• No results found

Cellular and molecular mechanisms for reduced interleukin 4 and interferon gamma production by neonatal T cells

N/A
N/A
Protected

Academic year: 2020

Share "Cellular and molecular mechanisms for reduced interleukin 4 and interferon gamma production by neonatal T cells"

Copied!
10
0
0

Loading.... (view fulltext now)

Full text

(1)

Cellular and molecular mechanisms for reduced

interleukin 4 and interferon-gamma production

by neonatal T cells.

D B Lewis, … , S J Kahn, C B Wilson

J Clin Invest.

1991;

87(1)

:194-202.

https://doi.org/10.1172/JCI114970

.

The mechanisms by which T lymphocytes acquire the capacity to produce interleukin 4

(IL-4) and other lymphokines during intrathymic and extrathymic development are poorly

understood. To gain insight into this process, we determined the capacity of human

neonatal and adult T lineage cell populations to produce IL-4 after polyclonal activation. IL-2

and interferon-gamma (IFN-gamma) production were studied in parallel, since their

production by neonatal T cells is known to be similar or diminished, respectively, compared

to adult T cells. Production of IL-4 by neonatal CD4+ T cells and IFN-gamma by neonatal

CD4+ and CD8+ T cells was markedly lower compared with analogous adult cell

populations, whereas IL-2 production was similar. Transcription of IL-4, as determined by

nuclear run-on assays, and IL-4 mRNA-containing cells, as determined by in situ

hybridization, were undetectable in neonatal T cells, whereas both were detectable in adult

T cells. IFN-gamma transcription and IFN-gamma mRNA-containing cells were reduced in

neonatal T cells compared with adult T cells. Reduced lymphokine production by neonatal

T cells correlated with their lack of a CD45R- (putative memory T cell) population; cells with

this surface phenotype comprised 30-40% of the adult CD4+ T cells and were highly

enriched for IL-4 and IFN-gamma, but not IL-2 production. IL-4, IFN-gamma, and IL-2 mRNA

expression by neonatal CD4+CD8- thymocytes was similar to that found in […]

Research Article

(2)

Cellular and Molecular Mechanisms for Reduced Interleukin 4

and

Interferon-'y

Production by Neonatal T Cells

DavidB. Lewis,* Charles C.Yu,*JeffMeyer,*B. KeithEnglish,*StuartJ. Kahn,* andChristopherB. Wilson*$

*Divisions ofImmunology/Rheumatology and Infectious Diseases, Departments ofPediatrics and

Immunology,*

University of Washington and Children's Hospital and Medical Center, Seattle, Washington 98105; and tImmunex Corporation, Seattle, Washington 98101

Abstract

Themechanisms by which T lymphocytes acquire the capacity

toproduceinterleukin4(IL-4)andotherlymphokinesduring intrathymic and extrathymic development are poorly under-stood. Togain insight into this process, we determined the

ca-pacity ofhuman neonatal and adult T lineage cell populations to produce IL4after polyclonal activation. IL-2 andinterferon-y

(IFN-'y)

production were studied in parallel, since their produc-tion by neonatal T cells is known to besimilarordiminished, respectively, compared to adult T cells. Production of IL4 by neonatal

CD4'

T cells andIFN-yby neonatalCD4'andCD8+ T cells was markedly lower compared with analogous adult cell populations, whereas IL-2production wassimilar.

Transcrip-tion ofIL4, as determined by nuclear run-on assays, and IL4 mRNA-containing cells, as determined by in situ hybridization, wereundetectable in neonatal Tcells, whereas both were detect-able in adult T cells.

IFN-'y

transcription and IFN--y

mRNA-containingcellswerereduced in neonatal T cells compared with adult T cells. Reduced lymphokine productionby neonatal T cellscorrelatedwith their lack of aCD45R-(putative memory Tcell) population; cellswith this surfacephenotypecomprised

30-40%of the adult

CD4'

Tcells andwerehighlyenriched for IL4 and IFN--y, but not IL-2 production. IL-4, IFN-y, and IL-2 mRNA expression by neonatal CD4+CD8- thymocytes wassimilartothatfoundincirculatingneonatalCD4+Tcells. Taken together, these findings suggest that the extrathymic

generation ofmemory Tcellsduring postnatal lifemay result in

an increasedcapacity forIL4 andIFN-ygeneexpression.In

addition,IFN-,y and IL-2 mRNAweresignificantlymore abun-dant than IL4 mRNA inactivated neonatalCD4+CD8-

thy-mocytes and CD4+ Tcells,aswellasadult

CD4'

CD45R-T cells.Therefore, thecapacity ofT

lineage

cellstoexpress the IL4 gene may be more restricted comparedtoother

lympho-kine genesbeginning in intrathymic development.This

re-strictedcapacityappearstopersistduring postnatal extrathy-mic maturation ofTcells.(J. Clin.Invest.1991.87:194-202.).

Key words: CD45 *

interferon-'y

*interleukin4 * memoryTcells * neonatal T cells

Introduction

Secretionof

lymphokines

suchasthe interleukins(IL)IL-2 and

IL-4 and

interferon-y (IFN-T) by

activated T cells isan

essen-Addressreprintrequests to Dr.Lewis, Division of Immunology/Rheu-matology, Children'sHospital and MedicalCenter, P.O.BoxC-5371,

Seattle,WA98105.

Receivedfor publication 15 December 1989 and inrevisedform8

July1990.

tial component ofantigen-specific immunity (1-3). HowT

cellsacquire, during their intrathymic and extrathymic matura-tion, the capacity to produce particular lymphokines remains largely unknown. Recently we have shown that virtually all IL-4 and most

IFN-y

produced by polyclonally activated adult CD4+ T cells is mediated by a CD45R- subset, whereas CD45R+ and CD45R- CD4+ T cells produce comparable

amountsofIL-2(4); theCD45R- subset, which is enriched in functional memory cells and is UCHL

I',

appears tobe derived from CD45R+ (virgin) T cells which have been activated in vivo, presumably by suitably presented cognate antigen (5-9). We reasoned thatifprevious in vivo activation is a major deter-minantof the capacityforTcells to produce IL-4 andIFN-y,

thenneonatalTcells, whichpresumably have had a minimal exposure to exogenousantigens, should exhibit aprofileof lymphokine production similar to that of adult virgin T cells. In agreement withthisprediction, reduced

IFN-'y

production byneonatal T cellscomparedtoadult T cellshas beennotedby

usand other workers(10-14).We now reportthat such a selec-tivereduction in both IL-4aswellasIFN-,y geneexpression by activated neonatal T cells doesoccur.This pattern of lympho-kine productioncorrelateswitha uniformlyCD45R+ surface phenotypeonneonatalTcells, and is largely transcriptionally regulated. A similarselective reduction in IL-4 and IFN-'y compared with IL-2 gene expression was observed in CD4+, CD8-thymocytes, suggestingthat the results withneonatalT

cells may reflect in part their antigenically naive status.

Methods

Monoclonal antibodies(MAb). Thefollowingmurine anti-human MAbwereused for either cellpurificationorflowcytometricanalysis: 9.6,CD2, IgG2a (15); OKT4, CD4, IgG2b (16); 66.1, anti-CD4, IgM (17); OKT8,anti-CD8,IgG2a (18); FC-l,anti-CD16, IgM (19);3AC5,anti-CD45R(200-,220-kDisoforms), IgG2a (20, 21);and

UCHL1, anti-CD45 (180-kDisoform), IgG2a (21, 22).The 9.6 and

66.1MAbwereprovided byDr.PaulMartin,FredHutchinson Cancer ResearchCenter, Seattle, WA;FC-l and 3AC5 MAbbyDr.Edward Clark,UniversityofWashington,Seattle, WA;andUCHL1 MAbby

Dr.PeterC.L.Beverley,ImperialCancer Research Fund-Human

Tu-mourImmunologyUnit,London, UK. The OKT4 and OKT8

hybrid-omas were purchased from the AmericanType CultureCollection, Rockville,MD.RPC-5(IgG2a) and MOPC- 195 (IgG2b) murineMAb,

whichserved asisotype-specificnegative controls for flowcytometric

analysiswerepurchased from Litton Bionetics, Kensington, MDas

affinity-purifiedpreparations.Allother MAbwereused in the form of sterileascitesexceptfor UCHL1 whichwasusedas asterilehybridoma

culturesupernatant.

Cell preparations. Peripheralblood mononuclearcells (PBMC)

wereisolated from theperipheralbloodofadultvolunteersorumbilical cord bloodof healthyterm neonatesusing Ficoll-Hypaquedensity

gra-dientcentrifugation (23),and Tcellswereprepared bytreatmentof PBMCwithTcellLymphokwik, (One Lambda,LosAngeles, CA)as

previouslydescribed (4). Tcell preparationswere routinely. 97%

194 Lewisetal.

J.Clin.Invest.

©TheAmerican SocietyforClinicalInvestigation,Inc.

0021-9738/91/01/0194/09 $2.00

(3)

CD2-positiveasassessedby flow cytometry after indirect immunofluo-rescentstaining with MAb 9.6. CD4'andCD8'T cell subsetswere

purified by negative selection using MAb and complement: T cells wereincubated with saturating concentrations of either OKT8or66.1 MAb,respectively, for 30 minat4VC in RPMI-1640 medium (Medi-atech, Washington, DC) containing 2% FCS (Hyclone Laboratories, Logan,UT). The anti-CD 16 MAb, FC-Iwas included in these incuba-tionstodeplete residual NK cells. After washingat4VC,the cellswere

resuspended inRPMImedium with 20%(vol/vol) HLA-typing grade rabbit complement (Pel-Freez Biologicals, Brown Deer, WI) and incu-batedat370C forIh withgentle agitation. Remaining viable cellswere

collected by Ficoll-Hypaque density gradient centrifugation and washedtwice in RPMI medium with 2% fetal calf serum (FCS) before

use.AdultCD4'Tcellswerefurther fractionated intoCD45R'and CD45R-populationsby incubation with MAb UCHLI or3AC5,

re-spectively, followed by negativeselection with indirectpanning (24) usingplastic petridishes(BaxterScientificProducts,McGawPark,IL) previously coated with affinity-purified goat anti-mouse IgG (Tago Inc., Burlingame, CA). Human thymus sampleswereobtained from

neonatesundergoingcardiac surgery for congenital heart disease who werefree of other significant congenital anomalies. After removal of fibrous capsular tissue and grossly visible blood vessels, sampleswere

finely minced and pressed through sieves,andthe viablethymocytes

werecollected by Ficoll-Hypaque centrifugation. CD4'CD8-

single-positivethymocyteswereprepared by incubating unfractionated thy-mocytes with MAb OKT8 for 30 minat4VC inRPMImedium with 2% FCS. The cells binding MAbwerelysed usingrabbitcomplementand theremainingviable cellssubsequentlyisolatedasdescribed above for thepurification oftheCD4'andCD8' peripheralTcells. Thepurity of

all Tcell andthymocyte subsetswas>95%asdetermined by immuno-fluorescent flow cytometry after indirect staining with

appropri-ateMAb.

Tcell andthymocyteactivation. Cells(5 x 106/ml)were incubated with either 0.5

,1M

ionomycin(Calbiochem-BehringCorp., SanDiego,

CA), 25 ,ug/ml concanavalinA(Con A; Pharmacia Fine Chemicals,

Piscataway, NJ), orOKT3 (anti-CD3)MAb (18) (1:25 [vol/vol] of sterileazide-freeascites)incombination with 50ng/ml phorbol

myris-tate acemyris-tate(PMA;SigmaChemicalCo.,St.Louis,MO); these

concen-trationswerefoundtobeoptimalforlymphokineinduction in

prelimi-naryexperiments. RPMI 1640 medium containing 5% (vol/vol)

hu-manAB serumwasused in all cases except for experiments assaying

IL-4and IFN-y protein in cell culture supernatants, for which AIM-V,

acommercialserum-free medium (Gibco Laboratories, Grand Island, NY),wasemployed. Both media were supplemented with 2 mM

L-glu-tamine,50U/mlpenicillinG, and 50

Ag/ml

streptomycin.

IL-4 and

IFN-y

protein assay. T cell culture supernatants were collectedafter 24 hofionomycin andPMA treatmentand

concen-tratedusingaCentricon-1Oapparatus(AmiconCorp., Danvers, MA)

Tablel. IL-4and IFN-,y Protein Concentrationsin Supernatants

ofActivated Adult and NeonatalT Cell Cultures

Cell type IL-4 IFN-y pM

Adult

A 14.6 1762

B 34.6 4164

C 26.8 4766

D 46.0 ND

Neonate

A <3.5 30.0

B <3.5 29.5

C <3.5 26.5

D <3.5 ND

aspreviously described (4). The IL-4 and IFN-'y content of concen-trated supernatants wasdetermined byradioimmunoassay (RIA)as

described (4, 25). Thefinal IL-4 andIFN-'yconcentrations reported are calculated for cell culture supernatants before concentration, assuming 100% recovery. The recovery of samples spiked with recombinant hu-man IL-4 orIFN-yand concentrated was - 70-80%. The results for

IL-4 and IFN-,yprotein concentrations for three of these adult T cell supernatants have been reported previously (4). For both IL-4 and

IFN-y RIA determinations, neonatal and adult T cell samples were analyzed in parallel.

In vitrotranscriptionwith isolated nuclei. Transcription assays were performed as previously described (26), using nuclei isolated fromT

cells after 1.5 or 3 h of incubation with ionomycin and PMA. RNA blot analysis. Total cellular RNA was isolated by the guani-diniumthiocyanate/CsCl method (27), electrophoresed in 2.2 M form-aldehyde, 1% agarose gels, blotted, hybridized, and washed as previously described (4, 14). 32P-labeled RNA probes were synthesized from human IL-4 (28), IFN-y (29), and IL-2 (30) cDNAs subcloned into transcription vectors, using a commercially available kit (Promega Biotech, Madison, WI). The human elongation factor I-a (EF)' cDNA probe (provided by R. Perlmutter, Univeisity of Washington) was 3p labeled by the random hexamer primer method (31). The cDNAs were allfull length except forIFN-yfor which a 0.4-kb internal fragment was used(14).

In situ hybridization. In situ hybridization was performed exactly according to the method of Pardoll et al. (32), using 35S-labeled single-stranded anti-sense or sense RNA probes prepared from the IL-4,

IFN-y,and IL-2 cDNAs as previously described (4). After counter staining with Diff-Quik reagent according to the manufacturer's instructions (Dade Diagnostics, Aguada, Puerto Rico) slides were permanently mounted and coded. For each slide, the number of grains over 300 individual cells was determined by counting 75 cells in each of four randomly selected fields free of obvious artifacts. Cells were considered positive for a mRNA species if their grain counts after hybridization with the antisense probe exceeded that of 99% of cells hybridized with thecontrol sense probe.

Indirect immunofluorescentstaining andflowcytometric analysis.

Unless otherwise indicated, PBS (pH 7.4) containing 1% BSA and 0.1%

sodiumazide was used to dilute all reagents and for cell washes. Cells were washed once and then incubated with saturating concentrations of diluted sterile MAb for 30 min at 4°C. After two washes, the cells were incubated with affinity purified fluorescein isothiocyanate

(FITC)-labeledgoatanti-mouse IgG orIgM F(ab')2 fragments (Tago Inc.) diluted 1:40 (vol/vol). The cells were subsequently washed three times and fixed in 2% paraformaldehyde/PBS for 5 min at room tem-perature, and then transfered to PBS. At least 3 x I03 cells per sample were analyzed using a flow cytometer (Epics C, Coulter Electronics, Inc., Hialeah, FL) equipped with a 2-W argon laser.

Results

Decreased production of IL-4 and

IFN-,y

by neonatal T cells. After treatment with ionomycin and PMA, which induces

maximal IL-4 and

IFN-y

production by adult T cells(4),IL-4 concentrations exceeded 10 pM in cell culture supernatants from three different adult donors butwere below the 3.5 pM

limit of detectability in all neonatal samples (Table I). Unlike IL-4,

IFN-'y

wasdetectable in all neonatal T cell culture super-natants, but in concentrations(25-30 pM) that were uniformly

<10% of those measured in adult T cell supernatants (Table I), asreported previously (4, 13, 14). Nevertheless,

IFN-'y

produc-tion by neonatal T cells was significantly greater than IL-4

pro-1.Abbreviations used in this paper: EF, elongation factor 1-a; LCA,

(4)

duction on a molarbasis, as previously reported for adult T cells (4).

Decreased IL-4 and

IFN-'y

production reflects diminished gene transcription and mRNA accumulation. The production of IFN-yand IL-2byTcellsappears tobeprimarilyregulated by the rate at which these lymphokine genes aretranscribed

(33-37). Todeterminewhether decreased IL-4 and

IFN-'y

pro-tein production by neonatal T cells was also regulated inthis

manner, thetranscription rates ofthe IL-4 and IFN-y genes werecompared in cell nucleipurified from activatedadultand neonatalTcells(Fig. 1). The IL-2 gene, whose productis ex-pressed in similar amounts by adult and neonatal T cells

(I12-14), and the EF gene, whose transcript comprises - 0.5% of

mRNAinmost typesof mammalian cells (4), were transcribed atsimilarratesbynuclei fromadult and neonatal T cells and served aspositivecontrols in these and subsequent

experi-ments. IL-4 gene transcriptionbyactivatedadult T cells was

detectableasshownby the greatersignal withtheslot-blotted IL-4cDNAplasmidthanthenegative control plasmid (Fig. I B). However, the IL-4 gene appeared to betranscribedin adult Tcellsat amarkedlylower ratethanthe

IFN-'y,

IL-2,or EF genes, and, unliketheseothergenes,

required

high

concentra-tions oflabelled nuclear RNAtodetectitstranscription (Fig. 1,

Avs.B). In contrast, IL-4 genetranscriptionby neonatal Tcells

was not clearly detectable (as shown by a signalessentially equaltothatofthenegativecontrol) even whenhigh

concen-trations of labelledtranscriptswereusedforhybridization

(Fig.

IB), and therateof

IFN--y

genetranscription, although

measur-able,wasmuch lower than in adultTcells(Fig. 1A). Thus,the

1

2

1

2

EF-

w

PBS-IFNy-

-W

It2

- -am

IL4-A

if-'

B

Figure1.Transcriptionratesof theIL-4, IFN--y,IL-2,andEFgenes in nuclei from activated adult and neonatalTcells;PBSreferstothe

negative control,theplasmid,pBluescribe. (A)Nuclei from adult (lane 1)orneonatalTcells(lane2) incubated with

ionomycin

and

PMAfor 3 h. Both filterswerehybridizedwith 1.0X 106 dpm/mlof

32P-labeled transcripts. (B)Nuclei from adult(lane 1)orneonatalT

cells(lane 2) incubated with

ionomycin

and PMA for 1.5h. Both filterswerehybridizedwith 2.5 X 106 dpm/mlof32P-labeled

transcripts. Integrated densitometryof the bands done witha

densitometer(Visage60,BioImage,AnnArbor,

MI)

aftersubtraction

of the values for the PBSnegative control,gave the

following

results

(arbitrary units): experimentA-adult(lane 1)EF7.0,IFN-'y

2.3,

IL-25.8,IL-4 -; neonatal(lane2)EF 10.1, IFN-,y 0.5,IL-26.5,IL-4 -;experimentB-adult(lane1)IL-4 1.03,IL-2 11.93;neonatal

(lane2)IL-40.16,IL-2 6.35.

transcription

ratesofthe IL-4 and IFN--y genes in adult and

neonatal Tcells correspondedwell withtheamountsofIL-4 andIFN-y proteinthese cellssecreted,

suggesting

that produc-tionof these

lymphokines

was

largely regulated

at a

transcrip-tionallevel in both cell types.

Thedecreased

transcription

ofthe IL-4 and

IFN-'y

genesby neonatal T cellswas associated with diminished steady-state levels of these

lymphokine

mRNAs

(Fig.

2A). AdultT cells contained markedlymoreIL-4 mRNAthandid neonatal cells, in whichsuch

transcripts

wereinsome cases

completely

unde-tectable

(Fig.

2,AandB).

IFN-'y

mRNAwas

routinely

measur-ableafterashorterperiod ofexposureofautoradiographsthan wasnecessaryforthedetection ofIL-4mRNA,consistent with

therelativeconcentrations ofthese

lymphokines

in culture

su-pernatants. AlthoughTcellsfrom different individuals varied

intheirlevelsofIL-2mRNA,ingeneral,theamountsofthese

transcripts in adult and neonatal T cellswere comparable as

previously

reported (4, 13, 14). Similarresultsfor IL-4,

IFN--y,

and IL-2 mRNAaccumulationwerealsoobserved

using

Tcells fromanadditional sixadult andsixneonatalsubjects (datanot

shown), andwhenotherstimuliwereused

(either

ConAand PMAoranti-CD3 mAband PMA-see

Fig.

2,Band

C),

indi-cating

that thesedifferenceswerenot

unique

to

particular

indi-vidualsortheconditionsusedforcellactivation.Because these

other stimuliwere

generally

less effectivethan

ionomycin

and

PMA, in subsequent

experiments

cells were stimulated with

ionomycin

and PMA to maximizedetection ofthe IL-4and IFN-,y geneproducts.

ThemarkedlyreducedlevelsofIL-4and IFN-,y mRNA in

neonatal T cells were not attributable to kinetic differences

between adult and neonatalcellsin theaccumulation of these

transcripts

(Fig.

2

C).

Inboth cell types,

lymphokine

mRNA

wasundetectablebeforeinvitro activation

(lanes

I and

8)

and

peak

levels of

IL-4, IFN-y,

and IL-2 mRNAwereachieved

by

- 6 hof incubation with

ionomycin

and PMAorConAand

PMA.InneonatalT

cells,

IL-4mRNA,if

measurable,

was

only

observedat6 hofincubationandnot atlatertimes

(data

not

shown). After

hybridization

with a

full-length

IL-2 cDNA

probeasshown in

Fig.

2C,

transcripts

wereobservedat24 hof incubation in neonatal and adult Tcells,which had alower apparent molecular

weight

than thosewhich accumulated

by

6 hof incubation (lanes

4, 7,

and

11).

However, these

late-ap-pearing transcripts

wereundetectableafter

hybridization

with

an IL-2 cDNA

probe

which lacked the last 100

bp

ofthe 3'

untranslated

region

but included the entire

coding

sequence

(30),whereas the

earlier-appearing transcripts

werestillpresent

(data

not

shown). Thus,

these

late-appearing transcripts

pre-sumably

didnotencode

IL-2,

but

cross-hybridized

witha

por-tionofthe 3'untranslated

region

of the IL-2 gene. EF mRNA

waspresent in activatedneonataland adultTcells insimilar

amounts,

indicating

thatthedifferencesobserved in

lympho-kinemRNAaccumulationwerenotattributabletodifferences

in theamountofRNAanalyzedin each lane.

Expression

of

IL-4, IFN-,y, andIL-2mRNA

by

Tcell

sub-sets.Because IL-4mRNAwasmeasurable

by

RNA

blotting

in

atleast someactivated neonatal Tcell samplesunder condi-tionsinwhichIL-4 gene

transcription

orsecretedIL-4

protein

wasundetectable,weused thismoresensitive

technique

to

compare

lymphokine

gene

expression by

adult and neonatal

CD4+

and

CD8+

Tcellsafter

ionomycin

and PMAactivation

(Fig.

3). Inagreement with

previous

observations (4), among adultTcells, the

CD4'

subsetwasenriched forIL-2 and IL-4

(5)

A

Exposure

Probe

tme

(h

)

1

2

3

4

5

B

C

1

2

3

4

1

2 3

4

5 6

7

8 9 10 11 12 13 14

I1l4 72-120 _

IFNy 2-5 *

$

* *

IL-2 4-8 _

ii,

II,

-il

EF 12-16

Figure 2. Total RNAsamples hybridized with IL-4,IFN-y,IL-2, and EF probes. All blotsarerepresentative ofatleast three different experiments. (A) T cellslymphokine mRNA accumulation after ionomycin and PMA incubation for 6 h. Lanes 1-3, adult T cells from three differentindividuals. Lanes 4 and 5, neonatal T cells fromtwodifferent individuals. (B) Lymphokine mRNA accumulation in adultvs.

neonatal T cells under different conditions of activation. Lanes I and 2, adult T cells incubated for 6 h with ionomycin and PMAorOKT3 and PMA, respectively. Lanes 3 and 4, neonatal T cells incubated for 6 h with ionomycin and PMAorOKT3 and PMA, respectively. (C) Kinetics

oflymphokine mRNAaccumulation in adultvs.neonatal T cells. Lanes 1-4, adult T cells incubated for 0, 6, 12, and 24 h, respectively, with

ConAand PMA. Lanes5-7, adultT cells incubated for 6, 12, and 24 h, respectively, with ionomycin and PMA. Lanes 8-11, neonatal T cells incubated for0, 6, 12, and 24 h, respectively, with Con A and PMA. Lanes 12-14, neonatal T cells incubated for 6, 12, and 24 h, respectively, with ionomycin and PMA. All laneswereloadedwith 5 ,g of total RNA,exceptfor lanes 3 and 4 of B in which10 ugof total RNAwasloaded.

mRNA,whereas theCD8' subsetwasenriched for

IFN-,y

tran-scripts (lanes1 and2).The markedreduction inIL-4 mRNA, but similaramounts ofIL-2 mRNA observed in

unfraction-Exposure

time

(h)

120

1

2 3 4

5 6 7

_

12

4

6

2

8

_

16

0 0

to"w

Figure3.Total RNAsamplesfrom subsetsof adultand neonatal T

cells andneonatal thymocytes after incubation with ionomycinand PMAfor 6h, hybridized with IL-4, IFN-y, IL-2,and EF probes. This blot isrepresentative oftwo separateexperiments. Lanes I and 2, adultCD4+ and CD8+ T cells, respectively. Lanes 3 and 4, adult

CD4+ CD45R+ and CD4+ CD45R- T cells, respectively.Lanes5and

6, neonatal CD4+ and CD8+ T cells, respectively.Lane 7, neonatal

CD4+CD8-thymocytes. All laneswereloaded with5ggoftotal RNA.

ated neonatal Tcellscomparedtoadultcells,wasalso found in

purifiedneonatal

CD4'

Tcells(lane1 vs.lane5).Like adult T cells, the

CD8'

subsetof neonatalTcellswasenriched in

IFN-y mRNAand the

CD4'

subsetwasenriched forIL-2 mRNA,

butdifferencesbetween

CD4'

and

CD8'

neonatalTcellsand between

CD8'

adult andneonatalTcells in IL-4 mRNA levels couldnotbe evaluated because of their very lowto undetect-ableamountsofIL-4transcripts.

Therelativelylow levelsofIL-4and

IFN-'y

mRNA in acti-vated neonatalTcellscould beexplainedbyeitherareduced numberofcells expressing these mRNAs or a reduced levelof

IL-4 orIFN--y transcriptsper cell orboth. Todistinguish

be-tweenthese

possibilities,

we

performed

in situ hybridization.

Results representative ofone ofthree such experiments are shown in Fig. 4 and Table II. Among activated adult cells,

IFN-yand IL-2 mRNAwasexpressed by - 40%ofTcells, and

inagreaterpercentageof CD8+andCD4+ T cellsrespectively,

whereas IL-4 mRNAexpressionappearedrestrictedto asmall

fraction(<5%)of CD4+Tcells, aspreviously described(4). In

striking

contrast, IL-4 mRNAwasundetectablein all neonatal Tcellfractions (Fig.4andTableI),evenafter autoradiographic

exposureofslidesforupto5 wk(datanotshown).Allneonatal

Tcell fractionsalso hadmarkedly fewercellswithdetectable

IFN-'y

mRNA. Inthose neonatalCD4+ orCD8+Tcells which werepositive for

IFN-'y

transcripts, the average amount of IFN-,ymRNApercell, as assessed bygraincounts, was lower than in the adult cellpopulation (Fig. 4,datanotshown). Like

adultcells,neonatalCD4+andCD8+Tcellswereenrichedfor

IL-2 and

IFN-'y

transcripts respectively, compared to

unfrac-tionatedT cells, in agreement withthe resultsobtainedby RNAblotting.

S.

I.

Probe

It-4

IFNr

It-2

(6)

A

*

.. 7

211

X * ...

*s *ff - #;

50 40

~30-20 10

10

0 10 20 30

Grains /Cell

C ' 9-<*, ,~~~~~~W

A_

50

w

40 /I a) 6 11

Cu I

0 20-

10-0

&-0 10 20 30

Grains/CeIl

Ps ts ~~*

* k .

#3i

50

40-U)

w 30

0 20

0-

I

10 n

I

I1

I

0 10 20

Groins /Cell

30

.1

UL)

a)

02 >0

10 30 10 II/

0 10 20 30

Grains /Cell

Figure 4. AdultorneonatalTcellsafter incubation with ionomycin andPMA for 6 h,hybridizedinsitu forIL-4andIFN--ymRNA.

Photomicrographs(X24) of adult (A and C)orneonatal (B andD)T

cellshybridizedwithananti-sense probe forIL-4(A andB)orIFN-y

(C and D).Directlybeloweachphotomicrographis thecorresponding histogramof the number ofgrainsper cell afterhybridizationwith

anantisense probe (-) for the slide depicted,orwithacontrol sense probe (o) by use of a duplicate slide processed in parallel.Allslides were exposed for7dbefore development.

Expression

of

IL-4,

IFN-Ty,

and IL-2mRNA

by

CD45R+

or

CD45R-CD4+

Tcells. Because IL-4

production by

adult T cellsappearedtobe almost

exclusively

limitedto asmall frac-tion of CD4+Tcellswhichwasundetectable intheperipheral

bloodofneonates,wesoughttofurtherdefinethispopulation. Previous studieshave shown thatCD4+Tcells that augment B cell

immunoglobulin production

are enrichedinproliferative

responsestorecallantigensand express the 180-kDisoformof

the leukocyte common

antigen

(LCA)

(recognized

by MAb UCHL

1),

whereas themutually exclusivesubsetof

CD4+

cells

that expresses the 200- and 220-kD isoforms of the LCA, CD45R (recognized byMAb3AC5),areineffectivein provid-ingsuchhelpandarenotenrichedinfunctionalmemorycells

(6-9,

38,

39).

ThesetwoLCAisoforms donotappearto

iden-tify

distinct

lineages

but rather T cells which have

(CD45R-,

Table II. IL-4, IFN--y, and IL-2 mRNA Expression in Adult orNeonatal T Cells and T Cell Subsets after Activation

with Ionomycin andPMAfor6h

Cellspositivefor mRNA Cell type IL-4 IFN-y IL-2

UnfractionatedTcells

Adult 3 41 42

Neonate <1 3 35

CD4'T cells

Adult 4 34 51

Neonate <1 2 45

CD8'Tcells

Adult 1 58 20

Neonate <1 4 16

The resultspresented arerepresentative of three (unfractionated T cells)ortwo(CD4'andCD8'Tcells) separateexperiments. Slides hybridized with the IL-4 probe were exposed for14d, whereas those hybridized with theIFN-,yand IL-2 probes were exposedfor7d.

UCHL1I) orhave not

(CD45R',

UCHLI-) been previously primed bytheir exposure to cognate antigen in vivo (7-9,40,

41), althoughthisinterpretationhas been disputedby others (42). TheCD45R- subpopulation of adult CD4' Tcells is enrichedinthe production of IL-4 as well as

IFN-y

(4). Re-cently others have found that nearly all neonatal T cells are

CD45R+, UCHLI- (5, 43, 44). By immunofluorescent flow cytometryweconfirmed that bothCD4+andCD8+neonatal T cellswerealmost all CD45R (3AC5)-positive and UCHLI -neg-ativeordull;in contrast, similarpopulations ofadult T cells demonstrated the previously described (7-9, 40, 41) bimodal patternof surfacestaining: - 40%ofcells expressed UCHL1

and - 60% of cells expressed CD45R inamutually exclusive manner(data notshown).

Thus,todirectlycomparetheirlymphokinemRNA expres-sion with thatbyneonatalCD4+Tcells, adultCD4+Tcells were fractionated into CD45R+ (UCHL1-) and

CD45R-(UCHLI+)subsets.Cellpurificationwasaccomplished by nega-tiveselectionusingMAb3AC5andUCHL1, sincethe activa-tioncharacteristics of T cellscanbe alteredbytreatmentwith MAb directed against LCA (19). IL-4 mRNA expression in adultCD4+Tcellswasvirtuallylimitedtothe CD45R- frac-tion andIFN-,ymRNAexpressionwasalsohighlyenrichedin

this fraction(Fig. 3,lanes I and4).Activated neonatal CD4+ and adultCD4+ CD45R+ T cells contained similarly lowto

undetectableamountsof IL-4 mRNA andmarkedlylower lev-elsofIFN-ymRNA(lanes3 and5)comparedtothe CD45R-fraction ofadultCD4+Tcells(lane 4).SimilaramountsofIL-2

mRNA werefoundin theCD45R+and CD45R-fractions of

adultCD4+Tcells,aswellasin neonatal CD4+ T cells.

Production

ofIL-4,

IFN-,y, andIL-2mRNA

by

CD4+

CD8-thymocytes. These resultssuggested that diminished produc-tion ofIL-4 and

IFN-'y

by neonatal CD4+ andadult CD4+

CD45R+

Tcells

might

reflecttheir

antigenically

naiveorvirgin

status. Ifthiswere so, then a similarpattern of

lymphokine

expression

might

be

expected

for

CD4+

CD8-

thymocytes,

a

relatively mature populationwhich includesCD4+T lineage

(7)

cells destined to emigrate into the circulation

(45).

Thiswas

largely true, in that

CD4'

CD8- neonatal thymocytes

com-pletelylackeddetectable IL-4 mRNAand had lowamountsof

IFN-'y

mRNA similar to those of neonatal

CD41

Tcells. In

addition, this thymocytesubset had slightlyreducedlevels of IL-2 andEF mRNAcomparedwithadultorneonatal

CD4'

T cells (Fig. 3, lanes 5 and 7). The staining ofthis filter with methylene blueafterhybridizationswerecompletedindicated thatequivalentamountsofribosomal RNAwerepresent in all lanes (data not shown). Similar results were obtained using CD4'CD8- thymocytesfromolderchildren (datanotshown).

Discussion

We found that IL-4productionbyunfractionated and

CD4'

neonatalTcells wasmarkedlyandconsistentlyreduced

com-pared to adult cellsafter polyclonal activation.Wealso found

that IFN-,y production both by

CD4'

and

CD8'

neonatal T cellswasdiminished comparedtoadultcells,as

previously

re-ported for unfractionated T cells (4,

13,

14). The decreased IL-4 and

IFN--y

productionis selective inthatproduction by

neonatal and adult T cellsofIL-2(thisreport andreferences 4,

and 12-14)andoftumornecrosis factor-aand

-/B

and the IL-2 receptorp55 chain(13, 46)arecomparable.IL-4

production

by

neonatal Tcells wasusuallyundetectable, apparentlybecause

ofthe complete absenceofa

relatively

rare

subpopulation

pres-entincirculatingadultTcells thatproduced this

lymphokine.

IFN-y-producing neonatal Tcells, although detectable, were

decreased in number and produced lower amountsof

IFN-'y

per cellcomparedtoadultTcells. DecreasedIL-4and

IFN-'y

production by neonatal CD4+ andCD8+ Tcells correlated with the complete absence of CD45R- T cells which are highly

enriched forIL-4 and

IFN-y-producing

cells(4). Theresults obtained for lymphokine production byneonatalCD4+ CD8-thymocytes paralleled thosefor circulatingneonatalCD4+ T

cells, and suggested that absent IL-4productionandreduced

IFN--y productionby neonatalTcells wasafunctional pheno-type established during intrathymic development. The unde-tectableamountofIL-4andreduced amountof IFN-y protein

secretedbyneonatalTcellswasparalleled byproportional

de-creasesin thetranscriptionratesofthe IL-4 and

IFN-'y

genes as well as accumulation of their cognate mRNA, indicating that

thisdecreased

lymphokine

productionwasprimarily transcrip-tionally mediated.

Inpreviousreports,reduced

IFN-'y

productionbymitogen activated neonatal leukocytesor Tcells has beenvariously

at-tributed to suppression mediated by radiosensitive CD4+ or

CD8+Tcellpopulations(47,48).However, ourstudies of IFN--y mRNA expression in bulk cultures or at theindividualcell leveldidnotcorroborate thesefindingsin that purified CD4+

orCD8+ neonatal Tcellsdid notaccumulate substantially

moreIFN-,yorIL-4mRNAthandidunfractionated neonatal

Tcells.Althoughourresultsalso do not exclude the possibility ofsuppression ofneonatal T cell

IFN-'y

andIL-4production by

a population found in both the CD4+ and CD8+ subsets, a moredirect explanation is that reduced expression of these

lymphokines isafeature intrinsictonaive T cells or, at least, T cellsatthisstage ofdevelopment.

There isincreasingevidence that memory T cells, defined

asthosewhich have previously beenactivatedor"primed" by antigen,arecapable of increased levels of

Iymphokine

produc-tion,

compared

to

antigenically

naive, virgin

T cells. T cell

fractionationstudies

making

useof themutuallyexclusive

ex-pression

of CD45Randthe 180-kD LCA isoform

(recognized

by

MAb UCHL1 in

humans),

ormarkerscoexpressedwith the 180-kD isoform,such as 4B4 (CDw29), LFA-3, and in the mouse,PgP-1 (CD44),have demonstrated that

postimmuniza-tion

proliferative

recall responsestosolubleantigens,

viruses,

and

alloantigens

aremediated

by

CD45R- T cell

populations

(6-9,

38,

49,

50).

In humans, this memory T cell-enriched

CD45R-subsetaccountsforvirtuallyallIL-4 andmost

IFN-y

producedby polyclonallyactivatedadult

CD4'

T cells (4), and IL-4 production byadult

CD8'

Tcells also appearstobe

re-strictedto a CD45R- subpopulation (D. B. Lewis and C. B. Wilson, unpublished observations). Similarly in mice,

CD4'

or

CD8'

Tcellswith high surface levels ofPgP-I are markedly enrichedin

IFN-,y

production comparedtoPgP-1 dullor

un-fractionatedTcells(51). Thisenhanced lymphokine

produc-tion is selectivein that

CD45R'

andCD45R-, orPgP-l dull

and PgP-1 brightTcellpopulations producesimilaramounts

ofIL-2(4,9,50).The observation originally made by Tedderet

al.(5)and confirmedbyothers(43,44),thathuman neonatal T

cellsare uniformly

CD45R'

whereas circulating adult T cells

are- 60%

CD45R',

led themtopropose that

CD45R'

Tcells

areconverted to aCD45R- surface phenotype bya post-thy-mic maturationalprocessinvolving,atleast in part, T cell

acti-vation by cognate antigen (6). The present study confirmed that both CD4+ and CD8+ neonatal T cells had a CD45R+ UCHLI-phenotype, consistent with their presumed

antigeni-callynaive status. This model is also supported by the fact that human neonatal or adult CD45R+ T cells polyclonally acti-vatedin vitro acquire a CD45R-UCHL1+ surface phenotype

(7, 9, 40, 41) and by recent adoptive transfer studies in therat

directlydemonstrating that CD45R+ T cells are the precursors

ofthe antigen specific CD45R- memory population (52).

Taken together, thesefindings strongly suggest that the similar amountof IL-2, but markedly diminished amounts of IL-4 and

IFN-,y

produced by neonatal T cells compared to adult T cells

is due to the absence of an antigenically primed, or at least previously activated, memory T cell subpopulation.

Studies of postnatal T cell maturation also supportamodel in which increased lymphokine production is induced by

anti-genic priming. Mitogen-induced

IFN-'y

production by

periph-eral leukocytes increases with postnatal age, as does the

num-ber ofcirculatingCD45R- T cells (43, 53), although a direct connection between these events has not been established. Ad-ditional evidence comes from longitudinal studies of

IFN--y

production in humans after neonatal or adult infection with herpes simplex virus (HSV): peripheral blood mononuclear cells from 2-mo-old infants who were infected as neonates, produced as much

IFN-'y

in response to HSV antigen as cells

from adults 2 moafterprimary HSV infection, suggesting that once neonates developed T cells responsive to recall HSV anti-gens, the capacity of these cells to secrete IFN-'y was similar to that of T cells responding to the same antigen from infected adults (Burchett, S. K., K. Mohan, J. Westall, R. Ashley, L. Corey, and C. B. Wilson, manuscript submitted for publica-tion). In cells from these infants,

IFN-,y

production after HSV antigen stimulation was equivalent to that observed after poly-clonal activation with Con A, consistent with the notion that virtually all T cell-mediated

IFN--y

production was by memory cells. In contrast, adult T cells produced much larger amounts of IFN-y after Con A treatment than after HSV antigen

(8)

num-bers ofmemory T cellsdirected against antigens other than HSVthat would be expected to be present in adults but not in neonates.Whether antigen-specific IFN-,y production after neonatal or adult infection is predominantly mediated by

CD45R-Tcells, andantigen-specificIL-4production parallels that for

IFN-,y

remains to be determined.

Although the above studies suggest that humans can de-velop functional memory T cells as early as 2 mo of age, they do notexclude the possibilitythatthis ability is impaired at

birthbutrapidlymaturesduringpostnatal development.

Sev-eral observations areconsistent with this idea. First, in cells

frominfants infectedwith HSV asneonates, peak

IFN-y

pro-duction

by

memory T cells doesnot occuruntil - 6 wk after

infection,whereas adult cells achieve peak responses after only 2weeks (Burchett, S.K., K. Mohan, J. Westall, R. Ashley,L.

Corey, and C. B. Wilson, manuscript submitted for publica-tion). Whetherthis delay reflects differences in events

occur-ringproximalto T cellactivation,such as inantigen-processing andpresentation, or reflectsa decrease inantigen-specific T

cells ortheir ability to be clonally expanded, remains to be

determined. Inprevious studies,theantigen-presenting ability ofneonatal and adult mononuclear cellsinvitrohasbeen

com-parable (54). Secondly, immunization of full-term infants with T-dependent antigens is often ineffective duringthefirst 2 wk

oflife, but results in significant antibodyresponses when

per-formedat 2 moofage(55). Although decreased neonatalBcell

functionorinterference bymaternalantibodieshas been

pro-posed to account for these results, a maturational defect in cognatehelp by Tcellsisplausible. Finally, neonatal T cells

obtained atbirthexpress uniformly high levels of the CD38 antigen (recognized by MAb OKT10), which isalsofoundon

virtuallyall thymocytes but only very small numbersof circu-latingadult

CD45R'

Tcells(56, 57, and D. B.LewisandC. B.

Wilson,unpublishedobservations), suggestingthat they are,in

part, animmature, transitional

population.

Thekineticsof

dis-appearance for circulating

CD38'

T cellspostnatally, and whetherthis population acquires a CD38- phenotype, or is replacedby a newpopulation of CD38-Tcells, remainstobe

determined.Becausesurface

expression

oftheCD38

antigen

is also increased after T cell activation

(58),

this may limit its usefulness asa markerwith whichtofollow the

extrathymic

maturation of neonatalT cells.

A directcomparison ofthe ability of neonatal and adult

CD45R'

T cells to develop into functional memory T cells

would, therefore, beof interest. One approach istocompare

lymphokine production

by neonataland adult

CD45R'

popu-lations afterinvitropriming. Priming of adult murineTcells

using antigens, mitogens,

oranti-CD3 MAb increasesIL-4and

IFN--y,

but not IL-2

production;

thisappearsto mimicthein vivoprocessby which antigen activation selectivelyenhances

lymphokine

production

(59-62).Thesestudiesareinprogress.

Stimulated neonatal CD4+CD8-

thymocytes

contained

un-detectableamountsofIL-4 mRNA, and low levelsof IFN--y

mRNAcomparedtoadultCD45R-CD4+Tcells,apatternof lymphokine mRNA expression similartothatofcirculating neonatalCD4+oradult

CD45R'

CD4+Tcells. Inaddition,the greater capacity for IL-2 productionbyCD4+ and for

IFN-'y

production by CD8+

circulating

neonatal T cells, isalso ob-servedfortheCD4+CD8- and CD4-

CD8+

subsetsofhuman

thymocytes,

respectively

(D. B.

Lewis,

unpublished

observa-tions). Although

these

observations

suggest thatthe

capacity

for

lymphokine

gene

expression

by

circulating virgin

T cell

populations is largely determined during intrathymic differen-tiation,

CD4'

CD8- thymocytes also had modestly reduced

levelsofIL-2 mRNAcompared tocirculating

CD45R' CD4'

Tcells.Thismayreflectthe knownheterogeneity within CD4' CD8-thymocytes (41,45,63)oradditionalimmaturity in lym-phokine productionbythis population.

The molecular mechanism fordecreased IL-4 and IFN-,y genetranscriptionandproteinproduction byneonatal or adult

CD45R'

T cellscomparedtotheadult CD45R-Tcellsubset

remainstobedefined.Theintracellulareventsinvolved inthe

induction of lymphokine gene transcription in T cells have been bestcharacterized forthe IL-2 gene, and suggest that two

signals-an elevationinthe intracellular concentration of

cal-cium

[Ca2+]

and activation of protein kinase C (PKC-are required(64), although aPKC-independentpathway may also

exist (65). Neonataland adult T cellsappearedtosharesimilar requirements fortheinduction ofall threeoftheselymphokine

transcripts:

inthepresenceof PMA,adirect PKCactivator,the

calcium ionophore ionomycinwasconsistently superiorto

ConA,which,in turn,was moreefficientthananti-CD3 MAb for inducing IL-2, IFN-y,and IL-4 mRNA in either adultor

neonatal T cells. These results suggest that the initial signal transduction pathways for IL-2, IL-4,and

IFN-'y

gene

expres-sion are similar in neonatal and adult T cells. Because IL-2 mRNA expression was comparable in neonatal and adult T

cells under theconditions of activationweemployed,and

be-causereduced IL-4 andIFN-ygene

transcription

wasseenwith stimulithatwouldbypass earlysteps inthissignal pathway,the

differences observed may bemediated by alternative signal transduction events or events

occurring

after elevation of

[Ca2+]

andPKCactivation.Itisalso

possible

that reduced

tran-scription might

result from alterations in the interaction of trans-acting factors with regulatory regions ofthe IL-4and IFN-y genes. These

possibilities

are notmutually exclusiveand

are avenuesfor future

investigation.

We can onlyspeculateas to thepotential advantagetothe immune systemoflimiting most IFN-,y

production

and

vir-tuallyall IL-4

production

to amemory T cell

population.

Possi-bly, this might focustheeffects ofthese

lymphokines

onto suit-ably

responsive

effector cells,suchasBcellsor

cytotoxic

Tcells whichhave also beenpreviously primed by antigen.Itis

inter-esting

in this

regard

that exogenous IL-4 can inhibit B cell

proliferationand the

generation

ofTcell-mediated

cytotoxicity

when added to in vitro

lymphocyte

cultures at early

times,

whilelateraddition ofIL-4is actually stimulatory;in contrast, IL-2additiondoesnothavethistime-dependent

inhibitory

ef-fect (28, 66, 67).

Expression

of

high

amountsofIL-4or

IFN-,y

byTcells

during

the

primary

immuneresponseor

during

in-trathymic development

could have otherdeleterious

conse-quences. For

example,

IL-4andIFN-y

by

virtueoftheir

ability

toincreaseclass II MHC

expression

ona

variety

of

hematopoi-etic celltypes(1,

2),

could

predispose

toautoimmune

reactivity

at a critical stage of

immunological development.

Future in

vivoexperimentsinwhichIL-4or

IFN--y

are

overexpressed

by

Tlineagecells

during

earlyontogeny may

provide insight

as to

the

physiologic significance

of

restricting

most

production

of

these

lymphokines

to

secondary

Tcell responses.

WhereasbothIL-4andIFN-y

production

by

peripheral

T

cells appear dependent on whether these cells have been

previously activatedin

vivo,

ourresultsalsoindicatethat the

capacity

for IL-4

production

was

remarkably

restricted

com-pared

toIFN-y

throughout

Tcellontogeny. IL-4 mRNAwas

(9)

undetectable ormarkedly less abundant than

IFN-'y

or IL-2 mRNAinactivated

CD4'

CD8-thymocytes, as well as

circu-lating neonatal and adult T cells, and previous studies have demonstrated that this is also true for lymphoid tissue-asso-ciated T cells from mice orhumans(4,68).Based ontheresults

ofinsituhybridization for lymphokinemRNA,thislow level

ofproduction was due to the rarity of IL-4producing cells. Thesefindings suggestthatduringintrathymicaswellas

ex-trathymicTcellmaturation, the capacity forTcells to produce IL-4 is relatively restricted, indicating that IL-4 may play a more specialized role in mediating T cell effector functions

than does

IFN--y

or IL-2. Itremainstobe determined if

particu-lar Tcellsarepermanentlycommittedtoproduce IL-4and/or

IFN-y

and the extent towhich production ofthese

lympho-kinesbyprimaryTcells ismutually exclusive,asituation

com-monly observedamonglong-termmurine

CD4'

Tcellclones, butnotamongshort-termmurine or mosthuman

CD4'

T cell clones(69-73). Recently, mutually exclusive productionof IL-4andIL-2 hasbeen demonstrated forprimary murine

CD4'

splenicTcells, fractionatedonthe basis ofreactivitywith mAb

againsttwouncharacterizedTlineage determinants (74, 75).It will be ofinterest to determinewhich ofthese

CD4'

splenocyte

populations secrete

IFN-'y,

andwhether analogoussurface markerswillalsoidentifyhumanprimaryTcellswith specific

patternsoflymphokine production.

Acknowledaments

We thank Janice Hall and Colleen

McCluskey

for

performing

flow

cytometric analysisandMichael Weaver for his technical assistancein

carrying

outnuclear

transcription

andinsitu

hybridization

assayand

BarbaraLovseth for secretarial assistance.

This workwas

supported by

grants

AI/HD-26940, HD-06706,

HD-18184, and AI-16760 from the National Institutes of Health andby

grant R-336-83 from the United Cerebral

Palsy

Foundation.

References

1.Paul,W.E.,andJ.Ohara. 1987. B-cellstimulatory factor-l/interleukin4. Annu.Rev.Immunol.5:429-475.

2. Trinchieri, G.,and B. Perussia. 1985. Immune interferon:apleiotropic

lymphokinewithmultipleeffects. Immunol.Today.6:131-136.

3.Paul,W.E.1989.Pleiotropyandredundancy:Tcell-derivedlymphokines

in the immune response.Cell.57:521-524.

4.Lewis,D.B.,K. S.Prickett,A.Larsen,K.Grabstein,M.Weaver,andC. B. Wilson. 1988.Restrictedproductionof interleukin-4byactivatedhuman T cells. Proc.Nail. Acad. Sci. USA. 85:9743-9747.

5.Tedder,T.F.,L. T.Clement,andM. D.Cooper.1985. Humanlymphocyte

differentiationantigensHB-10 and HB-I 1.I.Ontogenyofantigen expression.J. Immunol. 134:2983-2988.

6.Tedder,T.F.,M. D.Cooper,andL.T.Clement.1985. Humanlymphocyte

differentiationantigensHB-10 and HB-11.II.Differentialproductionof B cell

growthanddifferentiation factorsbydistincthelperTcellsubpopulations.J. Immunol. 134:2989-2994.

7.Sanders,M.E.,M. W.Makoba,S.O.Sharrow,D.Stephany,T.A.Springer,

H. A.Young,and S.Shaw. 1988. Human memoryTlymphocytesexpress in-creasedlevels of three cell adhesion molecules(LFA-3, CD2,andLFA-I)and three other moleucles(UCHL1,CDw29,andPgp- I)and have enhanced IFN-y

production.J.Immunol. 140:1401-1407.

8.Sanders,M.E.,M. W.Makgoba,and S. Shaw. 1988. Human naive and memoryTcells:Reinterpretationofhelper-inducerandsuppressor-inducer

sub-sets.Immunol.Today.9:195-199.

9.Akbar,A.N.,L.Terry,A.Timms,P.C.L.Beverley,and G.Janossy.1988. Lossof CD4SRandgainof UCHLIreactivityisafeatureofprimedTcells. J. Immunol. 140:2171-2178.

10.Bryson,Y.J.,H.S.Winter,S.E.Gard,T.J.Fischer,and E. R.Stiehm. 1980.Deficiencyof immune interferonproduction by leukocytesofnormal new-borns.Cell.Immunol.55:191-200.

11.Wakasugi, N., and J.-L. Verelizier. 1985. DefectiveIFN-'yproduction in the human neonate. I. Dysregulation rather than intrinsic abnormality. J. Im-munol. 134:167-171.

12. Miyawaki, T., Seki,H.,Taga, K., Sato, H.,andTaniguchi, N. 1985. Dissociated production ofinterleukin-2 and immune('y)interferon by

phytohae-magglutinin stimulated lymphocytes in healthy infants. Clin. Exp. Immunol. 59:505-511.

13. Wilson, C. B., J. Westall, L. Johnston, D. B. Lewis, S. K. Dower, and A. R. Alpert. 1986. Decreasedproduction of interferon gamma by human neonatal cells: Intrinsic and regulatory deficiencies. J. Clin. Invest. 77:860-867.

14.Lewis,D.B.,A.Larsen, andC. B. Wilson. 1986. Reduced interferon-gammamRNA levels in human neonates: evidence for an intrinsic T cell defi-ciency independent of other genes involved in T cell activation. J. Exp. Med. 163:1018-1023.

15. Kamoun, M., P. J.Martin, J. A. Hansen, M. A. Brown, A. W.Diadak,and R. C.Nowinski. 1981. Identification of a human T lymphocyte surface protein associated with the E-rosette receptor. J. Exp. Med. 153:207-212.

16. Reinherz, E. L., P. C. Kung, G. Goldstein, and S. F. Schlossman. 1979. Separation of functional subsets of human T cells by a monoclonal antibody. Proc.Natl. Acad. Sci. USA. 76:4061-4065.

17.Hansen, J. A., P. J. Martin, P. G. Beatty, E. A. Clark, and J. A. Ledbetter. 1984. In LeucocyteTyping. A. Bernard, L.Boumsell, J. Pausset, C. Milstein, and S. F.Schlossman, editors.Springer-Verlag, Inc., New York. 195-212.

18.Reinherz, E. L., P. C. Kung, G. Goldstein, R. H. Lovey, and S. F. Schloss-man.1980.Discrete stages of human intrathymic differentiation: analysis of nor-malthymocytes andleukemiclymphoblasts of T-celllineage.Proc.Natl. Acad. Sci. USA. 77:1588-1592.

19.Ledbetter, J. A., L. M. Rose, C. E. Spooner, P. G. Beatty, P. J. Martin, and E. A.Clark. 1985. Antibodies to common leukocyte antigen p220 influence hu-man Tcellproliferationby modifying IL2 receptor expression. J. Immunol. 135:1819-1824.

20.Rose, L. M., A. H.Ginsberg, T. L. Rothstein, J.A.Ledbetter, and E. A. Clark. 1985.Selective loss ofa subset of T helper cells in active multiple sclerosis. Proc.Natl.Acad. Sci. USA. 2:7389-7393.

21.McMichael, A. J., and F. Gotch. 1987. T cellantigens: New and previously defined clusters. InLeucocyteTyping III. A. J. McMichael, editor. Oxford Univer-sity,Oxford,England. 57-58.

22.Smith, S. H., M. H. Brown, D. Rowe, R. E.Callard, and P. C. L. Beverley. 1986.Functional subsetsof humanhelper-inducer cells defined by a new mono-clonalantibody,UCHLI.Immunology.58:63-70.

23.Wilson,C. B., and J. S.Remington. 1979. Effects of monocytes from human neonatesonlymphocyte transformation. Clin. Exp. Immunol. 36:511-520.

24.Maryanski, J. L., E. De Plaen, and J. Van Snick. 1985.Asimplepanning methodfor the selectionof cellsurfaceantigentransfectants.J.Immunol. Meth-ods. 79:159-165.

25.Chang, T. W., S. McKinney, V. Liu, P. C. Kung, J. Vilcek, andJ. Le. 1984. Useof monoclonal antibodiesassensitive and specific probes for biologically active human y-interferon.Proc. NatL. Acad.Sci. USA. 81:5219-5222.

26. Burchett, S. K., W. M. Weaver, J. A. Westall, A. Larsen, S.Kronheim,and C. B.Wilson.1988. Regulationoftumornecrosis factor/cachectin and IL- I secre-tion in humanmononuclear phagocytes.J.Immunol. 140:3473-3481.

27.Glisin, V., R. Crkvenjakov, and C. Byus. 1974.Ribonucleicacid isolated by cesiumchloridecentrifugation.Biochemistry 13:2633-2637.

28.Widmer,M.B., R. B. Acres, H. M.Sassenfeld,and K. H.Grabstein. 1987. Regulation of cytolytic cell populations from human peripheral blood by B cell

stimulatoryfactorI(interleukin4). J. Exp. Med. 166:1447-1455.

29.Gray, P. W., D. W. Leung,D.Pennica, E. Yelverton, R. Najarian, C. C.

Simonsen,R.Derynck, P. J.Sherwood, D. M. Wallace, S. L. Berger, et al. 1982. Expression of humaninterferon cDNA in E. coli and monkey cells. Nature

(Lond.).295:503-508.

30.Taniguchi, T., H. Matsui, T. Fujita, C. Takaoka, N. Kashima, R. Yoshi-moto, and J. Hamuro. 1983.Structure andexpression of a cloned cDNA for humaninterleukin-2. Nature (Lond.). 302:305-3 10.

31.Feinberg, A. P., and B. Vogelstein. 1983. A technique for radiolabeling DNArestrictionendonucleasefragments to high specific activity.Anal.Biochem. 132:6-13.

32. Pardoll, D., B. J. Fowlkes, R. Lechler, R. Germain, and R. Schwartz. 1987. Early events inTcelldevelopment analyzed by in situ hybridization. J. Exp. Med. 165:1624-1638.

33.Efrat, S., S. Pilo, and R. Kaempfer. 1982. Kinetics ofinduction and molec-ularsize ofmRNAsencodinghumaninterleukin-2 and -y-interferon.Nature

(Lond.).297:236-239.

34.Efrat,S., and R. Kaempfer. 1984. Control of biologically active interleu-kin 2 messengerRNAformation in induced humanlymphocytes. Proc.Natl. Acad. Sci. USA. 81:2601-2605.

(10)

36. Kronke, M.,W.J. Leonard, J. M. Depper, and W. C. Greene. 1985. Sequential expression of genes involved in human T lymphocyte growth and differentiation. J. Exp. Med. 161:1593-1598.

37.Taniguchi, T. 1988. Regulation of cytokine gene expression. Annu. Rev. Immunol. 6:439-464.

38. Morimoto, C., N. L. Letvin, A. W. Boyd, M. Hagan, H. M. Brown, M. M. Kornacki, and S. F. Schlossman. 1985. The isolation and characterization ofthe humanhelper inducerTcell subset. J. Immunol. 134:3762-3769.

39. Morimoto, C.,N. L. Letvin, J. A. Distaso, W. R. Aldrich, and S. F. Schlossman. 1985. The isolation and characterization of the human suppressor inducerTcell subset. J.Immunol. 134:1508-1515.

40. Clement, L. T.,N.Yamashita, and A. M. Martin. 1988. The functionally distinctsubpopulations of humanCD4'helper/inducer T lymphocytes defined by anti-CD45R antibodies derive sequentially fromadifferentiation pathway thatis regulated by activation-dependent post-thymic differentiation. J.

Im-munol. 141:1464-1470.

41.Serra, H. M., J. F. Krowka, J. A. Ledbetter, and L. M. Pilarski. 1988. Loss

ofCD45R (Lp 220) represents a post-thymic T cell differentiation event. J. Im-munol. 140:1435-1441.

42. Rothstein, D., S. Sohen, S. F. Schlossman, and C. Morimoto. 1989. CD45R is a functional and not a simple maturational marker for CD4 and CD8 cells in man. In Seventh International Congress of Immunology, Berlin (West). GustavFischer Verlag, Stuttgart. 248. (Abstr.)

43. Hayward, A. R., J. Lee, and P. C. L. Beverley. 1989. Ontogeny of expres-sion of UCHLI antigen onTCR-l'(CD4/8) and TCR6+ T cells. Eur. J.

Im-munol. 19:771-773.

44. Notarangelo, L. D., P. Panina, L. Imberti, P. Malfa, A. G. Ugazio, and A. Albertini. 1988. NeonatalT4'lymphocytes: analysis ofthe expression of4B4 and 2H4antigens. Clin. Immunol. Immunopathol. 46:61-67.

45.Fowlkes, B. J., and D. M. Pardoll. 1989. Molecular and cellularevents ofT celldevelopment. Adv. Immunol. 44:207-264.

46.English,B.K., S. K. Burchett,J.D.English, A. J. Ammann, D. W. Wara, andC.B.Wilson. 1988.Production of lymphotoxinand tumornecrosis factor by human neonatal mononuclear cells.Pediatr.Res.24:717-722.

47.Seki,H.,K.Taga,A.Matsuda, N. Uwadana,H.Masaki,T.Miyawaki, and N.Taniguchi.1986.Phenotypicandfunctional characteristics ofactive suppres-sorcellsagainst IFN-'y productioninPHA-stimulated cord bloodlymphocytes.J.

Immunol. 137:3158-3161.

48. Haas, A., S.Plager-Marshall,B. J.Ank,A.M.Martin,L.T.Clement, and E.R.Stiehm. 1987.PediatricRes. 21:408A.(Abstr.)

49.Budd,R.C., J.-C. Cerottini, and H. R. MacDonald. 1987.Phenotypic

identification of memory cytolyticTlymphocytesinasubset of Lyt-2+ cells. J.

Immunol.138:1009-1013.

50.Budd, R. C., J.-C.Cerottini,C.Horvath, C. Bron, T.Pedrazzini,R. C. Howe, and H. R. MacDonald. 1987.Distinctionofvirgin and memoryT

lym-phocytes: stableacquisitionofPgP-1 glycoproteinconcomitant withantigenic stimulation. J.Immunol.138:3120-3129.

51.Budd, R.C.,J.-C.Cerottini,andH. R.MacDonald. 1987. Selectively

increasedinterferon--y bysubsets ofLyt-2+andL3T4+Tcells identifiedby

ex-pressionofPgp-l.J.Immunol.138:3583-3586.

52. Powrie, F., andD. Mason.1989.TheMRC OX-22-CD4+Tcellsthathelp

Bcells insecondaryimmune responsesderive from naiveprecursorswiththe MRCOX-22+ CD4+phenotype.J.Exp. Med.169:653-662.

53. Frenkel, L., and V. Bryson. 1987.Ontogenyof

phytohemagglutinin-in-duced gammainterferon by leukocytesof healthinfantsand children:Evidence

fordecreasedproductionininfants younger than 2 months of age. J. Pediatr. 111:97-100.

54. Chilmonczyk,B.A., M. J.Levin,R.McDuffy,andA. R.Hayward.1985.

Characterization of the human newborn responsetoherpesvirusantigen.J. Immunol. 134:4184-4188.

55.Osborn,J.J., J.Dancis,and J. F. Julia. 1952. Studiesof theimmunology

of the newborn infant. I. Age andantibodyproduction.Pediatrics. 9:736-744.

56.Gerli, R.,P.Rambotti,C.Cernetti,A.Velardi,F.Spinozzi,A.Tabilio,M.

Martelli,F.Grignani,andS. Davis. 1984.A maturethymocyte-likephenotypic

patternonhuman cordcirculating T-lymphoidcells. J.Clin.Immunol. 4:461

-468.

57. Wilson, M.,F. S. Rosen, S.F.Schlossman, and E. L. Reinherz. 1985.

Ontogenyof humanTand Blymphocytes duringstressed and normalgestation:

Phenotypic analysisof umbilical cordlymphocytesfromtermand preterm in-fants.Clin. Immunol.Immunopathol.37:1-12.

58.Hercerd, T.,J.Ritz,S. F.Schlossman,andE.L.Reinherz. 1981.

Compara-tiveexpressionofT9,T10,and Iaantigensonactivated humanTcell subsets. Hum.Immunol.3:247-259.

59. Powers, G. D.,A.K.Abbas, R.A.Miller. 1988.Frequenciesof IL-2 and

IL-4-secretingTcells in naive andantigen-stimulatedlymphocytepopulations.J.

Immunol. 140:3352-3357.

60. Pure,E., K. Inaba, and J. Metlay. 1988.Lymphokine productionby

murineTcells in the mixedleucocyte reaction. J. Exp.Med. 168:795-800.

61.Swain,S. L., D.T.McKenzie,A. D.Weinberg,and W.Hancock. 1988.

CharacterizationofThelper I and 2cell subsets in normal mice. J.Immunol.

141:3445-3455.

62.Perussia,B.,L.Mangoni,H. D.Engers,andG. Trinchieri. 1980. Inter-feronproduction byhuman and murinelymphocytesin responsetoalloantigens.

J.Immunol. 125:1589-1595.

63.Lanier, L.L.,J. P.Allison,andJ. H.Phillips.1986.Correlation of cell surfaceantigen expressiononhumanthymocytes bymulti-colorflowcytometric analysis: implicationsfor differentiation.J.Immunol.137:2501-2507.

64.Crabtree,G. R. 1989.ContingentgeneticregulatoryeventsinT

lympho-cyteactivation. Science(Wash. DC).243:355-361.

65.Sussman,J.J.,M.Mercep,T.Saito,R.N.Germain,E.Bonvini,and J.D. Ashwell. 1988. Dissociation ofphosphoinositide hydrolysisandCa2"fluxesfrom thebiologicalresponsesofaT-cellhybridoma.Nature(Lond.).334:625-628.

66.Karray, S.,T. DeFrance,H.Merle-Beral,J.Banchereau,P.Debre,and P. Galanaud. 1988. Interleukin4counteractstheinterleukin2-induced

prolifera-tionof monoclonalBcells.J.Exp.Med. 168:85-94.

67.Jelinek,D.F.,andP. E. Lipsky. 1988.Inhibitoryinfluence ofIL-4on

human Bcellresponsiveness.J.Immunol. 141:164-173.

68.Sideras, P.,K.Funa,I.Zalcberg-Quintana,K.G.Xanthopoulos,P.

Kisie-low,andR.Palacios. 1988.Analysis byinsituhybridizationof cellsexpressing

mRNAforinterleukin4in thedeveloping thymusand inperipherallymphocytes

from mice. Proc.Nati.Acad.Sci. USA. 85:218-221.

69.Mosmann,T.R.,H.Cherwinski,M.W.Bond,M.A.Giedlin,andR.L. Coffman. 1986.Twotypes of murinehelperTcell clone.I.Definitionaccording

toprofilesoflymphokineactivities and secretedproteins.J. Immunol. 136:2438-2357.

70.Cherwinski,H.M.,J.D.Schumacher,K. D.Brown,and T. R. Mosmann. 1987.Two typesofmousehelperTcell clone.III.Further differences in

lympho-kine synthesisbetween Th1 andTh2clones revealedbyRNAhybridization,

functionallymonospecific bioassays,andmonoclonal antibodies. J.Exp.Med. 166:1229-1244.

71.Kelso,A., andN. M.Gough.1988.Coexpressionof

granulocyte-macro-phagecolony-stimulating factor,-yinterferon,and interleukins 3 and4is random in murine alloreactive T-lymphocyteclones. Proc.Natl. Acad. Sci. USA. 85:9 189-9 193.

72.Umetsu,D.T.,H. H.Jabara,R. H.DeKruyff,A. K.Abbas,J.S.Abrams,

and R.S. Beha. 1988. Functionalheterogeneityamonghuman inducerTcell clones. J.Immunol. 140:4211-4216.

73.Paliard,X., R. D. W.Malefijt, H.Yssel,D.Blanchard,I.Chretien,J. Abrams,J. D.Vries,andH.Spits. 1988. SimultaneousproductionofIL-2, IL-4,

andIFN--y byactivated humanCD4+andCD8+Tcell clones.J. Immunol. 141:849-855.

74.Hayakawa, K.,andR. R.Hardy. 1988. MurineCD4+Tcell subsets de-fined. J.Exp.Med. 168:1825-1838.

75.Hayakawa, K.,andR. R.Hardy. 1989.Phenotypicand functional alter-ation ofCD4+Tcellsafterantigenstimulation.J.Exp. Med. 169:2245-2250.

References

Related documents

53. Herbsleb M, Knudsen UB, Orntoft TF, Bichel P, Norild B, Knudsen A, Mogensen O. Telomerase Activity, Mib- 1, PCNA, HPV 16 and p53 as Diagnostic Markers for Cervical Intraepithelial

Abstract: Aim: This study aimed to evaluate temporomandibular joint reconstruction in Yemeni children with metatarsal bone graft after release of ankylosis.. Methodology: Ten

2, only the mean HDL-C was stable in postmenopausal women while the mean total cholesterol and LDL-C plasma concentrations were slightly increasing as the

Due to Kazakh‟s natural resources, the western companies were heavily involved in Kazakhstan.. and the key issue for American companies with investments in Kazakhstan has been at

By investigating TDP-43 expression patterns after stroke in young and 12-month-old mice, we showed (i) a marked increase and accumulation of TDP-43 in cytoplasmic compartment in

In the US the ISFSIs licensed (and those that will be built in future) store the spent fuel in dry casks approved for this purpose by the United States Nuclear Regulatory

IDS is nothing but the intrusion detection system. Intrusion is nothing but the attack. IDS is one type of tool that used to detect attack .That means IDS is used to prevent

Considering the effects of local buckling on the Optotrack strain calculation coupled with the fact that curvature is calculated on the basis of strain, the plastic hinge