l Triiodothyronine and l Reverse Triiodothyronine Generation in the Human Polymorphonuclear Leukocyte

l-Triiodothyronine and

l-Reverse-Triiodothyronine Generation in the Human

Polymorphonuclear Leukocyte

Kenneth A. Woeber, Betty A. Maddux

J Clin Invest. 1978;62(3):577-584. https://doi.org/10.1172/JCI109163.

Extrathyroidal monodeiodination of l-thyroxine (T₄) is the principal source of

l-triiodothyronine (T₃) and l-reverse-triiodothyronine (rT₃) production. To define some of the cellular factors involved, we examined T₃ and rT₃ generation from added nonradioactive T₄ in human polymorphonuclear leukocytes, using radioimmunoassays to quantify the T₃ and rT₃ generated. Under optimum incubation conditions which included a pH of 6.5 in sucrose-acetate buffer, the presence of dithiothreitol as a sulfhydryl-group protector, and incubation in an hypoxic atmosphere, significant net generation of T₃ and rT₃ was observed. Of the several subcellular fractions studied, the particulate fraction obtained by centrifugation at 27,000 g was found to possess the highest T₃- and rT₃-generating activities per unit quantity of protein. With respect to T₃ generation from substrate T₄, the K_m was 5 µM and the V_max was 7.2 pmol/min per mg protein. Propylthiouracil, methimazole, and prior induction of phagocytosis inhibited both T₃ and rT₃ generation, but T₃ generation was inhibited to a greater extent. rT₃, in a concentration equimolar to that of substrate T₄, did not alter T₃ generation, but inhibited T₃ generation when the molar ratio of rT₃ to T₄ approached 10:1. Under the incubation conditions employed, particulate fractions of leukocytes obtained from five cord blood samples displayed an essentially normal relationship between T₃- and rT₃ -generating activities, despite the distinctly divergent serum T₃ and rT₃ concentrations […]

Research Article

Find the latest version:

L-Triiodothyronine and

L-Reverse-Triiodothyronine

Generation

in

the

Human

Polymorphonuclear Leukocyte

KENNETH A. WOEBERwith the technical assistance ofBETTY A. MADDUX,

Department of Aledicine, Alount Zion Hospital and Medical Centerand University of California, Sant Francisco, California 94120

A BS TRAC T Extrathyroidal monodeiodination of L-thyroxine (T4) is the principal source of L-triiodothy-ronine

(T,)

and L-reverse-triiodothyronine

_(rT,)

pro-duction. To define some of the cellular factors in-volved, we examined T3 and rT3 generationfrom added nonradioactive T4 in human polymorphonuclear

leu-kocytes, using radioimmunoassays to quantify the T3 an-d rT3 generated. Under optimumii incubation condi-tions which included a pH of 6.5 in sucrose-acetate

bulffer, the presence of dithiothreitol as a

sulfhydryl-group protector, and incubation in an hypoxic at-mosphere, significant net generation of T3 and rT3was observed. Of the several subcellular fractions studied, the particulate fraction obtained by centrifugation at 27,000 g was found to possess the highest T3- and rT3-generatingactivitiesperunitquantity of protein. With respect to T3 generation from substrate T4,the K,m was 5

,MNI

and the

Vmax

was 7.2 pmol/min per mg protein. Propylthiouracil, methimazole, and prior induction of phagocytosis inhibited both T3 and rT3 generation,but

T3 generation was inhibited to a greater extent. rT3, inaconcentration equimolartothat of substrate T4, did not alter T3 generation, but inhibited T3 generation

when the molar ratio of rT3 to T4 approached 10:1. Under theincubation

coniditions

employed, particulate fractions of leukocytes obtained fromn five cord blood

samiples

displayed an essentially normal relationship

b)etween T3- and rT3-generating activities, despite the

distinctly divergent serum T3 and rT3 concentrations in these samples. From our findings, we draw the

fol-lowing conclusions: (a)the humanpolymorphonuclear

leukocyte possesses the ability to generate T3 and rT3

from substrate T4; (b) theT3- and rT3-generating activ-ities are associated principally with the 27,000 g par-ticulate fraction and display enzymic characteristics with asulfhydryl-group requirement;(c) T3-generating activity appears to be more susceptible to inhibitory influences thanrT3-generating activity; and (d) in cord

Receivedforpublicatiotn 5December 1977atndin revised

form 10I May 1.978.

blood leukocytes, the putative enzymes catalyzing

T:

and rT3 generation appear to be functionally intact un-der the experimental conditions employed.

INTRODUCTION

After the conclusive demonstration that the bulk of

L-triiodothyronine(T3)linman arisesthroughthe extra-thyroidal monodeiodination of the outer ring of L-thyroxine (T4) (1-4), there has been a resurgence of interest in the useof tissue preparations for examining this phenomenon in vitro. Thus, conversion of T4 to T3 has beendemonstrated to occurincultured human liver and kidney cells (5),incultured human fibroblasts (6), andin freshly isolatedhuman polymorphonuclear

leukocytes (7, 8). These studies involved the use of

radioiodine-labeled T4 as substrate,withdetectionand quantification oftheT3generated by chromatographic analysis. More recently, the availability of radioim-munoassay methods for measuring T3 has been ex-ploited for examining T4 to T3 conversioninvitro. Radio-immunoassay, in addition tobeingmore specific than

chromatographic analysis,permitsmoreprecise quanti-fication ofthe very small quantities ofT3 generated.

Using thisapproach, several investigators have reported the generation of T3 from added nonradioactive T4by

rattissues in vitro and have examined some of the char-acteristics of the T3-generating system (9-15).

Recent work has demonstrated convincingly that

extrathyroidal monodeiodination of T4 is also

respon-sible for almost all L-reverse T3 (rT3) production (16,

17), monodeiodination ofT4 in this instance occurring inthe inner ring. rT3 generationfrom T4 has alsobeen

demonstrated in rat liver in vitro (12, 13).

1_{Abbreviations used in this paper: DC, disrupted cell}

Wehave previously demonstratedthat intacthuman polymorphonuclear leukocytes, as wellasthe granule fraction isolated therefrom, are capable of

deiodinat-inglabeled T4 invitro, with the generationoflabeled

inorganic iodide and a labeled material that remains at the origin during

chromatography

(8, 18). In

addi-tion, wedetected labeledT3 as aproduct ofT4

deiodina-tion in intactcells_{by chromatographic analysis.} How-ever,thenetquantitiesofT3generated were verysmall,

and the-sensitivity ofchromatographic analysis was

in-sufficienttopermitquantitative definition of the

char-acteristicsof theT3-generatingsystem.Accordingly,we

undertook to study the generation ofT3 from added

nonradioactive T4 in the human polymorphonuclear leukocyte anditsvarious subcellular fractions, usinga

specific

radioimmunoassay

to quantify the T3

gen-erated. In addition, we examined the generation of rT3, using aspecific

radioimmunoassay

for quantifying the rT3 formed.

METHODS

Blood was collected, with 100 U/25 ml heparin as the

anti-coagulant, fromhealthy laboratorypersonnel. Bloodwasalso obtained _{from the umbilical cords of five full-term normal} infants.

Plasticware orsiliconized glasswarewas usedthroughout. The leukocytes were isolated from the blood as described previously (8);atleast90%of the cellssoobtainedwere poly-morphonuclear,3-8% were lymphocytes, <2% were

mono-cytes, and erythrocytes were virtually absent. Freshly

iso-latedleukocytes were employed for each experiment. Leukocyte fractionation. The isolated leukocytes were disrupted by homogenization in cold 0.34 M sucrose. This disruptedcell preparation (DC)wascentrifugedat 400 gfor

15 min at5°Cto removeunbroken cells,nuclei,andcelldebris, and theresulting supemate (S1) was saved. Thepellet was

washedthreetimeswithcold0.25Msucrose,0.02 M Tris, pH

7.2, containing 0.5% (vol/vol) Triton X-100 (Beckman

In-struments,Inc.,Fullerton, Calif.)toyieldanuclearfraction (N) whichwasthenwashed andsuspendedincold 0.25Msucrose,

0.02 M sodium acetate, pH 6.5 (sucrose-acetate buffer). SI

and,whendesired,DCin0.34Msucrose weredilutedwith

cold sodium acetate toyield final concentrations of0.25 M sucrose and0.02 Msodium acetate,pH 6.5. In some

experi-ments, Si wasdiluted withcold Tris when final pHvalues of7.0 and above were desired. SI was further resolved by centrifugationat27,000gfor15 min at5°Cinto aparticulate fraction(P) andsupernateS2.ThePfraction,which has been shown tobe composed ofgranules, plasmamembrane frag-ments, and occasional mitochondria (19), was suspended in the sucrose-acetate buffer.

When thePfraction from phagocytosingcells wasdesired, the intactleukocytes were incubatedat 37°C for 40min

be-forefractionation in Krebs-Ringer phosphate glucose buffer,

pH7.4, containing0.5mg/l x 107 cellszymosan(K & K Labo-ratories Inc., Plainview, N. Y.) that hadbeen opsonized by

coatingwithfresh plasma, as described in detailpreviously (8).Intactleukocytes fromthe sameisolatewerepreincubated concurrently in Krebs-Ringer phosphate glucose buffer alone to serve as control cells. The P fraction from the phagocytosing cells was free of zymosan particles, these sedimentingwiththe initial400 gpellet during fractionation,

and theyield, in terms ofproteincontent, was _very similar

tothat of the Pfraction fromthe control cells.

Protein concentrations of all fractions were measured

ac-cording

tothemethodof_Lowryetal. _(20).

MeasurementofT4conversiontoT3andrT3. The cell

frac-tionsin sucrose-acetate buffer(0.4ml,representingthe_yield

from 1x 107

leukocytes)

wereaddedto

_plastic

tubes

contain-ing 1 nmol ofnonradioactive T4 (Sigma Chemical Co., St. Louis, Mo.) and 50 nmol of dithiothreitol (DTT)

(Sigma

ChemicalCo.); the final volume of thereaction mixturewas

0.5ml,

yielding

finalconcentrations of_T4and DTT of 2 ,uM and 100 ,uM, respectively. For each fraction acontrol was

prepared in which the cell fraction, and the T4 plus DTT

wereincubatedinseparate tubesandthenmixedatthe end ofincubation. This control was used to correct for _T3 and

rT3contamination of theadded nonradioactiveT4,for

endog-enous T3and_rT3inthe cellfraction,andforcross-reactivity of the addedT4 with the _anti-T3 and anti-rT3 antisera used

in the radioimmunoassays. In some experiments, the

reac-tion mixtures were enriched with small quantities of

non-radioactive T3

(Sigma

Chemical Co.) and rT3

(obtained

throughthe_{courtesyof Dr. Robert I.} Meltzer from Warner-Lambert Research Institute, Morris Plains, N. J.) to assess

the _degradation ofT3 and rT3; here, reaction mixtures en-riched with _{identical quantities ofT3}and rT3atthe endof incubation servedascontrols.

The tubeswere incubated at37°C in a metabolicshaker.

An hypoxic atmosphere was attainedbycovering the tubes withagassinghood andgassingwith 100%nitrogen

through-outtheincubationperiod.Atthe end ofincubation,50

_Al

of

humanserumthat hadbeenstrippedofitsiodothyronine con-tentby overnightexposureto ananion_exchangeresin

(Amber-lite _IRA-400, Mallinckrodt Inc., St. Louis, Mo.) was added to each tube and the contents mixed. The stripped serum

was addedbecause itfacilitatedpelleting of the very small

amountof tissue in thereaction mixtureaftertheadditionof ethanol. This was followedby the addition of 1 ml of cold 95% ethanol and further _mixing. After standingat 50C for 20 _min, the tubes were centrifuged at 1,500g for 20 min and thesupernatantethanolextracts werecollected. In_eight

individual reaction mixtures, this extraction procedure

re-sulted in recovery of90±2%(mean±SE)of added nonradio-activeT3and_80±2% ofadded nonradioactiverT3.

The concentrations of_T3 and _rT3 in the ethanol extracts were measuredbydouble-antibody radioimmunoassay. For the_{T3radioimmunoassay, highspecificactivity[1251]T3(-300} isCi/nmol)wasobtained from Abbott Diagnostics,Diagnostic

Products, NorthChicago, Ill.,nonradioactive T3 from Sigma

Chemical_Co.,and rabbitanti-T3antiserum from Endocrine

Sciences,Tarzana,Calif. Cross-reactivitiesofrT3andT4inthis

assaydidnotexceed0.03%and0.3%,respectively. For the rT3radioimmunoassay, L-[125I]rT3, nonradioactive L-rT3,and rabbitanti-L-rT3antiserumwereobtained from Serono

Lab-oratories_Inc.,Boston,Mass.Cross-reactivities ofT3andT4in this_assaydidnotexceed0.003 and0.09%,respectively.The second antibody for both assays, goat anti-rabbit gamma globulin,wasobtainedfromAntibodies Inc., Davis,Calif. In

bothassays,ethanolwasroutinely addedtothe standardsto

yielda finalconcentration identical tothat in the samples;

additionof_{ethanol yielded}astandardcurvethat wasvirtually

superimposableonthatobtainedwithadditionofanethanol extractofiodothyronine-free serum, thecoefficientsof varia-tion intwoexperimentsbeing8.1 and7.3%. Foreach stand-ard or_sampleablank was preparedto contain all reagents

except the _anti-T3 or_-rT3 antiserum. The value for percent oftracerbound in the blankwas subtracted from the value

inthe corresponding standard orsampleto correctfor

specificbinding. The net quantity of T3 or rT3 generated in a given reaction mixture was derived by subtracting from the concentration ofT3 or rT3 in the ethanol extract thereof, the concentration ofT3 or rT3 in the ethanol extract of the cor-responding control in which the cell fraction and theT4plus DTT had been kept separated until the end of incubation. All valuesforT3and rT3 werecorrected for recovery, using the

coefficients cited earlier.

RESULTS

Determination ofoptimum incubation conditions.

Ininitialexperiments carriedoutinair,significantnet generationofT3 from 1nmol of added nonradioactive T4 was observed after 3h ofincubation with DC and with S1 derived therefrom. T3-generating activity was found to have apH optimum of 6.5 insucrose-acetate

buffer. Activity was abolished by prior boiling of the tissue for 30 min or by incubation at 5°C, and was greater at 37°Cthanat roomtemperature. Asdepicted

in Fig. 1,additionto thereaction mixture ofDTT in a

final concentration of100lM resulted in a moderate increase in net T3 generationwhenincubationwas car-ried outinair. However, whenincubation wascarried

outinanhypoxic atmosphereattainedbygassing with 100% N2, the presence of100

_AM

DTT resulted in a great increaseinnetT3generation andsignificant quan-tities ofrT3 were also detected. In the case of T3, the gross quantitygeneratedwas7.1+0.4pmol, (mean+SE,

n =6), andthe quantitypresentinthe controls, repre-senting T3 contaminationof the addedT4,endogenous

T3 inthetissue,andcross-reactivityof the addedT4 with

the anti-T3 antiserum, was 3.0+0.3 pmol. Thus, the net quantityofT3generatedwas4.1+0.3pmol. Inthe caseofrT3,the values forgross,control,andnet quan-tities generated were 3.3+0.2,

1.1+0.1,

and 2.2+0.1

51

UT3 EIrT3

4--1

3-0

2-

O--

0-DTT

(6)

(6)

II

0 + + 0

N2 0 0 + +

FIGURE 1 Influence of DTT and hypoxia attainedby gas-sing with 100% N2on netT3 and rT3 generation from 1 nmol of added nonradioactive T4 by S1 derived from -1 x 107

human polymorphonuclear leukocytes and suspended in sucrose-acetatebuffer,pH 6.5. Incubationwas carriedout at

37°C. Mean+SE and number of individualreaction mixtures

studied are indicated.

pmol, respectively. In the absence of DTT, incubation in N2 yielded values that were very similar to those obtained in air. As depicted in Fig. 2, a linear log con-centration-responserelationshipwasdemonstrable be-tween 10

,uM

and 1 mM DTT, and in this concentra-tionrange,the ratios of T3- to rT3-generating activities remained quite constant, ranging only between 2.3 and 2.6:1. At concentrations _ 1 mM DTT, significant gen-eration ofT3 from T4 was sometimes observed in the absence oftissue, and at>1-mMconcentrations no further increase in generation by tissue was observed. Replace-ment of sucrose-acetate buffer with 0.14 M KCI-0.02 M sodium acetate buffer, pH 6.5, as the suspending medium, resulted in 54% reduction of T3 generation and 59% reduction of rT3 generation. Accordingly, su-crose-acetate buffer, pH 6.5, enriched with 100lM DTT was routinely employed as the suspending

me-dium, and incubation was carried out in 100% N2 at 370C.

Comparison of activity in various cell fractions.

Fig. 3 depicts the net generation of T3 and rT3 from 1

nmol ofadded T4aftera 3-h incubation with the frac-tionsobtained from -=1 x 107cells, using the optimum

incubation conditions. The bulk of activity in the DC preparation was recovered in SI, although the

nu-clearfraction also possessedsomeactivity. Asdepicted in Fig. 4, the bulk of activity in the S1 fraction was

recoveredin Pderived therefromby centrifugation at

27,000g,with little activity remaining in the resulting supernate (S2). The T3- and rT3-generating activities of P fraction per unit quantity of protein were

approxi-mately four- to fivefold greater than those of the DC preparation.

Characteristics of activity of particulate fraction. Thenet generation ofT3 and rT3 from 1 nmol of added T4aftera3-h incubation with P fraction obtained from

12

~Z8

rT3

0,

10 30 I00 300 1000 DTT(MM)

FIGURE 2 NetT3 and rT3 generation from 1 nmol of added nonradioactive T4 by Si derived from _1 x 107 human

polymorphonuclear leukocytes as afunction of the

concen-trationof addedDTT.Thesuspendingmedium was sucrose-acetate buffer, pH 6.5, and incubation was carried out in

100% N2 at37°C. Each point represents the mean of

S- 0UT3 DrT3

qsIr) ~~~~~~~~(4)

'a

3-2 4)

Fraction DC N SI Protein, ig 628 166 380

FIGURE3 Comparison of net T3 and rT3 generation from 1

nmol of addednonradioactiveT4 by various fractionsobtained from 1 x107humanpolymorphonuclear leukocytes. In this and all the subsequent experimentsdepicted,thesuspending medium was sucrose-acetate buffer, pH 6.5, enriched with 100,uM DTT, and incubation was carried out in 100 % N2 at37°C. Mean±SE and number of individual reaction mixtures studied are indicated.

_1

x 107 cells, was much greateratpH 6.5 than atpH

7.4or 9.0. ThepmolofT3and rT3generatedwere,

re-spectively, 3.7 and 2.4 at pH 6.5, 1.2 and 0.9 at pH 7.4, and 1.1and 1.2atpH9.0. Using the optimum

in-cubation conditions at pH 6.5, 5.0+1.3%

(mean±+SE)

of added nonradioactiveT3and8.6± 1.4%of added non-radioactive rT3 were degraded after3 h in 10individual reaction mixtures. As depicted in Fig. 5, net T3 gen-eration

appeared

to increase as a linear function of

time during the first 2 h of incubation, and then

ap-peared

to level off during the 3rd h. In the case of rT3, net generation increased with time, but did not appeartolevel off during the 3rdh. Inan

experiment

inwhich incubation was carried out for 20 h, the net quantity ofT3

generated

after20 h was only 37%greater

thanthat after3 h,whereas thenetquantity ofrT3

gen-

5-,%.

4-i3I..

3-Q

Z.1

2-Fraction

Protein,m49

(4) XT3 0rT3

(4)

({4)

(4)

(3)

Si P _S2

316 120 154

FIGuRE 4 Comparison ofnetT3and rT3generation from 1 nmol ofadded nonradioactive T4 by Siand itsconstituentP

fraction and S2derived from 1 x 107human

polymorpho-nuclearleukocytes. Mean+SEand number of individual

reac-tion mixtures studied are indicated.

2-~

k

~~~~~~~rT3

0iI4

0 30 60 120 180 Time(min)

FIGURE 5 Time-course of net

T3

and

rT3

generation from 1 nmol of added nonradioactive

T4

by the P fraction derived from _1 x 107 human polymorphonuclear leukocytes. Each point represents the mean of duplicatedeterminationsof an individual reaction mixture.

erated was

116%

greater than that after 3 h. Fig. 6

depicts the P fraction dependence ofnet T3 and rT3 generation from 1 nmol of added T4 fora 3-h incuba-tionperiod. Net generation ofboth T3 and rT3 increased with an increase in the

concentration

ofP fraction in

the reaction mixture, and at all

concentrations,

theratios of

T3-

to

rT3-generating

activities

remained

constant

at 2:1. Here, the highest concentration ofPfraction of 326

,ug/ml

represented

the mean yield of fraction (163 ,ug) from

s1

x

107

cells

suspended

inafinal volume of

0.5ml.Fig.7depicts the

T4-concentration

dependence

ofnetT3 generation. A double

reciprocal

plot of the net quantity of T3 generated after 15 min

incubation

with 18,ug ofPfraction as afunction of the

concentra-tion of added T4,

suggested

that T3 generation con-forms to

Michaelis-Menten

kinetics with a 5-,uM

Km

and a 7.2-pmol

Vma.

of T3

generated

per minute per milligram protein. In the case of rT3, generation was greater with higher

concentrations

of added T4, but the

data obtained in several

experiments

could not be plottedinthe double reciprocal manner.

Consequently,

reliable values forKmand Vmax

could

notbe

derived.

5-0 ₀₀ 200 300 400

Pfraction

_(Ag/ml)

FIGURE 6 Net T3 and rT3 generation from 1 nmol of added

nonradioactive T4 as a function of the concentration of P fraction. Values represent

mean±SE

of fourindividual

reac-tionmixtures.

580 K. A. Woeber OJ

Km*SLM 0 0.2 0.4 0.6 0.8 1.0

T4

FIGURE 7 T4-concentration dependence ofnetT3 generation

by18,ug Pfraction. V measured in picomoles of T3 generated per 15 minute, and T4 measured in micromolar

concentra-tions. Values represent mean+SE of three individual

reac-tion mixtures.

Factors affecting activity of particulate fraction. Table Isummarizes the influence of several factorson the T3- and rT3-generating activities ofP fraction ob-tainedfrom =- x 107cells, employing 1 nmol of added

T4, a3-h incubation period,and the optimum

incuba-tion conditions. Addition to the reaction mixture of

either propylthiouracil (PTU) or methimazole (MMI)

in concentrations of 10, 30, and 100 ,uM was

accom-panied by a progressive inhibition ofnet T3 and rT3 generation. Similarly, prior induction of phagocytosis

in the leukocytes was accompanied by inhibition of

theT3- and rT3-generating activitiesofthe subsequently isolated P fraction. In the foregoing circumstances, T3 generationappearedtobedisproportionately depressed,

asjudged from the decreases in the ratio ofT3to rT3

generated. The presence of 100 ,uM PTU, 100 ,uM

MMI, or P fraction from phagocytosing cells did not

significantly affect thedegradation of added

nonradio-active T3orrT3.Thus, the inhibitionofnetT3 andrT3 generation in the foregoing circumstances could not

be attributed toincreased degradation of product gen-erated. Alsopresentedin Table I arethe results ofan experiment in which the influence ofadded rT3 on T3 generation was examined. Addition of rT3 in a concentration equimolartothat of the added substrate T4 did not interfere with net T3 generation, but inhibition was observed when a 10-fold greater con-centration of rT3was employed.

Activity of particulate fraction from cord blood. Table II summarizes the results of studies in which

T3 and rT3 generation from 1 nmol of added T4 was examined with P fraction of leukocytes obtained from cord blood. Also presented are the values for

concurrently studied P fraction from adult blood, as well as the values for serum T3 and rT3 concen-trations. Incubation conditions were as outlined be-fore. Becausethe yieldofPfraction varied among the

TABLE I

Influence ofVariousFactors on Net T3-and rT3-Generating

and T3- and rT3-DegradingActivitiesof Leukocyte ParticulateFraction

Degrading

Net generating activity activity

Experi-ment Factor T3 rT3 T3/rT3 T3 rT3

pmolI3 h %added

1 None 5.6 2.5 2.2 5.4 10.2

PTU, 10 ,uM 3.5 1.8 1.9 PTU, 30,uM 3.2 1.5 2.1

PTU, 100,uM 1.3 1.2 1.1 7.3 11.0 MMI, 10,tM 2.3 1.3 1.8

MMI, 30 ,uM 1.5 1.0 1.5

MMI, 100,UM 1.1 0.8 1.4 0 8.0

2 None 7.4 3.0 2.5 3.5 5.1

Phagocytosing 3.2 1.9 1.7 2.9 10.3

3 None 3.8

rT3, 2 ,uM 4.5 rT3,20uM 2.6

different blood samples, values for T3 and rT3 gen-eration per 100,ug protein arealsopresented.

In all five cord blood samples, the expected great increase in serum rT3 concentration and modest de-crease in serum T3 concentration relative to adult

blood were observed. For the group of five cord samples as a whole, T3- and rT3-generating activities per unit quantityofPfraction did notdiffersignificantly from those for adult blood. In addition, the ratios of T3-generating activitytorT3-generating activity, which areindependent ofthequantityofPfractionemployed,

were very similarin the cord and adultsamples.

DISCUSSION

Inthe present study, we have provided evidencethat thehumanpolymorphonuclear leukocytepossesses the

ability to generate both T3 and rT3 from substrate T4.

The pH optimum for T3 generation was found to be 6.5 which is very similar to that observed by others in rat liver homogenate (12) or rat liver microsomes (13). In contrast to these latter studies in which no rT3 generation could be detected at pH 6.5, this only

beingdetected at -pH 9.0, we found significant quan-tities of rT3 at pH 6.5. In fact, the quantity of rT3

TABLE II

ComparisonofSerum T3 and rT3 Concentrations and T3- and rT3-Generating Activities ofLeukocyte

ParticulateFractionfromCordBlood andfromNormal Adult Blood

Serum

concentration Netgenerating activity

Experi-ment State T, rT, Pfraction T, rT3 T3 rT, T3/rT3 ngldl Ag pmolI3 h pmolI3h/100Ag

1 Cord blood 42 233 175 2.9 1.7 1.7 1.0 1.7

Concurrentadult 133 12 201 3.6 1.9 1.8 0.9 2.0

2 Cordblood 88 168 105 0.6 0.4 0.6 0.4 1.5

Concurrentadult 124 14 191 4.3 2.1 2.2 1.1 2.0

3 Cord blood 43 160 88 1.2 0.6 1.4 0.7 2.0

Concurrentadult 116 11 105 2.6 1.0 2.5 1.0 2.5

4 Cord blood 64 135 179 4.0 1.6 2.2 0.9 2.4

Concurrentadult 116 12 196 4.2 1.8 2.1 0.9 2.3

5 Cord blood 42 130 130 2.3 1.2 1.8 0.9 2.0

Concurrent adult 110 19 217 3.2 1.9 1.5 0.9 1.7

Cordblood, Mean 1.5 0.8 1.9

±SE +0.3

±0.1

+0.2 Concurrentadult, Mean 2.0 1.0 2.1

±SE ±0.2 ±0.0 ±0.1

demonstrated that thioglycolate caused a threefoldor greater stimulation of T4 deiodination to inorganic

iodide inratliverhomogenate. Recently Hufner etal. (12) have demonstratedthat mercaptoethanolenhances

T4 conversion to T3 in rat liver homogenate, and

Visser et al. (23) and Chopra (24) have reported a

similar effect of DTT. Although, in our study, the

presenceofDTT in a concentrationof100 ,uMresulted

in a slight increase inT3 generation when incubation

wascarriedoutinair,the stimulatory effect ofDTTon T3andrT3generation wasclearly evident when

incuba-tion wascarried outin anhypoxic atmosphere. These

findings suggestthat SHgroups either serve as

cofac-tors or are a constituent of the putative enzymes

catalyzing T3 or rT3 generation, and because such groups arereadily oxidized inairto disulfides, an

hy-poxic atmosphere serves to maintain them in the

re-ducedstate.

The bulk of the T3- and rT3-generating activities in

the humanpolymorphonuclear leukocytewasfoundto beassociatedwith the particulatefraction obtainedby centrifugation at 27,000g. This finding is consonant

withourearlier work demonstratingthattheleukocyte granule fractionrepresents thesubcellular locus forthe

degradation of T4 to inorganic iodide and origin

ma-terial (18).Inthis earlierwork,theexperimental

condi-tions employed did not permit detection of either T3 or rT3 generation. In rat liver, on the other

hand, the subcellular locus ofT3 and rT3 generation has beenreported to beassociatedwiththemicrosomal

fraction(9, 13), althoughinratkidney,itappearstobe associated with a particulate fraction comprising

plasma membranes and mitochondria (15).

Recently,

Maciel etal. (25) have presented evidence suggesting

that the plasma membrane fraction is the subcellular locus of T3-generatingactivityinratliver.Accordingly,

entrapmentofactive

plasma

membranecomponentsin various subcellular fractions could accountfor the ap-parentdiscordanceinthereportedsubcellular

localiza-tionofT3-generating activity.Although our

particulate

fractioniscomprised largely of neutrophil granules, the

presence of plasma membrane components could be responsible for the activity observed.

The T3- and rT3-generating activities of the human leukocyte displayedenzymiccharacteristics, including

pH and temperature optima and tissue-concentration

andtime dependence.ThederivedvalueforKmofthe

putative enzyme responsible forT3generation was of the order of5 ,uM. This is similar to the Km ofthe

T3-generating system in rat liver and kidney ho-mogenates (12, 14, 15). On the other hand, the Vmax of7.2pmolof T3generatedperminuteper

milligram

of

particulate

fraction was much greater than that ofrat

liver homogenate(0.13pmol/minpermg)(14)orthatof

rat kidney homogenate (0.03 pmol/min per mg) (15), reflectingat least in parta greater degree of

purifica-tionof theputative enzyme. Inthe caseofrT3 genera-tion, reliable values for Km and Vmax could not be

derived fromthe data obtained.

A variety of physiologic and pathologic states are

accompanied by subnormal serum T3 concentrations

and by increased serum rT3 concentrations (26), and

this has ledtothe suggestionthatinnerandouter ring

monodeiodinationofT4 maybe regulatedina

recipro-cal manner. PTU has been shown to result in a de-crease in serum T3 concentration and an increase in serum rT3 concentration in vivo (27) and to in-hibit T3 generation from T4 by rat liver and kidney homogenates in vitro (10-12, 14, 15), but its effect on rT3 generation has not beenexamined. In common with this earlier work, we found that PTU inhibited T3 generation and, in addition, found that rT3 generation wasalso inhibited, butto alesserextent. MMI in a con-centration equimolarto that ofPTU was also found to inhibit both T3 and, to a lesserextent, rT3 generation

byparticulatefraction. This latterfindingis discordant with previous reports of a lack of an effect of MMI onT4 conversion to T3 in vivo (27) or invitro (11, 14, 15). This discordancemaybe due tothe large concen-trations ofMMI and the small quantity of tissue

em-ployedinourstudy relativetotheinvivo circumstance and to the previous in vitro studies. As was the case

withPTUandMMI, priorinduction ofphagocytosis also

resultedininhibition of bothT3and, to a lesser extent, rT3 generationbyparticulate fraction. Thus,noneof the foregoing factors resultedin a true dissociation of

T3-and rT3-generating activities, but T3 generation

ap-peared to be more susceptible to inhibition than rT3 generation. Because outerring monodeiodination of T4 is involved in the degradation ofrT3 aswell as repre-senting the pathway for T3 generation whereas inner ring monodeiodination has the converse actions, a greater inhibition of outer ring monodeiodination

might ultimately result in a net increase in rT3 and a netdecrease in T3 in the in vivo circumstance.

rT3 hasbeen reported to inhibitT4 conversion to T3 in rat liver homogenate (11, 12, 14), although the physiologic relevance of this finding has been

ques-tioned (28).Wealso found thatrT3inhibitedT3 genera-tion from T4, but only when the molarratio ofrT3 to T4 approached 10:1.

The veryearly newborn periodrepresents a physio-logic state in which serum T3 and rT3 concentrations are distinctly divergent. Despite this divergence, we

found that the T3- and rT3-generating activities of particulate fractionfrom cord bloodleukocytes did not

differ significantly from those ofnormal adult blood.

This discordance could be interpreted as suggesting

thatthe leukocyte isan unimportant siteofT3 and rT3 generation with little influence on overall T3and rT3 economyinvivo. Ontheotherhand,thereisan

alterna-tive explanation for our findings. Chopra (24) has recently reported that DTT restores T3-generating ac-tivity of liver homogenate from fetal sheep, and,

accordingly, suggested that deficient generation ofT3 infetal tissue was due not to a quantitative deficiency of the enzyme per se, but to the redox state of SH

groups in thetissue. Because our incubation conditions required the presence of DTT for detection of signifi-cant net T3and rT3 generation, our findings of normal activitywith particulate fraction from cord blood

leuko-cytes mayreflect correction of the putative abnormality in the redox state of SH groups. Thus, if it is granted that the leukocyte is an important site of T3 and rT3 generation,the results would suggest that, as in the case of fetal liver, the putative enzymes catalyzing T3 andrT3 generation appear to be functionally intact in the very earlynewborn period.

Thehumanpolymorphonuclear leukocyteresembles otherthyroid-hormone-responsive tissues in display-ing a calorigenic response tothyroid hormone and in possessing saturable nuclear binding sitesfor T3(29).

The present study demonstrates that this cell also has the capability of generating T3 and rT3 from

substrate T4.

ACKNOWLEDGMENTS

This study wassupportedinpartbyresearch grant AM-19081

fromthe National Institutes ofHealth, Bethesda, Maryland,

and in part by Mount Zion Hospital and Medical Center, San Francisco, California.

REFERENCES

1. Braverman, L. E., S. H. Ingbar, and K. Sterling. 1970. Conversionofthyroxine (T4) to triiodothyronine (T3) in

athyreotic humansubjects.J. Clin.Invest.49: 855-864. 2. Pittman, C. S., J. B.Chambers,Jr., andV. H.Read. 1971.

The extrathyroidal conversion rate of thyroxine to tri-iodothyronine in normalman.J. Clin. Invest.50:

1187-1196.

3. Surks, M. I., A. R. Schadlow, J. M. Stock, and J. H.

Oppenheimer. 1973. Determination ofiodothyronine

ab-sorption and conversion of L-thyroxine(T4) to

L-triiodo-thyronine (T3) using turnover rate techniques. J. Clin. Invest. 52: 805-811.

4. Braverman, L. E., A.Vagenakis, P. Downs, A. E. Foster, K.Sterling, andS. H.Ingbar. 1973.Effect of replacement doses ofsodium-L-thyroxine on the peripheral metab-olism ofthyroxine and triiodothyronine in man.J. Clin. Invest. 52: 1010-1017.

5. Sterling, K., M. A. Brenner, and V. F. Saldanha. 1973. Conversionofthyroxinetotriiodothyronine by cultured

human cells. Science(Wash.D. C.). 179: 1000-1001. 6. Refetoff,S., R.Matalon, and M. Bigazzi. 1972. Metabolism

ofL-thyroxine (T4)andL-triiodothyronine (T3) byhuman

fibroblastsintissueculture: evidencefor cellularbinding

proteins and conversion of T4 to T3. Endocrinology. 91:934-947.

7. Klebanoff, S. J., and W. L. Green. 1973. Degradation of

thyroidhormonesby phagocytosinghumanleukocytes.J. Clin.Invest. 52: 60-72.

8. Woeber, K. A., and S. H. Ingbar. 1973. Metabolism of

L-thyroxine byphagocytosing human leukocytes. J. Clin. Invest. 52: 1796-1803.

9. Hesch, R. D., G. Brunner, and H. D. Soling. 1975. Con-versionofthyroxine (T4)andtriiodothyronine (T3)and the subcellularlocalization of the converting enzyme. Clin. Chim. Acta. 59: 209-213.

tri-iodothyronine by rat liverhomogenate. Biochem.J. 150: 489-493.

11. Chopra, I. J. 1977. A study of extrathyroidal conversion ofthyroxine (T4) to3,3',5-triiodothyronine (T3) in vitro. Endocrinology. 101: 453-463.

12. Hufner, M., M. Grussendorf, and M. Ntokalou. 1977. Properties of thethyroxine(T4)monodeiodinatingsystem in rat liverhomogenate.Clin. Chim. Acta. 78: 251-259. 13. Hofken, B., R. Kodding, and R-D Hesch. 1977. Con-versionofT4 to T3 andrT3and their cytoplasmic binding:

pHdependency. Clin. Chim.Acta. 78: 261-266. 14. Kaplan, M. M., and R. D. Utiger. 1978. lodothyronine

metabolisminratliver homogenates.J. Clin. Invest. 61: 459-471.

15. Chiraseveenuprapund, P., U. Buergi, A. Goswami, and

I. N. Rosenberg. 1978. Conversion of L-thyroxine to

triiodothyronine in ratkidney homogenate.

Endocrinol-ogy. 102: 612-622.

16. Chopra,I.J. 1976. Anassessment ofdailyproductionand

significance of thyroidal secretion of 3,3',5'-triiodo-thyronine(reverse T3) in man.J.Clin. Invest.58: 32-40. 17. Gavin, L., J. Castle, F. McMahon, P. Martin, M.

Ham-mond, and R. R. Cavalieri. 1977. Extrathyroidal con-versionof thyroxineto3,3',5'-triiodothyronine (reverse-T3) and to 3,3',5-triiodothyronine (T3) in humans. J.

Clin. Endocrinol. Metab.44: 733-742.

18. Woeber,K. A. 1976. Agranule-associated L-thyroxine

de-iodinating system inthehumanleukocyte. Endocrinology

98: 802-806.

19. Babior, B. M., J. T. Curnutte, and B. J. McMurrich.

1976. The particulate superoxide-forming system from

humanneutrophils.J. Clin. Invest. 58: 989-996. 20. Lowry, 0. H., N. J. Rosebrough, A. F. Farr, and R. J.

Randall. 1951. Protein measurement with the Folin phenolreagent. J. Biol. Chem. 193: 265-275.

21. Cleland, W. W. 1964. Dithiothreitol, a new protective

reagent for SH groups.Biochemistry. 3: 480-482. 22. Nakagawa, S., and W. R. Ruegamer. 1967. Propertiesof

a rat tissue iodothyronine deiodinase and its natural

inhibitor.Biochemistry. 6: 1249-1261.

23. Visser, T. J., I. Van Der Does-Tobe, R. Docter, and G.

Hennemann. 1976. Subcellular localizationof arat liver enzyme convertingthyroxine intotri-iodothyronine and possibleinvolvement of essential thiol groups. Biochem.

J. 157: 479-482.

24. Chopra, I. J. 1978. Sulfhydryl groups and the

mono-deiodination of thyroxine to triiodothyronine. Science

(Wash.D. C.). 199: 904-906.

25. Maciel, R. M.B., Y. Ozawa,andI.J. Chopra. 1978.

Sub-cellular localization of T4 andreverseT3 (rT3)outer ring

monodeiodinaseactivities. Clin. Res. 26: 107A.(Abstr.)

26. Cavalieri, R. R., and B. Rapoport. 1977. Impaired peripheral conversion ofthyroxine to triiodothyronine.

Annu.Rev. Med. 28: 57-65.

27. Kaplan, M. M., M. Schimmel, and R. D. Utiger. 1977.

Changes in serum 3,3',5'-triiodothyronine (reverse T3)

concentrations with altered thyroid hormone secretion

andmetabolismJ. Clin. Endocrinol. Metab.45: 447-456. 28. Nicod, P., A. Burger, G. Strauch, A. G. Vagenakis, and L. E. Braverman. 1976. Thefailure of physiologic doses

ofreverse T3 toeffect thyroid-pituitary functionin man. J. Clin. Endocrinol. Metab.43: 478-481.

29. Woeber,K. A.1977.Observations concerningthe binding

of L-triiodothyronine in the human polymorphonuclear leukocyte.J. Clin. Endocrinol. Metab. 44: 62-68.