Membrane-Anchored 2-Microglobulin Stabilizes a Highly Receptive State of MHC Class I Molecules
http://www.100md.com
免疫学杂志 2005年第4期
Abstract
The magnitude of response elicited by CTL-inducing vaccines correlates with the density of MHC class I (MHC-I)-peptide complexes formed on the APC membrane. The MHC-I L chain, 2-microglobulin (2m), governs complex stability. We reasoned that genetically converting 2m into an integral membrane protein should exert a marked stabilizing effect on the resulting MHC-I molecules and enhance vaccine efficacy. In the present study, we show that expression of membranal human 2m (h2m) in mouse RMA-S cells elevates MHC-I thermal stability. RMA-S transfectants bind an exogenous peptide at concentrations 104- to 106-fold lower than parental RMA-S, as detected by complex-specific Abs and by T cell activation. Moreover, saturation of the transfectants’ MHC-I by exogenous peptide occurs within 1 min, as compared with 1 h required for parental cells. At saturation, however, level of peptide bound by modified cells is only 3- to 5-fold higher. Expression of native h2m only results in marginal effect on the binding profile. Soluble 2m has no effect on the accelerated kinetics, but the kinetics of transfectants parallel that of parental cells in the presence of Abs to h2m. Ab inhibition and coimmunoprecipitation analyses suggest that both prolonged persistence of peptide-receptive H chain/2m heterodimers and fast heterodimer formation via lateral diffusion may contribute to stabilization. In vivo, peptide-loaded transfectants are considerably superior to parental cells in suppressing tumor growth. Our findings support the role of an allosteric mechanism in determining ternary MHC-I complex stability and propose membranal 2m as a novel scaffold for CTL induction.
Introduction
To enable meticulous immune surveillance by armed effector CTL, MHC class I (MHC-I)3 molecules must efficiently bind and present peptides derived from intracellular proteins while minimizing acquisition of undesired peptides from the extracellular milieu. The MHC-I polymorphic H chain () is a transmembrane glycoprotein harboring three extracellular domains. The membrane distal 1 and 2 domains form the peptide-binding groove, typically capable of accommodating peptides of 8–10 amino acids. The H chain is noncovalently associated, mostly through 3, with the nonpolymorphic 2-microglobulin (2m) L chain, which is a single Ig-like domain, not anchored to the plasma membrane.
The 2m L chain plays an essential role both in promoting endogenous peptide binding at the endoplasmic reticulum (ER) and in diminishing peptide exchange at the cell surface. To allow binding of peptides transported to the ER from the cytosol by TAP, 2m must first associate with the H chain to induce a peptide-receptive conformation (1, 2). This heterodimer is unstable and is assembled and subsequently maintained at the peptide loading complex (PLC) in an open state with the concerted guidance of an array of ER chaperones, including calnexin, ERP57, calreticulin, and tapasin. Upon loading with an adequate peptide, the MHC-I molecule assumes a highly stable, closed conformation and is consequently released from the PLC to exit the ER toward the cell surface (for review, see Refs.3 and 4).
At the cell surface, the ternary MHC-I complex is apparently devoid of auxiliary proteins. Its stability appears to be controlled by an allosteric mechanism, as outlined in detail in a sequence of reports (5, 6, 7), and supported by numerous studies, which examined soluble as well as cell-surface MHC-I molecules. According to this model, dissociation of the peptide considerably decreases the affinity of the H chain for 2m, reducing the heterodimer lifespan to only several minutes or less. Free H chains, in turn, bind the peptide several orders of magnitude weaker than the 2m-coupled isoform, rendering functional rebinding negligible. Reciprocally, predissociation of 2m, rather than a peptide from the ternary complex, results in the same affinity decline and fast peptide detachment (8, 9). At the cell surface, the H chain monomer often denatures, shows strong propensity to oligomerize (10), and is later internalized (11). Hence, at physiological conditions in which extracellular concentrations of both peptide and 2m are limiting, binding an exogenous peptide is an unlikely, although not an improbable, event. In accord with this model are studies (e.g., Refs.12, 13, 14, 15, 16, 17) that showed that loading cells with an extracellular peptide is greatly facilitated in the presence of high concentration of exogenous 2m.
To evoke CTL, vaccines must target immunogenic peptides to MHC-I molecules on dendritic cells. This can be accomplished extracellularly, either by direct loading or through cross presentation of peptides derived from internalized immunogens, as well as via an endogenous route, which usually entails the use of a genetic approach. The appreciation that 2m is critical both for ternary complex assembly and for controlling its stability has prompted its incorporation into various vaccine designs. These include recombinant single-chain 2m/H chain MHC-I dimers (18, 19, 20, 21, 22), soluble (23, 24, 25) or cellular (23, 26) peptide/2m fusions, and cell-expressed single-chain peptide/2m/H chain trimers (27, 28).
We recently generated membrane-anchored chimeric 2m/peptide molecules fused with the intracellular activation domain of CD3 -chain (CD3). These polypeptides functionally associate with endogenous MHC-I H chains in transfected T cells and yield an exceedingly high complex density on the cell surface (29). We reasoned that attachment of 2m to the cell membrane offers a unique, universal tool for the generation and stabilization of immunogenic MHC-I/peptide complexes and may serve as a novel vaccine platform. As a preparatory step in the design and evaluation of CTL-inducing vaccines based on membrane-anchored 2m, we investigated biochemical and functional outcome of its expression in RMA-S cells.
Materials and Methods
Vectors and expression plasmids
The use of the pBJ1-Neo expression vector and the construction of plasmid 21-2, encoding chimeric human 2m (h2m)/CD3, have been described elsewhere (29).
In plasmid 323-5, the CD3 intracellular domain was replaced with that of H-2Kb to encode h2m/Kb as follows: first-strand DNA synthesis from mRNA prepared from RMA cells was performed with the reverse primer 5'-CGCGCGGCCGCAAGTCCACTCCAGGCAGC-3', and PCR was then conducted by adding the sense primer 5'-CCCTCGAGCTCCACTGTCTCCAACATGGCG-3'. The product, encoding the 3'-part of the peptide bridge and H-2Kb transmembrane and cytoplasmic portion, was cloned into pBJ1-Neo as an XhoI/NotI fragment, together with the XbaI/XhoI fragment from plasmid 21-2, encoding h2m leader, mature protein, and 5'-part of the peptide bridge.
Expression plasmid 845-6(n2m) encodes full-length, native h2m (n2m), which was amplified from Jurkat cells mRNA by reverse transcription-PCR with the sense primer 5'-GGGTCTAGAGCCGAGATGTCTCGCTCCGTG-3' and the reverse primer 5'-CGCGCGGCCGCTTACATGTCTCGATCCCACTTAAC-3' and inserted into pBJ1-Neo as a XbaI/NotI fragment.
Mice and cell lines
Eight to 12-wk-old C57BL/6 mice were purchased from The Jackson Laboratory and bred at the Weizmann Institute of Science facilities. Animals were maintained and treated according to the Weizmann Institute of Science animal facility and National Institutes of Health guidelines.
RMA is a Rauscher virus-transformed lymphoma cell line of C57BL/6 (H-2b) origin, and RMA-S is a RMA TAP-deficient mutant (30). Cells were grown in RPMI 1640 medium, supplemented with 10% heat-inactivated FCS, 2 mM L-glutamine, and 50 μm 2-ME. B3Z (31), an OVA257–264-specific, Kb-restricted CTL hybridoma, which expresses the NFAT-LacZ reporter gene, was a kind gift from Dr. N. Shastri (University of California, Berkeley, CA). MO5 is an OVA gene-transfected B16 murine melanoma clone of C57BL/6 origin. These cells were maintained in DMEM supplemented with 10% heat-inactivated FCS, 2 mM L-glutamine, 1% sodium pyruvate, 1% nonessential amino acids (Sigma-Aldrich), combined antibiotics, and 500 μg/ml G418 (both from Invitrogen Life Technologies).
Abs, proteins, and peptides
25-D1.16, a mAb specific to Kb-OVA257–264 (32), was a kind gift from Dr. R. Germain (National Institutes of Health, Bethesda, MD). mAb against h2m (clone BM-63) was from Sigma-Aldrich. Polyclonal rabbit anti-h2m Ab was from DakoCytomation. mAbs 20-8-4 and Y3, specific to H-2Kb, and 28-14-8, specific to H-2Db, were purified from hybridoma supernatants. Purified h2m was from Sigma-Aldrich. OVA257–264 was synthesized by Dr. M. Fridkin (Weizmann Institute of Science, Rehovot, Israel).
DNA transfection
A total of 0.8 ml of 4 x 106 RMA-S cells/ml was mixed in a 4-mm sterile electroporation cuvette (ECU-104; EquiBio) with 20 μg of SalI-linearized plasmid DNA. Transfection was performed with an Easyject Plus electroporation unit (EquiBio) at 350 V, 750 μF. Cells were resuspended in fresh medium and cultured for 24–48 h in 96-well plates before addition of G418 to 1 mg/ml. Resistant clones were expanded in 24-well plates and screened by flow cytometry for expression of h2m.
Flow cytometry
A total of 106 cells was washed with FACS buffer (PBS, 5% FCS, and 0.05% sodium azide) and incubated for 30 min on ice with 100 μl of first (or control) Ab at 10 μg/ml. Cells were then washed and incubated on ice with 100 μl of 1/100 dilution of goat anti-mouse IgG (Fab-specific)-FITC-conjugated polyclonal Abs (Sigma-Aldrich) for 30 min, washed, resuspended in PBS, and analyzed with FACSCalibur (BD Biosciences). Mean fluorescence intensity (MFI) was calculated using CellQuest software (BD Biosciences). Quantitative analysis of cell surface Ags was performed with QIFIKIT (DakoCytomation) according to the manufacturer’s instructions.
Peptide loading
Twenty-four hours before onset of the experiment, cells were washed three times with PBS and incubated in OptiMEM serum-free medium (Invitrogen Life Technologies) at 37°C. Cells were then transferred to fresh OptiMEM medium and coincubated with a peptide in 24-well plates at 1 x 106 cells/ml.
B3Z activation
Peptide-pulsed or control cells were washed three times with PBS, resuspended in nonselective medium at 5 x 105 cells/ml, and 50 μl of the cells were added to microtiter plates in triplicates. PBS-washed B3Z cells were resuspended in fresh growth medium at 5 x 105 cells/ml, added to wells at a 1:1 ratio, and coincubated for 6 h at 37°C. Cells were washed twice with PBS and fixed with 0.25% glutaraldehyde 5 min at 4°C, washed three times with PBS, incubated overnight with 100 μl of 5-bromo-4-chloro-3-indolyl -D-galactoside solution (0.2% X-Gal, 2 mM MgCl2, 5 mM K4Fe(CN)6 · 3H2O, and 5 mM K3Fe(CN)6 in PBS), and scored under the microscope for blue staining.
Immunoprecipitation
A total of 7 x 107 cells were harvested and washed twice with PBS. The cells were incubated with 15 μg of Ab for 2 h at 37°C. Cells were then washed three times with PBS, resuspended in cold lysis buffer (50 mM Tris-HCl (pH 8.0), 150 mM NaCl, 1% Nonidet P-40, 1 mM PMSF, 1 μg/ml aprotinin, and 1 μg/ml leupeptin), and incubated for 25 min on ice with gentle agitation. Cell lysate was collected by centrifugation (14,000 rpm, 5 min, 4°C). Thirty microliters of protein G-Sepharose 4 Fast Flow beads (Amersham Biosciences) were washed twice in cold lysis buffer and incubated with the cell lysate for 2 h at 4°C with gentle agitation. Nonbound proteins were removed by five washing cycles with lysis buffer (10,000 rpm, 1 min, 4°C). The precipitate was then eluted from the beads using elution buffer (0.1 M glycine (pH 2.7)) and immediately neutralized with 2 M Tris (pH 9). Recovered proteins were kept at –20°C.
Immunoblot analysis
Protein samples were boiled for 3 min, separated on a 10% nonreducing SDS polyacrylamide gel at 50 mA, and transferred onto a nitrocellulose membrane. The membrane was blocked with milk buffer (0.3 g of Na2HPO4, 2.19 g of NaCl, 25 ml of double-distilled water, and 225 ml of 1% low-fat milk) overnight at 4°C and 1 h at room temperature, washed twice with PBS, and incubated for 2 h with the primary Ab. The membrane was then washed 6 times for 6 min each with PBS, incubated for 1 h with the secondary peroxidase-conjugated Ab, washed 6 times for 6 min each with TPBS (PBS with 0.05% Tween 20), 6 times for 6 min each with PBS, and then developed using chemiluminescence kit (Pierce) and x-ray film 100NIF (Fuji).
Tumor immunotherapy
Ten mice in each experimental group were inoculated s.c. in the upper back with 105 MO5 cells/mouse. Local tumor diameter was measured with calipers. Starting 8 days later, when the tumor reached 3–4 mm in diameter, mice were immunized i.p. four times at 7-day intervals with 2 x 106 irradiated peptide-loaded cells, which were prepared as follows: RMA-S cells and transfectants were washed, resuspended at 5 x 106/ml in OptiMEM, and incubated with 50 μM synthetic peptide for 2 h at 26°C and then for 4 h at 37°C. The samples were irradiated (5000 rad) and washed. Mice were monitored daily and sacrificed when moribund. Survival was defined as the day when mice were sacrificed.
Results
Construction of h2m harboring two different membranal anchors
The original design of membrane-anchored h2m has already been delineated (29). Briefly, the carboxyl-terminal methionine residue of h2m was genetically linked to a peptide bridge, comprising the 13 membrane-proximal amino acids of the extracellular portion of HLA-A2, LRWEPSSNPTIPI (single-letter code), which encompasses the proline-rich connecting peptide. This sequence was tethered to the amino terminus of the transmembrane segment of the mouse CD3 to include the entire intracellular domain (h2m/CD3tc). At the TCR complex, CD3 primarily forms disulfide-bridged homodimers through the transmembranal cysteine residue. Indeed, although associated with endogenous MHC-I H chains in transfected T cells, these chimeric h2m/CD3tc polypeptides homodimerize (D. Berko, G. Cafri, and A. Margalit, unpublished observations). To rule out possible contribution of dimerization artifacts to the activity of h2m/CD3tc, we have assembled a similar construct, substituting the CD3 transmembrane and cytoplasmic portion with that of H-2Kb (h2m/Kbtc).
Expression of membranal 2m in RMA-S increases MHC-I thermal stability
RMA-S cells do not express functional TAP, and their MHC-I assembly pathway is thus cut from its major peptide supply. As a result, MHC-I molecules are mostly loaded with scarce, suboptimal peptides generated at the ER, and their surface level at 37°C is substantially reduced compared with RMA cells. However, this level can be elevated by incubating cells with high-affinity peptides or at lower temperatures (26–28°C), in which these peptide-receptive molecules are stabilized. We were interested in testing whether the mere association with membranal 2m could stabilize these thermally labile MHC-I molecules. Three RMA-S transfectants were generated: KD-21-4(tc) and KD-21-6(tc), expressing a relatively moderate and high level of h2m/CD3, respectively, and D-323-4(Kbtc), expressing the h2m/Kbtc construct. Fig. 1 shows that all three clones, but not the parental cells, stain brightly for h2m. Surface level of H-2Db at 37°C, which drops in RMA-S 1 log compared with 27°C or RMA, is almost completely restored in KD-21-6(tc), and to somewhat lesser, still noticeable, extent in KD-21-4(tc) and D-323-4(Kbtc). Interestingly, clones KD-21-6(tc) and D-323-4(Kbtc) express more surface h2m at 27°C than at 37°C unlike clone KD-21-4(tc), which displays comparable levels at both temperatures.
FIGURE 1. FACS analysis of transfectants. RMA, RMA-S, KD-21-4(tc), KD-21-6(tc), and D-323-4(Kbtc) cells were grown in serum-free medium for 24 h at 27 and 37°C. Cells were then incubated with anti-H-2Db (28-14-8) and anti-h2m (BM-63) mAbs, or no Ab as negative control, and then with FITC-conjugated goat anti-mouse Fab Abs and analyzed by FACS. Fluorescence intensity is presented in logarithmic scale.
We went on to obtain a quantitative evaluation of the dual effect of h2m expression and exogenous peptide binding on the level of MHC-I molecules in RMA-S cells. For this purpose, we used a commercial kit for quantitative analysis of cell-surface Ags with mouse Abs, and we measured binding of the synthetic chicken OVA peptide OVA257–264 to H-2Kb, using the Kb-OVA257–264 complex-specific 25-D1.16 mAb. This analysis is summarized in Table I. Expression of h2m elevates the level of H-2Kb 5- to 8.5-fold, whereas addition of the peptide results in an additional, relatively moderate increase of 1.3- to 1.5-fold of H-2Kb, with no effect on H-2Db. In comparison, incubation of RMA-S with a peptide under these conditions raised the surface level of H-2Kb 4-fold. The somewhat weaker effect of h2m expression on the level of H-2Db (1.9- to 5.7-fold increase) is in agreement with the lower ability of h2m to stabilize cell surface H-2Db compared with Kb (33).
Table I. Absolute numbers of cell surface MHC-I molecules expressed on RMA-S cells and transfectants, with or without OVA257–264 (Pep)
Membranal 2m greatly augments the ability of RMA-S to bind an exogenous peptide
We then examined whether these thermally stable MHC-I molecules possess enhanced ability to bind synthetic OVA257–264. Fig. 2A depicts a typical binding experiment. Threshold for detection of OVA257–264 binding to RMA-S was at a concentration of 1 ng/ml, which is in good agreement with a previous report (32). However, significant peptide binding to KD-21-6(tc) was evident even at 1 pg/ml, and residual binding at this concentration could also be detected in D-323-4(Kbtc). Such results could in fact be attributed both to the overall increase in the level of 2m expressed by these cells and to the higher affinity of h2m than mouse 2m for H-2Kb (33, 34). To test this possibility, we transfected RMA-S cells with n2m. Unlike membranal 2m, which is expected to be present on the cell surface also as a noncoupled monomer and reach very high density, level of n2m entirely depends on available MHC-I H chains. To carry out a reliable comparison of peptide loading, we screened a large number of transfectants for the highest expresser and chose clone D-845-6(n2m). Preliminary analysis of thermal stability of H-2Kb and Db on the surface of these cells showed a phenotype resembling that of the parental RMA-S cells, rather than KD-21-6(tc) or D-323-4(Kbtc) (data not shown). Fig. 2B indeed reveals a slight increase in the ability of D-845-6(n2m) to bind OVA257–264 and in its saturation level compared with RMA-S cells, but it is still significantly lower than for KD-21-6(tc).
FIGURE 2. Analysis of peptide binding by transfectants. A, Comparison of KD-21-6(tc) and D-323-4(Kbtc) with parental RMA-S cells. B, Comparative analysis of KD-21-6(tc) against 845-6(n2m) and RMA-S. For both experiments, indicated cells were grown at 37°C for 24 h in serum-free medium and were then incubated for 2 h with serial dilutions of synthetic OVA257–264. Cells were stained with the anti-H-2Kb-OVA257–264 mAb 25-D1.16, and FACS analysis was performed with FITC-conjugated anti-mouse Fab Abs. MFI was calculated using the CellQuest FACS program.
Such a dramatic increase in the ability to bind an exogenous peptide is of particular relevance to the design of cell vaccines expressing membrane-anchored 2m. Therefore, we evaluated the sensitivity with which a peptide-specific T cell hybridoma can respond to transfectants vs parental cells following peptide loading. Fig. 3 presents the peptide dose response of the Kb-restricted, OVA257–264-specific B3Z hybridoma (31). In good agreement with the FACS analysis, B3Z cells could be activated by KD-21-6(tc) and D-323-4(Kbtc) cells pulsed with as little as 1 and 100 fg/ml of the peptide, respectively, whereas RMA-S cells charged with as high as 1 ng/ml failed to activate these cells detectably under the same experimental conditions.
FIGURE 3. Dose-dependent response of the B3Z T cell hybridoma to transfectants KD-21-6(tc) and D-323-4(Kbtc) and parental RMA-S cells loaded with OVA257–264. Indicated APC were grown for 24 h in serum-free medium and incubated for 2 h with serial dilutions of the peptide. Peptide-loaded cells were then incubated overnight in triplicates at a 1:1 ratio (4 x 105 cells each) with B3Z cells, a CTL hybridoma specific to the OVA257–264-H-2Kb complex, which expresses the NFAT-LacZ reporter gene. T cell activation was monitored by intracellular 5-bromo-4-chloro-3-indolyl -D-galactoside staining. Percentage of cells stained blue was evaluated under a light microscope and scored as an average of 16 fields. Fifty-four percent of B3Z cells were activated with the anti-TCR Ab 2C11 (anti-CD3, data not shown). As activation by transfectants was conducted at 1:1 ratio, a score of 27% activated B3Z cells was considered 100% of potentially activated cells, and all results were normalized accordingly.
Membranal 2m increases on-rate of peptide binding
We next asked whether this marked effect on peptide binding is also manifested in a significant change in binding kinetics. To address this question, we designed a flow cytometry-based binding assay, which scores binding of the peptide at a saturating concentration following incubation for different time intervals. Results are shown in Fig. 4A and indicate that saturation of transfectant KD-21-6(tc) and near saturation of D-323–4(Kbtc) are already achieved within 1 min, whereas D-845-6(n2m) and RMA-S require 1 and 2 h, respectively. To assess peptide dissociation rates, we incubated cells with a saturating amount of peptide and similarly followed persistence of specific complexes on the cell surface. Fig. 4B shows no significant difference in dissociation rate between KD-21-6(tc), D-323-4(Kbtc), and RMA-S.
FIGURE 4. Peptide-binding kinetics of transfectants KD-21-6(tc), D-323-4(Kbtc), and 845-6(n2m). A, Cells were grown at 37°C for 24 h in serum-free medium and were then incubated at 37°C for different time intervals with 2 μg/ml OVA257–264. After incubation, cells were washed several times and stained with the complex-specific mAb 25-D1.16. FACS analysis was performed with anti-mouse FAB-FITC, and MFI was derived. One hundred percent saturation was determined as the MFI value obtained after 150 min of incubation with the peptide. B, Cells were incubated for 2 h with 2 μg/ml OVA257–264 in the presence of serum, washed, and incubated at 37°C in complete medium for 24 h. After each time, interval cells were washed and fixed with 0.5% paraformaldehyde. All samples were stained and analyzed in parallel as in A.
Exogenous 2m enhances peptide binding by RMA-S but not transfectants
We considered the possibility that the chimeric 2m polypeptide completely saturates MHC-I H chains on the surface of transfected cells. In this case, addition of exogenous 2m would only have a negligible effect on the kinetics of peptide binding to these clones but is expected to substantially increase on-rate of binding to the parental cells. To test this assumption, we used the experimental design of the peptide-binding assay but now examined the effect of preincubation of cells in the presence of a high concentration of soluble h2m. Indeed, Fig. 5 shows no effect on KD-21-6(tc) but demonstrates a striking increase in the on-rate of binding to RMA-S, which now appears similar to that of the transfectant, so that saturation occurs within the first minute of incubation with the peptide.
FIGURE 5. Effect of exogenous h2m on peptide-binding kinetics. KD-21-6(tc) and parental RMA-S cells were grown at 37°C for 24 h in serum-free medium and were then incubated in the presence and in the absence of 10 μg/ml purified n2m for 1 h at 37°C. Cells were then incubated for different time intervals with 2 μg/ml OVA257–264 at 37°C. After incubation, cells were washed three times and were then stained with the complex-specific 25-D1.16 mAb and analyzed by FACS.
Abs to h2m inhibit peptide binding to cells expressing membranal 2m
Results obtained so far imply that availability of 2m on the cell surface is a key factor, which governs the unique peptide-binding properties of the transfected cells. To assess the contribution of membranal 2m directly, we used polyclonal Abs to h2m. When applied along with an exogenous peptide for even a short period, these Abs should significantly diminish de novo formation of heterodimers but would have a smaller effect on pre-existing ones. In contrast, longer preincubation with the Abs is expected to result in a more pronounced effect also on pre-existing heterodimers. Fig. 6A presents flow cytometry analysis of the amount of peptide bound to KD-21-6(tc) cells following 1 min of incubation with a saturating amount of peptide in the presence of different dilutions of the Abs: in one set, the Ab was added to the assay along with the peptide, and in the other set, the Abs were preincubated for 1 h. These results suggest that the faster kinetics can largely be attributed to pre-existing heterodimers because only a minimal effect is evident following short incubation. Longer preincubation with Ab, nevertheless, severely diminishes binding, implying reformation of heterodimers is important. We then monitored ternary complex formation, following 1 h of preincubation with an effective concentration of these inhibiting Abs. Fig. 6B shows that the increase in binding on-rate is completely abolished under these conditions, and the peptide-binding profile is practically identical with that of RMA-S cells, which, as expected, are unresponsive to the presence of these Abs.
FIGURE 6. Effect of anti-h2m Abs on peptide-binding kinetics. A, KD-21-6(tc) cells were grown at 37°C for 24 h in serum-free medium and then incubated with serial dilutions of polyclonal rabbit anti-h2m Ab (DakoCytomation), either for 1 h at 37°C or for 1 min at room temperature. Cells were then incubated with 2 μg/ml OVA257–264 for 1 min, washed, stained with 25-D1.16, and analyzed by FACS. B, KD-21-6(tc) and parental RMA-S cells were similarly grown and incubated in the presence or in the absence of 1/62 dilution of the anti-h2m Ab for 1 h at 37°C. Cells were then incubated for different time intervals with 2 μg/ml OVA257–264 at 37°C, washed, and similarly analyzed by FACS.
Peptide binding primarily recruits pre-existing membranal 2m molecules
To obtain direct evidence for the involvement of membranal 2m in peptide binding, we performed a coimmunoprecipitation analysis of transfectant D-323-4(Kbtc), using the anti-H-2Kb Ab Y3. This experiment compared the amount of coprecipitated h2m before and after peptide loading, relative to the increase in total surface H-2Kb during the incubation period, as monitored by FACS. Results presented in Fig. 7A reveal a vast increase in the amount of h2m coprecipitated with H-2Kb H chain from the cell membrane following peptide binding, which is accompanied by only a 44% increase of total H-2Kb (Fig. 7B).
FIGURE 7. Coimmunoprecipitation of h2m with surface H-2Kb molecules from D-323-4(Kbtc) cells. Cells were grown 24 h in serum-free medium and incubated for 2 h with (+) or without (–) 2 μg/ml OVA257–264. A, Y3 Ab was incubated with cells for 2 h, cells were washed and lysed with 1% Nonidet P-40, and immune complexes were immunoprecipitated with protein G-Sepharose beads. Immunoblots of eluted samples were analyzed by ECL using HRP-conjugated rabbit anti-h2m polyclonal Abs. Results of three exposures times (minutes) are shown. Under these experimental conditions, capturing of intracellular H-2Kb molecules is minimized but cannot be entirely ruled out. B, FACS analysis of the same cells in the same experiment, performed with the indicated Abs. Results are shown as MFI.
Transfectants are superior to parental cells in treating tumor-bearing mice
To demonstrate the potential of membrane-anchored 2m as a new platform for CTL induction in vivo, we performed a tumor suppression experiment. MO5 is a transfectant of the B16 melanoma cell line (H-2b), which expresses chicken OVA and presents OVA257–264 in the context of H-2Kb. C57BL/6 mice were challenged with 105 MO5 cells each. Starting 8 days later, mice were subjected to an immunization regimen with either irradiated D-323-4(Kbtc) or parental RMA-S cells, both pulsed with OVA257–264, or with no cells as control. Tumor growth was substantially delayed in mice vaccinated with D-323-4 compared with RMA-S (Fig. 8A), and 80% of mice in the D-323-4(Kbtc) vaccinated group remained alive (five tumor-free), compared with only 40% (three tumor-free) in the group immunized with RMA-S 6 wk after the tumor challenge (Fig. 8B).
FIGURE 8. Inhibition of tumor growth by peptide-pulsed D-323-4(Kbtc) and RMA-S cells. Ten C57BL/6 mice in each experimental group were challenged with MO5 tumor cells. Eight days later, mice were vaccinated four times at 7-day intervals with D-323-4(Kbtc) or RMA-S cells pulsed with a saturating concentration of OVA257–264 or with no cells, as described in Materials and Methods. A, Tumor progression during 6 wk after tumor challenge expressed as mean tumor diameter in each of the three groups. Arrows indicate immunizations. B, Survival curves of mice in the three groups during the same period. Mice were monitored daily and sacrificed when moribund, which occurred when tumor reached a diameter of 20 mm.
Discussion
In this study, we expressed membrane-anchored h2m in mouse RMA-S cells and showed up to 106-fold increase in the ability of transfected cells to bind an exogenous peptide and at least 50-fold faster binding kinetics. The affinity of 2m for the MHC-I H chain determines heterodimer stability and, as a result, peptide-binding capacity (35, 36). Indeed, h2m possesses a higher affinity for most mouse MHC-I H chains than does mouse 2m, including H-2Kb and H-2Db, which are expressed by RMA-S (33, 34). Hence, these observations could be solely attributed to the elevated level of 2m expressed by transfected cells and to the higher affinity of the human L chain for MHC-I H chains. However, several findings suggest a major contribution of yet another factor. First, the combined effect of affinity with expression level, as revealed by the quantitative FACS analysis (Table I), mounts, at most, to only 8.5-fold increase of cell surface H-2Kb in transfectants, a value that is too small to account for the magnitude of the observed phenomena. Second, clone D-845-6(n2m), which expresses n2m, exhibits only a mild shift in peptide-binding profile (Figs. 2 and 4) compared with transfectants expressing the membranal derivative. Third, at peptide saturation, the total level of H-2Kb-OVA257–264 complexes formed on the surface of transfectants is only 3- to 5-fold higher than on parental RMA-S cells (Figs. 2 and 4–6).
We repeatedly observed higher level of membrane-anchored h2m at 27°C than at 37°C (Fig. 1). This difference may reflect a remaining degree of thermal instability of MHC-I molecules expressed by transfectants, and suggests that exit of membranal h2m to the cell surface from the ER still depends on available peptides.
A large fraction of MHC-I molecules on RMA-S cells cultured at 37°C are either H chain monomers or short-lived heterodimers (9, 37) so that efficient peptide binding requires equilibrium with exogenous 2m. Of particular importance is the observation that D-845-6(n2m) cells reach saturation only after 1 h of incubation with the peptide, whereas RMA-S cells are fully saturated within 1 min in the presence of soluble h2m. This result indicates that the contribution of exogenous 2m to peptide-binding kinetics far exceeds the effect exerted by cellular expression of n2m. Together with the similar binding kinetics monitored for RMA-S in the presence of 2m and for KD-21-6(tc), our data imply that it is the membranal anchorage of 2m, which is primarily responsible for the pronounced effect on transfected cells. The ability of transfectants to functionally bind an exogenous peptide at concentrations up to 106-fold lower than RMA-S cells supports this conclusion. It coincides with the notion that peptide binding to 2m/H chain heterodimers occurs at far lower concentrations than to free H chains (38, 39), which is probably the prevalent species in RMA-S.
Several nonmutually exclusive explanations may account for this new phenotype. First, the appended polypeptide sequences form additional contacts with the H chain in a manner, which secures a receptive conformation of the peptide-binding groove. One of the conjectures of the allosteric model (6, 7) is that the empty heterodimers exist in two different conformations, only one of which is capable of binding a peptide. Conversion to the active state constitutes a rate-limiting step in the binding reaction. Although our experiments do not directly address this possibility, the accelerated kinetics of RMA-S supplied with native exogenous 2m seem to preclude a major role for such an effect. Second, membrane anchorage of 2m stabilizes empty heterodimers and substantially prolongs their persistence on the cell surface in a peptide-receptive conformation. It is interesting to note in this regard that an additional membrane anchor is in fact provided to the 2m/H chain dimer at the PLC by tapasin, which may also contribute to stability of the open conformation. Third, whereas following dissociation from the MHC-I complex n2m is practically lost from the cell, its membranal attachment retains it in the cell membrane so that it is capable of rebinding to H chains by lateral diffusion. This is a highly likely scenario, which is supported by the observation that preincubation of KD-21-6(tc) cells with Abs to h2m for 1 h severely impairs their peptide-binding capacity, reducing it to that of parental RMA-S cells (Fig. 6B). Although disruption of only partially and transiently denatured dimers cannot be ruled out, Ab blocking of 2m rebinding to form peptide-receptive heterodimers is a conceivable interpretation of this result. The half-life of the empty heterodimer formed between H-2Kb and n2m has been estimated to be in the order of several minutes (6, 7). The finding that saturation by peptide is achieved already within 1 min sets an upper limit on the lifespan of noncoupled H chains. Taken together, these data imply that in the excess of membranal 2m and in the absence of peptide, at any given time point the majority of H chains are present on the transfectants cell surface as heterodimers rather than noncoupled monomers. Hence, the marked increase in the amount of chimeric h2m coimmunoprecipitated with H-2Kb in the presence of peptide (Fig. 7A) may primarily reflect the corresponding increase in affinity for the H chain, which prevents heterodimer disruption by the detergent, rather than recruitment of 2m by lateral diffusion. In fact, both the second and third accounts predict markedly elevated persistence of preformed h2m/H chain heterodimers at the cell surface, and in this regard, they are indistinguishable. Collectively, our findings provide yet another demonstration of the allosteric control underlying MHC-I stability. Genetic input of 2m with high accessibility to MHC-I H chains results in marked stabilization of a peptide-receptive conformation, as manifested both by vast acceleration of binding kinetics and the ability to bind a peptide at extremely low concentrations.
Although the level of h2m expressed by KD-21-6(tc) cells is comparable to that of D-323-4(Kbtc) cells, the former display a more pronounced phenotype. In particular, they functionally present synthetic peptide to B3Z T cells at a concentration at least 100-fold lower than D-323-4(Kbtc), as shown in Fig. 3. In fact, under the experimental conditions used in this experiment, a concentration of 1 fg/ml peptide translates into an average of 2.5 peptides/cell. Another interesting finding is the plateau reached during dissociation after 10 h (Fig. 4B), which was observed for KD-21-6(tc) but not for D-323-4(Kbtc). The observation was reproducible and may reflect binding equilibrium with the peptide released into the culture medium in the course of incubation, achieved as a result of the enhanced ability of KD-21-6(tc) cells to bind the peptide at exceedingly low concentrations. We were unable to detect binding of the complex-specific Ab in other experiments performed in the presence of serum and in the absence of synthetic peptide (data not shown). This rules out contribution of cross-reactive peptide(s) from either an exogenous or an endogenous source to this observation. We tend to attribute this difference to the natural propensity of the CD3-derived anchor in KD-21-6(tc) to homodimerize, which may confer yet greater stability on the resulting MHC-I molecules.
The properties endowed on MHC-I molecules by expression of membrane-anchored 2m bear obvious implications to vaccine development. Two versions of this polypeptide can be envisaged. The first is the one described in this study, namely, a peptide-less mode. Introduction of the gene into dendritic cells is expected to endow them with the capacity to be effectively loaded in vivo or ex vivo with a desired combination of immunogenic peptides, applied at increasingly low concentrations. The fact that 2m is monomorphic and can pair with all human MHC-I alloforms renders the use of such a construct universal in essence. The preliminary in vivo evaluation of this design with the MO5 tumor model described in this report (Fig. 8) underscores the potential advantages of this modality in stimulating CTL for tumor immunotherapy. The second approach is to genetically fuse an antigenic peptide to the amino terminus of membranal 2m via a synthetic linker (see Refs.23, 24, 25, 26, 27, 28, 29). This strategy combines the overriding of bottlenecks associated with the conventional processing and presentation pathway with the stabilization effect described herein, while requiring only one expression cassette for all immunogens. In vivo studies indeed show that cells expressing such genes are superior to cells saturated with synthetic peptides in their ability to stimulate CTL generation (A. Margalit et al., manuscript in preparation).
In summary, the effect of membrane-anchored 2m on the resulting MHC-I molecules strongly supports the allosteric model for ternary complex formation. This 2m derivative offers a highly sensitive tool for studying peptide loading onto MHC-I molecules and provides a novel and widely applicable genetic platform for CTL induction.
Acknowledgments
We thank Dr. A. Admon for helpful discussions.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This study was supported by a project grant from the Israel Cancer Research Fund and by the Chief Scientist of the Ministry of Industry, Trade, and Labor (Israel).
2 Address correspondence and reprint requests to Dr. Gideon Gross, MIGAL–Galilee Technology Center, P.O. Box 831, Kiryat Shmona 11016, Israel. E-mail address: gidi@migal.org.il
3 Abbreviations used in this paper: MHC-I, MHC class I; 2m, 2-microglobulin; ER, endoplasmic reticulum; PLC, peptide loading complex; CD3, CD3 -chain; h2m, human 2m; n2m, native h2m; MFI, mean fluorescence intensity.
Received for publication May 13, 2004. Accepted for publication December 10, 2004.
References
Williams, D. B., B. H. Barber, R. A. Flavell, H. Allen. 1989. Role of 2-microglobulin in the intracellular transport and surface expression of murine class I histocompatibility molecules. J. Immunol. 142:2796.
Danliczyk, U. G., T. L. Delovitch. 1994. 2-Microglobulin induces a conformational change in an MHC class I H chain that occurs intracellularly and is maintained at the cell surface. J. Immunol. 153:3533.
Cresswell, P., N. Bangia, T. Dick, G. Diedrich. 1999. The nature of the MHC class I peptide loading complex. Immunol. Rev. 172:21.
Williams, A., C. A. Peh, T. Elliott. 2002. The cell biology of MHC class I antigen presentation. Tissue Antigens 59:3.
Gakamsky, D. M., P. J. Bjorkman, I. Pecht. 1996. Peptide interaction with a class I major histocompatibility complex-encoded molecule: allosteric control of the ternary complex stability. Biochemistry 35:14841.
Gakamsky, D. M., L. F. Boyd, D. H. Margulies, D. M. Davis, J. L. Strominger, I. Pecht. 1999. An allosteric mechanism controls antigen presentation by the H-2Kb complex. Biochemistry 38:12165.
Gakamsky, D. M., D. M. Davis, J. L. Strominger, I. Pecht. 2000. Assembly and dissociation of human leukocyte antigen (HLA)-A2 studied by real-time fluorescence resonance energy transfer. Biochemistry 39:11163.
Parker, K. C., M. DiBrino, L. Hull, J. E. Coligan. 1992. The 2-microglobulin dissociation rate is an accurate measure of the stability of MHC class I heterotrimers and depends on which peptide is bound. J. Immunol. 149:1896
Cook, J. R., N. B. Myers, T. H. Hansen. 1996. The mechanisms of peptide exchange and 2-microglobulin exchange on cell surface Ld and Kb molecules are noncooperative. J. Immunol. 157:2256.
Matko, J., Y. Bushkin, T. Wei, M. Edidin. 1994. Clustering of class I HLA molecules on the surfaces of activated and transformed human cells. J. Immunol. 152:3353
Machold, R. P., H. L. Ploegh. 1996. Intermediates in the assembly and degradation of class I major histocompatibility complex (MHC) molecules probed with free heavy chain-specific monoclonal antibodies. J. Exp. Med. 184:2251.
Vitiello, A., T. A. Potter, L. A. Sherman. 1990. The role of 2-microglobulin in peptide binding by class I molecules. Science 250:1423.
Rock, K. L., L. E. Rothstein, S. R. Gamble, B. Benacerraf. 1990. Reassociation with 2-microglobulin is necessary for Kb class I major histocompatibility complex binding of exogenous peptides. Proc. Natl. Acad. Sci. USA 87:7517
Rock, K. L., S. Gamble, L. Rothstein, B. Benacerraf. 1991. Reassociation with 2-microglobulin is necessary for Db class I major histocompatibility complex binding of an exogenous influenza peptide. Proc. Natl. Acad. Sci. USA 88:301.
Kozlowski, S., T. Takeshita, W. H. Boehncke, H. Takahashi, L. F. Boyd, R. N. Germain, J. A. Berzofsky, D. H. Margulies. 1991. Excess 2-microglobulin promoting functional peptide association with purified soluble class I MHC molecules. Nature 349:74.
Kane, K. P., L. A. Sherman, M. F. Mescher. 1991. Exogenous 2-microglobulin is required for antigenic peptide binding to isolated class I major histocompatibility complex molecules. Eur. J. Immunol. 21:2289
Otten, G. R., E. Bikoff, R. K. Ribaudo, S. Kozlowski, D. H. Margulies, R. N. Germain. 1992. Peptide and 2-microglobulin regulation of cell surface MHC class I conformation and expression. J. Immunol. 148:3723.
Lee, L., L. McHugh, R. K. Ribaudo, S. Kozlowski, D. H. Margulies, M. G. Mage. 1994. Functional cell surface expression by a recombinant single-chain class I major histocompatibility complex molecule with a cis-active 2-microglobulin domain. Eur. J. Immunol. 24:2633
Mottez, E., P. Langlade-Demoyen, H. Gournier, F. Martinon, J. Maryanski, P. Kourilsky, J. P. Abastado. 1995. Cells expressing a major histocompatibility complex class I molecule with a single covalently bound peptide are highly immunogenic. J. Exp. Med. 181:493.
Toshitani, K., V. Braud, M. J. Browning, N. Murray, A. J. McMichael, W. F. Bodmer. 1996. Expression of a single-chain HLA class I molecule in a human cell line: presentation of exogenous peptide and processed antigen to cytotoxic T lymphocytes. Proc. Natl. Acad. Sci. USA 93:236.
Lone, Y. C., I. Motta, E. Mottez, Y. Guilloux, A. Lim, F. Demay, J. P. Levraud, P. Kourilsky, J. P. Abastado. 1998. In vitro induction of specific cytotoxic T lymphocytes using recombinant single-chain MHC class I/peptide complexes. J. Immunother. 21:283.
Chung, D. H., J. Dorfman, D. Plaksin, K. Natarajan, I. M. Belyakov, R. Hunziker, J. A. Berzofsky, W. M. Yokoyama, M. G. Mage, D. H. Margulies. 1999. NK and CTL recognition of a single chain H-2Dd molecule: distinct sites of H-2Dd interact with NK and TCR. J. Immunol. 163:3699.
Uger, R. A., B. H. Barber. 1998. Creating CTL targets with epitope-linked 2-microglobulin constructs. J. Immunol. 160:1598.
Uger, R. A., S. M. Chan, B. H. Barber. 1999. Covalent linkage to 2-microglobulin enhances the MHC stability and antigenicity of suboptimal CTL epitopes. J. Immunol. 162:6024.
White, J., F. Crawford, D. Fremont, P. Marrack, J. Kappler. 1999. Soluble class I MHC with 2-microglobulin covalently linked peptides: specific binding to a T cell hybridoma. J. Immunol. 162:2671.
Tafuro, S., U. C. Meier, P. R. Dunbar, E. Y. Jones, G. T. Layton, M. G. Hunter, J. I. Bell, A. J. McMichael. 2001. Reconstitution of antigen presentation in HLA class I-negative cancer cells with peptide-2m fusion molecules. Eur. J. Immunol. 31:440.
Yu, Y. Y., N. Netuschil, L. Lybarger, J. M. Connolly, T. H. Hansen. 2002. Cutting edge: single-chain trimers of MHC class I molecules form stable structures that potently stimulate antigen-specific T cells and B cells. J. Immunol. 168:3145.
Lybarger, L., Y. Y. Yu, M. J. Miley, D. H. Fremont, N. Myers, T. Primeau, S. M. Truscott, J. M. Connolly, T. H. Hansen. 2003. Enhanced immune presentation of a single-chain major histocompatibility complex class I molecule engineered to optimize linkage of a C-terminally extended peptide. J. Biol. Chem. 278:27105.
Margalit, A., S. Fishman, D. Berko, J. Engberg, G. Gross. 2003. Chimeric 2-microglobulin/CD3 polypeptides expressed in T cells convert MHC class I peptide ligands into T cell activation receptors: a potential tool for specific targeting of pathogenic CD8+ T cells. Int. Immunol. 15:1379.
Ljunggren, H. G., K. Karre. 1985. Host resistance directed selectively against H-2-deficient lymphoma variants: analysis of the mechanism. J. Exp. Med. 162:1745.
Karttunen, J., S. Sanderson, N. Shastri. 1992. Detection of rare antigen-presenting cells by the lacZ T-cell activation assay suggests an expression cloning strategy for T-cell antigens. Proc. Natl. Acad. Sci. USA 89:6020.
Porgador, A., J. W. Yewdell, Y. Deng, J. R. Bennink, R. N. Germain. 1997. Localization, quantitation, and in situ detection of specific peptide-MHC class I complexes using a monoclonal antibody. Immunity 6:715.
Shields, M. J., L. E. Moffat, R. K. Ribaudo. 1998. Functional comparison of bovine, murine, and human 2-microglobulin: interactions with murine MHC I molecules. Mol. Immunol. 35:919.
Schmidt, W., H. Festenstein, P. J. Ward, A. R. Sanderson. 1981. Interspecies exchange of 2-microglobulin and associated MHC and differentiation antigens. Immunogenetics 13:483.
Pedersen, L. O., A. Stryhn, T. L. Holter, M. Etzerodt, J. Gerwien, M. H. Nissen, H. C. Thogersen, S. Buus. 1995. The interaction of 2-microglobulin (2m) with mouse class I major histocompatibility antigens and its ability to support peptide binding: a comparison of human and mouse 2m. Eur. J. Immunol. 25:1609(Dikla Berko, Yaron Carmi,)
The magnitude of response elicited by CTL-inducing vaccines correlates with the density of MHC class I (MHC-I)-peptide complexes formed on the APC membrane. The MHC-I L chain, 2-microglobulin (2m), governs complex stability. We reasoned that genetically converting 2m into an integral membrane protein should exert a marked stabilizing effect on the resulting MHC-I molecules and enhance vaccine efficacy. In the present study, we show that expression of membranal human 2m (h2m) in mouse RMA-S cells elevates MHC-I thermal stability. RMA-S transfectants bind an exogenous peptide at concentrations 104- to 106-fold lower than parental RMA-S, as detected by complex-specific Abs and by T cell activation. Moreover, saturation of the transfectants’ MHC-I by exogenous peptide occurs within 1 min, as compared with 1 h required for parental cells. At saturation, however, level of peptide bound by modified cells is only 3- to 5-fold higher. Expression of native h2m only results in marginal effect on the binding profile. Soluble 2m has no effect on the accelerated kinetics, but the kinetics of transfectants parallel that of parental cells in the presence of Abs to h2m. Ab inhibition and coimmunoprecipitation analyses suggest that both prolonged persistence of peptide-receptive H chain/2m heterodimers and fast heterodimer formation via lateral diffusion may contribute to stabilization. In vivo, peptide-loaded transfectants are considerably superior to parental cells in suppressing tumor growth. Our findings support the role of an allosteric mechanism in determining ternary MHC-I complex stability and propose membranal 2m as a novel scaffold for CTL induction.
Introduction
To enable meticulous immune surveillance by armed effector CTL, MHC class I (MHC-I)3 molecules must efficiently bind and present peptides derived from intracellular proteins while minimizing acquisition of undesired peptides from the extracellular milieu. The MHC-I polymorphic H chain () is a transmembrane glycoprotein harboring three extracellular domains. The membrane distal 1 and 2 domains form the peptide-binding groove, typically capable of accommodating peptides of 8–10 amino acids. The H chain is noncovalently associated, mostly through 3, with the nonpolymorphic 2-microglobulin (2m) L chain, which is a single Ig-like domain, not anchored to the plasma membrane.
The 2m L chain plays an essential role both in promoting endogenous peptide binding at the endoplasmic reticulum (ER) and in diminishing peptide exchange at the cell surface. To allow binding of peptides transported to the ER from the cytosol by TAP, 2m must first associate with the H chain to induce a peptide-receptive conformation (1, 2). This heterodimer is unstable and is assembled and subsequently maintained at the peptide loading complex (PLC) in an open state with the concerted guidance of an array of ER chaperones, including calnexin, ERP57, calreticulin, and tapasin. Upon loading with an adequate peptide, the MHC-I molecule assumes a highly stable, closed conformation and is consequently released from the PLC to exit the ER toward the cell surface (for review, see Refs.3 and 4).
At the cell surface, the ternary MHC-I complex is apparently devoid of auxiliary proteins. Its stability appears to be controlled by an allosteric mechanism, as outlined in detail in a sequence of reports (5, 6, 7), and supported by numerous studies, which examined soluble as well as cell-surface MHC-I molecules. According to this model, dissociation of the peptide considerably decreases the affinity of the H chain for 2m, reducing the heterodimer lifespan to only several minutes or less. Free H chains, in turn, bind the peptide several orders of magnitude weaker than the 2m-coupled isoform, rendering functional rebinding negligible. Reciprocally, predissociation of 2m, rather than a peptide from the ternary complex, results in the same affinity decline and fast peptide detachment (8, 9). At the cell surface, the H chain monomer often denatures, shows strong propensity to oligomerize (10), and is later internalized (11). Hence, at physiological conditions in which extracellular concentrations of both peptide and 2m are limiting, binding an exogenous peptide is an unlikely, although not an improbable, event. In accord with this model are studies (e.g., Refs.12, 13, 14, 15, 16, 17) that showed that loading cells with an extracellular peptide is greatly facilitated in the presence of high concentration of exogenous 2m.
To evoke CTL, vaccines must target immunogenic peptides to MHC-I molecules on dendritic cells. This can be accomplished extracellularly, either by direct loading or through cross presentation of peptides derived from internalized immunogens, as well as via an endogenous route, which usually entails the use of a genetic approach. The appreciation that 2m is critical both for ternary complex assembly and for controlling its stability has prompted its incorporation into various vaccine designs. These include recombinant single-chain 2m/H chain MHC-I dimers (18, 19, 20, 21, 22), soluble (23, 24, 25) or cellular (23, 26) peptide/2m fusions, and cell-expressed single-chain peptide/2m/H chain trimers (27, 28).
We recently generated membrane-anchored chimeric 2m/peptide molecules fused with the intracellular activation domain of CD3 -chain (CD3). These polypeptides functionally associate with endogenous MHC-I H chains in transfected T cells and yield an exceedingly high complex density on the cell surface (29). We reasoned that attachment of 2m to the cell membrane offers a unique, universal tool for the generation and stabilization of immunogenic MHC-I/peptide complexes and may serve as a novel vaccine platform. As a preparatory step in the design and evaluation of CTL-inducing vaccines based on membrane-anchored 2m, we investigated biochemical and functional outcome of its expression in RMA-S cells.
Materials and Methods
Vectors and expression plasmids
The use of the pBJ1-Neo expression vector and the construction of plasmid 21-2, encoding chimeric human 2m (h2m)/CD3, have been described elsewhere (29).
In plasmid 323-5, the CD3 intracellular domain was replaced with that of H-2Kb to encode h2m/Kb as follows: first-strand DNA synthesis from mRNA prepared from RMA cells was performed with the reverse primer 5'-CGCGCGGCCGCAAGTCCACTCCAGGCAGC-3', and PCR was then conducted by adding the sense primer 5'-CCCTCGAGCTCCACTGTCTCCAACATGGCG-3'. The product, encoding the 3'-part of the peptide bridge and H-2Kb transmembrane and cytoplasmic portion, was cloned into pBJ1-Neo as an XhoI/NotI fragment, together with the XbaI/XhoI fragment from plasmid 21-2, encoding h2m leader, mature protein, and 5'-part of the peptide bridge.
Expression plasmid 845-6(n2m) encodes full-length, native h2m (n2m), which was amplified from Jurkat cells mRNA by reverse transcription-PCR with the sense primer 5'-GGGTCTAGAGCCGAGATGTCTCGCTCCGTG-3' and the reverse primer 5'-CGCGCGGCCGCTTACATGTCTCGATCCCACTTAAC-3' and inserted into pBJ1-Neo as a XbaI/NotI fragment.
Mice and cell lines
Eight to 12-wk-old C57BL/6 mice were purchased from The Jackson Laboratory and bred at the Weizmann Institute of Science facilities. Animals were maintained and treated according to the Weizmann Institute of Science animal facility and National Institutes of Health guidelines.
RMA is a Rauscher virus-transformed lymphoma cell line of C57BL/6 (H-2b) origin, and RMA-S is a RMA TAP-deficient mutant (30). Cells were grown in RPMI 1640 medium, supplemented with 10% heat-inactivated FCS, 2 mM L-glutamine, and 50 μm 2-ME. B3Z (31), an OVA257–264-specific, Kb-restricted CTL hybridoma, which expresses the NFAT-LacZ reporter gene, was a kind gift from Dr. N. Shastri (University of California, Berkeley, CA). MO5 is an OVA gene-transfected B16 murine melanoma clone of C57BL/6 origin. These cells were maintained in DMEM supplemented with 10% heat-inactivated FCS, 2 mM L-glutamine, 1% sodium pyruvate, 1% nonessential amino acids (Sigma-Aldrich), combined antibiotics, and 500 μg/ml G418 (both from Invitrogen Life Technologies).
Abs, proteins, and peptides
25-D1.16, a mAb specific to Kb-OVA257–264 (32), was a kind gift from Dr. R. Germain (National Institutes of Health, Bethesda, MD). mAb against h2m (clone BM-63) was from Sigma-Aldrich. Polyclonal rabbit anti-h2m Ab was from DakoCytomation. mAbs 20-8-4 and Y3, specific to H-2Kb, and 28-14-8, specific to H-2Db, were purified from hybridoma supernatants. Purified h2m was from Sigma-Aldrich. OVA257–264 was synthesized by Dr. M. Fridkin (Weizmann Institute of Science, Rehovot, Israel).
DNA transfection
A total of 0.8 ml of 4 x 106 RMA-S cells/ml was mixed in a 4-mm sterile electroporation cuvette (ECU-104; EquiBio) with 20 μg of SalI-linearized plasmid DNA. Transfection was performed with an Easyject Plus electroporation unit (EquiBio) at 350 V, 750 μF. Cells were resuspended in fresh medium and cultured for 24–48 h in 96-well plates before addition of G418 to 1 mg/ml. Resistant clones were expanded in 24-well plates and screened by flow cytometry for expression of h2m.
Flow cytometry
A total of 106 cells was washed with FACS buffer (PBS, 5% FCS, and 0.05% sodium azide) and incubated for 30 min on ice with 100 μl of first (or control) Ab at 10 μg/ml. Cells were then washed and incubated on ice with 100 μl of 1/100 dilution of goat anti-mouse IgG (Fab-specific)-FITC-conjugated polyclonal Abs (Sigma-Aldrich) for 30 min, washed, resuspended in PBS, and analyzed with FACSCalibur (BD Biosciences). Mean fluorescence intensity (MFI) was calculated using CellQuest software (BD Biosciences). Quantitative analysis of cell surface Ags was performed with QIFIKIT (DakoCytomation) according to the manufacturer’s instructions.
Peptide loading
Twenty-four hours before onset of the experiment, cells were washed three times with PBS and incubated in OptiMEM serum-free medium (Invitrogen Life Technologies) at 37°C. Cells were then transferred to fresh OptiMEM medium and coincubated with a peptide in 24-well plates at 1 x 106 cells/ml.
B3Z activation
Peptide-pulsed or control cells were washed three times with PBS, resuspended in nonselective medium at 5 x 105 cells/ml, and 50 μl of the cells were added to microtiter plates in triplicates. PBS-washed B3Z cells were resuspended in fresh growth medium at 5 x 105 cells/ml, added to wells at a 1:1 ratio, and coincubated for 6 h at 37°C. Cells were washed twice with PBS and fixed with 0.25% glutaraldehyde 5 min at 4°C, washed three times with PBS, incubated overnight with 100 μl of 5-bromo-4-chloro-3-indolyl -D-galactoside solution (0.2% X-Gal, 2 mM MgCl2, 5 mM K4Fe(CN)6 · 3H2O, and 5 mM K3Fe(CN)6 in PBS), and scored under the microscope for blue staining.
Immunoprecipitation
A total of 7 x 107 cells were harvested and washed twice with PBS. The cells were incubated with 15 μg of Ab for 2 h at 37°C. Cells were then washed three times with PBS, resuspended in cold lysis buffer (50 mM Tris-HCl (pH 8.0), 150 mM NaCl, 1% Nonidet P-40, 1 mM PMSF, 1 μg/ml aprotinin, and 1 μg/ml leupeptin), and incubated for 25 min on ice with gentle agitation. Cell lysate was collected by centrifugation (14,000 rpm, 5 min, 4°C). Thirty microliters of protein G-Sepharose 4 Fast Flow beads (Amersham Biosciences) were washed twice in cold lysis buffer and incubated with the cell lysate for 2 h at 4°C with gentle agitation. Nonbound proteins were removed by five washing cycles with lysis buffer (10,000 rpm, 1 min, 4°C). The precipitate was then eluted from the beads using elution buffer (0.1 M glycine (pH 2.7)) and immediately neutralized with 2 M Tris (pH 9). Recovered proteins were kept at –20°C.
Immunoblot analysis
Protein samples were boiled for 3 min, separated on a 10% nonreducing SDS polyacrylamide gel at 50 mA, and transferred onto a nitrocellulose membrane. The membrane was blocked with milk buffer (0.3 g of Na2HPO4, 2.19 g of NaCl, 25 ml of double-distilled water, and 225 ml of 1% low-fat milk) overnight at 4°C and 1 h at room temperature, washed twice with PBS, and incubated for 2 h with the primary Ab. The membrane was then washed 6 times for 6 min each with PBS, incubated for 1 h with the secondary peroxidase-conjugated Ab, washed 6 times for 6 min each with TPBS (PBS with 0.05% Tween 20), 6 times for 6 min each with PBS, and then developed using chemiluminescence kit (Pierce) and x-ray film 100NIF (Fuji).
Tumor immunotherapy
Ten mice in each experimental group were inoculated s.c. in the upper back with 105 MO5 cells/mouse. Local tumor diameter was measured with calipers. Starting 8 days later, when the tumor reached 3–4 mm in diameter, mice were immunized i.p. four times at 7-day intervals with 2 x 106 irradiated peptide-loaded cells, which were prepared as follows: RMA-S cells and transfectants were washed, resuspended at 5 x 106/ml in OptiMEM, and incubated with 50 μM synthetic peptide for 2 h at 26°C and then for 4 h at 37°C. The samples were irradiated (5000 rad) and washed. Mice were monitored daily and sacrificed when moribund. Survival was defined as the day when mice were sacrificed.
Results
Construction of h2m harboring two different membranal anchors
The original design of membrane-anchored h2m has already been delineated (29). Briefly, the carboxyl-terminal methionine residue of h2m was genetically linked to a peptide bridge, comprising the 13 membrane-proximal amino acids of the extracellular portion of HLA-A2, LRWEPSSNPTIPI (single-letter code), which encompasses the proline-rich connecting peptide. This sequence was tethered to the amino terminus of the transmembrane segment of the mouse CD3 to include the entire intracellular domain (h2m/CD3tc). At the TCR complex, CD3 primarily forms disulfide-bridged homodimers through the transmembranal cysteine residue. Indeed, although associated with endogenous MHC-I H chains in transfected T cells, these chimeric h2m/CD3tc polypeptides homodimerize (D. Berko, G. Cafri, and A. Margalit, unpublished observations). To rule out possible contribution of dimerization artifacts to the activity of h2m/CD3tc, we have assembled a similar construct, substituting the CD3 transmembrane and cytoplasmic portion with that of H-2Kb (h2m/Kbtc).
Expression of membranal 2m in RMA-S increases MHC-I thermal stability
RMA-S cells do not express functional TAP, and their MHC-I assembly pathway is thus cut from its major peptide supply. As a result, MHC-I molecules are mostly loaded with scarce, suboptimal peptides generated at the ER, and their surface level at 37°C is substantially reduced compared with RMA cells. However, this level can be elevated by incubating cells with high-affinity peptides or at lower temperatures (26–28°C), in which these peptide-receptive molecules are stabilized. We were interested in testing whether the mere association with membranal 2m could stabilize these thermally labile MHC-I molecules. Three RMA-S transfectants were generated: KD-21-4(tc) and KD-21-6(tc), expressing a relatively moderate and high level of h2m/CD3, respectively, and D-323-4(Kbtc), expressing the h2m/Kbtc construct. Fig. 1 shows that all three clones, but not the parental cells, stain brightly for h2m. Surface level of H-2Db at 37°C, which drops in RMA-S 1 log compared with 27°C or RMA, is almost completely restored in KD-21-6(tc), and to somewhat lesser, still noticeable, extent in KD-21-4(tc) and D-323-4(Kbtc). Interestingly, clones KD-21-6(tc) and D-323-4(Kbtc) express more surface h2m at 27°C than at 37°C unlike clone KD-21-4(tc), which displays comparable levels at both temperatures.
FIGURE 1. FACS analysis of transfectants. RMA, RMA-S, KD-21-4(tc), KD-21-6(tc), and D-323-4(Kbtc) cells were grown in serum-free medium for 24 h at 27 and 37°C. Cells were then incubated with anti-H-2Db (28-14-8) and anti-h2m (BM-63) mAbs, or no Ab as negative control, and then with FITC-conjugated goat anti-mouse Fab Abs and analyzed by FACS. Fluorescence intensity is presented in logarithmic scale.
We went on to obtain a quantitative evaluation of the dual effect of h2m expression and exogenous peptide binding on the level of MHC-I molecules in RMA-S cells. For this purpose, we used a commercial kit for quantitative analysis of cell-surface Ags with mouse Abs, and we measured binding of the synthetic chicken OVA peptide OVA257–264 to H-2Kb, using the Kb-OVA257–264 complex-specific 25-D1.16 mAb. This analysis is summarized in Table I. Expression of h2m elevates the level of H-2Kb 5- to 8.5-fold, whereas addition of the peptide results in an additional, relatively moderate increase of 1.3- to 1.5-fold of H-2Kb, with no effect on H-2Db. In comparison, incubation of RMA-S with a peptide under these conditions raised the surface level of H-2Kb 4-fold. The somewhat weaker effect of h2m expression on the level of H-2Db (1.9- to 5.7-fold increase) is in agreement with the lower ability of h2m to stabilize cell surface H-2Db compared with Kb (33).
Table I. Absolute numbers of cell surface MHC-I molecules expressed on RMA-S cells and transfectants, with or without OVA257–264 (Pep)
Membranal 2m greatly augments the ability of RMA-S to bind an exogenous peptide
We then examined whether these thermally stable MHC-I molecules possess enhanced ability to bind synthetic OVA257–264. Fig. 2A depicts a typical binding experiment. Threshold for detection of OVA257–264 binding to RMA-S was at a concentration of 1 ng/ml, which is in good agreement with a previous report (32). However, significant peptide binding to KD-21-6(tc) was evident even at 1 pg/ml, and residual binding at this concentration could also be detected in D-323-4(Kbtc). Such results could in fact be attributed both to the overall increase in the level of 2m expressed by these cells and to the higher affinity of h2m than mouse 2m for H-2Kb (33, 34). To test this possibility, we transfected RMA-S cells with n2m. Unlike membranal 2m, which is expected to be present on the cell surface also as a noncoupled monomer and reach very high density, level of n2m entirely depends on available MHC-I H chains. To carry out a reliable comparison of peptide loading, we screened a large number of transfectants for the highest expresser and chose clone D-845-6(n2m). Preliminary analysis of thermal stability of H-2Kb and Db on the surface of these cells showed a phenotype resembling that of the parental RMA-S cells, rather than KD-21-6(tc) or D-323-4(Kbtc) (data not shown). Fig. 2B indeed reveals a slight increase in the ability of D-845-6(n2m) to bind OVA257–264 and in its saturation level compared with RMA-S cells, but it is still significantly lower than for KD-21-6(tc).
FIGURE 2. Analysis of peptide binding by transfectants. A, Comparison of KD-21-6(tc) and D-323-4(Kbtc) with parental RMA-S cells. B, Comparative analysis of KD-21-6(tc) against 845-6(n2m) and RMA-S. For both experiments, indicated cells were grown at 37°C for 24 h in serum-free medium and were then incubated for 2 h with serial dilutions of synthetic OVA257–264. Cells were stained with the anti-H-2Kb-OVA257–264 mAb 25-D1.16, and FACS analysis was performed with FITC-conjugated anti-mouse Fab Abs. MFI was calculated using the CellQuest FACS program.
Such a dramatic increase in the ability to bind an exogenous peptide is of particular relevance to the design of cell vaccines expressing membrane-anchored 2m. Therefore, we evaluated the sensitivity with which a peptide-specific T cell hybridoma can respond to transfectants vs parental cells following peptide loading. Fig. 3 presents the peptide dose response of the Kb-restricted, OVA257–264-specific B3Z hybridoma (31). In good agreement with the FACS analysis, B3Z cells could be activated by KD-21-6(tc) and D-323-4(Kbtc) cells pulsed with as little as 1 and 100 fg/ml of the peptide, respectively, whereas RMA-S cells charged with as high as 1 ng/ml failed to activate these cells detectably under the same experimental conditions.
FIGURE 3. Dose-dependent response of the B3Z T cell hybridoma to transfectants KD-21-6(tc) and D-323-4(Kbtc) and parental RMA-S cells loaded with OVA257–264. Indicated APC were grown for 24 h in serum-free medium and incubated for 2 h with serial dilutions of the peptide. Peptide-loaded cells were then incubated overnight in triplicates at a 1:1 ratio (4 x 105 cells each) with B3Z cells, a CTL hybridoma specific to the OVA257–264-H-2Kb complex, which expresses the NFAT-LacZ reporter gene. T cell activation was monitored by intracellular 5-bromo-4-chloro-3-indolyl -D-galactoside staining. Percentage of cells stained blue was evaluated under a light microscope and scored as an average of 16 fields. Fifty-four percent of B3Z cells were activated with the anti-TCR Ab 2C11 (anti-CD3, data not shown). As activation by transfectants was conducted at 1:1 ratio, a score of 27% activated B3Z cells was considered 100% of potentially activated cells, and all results were normalized accordingly.
Membranal 2m increases on-rate of peptide binding
We next asked whether this marked effect on peptide binding is also manifested in a significant change in binding kinetics. To address this question, we designed a flow cytometry-based binding assay, which scores binding of the peptide at a saturating concentration following incubation for different time intervals. Results are shown in Fig. 4A and indicate that saturation of transfectant KD-21-6(tc) and near saturation of D-323–4(Kbtc) are already achieved within 1 min, whereas D-845-6(n2m) and RMA-S require 1 and 2 h, respectively. To assess peptide dissociation rates, we incubated cells with a saturating amount of peptide and similarly followed persistence of specific complexes on the cell surface. Fig. 4B shows no significant difference in dissociation rate between KD-21-6(tc), D-323-4(Kbtc), and RMA-S.
FIGURE 4. Peptide-binding kinetics of transfectants KD-21-6(tc), D-323-4(Kbtc), and 845-6(n2m). A, Cells were grown at 37°C for 24 h in serum-free medium and were then incubated at 37°C for different time intervals with 2 μg/ml OVA257–264. After incubation, cells were washed several times and stained with the complex-specific mAb 25-D1.16. FACS analysis was performed with anti-mouse FAB-FITC, and MFI was derived. One hundred percent saturation was determined as the MFI value obtained after 150 min of incubation with the peptide. B, Cells were incubated for 2 h with 2 μg/ml OVA257–264 in the presence of serum, washed, and incubated at 37°C in complete medium for 24 h. After each time, interval cells were washed and fixed with 0.5% paraformaldehyde. All samples were stained and analyzed in parallel as in A.
Exogenous 2m enhances peptide binding by RMA-S but not transfectants
We considered the possibility that the chimeric 2m polypeptide completely saturates MHC-I H chains on the surface of transfected cells. In this case, addition of exogenous 2m would only have a negligible effect on the kinetics of peptide binding to these clones but is expected to substantially increase on-rate of binding to the parental cells. To test this assumption, we used the experimental design of the peptide-binding assay but now examined the effect of preincubation of cells in the presence of a high concentration of soluble h2m. Indeed, Fig. 5 shows no effect on KD-21-6(tc) but demonstrates a striking increase in the on-rate of binding to RMA-S, which now appears similar to that of the transfectant, so that saturation occurs within the first minute of incubation with the peptide.
FIGURE 5. Effect of exogenous h2m on peptide-binding kinetics. KD-21-6(tc) and parental RMA-S cells were grown at 37°C for 24 h in serum-free medium and were then incubated in the presence and in the absence of 10 μg/ml purified n2m for 1 h at 37°C. Cells were then incubated for different time intervals with 2 μg/ml OVA257–264 at 37°C. After incubation, cells were washed three times and were then stained with the complex-specific 25-D1.16 mAb and analyzed by FACS.
Abs to h2m inhibit peptide binding to cells expressing membranal 2m
Results obtained so far imply that availability of 2m on the cell surface is a key factor, which governs the unique peptide-binding properties of the transfected cells. To assess the contribution of membranal 2m directly, we used polyclonal Abs to h2m. When applied along with an exogenous peptide for even a short period, these Abs should significantly diminish de novo formation of heterodimers but would have a smaller effect on pre-existing ones. In contrast, longer preincubation with the Abs is expected to result in a more pronounced effect also on pre-existing heterodimers. Fig. 6A presents flow cytometry analysis of the amount of peptide bound to KD-21-6(tc) cells following 1 min of incubation with a saturating amount of peptide in the presence of different dilutions of the Abs: in one set, the Ab was added to the assay along with the peptide, and in the other set, the Abs were preincubated for 1 h. These results suggest that the faster kinetics can largely be attributed to pre-existing heterodimers because only a minimal effect is evident following short incubation. Longer preincubation with Ab, nevertheless, severely diminishes binding, implying reformation of heterodimers is important. We then monitored ternary complex formation, following 1 h of preincubation with an effective concentration of these inhibiting Abs. Fig. 6B shows that the increase in binding on-rate is completely abolished under these conditions, and the peptide-binding profile is practically identical with that of RMA-S cells, which, as expected, are unresponsive to the presence of these Abs.
FIGURE 6. Effect of anti-h2m Abs on peptide-binding kinetics. A, KD-21-6(tc) cells were grown at 37°C for 24 h in serum-free medium and then incubated with serial dilutions of polyclonal rabbit anti-h2m Ab (DakoCytomation), either for 1 h at 37°C or for 1 min at room temperature. Cells were then incubated with 2 μg/ml OVA257–264 for 1 min, washed, stained with 25-D1.16, and analyzed by FACS. B, KD-21-6(tc) and parental RMA-S cells were similarly grown and incubated in the presence or in the absence of 1/62 dilution of the anti-h2m Ab for 1 h at 37°C. Cells were then incubated for different time intervals with 2 μg/ml OVA257–264 at 37°C, washed, and similarly analyzed by FACS.
Peptide binding primarily recruits pre-existing membranal 2m molecules
To obtain direct evidence for the involvement of membranal 2m in peptide binding, we performed a coimmunoprecipitation analysis of transfectant D-323-4(Kbtc), using the anti-H-2Kb Ab Y3. This experiment compared the amount of coprecipitated h2m before and after peptide loading, relative to the increase in total surface H-2Kb during the incubation period, as monitored by FACS. Results presented in Fig. 7A reveal a vast increase in the amount of h2m coprecipitated with H-2Kb H chain from the cell membrane following peptide binding, which is accompanied by only a 44% increase of total H-2Kb (Fig. 7B).
FIGURE 7. Coimmunoprecipitation of h2m with surface H-2Kb molecules from D-323-4(Kbtc) cells. Cells were grown 24 h in serum-free medium and incubated for 2 h with (+) or without (–) 2 μg/ml OVA257–264. A, Y3 Ab was incubated with cells for 2 h, cells were washed and lysed with 1% Nonidet P-40, and immune complexes were immunoprecipitated with protein G-Sepharose beads. Immunoblots of eluted samples were analyzed by ECL using HRP-conjugated rabbit anti-h2m polyclonal Abs. Results of three exposures times (minutes) are shown. Under these experimental conditions, capturing of intracellular H-2Kb molecules is minimized but cannot be entirely ruled out. B, FACS analysis of the same cells in the same experiment, performed with the indicated Abs. Results are shown as MFI.
Transfectants are superior to parental cells in treating tumor-bearing mice
To demonstrate the potential of membrane-anchored 2m as a new platform for CTL induction in vivo, we performed a tumor suppression experiment. MO5 is a transfectant of the B16 melanoma cell line (H-2b), which expresses chicken OVA and presents OVA257–264 in the context of H-2Kb. C57BL/6 mice were challenged with 105 MO5 cells each. Starting 8 days later, mice were subjected to an immunization regimen with either irradiated D-323-4(Kbtc) or parental RMA-S cells, both pulsed with OVA257–264, or with no cells as control. Tumor growth was substantially delayed in mice vaccinated with D-323-4 compared with RMA-S (Fig. 8A), and 80% of mice in the D-323-4(Kbtc) vaccinated group remained alive (five tumor-free), compared with only 40% (three tumor-free) in the group immunized with RMA-S 6 wk after the tumor challenge (Fig. 8B).
FIGURE 8. Inhibition of tumor growth by peptide-pulsed D-323-4(Kbtc) and RMA-S cells. Ten C57BL/6 mice in each experimental group were challenged with MO5 tumor cells. Eight days later, mice were vaccinated four times at 7-day intervals with D-323-4(Kbtc) or RMA-S cells pulsed with a saturating concentration of OVA257–264 or with no cells, as described in Materials and Methods. A, Tumor progression during 6 wk after tumor challenge expressed as mean tumor diameter in each of the three groups. Arrows indicate immunizations. B, Survival curves of mice in the three groups during the same period. Mice were monitored daily and sacrificed when moribund, which occurred when tumor reached a diameter of 20 mm.
Discussion
In this study, we expressed membrane-anchored h2m in mouse RMA-S cells and showed up to 106-fold increase in the ability of transfected cells to bind an exogenous peptide and at least 50-fold faster binding kinetics. The affinity of 2m for the MHC-I H chain determines heterodimer stability and, as a result, peptide-binding capacity (35, 36). Indeed, h2m possesses a higher affinity for most mouse MHC-I H chains than does mouse 2m, including H-2Kb and H-2Db, which are expressed by RMA-S (33, 34). Hence, these observations could be solely attributed to the elevated level of 2m expressed by transfected cells and to the higher affinity of the human L chain for MHC-I H chains. However, several findings suggest a major contribution of yet another factor. First, the combined effect of affinity with expression level, as revealed by the quantitative FACS analysis (Table I), mounts, at most, to only 8.5-fold increase of cell surface H-2Kb in transfectants, a value that is too small to account for the magnitude of the observed phenomena. Second, clone D-845-6(n2m), which expresses n2m, exhibits only a mild shift in peptide-binding profile (Figs. 2 and 4) compared with transfectants expressing the membranal derivative. Third, at peptide saturation, the total level of H-2Kb-OVA257–264 complexes formed on the surface of transfectants is only 3- to 5-fold higher than on parental RMA-S cells (Figs. 2 and 4–6).
We repeatedly observed higher level of membrane-anchored h2m at 27°C than at 37°C (Fig. 1). This difference may reflect a remaining degree of thermal instability of MHC-I molecules expressed by transfectants, and suggests that exit of membranal h2m to the cell surface from the ER still depends on available peptides.
A large fraction of MHC-I molecules on RMA-S cells cultured at 37°C are either H chain monomers or short-lived heterodimers (9, 37) so that efficient peptide binding requires equilibrium with exogenous 2m. Of particular importance is the observation that D-845-6(n2m) cells reach saturation only after 1 h of incubation with the peptide, whereas RMA-S cells are fully saturated within 1 min in the presence of soluble h2m. This result indicates that the contribution of exogenous 2m to peptide-binding kinetics far exceeds the effect exerted by cellular expression of n2m. Together with the similar binding kinetics monitored for RMA-S in the presence of 2m and for KD-21-6(tc), our data imply that it is the membranal anchorage of 2m, which is primarily responsible for the pronounced effect on transfected cells. The ability of transfectants to functionally bind an exogenous peptide at concentrations up to 106-fold lower than RMA-S cells supports this conclusion. It coincides with the notion that peptide binding to 2m/H chain heterodimers occurs at far lower concentrations than to free H chains (38, 39), which is probably the prevalent species in RMA-S.
Several nonmutually exclusive explanations may account for this new phenotype. First, the appended polypeptide sequences form additional contacts with the H chain in a manner, which secures a receptive conformation of the peptide-binding groove. One of the conjectures of the allosteric model (6, 7) is that the empty heterodimers exist in two different conformations, only one of which is capable of binding a peptide. Conversion to the active state constitutes a rate-limiting step in the binding reaction. Although our experiments do not directly address this possibility, the accelerated kinetics of RMA-S supplied with native exogenous 2m seem to preclude a major role for such an effect. Second, membrane anchorage of 2m stabilizes empty heterodimers and substantially prolongs their persistence on the cell surface in a peptide-receptive conformation. It is interesting to note in this regard that an additional membrane anchor is in fact provided to the 2m/H chain dimer at the PLC by tapasin, which may also contribute to stability of the open conformation. Third, whereas following dissociation from the MHC-I complex n2m is practically lost from the cell, its membranal attachment retains it in the cell membrane so that it is capable of rebinding to H chains by lateral diffusion. This is a highly likely scenario, which is supported by the observation that preincubation of KD-21-6(tc) cells with Abs to h2m for 1 h severely impairs their peptide-binding capacity, reducing it to that of parental RMA-S cells (Fig. 6B). Although disruption of only partially and transiently denatured dimers cannot be ruled out, Ab blocking of 2m rebinding to form peptide-receptive heterodimers is a conceivable interpretation of this result. The half-life of the empty heterodimer formed between H-2Kb and n2m has been estimated to be in the order of several minutes (6, 7). The finding that saturation by peptide is achieved already within 1 min sets an upper limit on the lifespan of noncoupled H chains. Taken together, these data imply that in the excess of membranal 2m and in the absence of peptide, at any given time point the majority of H chains are present on the transfectants cell surface as heterodimers rather than noncoupled monomers. Hence, the marked increase in the amount of chimeric h2m coimmunoprecipitated with H-2Kb in the presence of peptide (Fig. 7A) may primarily reflect the corresponding increase in affinity for the H chain, which prevents heterodimer disruption by the detergent, rather than recruitment of 2m by lateral diffusion. In fact, both the second and third accounts predict markedly elevated persistence of preformed h2m/H chain heterodimers at the cell surface, and in this regard, they are indistinguishable. Collectively, our findings provide yet another demonstration of the allosteric control underlying MHC-I stability. Genetic input of 2m with high accessibility to MHC-I H chains results in marked stabilization of a peptide-receptive conformation, as manifested both by vast acceleration of binding kinetics and the ability to bind a peptide at extremely low concentrations.
Although the level of h2m expressed by KD-21-6(tc) cells is comparable to that of D-323-4(Kbtc) cells, the former display a more pronounced phenotype. In particular, they functionally present synthetic peptide to B3Z T cells at a concentration at least 100-fold lower than D-323-4(Kbtc), as shown in Fig. 3. In fact, under the experimental conditions used in this experiment, a concentration of 1 fg/ml peptide translates into an average of 2.5 peptides/cell. Another interesting finding is the plateau reached during dissociation after 10 h (Fig. 4B), which was observed for KD-21-6(tc) but not for D-323-4(Kbtc). The observation was reproducible and may reflect binding equilibrium with the peptide released into the culture medium in the course of incubation, achieved as a result of the enhanced ability of KD-21-6(tc) cells to bind the peptide at exceedingly low concentrations. We were unable to detect binding of the complex-specific Ab in other experiments performed in the presence of serum and in the absence of synthetic peptide (data not shown). This rules out contribution of cross-reactive peptide(s) from either an exogenous or an endogenous source to this observation. We tend to attribute this difference to the natural propensity of the CD3-derived anchor in KD-21-6(tc) to homodimerize, which may confer yet greater stability on the resulting MHC-I molecules.
The properties endowed on MHC-I molecules by expression of membrane-anchored 2m bear obvious implications to vaccine development. Two versions of this polypeptide can be envisaged. The first is the one described in this study, namely, a peptide-less mode. Introduction of the gene into dendritic cells is expected to endow them with the capacity to be effectively loaded in vivo or ex vivo with a desired combination of immunogenic peptides, applied at increasingly low concentrations. The fact that 2m is monomorphic and can pair with all human MHC-I alloforms renders the use of such a construct universal in essence. The preliminary in vivo evaluation of this design with the MO5 tumor model described in this report (Fig. 8) underscores the potential advantages of this modality in stimulating CTL for tumor immunotherapy. The second approach is to genetically fuse an antigenic peptide to the amino terminus of membranal 2m via a synthetic linker (see Refs.23, 24, 25, 26, 27, 28, 29). This strategy combines the overriding of bottlenecks associated with the conventional processing and presentation pathway with the stabilization effect described herein, while requiring only one expression cassette for all immunogens. In vivo studies indeed show that cells expressing such genes are superior to cells saturated with synthetic peptides in their ability to stimulate CTL generation (A. Margalit et al., manuscript in preparation).
In summary, the effect of membrane-anchored 2m on the resulting MHC-I molecules strongly supports the allosteric model for ternary complex formation. This 2m derivative offers a highly sensitive tool for studying peptide loading onto MHC-I molecules and provides a novel and widely applicable genetic platform for CTL induction.
Acknowledgments
We thank Dr. A. Admon for helpful discussions.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This study was supported by a project grant from the Israel Cancer Research Fund and by the Chief Scientist of the Ministry of Industry, Trade, and Labor (Israel).
2 Address correspondence and reprint requests to Dr. Gideon Gross, MIGAL–Galilee Technology Center, P.O. Box 831, Kiryat Shmona 11016, Israel. E-mail address: gidi@migal.org.il
3 Abbreviations used in this paper: MHC-I, MHC class I; 2m, 2-microglobulin; ER, endoplasmic reticulum; PLC, peptide loading complex; CD3, CD3 -chain; h2m, human 2m; n2m, native h2m; MFI, mean fluorescence intensity.
Received for publication May 13, 2004. Accepted for publication December 10, 2004.
References
Williams, D. B., B. H. Barber, R. A. Flavell, H. Allen. 1989. Role of 2-microglobulin in the intracellular transport and surface expression of murine class I histocompatibility molecules. J. Immunol. 142:2796.
Danliczyk, U. G., T. L. Delovitch. 1994. 2-Microglobulin induces a conformational change in an MHC class I H chain that occurs intracellularly and is maintained at the cell surface. J. Immunol. 153:3533.
Cresswell, P., N. Bangia, T. Dick, G. Diedrich. 1999. The nature of the MHC class I peptide loading complex. Immunol. Rev. 172:21.
Williams, A., C. A. Peh, T. Elliott. 2002. The cell biology of MHC class I antigen presentation. Tissue Antigens 59:3.
Gakamsky, D. M., P. J. Bjorkman, I. Pecht. 1996. Peptide interaction with a class I major histocompatibility complex-encoded molecule: allosteric control of the ternary complex stability. Biochemistry 35:14841.
Gakamsky, D. M., L. F. Boyd, D. H. Margulies, D. M. Davis, J. L. Strominger, I. Pecht. 1999. An allosteric mechanism controls antigen presentation by the H-2Kb complex. Biochemistry 38:12165.
Gakamsky, D. M., D. M. Davis, J. L. Strominger, I. Pecht. 2000. Assembly and dissociation of human leukocyte antigen (HLA)-A2 studied by real-time fluorescence resonance energy transfer. Biochemistry 39:11163.
Parker, K. C., M. DiBrino, L. Hull, J. E. Coligan. 1992. The 2-microglobulin dissociation rate is an accurate measure of the stability of MHC class I heterotrimers and depends on which peptide is bound. J. Immunol. 149:1896
Cook, J. R., N. B. Myers, T. H. Hansen. 1996. The mechanisms of peptide exchange and 2-microglobulin exchange on cell surface Ld and Kb molecules are noncooperative. J. Immunol. 157:2256.
Matko, J., Y. Bushkin, T. Wei, M. Edidin. 1994. Clustering of class I HLA molecules on the surfaces of activated and transformed human cells. J. Immunol. 152:3353
Machold, R. P., H. L. Ploegh. 1996. Intermediates in the assembly and degradation of class I major histocompatibility complex (MHC) molecules probed with free heavy chain-specific monoclonal antibodies. J. Exp. Med. 184:2251.
Vitiello, A., T. A. Potter, L. A. Sherman. 1990. The role of 2-microglobulin in peptide binding by class I molecules. Science 250:1423.
Rock, K. L., L. E. Rothstein, S. R. Gamble, B. Benacerraf. 1990. Reassociation with 2-microglobulin is necessary for Kb class I major histocompatibility complex binding of exogenous peptides. Proc. Natl. Acad. Sci. USA 87:7517
Rock, K. L., S. Gamble, L. Rothstein, B. Benacerraf. 1991. Reassociation with 2-microglobulin is necessary for Db class I major histocompatibility complex binding of an exogenous influenza peptide. Proc. Natl. Acad. Sci. USA 88:301.
Kozlowski, S., T. Takeshita, W. H. Boehncke, H. Takahashi, L. F. Boyd, R. N. Germain, J. A. Berzofsky, D. H. Margulies. 1991. Excess 2-microglobulin promoting functional peptide association with purified soluble class I MHC molecules. Nature 349:74.
Kane, K. P., L. A. Sherman, M. F. Mescher. 1991. Exogenous 2-microglobulin is required for antigenic peptide binding to isolated class I major histocompatibility complex molecules. Eur. J. Immunol. 21:2289
Otten, G. R., E. Bikoff, R. K. Ribaudo, S. Kozlowski, D. H. Margulies, R. N. Germain. 1992. Peptide and 2-microglobulin regulation of cell surface MHC class I conformation and expression. J. Immunol. 148:3723.
Lee, L., L. McHugh, R. K. Ribaudo, S. Kozlowski, D. H. Margulies, M. G. Mage. 1994. Functional cell surface expression by a recombinant single-chain class I major histocompatibility complex molecule with a cis-active 2-microglobulin domain. Eur. J. Immunol. 24:2633
Mottez, E., P. Langlade-Demoyen, H. Gournier, F. Martinon, J. Maryanski, P. Kourilsky, J. P. Abastado. 1995. Cells expressing a major histocompatibility complex class I molecule with a single covalently bound peptide are highly immunogenic. J. Exp. Med. 181:493.
Toshitani, K., V. Braud, M. J. Browning, N. Murray, A. J. McMichael, W. F. Bodmer. 1996. Expression of a single-chain HLA class I molecule in a human cell line: presentation of exogenous peptide and processed antigen to cytotoxic T lymphocytes. Proc. Natl. Acad. Sci. USA 93:236.
Lone, Y. C., I. Motta, E. Mottez, Y. Guilloux, A. Lim, F. Demay, J. P. Levraud, P. Kourilsky, J. P. Abastado. 1998. In vitro induction of specific cytotoxic T lymphocytes using recombinant single-chain MHC class I/peptide complexes. J. Immunother. 21:283.
Chung, D. H., J. Dorfman, D. Plaksin, K. Natarajan, I. M. Belyakov, R. Hunziker, J. A. Berzofsky, W. M. Yokoyama, M. G. Mage, D. H. Margulies. 1999. NK and CTL recognition of a single chain H-2Dd molecule: distinct sites of H-2Dd interact with NK and TCR. J. Immunol. 163:3699.
Uger, R. A., B. H. Barber. 1998. Creating CTL targets with epitope-linked 2-microglobulin constructs. J. Immunol. 160:1598.
Uger, R. A., S. M. Chan, B. H. Barber. 1999. Covalent linkage to 2-microglobulin enhances the MHC stability and antigenicity of suboptimal CTL epitopes. J. Immunol. 162:6024.
White, J., F. Crawford, D. Fremont, P. Marrack, J. Kappler. 1999. Soluble class I MHC with 2-microglobulin covalently linked peptides: specific binding to a T cell hybridoma. J. Immunol. 162:2671.
Tafuro, S., U. C. Meier, P. R. Dunbar, E. Y. Jones, G. T. Layton, M. G. Hunter, J. I. Bell, A. J. McMichael. 2001. Reconstitution of antigen presentation in HLA class I-negative cancer cells with peptide-2m fusion molecules. Eur. J. Immunol. 31:440.
Yu, Y. Y., N. Netuschil, L. Lybarger, J. M. Connolly, T. H. Hansen. 2002. Cutting edge: single-chain trimers of MHC class I molecules form stable structures that potently stimulate antigen-specific T cells and B cells. J. Immunol. 168:3145.
Lybarger, L., Y. Y. Yu, M. J. Miley, D. H. Fremont, N. Myers, T. Primeau, S. M. Truscott, J. M. Connolly, T. H. Hansen. 2003. Enhanced immune presentation of a single-chain major histocompatibility complex class I molecule engineered to optimize linkage of a C-terminally extended peptide. J. Biol. Chem. 278:27105.
Margalit, A., S. Fishman, D. Berko, J. Engberg, G. Gross. 2003. Chimeric 2-microglobulin/CD3 polypeptides expressed in T cells convert MHC class I peptide ligands into T cell activation receptors: a potential tool for specific targeting of pathogenic CD8+ T cells. Int. Immunol. 15:1379.
Ljunggren, H. G., K. Karre. 1985. Host resistance directed selectively against H-2-deficient lymphoma variants: analysis of the mechanism. J. Exp. Med. 162:1745.
Karttunen, J., S. Sanderson, N. Shastri. 1992. Detection of rare antigen-presenting cells by the lacZ T-cell activation assay suggests an expression cloning strategy for T-cell antigens. Proc. Natl. Acad. Sci. USA 89:6020.
Porgador, A., J. W. Yewdell, Y. Deng, J. R. Bennink, R. N. Germain. 1997. Localization, quantitation, and in situ detection of specific peptide-MHC class I complexes using a monoclonal antibody. Immunity 6:715.
Shields, M. J., L. E. Moffat, R. K. Ribaudo. 1998. Functional comparison of bovine, murine, and human 2-microglobulin: interactions with murine MHC I molecules. Mol. Immunol. 35:919.
Schmidt, W., H. Festenstein, P. J. Ward, A. R. Sanderson. 1981. Interspecies exchange of 2-microglobulin and associated MHC and differentiation antigens. Immunogenetics 13:483.
Pedersen, L. O., A. Stryhn, T. L. Holter, M. Etzerodt, J. Gerwien, M. H. Nissen, H. C. Thogersen, S. Buus. 1995. The interaction of 2-microglobulin (2m) with mouse class I major histocompatibility antigens and its ability to support peptide binding: a comparison of human and mouse 2m. Eur. J. Immunol. 25:1609(Dikla Berko, Yaron Carmi,)