Dis inc subpopula ions o gyT cells a e p esen in no mal and
umo -bea ing human li e
Tony Kenna,
a
Lucy Golden-Mason,
a
Suzanne No is,
b
John E. Hega y,
c,d,1
Cliona O’Fa elly,
a,d,1
and De ek G. Dohe y
e,
*
,1
a
Educa ion and Resea ch Cen e, S . Vincen ’s Uni e si y Hospi al, Dublin, I eland
b
Hepa ology Cen e, S . James’s Hospi al, Dublin, I eland
c
Li e Uni , S . Vincen ’s Uni e si y Hospi al, Dublin, I eland
d
Conway Ins i u e o Biomolecula and Biomedical Resea ch, Uni e si y College, Dublin, I eland
e
Ins i u e o Immunology and Depa men o Biology, Na ional Uni e si y o I eland, Maynoo h, Co., Kilda e, I eland
Recei ed 2 Feb ua y 2004; accep ed wi h e ision 17 May 2004
Abs ac
gyT cells a e hough o media e immune esponses a epi helial su aces. We ha e quan i ied and cha ac e ized hepa ic and pe iphe al
blood gyT cells om 11 no mal and 13 un esol ed umo -bea ing human li e specimens. gyT cells a e en iched in no mal li e (6.6% o T
cells) ela i e o ma ched blood (0.9%; P= 0.008). The majo i y exp ess CD4

CD8

pheno ypes and many exp ess CD56 and/o CD161. In
i o, hepa ic gyT cells can be induced o kill umo cell lines and elease in e e on-g, umo nec osis ac o -a, in e leukin-2 and in e leukin-
4. Analysis o Vgand Vychain usage indica ed ha Vy3
+
cells a e expanded in no mal li e s (21.2% o gyT cells) compa ed o blood (0.5%;
P= 0.001). Tumo -bea ing li e s had signi ican expansions and deple ions o gyT cell subse s bu no mal cy oly ic ac i i y. This s udy
iden i ies no el popula ions o li e T cells ha may play a ole in immuni y agains umo s.
D2004 Else ie Inc. All igh s ese ed.
Keywo ds: Tumo immuni y; Li e ; Human; gyT cells; Cy okines; Cy o oxici y
In oduc ion
gyT cells accoun o small numbe s o pe iphe al blood T
cells bu accumula e a epi helial su aces and a si es o
in ec ion [1]. Mos gyT cells a e nega i e o CD4 and CD8
and a e capable o ecognizing an igen wi hou he need o
majo his ocompa ibili y complex (MHC) es ic ion
( e iewed in Re s. [2,3]. Thei T cell ecep o s (TCRs)
ecognize ela i ely conse ed s uc u es on pa hogens and
hos cells including non-pep ide me aboli es and hea shock
p o eins. Pe iphe al blood gyT cells can apidly lyse an igen-
bea ing a ge cells and can kill a ange o umo cell lines in
i o. They also can apidly elease Th1 o Th2 cy okines
which can egula e di e en ia ion and ac i a ion o compo-
nen s o he adap i e immune sys em. gyT cells a e he e o e
hough o b idge inna e and adap i e immune esponses [2,3].
Dis inc subse s o gyTcells, based on hei TCR g- and y-
chain a iable gene segmen usage, a e di e en ially dis i-
bu ed in di e en issues and show d ama ic changes wi h
age. In adul humans, he Vg9Vy2 TCR is p edominan ly
ound among pe iphe al blood gyT cells, whe eas he
majo i y o in es inal gyTcells exp ess Vy1 chains associa ed
wi h one o se e al g-chains [4–6]. Mu ine and human
s udies ha e p o ided e idence ha he selec i e accumula-
ions o gyT cell subse s a di e en body loca ions a e he
esul o pe iphe al selec ion and expansion by locally
exp essed an igens [4,7,8]. O he s udies, howe e , ha e
sugges ed ha selec ion is no equi ed o p oduce in a ian
gyTCRs and ha hese ea angemen s a e p og ammed
[9,10].Vg9Vy2 T cells ecognize o ganic phosphoes e s,
alkylamines, and nucleo ide conjuga es ha a e cons i u i ely
exp essed by hos cells and mic obial pa hogens [11–13].
They also ecognize hea shock p o eins and pu a i e umo
an igens [14–16].Vy1
+
gyT cells can ecognize he s ess-
1521-6616/$ - see on ma e D2004 Else ie Inc. All igh s ese ed.
doi:10.1016/j.clim.2004.05.003
* Co esponding au ho . Ins i u e o Immunology and Depa men o
Biology, Na ional Uni e si y o I eland, Maynoo h, Maynoo h, Co.,
Kilda e, I eland. Fax: +353-1-7083845.
E-mail add ess: de ek.g.dohe [email protected] (D.G. Dohe y).
1
These au ho s con ibu ed equally o he di ec ion o his s udy.
www.else ie .com/loca e/yclim
Clinical Immunology 113 (2004) 56– 63
inducible MICA and MICB molecules ha a e up egula ed as
a esul o umo ans o ma ion and i us in ec ion [17,18]
al hough i appea s ha his ecogni ion is no media ed by he
TCR [19].Vy1
+
Tcells ha e also been shown o ecognize he
lipid-p esen ing molecule, CD1c [20].
The li e appea s o ha e an impo an immuno egula-
o y ole, being able o selec i ely induce immuni y o
ole ance o an igens [21]. The adul human li e con ains
se e al popula ions o lymphocy es ha exhibi apid an i-
gen-non-speci ic cy o oxici y and Th1/Th2 and egula o y
cy okine sec e ion [22–24]. These include many gyT cells.
He e, we ha e quan i ied and cha ac e ized hepa ic gyT
cells om no mal and umo -bea ing human li e speci-
mens, wi h espec o Vgand Vy-chain usage, T and NK cell
ma ke exp ession, cy okine sec e ion, and cy o oxic po en-
ial. Ou da a indica e ha he li e has unique epe oi es o
gyT cells ha include cells capable o an i umo cy o o-
xici y in i o and whose numbe s a e changed in pa ien s
wi h hepa ic malignancy.
Ma e ials and me hods
Tissue specimens
Wedge li e biopsies (50–100 mg) we e ob ained om
11 heal hy dono o gans (8 male and 3 emale; mean age
39.5 yea s; age ange 18–63 yea s) a he ime o li e
ansplan a ion. Li e biochemis y and his ology we e
no mal in all cases. Li e issue was ob ained om 13
pa ien s unde going esec ion o hepa ic me as ases o
colonic o igin (7 male and 6 emale; mean age 60.2 yea s;
age ange 35–75 yea s). Wedge biopsies om umo -bea -
ing issue we e aken app oxima ely 10 cm om he umo
ma gin and appea ed his ologically no mal. All o gan
dono s and pa ien s we e nega i e o hepa i is A, B, C, y,
and E i uses. E hical app o al o his s udy was ob ained
om he Resea ch and E hics Commi ee a S . Vincen ’s
Uni e si y Hospi al, Dublin, I eland. Single cell suspensions
o hepa ic mononuclea cells (HMC) we e p epa ed as
p e iously desc ibed [25].gyT cells we e isola ed om
esh HMC p epa a ions using mAb-coa ed magne ic beads
(Mil enyi Bio ec, Be gisch Gladbach, Ge many) acco ding
o manu ac u e ’s p o ocols. The pu i y o isola ed ac ions
was assessed by low cy ome y and only p epa a ions wi h
pu i ies o 95% o g ea e we e used o unc ional s udies.
An ibodies and low cy ome y
Monoclonal an ibodies (mAb) speci ic o human CD3
(clone SK7), CD4 (SK3), CD8ah(SK1), CD56
(NCAM16.2), gyTCR (11F2), ahTCR (WT31), CD161
(DX12), IFN-g(25723.11), TNF-a(6401.1111), IL-2
(5344.1111), o IL-4 (8D4-8) we e ob ained om Bec on
Dickinson (Ox o d, UK). Unconjuga ed mAbs speci ic o
TCR V- egions Vy1 (clone 9R.12), Vy2 (Immu389), Vy3
(P11.5B), Vg2-3-4 (23D12), Vg8(R4.5),andVg9
(Immu360) we e ob ained om Immuno ech (Ma seille,
F ance). The exp ession o hese cell su ace an igens by
eshly isola ed HMC and pe iphe al blood mononuclea
cells (PBMC) was de ec ed by mAb s aining and h ee-colo
low cy ome y (FACScan, Bec on Dickinson) and analyzed
using CellQues so wa e (Bec on Dickinson). To a oid
compe i ion o binding si es on he gyTCR, he mAb ha
ecognizes all gyTCRs was no used in conjunc ion wi h
mAbs speci ic o indi idual Vgo VyTCR chains. Ins ead,
cells we e cos ained wi h he pan-gyand an i-CD3 mAbs, and
an i-CD3 wi h indi idual Vgo VymAbs in sepa a e ubes.
De ec ion o in acellula cy okines
Cy okine p oduc ion by gyT cells a e s imula ion in
i o was examined by low cy ome y. To elimina e non-
speci ic au o luo escence om o he li e cell popula ions,
gyT cells we e i s pu i ied om o al HMC by immuno-
magne ic bead sepa a ion. Hepa ic gyT cells (1 10
6
cells/
ml) we e s imula ed wi h 10 ng/ml pho bol my is a e ace a e
(PMA) (Sigma Co., Poole, UK) plus 1 Ag/ml ionomycin
(Sigma) o 4 h in 24-well pla es a 37jCin5%CO
2
.As
con ols, uns imula ed cells we e ea ed simila ly. B e eldin
A (10 Ag/ml, Sigma), an inhibi o o p o ein ansloca ion
om he endoplasmic e iculum o he Golgi appa a us, was
added o he cells o he 4-h incuba ion. Cells we e hen
s ained wi h mAbs speci ic o su ace CD3 and in acellula
IFN-g, TNF-a, IL-2, o IL-4 and de ec ed by h ee-colo
low cy ome y as desc ibed p e iously [24].
Cy o oxici y assays
Magne ic bead-pu i ied gyT cells we e used as e ec o
cells in cy o oxici y assays agains K562 and Daudi a ge
cell lines. Na u al cy o oxici y and lymphokine-ac i a ed
killing (LAK) we e examined using gyT cells ha we e
p ecul u ed o 3 days in he absence o p esence o 50 uni s/
ml IL-2, espec i ely. Cy o oxici y was assessed in s anda d
4-h
51
C elease assays using 2000
51
C -labeled a ge cells
incuba ed wi h pu i ied gyT cells a E/T a ios o 1:1, 5:1,
25:1, and 50:1 [24]. Speci ic lysis was calcula ed om he
amoun s o
51
C eleased in o supe na an s using he o -
mula: % speci ic lysis = (cpm o sample cpm o
spon aneous elease) 100 / (cpm o maximum elease 
cpm o spon aneous elease). Spon aneous elease was
de e mined by incuba ing he a ge cells in he absence o
e ec o cells and maximum elease was ob ained by incu-
ba ion o a ge s wi h 0.1% T i on X-100.
S a is ical analysis
Resul s we e exp essed as median and anges. The
Spea man Rank and Mann–Whi ney U es s o non-pa a-
me ic da a we e used o analyze esul s. P alues o <0.05
we e conside ed o be signi ican .
T. Kenna e al. / Clinical Immunology 113 (2004) 56–63 57
Resul s
cd T cells accumula e in human li e
Flow cy ome y was used o de e mine he equencies o
human T cells ha exp ess he TCR gyhe e odime in 11
his ologically no mal dono li e s and ma ched blood sam-
ples (Fig. 1A). Fo his pa o he s udy, cells we e s ained
ex i o wi h no p io manipula ion. Fig. 1B shows ha he e
is a signi ican en ichmen o gyT cells in no mal human
li e ela i e o blood, since his TCR was ound o be
exp essed by a median o 0.93% ( ange 0.24–2.7%) o
pe iphe al blood CD3
+
cells and 6.6% ( ange 4.0–13.8%) o
hepa ic CD3
+
cells (P= 0.008).
Pheno ypic cha ac e iza ion o cd T cells in no mal human
li e
The pheno ypes o gyT cells in eshly isola ed ma ched
li e and pe iphe al blood samples we e u he analyzed by
h ee-colo low cy ome y. Simila o ci cula ing gyT cells,
he majo i y o hepa ic gyT cells we e nega i e o CD4 and
CD8 (median 85.5%; ange 79.1–92.0%; Table 1). Less han
5% o hepa ic gyT cells exp essed CD4 (4.2%; ange 0.8–
12.6%), while a signi ican ly highe p opo ion o hepa ic gy
T cells we e CD8
+
(8.9%; ange 6.1%–15.8%) compa ed
wi h hose ound in he ci cula ion (1.5%; ange 0.6–2.8%)
(P< 0.001; Table 1). A high p opo ion o bo h ci cula ing
and hepa ic gyT cells exp essed he NK cell ma ke s CD56
(38.9%; ange 21.3–57.9% in blood and 38.8%; ange 21.1–
58.4% in li e ) and CD161 (36.8%; ange 9.1–59.2% in
blood and 36.2%; ange 17.0–54.6% in li e ; Table 1).
TCR cand dchain gene segmen usage
To de e mine whe he hepa ic gyT cells display any
‘ issue opism’, we examined he exp ession by hepa ic
and pe iphe al blood gyT cells o he Vgand Vychains
ha ha e p e iously been epo ed o be he mos equen -
ly ound in blood, namely Vg2-3-4, Vg8, Vg9, Vy1, Vy2,
and Vy3 chains [26,27] (Fig. 2). The Vg2-3-4, Vg8, and
Vg9 chains we e ound o be exp essed by simila e-
quencies o gyT cells in blood and li e (Fig. 2A). The
majo i y o hepa ic gyT cells exp essed Vy2chains
(56.4%; ange 29.0–86.4%) as ound in blood, while use
o he Vy1 chain by hepa ic gyT cells (8.9%; ange 3.4–
32.7%) was signi ican ly educed compa ed wi h ha seen
in he blood (42.7%; ange 5.7–74.2%; P= 0.035; Fig.
2A). S ikingly, a signi ican p opo ion o hepa ic gyT
cells we e ound o exp ess he Vy3 chain (21.2%; ange
0–37.2%). In con as , his y-chain is almos ne e
exp essed by ci cula ing lymphocy es (0.5%; ange 0–
0.9%; P= 0.001; Figs. 2A and B).
Cy okine p oduc ion by hepa ic cd T cells
Flow cy ome y was used o examine he cy okine
sec e ion p o iles o hepa ic gyT cells in esponse o in
i o s imula ion wi h PMA and ionomycin (Fig. 3).To
enhance cla i y o mAb s aining, gyT cells we e i s
posi i ely selec ed om o al HMC using magne ic beads.
Hepa ic gyT cells displayed a s ong Th1 bias. Signi ican
p opo ions (median 29.2%; ange 14.6–54.9%) o hepa ic
gyT cells p oduced IFN-gin esponse o s imula ion in i o
wi h PMA and ionomycin. Smalle p opo ions p oduced
IL-2 (12.9%; ange 8.0–17.1%) o TNF-a(5.3%; ange
1.4–15.6%). A median o 4.6% (2.8–6.2%) o hepa ic gyT
cells p oduced he Th2 cy okine IL-4. In addi ion, 6.2%
(1.7–14.7%) o hepa ic gyT cells om heal hy li e
Fig. 1. gyT cell ecep o (TCR) exp ession by eshly isola ed human
pe iphe al blood and hepa ic T cells. (A) Rep esen a i e low cy ome y do
plo showing ahand gyTCR exp ession by ga ed CD3
+
mononuclea cells
om he blood and li e o a li e ansplan dono . Numbe s show he
pe cen ages o CD3
+
cells ha exp ess gyTCRs. (B) Pe cen ages o CD3
+
cells
in blood and li e s o 11 li e ansplan dono s exp essing gyTCRs. Medians
a e shown as ho izon al lines wi h s anda d e o s as e o ba s. *P= 0.008.
Table 1
Pheno ypic analysis o pe iphe al blood and hepa ic gyT cells in 11
his ologically no mal li e ansplan dono s
Pheno ype F equency in blood F equency in li e P alue
% Range % Range
CD4 3.6 0.8–5.4 4.2 0.8–12.6
CD8 1.5 0.6–2.8 8.9 6.1–15.8 0.001
DN 92.8 91.9–98.7 85.5 79.1–92.0
CD56 38.8 21.3–57.9 38.9 21.1–58.4
CD161 36.8 9.1–59.2 36.2 17.0–54.6
T. Kenna e al. / Clinical Immunology 113 (2004) 56–6358
spon aneously p oduced IFN-gin he absence o s imula ion
while 7.3% (0.5–12.1%) o uns imula ed cells p oduced IL-
2. This spon aneous cy okine p oduc ion may be due o
backg ound le els o in i o ac i a ion, o al e na i ely, i
may be due o low le el ac i a ion by he mAb-coa ed
magne ic beads used o gyT cell pu i ica ion.
cd T cells in umo -bea ing li e
The numbe s, pheno ypes, and unc ions o hepa ic and
pe iphe al blood gyT cells we e also compa ed be ween
li e dono s and pa ien s wi h hepa ic malignancy. Fig. 4A
shows ha gyT cells a e signi ican ly expanded in he blood
o pa ien s wi h hepa ic malignancy (3.9% o o al CD3
+
cells s. 0.9%; P= 0.008). While umo -bea ing li e also
showed inc eased p opo ions o gyT cells (10.4% s.
6.6%) his inc ease was no signi ican (P= 0.3). The
exp ession o he NK cell ma ke s CD56 and CD161 by
hepa ic gyT cells was inc eased in malignancy. CD56 was
exp essed by a median o 38.9% o gyT cells om
his ologically no mal li e s and 59.7% in malignan issue
(P= 0.04). CD161 was exp essed by 36.2% o gyT cells
om no mal li e s and 47.5% o gyT cell om umo -
bea ing li e s ( P= 0.05; Fig. 4B).
Analysis o Vgand Vychain usage in umo -bea ing
li e e ealed signi ican ly di e en usage compa ed wi h
heal hy li e (Fig. 4C). Tumo -bea ing li e displayed
p e e en ial usage o he Vy1 gene p oduc s (26.8% o
o al gyT cells; P= 0.03), while Vy2 (21.2%) and Vy3
(9.61%) exp ession was signi ican ly educed ( P= 0.03
and 0.02, espec i ely). While almos all gyT cells in he
blood and li e o li e ansplan dono s consis en ly
s ained posi i e o mAbs speci ic o ei he Vy1, Vy2,
o Vy3, app oxima ely 40% o gyT cells in umo -bea ing
li e s did no s ain o any o hese mAbs, sugges ing ha
agyT cell wi h a p e iously un ecognized TCR y-chain
is expanded in he li e s o pa ien s wi h hepa ic malig-
nancy. No di e ences we e seen be ween Vgchain usage
be ween no mal and umo -bea ing li e s (da a no
shown).
Cy o oxic ac i i y o hepa ic cd T cells
gyT cells om heal hy (n= 4) and umo -bea ing (n=4)
li e samples we e pu i ied om eshly isola ed HMC
using immunomagne ic beads (Fig. 5A) and used as e ec-
Fig. 3. Cy okine p oduc ion by hepa ic gyT cells. Pe cen ages o hepa ic gy
T cells ha s ain posi i e o in e e on-g(IFN-g), in e leukin-2 (IL-2),
umo nec osis ac o -a(TNF-a) and IL-4 which we e ei he uns imula ed
() o s imula ed o 4 h wi h PMA and ionomycin (+). Ho izon al lines
indica e medians o ou indi idual samples.
Fig. 2. T cell ecep o gand ychain usage by eshly isola ed human
pe iphe al blood and hepa ic T cells. (A) Box plo s showing medians
(ho izon al lines), in e qua ile anges (shaded a eas) and anges (e o ba s)
o pe cen ages o gyT cells in blood and li e s o 11 li e ansplan dono s
exp essing Vy1, Vy2, Vy3, Vg2-3-4, Vg8, and Vg9. *P= 0.035; **P=
0.001. (B) Rep esen a i e low cy ome y do plo showing Vy3 exp ession
by CD3
+
T cells om he blood and li e o a li e ansplan dono . The
numbe s show he pe cen ages o CD3
+
T cells ha exp ess Vy3.
T. Kenna e al. / Clinical Immunology 113 (2004) 56–63 59
o s in cy o oxici y assays agains K562 and Daudi a ge
cells. When eshly isola ed gyT cells we e used as
e ec o s, hey we e unable o lyse ei he a ge cell line
(Fig. 5B). Howe e , a e incuba ion o 3 days wi h 50
uni s/ml IL-2, gyT cells om bo h no mal and umo -
bea ing li e s exhibi ed cy o oxici y agains K562 and
Daudi a ge s (Fig. 5B), indica ing ha hepa ic gyT cell
popula ion con ain p ecu so s o LAK cells. No signi ican
di e ence in he cy o oxic ac i i y o gyT cells om no mal
o umo -bea ing li e s agains K562 o Daudi a ge cells
was obse ed.
Discussion
gyT cells a e p esen in signi ican numbe s in he
epi helia o he small in es ine, skin, lungs and he p egnan
u e us bu hey ep esen only a small p opo ion (<5%) o
he T cells in pe iphe al blood [2,3]. The esul s o he
p esen s udy indica e ha he adul li e also is a si e o
accumula ion o gyT cells and ha his o gan has a dis inc
dis ibu ion o he mos common TCR g- and y-chains. The
majo i y o hepa ic gyT cells exp ess double-nega i e
CD4

CD8

pheno ypes in keeping wi h hei abili y o
ecognize an igen wi hou he need o MHC- es ic ion, bu
CD4
+
and CD8
+
gyT cells we e also seen. A signi ican
p opo ion exp ess he NK cell-associa ed ecep o s (NKR)
CD56 and CD161. Hepa ic gyT cells can be induced by
cul u e wi h IL-2 o kill he umo cell lines, K562 and
Daudi, and upon ac i a ion in i o, hey p oduce a ange o
cy okines, including IFN-g, TNF-a, IL-2, and IL-4.
Fig. 4. Pe iphe al blood and hepa ic gyT cell pheno ypes in pa ien s wi h
hepa ic malignancy. (A) Box plo s showing medians (ho izon al lines),
in e qua ile anges (shaded a eas) and anges (e o ba s) o pe cen ages o
CD3
+
T cells in blood and li e s o 11 his ologically no mal and 13 umo -
bea ing li e specimens ha exp ess gyT cell ecep o s. (B and C)
Exp ession o CD56, CD161 and CD8 (B) and Vy1, Vy2 and Vy3 T cell
ecep o chains (C) by gyT cells om he li e s o 11 his ologically no mal
and 13 umo -bea ing li e specimens. *P< 0.05 in all cases.
Fig. 5. Hepa ic gyT cells a e capable o lymphokine-ac i a ed killing o
K562 and Daudi a ge cells. (A) Rep esen a i e low cy ome y his og am
showing gyT cell ecep o (TCR) exp ession by magne ic bead isola ed
hepa ic gyT cells (shaded his og am) and unsepa a ed hepa ic mononuclea
cells (unshaded). The numbe indica es he pu i y o he gyT cell
p epa a ion. (B) Na u al cy o oxici y (no IL-2) and lymphokine ac i a ed
killing (+IL-2) o K562 and Daudi a ge cells by magne ic bead-pu i ied gy
T cells. The igu e is ep esen a i e o expe imen s using ou his ologically
no mal and ou umo -bea ing li e samples.
T. Kenna e al. / Clinical Immunology 113 (2004) 56–6360

T cells ha exp ess NKRs and display po en MHC-
un es ic ed cy o oxici y and apid cy okine sec e ion ac-
coun o abou one hi d o all hepa ic lymphocy es bu
<5% o pe iphe al blood lymphocy es [22–24].These
include CD1d- es ic ed NKT cells wi h in a ian ahTCRs
ha ecognize lipid an igens [28,29]. He e, we show ha
hey also include gyT cells. Thus, mul iple ypes o semi-
in a ian T cells ha media e immedia e and egula o y
immune unc ions accumula e in he li e .
In con as o he human ahTCR epe oi e, he ge mline
encoded gyTCR epe oi e is small [26,30,31]. As p e i-
ously epo ed [4,26,27], we ound ha almos all pe iphe al
blood and hepa ic gyT cells a e posi i e o ei he Vg2-3-4,
Vg8, o Vg9 and ei he Vy1, Vy2, o Vy3 TCR chains. In
bo h issues, he majo i y (usually be ween 50% and 90%)
o gyT cells exp ess Vg9andVy2. Howe e , we ound ha
Vy3
+
T cells accoun o app oxima ely 20% o hepa ic gyT
cells bu only abou 0.5% o pe iphe al gyT cells in heal hy
indi iduals. The eason o he ex ensi e use o he Vy3
elemen in he li e is unclea . I is known ha he Vy3 gene
segmen equen ly ecombines wi h Jaand Cagene seg-
men s and ha he Vy3 TCR chain does no appea o
selec i ely pai wi h any pa icula Vgchain [26].gyTCR
ea angemen s occu ea ly in childhood and pa icula gyT
cell subse s accumula e a pa icula body loca ions wi h age
[4–6]. Such accumula ions may be p og ammed [9,10] o
an igen-d i en [7,8,27] and i he la e is ue, he Vy3 TCR
chain is likely o espond o an an igen ound in he li e .
Vy3
+
T cells a e epo ed o expand in he pe iphe al blood
o enal ansplan ecipien s ha de elop cy omegalo i us
in ec ion [27]. I is unknown whe he Vy3
+
T cells ha e
unc ions ha a e di e en om hose o o he gyT cells bu
p e ious s udies ha e sugges ed ha di e en gyT cells
subse s ha e dis inc cy okine sec e ion pa e ns [3]. While
he biased usage o gyTCRs in he li e will e lec a
dis inc speci ici y o an igen ecogni ion by gyT cells,
u u e s udies using pu i ied Vy3
+
T cells a e equi ed o
iden i y hei unc ions.
The e is a la ge body o e idence o sugges ha gyT
cells play a ole in hos de ence agains umo s [32,33].gyT
cells display an i umo cy o oxici y in i o [34–36].
Vg9Vy2
+
T cells ecognize an igens ound on myeloma
and lymphoma cells [15,16] and Vy1
+
T cells ecei e
cos imula o y signals h ough NKG2D, a ecep o o he
nonclassical MHC class I-molecule, MICA, which is up e-
gula ed on umo cells o epi helial o igin [17,18].gyT cells
a e p esen among umo in il a ing lymphocy es in b eas
ca cinoma [37,38], enal ca cinoma [39] panc ea ic cance
[40], epi helial umo s [41], melanoma, and sa coma [42].
Addi ionally, di e en pa e ns o Vygene ea angemen
ha e been obse ed in pai ed blood and umo -in il a ing
lymphocy e samples om pa ien s wi h di e en cance s
[43]. Seki e al. [44,45] epo ed ha ac i a ed gyT cells
accumula e in he li e s o mice and humans wi h umo s.
We ha e ound small o e all expansions o gyT cells
exp essing NKRs in he blood and li e s o pa ien s wi h
hepa ic malignancy. Howe e , on examina ion o gyT cell
subse s, we ound ha he p opo ions o hepa ic gyT cells
exp essing he Vy1
+
TCR chain, hough o play a ole in he
esponse o epi helial umo s [17,18], a e signi ican ly
expanded in he li e . Vy1
+
T cells a e also expanded in
he li e s o pa ien s wi h ch onic hepa i is C i us in ec ion
[46], a condi ion ha ca ies an inc eased isk o de eloping
hepa ocellula ca cinoma. In con as , Vy2
+
cells which a e
he mos abundan hepa ic gyT cell subse , and Vy3
+
cells
which appea o be unique o he li e we e ound in
signi ican ly lowe equencies in umo -bea ing li e s.
In e es ingly, he p opo ions o gyT cells exp essing
Vy1, Vy2, o Vy3 TCR chains in he blood and li e o
li e ansplan dono s consis en ly added up o app oxi-
ma ely 100%, bu app oxima ely 40% o gyT cells in
umo -bea ing li e s did no s ain posi i e o any o hese
mAbs, sugges ing ha a gyT cell wi h a p e iously
un ecognized TCR y-chain is expanded in he li e s o
pa ien s wi h hepa ic malignancy. I is unclea om he
p esen s udy whe he he dec eases in Vy2
+
and Vy3
+
cell
numbe s in umo -bea ing li e s a e a esul o in luxes o
expansions o Vy1
+
and o he gyT cells. I is possible ha
changes in he numbe s o gyT cell subse s may ei he
p edispose an indi idual o me as a ic li e disease o occu
in esponse o he umo . Al hough he p opo ions o gyT
cell subse s we e di e en in his ologically no mal and
umo -bea ing li e s, o al gyT cells pu i ied om hese
issues displayed simila LAK cy o oxic ac i i ies agains
K562 and Daudi a ge s in i o.
In conclusion, he esul s o he p esen s udy p o ide
u he e idence o he p esence o cy o oxic and Th1/Th2
cy okine- eleasing T cells wi h semi-in a ian TCRs in he
adul human li e . Hepa ic gyT cells a e p edominan ly
Th1-like cells and show biased TCR usage compa ed o
blood. Ou da a sugges a ole o hepa ic gyT cells in
immuni y agains umo s.
Acknowledgmen s
This wo k was suppo ed by g an s awa ded by he I ish
Heal h Resea ch Boa d and En e p ise I eland. We a e
g a e ul o ou su gical colleagues, Ge y McEn ee, Osca
T ayno , Raghu Va ada ajan, Jus in Geoghegan and he li e
ansplan coo dina o s, Sheila O’Toole, Aoi e Co ey and
Jenni e Fleming, a S . Vincen ’s Uni e si y Hospi al o
assis ance in ob aining li e biopsy samples.
Re e ences
[1] R.L. Modlin, C. Pi mez, F.M. Ho man, V. To igian, K. Uyemu a, T.H.
Rea, B.R. Bloom, M.B. B enne , Lymphocy es bea ing an igen-spe-
ci ic gyT-cell ecep o s accumula e in human in ec ious disease
lesions, Na u e 339 (1989) 544–548.
[2] A.C. Hayday, gycells: a igh ime and a igh place o a conse ed
hi d way o p o ec ion, Annu. Re . Immunol. 18 (2002) 975 –1026.
T. Kenna e al. / Clinical Immunology 113 (2004) 56–63 61
[3] S.R. Ca ding, P.J. Egan, gyT cells: unc ional plas ici y and he e o-
genei y, Na . Re ., Immunol. 2 (2002) 336–345.
[4] C.M. Pa ke , V. G oh, H. Band, S.A. Po celli, C. Mo i a, M. Fabbi, D.
Glass, J.L. S ominge , M.B. B enne , E idence o ex a hymic
changes in he T cell ecep o gamma/del a epe oi e, J. Exp. Med.
171 (1990) 1597–1612.
[5] C.T. Mo i a, C.M. Pa ke , M.B. B enne , H. Band, TCR usage and
unc ional capabili ies o human gyT cells a bi h, J. Immunol. 153
(1994) 3979–3988.
[6] A. Buch , K. Sode s om, S. Esin, J. G unewald, S. Hagelbe g, I.
Magnusson, H. Wigzell, A. G onbe g, R. Kiessling, Analysis o gy
V egion usage in no mal and diseased human in es inal biopsies and
pe iphe al blood by polyme ase chain eac ion (PCR) and low
cy ome y, Clin. Exp. Immunol. 99 (1995) 57– 64.
[7] G. De Libe o, G. Caso a i, C. Giachino, C. Ca bona a, N. Migone, P.
Ma zinge , A. Lanza ecchia, Selec ion by wo powe ul an igens may
accoun o he p esence o he majo popula ion o human pe iphe al
g/yT cells, J. Exp. Med. 173 (1991) 1311–1322.
[8] C.A. Mallick-Wood, J.M. Lewis, L.I. Richie, M.J. Owen, R.E.
Tigelaa , A.C. Hayday, Conse a ion o T cell ecep o con o ma-
ion in epide mal gycells wi h dis up ed p ima y Vggene usage,
Science 279 (1998) 1729–1733.
[9] D.M. Asa now, D. Cado, R.H. Raule , Selec ion is no equi ed o
p oduce in a ian T-cell ecep o g-gene junc ional sequences, Na u e
362 (1993) 158–160.
[10] S. I oha a, P. Mombae s, J. La aille, J. Iacomini, A. Nelson, A.R.
Cla ke, M.L. Hoope , A. Fa , S. Tonegawa, T cell ecep o ygene
mu an mice: independen gene a ion o ahT cells and p og ammed
ea angemen s o gyTCR genes, Cell 72 (1993) 337–348.
[11] P. Cons an , F. Da odeau, M.A. Pey a , Y. Poque , G. Puzo, M. Bon-
ne ille, J.J. Fou nie, S imula ion o human gyT cells by nonpep idic
mycobac e ial ligands, Science 264 (1994) 267–270.
[12] Y. Tanaka, C.T. Mo i a, E. Nie es, M.B. B enne , B.R. Bloom, Na-
u al and syn he ic non-pep ide an igens ecognized by human gyT
cells, Na u e 375 (1995) 155–158.
[13] J.F. Bukowski, C.T. Mo i a, M.B. B enne , Human gyT cells ecog-
nize alkylamines de i ed om mic obes, edible plan s, and ea: impli-
ca ions o inna e immuni y, Immuni y 11 (1999) 57–65.
[14] A. Ha egewoin, G. Soman, R.C. Hom, R.W. Finbe g, Human gy
+
T
cells espond o mycobac e ial hea -shock p o ein, Na u e 340 (1989)
309–312.
[15] E. S u m, E. B aakman, P. Fisch, R.J. V eugdenhil, P. Sondel, R.L.
Bolhuis, Human Vg9-Vy2 T cell ecep o -gylymphocy es show spec-
i ici y o Daudi Bu ki ’s lymphoma cells, J. Immunol. 145 (1990)
3202–3208.
[16] P. Fisch, M. Malko sky, S. Ko a s, E. S u m, E. B aakman, B.S.
Klein, S.D. Voss, L.W. Mo issey, R. DeMa s, W.J. Welch, R.J.L.
Bolhuis, P.M. Sondel, Recogni ion by human Vg9/Vy2 T cells o a
G oEL homolog on Daudi Bu ki ’s lymphoma cells, Science 250
(1990) 1269–1273.
[17] V. G oh, A. S einle, S. Baue , T. Spies, Recogni ion o s ess-induced
MHC molecules by in es inal epi helial gyT cells, Science 279 (1998)
1737–1740.
[18] V. G oh, R. Rhineha , H. Sec is , S. Baue , K.H. G abs ein, T. Spies,
B oad umo -associa ed exp ession and ecogni ion by umo -de i ed
gyT cells o MICA and MICB, P oc. Na l. Acad. Sci. U. S. A. 96
(1999) 6879–6884.
[19] S. Baue , V. G oh, J. Wu, A. S einle, J.H. Phillips, L.L. Lanie , T.
Spies, Ac i a ion o NK cells and T cells by NKG2D, a ecep o o
s ess-inducible MICA, Science 285 (1999) 727 – 729.
[20] F.M. Spada, E.P. G an , P.J. Pe e s, M. Sugi a, A. Melian, D.L. Leslie,
H.K. Lee, E. an Donselaa , D.E.A. Hanson, A.M. K ensky, O. Maj-
dic, S.A. Po celli, C.T. Mo i a, M.B. B enne , Sel - ecogni ion o
CD1 by gyT cells: implica ions o inna e immuni y, J. Exp. Med.
191 (2000) 937–948.
[21] R.Y. Calne, Immunological ole ance—The li e e ec , Immunol.
Re . 174 (2000) 280–282.
[22] S. No is, C. Collins, D.G. Dohe y, F. Smi h, G. McEn ee, O.
T ayno , N. Nolan, J.E. Hega y, C. O’Fa elly, Residen hepa ic
lymphocy es a e pheno ypically di e en om ci cula ing lympho-
cy es, J. Hepa ol. 28 (1998) 84–90.
[23] S. No is, D.G. Dohe y, C. Collins, G. McEn ee, O. T ayno , J.E.
Hega y, C. O’Fa elly, Na u al T cells in he human li e : cy o oxic
lymphocy es wi h dual T cell and na u al kille cell pheno ype and
unc ion a e pheno ypically he e ogenous and include Va24-JaQ and
gyT cell ecep o bea ing cells, Hum. Immunol. 60 (1999) 20–31.
[24] D.G. Dohe y, S. No is, L. Mad igal-Es ebas, G. McEn ee, O.
T ayno , J.E. Hega y, C. O’Fa elly, The human li e con ains mul-
iple popula ions o NK cells, T cells, and CD3
+
CD56
+
na u al T
cells wi h dis inc cy o oxic ac i i ies and Th1, Th2, and Th0 cy o-
kine sec e ion pa e ns, J. Immunol. 163 (1999) 2314–2321.
[25] M.P. Cu y, S. No is, L. Golden-Mason, D.G. Dohe y, T. Deignan,
C. Collins, O. T ayno , G.P. McEn ee, J.E. Hega y, C. O’Fa elly,
Isola ion o lymphocy es om no mal adul human li e sui able o
pheno ypic and unc ional cha ac e iza ion, J. Immunol. Me hods 242
(2000) 21–31.
[26] M.A. Pey a , F. Da odeau, I. Houde, F. Romagne, A. Necke , C.
Lege , J.P. Ce oni, N. Ce -Bensussan, H. Vie, M. Bonne ille,
M.M. Halle , Repe oi e analysis o human pe iphe al blood lympho-
cy es using a human Vy3 egion-speci ic monoclonal an ibody. Cha -
ac e iza ion o dual T cell ecep o (TCR) y-chain exp esso s and ah
T cells exp essing Vy3JaCa-encoded TCR chains, J. Immunol. 155
(1995) 3060–3067.
[27] J. Dechane , P. Me ille, A. Lim, C. Re ie e, V. Pi a d, X. La a ge, S.
Michelson, C. Me ic, M.M. Halle , P. Kou ilsky, L. Po aux, M. Bon-
ne ille, J.F. Mo eau, Implica ion o gammadel a T cells in he human
immune esponse o cy omegalo i us, J. Clin. In es . 103 (1999)
1437–1449.
[28] M.A. Exley, Q. He, O. Cheng, R.J. Wang, C.P. Cheney, S.P. Balk,
M.J. Koziel, Cu ing edge: compa men aliza ion o Th1-like non-
in a ian CD1d- eac i e T cells in hepa i is C i us-in ec ed li e ,
J. Immunol. 168 (2002) 1519–1523.
[29] T. Kenna, L. Golden-Mason, S.A. Po celli, Y. Koezuka, J.E. Hega y,
C. O’Fa elly, D.G. Dohe y, NKT cells om no mal and umo -
bea ing human li e s a e pheno ypically and unc ionally dis inc
om mu ine NKT cells, J. Immunol. 171 (2003) 1775–1779.
[30] S. Ha a, M. Clabby, P. De lin, H. Spi s, J.E. De V ies, M.S. K angel,
Di e si y and o ganiza ion o human T cell ecep o y a iable gene
segmen s, J. Exp. Med. 169 (1989) 41 –57.
[31] X.M. Zhang, C. Tonnelle, M.P. Le anc, S. Huck, T cell ecep o g
cDNA in human e al li e and hymus: a iable egions o gamma
chains a e es ic ed o Vg1o Vg9, due o he absence o splicing o
he V10 and V11 leade in on, Eu . J. Immunol. 24 (1994) 571– 578.
[32] A. Mo is, S. Ro hen usse , E. Meue , R. Hang e inge , P. Fisch, Role
o gyT cells in umo immuni y and hei con ol by NK ecep o s,
Mic obes In ec . 1 (1999) 227–234.
[33] M. Fe a ini, E. Fe e o, L. Dagna, A. Poggi, M.R. Zocchi, Human gy
T cells: a non edundan sys em in he immune-su eillance agains
cance , T ends Immunol. 23 (2002) 14–18.
[34] A.S. Ensslin, B. Fo mby, Compa ison o cy oly ic and p oli e a i e
ac i i ies o human gyand ahT cells om pe iphe al blood agains
a ious human umo cell lines, J. Na l. Cance Ins . 83 (1991)
1564–1569.
[35] L. Di Fab izio, Y. Kimu a, R. Wa e, L. Rogozinski, L. Chess, Speci ic
igge ing o gyT cells by K562 ac i a es he gyT cell ecep o and
may egula e na u al kille -like unc ion, J. Immunol. 146 (1991)
2495–2503.
[36] I. Kau , J. de Jong, K. Schell, J. Hank, P. Fisch, P.M. Sondel, Human
pe iphe al gyT cells a e s imula ed by Daudi Bu ki ’s lymphoma and
no by any o he Bu ki ’s lymphoma es ed, Cell. Immunol. 156
(1994) 54–61.
[37] S.M. Alam, J.S. Cla k, V. Leech, P. Whi o d, W.D. Geo ge, A.M.
Campbell, T cell ecep o gyexp ession on lymphocy e popula ions o
b eas cance pa ien s, Immunol. Le . 31 (1992) 279– 283.
T. Kenna e al. / Clinical Immunology 113 (2004) 56–6362
[38] I. Bank, M. Book, M. Husza , Y. Ba am, I. Schni e , H. B enne , Vy2
+
gyT lymphocy es a e cy o oxic o he MCF 7 b eas ca cinoma cell
line and can be de ec ed among he T cells ha in il a e b eas umo s,
Clin. Immunol. Immunopa hol. 67 (1993) 17 –24.
[39] D. Mi opoulos, S. Kooi, J. Rod iguez-Villanue a, C.D. Pla soucas,
Cha ac e iza ion o esh (uncul u ed) umou -in il a ing lympho-
cy es (TIL) and TIL-de i ed T cell lines om pa ien s wi h enal cell
ca cinoma, Clin. Exp. Immunol. 97 (1994) 321–327.
[40] J. Ki ayama, Y. A omi, H. Nagawa, A. Ku oda, T. Mu oh, M. Minami,
T. Juji, Func ional analysis o TCR gy
+
T cells in umou -in il a ing
lymphocy es (TIL) o human panc ea ic cance , Clin. Exp. Immunol.
93 (1993) 442–447.
[41] S. Miesche , M. Sch eye , C. Ba as, P. Capasso, V. on Fliedne ,
Spa se dis ibu ion o gyT lymphocy es a ound human epi helial
umo s p edominan ly in il a ed by p imed/memo y T cells, Cance
Immunol. Immuno he . 32 (1990) 81–87.
[42] M. Nanno, H. Seki, G. Ma hioudakis, R. Suzuki, K. I oh, C.G. Ioan-
nides, S. Suzuki, P.F. Chen, C.D. Pla soucas, gyT cell an igen ecep-
o s exp essed on umo -in il a ing lymphocy es om pa ien s wi h
solid umo s, Eu . J. Immunol. 22 (1992) 679–687.
[43] D. Kowalczyk, W. Sko upski, M. D ews, J. Nowak, Di e en pa e n
o T cell ecep o del a gene ea angemen in umou -in il a ing
lymphocy es and pe iphe al blood in pa ien s wi h solid umou s,
Cance Immunol. Immuno he . 39 (1994) 178–275.
[44] S. Seki, T. Abo, T. Masuda, T. Oh eki, A. Kanno, K. Takeda, H.
Rikiishi, H. Nagu a, K. Kumagai, Iden i ica ion o ac i a ed T cell
ecep o gylymphocy es in he li e o umo -bea ing hos s, J. Clin.
In es . 86 (1990) 409–415.
[45] S. Seki, T. Abo, K. Sugiu a, T. Oh eki, T. Koba a, H. Yagi a, K.
Okumu a, H. Rikiishi, T. Masuda, K. Kumagai, Recip ocal T cell
esponses in he li e and hymus o mice injec ed wi h syngeneic
umo cells, Cell. Immunol. 137 (1991) 46–60.
[46] C. Ag a i, G. D’O izi, P. Na ciso, S. Ab ignani, G. Ippoli o, V. Co-
lizzi, F. Poccia, Vy1 T lymphocy es exp essing a Th1 pheno ype a e
he majo gyT cell subse in il a ing he li e o HCV-in ec ed pe -
sons, Mol. Med. 7 (2001) 11–19.
T. Kenna e al. / Clinical Immunology 113 (2004) 56–63 63