterol (11, 13). Recent analysesof the cholesterol ester composition of each humanplasma lipoprotein have demonstrated strong simi-larities in the cholesterol ester composition of thedifferent lipoproteins (13). Small but statisticallysignificant differences do, however, exist betweenthe different lipoproteins. Larger differences be-tween lipoprotein cholesterol ester patterns havebeen reported by Lindgren and Nichols (14).
Previous studies on the turnover of cholesterolin human plasma have treated plasma cholesterolester as a single homogeneous entity (15-17).Since plasma cholesterol ester exists as a mixture
* Submitted for publication April 2, 1964; acceptedJuly 2, 1964.
Supported by research grants AM-05986 and HE-05741from the National Institutes of Health.
t Recipient of Career Scientist Award of the HealthResearch Council of the City of New York under con-tract I-399.
of different esters distributed among several dif-ferent lipoproteins, however, multiple possibilitiesexist for the heterogeneous metabolism of plasmacholesterol esters. Thus, the turnover rates ofdifferent cholesterol esters, or of the same or dif-ferent esters within different lipoproteins, may notbe the same. Recent studies on the turnover ofindividual cholesterol esters in rat plasma have, infact, demonstrated substantial differences in theturnover rates of the different esters (18, 19).
The present study was undertaken to examinethe turnover of each different cholesterol esterwithin each of the three major plasma lipoproteinfractions in man. A preliminary report of someof these results has been published (19).
Methods
Forty-four microcuries of DL-2-C'4-mevalonic acid, insolution in isotonic saline, was injected intravenously intoeach of two fasting normal men. The C'4-mevalonicacid 1 was prepared for injection as described earlier(18). The subjects, a 39-year-old Negro (subject WB)and an 18-year-old Caucasian (subject EH), ate theirusual diets except for a slight reduction in fat intake forone week preceding, and during, the period of the study.Serial blood samples were collected in syringes moistenedwith a solution of heparin, and plasma samples were col-lected and processed as described in detail elsewhere(13). In brief, small samples of plasma were extractedimmediately, and larger volumes of plasma were usedfor the serial separation of lipoproteins as discussed byHavel, Eder, and Bragdon (10). Three lipoprotein frac-tions were collected from each large sample by ultracen-trifugation at densities (before centrifugation) of 1.019,1.063, and 1.21. The lipoprotein and whole plasma sam-ples were extracted with CHCl3: MeOH, 2: 1 (vol/vol),and the total lipid so obtained was separated into hydro-carbon, cholesterol ester, free cholesterol + triglyceride,and phospholipid fractions by silicic acid column chroma-tography as previously described (13). The separationswere checked repeatedly to ensure a quantitative re-
covery of cholesterol esters uncontaminated by triglycer-ides in the second column fraction. Portions of each
cholesterol ester and free cholesterol fraction were ana-lyzed for cholesterol mass by the method of Sperry andWebb (20). Duplicate analyses were carried out oneach sample; the results of these analyses differed byless than 5% in every case. Other portions of the cho-lesterol ester and free cholesterol fractions were as-sayed for C"4 with a Packard liquid scintillation spectrom-eter, using 0.5% diphenyloxazole in toluene as solvent.The counting efficiency in this system was 55%. Thespecific radioactivity of the free cholesterol, and theaverage specific radioactivity of the esterified cholesterolin each whole cholesterol ester fraction, were then cal-culated from these data.
The composition of each cholesterol ester fraction wasdetermined by gas-liquid chromatography (GLC) of thecholesterol ester fatty acids as their methyl esters, asdetailed earlier (13). The GLC distributions were cor-rected for the different molecular weights of the differentfatty acid methyl esters, and the distribution of choles-terol esters was tabulated in molar terms for each sample.This is equivalent to a tabulation in terms of the distri-bution of cholesterol mass among the different esters.The complete GLC data on the fatty acid composition ofeach lipid class in whole plasma and in each plasma lipo-protein for the samples that comprised the present studyhave been presented elsewhere (13). Both subj ectsshowed only extremely small variation from sample tosample in the concentration and composition of each lipidclass in each lipoprotein.
The distribution of radioactivity among the differentcholesterol esters in each sample was determined by thinlayer chromatography (TLC) on silica gel G impregnatedwith AgNO3, using benzene-hexane, 1: 1 (vol/vol), asascending solvent, and as previously described in detail(18). This method separates cholesterol esters differingin the number of double bonds in the fatty acid moiety.TLC resolved each sample into four widely separatedzones, comprising the saturated cholesterol esters (mainlycholesteryl palmitate, Rf -~- 0.8), the monounsaturatedesters (mainly cholesteryl oleate, Rf r 0.65), the diun-saturated esters (cholesteryl linoleate, Rf , 0.4), andthe tetraunsaturated esters (cholesteryl arachidonate,Rf _t 0.1). Each zone was separately scraped on to afilter funnel, eluted with 15 ml CHC13, and assayed forC14. Further elution with hot benzene: ethyl ether, 1: 1,was carried out to recover completely cholesteryl arachi-donate from the A4 zone; the other more saturated es-ters were satisfactorily eluted with CHC1s alone.
The validity of this TLC method for the determina-tion of the distribution of cholesterol ester radioactivityhas been verified by the analysis of various standard mix-tures of pure labeled cholesterol esters (18). DuplicateTLC analyses were carried out on each cholesterol estersample in the present study. The values for the per-centage of distribution of radioactivity in the saturatedand mono- and diunsaturated zones agreed within 10%(usually within 5%) in almost every pair of analyses.In the very few instances where differences greater than10% of the value were observed, one or more additional
analyses were carried out. The values for the percent-age of total radioactivity in the A4 zones agreed within20% of each other on duplicate analysis. This greaterrelative error in the A4 values largely derived from therelatively small amounts of cholesteryl arachidonate pres-ent in the samples. Only the average values for the du-plicate analyses have been tabulated in the results thatfollow.
To compare the distribution of cholesterol ester mass,as determined by GLC, with the C'4 distribution as de-termined by TLC, the saturated and monounsaturatedester mass values were summed for each sample, andthe distribution of esterified cholesterol mass was ex-pressed in terms of saturated, Al, A2, and A4 fractions.Almost identical values were obtained, from sample tosample, for the mass distributions for a given plasmalipoprotein (or for whole plasma). This is indicated bythe extremely small standard errors tabulated with theGLC data in earlier work (13). Since the small varia-tions from sample to sample were probably more tech-nical than biological in origin, the values for the varioussamples were averaged, and only the average values wereused as representing the distribution of esterified choles-terol mass in all samples of a given lipoprotein. Thisprocedure thus minimized any small variation from sampleto sample that might have been derived from small tech-nical errors in the GLC analyses.
The specific radioactivity of each cholesterol esterfraction in each sample was calculated from the pre-ceding data by multiplying the average specific radio-activity of the whole ester fraction by the ratio of thepercentage of total C" to the percentage of total mass.
Results
The values of the specific radioactivity of freeand esterified cholesterol in the whole plasma sam-ples of each subject, at various time intervals af-ter the injection of DL-2-C14-mevalonate, are listedin Table I. Labeled free cholesterol rapidly ap-peared in the plasma, with the maximal specificradioactivity of free cholesterol occurring at ap-proximately 3 hours in both subjects. The timeand levels of the peak specific radioactivity ofplasma free cholesterol seen here are comparableto those observed by Gidez and Eder in a patientgiven intravenous C14-mevalonate (17). Thepeak specific radioactivity for plasma ester cho-lesterol occurred at approximately 48 hours inboth subjects. Equilibration between the free andester cholesterol occurred more rapidly in subjectEH than in subject WB. Thus, after 24 hoursthis equilibration was 76% complete (ester/freeSA ratio, 0.76) in subject EH but only 50% com-plete in WB. After 48 hours equilibration was
2027
DEWITT S. GOODMAN
TABLE I
Cholesterol specific radioactivity in whole plasma after intravenous 2-C14-mevalonate
The urine of subject EH was collected andpooled during the first 48 hours after the mevalo-nate injection. One ml of urine was added to 10ml of the scintillation mixture described by Bray(21) and was assayed for C14; the observedcounts per minute were corrected to the usualefficiency by use of an appropriate internal stand-ard. The pooled urine contained 55% of the totalamount of C14 injected into subject EH. A simi-lar urinary excretion of radioactivity, mainly oc-curring in the first 12 hours, was also seen byGidez and Eder in the patient mentioned above(17). Most of this urinary C14 probably consisted
of L(-)miievalonic acid., since it has been shownthat only the D(+) isomer is biologically active(22) and that the L(-) isomer is primarily ex-creted in the urine (23). The urinary excretionin excess of 50% probably reflects the excretion ofsome of the active isomer during the first hour ortwo, together with the excretion of some labeledbicarbonate, derived from the C1402 released dur-ing the course of cholesterol biosynthesis (24).
The specific radioactivity of the free and esteri-fied cholesterol in each plasma lipoprotein, at vari-ous time intervals after C14-mevalonate, is listedin Table II. Lipoprotein fractionation was carriedout on all the samples from subject EH but ononly four of the plasma samples of subject W\B.
TABLE II
Cholesterol specific radioactivity in plasma lipoproteins after intravenous 2-CI4-mevalonate
FIG. 1. THE SPECIFIC RADIOACTIVITY OF FREE AND ESTERIFIED CHOLESTEROL
IN THE THREE PLASMA LIPOPROTEIN FRACTIONS OF SUBJECT EH AT VARIOUS
TIMES AFTER INTRAVENOUS2-C14-MEVALONATE. VLDL = very low density,LDL = low density, and HDL= high density lipoproteins.
The free cholesterol specific radioactivities in thethree lipoproteins derived from a given plasmasample were nearly identical in all samples. Incontrast, marked differences were seen in the spe-
cific radioactivities of the esterified cholesterol ofthe different lipoproteins. In both subjects, ra-
dioactivity appeared much more rapidly in theester cholesterol of the high density (density1.063 to 1.21) lipoprotein than in the ester cho-lesterol of the other lipoproteins. Radioactivityappeared least rapidly in the low density (density1.019 to 1.063) lipoprotein ester cholesterol.Most of the data for subject EH are also showngraphically in a log-log plot in Figure 1. Thedifferences between the three lipoproteins were
most marked in the early hours of the study andconsisted of specific radioactivity differences ofseveral hundred per cent. After 2 days the dif-ferences between the lipoproteins were virtuallygone. In the early periods of these studies, thedistribution of radioactivity in ester cholesterolin the three lipoproteins differed markedly fromthe distribution of mass. Both subjects had ap-
proximately 25% of their plasma ester cholesterol
mass in the high density lipoproteins and approxi-mately 60%o of the ester cholesterol in the density1.019 to 1.063 lipoproteins. In both subjects,however, during the first 5 hours more than 50%oof the ester cholesterol C14 was found in the highdensity lipoprotein. WVith increasing time the dis-tribution of ester cholesterol C14 approached themass distribution, as equilibration proceeded among
the various cholesterol pools.The data obtained with the isolated lipoprotein
fractions were used to calculate the expected spe-
cific radioactivity of the esterified cholesterol inthe corresponding samples of whole plasma. Thecalculated values for whole plasma agreed withthe observed values (Table I) within 6% in every
sample. Since the ester cholesterol specific activi-ties in the different lipoproteins differed by severalhundred per cent, the close agreement between thecalculated and observed values for whole plasmaestablished that no changes or selective losses oc-
curred during the 3 days involved in the lipo-protein fractionations. The specific radioactivitydifferences observed between the ester cholesterolof the different lipoproteins therefore must have
1500
E 1000a,E 800
% 600
i.- 400
C,I-
04
0 2000.(j) 100
-J 800w 60w(f)w 40-J0
)I
F-
201-
10
2029
DEWITT S. GOODMAN
TABLE III
Distribution of cholesterol mass among four cholesterol esterfractions in plasma and plasma lipoproteins
Per cent of total esterifiedcholesterol*
Saturated Al A2 A4
% % % %Subject WB
Whole plasma 15.1 26.4 50.9 7.6Lipoprotein of density:
* These are the average values for all the samples of whole plasma or ofa given lipoprotein, as determined by gas-liquid chromatography of thecholesterol ester fatty acid methyl esters [see text and (13)].
existed at the time the plasma samples were col-lected, and hence reflected true in vivo differences.In vitro esterification of free cholesterol (25)probably did not occur during lipoprotein separa-tions because of the low temperatures employedfor ultracentrifugation.
The distribution of esterified cholesterol massamong the saturated and mono-, di-, and tetraun-saturated cholesterol esters, for whole plasma andfor each plasma lipoprotein in each subject, isshown in Table III. As described under Methods,these data are derived from the complete GLCdata given earlier (13). Cholesteryl palmitate andoleate comprised the predominant saturated andmonounsaturated esters, respectively. Cholesteryllinoleate and arachidonate were, respectively, theonly di- and tetraunsaturated esters.
The data listed in Table III demonstrate foreach subject the strong similarities seen in thecholesterol ester compositions of the different lipo-proteins. As discussed previously (13), smallbut statistically significant differences did, how-ever, exist between the different lipoproteins. Forboth subjects these differences included the oc-currence of significantly more oleate in the verylow density (density < 1.019) lipoprotein cho-lesterol esters than in the cholesterol esters of theother lipoproteins. There was also significantlyless arachidonate in the very low density lipopro-tein cholesterol esters of subject WBcompared tohis other lipoproteins. None of these differenceswas sizeable.
Table IV lists the distribution of radioactivity
among the four cholesterol ester fractions, as de-termined by TLC, for each sample of whole plasmaor plasma lipoprotein. Table IV also lists thespecific radioactivity ratio for each ester fractionrelative to the free cholesterol specific radioactivityin the same sample.
Some of the relative specific radioactivity datafrom Tables II and IV are also shown graphicallyin Figure 2 for the lipoprotein samples of subjectEH. In Figure 2, each vertical column shows theresults obtained with one lipoprotein sample. Ineach column the height of the horizontally stripedbar represents the ratio of the average specific ra-dioactivity of the whole cholesterol ester fractionto that of the free cholesterol in the same lipopro-tein sample (from Table II). The relative spe-cific radioactivities of the four individual choles-
1.0 r
0.8
O.6h
0.4
0.2
0
1.0 r0
4)
01
0
01
0.8 F
0.6 L
0.4 P
0.2 [
P= WHOLE ESTERFRACTION
* = A l 0AzA2 VV * A4
-* SATURATED
LIPOPROrEI/N
o1 iii
IAz
A4
LIPOPROTiE/NtO9,-1(/.063 V
0%f-VTiEV ION 'FE I
1.0 *0
a. 0.8 ALIPOPROTEI/N
0.6 I.063</,o</2/ I1
2.92 4.83 7.75 10.8 15.8 24.8 48.8
TIME in HOURS
FIG. 2. THE RELATIVE SPECIFIC RADIOACTIVITY OF CHO-
LESTEROLESTERS IN THE THREEPLASMALIPOPROTEIN FRAC-
TIONS OF SUBJECT EH AFTER INTRAVENOUS2-C'4-MEVALO-NATE.
The distribution of cholesterol radioactivity among, and the relative specific radioactivity of, the fourcholesterol ester fractions in plasma and plasma lipoproteins
%Distribution of cholesterol-C'4 Relative SA (free cholesterol = 1.00)
terol ester fractions in each sample (from TableIV) are plotted as points in the same vertical col-umn. The potential error in the relative specificradioactivity values for the individual cholesterolesters (Table IV) is probably of the order of+ 10% for the saturated and mono- and diunsatu-
rated esters. The uncertainty in the A4 relativespecific activity values is probably + 20%. Thelesser reliability of the A4 values in Table IV de-rives from the small amount of radioactivity foundin the A4 zones; a small amount of contaminationof radioactivity from the preceding zones wouldintroduce a relatively large error in the zA4 value.
In both subjects, for each lipoprotein consideredseparately and for whole plasma, radioactivity ap-peared in the different cholesterol esters at a rateproportional to the relative mass of each cholesterolester. Thus, in each lipoprotein and in wholeplasma, the relative rate of appearance of radio-activity was the same for all individual cholesterolesters. This is most clearly indicated by the rela-tive specific radioactivity data of Table IV andFigure 2. At all time intervals the relative spe-cific radioactivities of the different cholesterolesters in each sample were virtually identical.The only exception occurred with the very lowdensity (density < 1.019) lipoprotein of subjectWB, which showed relatively more radioactivityin the monounsaturated esters in the earliest twosamples analyzed. This difference was not largeand was very transient; it was gone by 7.25 hours,although the extent of equilibration of free andester cholesterol in this lipoprotein was less than20% at that time. In contrast, large differenceswere seen previously in the relative rates of la-beling of the different cholesterol esters of ratwhole plasma, and these differences extendedthroughout the period of equilibration of free andester cholesterol (18, 19). In all the other sam-ples of both subjects WBand EH, the small dif-ferences seen in the relative specific radioactivityvalues for the different cholesterol esters werewithin the limits of error as discussed above.
Discussion
The studies reported here demonstrate markeddifferences in the relative rates of appearance ofradioactivity, after intravenous C14-mevalonate,inl the cholesterol esters of the various human
plasma lipoproteins. These differences probablyreflect comparable differences in the turnoverrates of the cholesterol esters of the different lipo-proteins, since a steady state probably existed inthese postabsorptive subjects with unchangingplasma lipid concentrations. Both subjects showedthe greatest fractional turnover rate in the highdensity lipoprotein cholesterol esters and the small-est fractional turnover rate in the low density(density 1.019 to 1.063) lipoprotein cholesterolesters. The fractional turnover of the very lowdensity (density < 1.019) lipoprotein cholesterolesters, which comprised only a very small por-tion of the total plasma cholesterol ester mass, wasintermediate in extent.
Similar data on the in vivo turnover of estercholesterol in different plasma lipoproteins arenot available for experimental animals or fromother human studies. Kritchevsky, Shapiro, andWerthessen studied the turnover of the serum freeand ester cholesterol, and of the total cholesterolof the a- and ,8-lipoproteins, in a baboon givenC'4-mevalonate by stomach tube (26). Theyfound that the specific radioactivity-time curveswere practically identical for the two lipoproteins.Unfortunately, the blood samples were not largeenough to permit separate determinations forfree and ester cholesterol in the two lipoproteinfractions. Gidez and Eder stated in an abstract(27) that in a patient given C14-mevalonate thespecific activity of the cholesterol esters of thelipoprotein fraction with density 1.019 to 1.063was lower than that of the cholesterol esters ofthe other lipoprotein fractions; the details of thiswork have not been published. Information isavailable about the relative turnover of choles-terol esters in different rat plasma lipoproteinsin vitro from the rat liver perfusion studies ofRoheim and his associates (28). During perfu-sion of livers containing labeled cholesterol fromeither normal or cholesterol-fed rats, the fastestrelative turnover of ester cholesterol occurred inthe very low density (< 1.019) lipoprotein frac-tion. This observation was consonant with theresults of previous rat liver perfusion studies bythese workers, which showed the fastest relativeturnover of lipoprotein protein in the very lowdensity fraction (29).
In the present work, the specific radioactivityof the free cholesterol in the different lipoproteins
was nearly identical in all samples of plasma.This finding was not surprising, since free cho-lesterol has been shown to undergo rapid and com-plete in vitro exchange between different lipo-proteins (28, 30, 31). This exchange also oc-curs between the free cholesterol of plasma andred cells, and plasma and liver, and presumablyaccounts for the observed rapid appearance of la-beled plasma free cholesterol subsequent to thebiosynthesis of labeled cholesterol from C14-meva-lonate in the liver. A slower exchange has alsobeen demonstrated with plasma phosphatides (32,33) and with triglycerides (34). The maximalextent of equilibration that could result from invitro exchange has not been established for thesetwo lipid classes. Although the mechanisms in-volved in the exchange process have not been de-fined, the fact of exchange tends to obscure thedemonstration of metabolic differences that mightexist between different lipoproteins for these lipidclasses.
In contrast to free cholesterol, virtually no ex-change of ester cholesterol occurs in vitro (28, 31).The absence of ester cholesterol exchange was re-flected, in the present study, in the very differentester cholesterol specific radioactivities seen in thedifferent lipoproteins. The finding of these dif-ferences established the absence of exchange invivo as well as in vitro and permits the interpre-tation of specific radioactivity changes in eachlipoprotein in terms of relative turnover rates.It has been suggested (28) that since ester cho-lesterol does not exchange among lipoproteins,study of ester cholesterol turnover in the variouslipoproteins may provide information about themetabolism of the entire lipid portion of the vari-ous lipoproteins. Although this may be true, thepresent finding of the highest fractional turnoverrate in cholesterol esters of high density lipopro-teins should not be applied to other portions of thelipoprotein molecules without further information.Potential discrepancies are suggested by the workof Gitlin and co-workers (35), who observed aslower fractional turnover of I131-labeled proteinin high density compared to low density humanlipoproteins. In addition, Havel (36) has re-ported in human subjects the most rapid relativeincorporation of circulating free fatty acids intovery low density (< 1.006) lipoprotein triglycer-ides. Further work is clearly needed on the me-
tabolism of the various moieties comprising eachhuman plasma lipoprotein.
The results presented here also show that alldifferent cholesterol esters were turning over atthe same fractional rate within each lipoprotein.Since the turnover rate of the whole cholesterolester fraction of each lipoprotein differed from thatof the other lipoproteins, the fractional turnoverrate of each ester within each lipoprotein musthave correspondingly differed from the fractionalturnover rate of the same ester in the other lipo-proteins. For each lipoprotein, however, the entirecholesterol ester mixture behaved as a single ho-mogeneous pool. Since the cholesterol ester com-positions of the various lipoproteins were verysimilar, homogeneous turnover of all cholesterolesters in each lipoprotein was reflected in the find-ing that all the different cholesterol esters of wholeplasma were turning over with the same fractionalturnover rate.
Very different results were obtained in a simi-lar study on the turnover of individual cholesterolesters in rat whole plasma (18, 19). A much morerapid fractional turnover rate was seen in themonounsaturated compared to the other cholesterolesters in rat plasma. This finding suggests thatmajor variations exist in the metabolism of plasmacholesterol esters in man and the rat. More in-formation is needed on the turnover of differentcholesterol esters in each rat plasma lipoprotein.
Portman and Sugano have reported the resultsof experiments dealing with the metabolism ofindividual cholesterol esters in monkey wholeplasma (37). Six female Cebus monkeys wereinjected intraperitoneally with C14-mevalonate,and the distribution of cholesterol ester radioac-tivity among four cholesterol ester groups wasdetermined after 2 hours, 1 day, 3 days, and atintervals thereafter. The distribution of radio-activity in the different cholesterol esters agreedfairly well with the composition of the total cho-lesterol ester fatty acids as determined by GLC.Of interest is the fact that the plasma cholesterolester composition observed in these monkeys wassimilar to that seen in man. These results sug-gest that the different plasma cholesterol estersof Cebus monkeys are all turning over at the samerelative rate, similar to the results obtained in thepresent study with human subjects. This con-clusion should only be considered as tentative,
2033
DEWITT S. GOODMAN
however, because the analyses were carried outon only one early sample (at 2 hours) during theperiod when label was first accumulating in theester cholesterol fractions. Since the free andester cholesterol pools of plasma eventually comeinto isotopic equilibrium, meaningful metabolicconclusions can only be derived from early plasmasamples, before such equilibrium has beenachieved.
Uncertainty exists as to the mechanisms re-sponsible for the plasma cholesterol ester turn-overs seen in the present study. The processesthat might be involved in this turnover have beendiscussed in detail elsewhere (18). One possi-bility is that turnover depends upon continuinghydrolysis of plasma cholesterol esters, followedby esterification of free cholesterol, within plasmalipoproteins. If this occurs, hydrolysis probablytakes place during lipoprotein circulation throughthe liver and involves liver enzymes, since cho-lesterol ester hydrolytic enzyme activity has notbeen found in human plasma (38). Such hydro-lytic activity has been found only in dog serum,and not in the serum of a number of other species(38). The characteristics of the rat liver en-zymes involved in cholesterol ester hydrolysishave been defined (39), but similar informationis not available for other species. Esterificationof plasma free cholesterol, on the other hand, mightoccur either in the liver or in situ in the plasma.Cholesterol esterification in rat liver involves thereaction of free cholesterol with a fatty acylcoen-zyme A thiol ester (40). In contrast, cholesterolesterification seen in plasma in vitro has beenshown by Glomset (25) to primarily involve atransferase reaction between free cholesterol andthe /3 fatty acids of plasma lecithin.
As discussed previously (18), a second pos-sible mechanism involved in the plasma cholesterolester turnover observed here might be the re-mnoval of intact cholesterol esters during their cir-culation through tissues (including the liver),followed by replacement of the removed estersduring circulation through the liver. Finally,there is the possibility that the observed turnoverreflected equilibration of the plasma cholesterolesters, by exchange, with a turning-over pool ofliver cholesterol esters. It has been pointed out(18) that for such equilibration to occur it wouldbe necessary to postulate the occurrence of some
conformational change in the lipoprotein duringits circulation through the liver.
More information is needed to assess the rela-tive roles of cholesterol esterification in liver andplasma in the normal formation and turnover ofplasma cholesterol esters. The potential im-portance of the plasma mechanism is suggested bythe fact that the initial rate of transesterificationobserved in vitro (25) is of an order of magnitudesimilar to the turnover rate of cholesterol estersseen in human whole plasma in vivo. On theother hand, the different turnover rates of cho-lesterol esters seen here in different plasma lipo-proteins conflict with the properties of the plasmatransesterification reaction. Thus, after in vitroincubation of human plasma, the increments inester cholesterol in each lipoprotein were propor-tional to the original concentration of free choles-terol in each lipoprotein fraction (41). In con-trast, in vivo turnover is relatively much morerapid in the high density lipoprotein fraction.More information is needed about the relativeparticipation of different lipoproteins in the inIvitro transesterification reaction at short time in-tervals. It has also been suggested (25) that theoperation of the plasma reaction might explainthe high proportion of polyunsaturated fatty acidsseen in plasma cholesterol esters. For this to be sofor man, this reaction would have to display speci-ficity for certain particular fatty acids in the,l-position of lecithin. This conclusion derivesfrom the present finding that the different plasmacholesterol esters were all turning over with thesame fractional rate, despite the fact that the com-position of the plasnia cholesterol esters differedconsiderably- fromt that of the fatty acids attachedto the A8-position of lecithin. Unpublished studiesfrom this laboratory have demonstrated a muchhigher ratio of arachidonic to linoleic acid in the/3-position of lecithin than in cholesterol esters in
human whole plasma and plasma lipoproteins (seealso the phospholipid composition data in refer-ence 13). Experiments with rat plasma (25)have suggested that the plasma transesterificationreaction does not display specificity for particularfatty acids in the /3-position of lecithin. Com-parable experiments have not been carried out
with human plasma.Although the results presented here do not de-
fine the mechanisms involved in plasma cholesterol
ester turnover, they do indicate that the mecha-nisms responsible for the turnover of cholesterolesters in each lipoprotein apparently operatedequally on all the cholesterol esters in a given lipo-protein. Thus, for each human plasma lipopro-tein, the entire cholesterol ester mixture behavedas if it were a homogeneous pool. In addition,homogeneous turnover of cholesterol ester mix-tures in each lipoprotein, with similar ester com-positions in the different lipoproteins, suggeststhat the same mechanisms were responsible forcholesterol ester turnover in all the lipoproteins,but that these mechanisms operated at differentrates for the different lipoproteins.
Summary
Studies have been conducted of the turnoverrates of individual cholesterol esters in wholeplasma and in each of three plasma lipoproteinfractions in man. Two normal fasting men wereinjected intravenously with 2-C14-mevalonic acid,and plasma samples were collected at time inter-vals varying from 1 hour to 7 days. Most of theplasma samples were separated into three lipo-protein fractions by serial ultracentrifugation atdensities of 1.019, 1.063, and 1.21. The choles-terol ester and free cholesterol fractions were iso-lated by silicic acid column chromatography ofthe total lipid extracts, and the specific radioac-tivity of the free and esterified cholesterol in eachsample was determined. The distribution of cho-lesterol mass and radioactivity among the severaldifferent esters were determined, respectively, bygas-liquid chromatography of the cholesterol esterfatty acid methyl esters and by thin-layer chro-matography on silver nitrate impregnated silicagel. The specific radioactivity of each differentcholesterol ester (saturated, and mono-, di, andtetraunsaturated esters) was then calculated fromthese data.
Marked differences were observed in the rela-tive rates of appearance of radioactivity in thecholesterol esters of the various plasma lipopro-teins. Both subjects showed the greatest frac-tional turnover rate in the high density lipopro-tein cholesterol esters and the smallest fractionalturnover rate in the low density (density 1.019to 1.063) lipoprotein cholesterol esters. Withineach lipoprotein. all the different cholesterol es-
ters turned over at the same fractional rate.Heterogeneity among plasma cholesterol esters inman therefore exists between the different plasmalipoproteins, rather than between the differentesters in a given lipoprotein. The results sug-gest that the mechanisms responsible for the turn-over of plasma cholesterol esters operate equallyon all the cholesterol esters in a given lipoprotein,and that the same mechanisms operate, albeit atdifferent rates, in the different lipoproteins.
Acknowledgments
I wish to express my thanks to Mr. T. Shiratori andMiss C. Goldman for their expert assistance in carryingout this work.
References
1. Hfirthle, K. Uber die Fettsiure-Cholesterin-Esterdes Blutserums. Hoppe-Seylers Z. physiol. Chem.1896, 21, 331.
2. Bloor, W. R., and A. Knudson. Cholesterol and cho-lesterol esters in human blood. J. biol. Chem.1917, 29, 7.
3. Swell, L., H. Field, Jr., and C. R. Treadwell. Cor-relation of arachidonic acid of serum cholesterolesters in different species with susceptibility toatherosclerosis. Proc. Soc. exp. Biol. (N. Y.)1960, 104, 325.
4. Rehnborg, C. S., A. V. Nichols, and J. K. Ashikawa.Fatty acid composition of mouse lipids and lipo-proteins. Proc. Soc. exp. Biol. (N. Y.) 1961, 106,547.
5. Albers, H. J., and S. Gordon. Comparison of serumcholesterol esters in gerbil and rat. Proc. Soc exp.Biol. (N. Y.) 1962, 109, 860.
6. Leat, W. M. F. Fatty acid composition of the serumlipids of pigs given different amounts of linoleicacid. Biochem. J. 1963, 89, 44.
7. Hallgren, B., S. Stenhagen, A. Svanborg, and L.Svennerholm. Gas chromatographic analysis of thefatty acid composition of the plasma lipids in nor-mal and diabetic subjects. J. clin. Invest. 1960,39, 1424.
8. Swell, L., H. Field, Jr., and C. R. Treadwell. Re-lation of age and race to serum cholesterol esterfatty acid composition. Proc. Soc. exp. Biol.(N. Y.) 1960, 105, 129.
9. Oncley, J. L. Lipoproteins of human plasma. HarveyLect. 1954-55, 50, 71.
10. Havel, R. J., H. A. Eder, and J. H. Bragdon. Thedistribution and chemical composition of ultra-centrifugally separated lipoproteins in human se-rum. J. clin. Invest. 1955, 34, 1345.
11. Bragdon, J. H., R. J. Havel, and E. Boyle. Humanserum lipoproteins. I. Chemical composition offour fractions. J. Lab. clin. Med. 1956, 48, 36.
2035
DEWITT S. GOODMAN
12. Lindgren, F. T., A. V. Nichols, and N. K. Freeman.Physical and chemical composition studies on thelipoproteins of fasting and heparinized human sera.J. phys. Chem. 1955, 59, 930.
13. Goodman, DeW. S., and T. Shiratori. The fattyacid composition of human plasma lipoprotein frac-tions. J. Lipid Res. 1964, 5, 307.
14. Lindgren, F. T., and A. V. Nichols. Fatty acid com-position of the serum lipoproteins. Ann. N. Y.Acad. Sci. 1961, 94, 55.
15. Rosenfeld, R. S., and L. Hellman. The relation ofplasma and biliary cholesterol to bile acid synthe-sis in man. J. clin. Invest. 1959, 38, 1334.
16. Chobanian, A. V., B. A. Burrows, and W. Hollander.Body cholesterol metabolism in man. II. Measure-ment of the body cholesterol miscible pool andturnover rate. J. clin. Invest. 1962, 41, 1738.
17. Gidez, L. I., and H. A. Eder. Cholesterol turnoverin man in Effects of Drugs on Synthesis and Mo-bilization of Lipids, E. C. Horning, Ed. NewYork, Pergamon Press, 1963, p. 67.
18. Goodman, DeW. S., and T. Shiratori. The in vivoturnover of different cholesterol esters in rat liverand plasma. J. Lipid Res., in press.
19. Goodman, DeW. S., and T. Shiratori. The turnoverof individual cholesterol esters in human and ratplasma. Biochim. biophys. Acta (Amst.) 1964, 84,104.
20. Sperry, W. M., and M. Webb. A revision of theSchoenheimer-Sperry method for cholesterol de-termination. J. biol. Chem. 1950, 187, 97.
21. Bray, G. A. A simple efficient liquid scintillator forcounting aqueous solutions in a liquid scintillationcounter. Analyt. Biochem. 1960, 1, 279.
22. Wagner, A. F., and K. Folkers. The organic andbiological chemistry of mevalonic acid. Endeavour1960, 20, 177.
23. Cornforth, R. H., K. Fletcher, H. Hellig, and G.Popjak. Stereospecificity of enzymic reactions in-volving mevalonic acid. Nature (Lond.) 1960,185, 923.
24. Popja'k, G., and J. W. Cornforth. The biosynthesisof cholesterol. Advanc. Enzymol. 1960, 22, 281.
25. Glomset, J. A. The mechanism of the plasma cho-lesterol esterification reaction: plasma fatty acidtransferase. Biochim. biophys. Acta (Amst.)1962, 65, 128.
26. Kritchevsky, D., I. L. Shapiro, and N. T. Werthessen.Biosynthesis of cholesterol in the baboon. Bio-chim. biophys. Acta (Amst.) 1962, 65, 556.
27. Gidez, L. I., and H. A. Eder. Cholesterol metabo-lism in man (abstract). Biochem. Pharmacol.1961, 8, 86.
28. Roheim, P. S., D. E. Haft, L. I. Gidez, A. White,and H. A. Eder. Plasma lipoprotein metabolismin perfused rat livers. II. Transfer of free andesterified cholesterol into the plasma. J. clin. In-vest. 1963, 42, 1277.
29. Haft, D. E., P. S. Roheim, A. White, and H. A. Eder.Plasma lipoprotein metabolism in perfused ratlivers. I. Protein synthesis and entry into theplasma. J. clin. Invest. 1962, 41, 842.
30. Hagerman, J. S., and R. G. Gould. The in vitrointerchange of cholesterol between plasma and redcells. Proc. Soc. exp. Biol. (N. Y.) 1951, 78, 329.
31. Goodman, DeW. S. The metabolism of chylomicroncholesterol ester in the rat. J. clin. Invest. 1962,41, 1886.
32. Kunkel, H. G., and A. G. Bearn. Phospholipid stud-ies of different serum lipoproteins employing P32.Proc. Soc. exp. Biol. (N. Y.) 1954, 86, 887.
33. Eder, H. A. The lipoproteins of human serum.Amer. J. Med. 1957, 23, 269.
34. Havel, R. J., J. M. Felts, and C. Nf. V\an Duyne.Formation and fate of endogenous triglycerides inblood plasma of rabbits. J. Lipid Res. 1962, 3,297.
35. Gitlin, D., D. G. Cornwell, D. Nakasato, J. L. Oncley,W. L. Hughes, Jr., and C. A. Janeway. Studieson the metabolism of plasma proteins in thenephrotic syndrome. II. The lipoproteins. J. clin.Invest. 1958, 37, 172.
36. Havel, R. J. Conversion of plasma free fatty acidsinto triglycerides of plasma lipoprotein fractionsin man. Metabolism 1961, 10, 1031.
37. Portman, 0. W., and M. Sugano. The hydrolysisand fatty acid exchange of cholesterol esters ad-ministered in lipoproteins to Ce'bus monkeys. Arch.Biochem. 1963, 102, 335.
38. Swell, L., and C. R. Treadwell. Cholesterol esterases.III. Occurrence and characteristics of cholesterolesterase of serum. J. biol. Chem. 1950, 185, 349.
39. Deykin, D., and DeW. S. Goodman. The hydrolysisof long-chain fatty acid esters of cholesterol withrat liver enzymes. J. biol. Chem. 1962, 237, 3649.
40. Goodman, DeW. S., D. Deykin, and T. Shiratori.The formation of cholesterol esters with rat liverenzymes. J. biol. Chem. 1964, 239, 1335.
41. Glomset, J. A. Further studies of the mechanism ofthe plasma cholesterol esterification reaction. Bio-chim. biophys. Acta (Amst.) 1963, 70, 389.
![and Effective Antibacterial Activity Electronic Supplementary ... · maleate]∙AgNO3, AgNO3 and PVM/MA were prepared with LB broth in the 15 mL of cultivation tubes. Then, 5 L of](https://static.documents.pub/doc/80x56/6074391d9303e76b8015f624/and-effective-antibacterial-activity-electronic-supplementary-maleateaagno3.jpg)
