This information is current as
of June 15, 2017.
Cutting Edge: Expression of Functional
CD94/NKG2A Inhibitory Receptors on Fetal
NK1.1+Ly-49− Cells: A Possible Mechanism
of Tolerance During NK Cell Development
P. V. Sivakumar, A. Gunturi, M. Salcedo, J. D. Schatzle, W.
C. Lai, Z. Kurepa, L. Pitcher, M. S. Seaman, F. A.
Lemonnier, M. Bennett, J. Forman and V. Kumar
J Immunol 1999; 162:6976-6980; ;
http://www.jimmunol.org/content/162/12/6976
Subscription
Permissions
Email Alerts
This article cites 22 articles, 12 of which you can access for free at:
http://www.jimmunol.org/content/162/12/6976.full#ref-list-1
Information about subscribing to The Journal of Immunology is online at:
http://jimmunol.org/subscription
Submit copyright permission requests at:
http://www.aai.org/About/Publications/JI/copyright.html
Receive free email-alerts when new articles cite this article. Sign up at:
http://jimmunol.org/alerts
The Journal of Immunology is published twice each month by
The American Association of Immunologists, Inc.,
1451 Rockville Pike, Suite 650, Rockville, MD 20852
Copyright © 1999 by The American Association of
Immunologists All rights reserved.
Print ISSN: 0022-1767 Online ISSN: 1550-6606.
Downloaded from http://www.jimmunol.org/ by guest on June 15, 2017
References
●
Cutting Edge: Expression of Functional
CD94/NKG2A Inhibitory Receptors on
Fetal NK1.11Ly-492 Cells: A Possible
Mechanism of Tolerance During NK
Cell Development1
P. V. Sivakumar,2* A. Gunturi,† M. Salcedo,‡
J. D. Schatzle,* W. C. Lai,* Z. Kurepa,† L. Pitcher,†
M. S. Seaman,† F. A. Lemonnier,‡ M. Bennett,*
J. Forman,† and V. Kumar*
M
urine NK cell effector function is controlled by both
positive and negative signaling receptors. These include Ly-49, NKR-P1, and the CD94/NKG2 receptors
(1). In the mouse, the ability of NK cells to lyse class I-deficient or
class I-allogeneic targets has been largely attributed to the presence or absence of positive and negative signaling receptors belonging to the Ly-49 family (2). Thus, NK cells lyse those targets
*Department of Pathology and †Center for Immunology, University of Texas Southwestern Medical Center, Dallas, TX 75235; and ‡Unité de Biologie Moléculaire du
Gène, Institut National de la Santé et de la Recherche Médicale U277, Institut Pasteur,
Paris, France
Received for publication March 17, 1999. Accepted for publication April 20, 1999.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
1
This work was supported by Grants AI20451, AI38938, CA36922, and CA70134 to
V.K. and M.B and by Grants AI37942 and AI37818 to J.F.
2
Address correspondence and reprint requests to P. V. Sivakumar, Department of
Pathology, University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, TX 75235-9072. E-mail address: [email protected]
Copyright © 1999 by The American Association of Immunologists
●
that do not express an inhibitory ligand for the Ly-49 molecules
expressed on effector cells.
We earlier described the generation of NK1.11 cells from murine fetal liver and thymus (3–5). These cells do not express detectable levels of known Ly-49 receptors (A, C, G2, D, H, and I)
on the surface. However, they do contain transcripts of certain
Ly-49 molecules, especially Ly-49E (6). Whether these are translated and are expressed on the cell surface is unknown. During
ontogeny, Ly-49 receptor expression is first detected on splenic
NK1.11 cells on days 3–5 after birth and reaches adult frequencies
by days 18 –21 (5). Despite the absence of known Ly-49 molecules, fetal- and neonatal-derived NK1.11Ly-492 cells distinguish
between class Ihigh- and class Ilow-expressing cells (5, 6). This
suggested that immature NK cells express class I inhibitory receptors that are different from the known Ly-49 molecules. In an effort
to identify the inhibitory receptor(s), we report here that a majority
of fetal NK1.11Ly-492 cells express the CD94/NKG2 receptor
complex. Further, lysis of targets by fetal NK1.11Ly-492 cells, as
well as mature splenic NK1.11Ly-492 cells, is inhibited by the
nonclassical class I molecule Qa1b. More importantly, we demonstrate that, during NK cell development, the CD94/NKG2A receptors are expressed earlier than Ly-49 molecules and may serve to
maintain tolerance to self in the absence of Ly-49 receptors.
Materials and Methods
PCR
CD94 was cloned from fetal and adult NK cells using primers designed
from sequences found in the EST database. NKG2A primers were designed
from sequences kindly provided by Dr. Fumio Takei (Terry Fox Laboratory, Vancouver, Canada). PCR products were TA cloned into pGEM TEasy vector (Invitrogen, San Diego, CA) and sequenced using T7 primers.
The following primers were used: CD94-1 (F), 59-GGATCCCTTCTCAT
GGCAGTTTCTAGG-39; CD94-3 (R), 59-GGATCCTTAAATAGGCAG
TTTCTTACA-39; NKG2A-1 (F), 59-GGATCCATGAGTAATGAACG
CGTC-39; NKG2A-2 (R), 59-GAATTCCTACTTGTCATCGTCGTCCTT
GTAATCGATGGGGAATTTACA-39
Effector cells
Mouse fetal thymic- and liver-derived NK populations were derived by
culture in 500 U/ml IL-2 for 10 –14 days as previously described (3–5).
Cells were then phenotyped for expression of NK1.1, Ly-49 (A, C and I,
G2 and D) and their ability to bind Qa1b tetramers. These cells were used
as effectors in cytotoxicity assays or for Western analysis as described.
0022-1767/99/$02.00
Downloaded from http://www.jimmunol.org/ by guest on June 15, 2017
Fetal liver- and thymus-derived NK1.11 cells do not express
known Ly-49 receptors. Despite the absence of Ly-49 inhibitory receptors, fetal and neonatal NK1.11Ly-492 cells can distinguish between class Ihigh and class Ilow target cells, suggesting the existence of other class I-specific inhibitory receptors.
We demonstrate that fetal NK1.11Ly-492 cell lysates contain
CD94 protein and that a significant proportion of fetal NK
cells are bound by Qa1b tetramers. Fetal and adult NK cells
efficiently lyse lymphoblasts from Kb2/2Db2/2 mice. Qa1bspecific peptides Qdm and HLA-CW4 leader peptide specifically inhibited the lysis of these blasts by adult and fetal NK
cells. Qdm peptide also inhibited the lysis of Qa1b-transfected
human 721.221 cells by fetal NK cells. Taken together, these
results suggest that the CD94/NKG2A receptor complex is the
major known inhibitory receptor for class I (Qa1b) molecules
on developing fetal NK cells. The Journal of Immunology,
1999, 162: 6976 – 6980.
The Journal of Immunology
6977
Lymphokine-activated killers were generated by culturing C57BL/6 spleen
cells in 500 U/ml IL-2 for 4 days.
Target cells
T cell blasts from Kb2/2Db2/2 mice (7) were generated as previously
described (8). The classical class I-deficient human B cell tumor 721.221
was a kind gift from Dr L. Lanier (DNAX, Palo Alto, CA). 721.Qa1b cells
were obtained by transfecting 721.221 cells with a mQa1b construct
(pCDNA3.1 expression vector; Invitrogen), selected in 1 mg/ml Geneticin
(Life Technologies, Gaithersburg, MD) and analyzed by cell staining as
specified below.
Abs, cell staining, and Western analysis
Peptides and killing assays
The Qdm (AMAPRTLLL, Qa1b-specific), HLA-CW4 (VMAPRTLIL,
Qa1b-specific), FLU (ASNENMETM, Db-specific), OVA (SIINFEKL, Kbspecific), and HIV-RT (ILKEPVHGV, HLA-A2-specific) peptides were
synthesized on an ABS 432A peptide synthesizer and purified using HPLC
in our laboratory. Target cells were incubated overnight at 26°C with or
without peptides. In some experiments 100 mM peptides were added overnight. In others, specified concentrations of peptides were added to cells for
1–2 h at room temperature before addition to effectors. Standard 4-h chromium release assays were done as described (8).
Results
Fetal liver- and thymus-derived NK cells express CD94 and
NKG2A transcripts
We considered the murine CD94/NKG2A heterodimer as a candidate inhibitory receptor on fetal and neonatal NK1.11Ly-492
cells. Using primers designed from the EST database, we cloned
murine CD94 from C57BL/6 fetal liver- and thymus-derived NK
cells (fetal liver NK (FL-NK)3 and fetal thymic NK (FT-NK),
respectively) and also from adult splenic NK cells. FL-NK and
FT-NK cells expressed CD94 transcripts, and the sequence obtained was identical to the published sequence (12, 13) (data not
shown). We also verified that FL-NK and FT-NK cells expressed
mNKG2A by using primers designed from an NKG2A sequence
that was kindly shared by Dr. Takei (Terry Fox Lab, Vancouver,
BC, Canada)(data not shown).
FL-NK, FT-NK, and adult splenic NK cell lysates express a
protein that is identified by an anti-CD94 antiserum
To obtain a rabbit Ab against CD94, we synthesized two peptides
from the extracellular domain of mCD94 as described under Materials and Methods. Sera from rabbits immunized with the two
different peptides (CD94 –2 and CD94 –3) reacted against a CD943
Abbreviations used in this paper: FL-NK, fetal liver NK; FT-NK, fetal thymic NK;
NFDM, nonfat dry milk.
FIGURE 1. FL-NK, FT-NK, and scid NK cells express CD94 protein.
Lysates from 5 million cell equivalents were separated on a 4 –20% SDS
gel and transferred to nitrocellulose membrane. After blocking in 3%
NFDM in TBST, anti-CD94-2 or anti-CD94-3 antisera were added at a
1:200 dilution. Donkey anti-rabbit HRP was used at 1:1000. CD94-GST
fusion protein was used as a positive control. Control NKR-P1B-GST and
control-GST fusion protein did not react with antisera (data not shown).
GST fusion protein by ELISA (data not shown) and Western analysis (Fig. 1). FL-NK, FT-NK, and adult splenic NK cell lysates
were also analyzed for reactivity to anti-CD94 antisera. As shown
in Fig. 1, both anti-CD94-2 and anti-CD94-3 antisera reacted with
a band that corresponds to the predicted molecular mass of mCD94
(;28 –30 kDa) (Fig. 1). Anti-CD94-3 also reacted with some
smaller size bands (Fig. 1, bottom). Although these bands could
have resulted from nonspecific binding or degradation products,
we believe that they might be alternatively spliced forms of CD94
(P. V. Sivakumar and V. Kumar, unpublished data). The antiCD94 antiserum did not cross-react with rat CD94 expressed in the
rat leukemia cell line RNK-16 (14). Mouse T cell tumors BW5147
and EL-4 were also CD94 negative (Fig. 1). There was no staining
of NK cells as determined by FACs analysis, suggesting that the
anti-CD94 antiserum was unable to recognize protein in the native
configuration.
Qa1b tetramers bind to FL-NK, FT-NK, and adult splenic NK
cells
The nonclassical class I molecule Qa1b has been identified recently
to be the ligand for mCD94/NKG2A (15). The Qa1b tetramer binds
to the CD94/NKG2A complex but not to CD94 alone (15). Qa1b
tetramers folded in the presence of Qdm peptide (Qa1b-specific,
high affinity Db, Dd leader peptide) (10) bound to ;50 – 60% of
NK1.11Ly-492 cells derived from fetal liver and thymus (Fig. 2).
As previously reported, it also bound to ;60% of C57BL/6 splenic
NK1.11 cells (9, 15) (Fig. 2). A control HLA-A2 tetramer refolded
with the HIV-GAG peptide did not bind to splenic or fetal NK1.11
cells (data not shown). In multiple experiments, between 40 and
70% of IL-2-cultured fetal NK cells were bound by the tetramer.
Thus, a significant proportion of fetal liver- and thymus-derived
NK1.11Ly-492 cells express CD94/NKG2 on the surface. The
Qa1b tetramers also bound to 40 – 60% of NK1.11 cells from fresh
day 15 fetal liver and thymus (Fig. 2, and data not shown), suggesting that culture in IL-2 (FL-NK and FT-NK) does not alter the
frequencies of NK1.11 cells bound by tetramer. Based on data
from Vance et al. (15) and Braud et al. (16), the detection of
NKG2A transcripts in FL-NK, and the functional data presented
below (Figs. 3 and 4), we believe that CD94 is in a heterodimeric
complex with NKG2A or another inhibitory NKG2 receptor.
Downloaded from http://www.jimmunol.org/ by guest on June 15, 2017
Anti-CD94 polyclonal sera were generated, by immunizing rabbits with
keyhole limpet hemocyanin (KLH)-conjugated peptides CD94 –
2(CIVYSPSKSVSAESCENKNRYICKK) (aa 152–176) or CD94 –3 (CS
KEEKSWKRSRDFC) (aa 76 – 89). Rabbit sera were then analyzed for
specificity by ELISA using a CD94-GST fusion protein control and also by
Western blot, as described. The CD94-GST fusion protein was generated
by cloning full length mCD94 into the pESP-1 expression vector (Stratagene, La Jolla, CA). For cell staining, the FcR on the NK cells was
blocked using mouse serum (1:10 dilution) for 20 min. Red-670-conjugated Qa1b tetramers were derived as previously described (9). After
blocking, cells were incubated with Qa1b tetramer, control HLA-A2 tetramer, or Red-670 alone for 45 min, washed once and incubated with
PE-conjugated NK1.1 (PharMingen, San Diego, CA) for 30 min. Cells
were then washed and analyzed by FACScan (BD Systems, Mountain
View, CA). 721.221 or 721.Qa1b cells were stained with biotinylated antiQa1b or rat isotype control Ab followed by streptavidin-conjugated PE.
Anti-Qa1b Ab was generated by immunizing rats with soluble mouse Qa1b
produced in Drosophila cells (10). This mAb, 910, recognizes plate-bound
Qa1b as well as Qa1b expressed on transfectants and mouse lymphoblasts
(see Figs. 3 and 4; A. Gunturi, Z. Kurepa, and J. Forman, unpublished
data). Western analysis was done as described (11).
6978
Qa1b-specific peptides inhibit the lysis of Qa1b-transfected
tumor cells by FL-NK and FT-NK cells
Having established that the Qa1b tetramer binds to FL-NK and
FT-NK cells, we analyzed their ability to receive negative or positive signals from the nonclassical class I molecule Qa1b. There are
no Abs available at present against Qa1b and CD94/NKG2 that can
be used to block interactions. Therefore, we used peptide stability
assays to look for the ability of Qa1b to inhibit NK cell lysis. For
these studies, we chose the human classical class Ia-deficient B cell
line 721.221, for several reasons. First, a human cell line would not
express mouse surface receptors, thereby minimizing cross-reactivity. Second, these cells do not express the dominant leader peptides that bind to Qa1b since they are deficient for human classical
FIGURE 4. Lysis of class I-deficient T cell blasts is inhibited by Qa1bspecific peptides. A, Spleen cells from wild-type C57BL/6 (H2b), TAP2/2
(H2b), and Kb2/2Db2/2 (H2b) mice were cultured for 2 days in 3 mg/ml
Con A. Viable cells were then used for FACS analysis using no Ab (light,
unshaded), isotype control (dark, unshaded), or anti-Qa1b mAb 910 (dark,
shaded) followed by streptavidin–PE. B, Viable T cell blasts were labeled
with 51Cr at 37°C in the presence of 100 mM peptide, washed, and plated
at 1000 targets/well with 1000 ng of peptide for 1 h. Effector cells were
then added (FT-NK 200:1, adult splenic lymphokine-activated killers 100:
1). A standard 4-h chromium release assay was performed.
class Ia molecules. Although they do express the human class Ib
molecule HLA-E, its leader should not bind Qa1b. 721.221 cells
were transfected with mQa1b (721.Qa1b) and analyzed for Qa1b
expression by flow cytometry using an anti-Qa1b mAb (Fig. 3A).
FL-NK cells were used as effectors against these two cell lines
(wild-type 721.221 and 721.Qa1b) in the absence or presence of
different peptides. FL-NK cells efficiently lysed untransfected
721.221 cells. Addition of the Qa1b-binding Qdm peptide (AMAPRTLLL) or control peptides FLU, OVA, and HIV-RT did not
have any effect on NK killing (Fig. 3B, left). 721.Qa1b cells not
pulsed with peptides were lysed equally well. However, addition
of Qdm (but not FLU, OVA, and HIV-RT) peptides dramatically
inhibited lysis of 721.Qa1b targets (Fig. 3B, right). Similar results
were obtained when FT-NK or adult splenic NK cells were used as
effectors (data not shown). Another Qa1b-binding peptide (HLACW4 leader) (17) also inhibited lysis (data not shown). These data
provide strong evidence that the predominant class Ib-recognizing
inhibitory receptor on these fetal NK cells was CD94/NKG2A.
Similar results were obtained using human T2 cells transfected
with murine Qa1b (data not shown).
Qa1b-specific peptides inhibit lysis of T cell blasts by FT-NK
and adult splenic NK cells
FIGURE 3. The Qa1b-specific peptide Qdm inhibits lysis of 721.Qa1b
cells by FL-NKs. A, The human class I-deficient (except HLA-E) 721.221
cells were transfected with mQa1b, selected in Geneticin, and analyzed for
Qa1b expression by flow cytometry. Cells were stained with biotinylated
isotype control (Rat IgG2) (gray, shaded) or with biotinylated anti-Qa1b Ab
(1:50) (black, unshaded), followed by streptavidin-PE. Untransfected cells
(left) are not bound by anti-Qa1b Ab whereas 721.Qa1b cells (right) express
Qa1b. B, Qdm peptide specifically inhibits lysis of 721.Qa1b but not
721.221 cells by FL-NKs. FL-NK cells cultured in 500 U/ml of IL-2 for 10
days were used as effectors in a standard 4-h chromium release assay
against wild-type 721.221 (left panel) and 721.Qa1b (right panel) cells.
Target cells were incubated overnight at 26°C with 100 mM of peptide.
Cells were labeled with 51Cr for 2 h at room temperature and used as
targets. Peptides (500 ng/well) were also added to the wells during the
assay.
To further establish that FL-NK and FT-NK cells can be inhibited
by Qa1b, we used T cell blasts as targets. For this, we chose cells
from Kb2/2Db2/2 mice (7). These mice offer two advantages.
First, they lack classical class I molecules and, therefore, are lysed
efficiently by adult splenic NK cells as well as fetal NK cells (Fig.
4B, and data not shown). Second, they lack the dominant Qa1b
leader peptide (Db leader, Qdm). We also infer this from our observations that Qdm-specific T cell clones do not lyse blasts from
Kb2/2Db2/2 mice unless Qdm is added to the assay (M. S. Seaman and J. Forman, unpublished data). Qdm peptide can therefore
be loaded on the Qa1b heavy chain molecules, presumably by displacing other lower affinity peptides. We first analyzed cell surface
expression of Qa1b on LPS blasts from Kb2/2Db2/2 (H2b) mice.
As shown in Fig. 4A, wild-type C57BL/6 (H2b) mice expressed
Qa1b but TAP2/2 (H2b) mice do not. Even in the absence of the
dominant leader peptide (Qdm), Kb2/2Db2/2 lymphoblasts express Qa1b on the cell surface (Fig. 4A), confirming observations
Downloaded from http://www.jimmunol.org/ by guest on June 15, 2017
FIGURE 2. Qa1b tetramers bind to fetal and adult NK cells. C57BL/6
fresh fetal liver (a), IL-2-cultured FL-NK (b), and FT-NK (c) cells were
stained with Red-670 alone (gray, shaded) or Red-670 conjugated Qa1b
tetramers (dark line, unshaded)(1:100 dilution) followed by PE-conjugated
NK1.1. Fresh C57BL/6 spleen cells were used as a control. Qa1b tetramer
binding is shown on gated NK1.11 cells in all the panels. A control
HLA-A2 tetramer refolded with the HIV-GAG peptide did not bind to
murine NK cells (data not shown).
CUTTING EDGE
The Journal of Immunology
Discussion
Fetal and neonatal NK1.11Ly-492 cells can distinguish between
class Ihigh and class Ilow targets. They are unable to lyse RMA cells
but lyse the TAP-deficient derivative RMA-S cells efficiently.
They are also unable to lyse wild-type syngeneic (H2b) or allogeneic (H2d) class I-positive lymphoblasts but lyse congenic class
Ilow T cell blasts from TAP2/2 (H2b) mice. This strongly suggests
the existence of non-Ly-49 receptors that can receive inhibitory
signals from class I on these cells. Several pieces of evidence
indicate that the CD94/NKG2 receptor complex acts as the major
inhibitory receptor complex for class I on these NK cells. First,
fetal liver- and thymus-derived NK cells express CD94 protein
(Fig. 1), and a significant proportion (between 40 –70%) of IL-2activated fetal NK cells as well as 60% of fresh fetal NK1.11 cells
(Fig. 2) are bound by Qa1b tetramers. Second, we can detect
NKG2A transcripts in these cells. Based on this as well as the data
of Vance et al. (15), we suggest that CD94 is expressed in a complex with NKG2A on the surface. However, it is possible that other
inhibitory NKG2 receptors may also be expressed. Third, we
found using two different systems (tumor cell lines as well as
blasts) that Qa1b-specific peptides greatly inhibit lytic activity
of fetal NK cells. Lysis of the Qa1b transfectant 721.Qa1b is reduced by Qdm but not control peptides. Additionally, lysis of
Kb2/2Db2/2 blasts was almost completely abrogated when pulsed
with Qdm and HLA-CW4 peptides. It would therefore seem that in
the absence of any known Ly-49-class I-mediated inhibition, the
Qa1b-CD94/NKG2 receptor complex can mediate strong negative
signals to NK cells. This has implications not only for NK cell
function but also for the ontogeny of NK cell receptors.
During development, the CD94/NKG2 receptor complex is expressed well before the Ly-49 receptor system. We and others have
demonstrated that Ly-49 receptor expression can be first detected
only after birth and reaches adult levels by days 18 –21 (5, 18). We
have also shown that fetal and neonatal NK1.11Ly-492 cells are
functional and can lyse class I-deficient targets (5, 6). It would
therefore be necessary to have a tolerance mechanism in the absence of Ly-49-class I interactions. This seems to be provided by
the CD94/NKG2A receptor interacting with Qa1b. Support for this
is strengthened by the fact that immature NK1.11Ly-492 cells
derived from marrow progenitors (19) can also distinguish between class Ihigh and class Ilow targets, and these cells are bound by
Qa1b tetramer (N. S. Williams, unpublished data). Because the
frequency of cells expressing surface CD94/NKG2 (as measured
by Qa1b tetramer binding) is the same in Ly-492 fetal NK cells
and Ly-491 adult NK cells, there does not seem to be a developmental regulation in the frequency of cells expressing CD94/
NKG2. However, the relative contribution of these two receptor
systems in inhibition of NK function in adult Ly-491 cells is unknown. Both receptor systems could contribute to self tolerance.
Additional experiments are needed to address this. However, in
fetal life, in the absence of any known Ly-49 receptors, the CD94/
NKG2 inhibitory system could act as the major inhibitory system
to maintain tolerance. One other important aspect that needs to be
considered is the presence, of other class I-specific inhibitory receptors on fetal NK cells. Approximately 50% of fetal NK cells are
Qa1b tetramer positive. Does that mean that the other 50% do not
distinguish between class I expressing cells? This seems unlikely.
One possibility is that these cells express CD94 complexed with
other NKG2 receptor family members that are not recognized by
Qdm-bound Qa1b tetramers. It should be noted that all tetramer
staining done so far has used refolded Qa1b in the presence of Qdm
peptide. It is possible that there are other peptides that are important in the recognition by CD94/NKG2B or CD94/NKG2C in contrast to CD94/NKG2A. Alternatively, the Qa1b tetramer-negative
cells express other class I-specific inhibitory receptors. Potential
candidate receptors would include the murine homologue of ILT2
(LIR1) and gp49B1 (20, 21). Another possible candidate could be
the Ly-49E receptor. Toomey et al. have reported that fetal NK
cells show higher expression of Ly-49E transcripts when compared with adult NK cells (6). It is possible that Ly-49E acts as
another class I-specific inhibitory receptor. The absence of Ly-49E
Abs prevents testing this possibility at this time.
To summarize, it seems that the CD94/NKG2A receptor complex acts as a dominant inhibitory receptor complex during NK
cell development in the mouse. This conclusion is supported by the
fact that NK cells derived from immature human fetal thymocytes
have been shown to express the CD94/NKG2A complex as the
only class I-specific receptor (22). We hypothesize therefore that
inhibitory CD94/NKG2 receptors, by interacting with the nonclassical class I molecule Qa1b, play an important role in maintaining
self tolerance in developing NK cells.
Acknowledgments
We thank Dr. Brian Gordon for providing SCID mice and Dr. Colin Brooks
for sharing unpublished data. We also thank Ying Zhang for technical
assistance and Maria and Sylvio Pena for breeding mice. We appreciate the
helpful discussions of Drs. Dorothy Yuan and Noelle Williams.
References
1. Lanier, L. L. 1998. NK cell receptors. Annu. Rev. Immunol. 16:359.
2. Yokoyama, W. M. 1997. What goes up must come down: the emerging spectrum
of inhibitory receptors. J. Exp. Med. 186:1803.
3. Brooks, C. G., A. Georgio, and R. K. Jordan. 1993. The majority of immature
fetal thymocytes can be induced to proliferate to IL-2 and differentiate into cells
indistinguishable from mature NK cells. J. Immunol. 151:6645.
4. Manoussaka, M., A. Georgiou, B. Rossiter, S. Shrestha, J. A. Toomey,
P. V. Sivakumar, M. Bennett, V. Kumar, and C. G. Brooks. 1997. Phenotypic and
functional characterization of long-lived NK cell lines of different maturational
status obtained from mouse fetal liver. J. Immunol. 158:112.
5. Sivakumar, P. V., M. Bennett, and V. Kumar. 1997. Fetal and neonatal NK1.11
Ly-492 cells can distinguish between major histocompatibility complex class
I(hi) and class I(lo) target cells: evidence for a Ly-49-independent negative signaling receptor. Eur. J. Immunol. 27:3100.
Downloaded from http://www.jimmunol.org/ by guest on June 15, 2017
of Perarnau et al. (7). Because mAb 910 recognizes a peptideindependent epitope on Qa1b, this result is not surprising. Further,
preliminary evidence indicates that Qa1b is associated with other
self peptides on these cells, which could account for the surface
expression of this molecule (M. Seaman and J. Forman, unpublished data). Addition of Qdm peptide did not significantly alter
Qa1b cell surface levels (data not shown). Fetal thymic and adult
splenic NK cells lysed blasts from Kb2/2Db2/2 mice. Addition of
Qdm or HLA-CW4 leader peptides greatly inhibited killing of the
blasts (Fig. 4B). These results suggest that the dominant Qdm peptide could displace other self peptides bound to Qa1b on these cells
and inhibit target cell lysis. The HLA-CW4 peptide that can bind
Qa1b with high affinity (17) could also inhibit NK cell lysis. This
demonstrates that mouse NK cells can be inhibited by peptides
from other species. Control FLU, OVA, and HIV-RT peptides
have no effect.
Similar results were obtained when B cell blasts were used as
targets (data not shown). In similar assays, there was no effect on
the lysis of TAP2/2 blasts by fetal or adult splenic NK cells by
addition of different peptides (data not shown). The inhibitory effects of the Qdm and the HLA-CW4 peptides could also be titrated
in this Kb2/2Db2/2 system (data not shown). This provides conclusive evidence that fetal NK cells can be inhibited by Qa1b and
that the CD94/NKG2–Qa1b interaction plays an important role in
preventing killing by Ly-492 NK cells.
6979
6980
15. Vance, R. E., J. R. Kraft, J. D. Altman, P. E. Jensen, and D. H. Raulet. 1998.
Mouse CD94/NKG2A is a natural killer cell receptor for the nonclassical major
histocompatibility complex (MHC) class I molecule Qa1b. J. Exp. Med. 188:
1841.
16. Braud, V. M., D. S. Allan, C. A. O’Callaghan, K. Soderstrom, A. D’Andrea,
G. S. Ogg, S. Lazetic, N. T. Young, J. I. Bell, J. H. Phillips, L. L. Lanier, and
A. J. McMichael. 1998. HLA-E binds to natural killer cell receptors CD94/
NKG2A, B and C. Nature 391:795.
17. Kurepa, Z., C. A. Hasemann, and J. Forman. 1998. Qa1b binds conserved class
I leader peptides derived from several mammalian species. J. Exp. Med. 188:973.
18. Dorfman, J. R., and D. H. Raulet. 1998. Acquisition of Ly49 receptor expression
by developing natural killer cells. J. Exp. Med. 187:609.
19. Williams, N. S., T. A. Moore, J. D. Schatzle, I. J. Puzanov, P. V. Sivakumar,
A. Zlotnik, M. Bennett, and V. Kumar. 1997. Generation of lytic natural killer
1.11, Ly-492 cells from multipotential murine bone marrow progenitors in a
stroma-free culture: definition of cytokine requirements and developmental intermediates. J. Exp. Med. 186:1609.
20. Colonna, M., F. Navarro, T. Bellon, M. Llano, P. Garcia, J. Samaridis,
L. Angman, M. Cella, and M. Lopezbotet. 1997. A common inhibitory receptor
for major histocompatibility complex class I molecules on human lymphoid and
myelomonocytic cells. J. Exp. Med. 186:1809.
21. Wang, L. L., I. K. Mehta, P. A. LeBlanc, and W. M. Yokoyama. 1997. Mouse
natural killer cells express gp49B1, a structural homologue of human killer inhibitory receptors. J. Immunol. 158:13.
22. Mingari, M. C., C. Vitale, C. Cantoni, R. Bellomo, M. Ponte, F. Schiavetti,
S. Bertone, A. Moretta, and L. Moretta. 1997. Interleukin-15-induced maturation
of human natural killer cells from early thymic precursors: selective expression
of CD94/NKG2-A as the only HLA class I-specific inhibitory receptor. Eur.
J. Immunol. 27:1374.
Downloaded from http://www.jimmunol.org/ by guest on June 15, 2017
6. Toomey, J. A., S. Shrestha, S. A. de la Rue, F. Gays, J. H. Robinson,
Z. M. Chrzanowska-Lightowlers, and C. G. Brooks. 1998. MHC class I expression protects target cells from lysis by Ly49-deficient fetal NK cells. Eur. J. Immunol. 28:47.
7. Peranau, B., M.-F. Saron, B. R. S. Martin, N. Bervas, M. Soloski, A. G. Smith,
J. M. Ure, J. E. Gairin, and F. A. Lemonnier. 1999. Single H2Kb,H2Db and
double H2KbDb knockout mice: peripheral CD81 T cell repertoire and antilymphocytic choriomeningitis virus cytolytic responses. Eur. J. Immunol. 29:
1243.
8. Yu, Y. Y., T. George, J. R. Dorfman, J. Roland, V. Kumar, and M. Bennett. 1996.
The role of Ly49A and 5E6(Ly49C) molecules in hybrid resistance mediated by
murine natural killer cells against normal T cell blasts. Immunity 4:67.
9. Salcedo, M., P. Bousso, H. G. Ljunggren, P. Kourilsky, and J. P. Abastado. 1998.
The Qa-1(b) molecule binds to a large subpopulation of murine NK cells. Eur.
J. Immunol. 28:4356.
10. Kurepa, Z., and J. Forman. 1997. Peptide binding to the class Ib molecule, Qa1b.
J. Immunol. 158:3244.
11. Nakamura, M. C., E. C. Niemi, M. J. Fisher, L. D. Shultz, W. E. Seaman, and
J. C. Ryan. 1997. Mouse Ly49A interrupts early signaling events in NK cell
cytotoxicity and functionally associates with the SHP-1 tyrosine phosphatase.
J. Exp. Med. 185:673.
12. Vance, R. E., D. M. Tanamachi, T. Hanke, and D. H. Raulet. 1997. Cloning of
a mouse homolog of CD94 extends the family of C-type lectins on murine natural
killer cells. Eur. J. Immunol. 27:3236.
13. Ho, E. L., J. W. Heusel, M. G. Brown, K. Matsumoto, A. A. Scalzo, and
W. M. Yokoyama. 1998. Murine NKG2D and CD94 are clustered within the
natural killer complex and are expressed independently in natural killer cells.
Proc. Natl. Acad. Sci. USA 95:6320.
14. Dissen, E., S. F. Berg, I. H. Westgaard, and S. Fossum. 1997. Molecular characterization of a gene in the rat homologous to human CD94. Eur. J. Immunol.
27:2080.
CUTTING EDGE