| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Structural Biology Section and
Laboratory of Immunogenetics, National Institute of Allergy and Infectious Disease, National Institutes of Health, Rockville, MD 20852; and
Institut fur Anthropologie und Humangenetik der Universitat Munchen, Munich, Germany
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The NKG2 proteins are a family of closely related polypeptides (NKG2A–F) that, like CD94, are type II integral membrane proteins of the C-type lectin superfamily (8, 9). Similar to other families of NK cell receptors that interact with class I molecules, such as KIR and the murine Ly49 receptors (10), some NKG2 family members appear to have diametrically opposed functions (11). NKG2A/B contains immunoreceptor tyrosine-based inhibitory motifs consistent with an inhibitory function, whereas NKG2C and E possess charged residues in their transmembrane domains that appear to be characteristic of other NK receptors, such as KIR2DS2, that interact with DAP12 to transmit an activation signal (12).
The first evidence that CD94/NKG2 receptors recognized HLA class I proteins came from transfection studies in which expression of certain HLA-B alleles protected target cells from NK cell-mediated lysis, and this protection was shown to be abrogated by CD94-specific mAb (5). Subsequent studies showed that transfection of many, but not all, HLA class I alleles, including the nonclassical HLA-G gene, could mediate protection from NK cell-mediated lysis that could be reversed by the addition of CD94-specific mAb or class I specific mAb (6, 7, 13, 14). Consequently, the CD94/NKG2 receptors were thought to interact with a large number of different HLA class I molecules. However, no obvious common structural features distinguish protective from nonprotective class I molecules. Recent studies have demonstrated an interaction between CD94/NKG2 receptors and HLA-E complexed with peptides derived from the leader sequence of some HLA class I heavy chains (15, 16, 17). These observations suggested that the CD94/NKG2 receptors specifically interact with HLA-E, but did not rule out the possibility of interactions with other class I molecules. The data presented here confirm a direct interaction between HLA-E and CD94/NKG2A and demonstrate that this interaction is peptide dependent and peptide specific. Moreover, no evidence of any interaction between CD94/NKG2A and HLA-A2, -A3, -B27, or -Cw3 was observed, suggesting that CD94/NKG2A does not recognize classical HLA class I Ags, but that HLA-E is its primary, if not sole, ligand.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The NK leukocytosis 221707 cells, which lack inhibitory KIR and express the CD94/NKG2A receptor, have been described previously (15). RMA-S cells transfected with human ß2m (hß2m), or hß2m together with HLA-E, HLA-A2, HLA-A3, and HLA-B27 and RMA-S cells transfected with HLA-Cw3 alone have been previously described (15, 18, 19). L cells transfected with HLA class I molecules were obtained from Centro di Biotecnologie (Avanzate, Italy) and were a gift from Dr. J. McCluskey (University of Melbourne, Melbourne, Australia). FITC-conjugated anti-HLA class I mAb (B9.12.1) was purchased from Immunotech (Westbrook, ME). Streptavidin-conjugated phycoerythrin (PE) was purchased from Molecular Probes (Eugene, OR), and streptavidin-conjugated horseradish peroxidase (HRP) was obtained from Amersham (Arlington Heights, IL). The NKG2A-specific mAb 1A12 (11) was provided by Dr. J. P. Houchins. Peptides were synthesized and purified as previously described (20).
Construction of baculovirus shuttle vectors
A cDNA encoding a hexahistidine sequence followed by a spacer region and the extracellular domain of NKG2A (amino acid residues 100–233) was amplified by PCR using the synthetic oligonucleotides GGGGGATCCCATCATCATCATCACCACGAAGGATCGGGAAGGCACAACAATTCTTCCCTG and GTGAATTCTAGATTACCTCTAAAGCTTATGCTTACAATG and was cloned into pAcGP67B (PharMingen, San Diego, CA) downstream of the GP67 leader sequence as a BamHI/EcoRI fragment to create pAc-sol2Ahis. The cDNA encoding the extracellular domain of CD94 (residues 34–179) was generated by PCR using the oligonucleotides GGGGGATCCTCTTTTACTAAACTGAGTATTGAGCCAGC and CCGGAATTCCCATGGAAACATTTAAATGAGCTGTTG and was also cloned into pAcGP67B as a BamHI/EcoRI fragment to create pAc-sol94. Annealed oligonucleotides encoding the substrate sequence for the BirA enzyme (GATCGCTGCATCATATTCTGGATGCACAGAAAATGGTGTGGAATCATCGTGGGTCCGGAA and GATCTTCCGGACCCACGATGATTCCACACCATTTTCTGTGCATCCAGAATATGATGCAGC) were then inserted by ligation into the BamHI site of pAc-sol94 to create pAc-sol94BirA. DNA fragments encoding both the entire GP67/BirA/CD94 and GP67/hexahistidine/NKG2A were then produced by PCR using the synthetic oligonucleotides GGGAGATCTATGCTAGTAAATCAGTCACACC and GGAAGATCTGGAAAGGATCAGATCTGCAGCGGC and were cloned into the dual promoter transfer vector pAcUW51 (PharMingen) at the BamHI and BglII sites, respectively, to create pAcsol2Ahis/94BirA. The sequence of the inserts was confirmed by dsDNA sequencing, and the transfer vector was cotransfected with Baculogold-linearized baculovirus DNA (PharMingen) into sf21 cells using Cellfectin (Life Technologies, Gaithersburg, MD) to create recombinant baculovirus. Recombinant baculovirus stock was obtained from the supernatant of cotransfected cells, and the recombinant baculovirus was cloned by limiting dilution. Cloned viral stocks were amplified and screened for expression of NKG2A by Western blot, and clones expressing the highest levels of NKG2A were selected for further amplification and production of recombinant protein.
Production of fluorescently labeled soluble CD94/NKG2A
Culture supernatants from sf21 cells infected with a recombinant
baculovirus directing the expression of soluble CD94/NKG2A were
concentrated through a hollow-fiber filter and initially purified using
Ni-NTA agarose (Qiagen, Santa Clarita, CA). The recombinant protein was
eluted in 10 mM Tris (pH 8.0), 50 mM NaCl, and 0.5 M imidazole;
dialyzed against 10 mM Tris (pH 8.0); and further purified by gel
filtration on a Superdex 200 HR column (Pharmacia, Uppsala, Sweden).
Purified protein was biotinylated by incubation at 30°C for 16–24 h
with BirA in a buffer containing 50 mM bicine (pH 8.3), 40 µM biotin,
10 mM magnesium acetate, and 10 mM ATP (Avidity, Denver, CO). The
biotinylated heterodimer was separated from free biotin by gel
filtration and then multimerized by the addition of
streptavidin-coupled PE (Molecular Probes). Multimeric CD94/NKG2A/PE
was purified from the heterodimeric species by gel filtration. The
fluorescently labeled CD94/NKG2A/PE was titrated for activity, and a
working dilution was established (generally corresponding to 
0.5
µg of protein/binding reaction).
Direct binding assay
RMA-S cells or their transfected derivatives were cultured with synthetic peptide (300 µM) at 26 or 37°C for 12–18 h. Cells were harvested, washed, and incubated with PE-labeled CD94/NKG2A/PE for 30 min on ice. Cells were then washed and analyzed by flow cytometry on a FACSort (Becton Dickinson, San Jose, CA). Experiments were performed at least four times, and single representative experiments are shown in the figures.
Cytotoxicity assays
Target cells were incubated overnight at the indicated temperature in the absence or presence of peptide (300 µM) and for the last 90 min were labeled with sodium [51Cr]chromate (DuPont, Boston, MA). The cytotoxic activity of 221707 NK cells was performed in 3-h chromium release assays as previously described using 2500 target cells/well and an E:T cell ratio of 60:1 (15). The radioactivity present in 100 µl of supernatant was measured using a 4/600 ME Plus gamma counter (ICN Biomedicals, Huntsville, AL). The percent specific cytotoxicity was calculated as previously described (19). Data shown are representative of multiple experiments at a variety of E:T cell ratios and are derived from the means of triplicate samples.
Immunoblot analyses
Western blots were performed as previously described (21). Purified protein preparations were separated by SDS-PAGE and transferred to nitrocellulose (Hybond C Extra, Amersham). To detect biotinylated proteins, membranes were blocked with 3% BSA in PBS before incubation with streptavidin-conjugated HRP (Amersham) diluted 1/3000 in PBS/0.1% Tween-20. The membranes were then washed with PBS containing 0.1% Tween-20/3% BSA. Biotinylated proteins were visualized using enhanced chemiluminescence (ECL, Amersham). To detect recombinant NKG2A, membranes were blocked with PBS containing 3% skim milk, before incubation with the NKG2A-specific mAb 1A12. Excess mAb was removed by washing the blot with PBS containing 0.1% Tween-20 and 3% skim milk before addition of goat anti-mouse IgG-conjugated HRP (Amersham). The blot was washed, and the proteins were visualized using enhanced chemiluminescence.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To examine the ligand specificity of CD94/NKG2A, a soluble form of
the receptor was produced. A recombinant baculovirus was engineered
that expressed both the NKG2A extracellular domain linked to a
hexahistidine sequence and the CD94 extracellular domain linked to a
substrate sequence for the BirA enzyme, each inserted behind the GP67
leader sequence (Fig. 1
A).
Recombinant protein from the supernatant of infected sf21 cells was
purified using immobilized metal chelate chromatography. The CD94
subunit of the purified recombinant protein was enzymatically labeled
with biotin, and the labeled protein was subjected to SDS-PAGE and
analyzed by Western blot (Fig. 1B). Incubation with
streptavidin-conjugated HRP to detect the CD94 polypeptide revealed a
single biotinylated protein of 24 kDa under reducing conditions and of
about 43 kDa under nonreducing conditions. Under reducing conditions,
parallel analysis with the NKG2A-specific mAb 1A12 revealed a single
NKG2A-reactive species slightly smaller in molecular mass than the
recombinant CD94. Under nonreducing conditions, a single band was
observed of the same molecular size as the protein revealed
by SA-HRP. These data indicate that the purified recombinant protein is
a disulfide-linked 43-kDa heterodimer that consists of the
extracellular domain of CD94 covalently associated with that of NKG2A.
|
). Moreover, binding of recombinant
CD94/NKG2A/PE to HLA-E was inhibited by the addition of either HLA
class I- or CD94-specific mAb. No binding of recombinant CD94/NKG2A/PE
to RMA-S/HLA-E transfectants was observed when cells were cultured
without synthetic peptide or in the presence of irrelevant peptide (see
Figs. 5 and 7). These data indicate that soluble recombinant
CD94/NKG2A/PE retains the specificity of the native receptor.
|
|
|
While it is evident that CD94/NKG2 receptors can interact with HLA-E (15, 16, 17), an interaction with other HLA class I molecules, as suggested in earlier studies (5, 7), has not been excluded. To determine whether CD94/NKG2A interacts primarily with HLA-E complexed to HLA class I leader sequence peptides or is broadly reactive with a large number of HLA class I molecules, recombinant CD94/NKG2A/PE was used in cell binding experiments with a panel of RMA-S transfectants expressing different classical HLA class I molecules or HLA-E.
RMA-S/HLA-class I transfectants were incubated at 26°C with synthetic
peptides and stained with either class I-specific mAb to detect cell
surface levels of HLA-E or recombinant CD94/NKG2A/PE to analyze its
ligand specificity. As expected, RMA-S/HLA class I transfectants after
incubation at 26°C without peptide showed significant levels of cell
surface HLA class I expression compared with cells expressing
hß2m alone (Fig. 3 and
Table I). HLA class I expression was
further enhanced by the addition of specific synthetic peptides known
to bind to each HLA class I molecule (19, 20, 22). No binding of
recombinant CD94/NKG2A/PE to RMA-S/HLA-E transfectants was observed in
the absence of class I leader sequence peptides despite the induction
of significant levels of cell surface HLA-E as a result of incubation
at 26°C. Binding of recombinant CD94/NKG2A/PE to RMA-S cells
expressing HLA-E was evident only when these cells were cocultured with
an appropriate synthetic leader sequence peptide. Transfection of
HLA-A2, -A3, and -Cw3 into 721.221 cells (which express HLA-E, but not
HLA-A, -B, or -C proteins) has previously been shown to be protect
these cells from NK-mediated lysis through an interaction with
CD94/NKG2 receptors (6). However, no detectable binding of recombinant
CD94/NKG2A/PE was observed to RMA-S cells expressing either
hß2m alone or HLA-A2, -A3, -B27, and -Cw3 in the absence
or the presence of specific HLA binding peptides. The failure to
observe binding to peptide-loaded RMA-S transfectants expressing
classical HLA class I molecules is unlikely to be due to the use of
inappropriate peptides, as soluble CD94/NKG2A/PE does not bind to L
cells transfected with HLA-A2, -A24, -A26, -B27, or -Cw7, each of which
complexes with a myriad of endogenous peptides (data not shown). Thus,
these data indicate that CD94/NKG2A is not broadly cross-reactive with
HLA class I molecules, but that the interaction with HLA-E is specific.
|
|
221707 NK cells express the inhibitory CD94/NKG2A receptor and do
not express inhibitory KIR (15). These cells were used in cytotoxicity
assays with the RMA-S/HLA-class I transfectants to functionally verify
the binding data obtained using recombinant CD94/NKG2A/PE. As expected
incubation of RMA-S/HLA-E transfectants with an HLA leader sequence
peptide resulted in reduced levels of NK-mediated lysis compared with
those in cells expressing hß2m alone (Fig. 4). In contrast, RMA-S transfectants
expressing HLA-A2, -A3, -B27, and -Cw3 that were incubated with
synthetic peptides specific for each HLA class I protein were lysed by
221707 NK cells at levels comparable to those in RMA-S cells
transfected with hß2m alone. These data are consistent
with the direct binding data reported above and strongly suggest that
CD94/NKG2A interacts specifically with HLA-E and not with a broad range
of HLA class I molecules.
|
Recent studies have shown that protection from NK cell lysis can
be mediated through an interaction between CD94/NKG2A and HLA-E
complexed with peptides corresponding to HLA class I leader sequence
peptides (residues 3–11) that have a Met, but not a Thr, at residue 4
(P2 of the HLA-E binding peptide) (15, 17). Binding data have indicated
that leader sequence peptides with Thr at P2 do not bind to HLA-E as
well as those with a Met at P2 (15, 16, 17, 23, 24). There are two possible
explanations for the failure of leader sequences containing Thr at
residue 4 (P2) to protect HLA-E-expressing target cells from lysis by
CD94/NKG2A+ NK cells. Firstly, CD94/NKG2A may not recognize
HLA-E complexed with leader sequence peptides that contain Thr at P2.
Alternatively, these peptides may fail to confer protection from NK
cell-mediated lysis because of their inability to maintain sufficient
levels of stable HLA-E/peptide complexes for recognition by CD94/NKG2A
over the course of a cytotoxicity assay. To discriminate between these
possibilities, the ability of the soluble CD94/NKG2 receptor to bind to
HLA-E complexed to peptides corresponding to HLA class I leader
sequences that differ only at P2 was tested. After coculture for
16 h in the presence or the absence of the indicated peptides,
RMA-S/HLA-E transfectants were incubated with anti-class I mAb or
soluble CD94/NKG2A/PE and were analyzed by flow cytometry. Incubation
with HLA leader sequence peptides containing either Met or Thr at P2
significantly increased the level of cell surface HLA-E expression with
respect to that in cells incubated without peptide or with a control
peptide, as detected by HLA class I-specific mAb (Fig. 5
A). However, those with Met
at P2 (VMAPRTVLL and VMAPRTLLL) consistently increased HLA-E expression
to levels higher than those with Thr at P2. Despite the fact that
higher levels of bound CD94/NKG2A/PE were consistently observed on
RMA-S/HLA-E transfectants cultured with peptides having Met at P2,
cells expressing HLA-E complexed to peptides with Thr at P2 clearly
bound soluble CD94/NKG2A/PE. No binding of soluble CD94/NKG2A/PE was
observed to RMA-S/HLA-E transfectants cultured in the absence of
peptide or with a control peptide.
The observation that the soluble CD94/NKG2A/PE receptor bound to RMA-S/HLA-E transfectants cocultured with leader sequence peptides with Thr at P2 is superficially in conflict with the finding that peptides with Thr at P2 failed to confer protection from NK-mediated lysis (15, 17). This might suggest that the specificity of the recombinant receptor is subtly different from its natural counterpart. Alternatively, this discrepancy may reflect differences between the conditions of the binding and cytotoxicity assays. Notably, the binding assays using the soluble form of the receptor were performed immediately after removal of the peptide from the RMA-S/HLA-E transfectants, whereas the functional assays typically involve a 3-h incubation at 37°C in the absence of peptide after extensive washing to remove unincorporated chromium and excess peptide. If the HLA-E/peptide complexes formed with peptides with Thr at P2 are not as stable as those with Met at P2, it is possible that they dissociated over the course of the cytotoxicity assay, thus leaving little, if any, properly folded HLA-E on the target cells available to interact with CD94/NKG2A on the effector cells.
Therefore, the stability of HLA-E complexed to peptides containing
either Thr or Met at P2 was tested using conditions similar to those
employed in the cytotoxicity assays. Cell surface expression of HLA-E
was maximized by overnight culture of RMA-S/HLA-E transfectants at
26°C with the indicated peptides. Cells were washed to remove excess
peptide and were returned to 37°C. At the indicated times, aliquots
of cells were sampled, and the amount of cell surface HLA-E was
determined by flow cytometry. As expected, culture at 26°C without
added peptide led to elevated levels of HLA-E expression (as a result
of the increased thermostability), which rapidly returned to background
levels when cultured at 37°C (Fig. 5B). A similar pattern
of HLA-E expression was seen with the control peptide KLFEKVYNY (which
binds HLA-A3). Overnight incubation at 26°C with HLA class I leader
sequence peptides initially resulted in higher levels of HLA-E
expression than that in cells incubated without peptide or with control
peptides. On incubation at 37°C, cell surface expression of
peptide-stabilized HLA-E was evident even after 2 h for cells
incubated with the leader sequence peptide, VMAPRTVLL. In contrast,
RMA-S/HLA-E transfectants incubated with the leader sequence peptide
VTAPRTLLL rapidly lost conformational HLA class I determinants over the
course of the 37°C incubation. Similar data were observed with
another peptide that contains Thr at P2, VTAPRTVLL (data not shown).
These data indicate that the leader sequence peptides with Thr at P2,
such as VTAPRTLLL, form complexes with HLA-E that are significantly
less stable than those formed with peptides containing Met at P2.
Consequently, the failure of peptides with Thr at P2 to protect
HLA-E-positive target cells apparently results from the rapid
dissociation of these complexes over the course of a cytotoxicity
assay. Therefore, it might be expected that the continued presence of
peptide with Thr at P2 throughout the course of the cytotoxicity assays
would allow for the continual formation of HLA-E/peptide complexes,
which would maintain cell surface HLA-E that could be recognized by
CD94/NKG2A, and thus reconcile the difference between the binding and
cytotoxicity data.
To examine this possibility, cytotoxicity assays were performed using
target cells that were cultured either with or without the continued
presence of peptide throughout the assay after cell surface HLA-E was
reconstituted by overnight culture at 26°C with the same peptides. As
expected, RMA-S cells that lacked HLA-E (hß2m alone) were
efficiently lysed regardless of whether peptide was maintained
throughout the assay (Fig. 5C). Overnight culture with the
leader sequence peptide with Met at P2, VMAPRTVLL, was sufficient to
protect RMA-S/HLA-E transfectants from lysis. As previously shown (15),
the peptide VTAPRTLLL was not able to confer protection from lysis
after overnight culture alone. However, if this peptide was maintained
throughout the assay, it was able to protect RMA-S/HLA-E transfectants
from NK-mediated lysis. The peptide VTAPRTVLL, like VTAPRTLLL, is also
capable of protecting target cells from lysis when maintained
throughout the coculture of target cells with NK cells (data not
shown). Furthermore, protection was specific to peptides that have some
capacity to interact with HLA-E, as control peptides KLFEKVYNY (Fig. 5C) and RRISGVDRS (data not shown), which do not bind HLA-E,
failed to induce protection under any conditions. These data indicate
that the CD94/NKG2A receptor does not functionally discriminate between
HLA-E complexed to leader sequence peptides containing either Met or
Thr at P2, but that this residue exerts a significant effect on
recognition by its contribution to the stability of the HLA-E/peptide
complexes.
Recognition of HLA-E by CD94/NKG2 is dependent on the sequence of the peptide associated with HLA-E
To examine the importance of other peptide residues in the
recognition of HLA-E by CD94/NKG2A, we tested a panel of peptides for
their ability to induce conformational class I epitopes on RMA-S/HLA-E
transfectants and for their ability to produce a ligand structure
recognized by CD94/NKG2A. Most human HLA class I alleles encode Val at
residue 3 of the leader sequence (P1 of the HLA-E binding peptide);
however, HLA-Aw34 encodes Ile, and murine class I molecules encode Ala
at this position. The substitution of Ile (HLA-Aw34) for Val at P1 had
little effect on either the ability to induce conformational
determinants on HLA-E as detected by the class I-specific mAb or on the
ability of soluble CD94/NKG2A/PE to bind to HLA-E compared with that of
the signal sequence peptide derived from HLA-A2 (Fig. 6). The peptide AMAPRTLLL derived from
murine H-2Dd and Ld differs from that derived
from HLA-A1 only at P1 (A vs V). This peptide was less effective in
stabilizing HLA-E on the surface of RMA-S/HLA-E cells than the
corresponding peptide derived from the leader sequence of HLA-A1.
However, recombinant CD94/NKG2A/PE was able to bind to HLA-E complexed
with AMAPRTLLL, and the levels of staining were reduced in
concordance with the lower levels of cell surface HLA-E. These data
indicate that Ile can be substituted for Val at P1 with little effect
on cell surface expression of HLA-E or on CD94/NKG2A recognition and
that while an alanine substitution at P1 reduces cell surface
expression of HLA-E, there is no apparent effect on the interaction
between HLA-E and CD94/NKG2A/PE.
|
A). Leader sequence
residue 10 (P8) also tends to be a hydrophobic residue, although there
is more variation than at residue 9. Val (HLA-A2), Ile (HLA Cw3), Leu
(HLA-A1), and Phe (HLA-G) are present in leader sequence peptides at
this position. No significant differences in the induction of HLA-E
expression could be attributed to the presence of Val, Ile, or Leu at
P8, but lower levels of HLA-E expression were observed using the
HLA-G-derived peptide that has Phe at P8 (Fig. 6). Nevertheless, all
these residues at P8 allowed for an interaction between HLA-E and
soluble CD94/NKG2A/PE. Thus, the presence of a hydrophobic residue
appears sufficient to allow for an interaction with soluble
CD94/NKG2A/PE.
Changes in P3 and P6 produced the most noticeable effects. The
leader sequence peptide corresponding to H-2Dk has a Val,
rather than Ala, at P3. This peptide had a reduced ability to stabilize
cell surface HLA-E on RMA-S cells with respect to the analogous peptide
from H-2Dd/Ld. Despite this, soluble
CD94/NKG2A/PE still bound to HLA-E complexed to the H-2Dk
leader sequence peptide, albeit at a reduced level. The leader sequence
of HLA-Cw7 is distinctive because it contains an Ala at P6. The other
leader sequence peptides, including the comparable peptide derived from
HLA-A1, encode Thr at P6. Consistent with previously described peptide
binding data (23), the Ala substitution leads to a slight decrease in
stabilization of cell surface HLA-E, but one that is not as marked as
that observed with peptides having Thr at P2 (see Fig. 5A).
However binding of soluble CD94/NKG2A/PE to HLA-E/Cw7 peptide was
markedly lower than that for other peptides that induced comparable
levels of HLA-E expression, suggesting that the substitution of Ala for
Thr at P6 has a significant effect on the interaction between
CD94/NKG2A and HLA-E.
Finally, a peptide derived from the EBV protein BZLF-1 (residues 39–47) that has recently been shown to bind HLA-E (25) was also tested for its ability to both bind HLA-E and produce the ligand structure recognized by CD94/NKG2A. While the sequence of this peptide is quite different from that of the prototypical HLA class I leader sequence, it clearly stabilized cell surface HLA-E with respect to control peptides. In marked contrast to the HLA-class I leader sequence-derived peptides, no binding of soluble CD94/NKG2A/PE to HLA-E/BZLF peptide complexes was observed. This suggests that the nature of the peptide associated with HLA-E plays a role in its recognition by CD94/NKG2 receptors.
CD94/NKG2A can functionally discriminate between different HLA-E binding peptides
For functional confirmation of the soluble CD94/NKG2A/PE
binding data described above, RMA-S/HLA-E transfectants were cultured
overnight in the presence of the indicated peptides, and then used as
targets with or without the continued presence of the peptide in
cytotoxicity assays with 221707 NK cells. As expected, protection from
lysis was dependent on coculture with an appropriate peptide, as cells
incubated without synthetic peptide were efficiently lysed by 221707
cells (Fig. 7). Incubation of RMA-S/HLA-E
transfectants with the peptide VMAPRTLIL was sufficient to protect
these cells from NK lysis regardless of whether the presence of the
peptide was maintained throughout the assay. The leader sequence
peptides derived from HLA-Cw7 and H-Dk had characteristics
similar to those of peptides with Thr at P2, such as VTAPRTVLL (see
Fig. 5C). These peptides do not protect RMA-S/HLA-E cells
under typical assay conditions, but do so if the presence of the
peptide is maintained throughout the assay. On the other hand, the
BZLF-1 peptide was unable to induce protection of RMA-S/HLA-E
transfectants regardless of whether the presence of the peptide was
maintained throughout the assay. These data are consistent with the
binding data obtained using the soluble receptor and indicate that the
CD94/NKG2A receptor can functionally discriminate between different
HLA-E/peptide complexes.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Previous studies have shown that class I leader sequence-derived peptides containing Met, but not Thr, at P2 complexed with HLA-E can act as a ligand for CD94/NKG2. It has also been shown that coculture of HLA-E-expressing cells with leader sequence peptides containing Thr at P2 increased HLA-E cell surface expression, albeit at reduced levels compared with those of similar peptides with Met at P2 (15, 17). The direct binding and functional data presented here clearly indicate that the CD94/NKG2A receptor can interact with HLA-E complexed to peptides with Thr at P2 and that these peptides can protect target cells from lysis by NK cells. However, these complexes are considerably less stable with respect to HLA-E complexed to peptides with Met at P2. Thus, the predominant factor in controlling CD94/NKG2 recognition of HLA-E with respect to the effects of substitutions at P2 of the bound peptide appears to be the affinity of the peptide for HLA-E, which, in turn, affects stable expression of HLA-E at the cell surface.
Analysis of a variety of other peptides derived from both human and mouse class I leader sequences showed that the naturally occurring variations at P1, P3, P7, and P8 had little obvious direct effect on the interaction between HLA-E and CD94/NKG2A. In general, substitutions that had an obvious effect on the stability of cell surface HLA-E lead to equivalent changes in the binding level of soluble CD94/NKG2A/PE. However, analysis of two peptides, the BZLF-1 peptide (residues 39–47) and the HLA-Cw7 leader sequence peptide, suggests that not all peptides that bind to HLA-E can efficiently form the ligand for CD94/NKG2A. While the BZLF-1 peptide does not bind HLA-E as well as most HLA class I leader sequence peptides tested, it is capable of stabilizing cell surface HLA-E. Despite this, no detectable binding of soluble CD94/NKG2A/PE was observed, nor was the peptide capable of protecting RMA-S/HLA-E transfectants from NK-mediated lysis even when its presence was maintained throughout the cytotoxicity assays.
The peptide VMAPRALLL derived from the HLA-Cw7 leader sequence differs from that of HLA-A1 only at P6 (Ala for Thr). This substitution appears to have a more profound effect on the interaction with CD94/NKG2A than on its ability to bind HLA-E. This peptide increased cell surface expression of HLA-E to levels similar to those of other class I leader sequence peptides containing Met at P2, yet was unable to confer protection from NK lysis to HLA-E-expressing RMA-S cells unless the peptide was maintained in culture during the assay. Moreover, the low level of binding of the soluble CD94/NKG2A/PE to HLA-E complexed with this peptide, relative to that of the other peptides that stabilize comparable levels of HLA-E, suggests that this substitution is affecting the conformation of the determinant recognized by CD94/NKG2A. The crystal structure of HLA-E (27) shows that that P6, a Thr in most HLA class I leader sequence peptides, makes significant contact with the C pocket, while the Arg at P5 is solvent exposed. The substitution of Ala for Thr in the HLA-Cw7 peptide may exert its affect on CD94/NKG2A recognition by altering the conformation of HLA-E around the C pocket or by altering the conformation of the peptide, which, in turn, may change the position of the exposed Arg at P5.
These data indicate that CD94/NKG2A recognition of HLA-E is potentially controlled at two levels. Firstly, there is a requirement for peptides that can bind HLA-E and allow its egress from the endoplasmic reticulum leading to subsequent cell surface expression. Studies with murine cells (28) and HLA class I-deficient human cells (17, 23) indicate that few self peptides are appropriate for this function, as cell surface expression of HLA-E in the absence of other HLA class I molecules is generally very low. Transfection of classical class I genes into cell lines that lack expression of classical class I genes but express HLA-E elevates cell surface expression of HLA-E (16, 17, 23). Thus, the signal sequences of class I molecules appear to be the predominant source of such peptides. Secondly, the sequence of the peptide also affects recognition of HLA-E. While most HLA class I signal sequence peptides produce the ligand for CD94/NKG2A when associated with HLA-E, the Cw7 peptide was poorly recognized by soluble CD94/NKG2A. Furthermore, no recognition of HLA-E complexed with the BZLF peptide was observed. These data suggest that even if self or viral peptides bind HLA-E sufficiently well to allow stable cell surface expression, recognition of HLA-E by CD94/NKG2A is further constrained by the sequence of the peptide.
While the precise role that CD94/NKG2 recognition of HLA-E plays
in NK cell biology is not understood, it is appears that this system
may function to complement the interaction of KIR with HLA class I
molecules. As KIR are both clonally distributed and restricted in their
interaction to particular HLA class I epitopes, the presence of a more
broadly reactive system may be important for NK cells that lack an
appropriate inhibitory KIR for recognition of self-HLA class I
molecules (29). In addition, up-regulation of cell surface HLA-E
expression through IFN-
induction of either HLA class I proteins
themselves or the Ag-processing/peptide-loading mechanisms (30) or
through induction of CD94/NKG2A expression through the action of IL-15
(31) may serve to make the recognition of HLA-E by CD94/NKG2 receptors
acutely responsive to the effects of cytokines produced in inflammatory
situations.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Abbreviations used in this paper: KIR, killer cell inhibitory receptors; hß2m, human ß-microglobulin; HRP, horseradish peroxidase; PE, phycoerythrin.
Received for publication June 23, 1998. Accepted for publication September 3, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  Y. KAWAGUCHI, K. KONO, Y. MIZUKAMI, K. MIMURA, and H. FUJII Mechanisms of Escape from Trastuzumab-mediated ADCC in Esophageal Squamous Cell Carcinoma: Relation to Susceptibility to Perforin-granzyme Anticancer Res, June 1, 2009; 29(6): 2137 - 2146. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Lo Monaco, L. Sibilio, E. Melucci, E. Tremante, M. Suchanek, V. Horejsi, A. Martayan, and P. Giacomini HLA-E: Strong Association with {beta}2-Microglobulin and Surface Expression in the Absence of HLA Class I Signal Sequence-Derived Peptides J. Immunol., October 15, 2008; 181(8): 5442 - 5450. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. J. Petrie, C. S. Clements, J. Lin, L. C. Sullivan, D. Johnson, T. Huyton, A. Heroux, H. L. Hoare, T. Beddoe, H. H. Reid, et al. CD94-NKG2A recognition of human leukocyte antigen (HLA)-E bound to an HLA class I leader sequence J. Exp. Med., March 17, 2008; 205(3): 725 - 735. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Ward, M. Bonaparte, J. Sacks, J. Guterman, M. Fogli, D. Mavilio, and E. Barker HIV modulates the expression of ligands important in triggering natural killer cell cytotoxic responses on infected primary T-cell blasts Blood, August 15, 2007; 110(4): 1207 - 1214. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Masilamani, C. Nguyen, J. Kabat, F. Borrego, and J. E. Coligan CD94/NKG2A Inhibits NK Cell Activation by Disrupting the Actin Network at the Immunological Synapse J. Immunol., September 15, 2006; 177(6): 3590 - 3596. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Derre, M. Corvaisier, B. Charreau, A. Moreau, E. Godefroy, A. Moreau-Aubry, F. Jotereau, and N. Gervois Expression and Release of HLA-E by Melanoma Cells and Melanocytes: Potential Impact on the Response of Cytotoxic Effector Cells. J. Immunol., September 1, 2006; 177(5): 3100 - 3107. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. S. Clements, L. Kjer-Nielsen, L. Kostenko, H. L. Hoare, M. A. Dunstone, E. Moses, K. Freed, A. G. Brooks, J. Rossjohn, and J. McCluskey Crystal structure of HLA-G: A nonclassical MHC class I molecule expressed at the fetal-maternal interface PNAS, March 1, 2005; 102(9): 3360 - 3365. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. I. Bonaparte and E. Barker Killing of human immunodeficiency virus-infected primary T-cell blasts by autologous natural killer cells is dependent on the ability of the virus to alter the expression of major histocompatibility complex class I molecules Blood, October 1, 2004; 104(7): 2087 - 2094. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. M. S. Huang, P. Borszcz, S. Sidobre, M. Kronenberg, and K. P. Kane CD1d1 Displayed on Cell Size Beads Identifies and Enriches an NK Cell Population Negatively Regulated by CD1d1 J. Immunol., May 1, 2004; 172(9): 5304 - 5312. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. A. Bland, M. K. Lemberg, A. J. McMichael, B. Martoglio, and V. M. Braud Requirement of the Proteasome for the Trimming of Signal Peptide-derived Epitopes Presented by the Nonclassical Major Histocompatibility Complex Class I Molecule HLA-E J. Biol. Chem., September 5, 2003; 278(36): 33747 - 33752. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. D. Miller, D. A. Weber, C. Ibegbu, J. Pohl, J. D. Altman, and P. E. Jensen Analysis of HLA-E Peptide-Binding Specificity and Contact Residues in Bound Peptide Required for Recognition by CD94/NKG2 J. Immunol., August 1, 2003; 171(3): 1369 - 1375. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. G. Shilling, K. L. McQueen, N. W. Cheng, J. A. Shizuru, R. S. Negrin, and P. Parham Reconstitution of NK cell receptor repertoire following HLA-matched hematopoietic cell transplantation Blood, May 1, 2003; 101(9): 3730 - 3740. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Gunturi, R. E. Berg, and J. Forman Preferential Survival of CD8 T and NK Cells Expressing High Levels of CD94 J. Immunol., February 15, 2003; 170(4): 1737 - 1745. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. O. Flores-Villanueva, H. Hendel, S. Caillat-Zucman, J. Rappaport, A. Burgos-Tiburcio, S. Bertin-Maghit, J. A. Ruiz-Morales, M. E. Teran, J. Rodriguez-Tafur, and J.-F. Zagury Associations of MHC Ancestral Haplotypes with Resistance/Susceptibility to AIDS Disease Development J. Immunol., February 15, 2003; 170(4): 1925 - 1929. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. O. Long and S. Rajagopalan Stress Signals Activate Natural Killer Cells J. Exp. Med., December 2, 2002; 196(11): 1399 - 1402. [Full Text] [PDF]
|
|
|
|
|
|
J. Michaelsson, C. Teixeira de Matos, A. Achour, L. L. Lanier, K. Karre, and K. Soderstrom A Signal Peptide Derived from hsp60 Binds HLA-E and Interferes with CD94/NKG2A Recognition J. Exp. Med., December 2, 2002; 196(11): 1403 - 1414. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Radaev, M. Kattah, Z. Zou, M. Colonna, and P. D. Sun Making Sense of the Diverse Ligand Recognition by NKG2D J. Immunol., December 1, 2002; 169(11): 6279 - 6285. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. C. Hsu, X.-R. Liu, A. Selvakumar, E. Mickelson, R. J. O'Reilly, and B. Dupont Killer Ig-Like Receptor Haplotype Analysis by Gene Content: Evidence for Genomic Diversity with a Minimum of Six Basic Framework Haplotypes, Each with Multiple Subsets J. Immunol., November 1, 2002; 169(9): 5118 - 5129. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. G. Shilling, N. Young, L. A. Guethlein, N. W. Cheng, C. M. Gardiner, D. Tyan, and P. Parham Genetic Control of Human NK Cell Repertoire J. Immunol., July 1, 2002; 169(1): 239 - 247. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Dulphy, C. Rabian, C. Douay, O. Flinois, S. Laoussadi, J. Kuipers, R. Tamouza, D. Charron, and A. Toubert Functional modulation of expanded CD8+ synovial fluid T cells by NK cell receptor expression in HLA-B27-associated reactive arthritis Int. Immunol., May 1, 2002; 14(5): 471 - 479. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Sharland, A. Patel, J. H. Lee, A. E. Cestra, S. Saidman, and G. L. Waneck Genetically Modified HLA Class I Molecules Able to Inhibit Human NK Cells Without Provoking Alloreactive CD8+ CTLs J. Immunol., April 1, 2002; 168(7): 3266 - 3274. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Barakonyi, K. T. Kovacs, E. Miko, L. Szereday, P. Varga, and J. Szekeres-Bartho Recognition of Nonclassical HLA Class I Antigens by {gamma}{delta} T Cells During Pregnancy J. Immunol., March 15, 2002; 168(6): 2683 - 2688. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. P. Shum, L. R. Flodin, D. G. Muir, R. Rajalingam, S. I. Khakoo, S. Cleland, L. A. Guethlein, M. Uhrberg, and P. Parham Conservation and Variation in Human and Common Chimpanzee CD94 and NKG2 Genes J. Immunol., January 1, 2002; 168(1): 240 - 252. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Matsumoto, K. Tajima, M. Mitsuki, and K. Yamamoto H-2 allele specificity of the NK cell C-type lectin-like MHC class I receptor Ly49A visualized by soluble Ly49A tetramer Int. Immunol., May 1, 2001; 13(5): 615 - 623. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. M. Gardiner, L. A. Guethlein, H. G. Shilling, M. Pando, W. H. Carr, R. Rajalingam, C. Vilches, and P. Parham Different NK Cell Surface Phenotypes Defined by the DX9 Antibody Are Due to KIR3DL1 Gene Polymorphism J. Immunol., March 1, 2001; 166(5): 2992 - 3001. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Gays, K. P. Fraser, J. A. Toomey, A. G. Diamond, M. M. Millrain, P. J. Dyson, and C. G. Brooks Functional Analysis of the Molecular Factors Controlling Qa1-Mediated Protection of Target Cells from NK Lysis J. Immunol., February 1, 2001; 166(3): 1601 - 1610. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. H. Chung, K. Natarajan, L. F. Boyd, J. Tormo, R. A. Mariuzza, W. M. Yokoyama, and D. H. Margulies Mapping the Ligand of the NK Inhibitory Receptor Ly49A on Living Cells J. Immunol., December 15, 2000; 165(12): 6922 - 6932. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Watzl, C. C. Stebbins, and E. O. Long Cutting Edge: NK Cell Inhibitory Receptors Prevent Tyrosine Phosphorylation of the Activation Receptor 2B4 (CD244) J. Immunol., October 1, 2000; 165(7): 3545 - 3548. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. R. Kraft, R. E. Vance, J. Pohl, A. M. Martin, D. H. Raulet, and P. E. Jensen Analysis of Qa-1bPeptide Binding Specificity and the Capacity of Cd94/Nkg2a to Discriminate between Qa-1-Peptide Complexes J. Exp. Med., September 5, 2000; 192(5): 613 - 624. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Ulbrecht, S. Martinozzi, M. Grzeschik, H. Hengel, J. W. Ellwart, M. Pla, and E. H. Weiss Cutting Edge: The Human Cytomegalovirus UL40 Gene Product Contains a Ligand for HLA-E and Prevents NK Cell-Mediated Lysis J. Immunol., May 15, 2000; 164(10): 5019 - 5022. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. H. Idris, H. R. C. Smith, L. H. Mason, J. R. Ortaldo, A. A. Scalzo, and W. M. Yokoyama The natural killer gene complex genetic locus Chok encodes Ly-49D, a target recognition receptor that activates natural killing PNAS, May 25, 1999; 96(11): 6330 - 6335. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. L. Lanier Natural killer cells fertile with receptors for HLA-G? PNAS, May 11, 1999; 96(10): 5343 - 5345. [Full Text] [PDF]
|
|
|
|
|
|
J. Stevens, R. C. Jones, R. S. Bordoli, J. Trowsdale, S. J. Gaskell, G. W. Butcher, and E. Joly Peptide Specificity of RT1-A1c, an Inhibitory Rat Major Histocompatibility Complex Class I Natural Killer Cell Ligand J. Biol. Chem., September 15, 2000; 275(38): 29217 - 29224. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. R. Snyder, J. Wang, J. F. Waring, and G. D. Ginder Identification of CCAAT Displacement Protein (CDP/cut) as a Locus-specific Repressor of Major Histocompatibility Complex Gene Expression in Human Tumor Cells J. Biol. Chem., February 9, 2001; 276(7): 5323 - 5330. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
