ISOTOPIC STUDIES OF PLASMACHOLESTEROLOFENDOGENOUSANDEXOGENOUSORIGINS'

By LEONHELLMAN, ROBERTS. ROSENFELD,MAXWELLL EIDINOFF,DAVID K FUKUSHIMA, AND T. F. GALLAGHER

ANDCHUN-I WANGAND DAVID ADLERSBERG

(From the Divions of Physics and Biophysics and Steroid Biochemistry and Metabolism,Sloan-Kettering Institute for Cancer Research, and the Department of Medicine,

The Mount Sinai Hospital, New York, N. Y.)

(Submitted for publication June 16, 1954; accepted September 8, 1954)

This study was undertaken in order to comparethe metabolic behavior of plasma cholesterol de-rived from the diet with that of cholesterol syn-thesized in vAvo from acetate (1). Cholesterollabeled with either isotopic carbon or hydrogenwas administered orally to human subjects andthe incorporation into plasma cholesterol was fol-lowed over an extended period. In certain in-stances, both endogenous and exogenous choles-terol metabolism were studied simultaneously bythe use of appropriately labeled acetate and cho-lesterol. These techniques were also applied to anexamination of the behavior of plasma cholesterolin four patients with hypercholesterolemia.

EXPERIMENTAL

Subjeces. The patients were hospitalized in the JamesEwing Hospital unit of Memorial Center; all were ingood clinical condition and a summary of pertinent dataappears in Table I. Of the seven patients reported, threehad normal plasma cholesterol levels and four had hyper-cholesterolemia of varied etiology.

Isotopic materials. The precise dosages of the labeledcompounds are recorded in Table I. Cholesterol, eitherradiohydrogen (tritium or H'), heavy hydrogen (deu-terium or H'), or radiocarbon (C4) labeled, was dis-solved in sesame oil and fed in gelatin capsules. Sodiumacetate labeled either with C1' or H' was dissolved in100 ml. of tap water and given orally. The quantity ofradiation delivered by C1' to the patients as well as pre-cautions to be observed in using this isotope in humanshave already been described (1). Approximately 6 mil-licuries (mc.) of Ha are required to deliver the tolerancedose of radiation or 0.3 rep. per week, if this isotopewere completely retained and equally distributed through

1 This investigation was carried out under contractAT(30-1)-910 with the United States Atomic EnergyCommission and supported in part by research grants(C-440) from the National Cancer Institute and (H-564)from the National Heart Institute, of the National In-stitutes of Health, United States Public Health Service.

the body. As the amount of H' administered was con-siderably below this level, and since a large fraction ofthe dose was not retained, the quantity of radiation thepatient receives from the isotope is far below tolerancelevels.

The tritium-labeled sodium acetate with a specific ac-tivity of 0.55 mc. per millimole was made in this labora-tory by exchange of malonic acid with tritium enrichedwater by the method of Halford and Anderson for prep-aration of deuterium-labeled acetate (2). Acetate-2-C'was obtained from a commercial source and had a spe-cific activity of 1 mc. per millimole. Cholesterol-4-C'was prepared in this laboratory from cholestenone4-C'by the method of Belleau and Gallagher (3). The cho-lestenone4-C' was obtained from the Oak Ridge Na-tional Laboratory, and had a specific activity of 2.3 mc.per millimole. Deuterium labeled cholesterol was pre-pared by catalytic exchange of cholesterol with heavywater in the manner described by Bloch and Rittenberg(4). Tritium-labeled cholesterol, with the isotope chieflyat positions C-14 and C-15, was made in this laboratoryby the following procedure which has not been describedpreviously.

A solution of one gram of &`-cholestene-3p-ol acetate(5), 5 ml. (10 curies) of tritium acetic acid, and 30 ml.of anhydrous ether was reduced with 254 mg. of Adams'catalyst and hydrogen gas until the absorption ceased.The crude cholestanyl acetate was treated with a dilutesolution of perbenzoic acid in benzene overnight at roomtemperature. Ten grams of non-isotopic cholestanyl ace-tate were added to the mixture and the diluted cholestane-3fi-ol-14,15-t acetate was purified by chromatography andrecrystallization. Saponification of the acetate followedby oxidation of cholestanol with sodium dichromate gaveeight grams of cholestane-3-one-14,15-t which was thenconverted to four grams of A'-cholestene-3-one (6). So-dium borohydride reduction (3) of the enol acetate pre-pared from the unsaturated ketone afforded cholesterol-14,15-t, m.p. 149 to 149.50.

Blood samples of thirty to forty ml. were collected atfrequent intervals in either oxalated or heparinizedtubes for the isolation of free and ester cholesterol as thedigitonide by the procedure described elsewhere (7).In those cases where cholesterol4-C' was administered,collections of expired carbon dioxide were made in orderto determine whether the steroid nucleus was broken

48

ISOTOPIC STUDIES OF PLASMACHOLESTEROL

TABLE I

Clinical data on patient

Plasmachestero

Ms./Iloo #iL. Labeled materialsWgt

Patient Sex Age (lbs.) Diao Total Free Compound Weight Activity

CHC-1 M 58 130 Pancreas 150 so Cholesterol-4-C0 15.4 mg. 91 pC.carcnoma Acetate-2-H' 98 mg. 700 pc.

CHC-2 F 30 115 Functioning 235 62 Cholesterol4.C4 10.1 mg. 60 pc.adrenal Acetate-2-Hs 67 mg. 480 pc.corticalcarcinoma

CHC-3 F 34 131 Cushing's 200 66 Cholesterol-4-C 9.5 mg. 56 jc.syndrome Acetate-2-Hs 67 mg. 480pac.

CHC4 F 56 115 Hypercholes- 410 106 Cholesterol-4-Cl' 20.6 mg. 122 pc.terolemiaXanthomatendinosum

CHC-5 F 41 116 Hypercholes- 408 122 Cholesterol-4-C4 10.7 mg. 63 c.terolemia Cholesterol-H' 1.0 g. 11.87AD*Xanthomatendinosum

CHC-6 M 43 160 Nephrotic 420 100 Cholesterol-4-C0 10.7 mg. 63 pc.syndrome

CHT-1 F 49 115 Hypercholes- 812 301 Acetate-2-C4 16 mg. 200 gc.terolemia Cholesterol-H' 7.5 mg. 113 gc.Xanthomatuberosum

* AD - atom per cent excess deuterium.

down to compounds that might be oxidized to carbondioxide. Urine and feces were obtained for variableperiods of time to measure the excretion of the admin-istered radioactivity through these routes, as well asfor the isolation of steroid metabolites. These latterdata will be presented in detail in a separate communica-tion, although preliminary results have been reported(8, 9).

Radioactiuity measurements. The methods used forthe measurement of radioactivity of cholesterol digitonidesamples containing ony C' have been previously re-ported (1). The tritium analyses were carried out oncholesterol digitonide by proportional counting in thegas phase by the method described by Eidinoff (10).

In those experiments in which acetate and cholesterolwere given simultaneously, the cholesterol digitonide iso-lated from the plasma contained both isotopes. Thedigitonide was burned in a combustion apparatus similarto that described by Glascock (11) and simultaneous col-lection of both water and carbon dioxide was made. Thewater was converted to hydrogen and counted in thegas phase in the proportional region (10). Radiocarbonactivity was assayed as carbon dioxide admixed withcarbon disulfide in the Geiger-Muller region (12).

The carbon-14 content of certain solid samples con-taining both carbon-14 and tritium was determined inwindowless flow gas counters with an aluminum foil(0.13 mg. per cm') placed over the planchet to absorb

the tritium beta particles. This foil removed approxi-mately 90 per cent of the tritium radiation, but permitted-about 80 per cent of the carbon beta particles to pass.An accurate estimation of the tritium content of a samplecould not be made by this procedure, but since a con-stant portion of the carbon radiation passed throughthe foil, it was adequate for the measurement of thecarbon-14 content of a sample containing both isotopes.2

C = C. Cs/Co - ft,lo -*t

C - true carbon count

C. = net count unshielded

Co = net count shieldedf- = fraction of C04 transmitted through shield

ft= = fraction of H' transmitted through shieldf. and ft, were determined once each week for each alumi-num shield. The f. was measured by counting a shieldedsample containing C14 only, while fh was determined inthe same manner with a sample containing only H'.

The CT content of urine was measured with 0.2 ml.of urine dried on a planchet of 4.9 cm' area containg apiece of lens paper of the same dimensions. The lens

2Calculation for determining radioxarbon by solid count-ing in the presence of; tritium:

49

5 HELLMANET AL.

TABLE II

Radioactivity of plasma chokstro,n subjecs wih normal cholesterol leveJs

CHC-1

From choleruoI4-C4

Time* Freet Estrt

0.125 22.0 3.180.35 61.0 43.00.73 81.2 85.41.2 135.5 137.51.7S 125 iSE.52.2 144 174.52.75 137 179.53.2 129.5 1673.75 129.0 173

CHC-2

From acetate-2-H'

Freet Easter

26.9 2.9516.4 6.6712.5 7.0411.8 6.75

9.16 8.449.21 7.849.21 5.577.80 7.656.48 8.05

From choesterol.4C14

Time* Freet Estert

O.t67 118 93.50.39 722 6100.94 1,500 1,0301.27 1,620 1,1102 1,420 1,2803 1,180 1,2804 941 1,2405 840 1,1207 643 865

10 535 711

From acetate2-H'

Freet Eskt

27.8 5.3218.7 5.37

8.71 7.388.69 7.72

10.92 9.658.86 9.236.49 8.446.17 8.15

6.554.98 3.96

CHC-3

From cholesterol4-Cu4

Tine* Freet Estert

0.184 134 117.50.417 446 4220.805 681 4141.11 622 5621.84 584 6012.18 554 6642.83 476 6124.0 452 5756 344 4887 286 4238 217 368

11 148 24013 139 19814 134 188

* Elapsed time in days after administration of labeled material.t) 108 DPMper mMcholesterol.

X103 DPMper mMH2.TABLE III

Radioactivity of plasma cholesrol after administration of cholesterol-4-C0 to hypercholesterolemic subjects

CHC4

Tit*& Freet

0.04 98.50.62 1,1121.1 1-7601.6 1,7502 1,5752.6 1,4803.1 1,4203.6 1,3355.6 9746.6 9827.6 8308.6 7419.6 728

10.16 66011.6 58712.6 55214.6 55616.6 44120.6 41321.6 39623.6 36326.6 32429.6 30233.6 23636.6 197

140.0 59.0

Estert

1,084

988864791718718597484457397381315326308

82.3

Time*

0.1670.3750.8451.061.441.44ratiot1.872.872.87ratio$3.95678

101313

ratiot151720222428354352667692

105144167

CHC-5

Freet

24.3126.721S246255

0.06050.24t

275210

0.04810.23t

210167162121.5116111

93.20.02210.24t

71.562.858.657.660.346.749.742.339.027.827.619.7

17.012.8

Ester

11.564.3

124.7204204

0.045f0.22:

214255

0.05710.22t

256207197185172152111

0.026§0.23t

10694.482.676.472.262.054.046.338.933.231.220.5

14.313.9

Time*

0.3312357

1014172227344148556269768397

110124138

CHC-6

Freet

320338417425357297239179154104101

69.059.456.643.333.832.333.639.224.020.717.512.5

* Elapsed time in days after administration of labeled material.t X1O' DPMper mMcholesterol.

Ratio of deuteriumn content X I0 to radiarbon content.I Atom per cent excess deuteriumn in cholesterol (AD).

Estert

408622604437404338266215169117.5

93.777.368.774.055.744.836.437.2

25.423.9

ISOTOPIC STUDIES OF PLASMACHOLESTEROL

paper facilitated even distribution over the surface of theplanchet to give plates of uniform thickness. Feces wereextracted three times in a Waring Blendor with. threevolumes of acetone. A portion of the acetone extractwas plated on lens paper in the manner described forurine.

The various counting procedures were intercalibratedwith a reference sodium carbonate-Cu standard obtainedfrom the National Bureau of Standards. The radioac-tivity of cholesterol samples contaming carbon-14 is ex-pressed in units of "disintegrations per minute per milli-mole" (DPM per mM). The tritium radioactivity dataare expressed in units of "disintegrations per minute permillimole of hydrogen" (DPM per mMH,).

RESULTS

Three patients, CHC-1, CHC-2, CHC-3, withnormal plasma cholesterol levels each receivedfrom 50 to 90 microcuries (jsc.) of cholesterol4-C4 and from 500 to 700 pLc. of tritium-labeled ace-tate. Three other patients, CHC4, CHC-5,CHC-6, with elevated plasma cholesterol levels,received from 60 to 120 plc. of cholesterol-4-C4;1.016 gm. of cholesterol containing 11.87 atomsper cent excess deuterium was given to subjectCHC-5 along with the C1' cholesterol. An addi-tional patient with hypercholesterolemia, CHT-1,received 113.2 p,c. of tritium labeled cholesteroland 200 pLc. of- acetate-2-C04.

The specific activities of free and ester choles-terol isolated from patients CHC-1, CHC-2,CHC-3 are listed in Table II. Table III containsthe specific activity data from subjects CHC4,CHC-5, CHC-6, while the results obtained inCHT-1 are listed in. Table IV.

The data obtained for four and a half days fol-lowing the simultaneous administration of acetate-Ha and cholesterol4-Cl' to subject CHC-1 areplotted in Figure 1. As shown on the lower rightportion of this figure, the specific activity of freecholesterol synthesized in vivo from acetate is ata maximum with the earliest sample and then de-clines rather rapidly. The ester cholesterol de-rived from acetate is at a minimum value with thefirst sample and then gradually rises until it in-tersects the free curve at approximately three days,after which the specific activity of the ester cho-lesterol exceeds that of the free. For comparisonwith these tritium data, specific activity curvespreviously reported from a patient who received

TABLE IV

Radioadivit7 of plasma cholesterol in a hypercholesteroemic-. sbect given aceate-2-C and chlesterol-H'

CHT-1

From acetate-2-C4

Time*

0.1250.33-0.80.1.261.732.733.734.736.738.73

10.7313.7317.7322252932363943496186

115

Freet

568535450372365263222180166154118108

88.482.461.4

53.652.850.044.843.336,628.920.4

Estert

18.153.785.7

122.8126.5154.2155.0172.0151.8158.2129.5117.5102.0

86.085.5

68.562.655.055.748.437.033.823.3

From cholesterol-Hs

Freet Ester:3.3 2.79

25.3 27.341.7 43.041.1 67.850.3 68.849.8 78.648.6 66.042.8 63.641.5 54.236.4 46.229.2 35.328.2 31.925.1 28.621.0 25.818.6 22.4

15.8 18.114.8 16.213.7 15.812.7 15.611.95 13.9

11.28.15 11.856.62 6.12

* Elapsed time in days after administration of labeledmaterial.

. X1O8 DPMper mMcholesterol.X10 DPMper mMH2.

C14-labeled acetate (1) are shown in the upperright portion of Figure 1. The behavior of cho-lesterol synthesized from acetate is similarwhether the acetate is labeled with either tritiumor CiS, and these data and others to be presenteddemonstrate the interchangeable use of these twoisotopes for the study of the conversion of acetateto cholesterol.

The activity of plasma free cholesterol origi-nating in the diet is at a minimum value with the3-hour sample, as illustrated on the left portionof Figure 1. The specific activity then rises to apeak at about 1.5 to 2 days and afterwards de-clines. The radioactivity of the "exogenous"plasma ester cholesterol is also at a minimum withthe first sample, rises less rapidly than that of thefree sterol, and reaches a peak value at about 2.5days, having intersected the free curve at about1 day. After this crossover point, the esterspecific activity is greater than that of the free forthe duration of the period of observation.

Figure 2 compares the incorporation of dietary

51

HELLMANET AL.

SIMUIANEOUSUTILIZATION OF DIETARYAT-H3FOR PLASMACLEROLSYNTHES AS COMAREDTO TION F1 ACETATE-2-C4

'°t EXOGENOUS INCORPORATION 2x I0n ENDOGENOUS SYNTHESIS

e,k-x --x----ax0-., ~ 0-

,;ox CHOLESTEROL-4-C

I x-T--< P bwm Fre (C") Cholp erolI 0---O Plosm Ester (C'4) Cholterol

I5

I

1 2Days

106

.C

I*(Z)

aI

*1

E

I

3 4

ACETATE-2-C04

x-x Plasma Fre (C14?Choleser0-o Plosma Ester(C ) Cholesterol

ACETATE-2-H3- ~~ ~AA

0 -0~~~~~~/ Plsmo Free (IH) oe l0-a Plasma Ester (H) Ch

I I I

1 2Days

FIG. 1. THE SIMULTANEOUSUTILIZATION OF INGESTED CHOLESTEOL-4-C' AND AcETATz-2-t FOR PLASMACHOLESTEROLSYNTHESIS AS COMPAREDWITH THE INCORPORATIONOF ACTATE-2-C"

cholesterol-4-C4 into plasma cholesterol of a sub-ject with a normal cholesterol level (CHC-2) witha patient who had hypercholesterolemia accom-

panying the xanthoma tendinosum syndrome(CHC-5). It can be seen that the mode of riseof free and ester cholesterol, the points of inter-section, and the subsequent rate of decline over a

ten-day period, are virtually indistinguishable.The specific activity of the plasma cholesterol inthe normocholesterolemic patient is higher; how-ever, this difference may be related to individualvariation in the quantity of labeled cholesterolthat had been absorbed. The dilution of the ad-ministered labeled sterol by the larger plasmacholesterol pool in the hypercholesterolemic mightalso explain the lower specific activity.

In Figures 3 and 4, results obtained over a

longer period are presented for comparison of therates of decline of endogenously and exogenouslyderived cholesterol. The plasma free cholesterolspecific activities measured after the administra-tion of labeled cholesterol to the hypercholestero-lemic subject, CHC-4, are plotted in Figure 3 to-gether with the specific activities of plasma free

cholesterol of a normocholesterolemic subject whoreceived labeled acetate. The acetate data havebeen previously reported (1). With the exceptionof the early differences in the behavior of endoge-nously or exogenously derived plasma cholesterolalready described above, it will be noted that theradioactivity of free cholesterol whether absorbedfrom the diet or made in vivo, declines at a simi-lar rate in these two patients during the 30-dayperiod of observation. In Figure 4, the equiva-lent data for ester cholesterol in the hypercholes-terolemic subject CHC-5 and the typical acetatesubject are shown and the similar rates of declineof sterol radioactivity in these dissimilar patientsare again apparent.

Subject CHT-1 who had hypercholesterolemiaassociated with the xanthoma tuberosum syn-drome received cholesterol-Ha and acetate-2-Cl'simultaneously. This patient was studied for sixmonths and Figure 5 is a comparison of the be-havior of cholesterol arising from these twosources for the first 43 days. The findings inthis patient differ in two respects from those ob-tained in normocholesterolemic subjects and in

2105

£I)NE

0

a

3 4

52

ISOTOPIC STUDIES OF PLASMAiCHOLESTEROL

INCORPORATION OF DIETARY CHOLESTEROL-4-C14 INTO PLASMACHOLETER IN NORMALAND HYPERCHOLESTEROLEMICSUBJECTS

2106~NORMAL

I. U , , ---' ---------b-

; 0 otJ- HYPEROHOLESTEROLEMIA ----___- o

/ .xPlas.Free Cholesterol fromCholesterol-4-C.4o----oPhasma Ester chobesterol from Cholesterol-4-C14

*)r~~~~~~~~~~~~~~~-

IO4 24 6 8 -10Days

FIG. 2. THZ INCORPORTIONOF IN4GTESD CHOLZSTnOL-4-CSINTO PLASMACHOLESTEOLIN SUBJECTSWITHNORAL CHOLESTEROLLEvELSAND Ix HYPERcHoLEs

INCX1OPIKARATION Of ACEATE .-C14 ANDDIETARYCOLSEO-444 KMPAM RE CHtiESTEROL

o9

-0 6 X-X Pka Fre CholOwW r Acdetf-2-e

10~~~~~~~~-.

o x

05 ---------P Fe C ro---o PlasmaEe C

l~~~~~~~~~~~~~~~~~~~~~~~

10 5 O 15 2026 38Days

FIe. 3., Tm-INCORPOATION OF AcErATz-2-C" AND INGESTED CHOLESTEROL-4-C' INTO PLASMAFR CHOLESTzOL

53

HEULMANET AL.

INCORPORATIONOF ACETATE-2-C4 AND DIETARYOHNE$MERL.4-C14 NTO PLASMAESTER CHOLESTEROL

0-o Plasma Ester Chosterol fomn Acsette-2-CI40---O Plsma Ester Chodsrol from Cholesterol-4-C4

Days

FIG. 4. THE INCORPORATIONOF AcrATE-2-VY AND INGESTED CHOLESTOL-4-C" INTO PLASMAEsTm CHOLESTEROL

hypercholesterolemic subjects with the xanthomatendinosum syndrome. Free cholesterol synthe-sized from acetate, although at a peak value withthe initial sample, declines more slowly duringthe first few days of observation, so that the risingspecific activity curve of ester cholesterol does not

aS

5x IC

105

zh2

intersect the free curve until about the sixthday. In all patients previously studied with ace-tate, this point of intersection occurred at two tothree days (1). Plasma cholesterol derived fromthe diet, differs in that the ester cholesterol spe-cific activities are initially at a higher level, rise

SIMULTANEOUS INCORPORATIONOF DETARY CHOLESTEROL- H3AND ACETATE-2-C4 INTO PLASMA CHOLESTEROL

IN HYPERCHOLESTEROLEMIA

ENOOGENOUSSYNTHESIS

x-x Plsma Free Cholsterol from Acetote-2-C40-O Plsma Ester Cholesterol from Acetate-2-C'4A.A Plasmo Free Cholesterol from Cholesterol-H3o---c Plasma Ester Cholesterol from Cholesterol-HS

I ~~~~~ ~ EXOGENOUSINCORPORATION

5 10 IS 20 25 30 35 40450r

54

.2

2

E

~l5

11

lgo

eV' 5 10 15 20 25 30 35 40 45Days

FIG. 5. SIMULTANEOUSINCORPOATIONOF INGESTED CKoszSTmOL-t ANDAczTATE-2-Cu INTO PLASMACHOLESTEROL Ix HYPEcHOLESTEOIMLA

ISOTOPIC STUDIES OF PLASMA CHOLESTEROL

INCORPORATIONOF DIETARY CHOLESTEROL-4-C14 INTOPLASMACHOLESTEROLIN THE NEPHROTIC SYNDROME

i7.

106- x----x Plasma Free Cholesterol from Cholesterol-4-C140-0 Plasma Ester Cholesterol from Cholesterol-4-C04P%.^~~~~~~~~~~~~~~~~~~~ ~~ ~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

,sv,

rX- -- -:- -- x

-- ~ ~ ~ ~

5 10 15 20 25 30 35 40Days

45 50

FIG. 6. INCOMR'oATnON OF INGESTED CHOLESTEROL-4-C1 INTO PLASMACHOLESTEROLIN THE NEPHROTIC SYNDROME

more rapidly, and exceed those of the free through-out the experiment. Since the free cholesterolspecific activities are always below the ester val-ues, the two specific activity curves do not in-tersect. The latter parts of the curves in Figure 5show parallel rates of decline of cholesterol radio-activity similar to those observed in other patientswith hypercholesterolemia and in patients withnormal plasma cholesterol levels (Figures 3 and4). These data derived after the simultaneous ad-ministration of labeled acetate and cholesterol showthat either labeled compound yields identical re-

sults when used for the study of long term as-

pects of the behavior of plasma cholesterol.Cholesterol-4-Cl" was also given to a patient,

CHC-6, who had hypercholesterolemia accom-

panying the nephrotic syndrome; the radioactivityof the plasma cholesterol is shown in Figure 6.Its behavior in this patient is similar to that ob-

served in the subject with xanthoma tuberosumCHT-1.

An analysis of the specific activity curves interms of component rate processes is presented inTable V and the similarities mentioned above forthe decline of cholesterol radioactivity during thelatter part of the experiments may be observed.

The radioactivity excreted in urine and feces in

patients who received C1' cholesterol, as well as

the levels of activity obtained in the blood, are re-corded in Table VI. The recovery of administeredradioactivity in feces by the fourth day varied from14 to 48 per cent. A much smaller proportion,from 0.35 to 1.76 per cent, of the administeredmaterial was excreted in the urine during thesame time interval. In the subjects whose plasmacholesterol specific activity curves exhibited acrossover of the free and ester cholesterol, 4 to 14per cent of the ingested C1' sterol was present incirculating cholesterol at the peak radioactivity ofplasma free cholesterol. In CHT-1 and CHC-6,where no crossover occurred, 27 and 12 per cent,

TABLE V

Component rates of decay curve of plasma chosaterolderied from labded dietary cholesterol

Decay rats

a bSubject ti t}

CHC-4 22 3.0CHC-5 22 2.5CHC-6 20 4.0CHT-1 19 1.9CHT-1' 17 1.4

AC-3t 23.0 2.3

* Components of curves derived from acetate-2-C4 givensimultaneously with cholesterol-Hs to subject CHT-1.

t Components of curve derived from acetate-2-C4 aspreviousty reported (1).

- I

0

0C

E

a.I

55

HELLMANET AL.

TABLE VI

Amount of labded cholesterol found in plasma as comparedwith the fraction creted in urine and feces

Intersection Total dose Four-day excretion

Subject Days In plasma* Urine Feces

~% %CHC-2 2.2 14 1.76 26CHC-3 1.9 7 1.24 14CHC-4 t 12 0.35 27CHC-5 2.9 4 1.35 48CHC-6 t 12 0.18 -CHT-1 t 27

* This value was calculated from the data obtained atthe intersection of the free and ester curves in subjectsCHC-2, -3, 4, and -5. Since the curves of subject CHC-6and CHT-1 did not intersect, the values are calculatedfrom the maximum observed specific activities of free andester cholesterol.

t As discussed in the text, the curves of subjects CHC-6and CHT-1 did not intersect. The early portions of the"ester" data in subject CHC-4 were lost due to a labora-tory accident.

respectively, of the labeled dietary cholesterol wasfound in the plasma cholesterol at the maximumspecific activity of the free fraction.

Expired carbon dioxide had no measureableradioactivity; as little as 0.1 per cent of the ad-ministered dose per day. could have been detectedby the method used.

DISCUSSION

Absorption and excretion of labeled cholesterol

The' specific activity of plasma cholesterol wouldnse rapidly to an early peak, if the process of cho-lesterol absorption from the intestine was rela-tively rapid. This would be analogous to the be-havior of plasma glucose after a glucose tolerancetest. However, the experimental results in manshow that the peak specific activity is not reacheduntil approximately the second or third day. Thissuggests either that absorption is slow, that ab-sorbed cholesterol is retained in the intestinalmucosa, or that newly absorbed cholesterol is rap-idly removed from the blood and then later rein-troduced' into the circulation. Biggs and hisco-workers (13) concluded that cholesterol absorp-tion was slow, since they observed a sustained fe-cal excretion of radioactivity after feeding tritiumlabeled cholesterol. However, these authors meas-ured either the total fecal radioactivity, or in oneinstance, the activity of a crude extract that con-tained cholesterol, coprosterol and possibly other

steroids. Therefore, the fraction of the total radio-activity in the form of unabsorbed cholesterol andthe fraction which represented cholesterol thathad been absorbed, transformed into other metab-olites and subsequently excreted in feces is un-known. The experiments of Siperstein, Jayko,Chaikoff, and Dauben (14) indicate that afteradministration of cholesterol the sustained fecalexcretion of radioactivity is chiefly in the form ofacidic transformation products, such as bile acids.Animal studies (15, 16) show that dietary cho-lesterol is absorbed within one day and is esteri-fied during the absorptive process. Both the freeand esterified sterol pass into the intestinal lym-phatics and enter the venous side of the circulation.Although the rate of cholesterol absorption inman has not been directly determined, it seems un-likely that slow absorption is responsible for therelatively delayed appearance of the maximumspecific activity.

No information is available about the possibleholdup of cholesterol in the intestinal mucosa, al-though this would seem unlikely in view of itsrapid appearance in lymph (15). With respect tothe possibility that newly absorbed cholesterol isinitially removed from the blood, animal experi-ments show that injected colloidal suspensions ofcholesterol resembling chylomicrons rapidly disap-pear from the circulation and that the cholesterolthus removed subsequently reappears in theplasma (17, 18). In addition, although choles-terol esterification during the process of intestinalabsorption is well documented (15, 16,'19), thereis no evidence that the esters thus formed areidentical with those normally found in the circu-lation. The slow rise observed in the plasmafree and ester specific activity curves can tenta-tively be ascribed to relatively rapid absorption ofthe sterol from the intestine, rapid removal of thenewly absorbed sterol from the blood, followed byits subsequent reappearance in the circulation.

The absorption of the fed cholesterol-4-Cvaried from 52 to 86 per cent (Table VI) as de-turmined by' the amount that was excreted in thefeces during the first four days. These.values arehigher than those reported for the absorption ofone granmi doses of tritium labeled cholesterol (13).With the exception of- CHC-5, the subjects in thisstudy received approximately 10 mg. of labeled

56

ISOTOPIC STUDIES OF PLASMACHOLESTEROL

cholesterol and the efficiency of absorption may berelated to the large difference in the amount fed.It should be noted that patient, CHC-5, who re-ceived one gram of cholesterol-d with the tracerdose of radioactive sterol showed the smallest ab-sorption of cholesterol (52 per cent). The por-tion of the fecal radioactivity derived from cho-lesterol-4-Cl' that had been absorbed, metabo-lized, and subsequently excreted, and that whichhad completely escaped absorption are not known.Since only 0.35 to 1.76 per cent of the cholesterol-4-C' appeared in the urine, the kidney is a minorroute for the excretion of the steroid nucleus; aconsiderable fraction of the urinary radioactivityis in the form of steroid hormone metabolites (9).

After the feeding of cholesterol, from 4 to 15per cent was present in the circulation at themaximum specific activity of free cholesterol(Table VI). Without considering any other tis-sues, the human liver contains about 4.5 gm. ofcholesterol in isotope equilibrium with plasmacholesterol (18) and an additional three grams ofsterol are in the red cells, which after 24 to 36hours are also in equilibrium with plasma freecholesterol (9). Thus approximately 8 to 33 percent of the administered dose is present in theliver and circulation, thereby accounting for asignificant fraction of the sterol absorbed from thediet.

The fact that no labeled carbon dioxide was de-tected after the administration of cholesterol-4-C1shows that carbon-4 remained attached to the restof the steroid nucleus. This suggests that thesteroid nucleus is not degraded and is probablyeliminated from the body in an intact form, eitherthrough the urine, feces or skin. This finding isin agreement with the results of Chaikoff and hiscolleagues (20) who have shown that the carbon-4of cholesterol-4-0C1 was not converted to carbondioxide by the rat, although cholesterol labeledat C-26 can be oxidized with the appearance ofC1' 02 in the expired air.

Plasma cholesterol radioactivity curves

In all the patients who received labeled cho-lesterol, except CHT-1 with xanthoma tuberosumand CHC-6 with the nephrotic syndrome, theplasma cholesterol specific activity curves behave

in the characteristic manner described. The dataof Biggs and his co-workers (13) who fed tritiumlabeled cholesterol are essentially similar to thosereported here except that the specific activity oftotal plasma cholesterol was measured rather thanthat of ester cholesterol. Since 60 to 75 per centof plasma cholesterol is esterified, measurementsof total cholesterol give a good approximation ofthe specific activity of the ester fraction. Warner(21) mentions that the specific activity of freecholesterol is higher than that of ester during the"period of active absorption."

The curves for plasma free and ester cholesterolderived from the diet decline at similar rates afterreaching their maxima and are comparable to thebehavior of plasma cholesterol synthesized in thebody from acetate. The ester specific activity re-mains higher than that of the free because thefree cholesterol is continually being diluted bynon-labeled cholesterol from the diet or by en-dogenous synthesis. A more complete descrip-tion of this process appears elsewhere (1). Thebehavior of the free and ester cholesterol shownin Figure 2, fits the criteria of Zilversmit, En-tenman, Fishler, and Chaikoff (22) for the rela-tionship between precursor (free) and product(ester). Inasmuch as free cholesterol was given,the free sterol must be the ultimate precursor ofthe plasma ester cholesterol.

The declining portion of the specific activitycurves may be resolved into the component ex-ponentiaL rates listed in Table V. Since bothdistribution phenomena and biochemical reac-tions contribute to the processes which accountfor the decline in specific activity, no unique sig-nificance can be attached to these rates; however,they are useful in characterizing the curves.

The congruity of the specific activity curves de-rived from either C' or tritium labeled acetate,as illustrated in Figure 1, shows that acetatelabeled with either isotopes may be used for thestudy of cholesterol metabolism. It is also note-worthy that the plasma cholesterol of subjectCHC-5 who received cholesterol-4-Cl' and cho-lesterol-d simultaneously possessed a constantdeuterium to C1' ratio at several points on thespecific activity curve where measurements weremade (Table III). This implies the identical

57

HELLMANET AL.

handling of cholesterol-4-0C and deuterium-la-beled cholesterol by the body.

If the rates describing the decline of plasmacholesterol derived from cholesterol-4-C0 arecompared with those representing the decline ofplasma cholesterol formed by endogenous syn-thesis from acetate-2-C14, it will be noted thatthey are remar kably similar. This is illustratedgraphically in Figures 3 and 4. Either labeledcholesterol or labeled acetate may be used inter-changeably to trace the decay curve of plasmacholesterol, provided it is recognized that theearly portions of the specific activity curves de-rived from these two tracer materials are different.

Studies in hypercholesterolemia

The altered relationship between free and estercholesterol in the subject with xanthoma tubero-sum (Figure 5) as compared with patients withxanthoma tendinosum or normocholesterolemicsubjects has been observed previously (23).While there are no comparable studies in nephro-sis, the similarity of the plasma cholesterol speci-fic activity curves in xanthoma tuberosum withthose of the nephrotic syndrome (Figure 6)should be emphasized. Since in both the xan-thoma tuberosum and the nephrotic syndromesbut, not in xanthoma tendinosum, the plasma ap-pears lipemic and excessive amounts of neutralfat, phospholipids and the abnormal Sf 10-400lipoproteins (24) are found, the higher initialspecific activity of plasma ester cholesterol mayperhaps be correlated with these findings whichare characteristic of essential hyperlipemia in ad-dition to hypercholesterolemia. Excessive esteri-fication during absorption or a modified esterifica-tion process during lipoprotein synthesis by livermay be responsible for the abnormal relation be-tween plasma free and ester cholesterol.

There are no findings in the data from the fourpatients with hypercholesterolemia that might aidin explaining the sustained elevation of the plasmacholesterol. The decline of the specific activitycurves in these patients is not appreciably differ-ent from those of patients with normal cholesterollevels. Since the component exponential processeswhich form the decay curves are many in number,and have not been individually evaluated by any

experimental procedure, small deviations thatnight be implicated in the maintenance of theelevated cholesterol levels may be "buried" in thecurve. Small sustained differences in the rate ofdecay could bring about large cumulative changesin the plasma cholesterol level.

Interrelations of dietary and biosynthetic cho-lesterol

Dietary cholesterol and cholesterol made invivo from acetate differ in metabolism only dur-ing the immediate post-absorptive and post-syn-thetic period in that biosynthetic cholesterolrapidly appears in the blood while the peak circula-tory level of dietary cholesterol is delayed. How-ever, the radioactivity data indicate conclusivelythat the two "species" of cholesterol merge by thethird or fourth day following synthesis or inges-tion. The fed, performed cholesterol is then in-distinguishable from that synthesized from acetateexcept by application of isotope techniques. Sincebiosynthetic cholesterol is constantly secreted inthe bile and mixes in the intestine with cholesterolpresent in food, a mixture of preformed and bio-synthetic cholesterol is always available forabsorption.

One difference between fed and biosynthesizedcholesterol may be in the mode of transport im-mediately after either formation or absorption.It is likely that biosynthetic cholesterol is secretedinto the circulation from the site of synthesis incombination with lipoprotein. The molecular formin which dietary cholesterol enters the circulationvia the lymph is unknown and may differ in physi-cal properties from lipoprotein-bound endogenouscholesterol. Since the physiological processesthat give rise to both "species" are continuous,both forms are present in the circulation at thesame time, and the possibility remains that theimmediate post-absorptive phase of cholesterolmetabolism is critical; further studies of these im-plications are in progress.

SUMMARY

1. Cholesterol and acetate, labeled with eitherradiocarbon or radiohydrogen, have been used totrace the metabolic behavior of plasma cholesterolsynthesized in vivo or derived from the diet in

58

ISOTOPIC STUDIES OF PLASMACHOLESTEROL

human subjects with normal and elevated plasmasterol levels.

2. From 4 to 27 per cent of a 10 mg. dose oflabeled cholesterol appeared in the plasma afteroral administration, 14 to 48 per cent was re-covered in the feces during the first four days andapproximately one per cent appeared in the urine.As suggested by the absence of measureable radio.activity in expired carbon dioxide, there was noevidence for the breakdown of the steroid nucleus.

3. Traces of orally administered cholesterol-4-C14 could be detected in the plasma within onehour and the maximum radioactivity of free andester cholesterol was attained in two to three days.This is in contrast to labeled plasma cholesterolsynthesized in vivo from radioactive acetate whichreaches peak specific activity within eight hours.

4. No differences in the patterns of incorpora-tion could be detected in two patients with hy-percholesterolemia accompanying the xanthomatendinosum syndrome when compared with sub-jects having normal sterol levels. Differences inthe relation of free to ester cholesterol were ob-served during the incorporation of dietary cho-lesterol in a patient with hypercholesterolemiaaccompanying the xanthoma tuberosum syndromeand in another subject with hypercholesterolemiaof the nephrotic syndrome.

5. Aside from early differences in the appear-ance of biosynthetic and dietary cholesterol inplasma cholesterol, the data suggest that cho-lesterol derived from the diet is eventually in-distinguishably mixed with cholesterol synthe-sized in the body.

ACKNOWLEDGMENTS

The authors are indebted for the devoted assistance ofVincent Abate, Mones Berman, William Considine,Hilda Denman, Albert Deutsch, Larrier Hendrickson,Joseph Knoll, Ruth Loevinger, and Barbara Weintraub.

REFERENCES

1. Hellman, L., Rosenfeld, R. S., and Gallagher, T. F.,Cholesterol synthesis from C'-acetate in man.J. Chin. Invest., 1954, 33, 142.

2. Halford, J. O., and Anderson, L. C., Organic deu-terium compounds. Acetic, malonic and succinicacids. J. Am. Chem. Soc., 1936, 58, 736.

3. Belleau, B., and Gallagher, T. F., The conversion ofA'-cholestene-3-one to cholesterol. J. Am. Chem.Soc., 1951, 73, 4458.

4. Bloch, K., and Rittenberg, D., The preparation ofdeuterio cholesterol. J. Biol. Chem., 1943, 149,505.

5. Schenck, F., Buchholz, K., and Wiese, O., Unter-suchungen am 7-Dehydro-cholesterin. Ber. ddeutsch. chem. Gesellsch., 1936, 69, 2696.

6. Rosenklranz, G., Mancera, O., Gatica, J., and Djer-assi, C., Steroids. IV. Alpha-iodoketones. Amethod for the conversion of allosteroids intoA'-3-ketosteroids. J. Am. Chem. Soc., 1950, 72,4077.

7. Rosenfeld, R. S., Hellman, L., Considine, W. J., andGallagher, T. F., Isolation of radioactive choles-terol from plasma. J. Biol. Chem., 1954, 208, 73.

8. Hellman, L., Rosenfeld, R. S., Fukushima, D. K.,Gallagher, T. F., and Dobriner, K., Utilization ofC'-acetate for cholesterol and steroid hormonesynthesis in the human. J. Clin. Endocrinol., 1952,12, 934.

9. Hellman, L., Rosenfeld, R. S., Fukushima, D. K,Eidinoff, M., and Gallagher, T. F., The majorpathways of steroid metabolism in man. J. Cin.Invest., 1953, 32, 573.

10. Eidinoff, M., Internal gas counting of radioactivecarbon and hydrogen. American Institute of Elec-trical Engineers, 2nd Annual Joint Conference onElectronics in Nucleonics and Medicine, New York:A. I. E. E., 1950, p. 65, (A. I. E. E. Special publi-cation No. 33).

11. Glascock, R. F., A combustion technique for theassay of tritium, 'C, and 14C in a single 10 mg.sample of biological material. Biochem. J., 1952,52, 699.

12. Eidinoff, M. L., Measurement of radiocarbon as car-bon dioxide inside Geiger-Muller counters. Analyt.Chem., 1950, 22, 529.

13. Biggs, M. W., Kritchevs-ky, D., Colman, D., Gofman,J. W., Jones, H. B., Lindgren, F. T., Hyde, G.,and Lyon, T. P., Observations on the fate of in-gested cholesterol in man. Circulation, 1952, 6,359.

14. Siperstein, M. D., Jayko, M. E., Chaikoff, I. L., andDauben, W. G., Nature of the metabolic productsof C"-cholesterol excreted in bile and feces. Proc.Soc. Exper. Biol. & Med., 1952, 81, 720.

15. Bollman, J. L., and Flock, E. V., Cholesterol in in-testinal and hepatic lymph in the rat. Am. J.Physiol., 1951, 164, 480.

16. Friedman, M., Byers, S. O., and Shibata, E., Obser-vations concerning the production and excretionof cholesterol in mammals. X. Factors affectingthe absorption and fate of ingested cholesterol.J. Exper. Med., 1953, 98, 107.

17. Bevans, M., Abell, L. L., and Kendall, F. C., Produc-tion of intimal atherosclerosis by intravenous in-jection of colloidal cholesterol into rabbits. Fed-eration Proc., 1948, 7, 269.

18. Unpublished results from this laboratory.

59

HELLMANET AL.

19. haikoff, I. L., Bloom, B., Siperstein, M. D., Kiyasu,J. Y., Reinhardt, W. 0., Dauben, W. -G., andEastham, J. F., C'-cholesterol. I. Lymphatictransport of absorbed cholesterol-4-C. J. Biol.Chem., 1952, 194, 407.

20. Chaikoff, I. L., Siperstein, M. D., Dauben, W. G.,Bradlow, H. L, Eastham, J. F., Tonkins, G. M.,Meier, J. R., Chen, R. W., Hotta, S., and Srere,P. A., Cu-cholesterol. II. Oxidation of carbons4 and 26 to carbon dioxide by the intact rat. 3.

Biol. Chem., 1952, 194, 413.21. Warner, I., Rate of appearance of dietary cholesterol

in blood. Federation Proc., 1952, 11, 306.

22. Zilversait, D. B., Entenman, C. Fishler, M. C, andChaikoff, I. L., The turnover rate of phospholipidsin the plasma of the dog as measured with radio-active phosphorus. J. Gen. Physiol., 1943, 26, 333.

23. Biggs, M. W., and Colman, D., A quantitative meta-bolic defect in lipid metabolism associated withabnormal serum lipoproteins in man. Circulation,1953, 7, 393.

24. Gofinan, J. W., Jones, H. B., Lindgren, F. T., Lyon,T. P., Elliott, H. A., and Strisower, B., Bloodlipids and human atherosclerosis. Circulation,1950, 2, 161.

60