The thyroid gland is a major source of circulating T3 in the rat

The thyroid gland is a major source of

circulating T3 in the rat.

J P Chanoine, … , S Dubord, J L Leonard

J Clin Invest.

1993;

91(6)

:2709-2713.

https://doi.org/10.1172/JCI116510

.

In rats, the respective contribution of the thyroid and peripheral tissues to the pool of T3

remains unclear. Most, if not all, of the circulating T3 produced by extrathyroidal sources is

generated by deiodination of T4, catalyzed by the selenoenzyme, type I iodothyronine

5'-deiodinase (5'D-I). 5'D-I in the liver and kidney is almost completely lost in selenium

deficiency, resulting in a marked decrease in T4 deiodination and an increase in circulating

T4 levels. Surprisingly, circulating T3 levels are only marginally decreased by selenium

deficiency. In this study, we used selenium deficiency and thyroidectomy to determine the

relative contribution of thyroidal and extrathyroidal sources to the total body pool of T3.

Despite maintaining normal serum T4 concentrations in thyroidectomized rats by T4

replacement, serum T3 concentrations remained 55% lower than those seen in intact rats. In

intact rats, restricting selenium intake had no effect on circulating T3 concentrations.

Decreasing 5'D-I activity in the liver and kidney by > 90% by restricting selenium intake

resulted in a further 20% decrease in serum T3 concentrations in the thyroidectomized, T4

replaced rats, suggesting that peripheral T4 to T3 conversion in these tissues generates

approximately 20% of the circulating T3 concentrations. While dietary selenium restriction

markedly decreased intrahepatic selenium content (> 95%), intrathyroidal selenium content

decreased by only 27%. Further, thyroid 5'D-I activity actually increased […]

Research Article

The Thyroid

Gland

Is a

Major

Source of

Circulating T3

in

the Rat

Jean-PierreChanoine, Lewis E. Braverman,Alan P.Farwell, MarjorieSafran,SharonAlex,

Susan Dubord, and Jack L. Leonard

DepartmentsofNuclear Medicine, Physiology, and Medicine, University ofMassachusetts MedicalCenter, Worcester, Massachusetts 01655

Abstract

In rats, the respectivecontributionofthethyroidand periph-eral tissues to the poolofT3remainsunclear.Most,ifnot all,of

thecirculatingT3 producedbyextrathyroidalsourcesis gener-ated by 5'-deiodination of T4,catalyzed bytheselenoenzyme, type Iiodothyronine 5'-deiodinase (5'D-I). 5'D-I intheliver

and kidney is almost completely lost in selenium deficiency,

resulting in a marked decrease in T4 deiodination and an in-crease incirculatingT4levels. Surprisingly, circulatingT3 lev-els are only marginally decreased by selenium deficiency. In

thisstudy, we usedseleniumdeficiencyandthyroidectomyto determine therelativecontribution of thyroidal and extrathyroi-dal sources to thetotal bodypoolofT3.Despite maintaining

normal serum T4 concentrations inthyroidectomizedrats by

T4

replacement, serum T3 concentrations remained 55% lower than those seen inintactrats. Inintact rats, restricting selenium intakehad noeffectoncirculating T3 concentrations.

Decreas-ing5D-1I activityin the liver andkidneyby >90%by restrict-ingseleniumintakeresulted in afurther 20%decreaseinserum T3 concentrations in the thyroidectomized, T4 replaced rats,

suggestingthatperipheral T4toT3 conversionin thesetissues

generatesapproximately 20% ofthecirculating T3

concentra-tions. Whiledietaryseleniumrestriction markedlydecreased

intrahepatic selenium content (>95%), intrathyroidal sele-nium content decreased by only 27%. Further, thyroid 5D-1I activity actually increased 25% in theselenium deficientrats,

suggestingthecontinued synthesis of this selenoenzyme over

selenoproteins in other tissues in selenium

deficiency.

These

data demonstratethat thethyroidis themajorsourceof T3 in

the ratandsuggestthatintrathyroidal T4toT3 conversionmay account for most oftheT3 released by the

thyroid. (J.

Clin. Invest. 1993.91:2709-2713.) Key words: deiodination* sele-nium-thyroid* hormonemetabolism*T4toT3 conversion

Introduction

In mammals, the relative contribution of

thyroidal

and ex-trathyroidalsources tothe total body pool of themetabolically

Partof this work has been presentedas anabstractatthe 20thmeeting

of theEuropean Thyroid Association, Dublin, Ireland, 20-25 June 1992.Addressreprintrequests to Dr. Jack L.Leonard,Departmentof NuclearMedicine, University of Massachusetts Medical School, 55 LakeAvenueNorth, Worcester,MA01655.

Receivedfor publication 23 October 1992and in revisedform8 January 1993.

activeiodothyronine, 3,5,3'-triiodothyronine (T3), is unclear. T3canbe derivedfrom conversion of the prohormone

thyrox-ine (T4) by outer ring (5'-) deiodination in the peripheral

tis-sues, by T4toT3 conversion within thethyroidgland, andby

direct secretion of de novo synthesized thyroidal T3.Estimates

ofthe contribution ofextrathyroidal T4toT3conversiontothe

totalT3 poolvaryfrom 20%to 100%in the rat ( 1-3) and the

thyroidaccountsfor the remainder of the T3produceddaily.

Laurberg used in situ thyroid perfusion to directly examinethe

contribution of the thyroidtothe T3 pool and reportedthat intrathyroidal T4 to T3 conversion accounted for a consider-able portion ofthe T3 secretedfrom thedog thyroid (4, 5).

Thus,both extrathyroidal andintrathyroidalT4 to T3 conver-sion appear to participate in the daily production of T3.

T4 toT3 conversion is catalyzed by the enzyme, iodothyro-nine5'-deiodinase. Two isozymes ofiodothyronine 5'-deiodin-ase havebeenidentified. Themostabundantform, type I

io-dothyronine 5'-deiodinase (5D-I), is found in liver, kidney,

andthyroid (6) and contains the rareamino acid selenocys-teine(7-9). Tissue content of 5 D-I in the liver and kidney is proportional to selenium intake (10). The other isozyme, type II iodothyronine 5'deiodinase (5D-IT), is abundant in the brain, pituitary, and brown adipose tissue and does not contain selenium (11, 12). T3 generated by 5 DI is released into the general circulation, while the majority of theT3produced by

5D-1I remains withthecell.Theabilitytomanipulate 5'D-I levels by alteringthedietary intake of selenium providesthe means to examine thecontributionof S'D-I to total T3 produc-tion.Selenium deficiencyleads to an almostcompleteloss of 5D-I in the liver and kidney and a 40-50% increase in the serumT4concentration( 10, 12-14). This incrementin

circu-latingT4 iscompletely accounted for by the prolonged meta-bolic half-life oftheiodothyroninedue to the loss of 5'D-I ( 14).

Paradoxically, serum T3 concentrations are not reciprocally affected and decreasebyno morethan20%,if at all ( 12-14).

While serum T3 concentrations are marginally depressed by selenium deficiency, circulating T3 sulfate concentrations in-creasenearlytwofold(12, 14).SerumTSH levels remain near normaldespite the elevated circulatingT4 in the selenium-defi-cient animal ( 12-14). Thus, despite the marked decrease in

hepatic and renal T4 to T3conversionin the absence of

sele-nium, othersourcesofT3 appear tobe madeavailablein

ani-malslacking 5D-I.

There are several possibilities to account for the discor-dance between the nearcomplete loss of5D-I and the

mar-ginal fall in

circulating

T3 observed in selenium deficiency. They include(a) diminished T3 clearance,

(b)

increased

thy-roidal13 secretion, and/or (c)enhanced recovery of T3 from

sulfo-conjugatesreleased into the gut in theenterohepatic cy-cle.Previous work has shown thatT3clearance isonly margin-allydecreasedby selenium

deficiency

and that the 20-25% in-crease inthemetabolic half-life of this

iodothyronine

is

insuffi-J.Clin.Invest.

©TheAmerican Society for ClinicalInvestigation,Inc. 0021-9738/93/06/2709/05 $2.00

cient to maintain the steady-state levels of T3 observed in serum( 14). The contribution of the thyroidtotheT3 pool in theselenium-deficient rat remains to be determined. Likewise, the contribution of enterohepatic recycling of T3orits conju-gates to thecirculating T3 pool is unclear.

Inthis study, we determined the source(s) of circulating T3 in selenium-deficient rats. The data show that the thyroid gland serves asamajorsourceofcirculating T3 in theratand suggest thatintrathyroidal T4toT3 conversionaccountsfor much of theT3 secreted by the thyroid.

Methods

Animals andreagents.WeanlingmaleSprague-Dawleyrats(40-50 g) suppliedby Charles River Laboratories(Wilmington, MA)wereused in allexperiments. The studywasapproved by the Animal Research Committee andcomplieswith theinstitutionalassurancecertificateof theUniversityof Massachusetts Medical Center. Ratswerefedatorula yeastbasedsemisyntheticdiet(TekladPremierLaboratory Diets, Mad-ison, WI) for5wk. Theselenium-deficient diet (TD86298)contains lessthan 16,gg/kgselenium and theselenium-repletediet(ID 91259)

is the same base diet supplemented with 200 ug/kg selenium as Na2SeO3. Rats were housed in stainless steel cages, and distilled water wasavailable ad lib. Body weight(BW)'was monitored biweekly.

Analytical proceduresand hormone assays. In allexperiments, ani-mals were killed by decapitation and exsanguinated, except where noted.Liverwashomogenized in4vol(wt/vol) of 20 mMpotassium

phosphatebuffer (pH 7.4), 150 mMNaCl,and in 4vol(wt/vol) of 250 mM sucrose, 20mMHepesbuffer (pH 7.0),1mMEDTA, and 1mM DTTandstored at -20°C for determination ofglutathione peroxidase

activity (GPx) and5D-I activity, respectively. Thyroidglands were weighed and homogenized in 800Mlof 250 mM sucrose, 20mMHepes

buffer(pH7.0), 1 mMEDTA, and 1 mM DTT for determination of

5'D-Iactivity.

The degreeof selenium deficiency in therats wasdetermined by the decrease inhepaticGPxactivity.GPxactivitywasdeterminedfrom the oxidation of NADPH in the presence of 0.35 mMt-butyl

hydroperox-ide monitoredspectrophotometricallyat340nm(15).Sampleswere runinduplicateand resultswereexpressedasnmol NADPHoxidized/

min per mgprotein. Hepatic GPx activities in intact, selenium-replete andthyroidectomized,T4replaced,selenium-repleteratswere598±58 nmol NADPHoxidized/ min per mg protein (n=9) and 906±53 nmol NADPHoxidized/min per mg protein (n = 12),respectively.

TypeIiodothyronine 5'-deiodinase activitywasdetermined by the releaseof radioiodide from 10MM['25I]rT3 in the presence of 20 mM DTT (5'D-I) (16). Samples wererun induplicate and results were expressedasunits/mgprotein; 1 unit of5'D-Ienzymeactivity repre-sentsthe releaseof 1 pmolradioiodine/minper mgproteinat37°C. Hepatic type I iodothyronine 5'-deiodinase activities in intact, sele-nium-replete andthyroidectomized,T4 replaced,selenium-replete rats were 224±15 U/mg protein (n= 12)and 114±8 U/mg protein (n =9),respectively.

Serum TSH was measured in duplicate byRIA using materials obtained fromtheNational Pituitary Agency, National Institutes of Health(Bethesda, MD). Serum T4 and T3 concentrations were deter-mined in duplicate by species-adapted specific RIAs.

Selenium was quantified by measuring the 162 KeV gamma ray producedduring the decay of radioactive"mSeafter irradiation of the sampleat aneutronflux. The sensitivity was 0.05 ppm (Research Reac-torFacility,University of Missouri-Columbia, Columbia, MO) (17).

Protein was measured by the method of Bradford (18).

1.Abbreviations used in this paper: BW, body weight; GPx, glutathione

peroxidaseactivity.

Experimental procedures

Effectofaltered selenium intakeon serumT4, T3, and TSH concentra-tions in T4 replaced,thyroidectomizedrats.35rats(22 selenium sup-plemented and 13 selenium deficient)wereused inthisexperiment.3 wkbefore being killed, 12 selenium-supplemented and 6 selenium-de-ficientratswereanesthetizedusing7 mg ketamine and 0.6 mg

xyla-zine/ 100 gBWand thethyroid glands removed. T4 replacementwas begunoneday afterthyroidectomybydailyintraperitoneal injections

of T4 ( 1.1 g/ 100 g BW) for 2 wk. For the final7dof theexperiment,

serumT4 concentrationsweremaintained ateuthyroidlevelsbythe useofanosmoticminipump(Alzet2001;AlzaCorp.,PaloAlto,CA)

calculatedtodeliver T4at a rateof 1.1 ug/ 100 g BW perday.T4(Sigma

Chemical Co., St. Louis, MO) used for hormone replacementwas

>99.7% pureasmeasuredby HPLC.

Effect of seleniumdeficiencyonthe selenium contentand 5D-I

activityin thethyroidgland. 15rats(8seleniumsupplemented,and 7 seleniumdeficient)wereanesthetizedasdescribed above andperfused

through theaortawith 40 ml of ice-cold saline. Theexsanguinated

thyroid glandandliverwereremoved and frozenat-70'C for subse-quent determination of seleniumcontentbyneutronactivation

analy-sis.In aparallel experiment,thyroidaland liver 5'D-I activitieswere

determined in 10selenium-supplementedand 10 selenium-deficient rats.

Effectsof seleniumdeficiencyonintrathyroidalmetabolism of 13 1.

Intrathyroidalmetabolism of13'Iwasdetermined in groups of 5rats(5

seleniumsupplementedand 5 seleniumdeficient).Animalswerekilled 2 hafter theintraperitonealinjectionof 10MCNa31I( 16.2

Ci/,Mg).

Thyroidglandswerethen dissected free of connective tissue and

ho-mogenizedin 500,lof sodium barbital buffer (pH8.6) containing20 mMmethimazole and4mM KI.The percentuptakeof radioiodide andthedistribution of the 1311 betweenMIT, DIT, T3, and T4was determined bydescendingpaperchromatography after pronase

diges-tion ofthyroidal homogenates accordingtoVagenakisetal.( 19).

Statistics

The results are presented as mean±SE. Statistical significance (P

<0.05)wasdeterminedusingthe Student'st testforunpairedvalues.

Results

Effects

of

seleniumstatus on serum T4, T3, and TSH

concen-trations inT4

replaced, thyroidectomized

rats.

During

the5-wk

experimental

period, body weights increasedfrom 55 to 330 g intheeuthyroidrats(intact) (n= _{17) and from}55 to289gin

the

thyroidectomized

ratsreplaced with T4

(14

replaced) (n

= _{18), indicating}that hormonereplacement was almost

com-plete in the thyroidectomized animals. No differences in

growth were observed between theselenium-deficientand

sele-nium-supplemented rats. Selenium deficiency resulted in a 97% decrease in hepatic GPx activityand a parallel 93% de-crease in hepatic 5'D-I

activity

inboth the intact and

T4-re-placedratsindicatingthattheanimalswereselenium deficient.

As shown inFig. 1, inselenium-supplemented rats,serum T4

concentrations were similar in the intact and T4 replaced groups(A). Selenium deficiencyresulted in theexpected

in-creasein the serum T4concentrationsin both theintactand T4

replacedgroups(P<0.05,A vs.B).

Althoughthe serumT4concentrationswereidenticalin the

selenium-supplementedintact and T4 replaced rats, serum T3

concentrationswere notnormalizedand T3concentrations in

the T4 replaced rats remained 55% lower than those in the

Se+

100

g 75

s

E 50 2 0

co

25

0

200

E 1.5

~-E 1.0

0

o

0.5

0.0

400

D300 E

i-200.

E

An100

0

100

75

E 50

(%25

IJLA 0

100

SE

:E 15-5

* EI 1.0.

0

CO

~~~~~o

0.5

.400-* , 300

E

co

I--200

E 100

01

F

I

E °6 c

oX, 0w

E*._

Ea

CO

*

Intact.

0 T4 Replaced

Figure1. Effect of selenium and thyroidectomyon serum

concentra-tions of T4,T3,and TSH intherat. Ratswerefedadefined diet,

thyroidectomized, andreplacedwithT4asdescribed inExperimental procedures.Bloodwasobtainedatthe time of killing and analyzed forT4,T3, and TSH.Intact, nonthyroidectomizedrats;T4 replaced,

ratsthyroidectomizedandreplacedwithexogenousT4. *P<0.05.

rats,respectively) but decreased theserum T3 concentrations

by 20%inthe T4 replacedgroup(C and D) (0.55±0.03 nMvs.

0.44±0.02nM,P<0.05selenium-supplementedvs.

selenium-deficientrats, respectively).

Inboth selenium-supplemented and selenium-deficient thy-roidectomized rats, T4 replacement did not normalize the

serum TSHconcentrations, andtheyremained 2 to 3 times

higher than those observed in the intactrats(Fig. 1, E and F). SeleniumdeficiencydidnotaffectserumTSH concentrations

inthe intactrats.

Effect ofseleniumdeficiencyonseleniumcontentand S'D-I

activityin thethyroid.The failure ofT4 replacementto normal-ize serum T3 concentrations in the selenium-supplemented

rats, despite the availability of normal T4 levelstothe 5D-I containingtissues, suggested that thethyroid glandmay

con-tributeupto55% of theT3 foundinthe circulation.Similarly,

thethyroid appearedtocontributeasmuchas60% ofthe

circu-lating T3inthe selenium-deficient rats, animals that lack the selenoprotein 5SD-Iintheliver andkidney.Since earlier work

suggestedthatintrathyroidal T4toT3conversionwas amajor

contributortotheT3 foundinthe dog thyroideffluent(4, 5),

Se+

Se-2.0

o II

0 1 1.5. E _se,

1.0.

Cn

0.5

0.0 Uver

Se-Figure 2.Effectof selenium deficiency on the selenium content in the

thyroidandliver.Rats werefed aselenium-supplemented(Se+)or

selenium-deficient(Se-)diet for 5 wk then werekilled. Selenium content wasdetermined asdescribed inExperimental procedures. *P <0.05.

and thyroidal59D-I appears to be a selenoprotein (7), the near normal serum T3concentrations found in intact, selenium-de-ficient animals raisedthe possibilitythatthyroidal5D-I was unaffected by selenium deficiency. Thus, we determined the

effects ofselenium deficiencyon thyroidal seleniumcontent (Fig. 2) and on 5 D-Iactivityin thethyroid (Fig. 3). Rats fed

theselenium-deficient diet hada >97%fall in selenium con-tent in the liverand acorresponding>93%decreasein liver

5D-Iactivity. However,inthethyroid,theselenium-deficient dietresultedinonlyamodest27%decrease inselenium con-tent, andparadoxically, the5 D-Iactivitywasincreased by25% (P < 0.05).

Effects ofselenium deficiency

on

intrathyroidal

metabolism

of"'

I.To evaluate theinfluence of altered selenium intakeon

intrathyroidal iodine metabolism,wedetermined the effects of selenium deficiencyon thethyroid gland's abilityto concen-trateand organify iodine.Thyroidal uptake of

"'lI

was unaf-fectedbyseleniumintake, and therewere nodifferences inthe

synthesis of the

13'1-labeled

iodotyrosines (MIT andDIT) or

iodothyronines (T4 andT3) between selenium-deficient and

selenium-supplemented

rats(TableI).

Discussion

Controversy

surrounds thecontribution of the various tissues

toT3

production

intherat.DiStefano (1) estimated that47% ofT3originatesfrom boththyroidal secretionand

extrathyroi-dalT4toT3 conversioninliver and

kidney,

while the

remain-a..C

z

CD

CM i

LonD

Se+

SF-*S; S

a ~E

se+

Figure 3. Effect of seleniumdeficiencyonSD-Iactivityin thethyroid

andliver.Rats werefedaselenium-supplemented(Se+)or sele-nium-deficient(Se-)diet for5 wkthenwerekilled. 5D-Iactivity

wasassayed as described inExperimental procedures. *P<0.05.

TableI.

Effect

ofSelenium on Intrathyroidal

13"I

Metabolism

Se+

Se-%Uptake 9.4±1.3 12.7±3.2

% MIT 27.4±0.6 27.0±2.2

%DIT 41.4±1.6 36.9±1.5

MIT/DIT 0.74±0.08 0.67±0.03

%T3 1.0±0.3 1.1±0.2

% T4 4.1±0.5 3.1±0.3

T3/T4 0.33±0.06 0.25±0.06

Intrathyroidal metabolism of'1'Iwasdetermined in 5

selenium-sup-plemented(Se+) and 5selenium-deficient(Se-)rats asdescribedin Experimental procedures. Resultsareexpressedasmean±SE.

ing53% comes fromT4toT3conversion in theslowly equili-brating pools, suchas

brain, skin,

and muscle. Kinlawetal.

(3)

revised theirearlier estimatethatonly 20% ofT3is produced by extrathyroidal T4toT3 conversionintherat

(2)

and suggested that

essentially

allT3 is producedby

T4

to

T3

conversion. They alsospeculatedthatthereis littleor nodirect secretionof T3

by

thethyroid.

The current datademonstratethat at least 55%of circulat-ingT3 comesfrom the

thyroid

gland in the

intact,

selenium-supplementedrat.Since liver and

kidney

5 D-Istill contribute

tothepool ofcirculating T3in these rats, it is difficultto esti-matetherelative contribution ofthe

thyroid

tototal body T3 concentrations. Moreover,thefailure ofT4 replacementto

nor-malizeserumT3 concentrationsin

thyroidectomized

rats,

re-sulting in reduced liver 5'D-I

activity,

further

complicates

the

analysis

ofthe sourcesof

circulating

T3. However, inthe sele-nium-deficientrat,liverand

kidney

5'D-I are

virtually

absent

(12), andthecontribution of the thyroidtothe serumT3 pool

can be

approximated.

Thecontribution of

thyroidal T3

derived

from thyroglobulinand thatfrom thyroidalT4 toT3

conver-sionto the

circulating

T3pool in these

animals,

asestimated from thedata presented in

Fig.

1 panel D, is

approximately

65% andtheresidualserumT3 concentrationspresentin these

ratslacking boththethyroidandextrathyroid 5'D-I is in

excel-lent agreement with previously measured values (12). The sourceof the residual T3in the serumof thyroidectomizedrats

withessentially undetectableserumT4concentrationsand no

hepatic and renal 5'D-I remains unclear. Taken together,

ex-trathyroidal T4toT3conversioncontributes 20-25%,and

thy-roidalT4 to T3conversionandT3derived from thyroglobulin contributes 55-65% of the total body pool of T3. Thus, nearly

all the

circulating

T3derivedfromT4 appears to be produced by the selenoenzyme, 5'D-I.

Thyroidectomy resulted in a marked increase in serum TSHconcentrationsin bothselenium-supplemented and

sele-nium-deficientratsdespite the maintenance of serum T4

con-centrationsidenticaltothose observed in the respective intact groups. The presence ofincreased serumTSHconcentrations,

decreased serum T3 concentrations, and normal or elevated

serumT4concentrations indicatesthatcirculatingT3 plays an

importantroleinregulating TSH secretion ( 12, 20, 21 ).

Previ-ously,we founda modestincrease in serum TSH

concentra-tions ( 12)in

selenium-deficient

ratsthat was proportional to thefall in circulatingT3suggestingthat TSH levels may

actu-allybeslightly elevatedforsomeperiod duringdietary restric-tion of selenium.

Dietary restriction ofseleniumintakehad only amarginal effectontheseleniumcontentandlevels oftheselenoenzyme

5'D-I in thethyroid gland. Thus, whilethere was a near com-plete loss in hepatic selenium content and 5DI activity, the

thyroidpreserves itsseleniumstoresand5D-Iactivitywas in-creasedby25% inselenium-deficientrats.Thisincreasein

thy-roidal 5D-I may be due to TSH mediated stimulation of 5'D-I

activity(22, 23) resulting from small increases incirculating

TSH that may occurinselenium-deficient rats. Thus,despite

the loss ofextrathyroidalT4 toT3 conversion, the thyroid

re-tains the ability to catalyze the 5'deiodination of T4.

Similar preservation of thyroidal selenium stores was

re-cently observed in rats fed a selenium-deficient diet fortwo

generations. Behne and co-workers have observed that the thy-roid gland and brain exhibited the highest avidity for selenium (24).Arthur et al. (25) recently found thatthyroidalGPx

activ-ity

fell

by

50%after5 wkof selenium

deficiency.

Since the 25%increase in thyroidal 5D-I occurred duringthissame time period in our studies, these data suggest that the thyroid is resistant to the effects of selenium deficiency andthatthyroidal 5'D-I is preserved over otherselenoproteinsin selenium-defi-cientanimals. The thyroid is notuniqueinthisregardsince the testis and brain also preferentially preserveseleniumover other organs(26).

In this study, intrathyroidal iodine metabolism was also unaffectedby selenium deficiency, whereas inconsistent results have been reported on the effects of selenium deficiency on intrathyroidal metabolism by others. Goldstein et al. observed adecrease in PB

13"I

after in vitro incubation of thyroid glands fromselenium-deficient ratswith

131I

(27).Arthur et al. (25) found a decrease in bothT4andT3content in thethyroidgland

from selenium-deficientrats,while Meinholdetal.(28)found nochange inT4andT3content inthethyroidfrom rats fed a

selenium-deficientdiet. Takentogether,these data suggest that secretion of de novosynthesized T3isunaffectedbyselenium deficiency. It is generally assumed that 70-80% of the

circulat-ingT3concentrations (29, 30) isderived fromT4 toT3

conver-sion in human and

similar, albeit,

morevariable estimateshave been made for the rat and that theliver,kidney, and thyroid

contain nearlyall the5D-Iactivityinthebody(6, 30).Since

extrathyroidal5'D-Icontributes 10-25% oftheT3production

(see above), then more than 50% of the T3 in the thyroidal effluent islikely to be derived from intrathyroidal T4to T3

conversion ofT4liberated fromthyroglobulin.

Thefindingthat themajority ofT3inthe ratderives from thethyroidprovidesa potential explanation for theapparent

discordance between the observedKmof5 D-I andthe

concen-tration ofT4 available to extrathyroidal tissues. Invitro

esti-matesofthe KmforT4 for 5D-Irange between 0.5 and 1

_IM

(6,

30), while the T4available tothetissues ("free hormone") is

3-4 ordersofmagnitudeless,indicatingthatcatalysisby5'D-I inperipheral tissuesisveryinefficient. However,intracellular T4levels in thethyroidwould beexpected to be much greater than thosein thecirculation,andtheKmforT4of 5D-I may

reflectthesubstrateavailable inthethyroidrather than thatin thecirculation.

Inconclusion,the current studydemonstratesthat the

thy-roid gland isamajorsourceof circulatingT3inrats,

contri-bution of intrathyroidal T4toT3conversion toT3 homeostasis appears tobeimportant, but theexactcontribution remainsto

bedetermined.

Acknowledgments

Wethank J. S.Morris,V. L.Spate, and C. L. Reamsfrom the Univer-sity of Missouri ResearchReactor,UniversityofMissouri-Columbia,

Columbia, MO, for determination of tissue seleniumcontent. J.-P.Chanoine is the recipient ofaPublic Health Service Fogarty International ResearchFellowship (I F05 TW04373-01 ) and Aspirant at the Fonds National de la RechercheScientifique, Belgium. This workwas supported by grants DK-38772, DK-18919, andDK-02005 from National Institute of Arthritis, Diabetes, and Digestive and Kid-neyDiseases, National Institutes of Health, Bethesda, MD, and by a grant fromNorthAtlantic TreatyOrganization.

References

1.DiStefano, J. J., M. Jang, T. K. Malone, and M. Broutman. 1982. Compre-hensive kinetics of T3 production, distribution and metabolism in blood and tissue pools of theratusing optimized blood sampling protocols.Endocrinology.

110:198-213.

2.Schwartz, H. L., M. I. Surks, and J. H.Oppenheimer.1971.Quantitation of extrathyroidalconversionof L-thyroxine to3,5,3'-triiodo-L-thyroninein the rat. J.Gin.Invest.50:1124-1130.

3.Kinlaw,W.B., H. L.Schwartz, and J. H.Oppenheimer. 1985. Decreased

serumtriiodothyronineinstarvingratsis dueprimarilytodiminishedthyroidal

secretion ofthyroxine. J. Clin.Invest.75:1238-1285.

4.Laurberg, P. 1981. lodothyroninesecretion from perfuseddogthyroid

lobesafterprolongedthyrotropin treatment in vivo.Endocrinology. 109:1560-1565.

5.Laurberg,P.1978.Selective inhibition of the secretion oftriiodothyronine from theperfused caninethyroid by propylthiouracil.Endocrinology. 103:900-905.

6.Kohrle,J., R. DieterHesch,and J. L. Leonard.1991. Intracellularpathways ofiodothyroninemetabolism.In WernerandIngbar's TheThyroid.L.E. Braver-manand R. D.Utiger,editors. J. B.LippincottCo.,Philadelphia,PA.144-189.

7.Berry, M. J., L. Banu, and P. R. Larsen. 1991.TypeIiodothyronine deio-dinase isaseleniumcontainingenzyme. Nature(Lond.).349:438-440.

8.Arthur,J.R., F.Nicol,andG. J. Beckett. 1990.Hepatic iodothyronine5' deiodinase. The roleofselenium.Biochem.J. 272:537-540.

9. Behne, D.,A.Kyriakopolous, H.Meinhold,and J.K6hrle. 1990. Identifi-cation oftype I iodothyronine 5'-deiodinase as a selenoenzyme. Biochem.

Biophys. Res. Commun. 173:1143-1149.

10. Beckett, G. J., D.A.MacDougall,F.Nicol,and J. R. Arthur. 1989. Inhibi-tion oftypeIand type IIiodothyroninedeiodinaseactivityinratliver, kidneyand brainproducedby seleniumdeficiency. Biochem.J. 259:887-892.

11.Safran,M.,A.P.Farwell, and J. L. Leonard. 1991. Evidence that type II 5' deiodinase isnot aselenoprotein.J.Biol.Chem. 266:13477-13480.

12.Chanoine,J.P., M.Safran,A.P.Farwell,P.Tranter,D.Ekenbarger,S. Dubord, S.Alex, J. R. Arthur, G. J.Beckett,L.E.Braverman,andJ. L.Leonard. 1992.Seleniumdeficiencyand type II5'-deiodinaseregulationin theeuthyroid

andhypothyroid rat: evidence of a direct effect of thyroxine. Endocrinology. 131:479-484.

13. Beckett, G. J., S. E. Beddows, P. C. Morrice, F. Nicol, and J. R. Arthur. 1987. Inhibition ofhepatic deiodination ofthyroxineis causedby selenium

defi-ciencyinrats.Biochem. J 248:443-447.

14.Chanoine, J. P., M. Safran, A. P. Farwell, S. Dubord, S. Alex, S. Stone, J.R.Arthur, L. E. Braverman, and J. L. Leonard. 1992. Effects of selenium

deficiencyonthyroidhormone economy in therat.Endocrinology. 131:1787-1792.

15.Chada, S., C. Whitney, and P. Newburger. 1989. Post-transcriptional

regulationof glutathione peroxidase geneexpressionby selenium in the HL-60 humanmyeloid cell line. Blood. 74:2535-2541.

16.Leonard, J. L., and I. N. Rosenberg. 1980. Thyroxine 5'-deiodinase from

ratkidney: substrate specificity and the5'-deiodinationof reverse triiodothyro-nine.Endocrinology. 107:1376-1383.

17.Morris, J. S., M. J. Stampfer, and W. Willett. 1983. Dietary selenium in humans:toenails as anindicator. Biol. TraceElem. Res. 5:529-537.

18.Bradford, M. M. 1976. A rapid and sensitive method for the quantitation ofmicrogram quantitiesofproteinutilizingtheprincipleof protein-dyebinding.

Anal. Biochem. 72:248-254.

19.Vagenakis, A. G., S. H. Ingbar, and L. E. Braverman. 1974. The relation-ship between thyroglobulin synthesis and intrathyroidal iodine metabolism as

indicatedby the effects of cycloheximide in the rat.Endocrinology. 94:1669-1680.

20. Abend, S. L., S. L. Fang, S. Alex, L. E. Braverman, and J. L. Leonard. 1991.Rapidalteration incirculating freethyroxine modulates pituitary 5' deio-dinase and basalthyrotropin secretionin the rat. J.Clin.Invest.88:898-903.

21. Emerson,C. H., R. Lew, L. E. Braverman, and W. J. DeVito. 1989. Serum

thyrotropinconcentrations are morehighlycorrelatedwith serum triiodothyro-nineconcentrationsthanwith serumthyroxineconcentrations in thyroid

hor-mone-infused thyroidectomizedrats.Endocrinology. 124:2415-2418. 22. Wu, S. Y. 1983.Thyrotropin-mediated induction of thyroidal iodothy-ronine monodeiodinases in the dog. Endocrinology. 112:417-424.

23.Erickson, V. J., R. R. Cavalieri, and L. L. Rosenberg. 1982. Thyroxine 5'-deiodinase ofratthyroid,butnotthatofliver,is dependent on thyrotropin.

Endocrinology. 11 1:434-440.

24.Meinhold,H., U. Haselbach, B. Zegenhagen, A.Kyriakopoulos, and D. Behne. 1992.Effectsofseleniumdeficiencyandgraded selenium supply on typeI

5' deiodinaseandglutathione peroxidase activitiesinvarious rat tissues. J. Endo-crinol. Invest. 15(Suppl.2):131. (Abstr.)

25.Arthur, J. R., F. Nicol, P. W. Rae, and G. J. Beckett. 1990. Effects of seleniumdeficiencyonthethyroid glandandonplasmaandpituitary

thyrotro-phin and growth hormone concentrations in the rat. Clin. Chem. Enzym. Comms. 3:209-214.

26.Behne, D.,H.Hilmert,S.Scheid, H. Gessner, andW.Elger. 1988. Evi-dencefor specificselenium target tissues andnewbiologically important

seleno-proteins. Biochim.Biophys.Acta.966:12-21.

27.Goldstein, J., B. Corvilain, F. Lamy,D.Paquer, and J.E.Dumont.1988. Effectsofaseleniumdeficientdietonthyroidfunction of normal andperchlorate

treatedrats.ActaEndocrinol. 118:495-502.

28.Meinhold,H.,A.Campos-Barros,and D. Behne. 1992.Effects of iodine

deficiencyoniodothyroninedeiodinases inbrain, thyroidandperipheraltissue.

ActaMed.Austriaca. 19(Suppl.1):8-12.

29.DeGroot,L. J., P.R.Larsen, S.Refetoff,and J. B.Stanbury. 1984.

Hor-monesynthesis, secretion,andaction. In TheThyroidand ItsDiseases. 5th ed. JohnWiley&Sons,NewYork,NY.36-117.