Download The NK2.1 receptor is encoded by Ly-49C and its

iim$$$0407
International Immunology, Vol. 9, No. 4, pp. 533–540
© 1997 Oxford University Press
The NK2.1 receptor is encoded by Ly-49C
and its expression is regulated by MHC
class I alleles
Pierre Gosselin1,3, Yvette Lusignan1, Jack Brennan2, Fumio Takei2 and
Suzanne Lemieux1
1Centre
de Recherche en Immunologie, Institut Armand-Frappier, Université du Québec, Laval, Québec,
Canada H7V 1B7
2Terry Fox Laboratory, British Columbia Cancer Agency, University of British Columbia, Vancouver,
British Columbia, Canada V5Z 1L3
3Present
address: Laboratory of Experimental Immunology, Division of Basic Sciences, NCI-FCRDC,
Frederick, MD, 21702-1201, USA
Keywords: Ly-49 receptor, murine MHC molecules, NK cell
Abstract
A dual receptor system composed of activation and inhibitory receptors apparently controls NK
cell-mediated lysis. In the C57BL/6 mouse, the NK1.1 molecule acts as an activation receptor
whereas Ly-49A, C and G2 can inhibit NK cell lysis of target cells expressing specific MHC class I
molecules. We previously reported that NK2.1 is an activation receptor sharing structural
properties with members of the NKR-P1 and Ly-49 receptor families. In this study, we have shown
that NK2.1 is encoded by the previously described Ly-49C gene. We also found that the expression
level of NK2.1/Ly-49C is modulated by H-2-dependent factors and that this regulation differs from
that previously described for Ly-49A. Flow cytometry analyses of NK-enriched spleen cells from
MHC congenic strains on C57BL/10 and BALB/c backgrounds indeed revealed that the level of
NK2.1/Ly-49C expression, but not the number of positive cells, is low in strains expressing H-2b
and H-2k haplotypes as compared to H-2d mice. Similar analyses of splenic NK cells from two
series of congenic and congenic recombinant strains on the C57BL/10 background indicate that
the main regulatory element(s) are most likely the H-2Kb and H-2Dk alleles. Together with our and
others previous observations, these results identify the NK2.1/Ly-49C antigen as a receptor for
MHC class I molecules whose expression is regulated by host MHC genes.
Introduction
NK cells can kill tumor cells and virus-infected cells without
the need for prior sensitization and without the requirement
for MHC molecule expression on target cells (1). Instead, it
is of general agreement that MHC class I molecules expressed
by target cells may interrupt the lytic process of NK cells,
following their interaction with specific inhibitory receptors
(2). Ly-49A, which selectively binds H-2Dd and Dk molecules
(3), is the first mouse NK cell antigen shown to have such
inhibitory properties (4). In addition to inhibition of NK cell
lysis, the interaction between membrane-bound H-2Dd or Dk
host molecules and Ly-49A was shown to result in downregulation of Ly-49A expression on individual NK cells
although the size of the Ly-49A1 subpopulation is unaltered
(5). On the other hand, NK cell-mediated killing can be
triggered through activation receptors such as NK1.1 (6)
which is a member of the NKR-P1 family (7). The physiologic
ligand of NK1.1 is still unknown and, contrary to Ly-49A, NK1.1
expression is not influenced by host MHC haplotypes (5).
Ly-49A and NK1.1 belong to two families of genes mapping
to the distal segment of mouse chromosome 6, named the
NK gene complex, and encoding type II integral membrane
proteins homologous to C-type animal lectins (8–15). Whereas
in positive strains NK1.1 is expressed by most NK cells (16),
Ly-49A defines a small subset of splenic NK cells (12). The
Ly-49 family also includes the 5E6 and LGL-1 molecules, also
expressed by NK cell subpopulations (17,18) and encoded
by Ly-49C and Ly-49G2 genes respectively (19–21). As
reported for Ly-49A1 cells, Ly-49G21 NK cells are inhibited
Correspondence to: S. Lemieux
Transmitting editor: H. R. MacDonald
Received 26 August 1996, accepted 25 December 1996
534 MHC-dependent regulation of NK2.1/Ly-49C expression
by MHC class I molecules expressed on target cells (21).
More recently, binding experiments showed that Ly-49C binds
to the H-2Kb, Kd and Dd molecules (22). Observations made
in hybrid resistance studies further suggested that Ly-49C
can inhibit NK cell lysis upon recognition of H-2Kb and H-2d
molecules (23). Interestingly, the different members of the Ly49 family appear to have distinct but overlapping class I
specificities.
Until recently, the nature and function of the mouse NK2.1
molecule, an NK cell antigen initially identified with an NZB
anti-BALB/c antiserum (24), were still unknown. Our laboratory
has produced an anti-NK2.1 mAb (25) and characterized the
NK2.1 antigen as a highly glycosylated disulfide-linked protein
dimer of 65 kDa subunits (26). We further showed that
immobilized anti-NK2.1 mAb induced granule exocytosis from
IL-2-activated cells and that soluble mAb as well as its F(ab9)2
and Fab fragments increased lysis of NK susceptible target
cells by resting or IL-2-activated cells (27). These results are
thus consistent with NK2.1 being a relevant activation receptor
for natural killing. Similar to Ly-49 receptors characterized
thus far, NK2.1 is expressed by a subpopulation rather than
all NK cells (25–28). In addition, the percentage of NK2.11
cells in nylon-wool non-adherent spleen cells as well as the
level of NK2.1 expression are variable among positive mouse
strains (28). In the present study, we report that NK2.1 is
encoded by the Ly-49C gene and that its cell surface expression is down-regulated by host MHC genes of H-2b and H-2k
haplotypes.
Methods
Mice
AKR/N, BALB/cAnN, C57BL/6N, C3H/HeN, DBA/2N and
(BALB/cAnN3C57BL/6N)F1 mice were purchased from
Charles River Canada (St Constant, Québec, Canada). 129/
Sv, A/J, C57BL/10SnJ, CBA/J, LP/J, B10.A/SgSnJ, B10.A(2R)/
SgSnJ, B10.A(5R)/SgSnJ, B10.BR/SgSnJ, B10.D2/nSnJ,
B10.D2(R103)/EgDvEgJ, B10.D2(R107)/EgDvEgJ, BALB.B/
LiMcdJ, BALB.K/LiMcdJ and NZB/BINJ mice were obtained
from the Jackson Laboratory (Bar Harbor, ME).
Antibodies
The 4LO3311 anti-NK2.1 mAb (mouse IgG3) was generated
in our laboratory (25). The hybridoma PK136 producing antiNK1.1 mAb (mouse IgG2a) (29) was purchased from ATCC
(Rockville, MD). These mAb were purified by affinity chromatography and biotinylated using standard methods. Biotinylated
5E6 mAb (mouse IgG2a) (17) was obtained from PharMingen
(San Diego, CA). Phycoerythrin (PE)-labeled A1 mAb (mouse
IgG2a) (8) was kindly provided by Dr K. P. Kane (University
of Alberta, Edmonton, Alberta, Canada). Irrelevant biotinylated
mouse IgG3 mAb, purchased from PharMingen, and mouse
IgG2a mAb, generously provided by Dr P. J. Talbot (Institut
Armand-Frappier), were used as isotype controls.
Enrichment of splenic NK cells
Splenic NK cells were enriched by selective depletion of T
lymphocytes from nylon-wool non-adherent cells using antiCD4 and anti-CD8 rat mAb and sheep anti-rat IgG-coated
magnetic beads (Dynal, Great Neck, NY) as previously
described (26). We reported elsewhere that .90% of these
cells react with anti-asialo GM1 antiserum (27).
NK cell staining and flow cytometry analysis
Splenic NK cells were incubated for 30 min on ice with optimal
concentrations of PE-labeled A1 mAb or biotinylated mAb.
Binding of biotinylated mAb was detected with PE-labeled
streptavidin (Becton Dickinson, Mountain View, CA). Control
samples were incubated with biotinylated isotype control mAb
and streptavidin–PE. In the phenotypic study of (1293C57BL/
6)F13129 backcross animals, samples from individual mice
were analyzed on a Coulter Epics C flow cytometer (Coulter
Electronics, Hialeah, FL) equipped with a water-cooled 5 W
argon laser of 488 nm. The analysis of NK2.1 expression in
inbred, congenic and hybrid mice was performed with a
Coulter Epics XL equipped with an air-cooled 15 mW argon
laser of 488 nm. Results are expressed as the percentage of
lymphocytes, defined by forward and side scatter gates,
which reacted with anti-NK mAb. Data analysis based on
collection of 10,000 events/sample was done with XL software.
The cytometers used for NK cell analysis were calibrated
daily with 10 µm Coulter DNA-check beads (Coulter, Hialeah,
FL) using a standard protocol.
COS cell transfections and analyses
cDNAs encoding individual members of the Ly-49 family
(Ly-49AB6, B, CBALB/CBA, DB6, EB6, FB6, G4B6 and H) were
subcloned into the expression vector pAX142 and COS cells
were transfected by the DEAE–dextran method as previously
described (19). The Ly-49EB6 and Ly-49FB6 cDNAs (15) were
obtained from Dr W. M. Yokoyama (Washington University, St
Louis, MO). Cells were analyzed for cell surface expression
by flow cytometry 72 h post-transfection. R-PE-conjugated
5E6 and FITC-labeled A1 were purchased from PharMingen
(San Diego, CA). Cell staining procedures were carried out
for 30 min at 4°C at concentrations of 107 cells/ml, followed
by two washes in PBS containing 2% FCS. Those samples
stained with biotinylated 4LO3311 required a secondary step
with streptavidin–FITC. Analysis was performed on a FACSort
(Becton Dickinson), and dead cells were stained with propidium iodide (1 µg/ml in final wash) and gated out.
Results
Linkage of the NK2.1 gene to the NK gene complex
To investigate whether the NK2.1 gene is associated with
those of the NK gene complex, the expression of NK1.1
(musNKR-P1C), 5E6 (Ly-49C) and NK2.1 on NK-enriched
spleen cells from 80 individual mice of the (1293C57BL/
6)F13129 backcross progeny was determined by flow cytometry analysis. This strain combination was suitable for a
linkage study since 129 mice are NK1.1–5E6–NK2.1– whereas
all three antigens are expressed in C57BL/6 (17,24,25,30).
Because only a low number of dimly fluorescent cells was
detected after incubation of C57BL/6 nylon-wool non-adherent
spleen cells with 4LO3311 anti-NK2.1 mAb (28), the backcross
analysis was performed on NK-enriched spleen cells in order
to get more reliable results. Our data showed that musNKR-
MHC-dependent regulation of NK2.1/Ly-49C expression 535
Table 1. Expression of NK1.1, 5E6 and NK2.1 antigens in the
(1293C57BL/6)F13129 backcross progenya
NK2.11
NK2.1–
Total
NK1.11/5E61
NK1.1–/5E6–
Total
36
0
36
0
44
44
36
44
80
aExpression of NK1.1, NK2.1 and 5E6 was determined by flow
cytometry analysis of NK-enriched spleen cells from individual animals
as described in Methods. Results shown correspond to the number
of mice expressing a given phenotype.
P1C, Ly-49C and NK2.1 genes never segregated in backcross
mice (Table 1). In agreement with these results, a separate
study from our laboratory revealed that none of the recombinant inbred mice of the 129XB6 series, kindly provided by Dr
J.-L. Guenet (Institut Pasteur, Paris, France), were NK1.1 or
NK2.1 single positive (unpublished observations). It thus
appears that the NK2.1 gene maps to chromosome 6 either
near or within the NK gene complex, in association with the
NKR-P1 and Ly-49 gene families.
Identification of NK2.1 as Ly-49C
Several features of NK2.1 initially suggested that it might be
encoded by a member of the NKR-P1 or Ly-49 gene families.
In addition to its chromosomal localization to the NK gene
complex, NK2.1 has been shown to be a dimeric cell surface
antigen expressed by a subset of NK cells (26). The fact that
NK2.1 is expressed in strains that lack NKR-P1 transcripts
(31) led us to first investigate the Ly-49 gene family. Individual
Ly-49 cDNAs were transiently expressed in COS cells which
were subsequently tested for reactivity with the anti-NK2.1
mAb. 4LO3311 was found to react specifically with Ly-49C
(Fig. 1) but none of the other Ly-49s tested (data not shown
for Ly-49B, D, E, F, G4 and H). As reported previously (10,19),
the A1 mAb specifically bound Ly-49A and the 5E6 mAb also
recognized Ly-49C (Fig. 1). Although both 5E6 and 4LO3311
mAb apparently recognize the same Ly-49 molecule, they
likely detect different epitopes of that polymorphic receptor.
This is suggested by the existence of 5E614LO3311– cells in
certain strains of mice (17,25) and the absence of competition
between 4LO3311 and 5E6 mAb for staining BALB/c NK cells
(data not shown).
Variations in NK2.1/Ly-49C expression among inbred mouse
strains
Our earlier flow cytometry analyses of nylon-wool non-adherent spleen cells from selected mouse strains have suggested
a strain variation in the level of NK2.1 expression (28). Since
expression of Ly-49A has been reported to be regulated by
host MHC class I alleles (5,32) and Ly-49C has recently been
shown to also be a receptor for MHC class I molecules
(19,22,23), it was of interest to determine whether NK2.1/Ly49C expression may be subject to a similar regulation. In
order to answer that question, we extended the flow cytometry
analysis of NK2.1/Ly-49C expression on NK-enriched spleen
cells to a number of mouse strains of different H-2 haplotypes.
As presented in Fig. 2 and Table 2, the percentage of NK2.1/
Fig. 1. Expression of NK2.1 on COS cells transfected with Ly-49C.
Ly-49 cDNAs were expressed in COS cells and stained with the
antibodies A1, 4LO3311 or 5E6. Solid histograms represent COS
cells transfected with Ly-49A or C and empty histograms are cells
transfected with vector alone.
Ly-49C1 cells and/or their mean fluorescence intensity (MFI)
was found to be low in C57BL/6 and C57BL/10 (H-2b) mice,
intermediate in AKR, C3H and CBA (H-2k) mice, and high in
BALB/c and DBA/2 (H-2d) mice. Interestingly, NK2.1/Ly-49C1
cells of A/J (H-2a) mice, which express H-2Kk and Dd molecules, showed a MFI in the range of H-2d mice. With an NK2.1/
Ly-49C1 cell population with a higher MFI than other H-2b
strains as well as a second peak of low intensity, LP mice
showed a unique phenotype compared to all other mouse
strains. Altogether, these results indicated that H-2-linked
factors might influence NK2.1/Ly-49C expression but also
strongly suggest that H-2-independent factors are involved.
Variations in NK2.1/Ly-49C expression in H-2-congenic mice
To better evaluate the putative modulation of NK2.1/Ly-49C
by H-2 haplotypes, we carried out a series of flow cytometry
analyses of NK2.1/Ly-49C expression on NK-enriched spleen
cells from selected MHC-congenic strains. Since C57BL/10
and BALB/c mice showed respectively the NK2.1/Ly-49C NK
cell populations with the lowest and highest MFI, we reasoned
that if NK2.1/Ly-49C expression is indeed H-2-dependent,
the NK2.1/Ly-49C phenotype of congenic mice would tend
towards the phenotype of the mouse strain contributing the
H-2 allele, and thus vary in opposite directions in B10 and
BALB/c congenic mice expressing a given haplotype. With
MHC-congenic strains on B10 background, the MFI of NK2.1/
Ly-49C1 cells was enhanced three to four times in B10.D2
(H-2d) mice (P , 0.0001 as calculated by the two tailed
Student’s t-test from three to four mice per group) but remained
unchanged in B10.BR (H-2k) mice (Fig. 3A). No significant
variation in the percentage of NK2.1/Ly-49C1 cells was
detected in these congenic mice. As reported by others (5),
we found no alteration of NK1.1 expression in B10 congenic
strains expressing H-2d or H-2k haplotypes, whereas the Ly49A expression detected by the A1 mAb was significantly
reduced (data not shown). As expected, when similar analyses
were done with MHC-congenic strains on the BALB/c background, a significant down-regulation of NK2.1/Ly-49C
expression level was observed in BALB.K (H-2k) and BALB.B
(H-2b) mice (P , 0.0001) (Fig. 3B). An increase in the
percentage of NK2.1/Ly-49C1 cells was detected in BALB.K
(H-2k) mice (P , 0.0001) but not in BALB.B (H-2b) mice. These
results strongly support the hypothesis that the expression of
NK2.1/Ly-49C is regulated by the host H-2 haplotype.
536 MHC-dependent regulation of NK2.1/Ly-49C expression
Fig. 2. NK2.1/Ly-49C expression on NK-enriched spleen cells from different inbred mouse strains. Nylon-wool non-adherent CD4–CD8– spleen
cells were stained with biotinylated anti-NK2.1 mAb (solid lines) or isotype control mAb (dotted lines) and streptavidin–PE, and then analyzed
by flow cytometry. The histograms are representative of three to five experiments.
NK2.1/Ly-49C expression is mainly modulated by H-2Kb and
H-2Dk
To clarify further the contribution of H-2 loci in regulating
NK2.1/Ly-49C expression, a series of B10 congenic strains
carrying a chromosome 17 differential segment of variable
length inherited from DBA/2 or A mice were included in the
study. Comparative flow cytometry analysis of NK-enriched
spleen cells revealed that the percentage of NK2.1/Ly-49C1
cells was almost the same in all MHC-congenic and recombinant lines tested (Table 3). However, it appeared that the MFI
of the NK2.1/Ly-49C1 cell population was markedly lowered
in all mice expressing H-2Kb as compared to H-2Kd. In mice
expressing H-2Kb, an haplotype change from b to d at the D
locus, as in B10.D2(R107) and B10.A(5R), did not further
change the level of NK2.1/Ly-49C expression. As NK2.1/Ly49C expression is low in B10.BR (Table 3), it is likely that
the H-2k haplotype also contributes to the down-regulation
observed. In this case, however, the H-2D rather than H-2K
locus appears to be determinant. Consistent with this hypothesis, the level of NK2.1/Ly-49C expression in B10.A was
only slightly reduced compared to B10.D2, whereas it is low
in B10.BR which differs from B10.A by expressing H-2Dk
rather than H-2Dd. These results suggest that as for Ly-49A,
the host MHC class I molecules which are the natural ligands
of NK2.1/Ly-49C are involved in regulating the level of expression of this NK cell receptor.
Two NK cell populations expressing different levels of NK2.1/
Ly-49C are present in (BALB/c3C57BL/6)F1 mice
Since NK2.1/Ly-49C is highly expressed in H-2d mice and
down-regulated in H-2b mice, it was of interest to determine
its expression level in H-2d/b heterozygous mice. As shown in
Fig. 4, two slightly overlapping NK2.1/Ly-49C1 cell populations
were observed in (BALB/c3C57BL/6)F1 mice. This detection
of a smaller C57BL/6- and a larger BALB/c-like population in
the hybrid mice with the 4LO3311 mAb further supports the
recent observation that Ly-49C is subject to allelic exclusion
(37). Interestingly, the expression level of NK2.1/Ly-49C on
the BALB/c-like population was half the one found in BALB/c
(H-2d) mice (MFI: 49 and 104 respectively). It was however
about twice the expression level found on BALB.B (H-2b) NK
cells (MFI: 25 as shown in Fig. 3). A similar pattern was
observed for the C57BL/6-like population. Whereas NK2.1/
Ly-49C was expressed on these cells at about twice the level
found in C57BL/6 (H-2b) mice (MFI: 18 and 10 respectively)
it was half the one found in B10.D2 (H-2d) mice (MFI: 38 as
shown in Table 3). The cell surface expression level of NK2.1/
Ly-49C thus appears to be down-regulated to a lower extent
MHC-dependent regulation of NK2.1/Ly-49C expression 537
Table 2. Expression of NK2.1/Ly-49C on NK-enriched spleen
cells from various inbred mouse strainsa
Strain
C57BL/6
C57BL/10
LPb
AKR
C3H
CBA
A
BALB/c
DBA/2
129
NZB
H-2 haplotype
b
b
b
k
k
k
a
d
d
b
d
Percent positive
cells 6 SD
MFI 6 SD
28 6
25 6
33 6
40 6
39 6
33 6
33 6
52 6
28 6
,1
,1
10 6 1
11 6 1
27 6 4
20 6 2
24 6 3
17 6 2
78 6 10
115 6 15
66 6 8
NAc
NA
2
6
8
5
3
2
2
5
1
aExpression of NK2.1 on NK-enriched spleen cells from three to
seven individual mice of each strain was analyzed by flow cytometry
as described in Methods.
bIn this mouse strain, an NK2.1int cell populations representing
13 6 2 % of the cells with a MFI of 4 6 1 was also clearly detectable
in every mouse tested.
cNot applicable.
when the regulatory MHC gene is expressed co-dominantly
with another allele.
Discussion
Our previous structural and functional studies of the NK2.1
antigen indicated that this molecule was a disulfide-linked
dimer, activation receptor specifically expressed by NK cells
(26,27). An activating function for NK2.1 was suggested by
the ability of anti-NK2.1 to augment the lysis of NK-sensitive
targets by fresh and IL-2-activated NK cells and to induce
granule exocytosis from IL-2-activated BALB/c NK cells (27).
It is therefore of interest that NK2.1 was found to be encoded
by Ly-49C because recent studies of this molecule have
suggested that it (23), like Ly-49A (4) and G2 (21), functions
as an inhibitory receptor. Indeed, Ly-49C1 NK cells, as
recognized by the mAb 5E6, acquire the ability to lyse
otherwise resistant H-2Kb bearing targets following the addition of either F(ab9)2 5E6 or anti-Kb antibodies (23). This
observation, together with the binding specificity of Ly-49C
for Kb (22) strongly suggests that the recognition of MHC
class I by this receptor results in the delivery of a negative
Fig. 3. NK2.1/Ly-49C expression on NK-enriched spleen cells from H-2 congenic mice on the B10 and BALB/c backgrounds. NK-enriched
spleen cells from inbred and congenic mice on the B10 (A) and BALB/c (B) backgrounds were stained with biotinylated 4LO3311 mAb (solid
lines) or isotype control mAb (dotted lines) and streptavidin–PE before being analyzed by flow cytometry. The histograms are representative
of three to five experiments. The percentage of NK2.1/Ly-49C1 cells and the corresponding MFI are indicated (% cells/MFI). These are mean
values from three to five mice.
538 MHC-dependent regulation of NK2.1/Ly-49C expression
Table 3. Expression of NK2.1/Ly-49C on NK-enriched spleen cells from MHC congenic and congenic recombinant micea
Strain
B10.D2
C57BL/10
B10.BR
B10.D2(R103)
B10.D2(R107)
B10.A
B10.A(2R)
B10.A(5R)
H-2
haplotype
d
b
k
g3
i7
a
h2
i5
Positive cells 6 SD
(%)
H-2 allelesb for
K
A
E
D
d
b
k
d
b
k
k
b
d
b
k
d
b
k
k
b
d
b
k
d
b
k
k
/ k
d
b
k
/ b
/ d
d
/ b
d
24
27
33
22
24
25
27
31
6
6
6
6
6
6
6
6
2
4
8
9
8
3
6
8
MFI 6 SD
MFI reduction (relative to H-2d)
(%)
38 6 3
11 6 3
863
24 6 7
963
28 6 7
18 6 2
13 6 3
N/Ac
71
79
37
76
26
53
66
a
b
See legend to Table 2.
For each strain, the haplotype of each MHC subregion is shown with a slash indicating the site of recombination when applicable.
c Not applicable
signal to an NK cell. The stimulatory effects that we have
seen with the anti-NK2.1 mAb are therefore provocative and
suggest that this receptor may in some cases have an
activating function. Although the enhanced lysis which follows
the addition of anti-NK2.1 mAb may also be interpreted in
terms of a blockage of negative signalling, the induced
granule exocytosis appears to be the result of an actual
triggering effect because immobilized isotype control antibodies (IgG3) had no such effect (27). Moreover, it was
previously reported that 5E6 anti-Ly-49C mAb was able to
trigger redirected lysis of FcR1 targets (33), an observation
consistent with activating properties.
Studies of human NK cell recognition have found that
activating and inhibitory functions may be shared by the same
type of receptor. The first such example described was that
of CD941 NK cell clones, some of which are activated and
some of which are inhibited in the presence of an anti-CD94
mAb (34). Although the mechanisms responsible for this
ambivalence remain unclear, high-level CD94 expression on
a NK cell appears to be required for inhibition. The p58 family
of molecules, members of which bind HLA-C and inhibit NK
cell killing, has also been found to encode receptors (p50)
which activate NK cells upon class I recognition. These
activation structures differ from the inhibitory receptors in the
transmembrane region and in their truncated cytoplasmic
domains (35). The occurrence of either activating or inhibitory
effects upon binding of NK cells to target MHC molecules
has also been documented in the rat (36), thus supporting
the idea that dual MHC antigen-induced signaling functions
may be more common than originally anticipated. In the
case of Ly-49, however, it remains unknown what factors or
experimental conditions may result in activation or inhibitory
signals.
Original studies of the Ly-49 gene family showed significant
restriction fragment length polymorphism among various
inbred strains of mice, suggesting that these genes are highly
polymorphic (12). In the case of Ly-49C, it was reported that
C57BL/6, NZB and 129 mice each express distinct alleles
that are different from the one shared by BALB/c, CBA and
A/Sn mice (20,37,38). The 4LO3311 and 5E6 mAb were both
derived from 129 mice immunized with C57BL/6 NK cells
(17,25) and they both recognize Ly-49C (20 and this study).
Fig. 4. NK2.1/Ly-49C expression in H-2d/b hybrid mice. NK-enriched
spleen cells from BALB/c (H-2d) (dotted line), C57BL/6 (H-2b) (thin
solid line) and (BALB/c3C57BL/6)F1 (H-2d/b) (thick solid line) mice
were stained with biotin-conjugated 4LO3311 mAb and streptavidin–
PE and then analyzed by flow cytometry. The histogram shown is
representative of four similar experiments.
Since 5E614LO3311– splenic NK cells were detected in
certain strains of mice (17, 25), it is likely that 4LO3311 and
5E6 mAb bind different epitopes of this polymorphic molecule.
Allelic variations may also result in some forms of Ly-49C that
are recognized by neither, both or only one of these antibodies.
Finally, it is also possible that in addition to react with Ly-49C,
one or both of these antibodies recognize other unidentified
Ly-49 molecules. We confirmed by competition studies that
5E6 and 4LO3311 mAb bind different epitopes of Ly-49C. In
a concurrent study, we demonstrated that in C57BL/6 mice,
5E6 but not 4LO3311 mAb detects two distinct but highly
related molecules (39). The one reacting with 5E6 but not
with 4LO3311 mAb was identified as the product of a gene
previously thought to be the B6 allele of Ly-49C, which was
renamed Ly-49I. From the comparison of the deduced amino
acid sequences of the cDNAs, it appeared that the BALB/c
and C57BL/6 Ly-49C gene products differ at only four
residues. While our paper was in press, the cloning of an
identical C57BL/6 Ly-49C cDNA and the recognition of its
gene product by the 5E6 mAb were reported by Sundbäck
et al. (38). Their observations and those of previous reports
MHC-dependent regulation of NK2.1/Ly-49C expression 539
(20,37) confirm that in C57BL/6 mice, the 5E6 mAb reacts
with two molecules. The 4LO3311 mAb thus appears to be
the only known antibody reacting only with Ly-49C. The
reactivity of the 5E6 mAb for Ly-49C and Ly-49I could therefore
certainly account for the recently reported inconsistencies
regarding Ly-49C expression in relation to host MHC class I
haplotypes (40).
The variation in NK2.1/Ly-49C expression on NK cells from
several inbred, MHC-congenic, and congenic recombinant
mouse strains raised the possibility that NK2.1/Ly-49C expression is regulated in a way similar to Ly-49A and Ly-49G2 in
hosts expressing their class I ligands (5,40). The enhanced
NK2.1/Ly-49C expression in B10.D2 (H-2d) compared to B10
(H-2b) and B10.BR (H-2k) and, conversely, the reduced NK2.1/
Ly-49C expression in BALB.B (H-2b) and BALB.K (H-2k)
compared to the BALB/c (H-2d) inbred partner strongly support an H-2-dependent regulation of this receptor. The low
MFI of NK2.1/Ly-49C1 cells in B10 (H-2Kb, H-2Db), B10.D2
(R107) (H-2Kb, H-2Dd) and B10.A (5R) (H-2Kb, H-2Dd) mice
are consistent with NK2.1/Ly-49C expression being downregulated by H-2Kb. In view of the recently described interactions between Ly-49C and H-2Kb (22), the expression of
NK2.1/Ly-49C on NK cells, like that of Ly-49A, appears to be
down-regulated in the presence of its MHC class I ligand.
Although the occurrence of NK2.1/Ly-49C1 cells with intermediate fluorescence intensity in LP mice appears inconsistent with the hypothesis that H-2Kb molecules down-regulate
NK2.1/Ly-49C expression, this is most probably explained by
allelic variation in Ly-49C which has resulted in a receptor
with reduced affinity for Kb and therefore shows little or no
down-regulation by it.
In our experiments, the highest NK2.1/Ly-49C expression
was seen in H-2d mice, and the lowest was in H-2b and H-2k
mice. All observations concerning NK2.1/Ly-49C calibration
are therefore made relative to these two levels. H-2 recombinant analysis has in fact shown that the level of NK2.1/Ly-49C
expression falls along a continuum between these high (H-2d)
and low (H-2b and H-2k) standards. In addition to the maximal
down-regulation associated with the K locus of the b haplotype
(66–76% MFI reduction), the D locus of this same haplotype
resulted in a moderate down-regulation (37%). With regards
to H-2k down-regulation (79%), the K locus alone was shown
to have only a modest effect (26%). It is therefore thought
that Dk may have a more significant role in NK2.1/Ly-49C
receptor modulation. This variability in the level of receptor
calibration is likely influenced by the affinity of an Ly-49 for a
given class I molecule, such that stronger interactions bring
about greater down-regulation. Because the H-2d haplotype
was associated with the highest NK2.1/Ly-49C expression
relative to those tested, it is unknown to what degree these
H-2 antigens affect NK2.1/Ly-49C expression. In light of
binding studies which have shown Ly-49C to bind to H-2b,
H-2k, as well as H-2d antigens (including Kd and Dd), it is
expected that expression of this receptor is affected by these
molecules, although weaker affinity interactions may result in
a relatively low degree of down-regulation.
The use of congenic strains has limitations since congenic
mice generally differ from their inbred partners at hundreds
of genes inherited from the donor strain, the exact length of
the differential segment often being unknown. Our data at
least clearly indicate that the main H-2-associated element
down-regulating NK2.1/Ly-49C expression is centromeric to
H-2D in H-2b mice, and telomeric to H-2K in H-2k mice,
although multiple H-2 regions appear to be involved and
their relative importance may be haplotype-dependent. It is
noteworthy that although NK2.1/Ly-49C expression in B10.D2
mice is enhanced compared to B10 mice, it does not reach
the expression level observed in the donor DBA/2 mice.
Similarly, the introduction of a B10 differential segment including the H-2 complex into BALB/c mice did not reduce NK2.1/
Ly-49C expression to the level found in B10 mice, thus
indicating that as suggested for other Ly-49 receptors (40)
non-H-2-dependent factors may also contribute to the regulation of NK2.1/Ly-49C expression and influence the Ly-49
repertoire. It is expected that polymorphism of an individual
Ly-49 is one such factor that will affect its regulation by the
H-2 complex.
Non-MHC-linked genes putatively influencing the expression level of NK2.1/Ly-49C may have contributed to the
particular pattern observed in (C57BL/63BALB/c)F1 mice.
However, the enhanced expression of NK2.1/Ly-49C on the
C57BL/6-like cell population detected in the hybrid mice and
its reduced expression on the BALB/c-like cell population,
relative to those found in the parental strains, may also
indicate that the C57BL/6 and BALB/c Ly-49C alleles are
similarly down-regulated by H-2b. This is further supported
by our recent observation that these two Ly-49C alleles have
indistinguishable MHC class I binding specificity (39). The
10-fold difference observed between BALB/c (H-2d) and
C57BL/6 (H-2b) mice, regarding the MFI of NK2.1/Ly-49C1
cells, has certainly facilitated the detection of intermediate
cell surface expression levels of this receptor in hybrid mice
expressing lower levels of H-2b. Although it was initially
reported that the decreased expression of Ly-49A was
inherited as a dominant trait (32), it is conceivable that several
Ly-49 molecules including Ly-49A may be subject to a H-2dependent fine tuning of cell surface receptor calibration as
shown here for Ly-49C. However, subtle variations in the
expression level of a given receptor might be difficult to see
in hybrid mice when differences in parental strains are ,2fold, as observed recently for Ly-49A and Ly-49G2 (40).
Abbreviations
MFI
PE
mean fluorescence intensity
Phycoerythrin
References
1 Trinchieri, G. 1989. Biology of natural killer cells. Adv. Immunol.
47:187.
2 Ljunggren, H.-G. and Kärre, K. 1990. In search of the ‘missing self’:
MHC molecules and NK cell recognition. Immunol. Today 11:7.
3 Kane, K. P. 1994. Ly-49 mediates EL4 lymphoma adhesion to
isolated class I major histocompatibility complex molecules.
J. Exp. Med. 179:1011.
4 Karlhofer, F. M., Ribaudo, R. K. and Yokoyama, W. M. 1992. MHC
class I alloantigen specificity of Ly-491 IL-2-activated natural killer
cells. Nature 358:66.
540 MHC-dependent regulation of NK2.1/Ly-49C expression
5 Karlhofer, F. M., Hunziker, R., Reichlin, A., Margulies, D. H. and
Yokoyama, W. M. 1994. Host MHC class I molecules modulate
in vivo expression of a NK cell receptor. J. Immunol. 153:2407.
6 Karlhofer, F. M. and Yokoyama, W. M. 1991. Stimulation of murine
natural killer (NK) cells by a monoclonal antibody specific for the
NK1.1 antigen. IL-2-activated NK cells possess additional specific
stimulation pathways. J. Immunol. 146:3662.
7 Ryan, J. C., Turck, J., Niemi, E. C., Yokoyama, W. M. and Seaman,
W. E. 1992. Molecular cloning of the NK1.1 antigen, a member
of the NKR-P1 family of natural killer cell activation molecules.
J. Immunol. 149:1631.
8 Nagasawa, R., Gross, J., Kanagawa, O., Townsend, K., Lanier,
L. L., Chiller, J. and Allison, J. P. 1987. Identification of a novel T
cell surface disulfide-bonded dimer distinct from the α/β antigen
receptor. J. Immunol. 138:815.
9 Chan, P.-Y. and Takei, F. 1989. Molecular cloning and
characterization of a novel murine T cell surface antigen, YE1/48.
J. Immunol. 142:1727.
10 Yokoyama, W. M., Jacobs, L. B., Kanagawa, O., Shevach, E. M.
and Cohen, D. I. 1989. A murine T lymphocyte antigen belongs
to a supergene family of type II integral membrane proteins.
J. Immunol. 143:1379.
11 Giorda, R. and Trucco, M. 1991. Mouse NKR-P1. A family of
genes selectively coexpressed in adherent lymphokine-activated
killer cells. J. Immunol. 147:1701.
12 Yokoyama, W. M., Kehn, P. J., Cohen, D. I. and Shevach, E. M.
1990. Chromosomal location of the Ly-49 (A1, YE1/48) multigene
family: genetic association with the NK1.1 antigen. J. Immunol.
145:2353.
13 Yokoyama, W. M., Ryan, J. C., Hunter, J. J., Smith, H. R. C., Stark,
M. and Seaman, W. E. 1991. cDNA cloning of mouse NKR-P1
and genetic linkage with Ly-49. Identification of a natural killer cell
gene complex on mouse chromosome 6. J. Immunol. 147:3229.
14 Wong, S., Freeman, J. D., Kelleher, C., Mager, D. and Takei, F.
1991. Ly-49 multigene family: new members of a superfamily of
type II membrane proteins with lectin-like domains. J. Immunol.
147:1417.
15 Smith, H. R. C., Karlhofer, F. M. and Yokoyama, W. M. 1994.
Ly-49 multigene family expressed by IL-2-activated NK cells.
J. Immunol. 153:1068.
16 Hackett, J., Jr, Tutt, M., Lipscomb, M., Bennett, M., Koo, G. and
Kumar, V. 1986. Origin and differentiation of natural killer cells. II.
Functional and morphologic studies of purified NK-1.11 cells.
J. Immunol. 136:3124.
17 Sentman, C. L., Hackett, J., Jr, Kumar, V. and Bennett, M. 1989.
Identification of a subset of murine natural killer cells that mediate
rejection of Hh-1d but not Hh-1b bone marrow grafts. J. Exp.
Med. 170:191.
18 Mason, L. H., Mathieson, B. J. and Ortaldo, J. R. 1990. Natural
killer (NK) subsets in the mouse. NK1.11/LGL-11 cells restricted
to lysing NK targets, whereas NK-1.11/LGL-1– cells generate
lymphokine-activated killer cells. J. Immunol. 145:751.
19 Brennan, J., Mager, D., Jefferies, W. and Takei, F. 1994. Expression
of different members of the Ly-49 gene family defines distinct
natural killer cell subsets and cell adhesion properties. J. Exp.
Med. 180:2287.
20 Stoneman, E. R., Bennett, M., An, J., Chesnut, K. A., Wakeland,
E. K., Scheerer, J. B., Siciliano, M. J., Kumar, V. and Matthew, P.
A. 1995. Cloning and characterization of 5E6 (Ly-49C), a receptor
molecule expressed on a subset of murine natural killer cells.
J. Exp. Med. 182:305.
21 Mason, L. H., Ortaldo, J. R., Young, H. A., Kumar, V., Bennett, M.
and Anderson, S. K. 1995. Cloning and functional characteristics
of murine large granular lymphocyte-1: a member of the Ly-49
gene family (Ly-49G2). J. Exp. Med. 182:293.
22 Brennan, J., Mahon, G., Mager, D. L., Jefferies, W. A. and Talei,
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
F. 1996. Recognition of class I major histocompatibility complex
molecules by Ly-49: specificities and domain interactions. J. Exp.
Med. 183:1553.
Yu, Y. Y. L., George, T., Dorfman, J. R., Roland, J., Kumar, V. and
Bennett, M. 1996. The role of Ly-49A and 5E6 (Ly-49C) molecules
in hybrid resistance mediated by murine natural killer cells against
normal T cell blasts. Immunity 4:67.
Pollack, S. B. and Emmons, S. L. 1982. NK-2.1: an NK-associated
antigen detected with NZB anti-BALB/c serum. J. Immunol.
129:2277.
Lemieux, S., Ouellet-Talbot, F., Lusignan, Y., Morelli, L., Labrèche,
N., Gosselin, P. and Lecomte, J. 1991. Identification of murine
natural killer cell subsets with monoclonal antibodies derived from
129 anti-C57BL/6 immune spleen cells. Cell. Immunol. 134:191.
Gosselin, P., Lusignan, Y. and Lemieux, S. 1993. The murine
NK2.1 antigen: a 130 kD glycoprotein dimer expressed by a
natural killer cell subset of the spleen, thymus, and lymph nodes.
Mol. Immunol. 30:1185.
Morelli, L. and Lemieux, S. 1993. Triggering of the cytotoxic
activity of murine natural killer and lymphokine-activated killer
cells through the NK2.1 antigen. J. Immunol. 151:6783.
Morelli, L., Lusignan, Y. and Lemieux, S. 1992. Heterogeneity of
natural killer cell subsets in NK-1.11 and NK-1.1– inbred mouse
strains and their progeny. Cell. Immunol. 141:148.
Koo, G. C. and Peppard, J. R. 1984. Establishment of monoclonal
anti-NK-1.1 antibody. Hybridoma 3:301.
Sentman, C. L., Kumar, V., Koo, G. and Bennett, M. 1989. Effector
cell expression of NK1.1, a murine natural killer cell-specific
molecule, and ability of mice to reject bone marrow allografts.
J. Immunol. 142:1847.
Giorda, R., Weisberg, E. P., Ip, T. K. and Trucco, M. 1992.
Genomic structure and strain-specific expression of the natural
killer cell receptor NKR-P1. J. Immunol. 149:1957.
Olsson, M. Y., Kärre, K. and Sentman, C. L. 1995. Altered
phenotype and function of natural killer cells expressing the major
histocompatibility complex receptor Ly-49 in mice transgenic for
its ligand. Proc. Natl Acad. Sci. USA 92:1649.
Sentman, C. L., Kumar, V. and Bennett, M. 1991. Rejection of
bone marrow cell allografts by natural killer cell subsets: 5E61
cell specificity for Hh-1 determinant 2 shared by H-2d and H-2f.
Eur. J. Immunol. 21:2821.
Perez-Villar, J. J., Morello, I., Rodriguez, A., Carretero, M.,
Aramburu, J., Sivori, S., Orengo, A. M., Moretta, A. and LopezBotet, M. 1995. Functional ambivalence of the Kp43 (CD94) NK
cell-associated surface antigen. J. Immunol. 154:5779.
Moretta, A., Sivori, S., Vitale, M., Pende, D., Morelli, L., Augugliaro,
R., Bottino, C. and Moretta, L. 1995. Existence of both inhibitory
(p58) and activatory (p50) receptor for HLA-C molecules in human
natural killer cells. J. Exp. Med. 182:875.
Naper, C., Vaage, J. T., Lambracht, D., Lovik, G., Butcher, G. W.,
Wonigeit, K. and Rolstad, B. 1995. Alloreactive natural killer cells
in the rat: complex genetics of major histocompatibility complex
control. Eur. J. Immunol. 25:1249.
Held, W., Roland, J. and Raulet, D. H. 1995. Allelic exclusion of
Ly-49-family genes encoding class I MHC-specific receptors on
NK cells. Nature 376:355.
Sundbäck, J., Kärre, K, and Sentman, C. L. 1996. Cloning of
minimally divergent allelic forms of the natural killer (NK) receptor
Ly-49C, differentially controlled by host genes in the MHC and
NK gene complexes. J. Immunol. 157: 3936.
Brennan, J., Lemieux, S., Freeman, J. D., Mager, D. L. and Takei,
F. 1996. Heterogeneity among Ly-49C natural killer (NK) cells:
characterization of highly related receptors with differing functions
and expression patterns. J. Exp. Med. 184:2085.
Held, W., Dorfman, J. R., Wu, M.-F. and Raulet, D. 1996. Major
histocompatibility class I-dependent skewing of the natural killer
cell Ly49 receptor repertoire. Eur. J. Immunol. 26:2286.