| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ludwig Institute for Cancer Research, Lausanne Branch, University of Lausanne, Epalinges, Switzerland
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Ab-mediated blocking of inhibitory receptor-MHC interaction is sufficient to enable NK cells to kill syngeneic, normal cells, suggesting that these receptors are important to prevent autoaggression (3). However, although syngeneic cells are normally not killed, NK cells are able to lyse MHC-different cells, indicating that tolerance toward self-MHC-expressing cells is acquired. Self-tolerance may require that each NK cell is inhibited by self-MHC class I. Thus, the random repertoire of inhibitory MHC receptors may be expected to adapt to the available self-MHC class I molecules. Indeed, the analysis of a large panel of human NK cell clones supports this model, because all clones expressed at least one self-MHC-specific inhibitory receptor (4, 5). Although it is currently not clear whether the same holds true for mouse NK cells (6), the MHC background can influence the specificity of developing murine NK cells. Thus, the introduction of a transgenic Dd class I molecule into H-2b mice results in the emergence of NK cells that are able to kill H-2b-positive but Dd-negative targets (7). Such cells, however, are not found in nontransgenic H-2b mice. Therefore, the presence of Dd may allow the development of a NK cell subset that expresses a Dd- but no H-2b-specific inhibitory receptor. Consistent with such a scenario, certain Ly49 receptors, such as Ly49A and Ly49G2, discriminate between Dd and H-2 b class I molecules (8, 9).
These findings suggest that NK cell self-tolerance is, at least in part, based on the adaptation of the inhibitory receptor repertoire to self-MHC class I. Nevertheless, the process of self-tolerance induction remains poorly understood. Even though MHC class I molecules are not required for the development of normal numbers of NK cells (10), it is conceivable that the engagement of inhibitory receptors on a developing NK cell may be required for NK cell maturation and/or differentiation events. Indeed, NK cells arising in the absence of class I MHC are considered anergic (11). Consequently, the interaction of inhibitory receptors with self-MHC class I may positively influence certain aspects of NK cell development. To test this hypothesis, we used a developmental assay that takes advantage of a transgenic mouse strain in which all NK cells express the H-2d-specific Ly49A receptor (12). In contrast, Ly49A is expressed only by a subset of NK cells in normal mice, where some developing NK cells may fail to acquire a self-MHC-specific inhibitory receptor (6, 13). Thus, in a competitive situation and in the presence of H-2d, the development of transgenic over nontransgenic NK cells should be favored if engagement of inhibitory receptors confers a selective advantage. Indeed, our results demonstrate that the expression of self-MHC-specific inhibitory receptors has a positive effect on NK cell development. Thus, our results provide evidence that self-inhibitory receptors generate crucial signals for the maturation of NK cells.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CD45.1 congenic C57BL/6 (B6)3 mice were purchased from The Jackson Laboratory (Bar Harbor, ME). B6 (H-2b), and B10.D2 (D2, H-2d) mice were obtained from Harlan (Zeist, The Netherlands). Ly49A transgenic mice (line 2) were described previously (12). Here we have used transgenic and nontransgenic littermate (LM) mice from B6 and D2 back-crosses 6 and 7, respectively. To establish H-2d CD45.1 congenic mice, (B6 CD45.1 x D2) F1 mice were back-crossed twice to D2. Offspring used for experiments were CD45.1/CD45.2, H-2d/d.
Abs
Anti-Ly49A (JR9-318), anti-Ly49C/I (SW5E6), and anti-Ly49G2 (4D11) have been described (14, 15, 16). Abs against NK1.1 (PK136), CD3 (145–2C11), CD45.1 (A20), and CD45.2 (104) were purchased from PharMingen (San Diego, CA). Anti-mouse IgG, which cross-reacts with IgM and thus detects all sIg-positive B cells was obtained from Caltag Laboratories (San Francisco, CA).
Flow cytometry
Single-cell suspensions were prepared from spleen, bone marrow,
lymph node, and liver. Spleen and bone marrow cell suspensions were
depleted of erythrocytes and thereafter passed over nylon wool columns.
Nylon wool-nonadherent cells were collected. Livers were passed through
a steel mesh. Mononuclear cells were isolated from the 40–80%
interface of a Percoll (Pharmacia, Uppsala, Sweden) step gradient after
centrifugation for 20 min at 2000 rpm. For flow cytometry, 1.5 x
106 cells were incubated with 2.4G2 hybridoma
supernatant (anti-CD16/32) for 20 min on ice to block nonspecific
Ab binding via Fc
R. For four-color flow cytometry, cells were
stained with FITC-labeled anti-Ly49 mAbs plus CD3 CyChrome (Cy) and
NK1.1 PE. After washing, cells were further incubated with biotinylated
CD45.1 or CD45.2, respectively, followed by
streptavidin-allophycocyanine. Samples to analyze B cells (not blocked
with 2.4G2 hybridoma supernatant) were incubated with FITC-labeled
anti-IgG (cross-reacts with IgM and thus detects all sIg-positive
cells) followed by mouse IgG to reduce background. The samples were
analyzed on a FACScalibur using CELLQuest for data evaluation (Becton
Dickinson, San Jose, CA). Dead cells were excluded by life gating of
forward light scatter and side light scatter, and 100,000 cells were
usually analyzed in each file.
Generation of bone marrow chimeras
Recipient mice were injected i.p. with 100 µg mAb PK136 to deplete NK cells, followed by lethal irradiation (1000 rad from a 137Cs source). Mice were then reconstituted by i.v. injection of a 50:50 mixture of wild-type (CD45.1 or CD45.1/2) and Ly49A-transgenic (CD45.2) bone marrow cells (1 to 2 x 107 cells in total). Before injection, bone marrow cells were depleted of Thy-1+ cells by Ab plus complement treatment. Chimeras were analyzed 8–10 wk after reconstitution. To distinguish between the two donor origins, cells were usually stained with CD45.1- and CD45.2-specific mAbs. When using CD45 heterozygous normal donor cells, only CD45.1-specific stainings were performed. In this case, cells derived from Ly49A-transgenic (or LM) mice that are identified as being CD45.1 negative are referred to as CD45.2 cells.
Statistics
Data are presented as means ± SD. The two-tailed Student t test was used for data evaluation. In addition, the rank-sum test was used to assess significant differences among ratios.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We have addressed the question of whether the acquisition of
self-MHC-specific inhibitory receptors can positively influence NK cell
development by means of mixed bone marrow chimeras. In a first set of
experiments, we generated syngeneic mixed bone marrow chimeras.
Lethally irradiated B6 (H-2b) or D2
(H-2d) mice were reconstituted with a mixture of
equal proportions of bone marrow cells from syngeneic Ly49A transgenic
and normal donor mice. The reconstitution of the lymphoid system was
analyzed by flow cytometry 8–10 wk later. All analyzed chimeras
were successfully repopulated with both donor cell types, which were
distinguished by an allelic difference at the CD45 locus. Both
transgenic (CD45.2) and normal (CD45.1) bone marrow cells contributed
about equally to the B cell compartment in spleen (data not shown) and
lymph nodes, thus reflecting the input of similar numbers of stem cells
(Figs. 1
and
2A). B cells were used as
reference because transgene expression has no obvious effect on B cell
development (data not shown). Furthermore, the Ly49A transgene is
expressed only in a subpopulation of B cells (9.0 ± 1.9% CD45.2
B cells are Ly49A+ in a
H-2b chimera). The size of this subset remains
unaltered over time and in the presence of ligand (7.7 ± 2.0% of
CD45.2 B cells are Ly49A+ in a
H-2d chimera), suggesting the absence of
selective pressure on B cells expressing the transgene.
|
and 2), suggesting that the expression of the Ly49A transgene in the
absence of its Dd ligand does not influence the
efficiency of NK cell development. In contrast, a significant and
selective skew toward transgenic NK cells was observed in the
corresponding H-2d chimeras that express the
Ly49A ligand (131 ± 15% compared with CD45.2 B cells) (Figs. 1 and 2). Staining with the anti-Ly49A mAb JR9-318 confirmed that
virtually all CD45.2 NK cells were Ly49A+
(94.2 ± 4.7%) and thus derived from transgenic bone marrow.
CD45.1 NK cells were derived from normal bone marrow based also on the
size of the Ly49A+ subset (22.1 ± 1.7%).
In addition to spleen, similar positive effects on transgenic NK cells
were observed in liver and bone marrow (139 ± 12% and 134
± 25% compared with CD45.2 B cells, respectively) (Fig. 2B). The latter observation suggests a positive impact of
inhibitory receptor-MHC ligand interaction during a relatively early
phase of NK cell development and the stable maintenance of the skewed
representation of transgenic NK cells thereafter.
|
The presence of the Ly49A-ligand on both hemopoietic and nonhemopoietic cells is required to favor the development of transgenic NK cells
Both hemopoietic and nonhemopoietic cells have been shown to
influence NK cell reactivity (17, 18). To investigate
whether the observed positive effect on NK cell development
required H-2d ligand expression on hemopoietic
cells (radiosensitive), nonhemopoietic cells (radioresistant), or both
cell types, we have generated allogeneic mixed bone marrow chimeras. To
this end, irradiated B6 (H-2b) or D2
(H-2d) hosts were repopulated with appropriate
mixtures of H-2d and H-2b
bone marrow cells, respectively. The analysis of the chimeras
revealed that H-2d ligand expression exclusively
on hemopoietic cells was not sufficient to confer a positive effect on
NK cell development (Fig. 3A).
In contrast, the presence of the H-2d ligand
exclusively on nonhemopoietic cells led to a variable, but
statistically significant developmental advantage of transgenic NK
cells. In this situation, however, transgenic NK cells were not more
abundant among bone marrow and liver NK cells (data not shown). Thus,
our results suggest that optimal NK cell development requires the
presence of H-2d ligand on both hemopoietic and
nonhemopoietic cells.
|
Analysis of Ly49A transgenic mice has revealed that the expression
of the H-2d ligand drastically reduces the
presence of NK cells expressing Ly49G2, another
H-2d (but not
H-2b)-specific inhibitory receptor. Similarly, we
have found that Ly49A-transgenic, Ly49G2+ NK
cells are underrepresented in spleens of H-2d
compared with H-2b bone marrow chimeras (2.9-fold
reduction, Fig. 3B). In agreement with earlier results
(19), NK cells positive for Ly49C/I, yet another
H-2d (but also
H-2b)-specific receptor, were present at normal
frequencies. Similar results were obtained among bone marrow and liver
NK cells (data not shown). A minor and a partial contraction of the
Ly49G2+ NK cell subset was observed when the
H-2d ligand was present exclusively on
nonhemopoietic or hemopoietic cells, respectively. However, the
presence of the ligand on both cell types was required to profoundly
reduce the number of NK cells expressing multiple self-MHC-specific
receptors. Therefore, the presence of ligand on all cells is required
to positively affect NK cell development as well as to form the
appropriate Ly49 receptor repertoire. Similarly, functional data showed
that MHC class I ligand on all cells of both types was required to
generate functional NK cells capable of reacting to normal cells
lacking MHC class I ligand(s) (17, 18, 20). Although the
failure to engage MHC-specific inhibitory receptors does not
necessarily prevent NK cell development, such cells may develop less
efficiently, and their functional maturation may be impaired (11, 20).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B) (19). Lastly, the positive effect on NK
cell development described is clearly MHC class I dependent (Fig. 2).
There exist at least two possibilities to explain the net positive
effect on transgenic NK cells expressing an inhibitory self-MHC
receptor. The so-called selection model proposes that NK cells that
express at least one but not too many self-MHC-specific receptors are
selected from a pool of NK cells expressing random combinations of Ly49
receptors (2). Thus, the observed positive effect may
result from the rescue of immature NK cells that fail to acquire a
sufficient number of self-specific inhibitory receptors and
consequently would not mature. The selection model also predicts that
NK cells expressing too many self-MHC-specific receptors are
eliminated. Indeed, we observed a reduction in the number of transgenic
NK cells expressing another H-2d-specific Ly49
receptor. Curiously, this putative elimination process required the
presence of the H-2d ligand on both
radioresistant and radiosensitive cells (Fig. 3B). However,
one may have predicted that elimination of NK cells due to inhibitory
receptor-MHC interaction is a dominant effect. Thus, the presence of
the MHC ligand on either cell type should be sufficient for the
elimination process to occur which is contrary to our results.
Alternatively, the sequential model proposes the cumulative acquisition of Ly49 receptors with ongoing testing for the presence of self-MHC-specific receptors (2). A sufficient number of self-MHC-specific receptors would abort further Ly49 acquisition and allow NK cell maturation. Ly49A transgenic NK cells may thus acquire a sufficient number of self-MHC receptors before nontransgenic NK cells. They would develop faster and therefore contribute more efficiently to the NK cell compartment in a mixed bone marrow chimera. Engagement of inhibitory receptors may abort further Ly49 acquisition, thus preventing the development of NK cells with too many self-MHC receptors.
To positively influence the development of Ly49A transgenic NK cells,
the H-2d ligand must be expressed on all cells.
When the H-2d ligand is absent from either
nonhemopoietic (radioresistant) cells or hemopoietic cells, the effect
is, respectively, strongly reduced or abolished (Fig. 3A).
In these situations, maturation events may be induced when immature NK
cells encounter H-2b or
H-2d haplotype cells. Thus, Ly49A transgenic
expression may no longer confer a sufficient developmental advantage
over nontransgenic cells. The availability of two haplotypes may
therefore obscure the observed positive effect on the Ly49A-transgenic
NK cells. The expression of ligand on all cells was shown to be
required to allow the development of functional NK cells (i.e., NK
cells which are able to kill ligand-negative, normal cells). In one
experimental system, a Dd transgene had to be
expressed on all cells of H-2b mice to allow the
emergence of NK cells, which were able to kill
Dd-negative H-2b target
cells. NK cells developing in Dd mosaic mice
(where Dd-negative H-2b
cells coexist with Dd-positive
H-2b cells) were unable to kill
Dd-negative H-2b cells
(7, 20). During development, these NK cells may be
selected either by Dd H-2b
or H-2b host cells (the former NK cells would be
responsible for killing Dd-negative
H-2b targets). A subsequent encounter of a
"Dd-selected" immature NK cell (expressing a
H-2Dd, but not
H-2b-specific inhibitory receptor(s)) with a
Dd-negative H-2b host cell
may render this NK cell anergic (i.e., unable to kill
Dd-negative H-2b target
cells) (20). Similarly, NK cells that develop in the
absence of MHC class I molecules are considered anergic
(11). Moreover, the presence of some MHC class I-deficient
cells during development results in NK cells that are unable to reject
MHC class I-deficient bone marrow grafts (18). Applied to
our experimental system, it is conceivable that
H-2d-selected Ly49A-transgenic NK cells are
retarded in their development on encounter of a
H-2b haplotype cell. Thus, homogeneous class I
ligand expression and consequently Ly49 engagement may be required for
the efficient development and complete functional maturation of NK
cells.
Whatever the precise basis of the positive effect, our data suggest that the engagement of inhibitory receptor during NK cell development provides important signals for NK cell maturation, differentiation, or development. We have recently proposed that similar to the reactivity of mature NK cells, immature NK cells may be activated via MHC-independent receptors during development. Inhibitory MHC receptors would then be acquired to balance these activation signals.4 In the absence of inhibitory receptor engagement, NK cell activation may eventually induce anergy in NK cells, which fail to acquire self-MHC-specific inhibitory receptors (11, 20). Signals derived from inhibitory receptors would thus prevent anergy induction. Alternatively, the interaction of inhibitory receptors with cognate class I molecules could itself induce NK cell maturation. Consistent with a positive role of inhibitory receptors, human Ig-like killer-inhibitory receptors were shown to recruit phosphatidylinositol-3-kinase and thus potentially couple to signaling pathways that provide growth and/or survival signals (21, 22).
Our data provide the first evidence that MHC-specific inhibitory receptors positively influence NK cell development. This property may help to ensure self-tolerance and permit further NK cell differentiation or maturation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Werner Held, Ludwig Institute for Cancer Research, Lausanne Branch, 1066 Epalinges, Switzerland.
3 Abbreviations used in this paper: B6, C57BL/6; Cy, CyChrome; D2, B10.D2; LM, littermate.
4 B. Lowin-Kropf and W. Held. A role for the src family kinase Fyn in NK cell activation and the formation of the repertoire of inhibitory Ly49 receptors. Submitted for publication.
Received for publication December 22, 1999. Accepted for publication April 13, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
ßTCR population in i-IEL with an extrathymic maturation. Int. Immunol. 4:699.This article has been cited by other articles:
|
|
 |
|
  A. W. MacFarlane IV, T. Yamazaki, M. Fang, L. J. Sigal, T. Kurosaki, and K. S. Campbell Enhanced NK-cell development and function in BCAP-deficient mice Blood, July 1, 2008; 112(1): 131 - 140. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. A. Benoit and R. Tan Xenogeneic beta2-Microglobulin Substitution Alters NK Cell Function J. Immunol., August 1, 2007; 179(3): 1466 - 1474. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Taki, S. Nakajima, E. Ichikawa, T. Saito, and S. Hida IFN Regulatory Factor-2 Deficiency Revealed a Novel Checkpoint Critical for the Generation of Peripheral NK Cells J. Immunol., May 15, 2005; 174(10): 6005 - 6012. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. E. Johansson, H. Hall, J. Bjorklund, and P. Hoglund Broadly impaired NK cell function in non-obese diabetic mice is partially restored by NK cell activation in vivo and by IL-12/IL-18 in vitro Int. Immunol., January 1, 2004; 16(1): 1 - 11. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.-H. Lee, A. Zafer, Y. de Repentigny, R. Kothary, M. L. Tremblay, P. Gros, P. Duplay, J. R. Webb, and S. M. Vidal Transgenic Expression of the Activating Natural Killer Receptor Ly49H Confers Resistance to Cytomegalovirus in Genetically Susceptible Mice J. Exp. Med., February 17, 2003; 197(4): 515 - 526. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-W. Wang, J. M. Howson, T. Ghansah, C. Desponts, J. M. Ninos, S. L. May, K. H. T. Nguyen, N. Toyama-Sorimachi, and W. G. Kerr Influence of SHIP on the NK Repertoire and Allogeneic Bone Marrow Transplantation Science, March 15, 2002; 295(5562): 2094 - 2097. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Ioannidis, J. Zimmer, F. Beermann, and W. Held Cre Recombinase-Mediated Inactivation of H-2Dd Transgene Expression: Evidence for Partial Missing Self-Recognition by Ly49A NK Cells J. Immunol., December 1, 2001; 167(11): 6256 - 6262. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. T. Young, M. Uhrberg, J. H. Phillips, L. L. Lanier, and P. Parham Differential Expression of Leukocyte Receptor Complex-Encoded Ig-Like Receptors Correlates with the Transition from Effector to Memory CTL J. Immunol., March 15, 2001; 166(6): 3933 - 3941. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
d.
B, Histograms illustrate Ly49A expression among
CD45.1- (CD45.2) or CD45.1 gated NK cells.
C, Histograms show the contribution of Ly49A transgenic
cells (CD45.2) to the NK cell (gated
CD3-NK1.1+) or B cell (gated sIg+)
(right) compartments in H-2b ([Tg
b+b]
