| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Laboratory of Experimental Immunology, Division of Basic Sciences, and
Intramural Research Support Program, Science Applications International Corp. Frederick, National Cancer Institute, Frederick, MD 21702
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The Ly49 family represents a group of receptors expressed on murine NK, NK T cells, and some memory CD8+ T cells (5, 6, 7). The most extensively studied repertoire belongs to C57BL/6 (B6) mice. In these mice, Ly49AB6-JB6 are known to be expressed, whereas only gene fragments and/or aberrant mRNAs have been detected for Ly49kB6, lB6, mB6, and nB6 (5, 6, 8, 9, 10). We have previously shown that other mouse strains possess Ly49 molecules not found in B6 mice. Ly49O and P were found to be expressed in 129/J mice and Ly49L was found in CBA/J and C3H/He mice, yet none of these were detected in B6 mice using immunoprecipitation or RT-PCR (11, 12). This suggests that the Ly49 repertoire is extremely polymorphic between inbred mouse strains, even those sharing the same MHC haplotype such as 129/J and B6 (H-2b). In fact, Southern blotting with Ly49 single-exon probes has shown striking differences in the Ly49 gene content between mouse strains (9). A survey of inbred mouse strains with Ly49-specific Abs revealed that Ly49 epitope expression varies greatly. For example, a mAb that recognizes Ly49G2B6 (4D11) was found to stain DX5+CD3- NK cells from nine of nine strains tested, whereas 4E5 (Ly49DB6) stained NK cells from only three strains (B6, 129/J, and SJL) (13). Whether this is a result of allelic divergence or absent/silent genes is unknown.
With respect to function, Ly49 molecules can be grouped into activating and inhibitory subgroups. Inhibitory Ly49 sequences can be identified by the presence of an immunoreceptor tyrosine-based inhibitory motif (ITIM; I/VxYxxL/V) in the intracellular domain that recruits Src homology 2 domain-containing phosphatase 1 upon phosphorylation (14, 15). Activating Ly49 molecules (DB6, HB6, LCBA/J, and P129/J) lack an intact ITIM sequence but possess an arginine in the transmembrane domain for association with the signal-transducing protein DAP12 (16, 17). The binding of inhibitory Ly49 receptors to MHC ligands results in an inhibition of cytotoxicity (18, 19). In contrast, NK killing can be triggered via Ly49DB6 by H-2Dd-expressing target cells (20, 21). Furthermore, cross-linking of activating Ly49 molecules by specific Ab or MHC ligand results in cytokine production and intracellular calcium ion mobilization (11, 20, 22). In addition to Dd, Ly49D+ B6 NK cells have been shown to interact with Dr and Dsp2-expressing target cells (23). A recent report has shown that, like Ly49DB6, the presence of the nonobese diabetic allele of Ly49P on killer cells facilitates the lysis of Dd-expressing target cells (24).
Soluble MHC tetramers have been used to confirm and extend the known ligand specificities of many B6 Ly49 proteins. Some receptors, such as Ly49G2B6, show a very fine specificity toward Dd, whereas Ly49CB6 has a much broader range of tetramer binding (Kb, Db, Kd, Dd, and Dk), and no ligand has been found for Ly49EB6 or FB6 (19, 25, 26). Ly49AB6 was found to react with Dd and Dk tetramers and has been reported to inhibit NK lysis of Dd- and Dp-positive target cells (25, 26, 27). One report has also shown that Ly49AB6 is bound by Db tetramers (26). Although the role of inhibitory Ly49 such as Ly49CB6/IB6 can be explained as the monitoring of "self" in B6 mice, the purpose of inhibitory Ly49 that bind to MHC not present in B6 mice, as well as of activating Ly49 in general, still remains controversial.
Because 129 embryonic stem (ES) cells are used for the production of knockout mice for immunologic research, knowledge of the 129/J Ly49 repertoire is important, especially for studies dealing with NK biology. The 129/J and other closely related 129 strains show very different immunological phenotypes compared with B6 mice, with which they share the same MHC background. For example, 129/J mice are more susceptible to intracellular pathogens, such as the WA1 piroplasm and Sendai virus, than B6 mice (28, 29). Also, 129/J mice are more prone to the induction of autoimmune diseases such as experimental autoimmune encephalomyelitis and show greater disease severity in diabetes models than B6 mice (30, 31). In contrast, 129/J mice are relatively resistant to tumor induction by various carcinogenic agents and display a lower spontaneous overall tumor incidence (32, 33, 34). It has long been known that 129/J mice are not as capable as B6 mice in rejecting allogeneic and xenogeneic bone marrow transplants (35). Interestingly, the regulation of murine transplant rejection is at least partly dependent on Ly49 expression by NK cells (36). Our initial observations of the 129/J strain show that the Ly49 repertoire is different from B6 mice (11); therefore, elucidation of the total 129/J Ly49 repertoire was conducted to facilitate comparison of B6 and 129/J NK cell function. In addition, we have tested the ability of the newly described receptors to bind seven different types of class I MHC.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B6 and 129/J mice were purchased from Jackson Laboratories (Bar Harbor, ME). All mice were kept under pathogen-free conditions until use at 20–25 wk of age. Liver NK cells were isolated as previously described (37). CD3-DX5+-sorted liver NK cells were expanded for 3–5 days in complete RPMI 1640 supplemented with 1000 Cetus U/ml IL-2 (Hoffman-LaRoche, Nutley, NJ). Routinely, 75–80% of the resulting cells were DX5+. Cell line 293T is a human kidney epithelial cell line. YB20 and YB/Dd rat cell lines were provided by Drs. J. Ryan and M. Nakamura (University of California Veterans Affairs Medical Center, San Francisco, CA).
Cloning of Ly49 cDNA from 129/J NK cells
A cDNA library from IL-2-activated 129/J liver NK cells was constructed in the pBK-CMV vector (Stratagene, La Jolla, CA) by Cell and Molecular Technologies (Phillipsburg, NJ). Library screening was conducted with Ly49eB6, dB6, g2B6, and o129/J probes (38). One hundred positive clones were analyzed, and the following Ly49 sequences were found: Ly49e129/J (AF247643), Ly49g1129/J (AF288375), Ly49g2129/J (AF288376), Ly49i129/J (AF237686), Ly49r129/J (AF288377), Ly49s129/J (AF288378), Ly49t129/J (AF288379), Ly49u129/J (AF288380), and Ly49v129/J (AF288381). Previously assigned GenBank files for Ly49p129/J and Ly49o129/J (AF146570 and AF146571, respectively) were updated to include 5' and 3' untranslated sequence identified in this study.
Plasmids and transfections
Ly49 cDNAs were PCR-subcloned into pEF6/V5-His (pEF6) using the TOPO TA Cloning Kit (Invitrogen, Carlsbad, CA). All subclones were sequenced. The day before transfection, 2.5 x 105 293T cells were seeded in six-well plates. The following day, 293T cells were transfected with FuGENE6 (Roche, Indianapolis, IN) using 1.25 µg each of the indicated combinations of Ly49/pEF6 and mDAP12/pSport (a gift from D. McVicar, National Cancer Institute, Frederick, MD) for activating receptors plus empty pEF6 vector to a total of 5 µg of DNA.
Abs and soluble class I MHC tetramers
The following mAbs were used for two-color flow cytometry:
FITC-labeled 12A8, which binds to
Ly49AB6/DB6
(39) and PE-labeled 4E5 (Ly49DB6)
(16). YE1/48, YE1/32 (Ly49AB6)
(40), 4D11 (Ly49G2B6)
(19), 5E6
(Ly49CB6/IB6)
(41), and A1 (Ly49AB6)
(42) were all used as primary staining reagents followed
by FITC-conjugated goat anti-rat IgG secondary (Kirkegaard & Perry
Laboratories, Gaithersburg, MD) or, in the case of 5E6 and A1,
FITC-conjugated goat anti-mouse IgG was used as a secondary
(Kirkegaard & Perry Laboratories). A generous gift from V. Kumar,
University of Chicago (Chicago, IL), 1F8
(Ly49CB6/IB6/HB6)
was FITC-labeled, and PE-12A8 was also used for single-color analyses.
Biotinylated 4G10 Ab, which recognizes phosphotyrosine, was purchased
from Upstate Biotechnology (Lake Placid, NY). Detection of
phosphorylated proteins was conducted as previously described
(11). DX5 (pan-NK cell) mAb was a kind gift from L. Lanier
(University of California at San Francisco). Dd
expression was analyzed using FITC-conjugated 34-5-8S (BD PharMingen,
San Diego, CA). Streptavidin-PE conjugated murine class I MHC tetramers
Db, Dd,
Dk, Ld,
Kb, Kd, and
Kk were provided by the National Institute of
Allergy and Infectious Diseases Tetramer Facility/National
Institutes of Health AIDS Research and Reference Reagent Program
(Yerkes Regional Primate Research Center, Atlanta, GA). Tetramers were
refolded in the presence of peptide previously shown to form a stable
tetrameric structure and are as follows: Db,
GP33–41 (KAVYNFATC) of lymphocytic choriomeningitis virus
(43); Dd, GP160–169 (RGPGRAFVTI) of
HIV-1 (44); Dk, MT389–397
(RRLGRTLLL) of polyoma virus (45);
Kb, N324–332 (FAPGNYPAL) of sendai virus
(46); Kd, NPP147–155 (TYQRTRALV) of
influenza virus (47); Kk, Ha255–262
(FESTGNLI) of influenza virus (48); and
Ld, NP118–126 (RPQASGVYM) of lymphocytic
choriomeningitis virus (49). Class I MHC H chain and human
2-microglobulin were both produced in bacteria using the
pET expression system.
Flow cytometry and cytotoxicity assays
NK cells and transfected 293T were stained with the Abs described above (for 30 min on ice) and MHC tetramers (for 30 min on ice or at 37°C) and analyzed on a FACSort (BD Biosciences, Mountain View, CA). Incubations longer than 30 min did not increase tetramer-binding levels. NK cells used for cytotoxic analyses were sorted on a MoFlo (Cytomation, Ft. Collins, CO). Cytotoxicity assays were conducted as previously described (19).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We have previously observed that anti-Ly49 immunoprecipitates
of 129/J NK cells yield different phosphotyrosine-banding patterns when
compared with NK cells from B6 mice (11). Specifically,
immunoprecipitation with mAb 4E5, which only binds the activating Ly49D
receptor in the B6 strain, results in the coimmunoprecipitation of both
activating and inhibitory Ly49 molecules in 129/J mice
(11). To better characterize the
Ly49DB6-like receptors in 129/J mice, two-color
flow cytometric analysis was conducted with 4E5 and 12A8. A mAb that
can bind to Ly49DB6 as well as
Ly49AB6, 12A8 has been shown to react with a
subpopulation of 129/J NK cells (13, 39). Two-color
staining with 12A8 and 4E5 of B6 NK cells results in four populations,
a double-negative, two double-positive, and a
4E5-12A8+ single-positive
population (Fig. 1
A). The
single-positive population in the lower right quadrant
represents Ly49A+D- NK
cells, whereas the upper right quadrant contains
Ly49A+D+
(12A8high) and
Ly49A-D+
(12A8low) populations. In contrast, when the same
Abs are used to stain 129/J NK cells, a very different pattern is
observed (Fig. 1C). The
4E5-12A8+ population is
not present, and instead a
4E5+12A8- population
appears (upper left quadrant). Although there are still two
double-positive populations in 129/J mice, the
12A8high population is gone, and instead a
4E5high population is now present. NK cells from
a first generation cross of B6 and 129/J mice show that all populations
of the parent mice are present, yet the single-positive populations are
diminished (Fig. 1B).
|
C, proteins were isolated from
pervanadate-treated NK cells, immunoprecipitated with Ly49-specific
mAb, separated by SDS-PAGE under nonreducing conditions, and detected
by anti-phosphotyrosine blotting. We have previously shown that
this method will detect inhibitory Ly49 proteins as a 
110-kDa band
due to a phosphorylated ITIM. In addition, a 28-kDa band (DAP12) is
coimmunoprecipitated from activating Ly49 molecules (22).
Activating Ly49 receptors have no ITIM; therefore, they are not
directly phosphorylated, and their presence is detected indirectly
through DAP12 association. The results of two such experiments are
shown in Fig. 2, A and
B. Immunoprecipitation of unsorted B6 splenic NK cells with
4E5 only detected the DAP12 molecule associated with the activating
receptor, Ly49D. In contrast to B6 mice, unsorted 129/J NK cells showed
both activating and inhibitory receptors when immunoprecipitated with
4E5 as previously reported (11). Due to concerns that
different organs might contain different NK populations, bulk spleen
and liver 129/J NK cells were compared and shown to contain both types
of 4E5-reactive receptors (Fig. 2A). The type of receptor
represented in the sorted populations of Fig. 1C was then
analyzed. The upper left quadrant containing
4E5+12A8- cells was found
to contain only inhibitory-type receptors (Fig. 2A). This
suggests that 129/J mice possess an inhibitory-type Ly49 that is
recognized by 4E5 but not 12A8.
|
A). The
4E5low12A8+ population was
found to contain an activating receptor that was immunoprecipitated by
both 4E5 and 12A8 (Fig. 2B). The appearance of a 60-kDa
band is not always reproducible, as noted by its absence in Fig. 2A, and is most likely an Ig H chain breakdown product. The
4E5high12A8+ population
contained the 4E5/12A8-reactive activating receptor as well as a
4E5+12A8- inhibitory
molecule (Fig. 2B), most probably the same one observed in
the 4E5+12A8- population
(Fig. 2A). These experiments suggest the existence of at
least two different, independently expressed
Ly49DB6-related molecules in 129/J mice. It
appears that 12A8 only reacts with the activating receptor, whereas 4E5
reacts with both. The existence of a 4E5-reactive inhibitory receptor
corroborates two-color flow cytometric analyses showing that 129/J, but
not B6, mice contain a small but measurable population of
4E5+CD3+ T cells (data not
shown). In B6 mice, expression of Ly49D is not detectable on T cells
(50). The discovery that 4E5 recognizes an inhibitory Ly49
protein in 129/J mice supports observations that activating Ly49
receptors are not expressed on T cells. The 129/J mice express at least 10 different Ly49-related genes
To further characterize the 129/J Ly49 repertoire, a cDNA library
was constructed using mRNA from IL-2-activated 129/J liver NK cells.
This library was then probed with Ly49eB6,
dB6,
g2B6, and
o129/J full-length coding regions, and
100 positive clones were sequenced. A total of 10 distinct Ly49-related
sequences were found. Table I shows the
percentage of amino acid identity between the new 129/J Ly49 cDNAs and
all known B6 Ly49 full-length coding regions. Only one sequence
(Ly49e129/J) was found to be highly similar
to a B6 Ly49 gene. Ly49e129/J contained
only one silent nucleotide difference when compared with the
Ly49eB6 nucleotide-coding region. Both
Ly49g1 and g2 isoforms were found in 129/J NK
cells, with the Ly49g1 insert sharing 37/39 nucleotides
between the 129/J and B6 alleles (data not shown).
|
.
Ly49R129/J and U129/J are
putative activating receptors that lack an ITIM domain and contain a
transmembrane arginine residue (Fig. 3).
Ly49R129/J and U129/J are
most similar to Ly49DB6 and
HB6, respectively, whereas the inhibitory
Ly49S129/J and V129/J are
most similar to Ly49FB6 and
AB6, respectively (Table I).
Ly49T129/J is distantly related to
Ly49G2B6.
|
Reactivity of 129-Ly49 proteins to Ly49-specific mAb
To identify the molecules responsible for the staining and
biochemical patterns observed in Figs. 1 and 2, expression constructs
were transiently transfected into 293T cells and stained with a panel
of Ly49-specific mAb. Three receptors,
Ly49O129/J, R129/J, and
V129/J, were found to react strongly with 4E5
and/or 12A8 (Fig. 4 and Table II). As predicted by the biochemical data
(Fig. 2B), the putative activating
Ly49R129/J was bound by both 4E5 and 12A8,
whereas the two putative inhibitory receptors,
Ly49O129/J and V129/J, were
recognized by only 4E5 and not 12A8 (Fig. 4). This contrasts with our
previous finding that Ly49O129/J did not bind to
4E5 (11). This discrepancy was resolved by sequencing of
the PCR-derived Ly49o129/J expression
construct, which was found to have a frameshift mutation. These
staining results, along with the immunoprecipitation data, suggest that
the 4E5+12A8- population
contains NK cells expressing Ly49O129/J and/or
V129/J but not Ly49R129/J.
The 4E5high12A8+ population
coexpresses Ly49O129/J and/or
V129/J and Ly49R129/J.
Finally, the
4E5low12A8+-activating
population is due to the singular expression of
Ly49R129/J (Fig. 1C).
|
|
). For example, 4D11 was found to bind to
129/J, Ly49G2129/J, and
Ly49T129/J. YE1/48 showed weak reactivity with
Ly49P129/J, V129/J,
T129/J, and R129/J. YE1/32
produced a reactivity pattern similar to YE1/48 but of lesser
intensity. A1, which has been shown to bind
Ly49AB6 and Ly49PNOD
(24, 52), was also found to react with
Ly49P129/J. A newly developed mAb, 1F8, which
binds to Ly49CB6, IB6, and
HB6 was found to react well with
Ly49I129/J and, to a slightly lesser extent,
Ly49U129/J. As expected, mAb 5E6 did not bind to
any 129/J Ly49 molecules. This Ab was produced in the 129/J strain and
was previously shown not to bind 129/J NK cells to any detectable
degree (13, 53). Soluble murine MHC class I tetramers reveal ligand specificities of the 129/J Ly49 repertoire
Fluorochrome-labeled soluble murine class I MHC "tetramers"
have previously been used to identify the specificity of various
members of the B6 Ly49 repertoire (25, 26). Therefore, to
investigate the MHC-binding characteristics of the 129/J Ly49
repertoire, the tetramer approach was used. To determine the level of
expression, 293T cells transfected with individual Ly49 expression
constructs were stained with specific mAb or PE-conjugated murine class
I tetramers Db, Dd,
Dk, Kb,
Kd, Kk, and
Ld (Fig. 5). No
control staining for Ly49S129/J expression is
shown, due to a lack of Ly49S-reactive Abs (Table II). Otherwise, all
transfected cells showed relatively high levels of appropriate Ly49
expression (Fig. 5). Observed differences in expression levels of
individual receptors may be due to differences in detection (FITC- vs
PE-labeling). Also, it cannot be assumed that the cross-reactive mAbs
used all share similar affinities for the 129/J Ly49 proteins. As a
control, tetramers were used to stain 293T cells transiently expressing
Ly49AB6, CB6, and
DB6. Tetramer binding to these three receptors
was found to be almost identical with that reported previously (data
not shown).
|
) (25).
Ly49G2129/J also bound to
Dk and Kd, but to a lesser
extent. Ly49O129/J,
Ly49R129/J, and Ly49V129/J
bound Dd and Dk. To our
knowledge, this is the first report of a MHC tetramer binding to an
activating receptor, Ly49R129/J. The tetramer
binding by Ly49O129/J and
V129/J was particularly strong, especially
Ly49V129/J to Dk (Fig. 5).
Surprisingly, Ly49O129/J and
Ly49V129/J also showed strong reactivity to
Ld (Fig. 5), something not seen with their
closest inhibitory B6 counterpart, Ly49AB6, which
was found to bind Dd and Dk
(25, 26).
In addition to binding Dd,
Dk, and Ld,
Ly49V129/J also bound strongly to
Db. The ability of
Ly49V129/J and Ly49O129/J
to bind the autologous Db haplotype suggests that
they can detect the presence of self-MHC as shown for the related
Ly49AB6. (26).
Ly49V129/J showed weaker reactivity toward
Kb, Kd, and
Kk (Fig. 5). Ly49V129/J is
reminiscent of Ly49CB6 in its ability to bind
many types of murine class I MHC (25).
Ly49S129/J, Ly49T129/J, and
Ly49U129/J showed no significant binding to any
MHC tetramers, whereas the third activator,
Ly49P129/J, only showed a low level of binding to
Dd (Fig. 5). Finally,
Ly49I129/J-transfected 293T showed the second
highest promiscuity with reactivity toward Dk,
Kb, Kd, and
Kk (Fig. 5). The MHC tetramer-binding data
obtained for Ly49O129/J and
V129/J from transiently transfected human cells
was confirmed in stably transfected rat RNK-16 cells (data not
shown).
The tetramer binding experiments described above were all conducted on
ice. However, this may not reflect the binding potential of Ly49 and
MHC proteins at physiological temperature. To test the
temperature-dependency of tetramer binding, duplicate samples of
Ly49-transfected 293T cells were incubated on ice or at 37°C and then
analyzed by flow cytometry. Tetramer binding at 37°C increased over
that seen on ice where it was already present and, in some cases,
appeared where it was not observed on ice. There were notable increases
in the binding of Db to
Ly49G2129/J, O129/J, and
R129/J (Table III).
However, in the majority of cases where "cold" tetramer binding was
not observed, there was no promiscuity induced at 37°C.
|
).
Because Ly49R129/J and
Ly49DB6 are closely related, this prompted us to
perform a similar experiment with
Ly49DB6-transfected 293T. No binding of
Ly49DB6-expressing 293T was found for any
tetramers when incubated on ice as previously reported
(24), but, at 37°C, the Dd,
Dk, and Ld tetramers were
bound (data not shown). This brings into question the ability of other
activating Ly49 proteins to bind MHC tetramers at 37°C, but
Ly49P129/J and U129/J did
not show enhanced tetramer binding at the higher incubation temperature
(Table III). Interestingly, there was no increase in tetramer binding
to Ly49I129/J at 37°C over that already seen on
ice. Also of interest was the significantly increased binding of the
H-2b,d,k alleles of H-2K for
Ly49V129/J (Table III). Incubation at
physiological temperature shows significant binding of
Ly49V129/J to all soluble class I MHC tetramers
used in this study. This suggests that Ly49V+
129/J NK cells monitor host cells for a global down-regulation of class
I MHC. Ly49O and/or V+ NK cell cytotoxicity is inhibited by class I MHC Dd
Identification of mAb that can bind to the novel 129/J Ly49
proteins allowed the possibility of isolating different NK
subpopulations for functional analysis to confirm the observations of
the tetramer-binding experiments. When transiently expressed in 293T or
stably expressed in RNK-16, Ly49O129/J and
V129/J induced binding of the
Dd tetramer (Fig. 5 and data not shown). Because
both of these receptors are predicted to be inhibitory in nature, the
Ly49O and/or V (hereafter referred to as Ly49O/V)-positive NK subset
should be suppressed by Dd-expressing target
cells. To test this hypothesis, 129/J NK cells were sorted into
4E5+12A8-
(Ly49O/V-positive and Ly49R-negative),
4E5+12A8+ (Ly49O/V-positive
and Ly49R-positive), and
4E5-12A8- (Ly49O and
V-negative and Ly49R-negative) populations and used as effector cells
against the BA/2-derived (H-2d) P815 target cell.
The single- and double-positive populations, which contain
Ly49O129/J and/or V129/J,
lysed P815 poorly (Fig. 6B).
In contrast, double-negative NK cells displayed a 4-fold increase in
killing against this H-2d target cell (Fig. 6B). The classical NK target cell YAC-1 was used as a
control for the cytotoxic potential of the differentially sorted
populations. All three populations lysed YAC-1 cells
efficiently at low E:T ratios in a fashion comparable to
unsorted splenic B6 NK cells (Fig. 6A). This indicates that
the 4E5-reactive subpopulation contains a receptor that can inhibit NK
cytotoxicity in the presence of a H-2d-derived
ligand.
|
C). Both
4E5+12A8-
(Ly49O/V-positive and R-negative) and
4E5-12A8-
(Ly49O/V-negative and R-negative) NK cells displayed similar levels of
cytotoxicity toward the YB20 parental line (Fig. 6D). In
contrast, lysis by the
4E5+12A8- subpopulation
was severely inhibited against the Dd-expressing
YB20 transfectant in comparison to the lysis by the
4E5-12A8- subpopulation
(Fig. 6E). This suggests that Dd is a
ligand for Ly49O129/J and/or
V129/J. Unsorted 129/J NK cells also displayed
decreased cytotoxicity toward YB/Dd compared with
the sorted 4E5-12A8-
population (Fig. 6E). This is consistent with the expression
of Ly49G, O, and V on unsorted 129/J NK cells, especially
Ly49G129/J, which is an abundant 129/J receptor
that can also bind the Dd tetramer (Fig. 5).
Cytotoxicity and cytokine secretion of Ly49D+ B6
NK cells in response to Dd-expressing target
cells can be blocked by preincubation with
4E5-F(ab')2 (20, 23). Because 4E5
also recognizes
Ly49O129/J/V129/J, the
inhibitory signal should also be affected by such treatment. Indeed,
blockade with 4E5-F(ab')2 resulted in the
restoration of cytotoxicity by the
4E5+12A8- population to
levels obtained by the
4E5-12A8- population
toward YB/Dd (Fig. 6F). We found that
4E5 F(ab')2 did not affect the lysis of
YB/Dd induced by
4E5-12A8- cells.
Collectively, these results indicate that
Ly49O129/J and/or V129/J
bind to murine class I Dd and signal in a manner
which inhibits NK cell cytotoxicity. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The ITIM of Ly49T129/J (VTYSTM) is different from
that of all other known inhibitory Ly49 proteins (VxYxxV) and is
identical with the binding motif for phosphatidylinositol
(PI)3-kinase (YxxM) (Fig. 3). To test for possible PI3-kinase
recruitment, we pervanadate-treated
L929-Ly49T129/J stable transfectants,
immunoprecipitated with a pan-Ly49 antiserum, separated proteins by
SDS-PAGE, and blotted with an anti-PI3 kinase (anti-p85
subunit) mAb. Although p85 was readily detectable in whole-cell lysate,
there was no evidence of its existence in immunoprecipitated samples.
We subsequently blotted with anti-phosphotyrosine and observed the
phosphorylated form of Ly49T129/J (data not
shown). Therefore, the ITIM of Ly49T129/J is
phosphorylated but does not seem to associate with PI3-kinase.
Compared with the divergence of the other nine 129/J Ly49 genes, the conservation of Ly49e between B6 and 129/J is intriguing. The conservation may emphasize an important and perhaps separate function for this Ly49 gene. It is interesting to note that Ly49e transcripts, unlike Ly49a–d, g, and i, are readily detectable in B6 fetal thymic and liver NK preparations (55). The divergence of the other Ly49 genes between 129/J and B6 may reflect the fact that these strains are very distantly related, despite sharing the same MHC haplotype (56).
Interestingly, Ly49I129/J bound
Kb tetramers, whereas
Ly49V129/J, G2129/J, and
O129/J recognized Db (Fig. 5 and Table III). Both tetramers represent self-MHC Ags of 129/J mice.
Ly49CB6 binds to both Kb
and Db in B6 mice (25). In contrast,
Ly49IB6 did not significantly bind to
H-2b tetramers (25) but, when
overexpressed as a transgene, inhibited the rejection of
H-2b bone marrow transplants (57).
These results show that both B6 and 129/J mice possess multiple
inhibitory receptors for self-MHC Ags. However, it is clear that even
at physiological temperatures there is wide variation in the binding of
different Ly49 molecules for the same class I MHC (Table III). Such
variation in binding ability may be a consequence of the peptide being
presented, as in the case of Ly49IB6
(25), or even a result of the presence or absence of
peptide altogether, as has recently been reported for
Ly49CB6 (58).
The three activating Ly49 cDNA species, Ly49P, R, and U,
expressed in 129/J mice did not bind the Kb
tetramer, but Ly49R129/J was able to bind
Db at 37°C (Fig. 5 and Table III). Despite this
apparent self-reactivity, Ly49R seems to be expressed at moderate
levels on 129/J NK cells (see
4E5low12A8+ population in
Fig. 1C). Therefore, to avoid autoreactivity, all
Ly49R129/J-expressing NK cells must also possess
one or more inhibitory receptors for self-MHC Ags, and, considering the
wide array of such receptors as discussed above, this is likely the
case. In contrast, no self-MHC class I binding was reported for
Ly49DB6 (25). Although MHC-binding
data for Ly49HB6 is unavailable, it is entirely
possible that Ly49HB6 may bind to an
H-2b ligand because the extracellular domain of
Ly49HB6 is very close to that of
Ly49CB6 and IB6
(51). Unlike the study by Hanke et al.
(25), where no detectable tetramer binding was
reported for Ly49DB6, we have found that
Ly49R129/J bound weakly to
Dd, Dk, and
Ld on ice (Fig. 5 and Table III). This suggests
that the affinity of different activating receptors for MHC ligands may
vary widely. It also appears that the tetramer binding by
Ly49R129/J is not an anomaly, because we have
found that the recently described activating receptor in CBA/J and C3H
mice, Ly49L (12), also bound weakly to
Dd tetramers (data not shown). The
well-documented ability of Ly49D+ NK cells to
confer rapid cytotoxicity and cytokine production in response to
Dd-expressing target cells reveals that
tetramer-binding data may actually understate the binding potential of
Ly49 molecules for in vivo-expressed MHC (20, 21). The
increased binding of Ly49R129/J to
Dd, Dk, and
Ld, as well as the additional binding to
Db at 37°C (Table III), and our finding that
Ly49DB6 binds to Dd at
physiological temperatures supports this position (data not shown).
Recently, the crystal structure of Ly49AB6 bound
to Dd was deduced by Tormo et al. and the sites
of contact between the two proteins predicted (59). The
Ly49AB6 residues required for site 1 interaction
with Dd, predicted to be the trans
binding site (between a ligand and receptor on different cells), are
fairly well conserved in those 129/J Ly49 proteins that were found to
interact with Dd tetramers. In this study,
Ly49G2129/J, O, V, R, and, to a much lesser
extent, P, were found to bind Dd (Fig. 5 and
Table III). Of the 12 Ly49AB6 residues thought to
be used for Dd binding in trans,
Ly49G2129/J has 7 of 12,
Ly49O129/J has 10 of 12,
Ly49V129/J has 9 of 12,
Ly49R129/J has 10 of 12, and
Ly49P129/J has in common 12 of 12. It is
interesting to note that the lowest (but still detectable)
Dd-tetramer binding was seen with
Ly49P129/J, which had all the predicted
trans binding sites in common with
Ly49AB6, but the best binding was seen with
Ly49V129/J, which had only 9 of 12 binding sites
conserved. Also, Ly49G2129/J had the least number
of residues in common with Ly49AB6 for
Dd binding, 7 of 12, but also bound the
Dd tetramer much better than
Ly49P129/J (Fig. 5 and Table III). Our study,
like that of Tormo et al., used bacterially expressed MHC and the same
peptide for presentation by Dd. This suggests
that other as yet unidentified residues may be important for MHC
binding by Ly49 proteins. Comparison of the predicted MHC-contacting
residues of the two pan-MHC-binding proteins,
Ly49CB6 and Ly49V129/J,
showed that they agreed at only 1 of 12 trans residues,
despite Ly49CB6 binding Dd
tetramers very well (25). This suggests that the
Ly49C-like family members use an altogether different combination of
residues for MHC-binding compared with Ly49A-like family members.
ES cells of 129/SvJ origin are the most widely used in current mouse gene-mutation studies, although most of the functional assays used to characterize the mutant mice were developed in non-129 mouse strains. Because 24% of 212 129/SvJ genomic sequence markers differ from the consensus 129 genotype (60), 129/SvJ is considered to be "contaminated." We have found that 129/SvJ splenocytes express transcripts identical with Ly49r129/J, s129/J, and v129/J (data not shown). This agrees with the prediction that chromosome 6, where the Ly49 gene cluster resides, is of 129 origin in 129/SvJ (60). Our current findings suggest that an ES cell of 129 origin would not be the appropriate choice for deletion of a gene mapping near the Ly49 locus where the phenotype of interest is immunological in nature. Even with many backcrosses to the phenotypically important strain, closely linked genes may be carried over and affect data interpretation. The divergence of the Ly49 repertoire of 129 mice, and the strain in general (56), from B6 mice requires a careful assessment of the use of this strain for the production of gene-targeted mice as it pertains to immunological studies, especially to the field of NK biology. The results presented here indicate that NK cells expressing a 129/J NK complex would possess a vastly different Ly49 repertoire with associated differences in MHC specificity and responses to these ligands. We predict that other inbred mouse strains may possess similarly divergent Ly49-repertoires, which may affect their respective NK cell functions.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Stephen K. Anderson, Science Applications International Corp. Frederick, National Cancer Institute, Frederick Cancer Research and Development Center, Building 560, Room 31-93, Frederick, MD 21702-1201.
3 Abbreviations used in this paper: KIR, killer cell Ig-like receptor; B6, C57BL/6; ITIM, immunoreceptor tyrosine-based inhibitory motif; ES, embryonic stem; PI, phosphatidylinositol.
Received for publication November 30, 2000. Accepted for publication February 14, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
secretion. J. Immunol. 164:603. interferon but not the immunoglobulin response in mice that vary in susceptibility to Sendai virus pneumonia. J. Virol. 69:5592.[Abstract]
gt recombinant clones by hybridization to single plaques in situ. Science 196:180.This article has been cited by other articles:
|
|
 |
|
  A. Rouhi, C. B. Lai, T. P. Cheng, F. Takei, W. M. Yokoyama, and D. L. Mager Evidence for high bi-allelic expression of activating Ly49 receptors Nucleic Acids Res., September 1, 2009; 37(16): 5331 - 5342. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Xie, M. D. Stadnisky, and M. G. Brown MHC Class I Dk Locus and Ly49G2+ NK Cells Confer H-2k Resistance to Murine Cytomegalovirus J. Immunol., June 1, 2009; 182(11): 7163 - 7171. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L.-H. Tai, M.-L. Goulet, S. Belanger, N. Toyama-Sorimachi, N. Fodil-Cornu, S. M. Vidal, A. D. Troke, D. W. McVicar, and A. P. Makrigiannis Positive regulation of plasmacytoid dendritic cell function via Ly49Q recognition of class I MHC J. Exp. Med., December 22, 2008; 205(13): 3187 - 3199. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. S. Goldszmid, A. Bafica, D. Jankovic, C. G. Feng, P. Caspar, R. Winkler-Pickett, G. Trinchieri, and A. Sher TAP-1 indirectly regulates CD4+ T cell priming in Toxoplasma gondii infection by controlling NK cell IFN-{gamma} production J. Exp. Med., October 29, 2007; 204(11): 2591 - 2602. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Pascal, N. R. Nathan, E. Claudio, U. Siebenlist, and S. K. Anderson NF-{kappa}B p50/p65 Affects the Frequency of Ly49 Gene Expression by NK Cells J. Immunol., August 1, 2007; 179(3): 1751 - 1759. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Kielczewska, H.-S. Kim, L. L. Lanier, N. Dimasi, and S. M. Vidal Critical Residues at the Ly49 Natural Killer Receptor's Homodimer Interface Determine Functional Recognition of m157, a Mouse Cytomegalovirus MHC Class I-Like Protein J. Immunol., January 1, 2007; 178(1): 369 - 377. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. H. Johansson, M. A. Taylor, M. Jagodic, K. Tus, J. D. Schatzle, E. K. Wakeland, and M. Bennett Mapping of Quantitative Trait Loci Determining NK Cell-Mediated Resistance to MHC Class I-Deficient Bone Marrow Grafts in Perforin-Deficient Mice J. Immunol., December 1, 2006; 177(11): 7923 - 7929. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. L. Rogers, A. Rouhi, F. Takei, and D. L. Mager A Role for DNA Hypomethylation and Histone Acetylation in Maintaining Allele-Specific Expression of Mouse NKG2A in Developing and Mature NK Cells J. Immunol., July 1, 2006; 177(1): 414 - 421. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. A. Benoit, J. Shannon, J. W. Chamberlain, and R. G. Miller Influence of Xenogeneic {beta}2-Microglobulin on Functional Recognition of H-2Kb by the NK Cell Inhibitory Receptor Ly49C J. Immunol., September 15, 2005; 175(6): 3542 - 3553. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Stevenaert, K. Van Beneden, V. De Colvenaer, A. S. Franki, V. Debacker, T. Boterberg, D. Deforce, K. Pfeffer, J. Plum, D. Elewaut, et al. Ly49 and CD94/NKG2 receptor acquisition by NK cells does not require lymphotoxin-{beta} receptor expression Blood, August 1, 2005; 106(3): 956 - 962. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Bubic, M. Wagner, A. Krmpotic, T. Saulig, S. Kim, W. M. Yokoyama, S. Jonjic, and U. H. Koszinowski Gain of Virulence Caused by Loss of a Gene in Murine Cytomegalovirus J. Virol., July 15, 2004; 78(14): 7536 - 7544. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. P. Makrigiannis, E. Rousselle, and S. K. Anderson Independent Control of Ly49g Alleles: Implications for NK Cell Repertoire Selection and Tumor Cell Killing J. Immunol., February 1, 2004; 172(3): 1414 - 1425. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. J. Ma, E. T. Silver, B. Hazes, and K. P. Kane Reciprocal Transfer of Class I MHC Allele Specificity between Activating Ly-49P and Ly-49W Receptors by Exchange of {beta}4-{beta}5 Loop Residues J. Immunol., November 15, 2003; 171(10): 5337 - 5344. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Voigt, C. A. Forbes, J. N. Tonkin, M. A. Degli-Esposti, H. R. C. Smith, W. M. Yokoyama, and A. A. Scalzo Murine cytomegalovirus m157 mutation and variation leads to immune evasion of natural killer cells PNAS, November 11, 2003; 100(23): 13483 - 13488. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. H. Mason, J. Willette-Brown, S. K. Anderson, W. G. Alvord, R. L. Klabansky, H. A. Young, and J. R. Ortaldo Receptor Glycosylation Regulates Ly-49 Binding to MHC Class I J. Immunol., October 15, 2003; 171(8): 4235 - 4242. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. T. Silver, K. J. Lavender, D.-E. Gong, B. Hazes, and K. P. Kane Allelic Variation in the Ectodomain of the Inhibitory Ly-49G2 Receptor Alters Its Specificity for Allogeneic and Xenogeneic Ligands J. Immunol., November 1, 2002; 169(9): 4752 - 4760. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. W. McVicar, R. Winkler-Pickett, L. S. Taylor, A. Makrigiannis, M. Bennett, S. K. Anderson, and J. R. Ortaldo Aberrant DAP12 Signaling in the 129 Strain of Mice: Implications for the Analysis of Gene-Targeted Mice J. Immunol., August 15, 2002; 169(4): 1721 - 1728. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. H. Lian, M. Maeda, S. Lohwasser, M. Delcommenne, T. Nakano, R. E. Vance, D. H. Raulet, and F. Takei Orderly and Nonstochastic Acquisition of CD94/NKG2 Receptors by Developing NK Cells Derived from Embryonic Stem Cells In Vitro J. Immunol., May 15, 2002; 168(10): 4980 - 4987. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Saleh, A. P. Makrigiannis, D. L. Hodge, and S. K. Anderson Identification of a Novel Ly49 Promoter That Is Active in Bone Marrow and Fetal Thymus J. Immunol., May 15, 2002; 168(10): 5163 - 5169. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Sundback, A. Achour, J. Michaelsson, H. Lindstrom, and K. Karre NK Cell Inhibitory Receptor Ly-49C Residues Involved in MHC Class I Binding J. Immunol., January 15, 2002; 168(2): 793 - 800. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
