journal 2 - t3 and ptu on rats

8/3/2019 Journal 2 - T3 and PTU on Rats

1/8

Effects of triiodothyronine and propylthiouracil onplasma lipoproteins in male ratsHenry G. Wilcox, William G. Keyes, Thomas A. Hale,' Richard Frank,Douglas W. Morgan, and Murray HeimbergDepartment of Pharmacology, University of Tennessee Center for the Health Sciences, Memphis,T N38163 and Department of Pharmacology, University of Missouri School of Medicine,'Columbia, MO 65212

Abstract Hyperalphalipoproteinemia, characterized by in-creased plasma concentrations of apoA-I and of HDL lipidand protein, was observed in rats treatedwith triiodothyronine(T,) for 7 days. The increase in the plasma HDL apoproteinswas general for apoC, apoE plus A-IV, and apoA-I, as deter-mined by isoelectric focusing. Hypotriglyceridemia, charac-terized by decreased concentrations of VLDL and apoB, wasalso observed in the hyperthyroid state. Although in the mildlyhypothyroid animals (propylthiouracil-treated), hepatic me-tabolism of free fatty acid is shifted toward esterification totriglyceride and VLDL formation, as we reported previously,plasma HDL and apoA-I concentrations were not differentfrom control plasma values, while the d 1.006-1.063 g/ml(IDL + LDL) lipoprotein fraction tended to be increased. Ingeneral, the proportion of apoE in the (IDL + LDL) fractionof the hypothyroid rat was greater than in controls and hy-perthyroid animals, while the proportion of apoE tended tobe ower n VLDL from both hypo- and hyperthyroid ratsthan in VLDL from controls. An enhanced release of apoA-I by perfused livers isolated rom rats treated with TSwas alsoobserved; this enhanced output of apoA-I may explain, in part,the hyperalphalipoproteinemia observed in these rats. Thedepressed net output of apoA-I in vitro by perfused liversfrom rats treated with propylthiouracil (PTU) was not ex-pressed in a statistically significant iminished plasma concen-tration of HDL or apoA-I in the intact animals. Treatmentwith T s also resulted in modification ofhe content of essentialfatty acids in various lipid classes. Linoleic acid residues weresignificantly reduced and arachidonic acid content was in-creased in plasma phospholipids and esterified cholesterol inTs-treated rats. However, the relative fatty acid compositionof unesterified fatty acids and triglyceride fatty acids was notaltered by TS reatment. PTU treatment had no effect on fattyacid distribution in any ofhe plasma lipids.Secretion of biliarylipids was increased in perfused livers from Ts-treated rats,while treatment with PTU did not affect release of lipids inthe bi1e.M These observations suggest a regulatory role forthyroid hormones that determine concentration and compo-sition ofplasma HDL andother lipoproteins.-Wilcox,H. G.,W.G. Keyes, T.A. Hale, R. Frank, D. W. Morgan,and M. Heimberg. Effects of riiodothyronine and propylthio-uracil on plasma lipoproteins in male rats. J. Lipid Res. 1982.23: 1159-1 166.

Thyroid hormones appear tobe important regulatorsof lipid metabolism. Hypercholesterolemiaor hypocho-lesterolemia is usually associated with hypothyroidismor hyperthyroidism, respectively (1). Considerable vari-ability has been reported in plasma concentrations oftriglyceride, which may be increased, decreased, or nor-mal in hyperthyroidism (2 , 3). In hypothyroidism, theplasma triglyceride concentration is usually increasedbut may be normal (3 , 4). We reported previously ( 5 ,6) that perfused livers from rats treated with T3 ester-ified less and oxidized more free fatty acid than didlivers from euthyroid animals. Treatment with propyl-thiouracil (P TU) sufficient to lower plasmaT3 levels to60% of normal, and to induce a mild hypothyroidism,had the opposite effect. One consequence of these di-vergent pathways ofhepatic fatty acid metabolisms thelower secretion rate of VLDL by perfused livers fromhyperthyroid rats, and the reverse in hypothyroidism( 5 , 6).

The influence of thyroid hormones on HDL metab-olism has no t previously been studied. It has been ob-served, however, that HDL cholesterol is moderatelydepressed along with other classes of plasma lipopro-teins n hyperthyroidism (7). In hypothyroidism, theconcentration of HDL cholesterol is variable, but usu-ally normal or slightly increased (7 , 8). Dory and Roh-eim (9)observed elevated plasma levels ofDL and LDLin the rat made hypothyroid by feeding PTU for 30days; decreased removal of these apoE-rich lipoproteinfractions from the circulation may have, in part, ac-

Abbreviations: Ts, triiodothyronine; PTU, propylthiouracil; VLDL,very lo w density lipoprotein; IDL, intermediate density lipoprotein;LDL, lo w density lipoprotein; HDL, high density lipoprotein; T G ,triglyceride; PL, phospholipid; C, cholesterol; CE, cholesteryl ester;SDS, sodium dodecyl sulfate; FFA, free fatty acid; RID, radial im-munodiffusion; LCAT. 1ecithin:cholesterol acvltransferase: PAGE.polyacrylamide gel electrophoresis. ~~ I ~.Supplementary key words thyroid hormones apolipoproteins A-I,B, C, and E biliary lipids fatty acids phospholipids cholesterol Medicine, Amarillo, T X 79109.' Present address: Department of Pediatrics, Texas Tech School of

triglycerides D. W. Morgan.

Journal of Lipid Research Volume 23, 1982 1159


2/8

counted for their observations. A modest elevation ofapoA-I was also observed (9). Experimental hyperthy-roidism was not studied by these investigators.

Our interest in evaluation of effects of hyroid statuson lipid metabolism, through hepatic mechanisms andestablishment of a reliable model for such studies, ledus to examine the concentrations and composition ofplasma lipids, lipoproteins, and apolipoproteins in ratsmade hyperthyroid with T3or hypothyroid with PTU.This was of particular interest since we had previouslyobserved important differences in hepatic free fatty acidmetabolism and VLDL formation in livers from suchanimals. Additionally, we wished to evaluate effects ofthyroid status on hepatic output of HDL and biliarylipids. A preliminary report of this work has been pre-sented (10).

METHODSMale Sprague-Dawley rats (Charles River BreedingLaboratories, Wilmington, MA), having initial weights

of 150-1 75 g, were housed in individual wire-bottomcages. Lighting was maintained between 0430 and 1630hr. Animals were made hyperthyroid by subcutaneousinjection of T3 10 pg/lOO g body wt per day) or madehypothyroid by subcutaneous injection of PTU (1 mg/100 g body wt per day). Injections were carried outbetween 1530 and 1630 hr. In another group, animalswere treated similarly, but injections were carried outat 1000-1 100 hr. Th e drugs were dissolved in 0.9%NaCl with 1 N NaOH and the pH was adjusted to 8 . 5 .Control (euthyroid) rats were injected with an equiva-lent volume of the vehicle at the same pH. All animalswere treated with drugs for 7 days and allowed freeaccess to food (powdered Purina Lab Chow) and water.Daily food consumption was measured and was equalamong all groups for the duration of treatment (5 , 6).Th e reason for selection of reatment periods and drugdosage was documented in previous publications fromour laboratory (5, 6). These treatments produced mildhyperthyroid and hypothyroid states in the rats and didnot affect food consumption. Under those experimentalconditions (5 , 6) , identical to those used in the presentstudy, the plasma concentrations of T3were 47% higherfollowing administration of T3, and were 60% lowerthan the euthyroid rats, treated with PTU , at the timeblood samples were taken for analysis (6); plasma thy-roxine (T4) oncentrations were 90% ower withT3 nd75% lower after treatment with PTU, than in the eu-thyroid animal.

At the end of the treatment period, rats were anes-thetized with diethyl ether between 0830 and 1030 hr,

17-24 hr after the last injection of drug. Blood wasobtained from the abdominal aorta using EDTA ( 1 mg/ml blood) as the anticoagulant. There were no discern-ible differences in the lipid and lipoprotein parametersmeasured between the morning and the afternoon in-jection groups, so they were treated as onegroup.Plasma was collected by low speed centrifugation to sed-iment the formed elements of the blood. Isolation oflipoproteins was started immediately after collection ofblood from individual randomly selected samples.

Perfusions of the isolated rat liver were carried outas reported previously from this laboratory (5, 6). Theperfusion medium consisted of Krebs-Henseleit bicar-bonate buffer (pH 7.4), 30% washedbovine erythro-cytes, glucose (100 mg/dl), and 3% delipidated bovineserum albumin. An oleic acid albumin complex was in-fused at a rate of 166 pmol of oleic acid per hour (6 gof albumin and 14 19pmol of oleicacid/dl of infusate).

VLDL, I D L + LDL, and HDL were isolated fromplasma by ultracentrifugation after sequential densityadjustments (11 ) with solid NaBr (12). VLDL was col-lected at a density of 1.006 g/ml after 18 hr centrifu-gation in a Type 40 rotor at 10C and 39,000 rpm ina Beckman-Spinco ultracentrifuge. Th e isolated VLDLwas recentrifuged as before to free he VLDL fromcontamination with higher density proteins (principallyalbumin). IDL + LDL was collected in the density range1.006- 1.063 g/ml following an 18-hr centrifugationand was not recentrifuged. HDL (d 1.063-1.21 g/ml)was isolated andrecentrifugedfor 24 hr under thesame conditions. The lipoproteins were dialyzed at4C overnight against 100 volumes of 0.00 1 M EDTA-0.02% NaN3.

An aliquot of plasma,bile, or isolated lipoprotein wasextracted with chloroform-methanol2: 1 (v/v) (13). In-dividual lipid classes were separated by thin-layer chro-matography ( 1 4) . Analysis of triglyceride ( 1 5), choles-terol (16),and phospholipid ( 1 7) was carried outon theappropriate silicic acid band. Lipoprotein-protein wasmeasured by the method of Lowryet al. ( 1 8)as modifiedby Markwell et al. (19), a method which uses sodiumdodecyl sulfate as the delipidating agent for clearingturbidity. Fatty acyl group analysis was carried out asdescribed previously (14, 20). A Packard-Becker 42 1Gas Chromatograph interfaced to a Perkin-Elmer Sigma10 integrator/computer was used for these analyses.Lipoproteins were delipidated for electrophoreticanalysis with chloroform-methanol 2: 1 (v/v) (2 1). Iso-electric focusing of the apolipoproteins was carried outin 16.0 X 0 .7 cm tubes using 50 pg HDL protein and50-1 00 pg of VLDL or IDL + LDL protein essentiallyas described by Gidez, Swaney, and Murnane (22). Am-pholines (LKB pH 4-6 and 3.5-10) were combined in

1160 Journal of Lipid Research Volume 2 3 , 1982


3/8

the ratio of 5:3 to provide a wide pH range, but withan expanded pH 4-6 region. Gels prepared with bothammonium persulfate and riboflavin as the catalyst forpolymerization were consistently more uniform. Elec-trophoresis was carried out for 16 hr at 400 volts and5C. The extruded gels were stained at 65C for 45-60 min in a solution of 0.15% Coomassie Blue R250,30% methanol, 12.5% trichloracetic acid, and 4% sul-fosalicylic acid. Destaining was carried out by diffusionin 10% ethyl acetate, 7% ethanol , and 5% acetic acidfor 48-72 hr and gels were stored in 5% methanol-7%acetic acid (23). Gels were scanned at 550 nm using aGilford 2520 gel scanner; peak areas were estimated bytriangulation and later by use of a Sigma 10 integrator(Perkin-Elmer). The areas of the individual poly-morphic forms of apoA-I, apoE (also containing apoA-IV), and apoC were combined andare eported asgroups of isoforms.

Plasma apoA-I levels were estimated by the techniqueof radial immunodiffusion (RID). Rat apoA-I, used asthe antigen standard and for antibody production, wasobtained from lyophilized rat plasma HDL (d 1.063-1.21 g/ml) delipidated with ethanol-ether 3:2 (v/v)(24). The delipidated HDL was dissolved in 0.2 M Tris-6 M urea buffer (pH 8.0) and passed through two Se-phacryl S 200 (Pharmacia) columns in tandem (2.6x 100 cm) at a rate of 10 ml/hr, using the same buffer.Recently, we obtained even more satisfactory separationusing Sephacryl 200 and 300 columns in tandem. T heeluted apoA-1 peak was dialyzed against 4 M urea andfurther purified using preparative flatbed isoelectric fo-cusing (LKB Multiphor 2100) on Sephadex-IEF in thepH range4-6 (Ampholines LKB). The final preparationof apoA-1 was judged to be pureby analytic isoelectricfocusing (Fig. 1) and SDS-PAGE (not shown).

Antiserum to rat apoA-I was produced in New Zea-landwhitemale rabbits by intradermal injections atmultiple sites on the back. One mg of apoA-I in 0 .9%NaCl mixed with complete Freunds adjuvant 1 1 (v/v)was used for injection. A booster injection was given 2weeks later and the ntisera was harvested after another2 weeks. Th e antiserum was monovalent as determinedby immunodiffusion (Ouchterlony); a ingle line of iden-t i ty between whole rat serum, poA-I, and rat HDLwasobserved. RID plates were prepared with 1% agarose-5% T-10 dextran25) in 0.05 M barbital buffer, pH 8.6(0.02% NaN3-0.001 M EDTA). Five-microliter samplesof plasma diluted 1 10 with 0.1 M sodium decyl sulfatewere placed in 4 mm wide X 2 mm deep wells in geladhering to Gelbond Film (Marine Colloids, Rockland,ME). Plates (8.5 X 9.5 cm) contained 25 evenly spacedwells. Linearity of apoA-I concentration was obtainedby plotting the square of the precipitin ring diameteragainst standard apoA-I concentrations (50 to 350 ng)in 0.5% bovine serum albumin.

ApoB3 and additional estimates of apoA-I4 were de-termined in plasma and perfusate samples by radioim-munoassay procedures developed in our laboratory us-ing Triton X-100 to dissociate the apoprotein from thenative lipoproteins.

Statistical significance was calculated using the un-paired Students t test. The value 2P < 0.05 was ac-cepted as an indication of statistical significance.

RESULTSThe concentrations of plasma HDL (d 1.063-1.2 10

g/ml) protein, phospholipid, and cholesterol were in-creased by treatment with T3, while treatment withPTU did not affect the plasma concentration of thesecomponents (Table1).Hyperthyroid animals had lowerconcentrations of protein and cholesterol in the VLDL(d < 1.006 g/ml) than did euthyroid control animals.The levelsofVLDL phospholipid and triglyceridetended to be lower in hyperthyroid rats, but did notquite reach (2P< 0.07) the assigned value for statisticalsignificance. The concentrations of cholesterol and tri-glyceride in the VLDL were lower in PTU-treated an-imals than in controls. Although the levels of proteinand phospholipid in VLDL tended to be lower in thePTU group, they did not reach statistical significance.Th e concentrations of protein, phospholipid, and cho-lesterol in the plasma IDL + LDL fraction (d 1.006-1.063 g/ml) of hyperthyroid animals did not differ fromcontrols. Phospholipid and cholesterol of the d 1.006-1.063 g/ml lipoprotein fraction in PTU-treated animalswere higher than in euthyroid animals, while proteintended to be higher but did not reach statistical signif-icance.

The plasma concentration of apoA-I also increasedin T3-treated animals in comparison to euthyroid con-trols (Table2), complementing the increase in levels ofHDL protein and lipids. Trea tment with PTU underou r experimental conditions, however, did not alter theplasma apoA-I level significantly.The apoB content wasdiminished in plasma of hyperthyroid animals and un-changed in rats treated with PTU in comparison to eu-thyroid controls. These differences must reflect thechanges in the VLDL and LDL lipidsand protein,sincethese lipoprotein classes constitute the major carriers ofapoB.

The isoelectric focusing patterns of the HDL apo-proteins are shown in Fig. 1. Estimation of the relativedistribution of the grouped apoC, apoE, and apoA-Iisoforms is presented in Table 3 for all plasma lipopro-

ale, T., H . G . Wilcox, and M . Heimb erg. Unpublished results.Frank, R . , H . G. Wilcox, T. Hale, andM . Heimberg. U npublishedresults.Wilcox et al . Lipoproteins in experimental hypo-andhyperthyroid rats 1161


4/8

TABLE 1. Effectsof treatment with triiodothyronine or propylthiouracil on the composition of rat plasma lipoproteinsLipoproteinlass Thyroid Status Proteinhospholipidholesterol Triglyceride

mgldl pmolldlVLDLutliyroid 6.8 f 0.91) 8.8 f 1.4 10) 11.1 f 2.2 (lop+ 35.3f 5.1 lo)4.3 f 0.5 (11) 5.5 f 0.99).1 f 0.8 lo) 3.7 f 3.1 (10)4.8 f 0.71).5 f 1.21).2 f 0.81.3 f 3.910)d < 1.006 TsPTU

Euthyroid 15.8 f 1.6 1 ) 3.6 f .1 (11) 2.1 f .0 (1)PTU 46.9 f 6.8 (10)19.0 f 1.610)8.1.79)8.0 f 3.7lo)

HDLuthyroid 60.3f 3.51)8.6 f 2.810)4.4 f 6.9 (1 157.2 f 4.0 1 )d 0.4 f 2.6 8.2 f 9.1 1 )d

IDL + LDLd 1.006-1.063 Ts 15.3f 2.11)7.0 f 1.910)

d 1.063-1.210 T3 87.0f 2.81 IFd 68.0f 5.61 lpdPTU 100.3 f 6.1 (lopdAll data are meansf SEM. Figures in parentheses indicate the number of observations.

* Represents the sum of separate analyses for free and esterified cholesterol.r.d Values within a particular lipoprotein class with the same superscript are statisticallydifferent (2P< 0.05) rom one another.Euthyroid, untreated animals; Ts, Ts-treated (hyperthyroid) animals; PTU, PTU-treated (hypothyroid)animals.

tein classes. The percentage distribution of apoproteinsin HDL did not differ between euthyroid controls andTs-treated animals. However, there was a lower per-centage of apoA-I and a higher percentageof apoE inHDL obtained from plasma of PTU-treated animalscompared to control or T3-treated animals.An n-creased percentage of apoE in the IDL+ LDL fractionin the PTU-treated rat was also suggested. The per-centage of apoA-I inhe IDL+LDL was relatively smalland did not appear to be different among the groups.ApoE appears to make up a higher proportion of theapoprotein of the VLDL in euthyroid controls than ineither PTU- or Ts-treated animals.

T o characterize further the lipid abnormalities in-duced by treatment with T S or PTU, w e examined thefatty acyl group composition of plasma esterified andunesterified fatty acids (Table 4). The relative distri-

TABLE 2. Effects of treatment with triiodothyronine orpropylthiouracil on concentrationsof ratplasma lipoprotein apoproteinsApolipoprotein

Thyroid Status ApA-1 ApoBmgldl

A. Euthyroid 41.4 f 2.0 24) 9.7 f 1.8 17)6T sPTU 58.7 f 2.2 (31). 11.5 f 0.6 (19)b.35.8 f 1.9 (21) 24.5 f 2.0 (18)

B. Euthyroid 49.7f 2.6 (a)* 30.1 f 2.7 (8)PTU 47.2f 4.2 (8) 28.2 f 2.8 (8)T S 87.3 f 7.68)4.3 f 1.1

Data presented in panel A were determined by radial immuno-diffusion. All data are means f SEM. Figures in parentheses indicatethe number of observations. Data presented in panel B represent anadditional eight animals for which both plasma apoA-Iand apoB wereestimated by radioimmunoassay.*. Values within a specific apoprotein class having the same super-script are statistically different (2P< 0.05) rom one another.

1162 Journal of Lipid Research Volume 23, 1982

bution of fatty acid residues in plasma triglyceride wasnot altered by the thyroid status of the animal. How-ever, therewas a significantly higher proportionof 18:in the plasma FFA in yperthyroid rats than in controls.Compensatory decreases occurred in the otheratty acidpercentages, but were of insufficient magnitude to bestatistically significant. The relative percent of linoleicacid was reduced and that of arachidonic acid was in-creased in cholesteryl esters and phospholipids n plasmaofTs-treated animals. PTU treatment and theesultantmild hypothyroid state didnot seem to alter theelative

2 3

I] Apo AI

Group

Apo C I1Apo C 1110ADOC 1111.2APo C1113

Fig.1. Effects of reatment with triiodothyronineor propylthiouracilon isoelectric focusing atterns of rat plasma lipoprotein apoproteins.Tube 1 depicts plasma HDL apoproteins from a PTU-treated rat.T u b e 2 shows HDL apoproteins from an untreated euthyroid rat.Tube 3 presents HDL apoproteins from a Ts-treated rat. Tube 4depicts purified apoA-I from ra t plasma HDL, and tube 5 shows thepurified apoE from plasma VLDL.


5/8

TA B LE 3. Effects of trea tmen t with triiodothyronine or propylthiouracilon the apop rotein composition of plasma lipoproteins"Lipoprotein Class Thy roid Status ApoA-1 A ~ E ~ A w C '

% of total58 f 443 f 5P T U (2) 40, 48

1D L + LDL Euthyroid (2) 10, 18 46, 52d 1.006-1.063 g/ml T s (2) 18 , 19 45, 59P T U (2) 14, 18 58,0HDLuthyroid (8) 72 f d 19 f 2dd 1.063-1.210 g/ml T s (8) 77 f 2' 15 f 2'

VLDLuthyroid (3)d < 1.006 g/ml T s (5)

P T U (9) 65 f d.' 26 f d,'

42 f57 f 553, 5936,4023, 3723, 279 f 18 f 1

1 0 f 1~ ~~~~~

" All data are means f SEM. Figures in parentheses indicate the number of observations.' ApoE also reflects the presence of apoA-IV which focuses in th e same region as the apoE

ApoC refers to the sum of the percentages of all the polymorphic forms of apoC-Ill andd. r Values within a specific apoprotein class having the same superscript are statistically dif-

Percentages were obtained from scans of isoelectric focusing gels.isoforms.apoC-11 as show n in Fig. 1.ferent from one another (2P c 0.05).

distribution of fatty acyl residues in plasmaphospholipidand cholesteryl esters in comparison to that of the eu-thyroid animal.

The net output of apoA-I by the perfused liver wasstimulated by hyperthyroidism and diminished by hy-pothyroidism. Treatment withT3 ncreased the net out-put of apoA-I (25.4* 1.7 vs. 45.4 f .6 pg/g liver per4 hr for the euthyroid and hyperthyroid animals, re-spectively). Treatment with PTU reduced the net out-put of apoA-I to 9.0 f .3 pg / g per 4 hr. In all cases,near linearity of apoA-I output over the 4-hreriod wasobserved. ApoA-I accumulating during thesecond 2-hrperiod was about 90% of that of the first 2-hr period

of perfusion. Furthermore, all three means were sig-nificantly different from each other (2P< 0.05). ApoA-I secretion by the liver probably is a good indicator forhepatic output of nascent HDL; this apoprotein is nota protein constituent of hepatic VLDL, the predomi-nant lipoprotein secreted by the liver. Unfortunately,the condition of the perfusions did not allow us to ex-amine the density fraction of the perfusion medium thatcould be properly identified as HDL. This resulted pri-marily from the use of bovine serum albumin contam-inated with bovine apoA-I in the perfusion medium (6).The d 1.063-1.210 g/ml fraction isolated from theperfusion medium, although containing protein and

TA B LE 4. Effect of treatment with triiodothyronine or propylthiouracil on fatty acyl gro up composition of plasma lipids"Fatty Acyl Group

Thyroid Status PlasmaLipids 16:O 18:O 18:1 18:Z 20:4'Z of total

Euthyroid (10) FF A 31.8 f 2.4 10.7 f 0.4 11.9 f 1.5b 10.0 f 1.0 10.0 f 1.032.4 f 0.9 3.4 f 0.2 38.2 f 2.2 17.0 f 1.8 1.1 f 0.229.4 f 2.3 19.8 f 0.7 6.1 f 0.2 26.9 f 0.4' 9.8 f 0.6'11.3 f 0.31% 11.9 f 1.3 32.8 f 0.6d 33.7 f O.gd

T GPLCETs-treated (10) FF A 29.4 f 1.7 9.1 f 0.6 17.9 f 1.5',' 12.9 f 0.9 10.3 f 1.4

13.7 f 0.5*.'8.2 f 0.4


6/8

TABLE 5. Effects of treatm ent with triiodo thyro nine or propylthiouracil on biliarylipid secretion by perfused liversThyroid Status Cholesterolhospholipid

pm ol l mi bile pmollgper 4 hr pmol I ml bile pmol lg per 4 hrEuthyroid (6) 0.090 f .020 0.019 f .004" 0.360 f 0.070 0.077 f .01 1Ts-treated (9) 0.140 f .023 0.049 f .010".* 0.425 f 0.058 0.118 f .016PTU-treated (9) 0.093 f .010 0.013 f .002* 0.256 f 0.042 0.036 f .009

Livers were perfus ed as described in th e te xt. Data are means f SEM. Figures in parentheses are the numberof observations.Values in the same column with the same superscript differ statistically from one another (2P < 0.05).

rich in lipid, did not contain apoA-I as estimated byRIA. All the apoA-I found in whole perfusates couldbe accounted for in the d > 1.2 10 g/ml infranatant so-lution. It is probable that the ack ofapoA-I in the HDLdensity fraction resulted from the displacement of ratapoA-I from the lipoprotein by the more plentiful andimmunochemically distinct bovine apoA-I. That thiscould occur was demonstrated by examining rat HDLreisolated by ultracentrifugation after incubation withthe bovine serum albumin. Th e reisolated rat HDL nolonger reacted with antisera to rat apoA-I by radioim-munoassay and the eactive apoprotein was recoverableentirely in the d > 1.210 g/ml fraction.

We reported previously (6) that the volume of bilesecreted by the perfused liver was increased in hyper-thyroidism andreduced in hypothyroidism. Table 5presents data which also indicate that the secretion ofbiliary cholesterol by perfused livers from rats treatedwith TSwas increased; phospholipid secretion tendedalso to increase, but failed to reach levels of statisticalsignificance. Biliary secretion of cholest.erol and phos-pholipid were not influenced by the mild hypothyroid-ism induced by treatment with PTU under conditionsof these experiments.

DISCUSSIONThe decreased concentrations of plasma VLDL that

we and others observed in hyperthyroid animals andpatients may have been predicted, in part, by our pre-vious observations of the reduced net rate of VLDLsecretion by livers from Ts-treated rats (6); thede-creased levels of plasma VLDL may, in part , also haveresulted from an enhanced rate of clearance of plasmatriglyceride (26). Treatment with PTU under our ex-perimental conditions increased hepatic esterification offatty acids and secretion of VLDL and decreased fattyacid oxidation by the isolated perfused liver ( 5 , 6). Inthe present study, however, the plasma concentrationof VLDL triglyceride was lower in PTU-treated ratsthan in controls. Since the plasma FFA concentration


is diminished in hypothyroidism (3), this factor per sewould tend to reduce the hepatic output of VLDL invivo. The IDL, partially metabolized VLDL and chy-lomicrons, and LDL have been shown to be increasedin hypothyroidism (9), perhaps due to decreased pe-ripheral catabolism. A similar trend for DL + LDL wassuggested in the present study with less severely hypo-thyroid animals.

A most interesting observation to us among thosereported herewas the increase in HDL protein and ipidin the T3-treated rats, accompanied by the decrease inplasma VLDL.Studies on concentration of plasmaHDLin thyroid pathology are sparse. Concentrations of HDLcholesterol are generally lower inhyperthyroid patientsthan in euthyroid controls (7, 8), while administrationof upraphysiologicdosesofL-thyroxine to normalmale volunteers was found o be without significanteffect (7).

Interestingly, it was reported in 1957 that he ad-ministration of triiodothyroacetic acid to euthyroid pa-tients with coronary disease increased a-lipoprotein cho-lesterol and decreased &lipoprotein cholesterol (27). Noother data have been reported suggesting a hyperal-phalipoproteinemic effect of thyroid hormone in man.Treatment of rats with thyroxine (28) or with TS 29)was reported to have little effect on lipoprotein levelsbut reduced cholesterol content in all plasma lipopro-tein classes (30) . Treatment of rats for 7 days with L-thyroxine (30 pg/d per 100 g) was reported to ncreaseHDL phospholipid by 12% although HDL cholesterolwas not changed, while VLDL and LDL protein andlipids were reduced by 20-48% (31).

Many of the apolipoproteins function as regulatorsof various metabolic processes (32). We reported pre-viously that livers from hypothyroid rats secreted VLDLparticles containing higher percentages of apoE andlower percentages of apoC than did the nascent VLDLfrom livers ofhyperthyroid animals (6).Th e differenceswere not expressed in the plasma VLDLof hypothyroidrats (Table 3). The increased proportion of apoE inVLDL newly secreted by livers from PTU-treated an-imals (6 )may not be maintained while recirculating in


7/8

the plasma in vivo, during which time apoE exchangereactions involving IDL and HDL occur. Increases inthe proportion of apoE in VLDL and LDL of hypothy-roid patients (33) and in the triglyceride-rich lipopro-teins of PTU-fed hypothyroid rats have been reportedby other workers (9).

The principal apoprotein associatedwith HDL inthe rat is apoA-I. This apoprotein probably is derivedfrom the nascent HDL secreted by the liver (34) andfrom chylomicrons originating in the gastrointestinaltract (35) .

We have shown in this study that the net secretionof apoA-I is increased in livers from T3-treated animalscompared to euthyroid controls, which may provide apartial explanation for the higher concentration ofplasma apoA-I and HDL in vivo. Hepatic apoA-I bio-synthesis may be stimulated in the animals treated withT3 nd diminished in those treated with PTU . However,the increased net secretion of apoA-I observed with therecycling liver perfusion system might also result fromreduced hepatic uptake and metabolism of the secretednascent lipoprotein. In preliminary studies, we observedthat HDLand plasma apoA-I levels tended topeak after5-7 days of treatment with 10 pg T3/day per 100 gbody weight, but then tended to return slowly towardcontrol levelswith continued treatment for an addi-tional 7 days. An early and acute effect of T 3 (and per-haps of T4)administration may be an appreciable in-crease in plasma HDL concentration. With longer ex-posure to high levels of hyroid hormone, however, thisacute rise may be normalized and even diminished asa result of enhanced catabolism and/or diminished he-patic apoA-I otuput. Theseconsiderations are being in-vestigated in our laboratory presently. The depressednet output f apoA-I by perfused livers from rats treatedwith PTU was not expressed in a diminished plasmaconcentration of HDL in the animal. The explanationfor this observation remains to be determined.

Th e increased concentration of plasma HDL seen inT3-treated rats s of interest in relationship to the cur-rent theory that an important role of HDL, in concertwith LCAT, is in the transport of cholesterol from pe-ripheral tissues to the liver for subsequent excretion(36). Although HDL cholesterol has been suggested asthe preferred source of biliary cholesterol (37) and bileacid (38), this has not yet been unequivocally estab-lished. Increased biliary excretion of cholesterol by per-fused livers from T3-treated rats was observed in ourpresent studies, in support of observations of others(39-41). Not all patients with hyperthyroid disease haveenhanced biliary cholesterol excretion, however, andenhanced excretion has been reported in some hypo-thyroid patients (42) .

Th e influence of the hyperthyroid state on metabo-

lism of essential fatty acids in rats (43) and in patientshas been observed previously. We confirmed the majorfindings of these studies in the hyperthyroid rat (44 ,45) . One of the principal effects of hyroid hormone onessential fatty acid metabolism may be enhanced con-version of 18:2 to 20:4 (43). It is unlikely that dietary18:2 was limited in the hypothyroid animal studied, forneither plasma triglyceride nor FFA contained a lowerproportion of 18:2. The mild hypothyroidism resultingfrom PTU treatment of the rat was not accompaniedby the alterations in essential fatty acid metabolism de-scribed previously in patients (44) .

Th e data reported in this manuscript clearly dem-onstrate that alterations in thyroid status can changelipoprotein apoprotein concentration and/or distribu-tion. These alterations may be a consequence of or thereason for the altered lipid metabo1ism.lThis research was supported by grants HL22812, HL25800,and HL27850 from the National Institutes of Health, U.S.Public Health Service, and by the American Heart Associa-tion, Missouri Affiliate. TH wa s a Postdoctoral Fellow of theNational Institutes of Health, HL05919. DW M was a Post-doctoral Fellow of the American Heart Association, MissouriAffiliate. W e wish to thank Shirley Frank, Peggy Bledsoe, andJudy Stein for their excellent technical assistance.Manuscript received 25 August 1981, in raised f o n n 11 March 198 2, andin re-raised form 9 une 1982.

1

2.

3.

4.

5.

6 .

7.

8.

REFERENCESCornwell, D. G. , F.A. Kruger, G . J. Hamwi, and J. B.Brown. 1961.Studies on characterization of h u m a nserum lipoproteins separatedby ultracentrifugationin adensity gradient. I. Serum lipoproteins n normal, hyper-thyroid, and hypercholesterolemic subjects. Am . J . Clin.Nu tr . 9 24-40.Tulloch, B. R. , B. Lewis, and T. R. Fraser. 1973. Tri-glyceride metabolism in thyroid disease. Lancet. i: 391-394.Nikkila, E. A ,, and M . Kekki. 1972. Plasma triglyceridemetabolism in thyroid disease.J . Clin. Invest. 51: 2103-2 1 1 4 .Kut ty , K. M., D. G.Bryant, and N. R . Farid. 1978.Serumlipids in hypothyroidism-a re-evaluation. J. Clin. Endo-crinol. Metab. 46: 55-60.Keyes, W . G.,and M. Heimberg. 1979.Influence of thy-roid status on lipid metabolism in the perfused liver. J .Clin. Invest. 64:182-190.Keyes, W. G . , H. G. Wilcox, and M. Heimberg. 1981.Formation of the very low density lipoprotein and me-tabolism of [l-'4C]-oleate by perfused livers from ratstreated with triiodothyronineor propylthiouracil.Metab-olism. 3 0 135-146.Agdeppa, D., C . Macaron, T. Mallik, and N. D . Schnuda.1979. Plasma high density lipoprotein cholesterol in thy-roid disease. J . Clin. Endocrinol. Metnb. 49: 726-729.Scottolini, A. G., N . V . Bhagavan, T. H. Oshiro, andS. Y.Abe. 1980. Serum high density lipoproteincholes-

Wilcox et a l . Lipoproteins in experimental hypo-andhyperthyroidrats 1165


8/8

terol concentrations in hypo- and hyperthyroidism. Clin.Chem. 2 6 584-587.9. Dory, L., and P. S. Roheim. 1981 . Rat plasma lipoproteins

and apolipoproteins in experimental hypothyroidism. J .Lipid Res. 2 2 287-296.10. Wilcox, H. G., W. G. Keyes, T. W. Hale, D. W. Morgan,

and M. Heimberg. 1981 . Effectsof thyroid status onplasma high density lipoproteins (HDL). Arteriosclerosis. 1:93.

1 1 . Havel, R., H. A. Eder, and J. H. Bragdon. 1955. Thedistribution and chemical composition of ultracentrifu-gally separated lipoproteins in human serum. J . Clin. Zn-vest. 3 4 1345-1353.

12. Radding, C. M., and D. Steinberg. 1962. Studies on thesynthesis and secretion of serum lipoproteins by ra t liverslices. 1.Clin. Invest. 39: 1560-1569.13. Folch, J., M. ees, and G. H. Sloane Stanley. 1957. Asimple method for the isolation and purification of totallipids from animal tissues.]. Biol. Chem. 2 2 6 497-509.

14. Wilcox, H. G., G. D. Dunn,and M. Heimberg.1978.Effects of several long chain fatty acids on the propertiesand lipid composition of the very low density lipoproteinsecreted by the perfused liver. Biochim. Biophys. Acta. 2 9 839-54.

15. Newman, H . A. I. , C. T. Liu, and D. B. Zilversmit. 1961.Evidence for the physiological occurrence of lysolecithinin ra t plasma. J . Lipid Res. 2: 403-4 1 1 .16. Rudel, L. L., and M.D. Morris. 1973. Determination ofcholesterol using o-phthalaldehyde.J . Lipid Res. 14: 364-366.17. Rieser, F., A. N. Siakotos, and S. Fleischer. 1966. Quan-titative analysis of phospholipids by thin-layer chroma-tography and phosphorus analysisof spots. Lipids. 1: 85-

86.18. Lowry, 0 . H., N. J. Rosebrough, A. L. Farr, and R. J.Randall. 1951 . Protein measurement with the Fohn phe-nol reagent. J. Biol. Chem. 193: 265-275.19. Markwell, M. A. L., S. M. Haas, L. L. Bieber, and N. E.Tolbert. 1978. A modification of the Lowry procedure

to simplify protein determination in membrane and li-poprotein samples. Anal. Biochem. 87: 206-2 10.20. Morrison, W. R., and L. M. Smith. 1964. Preparations

of fatty acid methyl esters and dimethylacetals from lipidswith boron fluoride-methanol. J. Lipid Res. 5 600-608.21. Windmueller, H. G., P. J. Herbert, and R.J. Lenz. 1973.Biosynthesis of ymph and plasma lipoprotein apoproteins

by isolated perfused rat liver and intestine. J . Biol. Chem.22. Gidez, L. I . ,J. B. Swaney, and S . Murnane. 1977.Analysisof rat serum apolipoproteins by isoelectric focusing. I.

Studies on the middle molecular weight subunits.J. LipidRes. 18: 59-68.23. Vesterberg, O., L. Hanscn, and A. Sjosten. 1977. Stainingof proteins after isoelectric focusing in gels by new pro-

cedures. Biochim. Biophys. Acta. 491: 160-166.24. Scanu, A. 1966. Forms of human serum high density li-poprotein protein. J . Lipid Res. 7: 295-306.25. Curry, M. P., . Alaupovic, and C. A. Suenram. 1976.Determination of apolipoprotein Aand its constitutive A-I and A-I1 polypeptides by separate electroimmunoassays.Clin. Chem. 22: 315-322.26. Abrams, J. J. , S. M. Grundy, and H. Ginsberg. 1981.Metabolism of plasma triglycerides in hypothyroidism andhyperthyroidism in man. J. Lipid Res. 22: 307-322.27. Oliver, M. F., and F. S . Boyd. 1957. The influence of

1 4 215-223.

triiodothyroacetic acid on the circulating lipids and li-popfoteins in euthyroid men with coronary disease. Lan-cet. I: 124.28. PalkoviE,M., and N. Skottovk 1978. Effect of thyroidhormones on theconcentration of lipids and lipoproteinsin the blood. Physiol. Bohemoslov. 27: 266.29. Malkoven,M.,nd V. Manniven. 1981. Failure of thyroidhormones to maintain the normal lipoprotein pattern inrats after removal of the pituitary gland. Atherosclerosis.30. Engelken, S. F., and R. P. Eaton. 1980. Thyroid hor-

mone-induced dissociation between plasma triglycerideand cholesterol regulation in the rat. Endocrinology. 107:3 1 . Muller, K. R., and R. G. Cortese. 1978. The value ofnormolipemic rat as an experimental animal in hypocho-

lesterolemic drug research. Artery. 4 564-577.32. Schaefer, E. J., S. Eisenberg, and R. I. Levy. 1978. Li-

poprotein apoprotein metabolism.J . Lipid Res. 19 667-687.

33. Clifford, C., A. F. Salel, B. Shore, V. Shore, and D. T.Mason. 1975. Mechanisms of lipoprotein alterations inpatients with idiopathic hypothyroidism. Circulation. 51:38.

34. Hamilton, R. L., M. C. Williams, C. J. Fielding, andR. J. Havel. 1976. Discoidal bilayer structure of nascenthigh density lipoproteins from perfused rat liver.J . Clin.Invest. 58: 667-680.

35. Glickman, R. M., and P. H. R. Green. 1977.Th e intestineas a source of apolipoprotein A-I. Proc. Natl. Acad. Sc i .36. Glomset, J. A., and K. R. Norum. 1973. Th e metabolicrole of 1ecithin:cholesterol acyltransferase: perspectivesfrom pathology. Adv. Lipid Res. 11: 1-65.37. Schwartz, C. C., L. G. Halloran, Z. R. Vlahcevic, D. H.Gregory, and L. Swell. 1978. Preferential utilization of

free cholesterol from high density lipoproteins for biliarycholesterol secretion in man. Science. 200: 62-64.38. Halloran, L. G., C. C. Schwartz, Z. R. Vlahcevic, R. M.Nisman, and L. Swell. 1978. Evidence for high densityfree cholesterol as the primary precursor for bile-acid syn-thesis in man. Surgery. 84: 1-6.

39. Roseman, R. H., S. 0.Byers, and M. Friedman. 1952.The mechanism responsible for the altered blood choles-terol content in deranged thyroid states. J . Clin. Endocri-nol. Metab. 12: 1287-1299.

40. Van Zyl, A. 1957. Note on the effects of thyroidectomyand thyroid hormone administration on the concentrationof bile cholesterol and cholic acid.J . Endocrinol. 1 6 2 13-216.41. Thompson, J. C . , and H. M. Vars. 1953. Biliary excretionof cholic acid and cholesterol in hyper-, hypo-, and eu-thyroid rat. Proc. SOC.Exp. Biol. Med. 83: 246-248.

42. Abrams, J. J. , and S. M. Grundy. 1981. Cholesterol me-tabolism in hypothyroidism and hyperthyroidism in man.J . Lipid Res. 22: 323-338.43. Ellefson, R. D., and H. L. Mason. 1964. Effects of defi-ciency and excess of thyroid hormone on the istributionof fatty acids in lipids of the rat. Endocrinology. 7 5 179-186.44. Jurand, J., and M. . Oliver. 1970. Effect of thyroid ac-tivity on fatty acid composition of serum lipids. Athero-sclerosis. 1 : 125- 140.45. Kirkeby, K. 1972. Fatty acid composition of serum lipidsin hyper- and hypothyroidism. Acta Endocrinol.71: 62-72.

38: 121-128.

208-2 14.

USA. 74: 2569-2573.


journal 2 - t3 and ptu on rats

Documents