| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Immunology, University of Toronto, Toronto, Canada
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
One potential approach to augmenting the surface display and immunogenicity of an epitope is to physically couple it to its presenting MHC molecule. The peptide-binding class I dimer is comprised of a polymorphic 44-kDa membrane bound heavy chain interacting with an invariant 12-kDa soluble light chain, ß2-microglobulin (ß2m)3 (9). In previous reports, peptide Ags have been tethered, via flexible polypeptide linkers, to the heavy chain of the mouse class I molecule Kd (10) and the human heavy chain from HLA-A2 (11). The resulting fusion proteins have been shown to elicit CTL responses when expressed in transfected cells. In addition, CTLs have been induced in vivo using a chemically modified, photoreactive peptide cross-linked to Kd complexes (12). However, there has been no attempt, to the best of our knowledge, to exploit the potential of the ß2m subunit for coupling peptide Ag.
Structurally, tethering a peptide to ß2m is less demanding than coupling Ag to the heavy chain, as the carboxyl end of the peptide and amino terminus of ß2m are positioned relatively close together (13). Since ß2m is a soluble molecule, it is amenable for use as a protein immunogen, unlike peptide/heavy chain fusions that must be cell surface bound. Additionally, ß2m protein has been observed to act as an "adjuvant" for enhancing peptide-specific CTL responses in vivo (14), presumably by assisting in the MHC loading of peptides, a phenomenon that has been extensively investigated in vitro (15, 16, 17). Therefore, an epitope-linked ß2m molecule could provide a simple and more efficient means to enhance the formation of defined MHC/peptide complexes.
In this report, we describe two different peptide-ß2m fusion proteins. One molecule, a human ß2m (hß2m) with a tethered Db-restricted influenza nucleoprotein (NP) epitope, was able to form CTL target structures endogenously through expression in transfected murine cell lines. A second protein, hß2m with a linked Kd-restricted influenza NP epitope, was expressed in Escherichia coli and was able to sensitize cells for peptide-specific lysis when added to target cells exogenously. In both scenarios, target cell killing was attributable to the intact epitope-linked ß2m and was not due to a free, uncoupled peptide. Thus, peptide-linked ß2m molecules, whether produced endogenously through the normal biosynthetic pathway or added as exogenous protein, can form specific MHC/peptide complexes that trigger CTL-mediated killing. These molecules therefore offer an attractive strategy for designing new CTL-priming vaccines.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
P815 (H-2d) is a murine mastocytoma (TIB-64, American Type Culture Collection, Rockville, MD). RMA-S (H-2b) is a mutagenized and immunoselected variant of the mouse T lymphoma cell line RBL-5, and is defective in class I surface expression due to a defect in the TAP-2 gene (18, 19). RMA (H-2b) is a mutagenized but nonselected control. RMA and RMA-S were cultured in complete medium (RPMI 1640 supplemented with 10% (v/v) FBS, L-glutamine, antibiotics, nonessential amino acids, and sodium pyruvate). P815 cells were cultured in defined serum-free conditions (10% FBS replaced by 2% Ultroser HY (Life Technologies, Grand Island, NY)). RMA and RMA-S transfectants were cultured in the presence of G418 medium (500 µg/ml). The anti-H-2Db Ab (IgG2b) B22.249.R1 (20, 21) and the anti-hß2m Ab (IgG2b) BBM.1 (22) have been described previously. Purified hß2m was purchased from Calbiochem (San Diego, CA).
Construction of eukaryotic expression plasmids
Full length hß2m (including signal sequence) was cloned by RT-PCR from mRNA isolated from the human lymphoblastoid cell line T2 (23). Amplification was performed with the 5' primer (5'-TCTAAGCTTGCCACCATGTCTCGCTCCGTG-3'), which includes a HindIII restriction site and Kozak sequence, and a 3' primer (5'-TATTCTAGATTACATGTCTCGATCCCA-3') encoding an XbaI site. The PCR product was cloned into the HindIII and XbaI sites of the pcDNA3 plasmid (Invitrogen, San Diego, CA) and sequenced to confirm its identity. A unique XhoI site was created by site-directed mutagenesis (Transformer Site-Directed Mutagenesis Kit, Clontech, Palo Alto, CA) to facilitate NP(366–374)-L8-hß2m creation.
NP(366–374)-L8-hß2m was generated by a two-step PCR process. Initially, wild-type (wt) hß2m was amplified using the 5' primer (5'-GGAGGAGGATCCGGAGGTGGCAGCATCCAGCGTACTCCAAAGAATCAGG-3'), which hybridizes to hß2m immediately adjacent to the signal sequence and encodes the eight-amino acid linker sequence GGGSGGGS (single-letter amino acid code). This PCR product was then used as a template for a subsequent PCR reaction, using the 5' primer (5'-GTACTCGAGGCTGCTTCCAATGAAAATATGGAGACTATGGGAGGAGGATCCGAGGTGGC-3'), which hybridizes to the region encoding the glycine/serine linker, and contains, as an overhang, the sequence encoding the NP366–374 epitope and an XhoI site. Both rounds of amplification used the 3' primer mentioned above, which encodes an XbaI site. The XhoI/XbaI fragment of wt hß2m was removed and replaced with the similarly digested PCR product, producing NP(366–374)-L8-hß2m.
Construction of bacterial expression plasmids
wt hß2m was generated by PCR using the ß2m.phn1 plasmid (a kind gift from Dr. D. Wiley, Howard Hughes Medical Institute, Harvard University, Cambridge, MA) as a template and using oligonucleotide primers (5'-TATCATATGATCCAGCGTACTCCA-3'), encoding an NdeI site, and (5'-TATGGATCCTTACATGTCTCGATCCCA-3'), encoding a BamHI site. The PCR product was initially cloned into the pNoTA/T7 vector (5 Prime-3 Prime, Inc., Boulder, CO), sequenced, and then cloned again by NdeI/BamHI digest into the pET-12a vector (Novagen, Madison, WI).
NP(147–155)-L12-hß2m was constructed in a two-step PCR process. Initially, the ß2m.phn1 plasmid was amplified with the oligonucleotides (5'-GGAGGAGGAGGATCTGGAGGAGGAGGATCTGGAGGAATCCAGCGTACTCCAAAGATTCAGGTT-3'), encoding the glycine/serine linker and hybridizing to the first 27 bases of hß2m, and (5'-TATGGATCCTTACATGTCTCGATCCCA-3'), encoding a BamHI site. This PCR product was then amplified in a second reaction, using the forward primer (5'-CATATGACCTACCAGCGTACCCGTGCTCGTGTTGGAGGAGGAGGATCTGGAGGAGGAGGATCT-3'), encoding the NP147–155 epitope and an NdeI site as an overhang, and the reverse primer from the first amplification. The final PCR product was cloned into the pNoTA/T7 vector (5 Prime-3 Prime, Inc.), sequenced, and then cloned again by NdeI/BamHI digest into the pET-12a vector (Novagen).
Transfections
Twenty micrograms of hß2m or NP(366–374)-L8-hß2m were linearized with PvuI, added to 1.5 x 107 cells (washed twice in serum-free medium) in a 0.4-cm cuvette, and pulsed at either 300 V, 900 µF (for RMA) or 250 V, 500 µF (for RMA-S) with a Gene Pulser apparatus (Bio-Rad, Richmond, CA). Cells were then incubated for 10 min at room temperature, 1 ml of complete medium was added, and cells were incubated for an additional 20 min. Transfectants were cultured for 2 days in normal medium before being transferred to G418 selection medium.
Flow cytometry
Cells (106) were incubated on ice for 1 h with 2 µg of primary Ab (in PBS containing 1% BSA and 0.1% sodium azide). Cells were then washed twice and incubated for 1 h with a goat anti-mouse FITC conjugate (Sigma Chemical Co., St. Louis, MO). Cells were washed three times and resuspended in 0.5 ml PBS (0.1% sodium azide), and paraformaldehyde fixative was added to obtain a final concentration of 1% (v/v). Samples were analyzed on a Coulter flow cytometer (Hialeah, FL).
CTL assays on transfectants
CTLs were primed by immunizing mice with 200 hemagglutinin units of the X-31 strain of influenza virus, which possesses the NP gene from the A/PR/8/34 virus (24). At least 4 wk postinfection, spleen cells were harvested and either restimulated for 5 days with virus-infected, irradiated spleen cells (at a 10:1 responder to stimulator ratio) or restimulated for 7 days in the presence of 2 µg/ml NP366–374 peptide (Alberta Peptide Institute, Alberta, Canada). The restimulated effectors were then used in a standard 51Cr release assay. Target cells (106) were labeled for 1.5 h with 100 µCi of Na2[51Cr]O4 (Amersham, Arlington Heights, IL) and washed repeatedly. When necessary, target cells were incubated with peptide (>100 nM) for at least 30 min at room temperature. Cells (104) were dispensed into 96-well plates, titrated effectors were added, and the plates were incubated at 37°C for 4 h. Supernatants were harvested using a filter system (Skatron, Lier, Norway) and radioactivity was measured by gamma-counting (Beckman, Fullerton, CA). Percent specific lysis was calculated, using the mean of triplicate samples, as: 100 x ([experimental cpm - spontaneous cpm]/[maximum cpm - spontaneous cpm]). Spontaneous cpm values were determined by incubating target cells alone in medium, and maximum values were determined by lysing target cells in 1% Triton X-100 (v/v).
Peptide elution
Cells (107) were incubated in a peptide-stripping buffer (0.13 M citric acid, 66 mM Na2HPO4, 150 mM NaCl, 17 µg/ml phenol red, pH 3.2) (25, 26) for 1 min at room temperature. Cells were then pelleted, and supernatants were recovered and neutralized by dropwise addition of NaOH. Peptides were filtered and stored at -20°C. The acid treatment had no effect on cell viability as assessed by trypan blue staining. As a positive control, RMA cells were first cultured overnight at 37°C with 1 µg/ml NP366–374 and then washed extensively in PBS before extraction.
Acid-eluted fractions were tested in a standard CTL assay, essentially as described above, with the following modifications. After 51Cr labeling and washing, target cells (RMA) were resuspended at 2 x 105/ml and hß2m was added to a final concentration of 5 µg/ml. The addition of exogenous hß2m was intended to compete out any peptide-linked hß2m molecules that may have been acid-stripped. Cells (750 µl) were transferred to 750 µl of medium containing 10-fold serial dilutions of the acid-eluted material (starting at 1:8 dilution) and incubated at room temperature for 20 min. Cells (100 µl) (104) were then plated out in 96-well plates containing 4 x 105 effector cells/well, thus achieving a 40:1 E:T ratio.
Preparation of bacterial lysates
Wild-type hß2m and NP(147–155)-L12-hß2m plasmids (in pET vectors) were transformed into the bacterial strain BL21(DE3)plysS (Novagen). Protein expression and lysate preparation were performed essentially as described (27). Briefly, bacterial cultures were grown in the presence of ampicillin (100 µg/ml) and chloramphenicol (34 µg/ml) until OD600 = 0.4 to 0.8. Cultures were then induced by the addition of isopropyl ß-D-thiogalactopyranoside to a final concentration of 0.4 mM and grown for 2 to 3 h. Bacteria were harvested by centrifugation and the cell pellet was resuspended in 10 mM Tris, pH 8.0, supplemented with 1 mM EDTA, 20 µg/ml DNase, 20 µg/ml RNase, and 50 µg/ml PMSF. Cells were lysed by repeated cycles of freeze-thaw, with or without sonication. Lysates were centrifuged (10,000 x g) for 20 min, and the pellet was washed with 10 mM Tris, pH 8.0. The pellet was then solubilized in 1/20-vol 8 M urea/100 mM Tris, pH 8.0, and centrifuged at 100,000 x g for 1 h at 4°C. The hß2m protein in the supernatant was then refolded by dialysis against 10 mM Tris, pH 7.0, and stored at -70°C.
Western blotting
Lysates were run on 10 to 20% gradient SDS-PAGE gels and transferred to nitrocellulose (Trans-Blot, Bio-Rad). Filters were probed with an anti-hß2m rabbit serum (1:500 dilution) and detected with a goat anti-rabbit horseradish peroxidase conjugate (Sigma, 1:1000 dilution) using the enhanced chemiluminescence Western Blotting Detection Reagents kit (Amersham) according to the manufacturer’s protocols.
CTL assays using bacterial lysates
CTLs from X-31-immunized BALB/c mice were restimulated in vitro for 7 days by culturing cells in the presence of 2 µg/ml NP147–155 peptide (Alberta Peptide Institute). Target cells (P815) were labeled with 51Cr as described above, except that labeling was performed using a defined serum replacement (Ultroser HY, Life Technologies), which is free of bovine ß2m. Targets were incubated with various dilutions of the NP(147–155)-L12-hß2m lysate for 1 h at 37°C, washed once, plated, and assayed as described above. For hß2m inhibition assays, target cells were incubated with wt hß2m bacterial lysate (1:20 final concentration) for 45 min at room temperature before treatment with the NP(147–155)-L12-hß2m lysate.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A mammalian expression vector was constructed that encodes a
recombinant hß2m with covalently attached peptide Ag
(Fig. 1
A). The
hß2m-linked Ag, NP366–374, is an optimal
H-2Db-restricted immunodominant epitope from influenza
virus NP (28). It was connected to the amino terminus of the mature
domain of hß2m via an eight-amino acid polypeptide linker
(GGGSGGGS, single-letter amino acid code). The epitope and linker were
inserted immediately following the hß2m signal sequence,
and thus the recombinant molecule should be targeted to the endoplasmic
reticulum (ER) for cell surface expression. Note that this
peptide-hß2m fusion, designated
NP(366–374)-L8-hß2m, is heterologous in nature, using a
mouse class I binding peptide and ß2m of human origin.
The mouse influenza virus model offers a convenient, well-characterized
system in which CTL target structure formation can be readily assessed,
while the human ß2m allows for the recombinant fusion
protein to be monitored amidst a background of mouse ß2m.
In addition, the hß2m subunit is known to interact with
mouse class I heavy chains with slightly higher affinity than mouse
ß2m (29), potentially enhancing the effectiveness of
NP(366–374)-L8-hß2m.
|
Expression and target structure formation in NP(366–374)-L8-hß2m transfectants
NP(366–374)-L8-hß2m and the control wt
hß2m vector were initially transfected into the murine
cell line RMA (H-2b). Stable, drug-resistant lines were
analyzed for expression of the transfected hß2m gene
product at the cell surface by flow cytometry using a mAb (BBM.1)
specific for hß2m. As shown in Figure 2A,
NP(366–374)-L8-hß2m transfectants showed only a modest
increase in surface staining (2- to 3-fold higher mean fluorescence)
compared with untransfected RMA cells. The fact that this staining was
lower than expected cannot be attributed to negative effects from the
presence of the epitope and/or linker, since RMA cells transfected with
a wt hß2m-encoding vector also showed only a slight
increase in mean fluorescence. Indeed, the low expression of
transfected hß2m at the cell surface is likely
attributable to an overall level of low protein production, as
determined by metabolic labeling and immunoprecipitation (data not
shown).
|
B).
Since RMA-S cells cannot supply peptides to nascent class I molecules,
only low amounts of properly folded class I molecules are expressed at
the cell surface (31). Thus, expression of transfected
NP(366–374)-L8-hß2m can also be assessed by monitoring
the level of H-2Db cell surface expression with the
conformation-sensitive Ab B22.249.R1 (B22). As shown in Figure 2C, RMA-S cells transfected with
NP(366–374)-L8-hß2m showed only a very slight increase
in B22 staining. Thus, the transfectants appear to be producing only
small amounts of the NP epitope. This is consistent with the above
results, and suggests that there is a low level of expression of the
transfected gene product.
Despite low cell surface expression of
NP(366–374)-L8-hß2m, RMA and RMA-S transfectants are
capable of efficiently generating a CTL target structure recognizable
by NP366–374-specific CTLs. In a standard cytotoxicity
assay using CTL effectors from influenza virus-immunized mice, RMA and
RMA-S cells transfected with NP(366–374)-L8-hß2m were
lysed at levels comparable with the positive control, represented by
untransfected cells pulsed with free peptide (Fig. 3, A and B).
The killing was specific for the tethered epitope, as RMA and RMA-S
cells transfected with wt hß2m exhibited only background
lysis, similar to untreated, nontransfected RMA and RMA-S. Thus,
transfection of mouse cell lines with a vector encoding
hß2m with covalently attached class I binding peptide
results in target structure generation and cell lysis. The observation
that RMA-S transfectants can generate recognizable CTL target
structures indicates that the presentation of the tethered NP Ag does
not require TAP transport, which is consistent with the Ag being
covalently linked to the ß2m subunit (see below). The
high level of specific lysis in the context of low transfected gene
product expression suggests a potent effect from the
NP(366–374)-L8-hß2m molecule.
|
While the observation that RMA-S cells transfected with
NP(366–374)-L8-hß2m are sensitized for CTL lysis
indicates that the NP epitope cannot originate from the cytosol, it
does not conclusively demonstrate that the peptide Ag remains
covalently tethered to the hß2m subunit throughout the
presentation pathway. Indeed, it is possible that the polypeptide
linker is cleaved, perhaps during assembly of the class I complex in
the ER or at the cell surface, and the resulting free peptide is
mediating the observed CTL killing. To address this issue, we attempted
to isolate free NP peptide from the surface of
RMA.NP(366–374)-L8-hß2m transfectants by acid elution.
In this experiment, transfectants were incubated briefly in a low pH
buffer that is sufficient to cause dissociation of surface class I
molecules yet gentle enough not to affect cell integrity. Free NP
peptide was detected by titrating the acid-eluted fraction in a
standard CTL assay. As a positive control, NP peptide was eluted from
RMA cells that had been preincubated with NP366–374 (and
washed extensively) before elution. The results are shown in Figure 4. There was no detectable NP peptide in
the material acid-eluted from NP(366–374)-L8-hß2m
transfectants or from the negative control, represented by
untransfected RMA cells. In contrast, peptides acid-eluted from the
positive control cells, RMA prepulsed with NP366–374, were
capable of sensitizing target cells for lysis over an extremely large
dilution range, titrating out near 105. Thus, there is
almost 100,000-fold less elutable free peptide on
RMA.NP(366–374)-L8-hß2m transfectants compared with
peptide-pulsed RMA cells. Since we expect 
104 NP
peptides on each peptide-pulsed RMA cell (10% occupancy of roughly
105 surface class I molecules (32)), there must be <1 free
NP peptide per NP(366–374)-L8-hß2m transfectant. Hence,
it is reasonable to conclude that target cell lysis is being mediated
through peptide-linked hß2m molecules and not through the
action of a free, uncoupled peptide.
|
In a related strategy, we have engineered a recombinant
hß2m molecule with covalently attached class I
binding peptide for expression in E. coli. This molecule,
NP(147–155)-L12-hß2m, has the Kd-restricted
influenza NP epitope NP(147–155) (28) tethered to hß2m
via a 12-amino acid glycine/serine linker (Fig. 1B).
Note that this molecule does not possess a signal sequence, and thus
should be localized to the bacterial cytoplasm. The epitope is preceded
by a methionine residue to initiate mRNA translation. This methionine
will likely be removed shortly after protein synthesis (33), and
therefore is not expected to be incorporated into the NP epitope.
NP(147–155)-L12-hß2m and wt hß2m
constructs were cloned into the pET-12a bacterial expression vector
(Novagen), which uses an inducible T7 RNA polymerase promoter. E.
coli transformants were grown and induced, and lysates were
prepared as described in Materials and Methods. Western blot
analysis of lysates using an anti-hß2m rabbit serum
reveled a single band from the wt hß2m transformants that
comigrates with a standard hß2m (Fig. 5). The
NP(147–155)-L12-hß2m transformant lysate shows a single
reactive band that is 2 kDa larger, consistent with the additional
mass from the linker and epitope. These bands are not visible in a
control lysate that is prepared from bacteria harboring an irrelevant
gene (ß-galactosidase). There are no smaller bands detected by the
anti-hß2m serum in the
NP(147–155)-L12-hß2m lysate, indicating that the
epitope-tethered hß2m molecule is not being significantly
degraded. Typically, 500 µg of NP(147–155)-L12-hß2m
protein are obtained per liter of bacteria culture, as determined by
Coomassie staining and competitive ELISA analysis (data not shown). The
yield of wt hß2m is dramatically higher (100 mg/L) for
reasons that are currently unknown.
|
The ability of a recombinant
NP(147–155)-L12-hß2m produced in E. coli
to form recognizable CTL target structures was assessed by a simple
cytotoxicity assay. P815 cells (H-2d) grown in defined
serum-free medium (to exclude any competition from bovine
ß2m) were pretreated with unpurified lysate from
NP(147–155)-L12-hß2m transformants before incubation
with CTL effectors derived from influenza-infected BALB/c mice. As
shown in Figure 6, exposing P815 cells to
the NP(147–155)-L12-hß2m lysate resulted in a high level
of target cell lysis, comparable with the killing obtained by pulsing
target cells with free NP147–155 peptide. This cytotoxic
effect was specific for the epitope, as incubation with lysates from wt
hß2m or control transformants failed to sensitize
targets. The NP(147–155)-L12-hß2m lysate was remarkably
potent. Less than 1 ng/ml of NP(147–155)-L12-hß2m
protein was capable of sensitizing target cells for CTL killing (data
not shown). Thus, a bacterially expressed recombinant
hß2m with covalently attached peptide is capable of
efficiently sensitizing target cells for lysis when added exogenously
as a component of an unpurified lysate.
|
As is the case with the mammalian-expressed
NP(366–374)-L8-hß2m construct, we wished to
determine whether the biologic activity observed with the
NP(147–155)-L12-hß2m protein was truly due to a
peptide-linked hß2m structure or was a result of a free
peptide generated through cleavage of the glycine/serine linker.
Although the NP(147–155)-L12-hß2m lysate was extensively
dialyzed during preparation, thus removing any contaminating free
peptide, it is possible that the recombinant protein was being degraded
during the course of the in vitro CTL assay. To address this issue, we
attempted to inhibit the activity of the
NP(147–155)-L12-hß2m lysate with wt hß2m
protein. If target structure formation is a result of an NP-tethered
hß2m molecule interacting with cell surface
Kd heavy chains, then the biologic activity of
NP(147–155)-L12-hß2m should be diminished by providing
an excess of competitor wt hß2m. Alternatively, if a free
peptide is mediating CTL sensitization, then the presence of additional
wt hß2m should have no inhibitory effect. In the
experiment shown in Figure 7, P815 target
cells were preincubated with wt hß2m lysate as a source
of competitor hß2m before exposure to low concentrations
of the NP(147–155)-L12-hß2m lysate. Pretreatment of
target cells with the wt lysate had a dramatic inhibitory effect on the
activity of NP(147–155)-L12-hß2m protein. In contrast,
when P815 cells were pretreated with wt lysate and then pulsed with
free NP147–155 peptide, no inhibitory effect was observed
(the "inhibitory" hß2m actually resulted in slightly
higher lysis of peptide-pulsed targets). Similarly, inhibition of low
concentrations (<1 ng/ml) of NP(147–155)-L12-hß2m could
also be accomplished using 20 µg/ml of purified, commercially
available hß2m (data not shown). Note that the inhibition
of NP(147–155)-L12-hß2m protein by wt hß2m
can only be observed at very low concentrations of the epitope-tethered
hß2m, presumably because of the potent activity of this
recombinant molecule.
|
) and
suggests that the peptide-linked hß2m molecules are
relatively resistant to degradation. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Physically coupling a peptide to its presenting class I MHC molecule is one such mechanism that has been investigated. The structural and biochemical data available on MHC/peptide interactions have demonstrated the importance of free peptide-termini in binding to class I molecules (34, 35, 36, 37) and have suggested that a linked peptide may not bind stably to MHC. Nevertheless, optimal binding class I peptides have been successfully tethered, via their carboxyl termini, to the amino termini of both mouse and human class I heavy chains (10, 11). These recombinant molecules were capable of presenting their linked epitope and inducing CTLs when expressed in transfected cell lines. While these studies demonstrate the feasibility of such an approach, their applications may be limited by the requirement to express the MHC/peptide fusions in transfectants. Alternatively, covalent class I-peptide complexes have been formed using a modified peptide that contains a photoreactive chemical cross-linker (12). This strategy, however, requires that the chemical modifications do not disturb peptide binding to MHC and can result in a CTL response that is primarily specific for the altered peptide.
In this report, we describe an alternative and as yet unexplored approach: tethering peptides to the ß2m subunit. Structurally, this linkage should prove less demanding than a peptide/heavy chain fusion, since the carboxyl terminus of a class I bound peptide is located significantly closer to the amino terminus of ß2m than to the amino terminus of the heavy chain (13, 35). As ß2m is a small soluble protein, a peptide-ß2m fusion molecule could be used as a soluble protein immunogen. In this context, it is noteworthy that ß2m protein has been observed to augment epitope-specific CTL responses in vivo when coinjected with peptide (14). This adjuvant effect was attributed to the ability of ß2m to enhance peptide loading of surface class I molecules, a well-defined in vitro phenomenon (15, 16, 17). We anticipated that a more potent immunogen could be generated if the peptide was covalently linked to ß2m, thereby restricting its diffusion and creating a molecule that contains two high affinity binding sites for class I heavy chains.
Toward this goal, we have created two unique peptide-linked
ß2m molecules that use different presentation pathways.
The first, NP(366–374)-L8-hß2m, is a recombinant
hß2m that possesses the H-2Db-restricted
influenza NP epitope NP366–374 tethered to its amino
terminus via an 8-amino acid polypeptide linker. This molecule was
expressed in transfected murine cell lines, and would presumably
function within the normal, endogenous class I biosynthetic pathway.
Despite only modest cell surface expression (Fig. 2), which we
attribute to a generally low level of protein production,
NP(366–374)-L8-hß2m transfectants, but not wt
hß2m transfectants, were efficiently lysed by
NP366–374-specific CTLs (Fig. 3). In this regard, it is
important to note that other studies have indicated that only small
numbers of class I MHC/peptide complexes (ranging from several hundred
to <10) are required to sensitize a cell for lysis (32, 38, 39).
To demonstrate that the observed lysis was due to a peptide-linked
ß2m structure and not a free peptide resulting from
protein degradation, acid elution experiments were performed (Fig. 4).
We were unable to liberate any detectable NP peptide from the surface
of NP(366–374)-L8-hß2m transfectants, in contrast to NP
peptide-pulsed control cells, from which the acid-eluted material could
sensitize targets for CTL lysis over a 105-fold dilution
range. Hence, the biologic activity of
NP(366–374)-L8-hß2m molecules in the mouse transfectants
is attributable to an antigenic peptide physically coupled to the
hß2m subunit.
The second recombinant molecule, NP(147–155)-L12-hß2m,
couples the H-2Kd-restricted influenza NP epitope
NP147–155 to the amino terminus of hß2m
using a 12-residue glycine/serine linker (Fig. 1B).
Rather than expressing this peptide-ß2m fusion molecule
endogenously in transfected cell lines, the recombinant
protein was produced in E. coli to permit CTL target
structure formation through an exogenous pathway. Using an isopropyl
ß-D-thiogalactopyranoside-inducible expression system,
NP(147–155)-L12-hß2m protein was produced in
BL21(DE3)plysS transformants. This NP(147–155)-L12-hß2m
molecule could be detected as a single band when bacterial lysates were
analyzed by Western blot using an anti-hß2m rabbit
serum (Fig. 5). When added to H-2d-expressing P815 target
cells exogenously, the unpurified NP(147–155)-L12-hß2m
lysate resulted in CTL-mediated lysis (Fig. 6). Target cell
sensitization by the NP(147–155)-L12-hß2m lysate could
be inhibited by pretreating cells with an excess of competitor wt
hß2m (Fig. 7), indicating that the activity of the
NP(147–155)-L12-hß2m lysate is a result of a
ß2m-linked epitope and not a free peptide. This is
consistent with the above mentioned observation that the
mammalian-expressed NP(366–374)-L8-hß2m molecule also
mediates CTL target structure formation via a linked
peptide-hß2m structure.
Our results have indicated that target cells can be rendered susceptible to CTL-mediated killing by exposure to a peptide-hß2m linked molecule. This strategy for forming defined class I MHC complexes demonstrates versatility in two respects. First, the route of presentation can be either endogenous, through expression of a peptide-hß2m fusion in transfected cells, or exogenous, via treatment of cells with bacterially derived peptide-hß2m protein. It is noteworthy that both routes of presentation appear quite potent. Transfected cell lines are lysed despite low levels of transfected gene expression (as judged by flow cytometry) and lysate treated target cells are sensitized by extremely small quantities (subnanomolar) of recombinant protein. Second, the strategy has proven successful for two different epitopes (influenza NP366–374 and NP147–155), restricted through different class I molecules (Db and Kd), using two different sizes of polypeptide linkers (8 and 12 residues). This suggests that a general peptide-ß2m fusion strategy could be extended to a variety of class I binding peptides.
In summary, we have demonstrated that CTL target structure formation can be accomplished in vitro by physically coupling peptide Ag to the ß2m subunit, either in the context of DNA transfection or as an exogenous, bacterially derived protein. We are currently investigating the ability of peptide-ß2m molecules to elicit primary CTL responses in vivo. Since ß2m has been observed to augment otherwise weak CTL responses to peptide immunogens (14), a peptide-linked ß2m strategy may offer a safe, convenient, and effective method of inducing CTL immunity to desired class I restricted epitopes. This may be achieved by plasmid DNA immunization (40) with an appropriate epitope-linked ß2m expression vector, or alternatively by saline injection of the purified epitope-linked ß2m protein. Success with either approach would offer a new alternative with respect to the induction of adjuvant-independent, epitope-specific CTL responses.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Brian H. Barber, Department of Immunology, University of Toronto, Medical Sciences Building, 1 King’s College Circle, Toronto, Canada, M5S-1A8.
3 Abbreviations used in this paper: ß2m, ß2-microglobulin; hß2m, human ß2-microglobulin; NP, nucleoprotein; wt, wild-type; ER, endoplasmic reticulum.
Received for publication August 15, 1997. Accepted for publication October 23, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  N. Tiwari, N. Garbi, T. Reinheckel, G. Moldenhauer, G. J. Hammerling, and F. Momburg A Transporter Associated with Antigen-Processing Independent Vacuolar Pathway for the MHC Class I-Mediated Presentation of Endogenous Transmembrane Proteins J. Immunol., June 15, 2007; 178(12): 7932 - 7942. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. J. Holst, C. Bartholdy, A. Stryhn, A. R. Thomsen, and J. P. Christensen Rapid and sustained CD4+ T-cell-independent immunity from adenovirus-encoded vaccine antigens J. Gen. Virol., June 1, 2007; 88(6): 1708 - 1716. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Margalit, H. M. Sheikhet, Y. Carmi, D. Berko, E. Tzehoval, L. Eisenbach, and G. Gross Induction of Antitumor Immunity by CTL Epitopes Genetically Linked to Membrane-Anchored {beta}2-Microglobulin J. Immunol., January 1, 2006; 176(1): 217 - 224. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Vidric, W.-K. Suh, U. Dianzani, T. W. Mak, and T. H. Watts Cooperation between 4-1BB and ICOS in the Immune Response to Influenza Virus Revealed by Studies of CD28/ICOS-Deficient Mice J. Immunol., December 1, 2005; 175(11): 7288 - 7296. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Berko, Y. Carmi, G. Cafri, S. Ben-Zaken, H. M. Sheikhet, E. Tzehoval, L. Eisenbach, A. Margalit, and G. Gross Membrane-Anchored {beta}2-Microglobulin Stabilizes a Highly Receptive State of MHC Class I Molecules J. Immunol., February 15, 2005; 174(4): 2116 - 2123. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. M. Bertram, W. Dawicki, B. Sedgmen, J. L. Bramson, D. H. Lynch, and T. H. Watts A Switch in Costimulation from CD28 to 4-1BB during Primary versus Secondary CD8 T Cell Response to Influenza In Vivo J. Immunol., January 15, 2004; 172(2): 981 - 988. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Margalit, S. Fishman, D. Berko, J. Engberg, and G. Gross Chimeric {beta}2 microglobulin/CD3{zeta} 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., November 1, 2003; 15(11): 1379 - 1387. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Lybarger, Y. Y. L. Yu, M. J. Miley, D. H. Fremont, N. Myers, T. Primeau, S. M. Truscott, J. M. Connolly, and T. H. Hansen 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., July 11, 2003; 278(29): 27105 - 27111. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Kawana-Tachikawa, M. Tomizawa, J.-i. Nunoya, T. Shioda, A. Kato, E. E. Nakayama, T. Nakamura, Y. Nagai, and A. Iwamoto An Efficient and Versatile Mammalian Viral Vector System for Major Histocompatibility Complex Class I/Peptide Complexes J. Virol., October 25, 2002; 76(23): 11982 - 11988. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. L. Cannons, E. M. Bertram, and T. H. Watts Cutting Edge: Profound Defect in T Cell Responses in TNF Receptor-Associated Factor 2 Dominant Negative Mice J. Immunol., September 15, 2002; 169(6): 2828 - 2831. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. M. Bertram, P. Lau, and T. H. Watts Temporal Segregation of 4-1BB Versus CD28-Mediated Costimulation: 4-1BB Ligand Influences T Cell Numbers Late in the Primary Respons |
1 µl) were added. After
pulsing, target cells were washed once and then incubated in a standard
4-h chromium release assay using CTLs derived from influenza
virus-immunized BALB/c mice. The entire assay was performed using a
defined serum substitute devoid of any bovine ß2m.