The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

The Journal of Immunology, 2008, 181, 7062 -7072
Copyright © 2008 by The American Association of Immunologists, Inc.
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Flobakk, M.
Right arrow Articles by Sandlie, I.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Flobakk, M.
Right arrow Articles by Sandlie, I.
Right arrowPubmed/NCBI databases
*Substance via MeSH

Processing of an Antigenic Sequence from IgG Constant Domains for Presentation by MHC Class II1

Morten Flobakk2,*,{ddagger}, Ingunn B. Rasmussen*,{ddagger}, Elin Lunde*,{ddagger}, Terje Frigstad{dagger},{ddagger}, Gøril Berntzen*,{ddagger}, Terje E. Michaelsen§, Bjarne Bogen{dagger},{ddagger} and Inger Sandlie3,*,{ddagger}

* Department of Molecular Biosciences, {dagger} Institute of Immunology, and {ddagger} Centre for Immune Regulation, University of Oslo, Oslo, Norway; and § Department of Vaccines, National Institute of Public Health, Oslo, Norway


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Targeting of T cell epitopes to APC enhances T cell responses. We used an APC-specific Ab (anti-IgD) and substituted either of 18 loops connecting β strands in human IgG constant H (CH) domains with a characterized T cell peptide epitope. All Ab-epitope fusion molecules were secreted from producing cells except IgG-loop 2(BC)CH1, and comparing levels, a hierarchy appeared with fusions involving CH2≥CH1>CH3. Within each domain, fusion at loop 6(FG) showed best secretion, while low secretion correlated with the substitution of native loops that contain conserved amino acids buried within the folded molecule. Comparing the APC-specific rAb molecules for their ability to induce T cell activation in vitro, the six mutants with epitope in CH2 were the most effective, with loop 4CH2 ranking on top. The CH1 mutants were more resistant to processing, and the loop 6CH1 mutant only induced detectable activation. The efficiency of the CH3 mutants varied, with loop 6CH3 being the least effective and equal to loop 6 CH1. Considering both rAb secretion level and T cell activation efficiency, a total of eight loops may carry T cell epitopes to APC for processing and presentation to T cells, namely, all in CH2 in addition to loop 6 in CH1 and CH3. Comparing loop 4CH2 with loop 6CH1 mutants after injection of Ab in BALB/c mice, the former was by far the most efficient and induced specific T cell activation at concentrations at least 100-fold lower than loop 6CH1.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Increased presentation of antigenic peptide in complex with MHC molecules enhances T cell responses, and Ab- mediated targeting of peptide or whole Ag to APC is a way to increase the number of peptide-MHC complexes on the surface of APC (1). This is important since major efforts are concentrated on the development of safe vaccines that generate strong, specific T cell responses. Abs with T cell epitopes within their V region CDR loops have been described previously (2). Although such rAbs may enter APC by way of Fc{gamma}Rs, rAbs with peptides or whole Ag added C-terminally to Fab (3) or to Ab (4, 5) may target a preferred and defined APC surface molecule by way of V region specificity. We have introduced peptide epitopes into loops connecting β strands in constant heavy (CH)4 domains, and denoted the epitope-loaded rAb "Troybodies." The loop-grafting experiments have involved the amino acid sequence 91–101 of the MOPC315 plasmacytoma {lambda}2 L chain (91–101 {lambda}2315), which represents a minimal stimulating T cell epitope (6) presented on I-Ed MHC class II for CD4+ T cells (7). In initial experiments, this epitope was exchanged with or inserted into either of three loops (BC, DE, or FG) in CH1 of human (h) IgG3 (8) and murine (m) IgG2b (9). For simplicity, these loops are hereby denoted loop 2CH1, 4CH1, and 6CH1, respectively (see Fig. 1A). The loop 2CH1 hIgG3 rAb mutant was retained, while all other single loop substitution mutants were secreted. We also showed that loop 6CH1 hIgG3 could be substituted with model epitopes that show great variation in amino acid sequence, length, and secondary structure, namely, aa 323–339 from OVA, aa 110–120 from hemagglutinin, and aa 46–61 from hen egg lysozyme (10). Initially, the epitope was grafted into hapten-specific rAbs (8) and, subsequently, the rAbs were equipped with V genes encoding APC specificity. Following in vitro targeting, the epitopes were excised from the rAbs, loaded on MHC class II molecules, and presented to specific T cells. Such Troybody targeting to murine IgDa resulted in a 103-fold improvement in presentation efficiency compared with rAbs with irrelevant specificity. Importantly, T cell activation was improved up to 105-fold compared with that achieved using synthetic peptide or whole protein (10, 11). The same results were obtained in experiments with (91–101 {lambda}2315) in loop 6CH1 where the target was MHC class II (I-E) (12). Furthermore, a 102-fold enhanced presentation was seen in vivo in mice (12).


Figure 1
View larger version (16K):
[in this window]
[in a new window]

  		FIGURE 1. A, Graphic representation of a model C domain. The loops are numbered 1–6, β strands are labeled A–F. B, Human C{gamma}3 gene (ID: X03604). Exons are marked as arrows. For simplicity, the four hinge exons are shown as one exon. Restriction sites used for subcloning are indicated.

 
Comparing the in vitro presentation efficiency of APC loaded with mouse and human rAbs with 91–101 {lambda}2315epitopes grafted in various loops in CH1 (8, 9), differences were found which prompted us to initiate a comprehensive analysis that involved grafting in all loops in all three CH domains. Each domain has six loops connecting β strands, offering 18 possibilities for loop replacement. To identify the loops that are best suited for epitope insertion, we exchanged every loop in the three hIgG3 CH domains with the amino acid sequence 89–105 from {lambda}2315 (89–105) {lambda}2315 and show here that a total of 17 such fusion molecules were secreted from transiently transfected 293E cells, although in different amounts. Levels of secretion are compared with variability as well as hydropathicity and solvent accessibility score for each amino acid within each domain.

For each domain, the loop 6 mutant was secreted at the highest level. Furthermore, the CH1 and CH2 mutants (except loop 2CH1) were secreted better than the CH3 domain mutants, of which the loop 6 mutant only was secreted well. In general, low secretion correlated with the removal of conserved amino acids that were buried either within a domain core or between two interacting domains.

All secreted rAbs were tested for the ability to stimulate CD4+ T cells in vitro in T cell activation and growth inhibition assays. Importantly, T cell epitope loop grafting on all three CH domains was compatible with efficient stimulation of T cells. Eight mutants were both secreted in good amounts and found to induce T cell activation, namely, loop 6CH1, all CH2 mutants and loop 6CH3. The difference in induction potential was large, as the best activator (loop 4CH2) induced T cell activation at a concentration that was 102-fold lower than the weakest (loop 6CH1). These two Abs were selected for further studies in vivo and both were injected into the tail vein of normal mice at various concentrations. Following in vivo targeting, isolated spleen cells as APC stimulated specific T cells in vitro and again the loop 4CH2 mutant was at least 102-fold more efficient than the loop 6CH1 mutant.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Mice, cell lines, and Abs

BALB/c mice were bred by Taconic Farms. The study was approved by the National Committee for Animal Experiments (Oslo, Norway). 293E (CRL-10852) cells were obtained from American Type Culture Collection. The T cell clone 7A10B2 recognizes aa 91–101 of {lambda}2315 in complex with I-Ed (7). Spleen cells from BALB/c mice were used as APC, as were A20 B lymphoma cells transfected with a 2,4,6-trinitrophenyl-specific IgD BCR (A20{delta}; gift from Dr. N. Hozumi, Department of Medical Genetics, University of Toronto, Toronto, Ontario, Canada) (13).

The 4B2A1 and 7A10B2 T cell clones recognize the same {lambda}2315 I-Ed complex (7). The T cell hybridoma BW4B2A1 was produced by fusing lymph node cells from 4B2A1 TCR-transgenic SCID mice (14) with the TCR-negative T cell hybridoma line BW51.47 {alpha}/β (15) using polyethylene glycol under standard conditions. Hypoxanthine/aminopterin/thymidine-resistant clones were selected for surface expression of the 4B2A1 TCR. Functionality was confirmed by coculturing hybridoma cells with irradiated BALB/c splenocytes (20 Gy) and synthetic {lambda}2315 peptide (89–107), with subsequent detection of IL-2 in supernatant by sandwich ELISA (data not shown).

All cells were cultured in DMEM (BioWhittaker) or RPMI 1640 supplemented with 10% heat-inactivated FCS (PAA), 2 mM L-glutamine (DMEM only), 25 µg/ml streptomycin, and 25 U/ml penicillin (both from BioWhittaker) under standard conditions. Abs for IL-2 detection were rat anti-mouse IL-2 (clone JES6-1A12) and biotin rat anti-mouse IL-2 (clone JES6-5H4), both from BD Pharmingen. Abs for IFN-{gamma} detection were rat anti-mouse IFN-{gamma} (AN18 (16)) and biotin rat anti-mouse IFN-{gamma} (XMG1.2-bio (17)). mAb HP-6050 is specific for hIgG3 hinge (Sigma-Aldrich). All other Abs used were produced by us using standard procedures. The mAb K13 is specific for the human {kappa} L chain (18). The mAbs 132c8 and HP-6050 (19) were compared for binding to all four hIgG isotypes side-by-side in ELISA and found to bind the hIgG3 hinge in a similar fashion. Affinity-purified sheep polyclonal Abs, s303 (20) and s12 (21), are specific for hIgG Fab and hIgG3 hinge, respectively.

Structure analysis of hIgG3 H chain constant domains

Secondary structure analysis (β strand or loop) was conducted using data and evaluation programs given in the PDB database (http://www.rcsb.org/pdb/) (22) and as described previously (9). Briefly, crystal structures of 1FC1 (hIgG1), 1IGY (mIgG1), and 1IGT (mIgG2a) were studied regarding both loop length and limits as determined by three-dimensional visualization and the program Structure Explorer (http://www.pdb.org). Loops were numbered 1–6 in each domain starting from the N-terminal end of the polypeptide chain. Variability of each amino acid position in Ig H chains was analyzed using the sequence analysis program provided by S. M. J. Searle (The Sanger Institute, Cambridge, U.K.) as described previously (9).

The relative total side chain accessibility of each amino acid in the CH1, CH2, and CH3 domains was investigated using Naccess (http://wolf.bms.umist.ac.uk/naccess/) and PDB ID: 1HZH (human IgG1) (23). Hydropathicity analysis of hIgG3 domains CH1, CH2, and CH3 was performed using the program ProtScale (window size = 9; http://www.expasy.org/cgi-bin/protscale.pl) (24) and the Kyte-Doolittle scale (25), as provided by the Swiss Institute of Bioinformatics.

Production of mutant hIgG3: construction of loop exchange mutants

The CH chain gene encoding the G3m (b0) allotype (26) was a gift from Dr. M. P. LeFranc (International ImMunoGeneTics Information System, Montpellier, France) and cloned as a 2.8-kb fragment into HindIII-SphI sites in the polylinker of pUC19 (Sigma-Aldrich). The resulting plasmid, pUC19{gamma}3wt, was template in all mutagenesis reactions. All predicted loop sequences in C{gamma}3 CH1, CH2, and CH3, were exchanged with aa 89–105 from {lambda}2315 (FAALWFRNHFVFGGGTK) in individual H chains. The cysteine in position 90 of the endogenous sequence was exchanged with alanine. Mutagenesis was performed by QuikChange mutagenesis (Stratagene). Table I shows loop sequences and Table II shows mutagenic primer sequences. Colonies were screened for the presence of the mutation by the simultaneous introduction of a silent DraIII restriction site and DraIII digestion of plasmid DNA. All 18 mutations were subsequently confirmed by sequencing by GATC. Sequenced fragments containing mutations were subcloned into unmanipulated vectors as HindII-BglII, BglII-PmlI, PmlI-NsiI, or PmlI-SphI (Fig. 1B) to exclude possible amplification errors outside the sequenced areas. Complete mutant H chain genes were assembled in pLNOH2IgD, which encode IgD-specific H chains, as described elsewhere (11, 27). The corresponding L chain gene, encoded on pLNOKIgD (11, 27) and each of the 18 pLNOH2IgD variants were transiently cotransfected in 293E cells (28). Portions of supernatant were harvested and replaced with fresh medium every 2–3 days for 14 days as described. A substitution of loop 4CH1 with 91–101 {lambda}2315 has been described elsewhere (8) and the complete H chain gene was assembled in pLNOH2IgD. The corresponding loop 6CH1 substitution in an IgD-specific H chain has been described previously (11). Both H chain genes were transiently cotransfected with pLNOKIgD and supernatant harvested as described above. The complete IgD-specific rAbs are denoted "loop 4 or loop 6CH1 (91–101)," respectively.


View this table:
[in this window]
[in a new window]

  		Table I. Loop positions in hIgG3

 

View this table:
[in this window]
[in a new window]

  		Table II. Mutagenesis primersa

 
Sandwich ELISA for detection of hIgG3 variant concentrations

The amounts of IgD-specific rAb mutants secreted after each transfection were measured as follows: 96-well microtiter plates were coated with a hIgG3-specific Ab and incubated overnight at room temperature (RT). Then, samples of 100 µl of diluted supernatants were added to each well and detected with a second hIgG3-specific Ab. A hIgG3wt preparation was diluted in a 3-fold series and used as standard. Three different Ab combinations were used as coat and detecting agent: s303 (2 µg/ml) and 132c8-bio (1/6000), s12 (10 µg/ml) and 132c8-bio (1/6000), as well as K13 (2 µg/ml) and s303-bio (1/6000), respectively. Detection was done with the substrate for alkaline phosphatase, p-nitrophenyl phosphate (Sigma-Aldrich) diluted in diethanolamine buffer to 1 mg/ml.

Isolation of mutant hIgG3

Proteins in supernatants were precipitated by 1:1 addition of portions of a saturated ammonium sulfate solution. Incubation at RT for 20 min was followed by a 10-min centrifugation at 17,000 x g using a Sorvall centrifuge (OneMed). Pellets were dissolved in dH2O and dialyzed three times to PBS/azide and two times to RPMI 1640 at 4°C before sterile filtration. Abs used in in vivo targeting experiments were affinity purified from cell supernatant by use of protein G-conjugated Sepharose columns.

Western blotting

Western blots were performed using Criterion XT Precast Gels (Bio-Rad). Briefly, the Ab samples were preheated at 95°C for 3 min before they were loaded onto the gel and separated at 140 V for 100 min. By Semi Dry transfer at 20 V for 30 min, proteins were transferred to an Immobilon-P polyvinylidene difluoride membrane (Sigma-Aldrich). Detection reagents were biotinylated mouse anti-hIgG3 (HP-6050) (Sigma-Aldrich) and HRP-conjugated streptavidin (GE Healthcare). Detection was performed using the ECL Plus Western Blotting Detection Reagents kit (GE Healthcare) as described by the manufacturer, and membranes were analyzed using ImageQuant TL version 2003.02 software (GE Healthcare). Individual Ab concentrations were normalized to a known concentration of purified hIgG3.

Binding to soluble Fc{gamma}RIIA or protein G

The extracellular domains of hFc{gamma}RIIA were cloned and expressed as soluble fusion to GST (Fc{gamma}RIIA-GST) as described elsewhere (29). ELISA plates were coated overnight at 4°C with mouse IgD at 4 µg/ml. The following day, the wells were blocked in 2% skim milk, washed three times in PBS/0.05% Tween 20, and incubated for 1 h at RT with the mutant Ab samples at 1 µg/ml. After washing, the wells were incubated for 1 h at RT with protein G-conjugated to HRP (VWR) or Fc{gamma}RIIA-GST at 1 µg/ml followed by anti-GST conjugated to HRP (GE Healthcare). After three washes, the plates were developed in ABTS (Sigma-Aldrich) in citrate buffer at pH 2.2. Absorbance at 405 nm was read after 15–60 min.

In vitro Ag presentation assays

Growth inhibition assay Growth inhibition assays were performed as described by Bogen et al. (7). All Ab mutants were diluted in 5-fold series starting at 1 µg/ml and added as triplicates in flat-bottom 96-well microtiter plates. 7A10B2 T cells on day 10 after the last stimulation were irradiated (20 Gy) and 103 cells were added to each well along with 5000 A20{delta} cells in exponential growth expressing the IgDa allotype (13). Synthetic 89–107 {lambda}2315 peptide was included as positive control. Supernatant from cells that had had been treated with transfection reagent without DNA addition was ammonium sulfate precipitated and used as medium control. Negative control was hIgG3wt with IgD specificity. After incubation for 24 h, the cultures were pulsed with 1 µCi of [3H]dThd for 16–24 h, harvested onto filters, and counted using a TopCount NXT scintillation counter (GMI).

T cell proliferation assay Samples of 5 x 105 irradiated (20 Gy) BALB/c spleen APC were cultured with 2 x 104 7A10B2 T cells and various amounts of rAb or synthetic 89–107 {lambda}2315 peptide in triplicates for 48 h. The cultures were then pulsed with 1 µCi of [3H]dThd for 16–24 h, harvested onto filters, and counted using the TopCount1 counter (GMI).

In vivo experiments

BALB/c mice were injected i.v. in the tail with titrated amounts of loop 6CH1 and loop 4CH2 rAbs. Two mice received PBS only. Ninety minutes after i.v. injections, the mice were killed by cervical dislocation and the spleens were removed. Irradiated (8 Gy) spleen cells (5 x 105/well) were cultured with responder T cells, namely, 7A10B2 or T cell hybridomas BW4B2A1 (both 2 x 104/well). An optimal concentration of the {lambda}2315 synthetic peptide (10 µg/ml) was added to the positive control. After 72 h, portions of 100 µl of supernatant were collected for cytokine measurements, and the cultures were pulsed for 24 h with 1 µCi of [3H]dThd. The cultures were harvested and incorporated [3H]dThd was measured using the TopCount1 counter. Detection of IFN-{gamma} and IL-2 in supernatants was performed in microtiter plates coated with AN18 or JES6-1A12 (both 2 µg/ml in PBS), respectively. XMG1.2-bio (1 µg/ml in PBST) was used as detection Ab for IFN-{gamma}, whereas JES6-5H4 (1 µg/ml in PBST) was used for IL-2, followed by streptavidin-alkaline phosphatase (1/3000) and 1 mg/ml p-nitrophenyl phosphate in diethanolamine buffer. Standard curves were prepared from a 3-fold dilution starting at 35 ng/ml IFN-{gamma} supernatant or a 2-fold dilution starting at 2 ng/ml rIL-2 (BD Pharmingen).

Prediction of asparaginyl endopeptidase (AEP) cleavage sites

Prediction of AEP cleavage sites within every mutated hIgG3 H chain was performed with NetAEP (http://theory.bio.uu.nl/kesmir/AEP/) as provided by C. Kesmir at Utrecht University (Utrecht, The Netherlands). Distribution of every asparagine residue, from those positively identified as AEP cleavage sites to totally overlooked residues, were positioned relative to the inserted epitope and within the original secondary structure.

Prediction of MHC II peptide presentation

The presence of I-Ed binding peptides in hIgG3 H chains was predicted using the PredBALB/c server (30). Briefly, each mutant IgG3 H chain was scanned for I-Ed binding nonameric peptides. Nonamers of lower score than aa 92–100 in the {lambda}2315 epitope (31) were excluded.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
Construction of mutant C{gamma}3 genes and secretion of recombinant Abs

Because Abs are stable molecules and may be given unique specificities that allow targeting to APC, they are ideal vehicles for delivery of amino acid sequences that contain T cell epitopes. We therefore wished to study how such a sequence of 17 aa that contains a model epitope could be introduced into a hIgG molecule. It is essential that the fusion proteins are secreted from producing cells. We focused on loop sequences that link β strands in the CH domains of hIgG3. The mutant molecules should necessarily pass the endoplasmic reticulum quality control and be secreted as complete H2 plus L2 Ig (32). We did an analysis of the domain architecture and then took an empirical approach and exchanged every loop in every CH domain with the model sequence and estimated the amounts of rAbs secreted (Figs. 2A, 3A, and 4A).


Figure 2
View larger version (31K):
[in this window]
[in a new window]

  		FIGURE 2. A, Secretion of rAb compared with wt, as determined by ELISA. Error bars indicate SD of triplicates. The graph is derived from one representative experiment of three. rAbs are named according to loop of epitope grafting and domain, i.e., 1CH1 = epitope in loop 1 of CH1. B, Hydropathicity; C, Variability; and D, accessibility analysis of the CH1 domain. Loop positions are indicated as L1–L6.

 

Figure 3
View larger version (34K):
[in this window]
[in a new window]

  		FIGURE 3. A, Secretion of rAb compared with wt, as determined by ELISA. Error bars indicate SD of triplicates. The graph is derived from one representative experiment of three. rAbs are named according to loop of epitope grafting and domain, i.e., 1CH2 = epitope in loop 1 of CH2. B, Hydropathicity; C, variability; and D, accessibility analysis of the CH2 domain. Loop positions are indicated as L1–L6. Amino acids in IgG known to be involved in effector molecule binding are indicated: *, FcRn; {checkmark}, Fc{gamma}R; {diamondsuit}, C1q.

 

Figure 4
View larger version (31K):
[in this window]
[in a new window]

  		FIGURE 4. A, Secretion of rAb compared with wt, as determined by ELISA. Error bars indicate SD of triplicates. The graph is derived from one representative experiment of three. rAbs are named according to loop of epitope grafting and domain i.e., 1CH3 = epitope in loop 1 of CH3. B, Hydropathicity; C, Variability; and D, accessibility analysis of CH3. Loop positions are indicated as L1–L6. Amino acids in IgG known to be involved in effector molecule binding are indicated: *, FcRn.

 
Since no crystal structure of a hIgG3 has been published, secondary structure analysis of individual C region domains was based on knowledge of the structure of representative IgGs, as described in Materials and Methods. Amino acids located in loops connecting β strands were determined and the results are presented in Table I. Although loops 1, 4, and 6 in each domain connect β strands within the same sheet, loops 2, 3, and 5 connect β strands on two opposing sheets. One-half of the 18 loops had a proline at or close to the N- or C-terminal boundary and these were defined as part of the loop rather than the framework. Hydropathicity and solvent exposure of each amino acids in all three domains were analyzed as described in Materials and Methods. The results are described for hydropathicity (Figs. 2B, 3B, and 4B) and solvent exposure (Figs. 2C, 3C, and 4C). We found that in CH1, four of six loops are mostly hydrophilic while two are hydrophobic. All loops in CH2 are either neutral or mostly hydrophilic. In CH3, five loops are mostly hydrophilic and one is neutral in character.

The loops in the CH2 domain have by far the largest number of amino acids that are solvent exposed, while the loops in the CH3 domain have the largest number of buried amino acids. The majority of the loop amino acid residues with low solvent exposure point toward the domain core rather than an interdomain interface. Exceptions are single amino acid residues in loop 2 and loop 4CH3 that point toward the CH3-CH3 interface and one amino acid located in loop 2CH1 that point into the V region. As expected, hydrophobic regions contain many buried amino acids, mainly in frameworks, while the loops, mostly hydrophilic in nature, for the most part are exposed to the solvent.

The T cell epitope selected for study, characterized by the specificity of the T cell line 7A10B2 (7), encompasses aa 91–101 of {lambda}2315. Notably, the aa 91–97 sequence constitutes the CDR3 loop of the VL chain of the myeloma protein of MOPC315 (33). In this study, this sequence was extended by two amino acids N-terminally and eight amino acids C-terminally. The cysteine in position 90, which normally participates in the intradomain disulfide bond, was changed to alanine, resulting in the sequence 89-FAALWFRNHFVFGGGTK-105, and denoted 89–105 {lambda}2315. Compared with the loop sequences that were exchanged, this is rather hydrophobic. Studies with synthetic peptides show that the Cys90Ala substitution does not influence T cell responsiveness (B. Bogen, unpublished data).

We assumed that residues involved in maintaining structural stability would be rather conserved in various Ab molecules and that amino acids that were highly variable might well be substituted. We therefore performed the variability analysis for all three CH region domains that has previously been described for CH1 (9) (Figs. 2D, 3D, and 4D). In short, 144 different sequences of all isotypes from 30 species were aligned and the variability at each position was calculated as the number of different residues observed divided by the frequency of the most common residue. We found that both framework and loop sequences in all three domains contained both variable and conserved amino acids. For the most part, the loop amino acids were more variable than the framework residues. Conserved loop residues that in addition had low solvent exposure were in particular phenylalanine in loop 2CH1, a tryptophan residue in loop 5CH2 in addition to residues in all loops in CH3.

All loops were substituted with aa 89–105 {lambda}2315. Initially, 18 different mutants were made. Supernatants from transfectants were ammonium sulfate precipitated and dialyzed. As observed in earlier substitution experiments involving CH1, the IgD specificity was retained in all mutants (data not shown). Secretion levels were detected by three different sandwich ELISAs using pairs of hIgG3-specific Ab. In each pair, one Ab was used as coat and the other, which was biotinylated, as detection reagent. The Abs s303, s12, and K13 used as coat had specificity for Fab, hinge, or {kappa} L chain, respectively, whereas the detecting Abs 132c8-bio and s303-bio had specificity for hinge or Fab, respectively. The results in Figs. 2A, 3A, and 4A are from a representative ELISA with anti-Fab as coat and anti-hinge as detection reagent. The secretion level of wild-type (wt) IgG3 was defined as 100% and compared with the mutants. ELISAs with the two other Ab pairs gave approximately the same results. We found that all mutants were secreted at levels below wt. One mutant only, that had the mutation in loop 2 of CH1, was completely retained. Samples of the secreted rAb run on SDS-PAGE and Western blot demonstrated that all were secreted as ~165-kDa proteins characteristic of complete H2L2 molecules (data not shown).

Comparing the secretion levels of the mutants, those with 89–105 {lambda}2315 in loop 6CH1 and loop 6CH2 were secreted in relatively high amounts. Excluding loop 2CH1, all mutants with substitutions in CH1 were secreted rather well at levels between 20 and 80% of wt. The CH2 mutants were all secreted at 20–60% of wt, whereas for CH3, the loop 6 mutant only was secreted in high amounts. The remaining CH3 rAbs were secreted at levels below 10% of wt. All in all, 17 of 18 loop positions may be exchanged without complete retention and 12 without more than a 5-fold reduction in secretion.

Activation of specific T cells

It is crucial that the mutants are internalized by APC so as to enter the Ag-processing pathway and that the specific epitopes are properly excised from the rAb carrier to bind MHC and transported as peptide-MHC complexes to the cell surface. In this study, the mutant hIgG3s were tested in two different in vitro T cell activation assays, namely, a growth inhibition assay and a T cell proliferation assay.

Growth inhibition assay

The TH 1 cell clone used (7A10B2) has cytotoxic activity toward A20 lymphoma APC upon ligand recognition (7). The assay readout, incorporated radioactivity, reflects growth of the lymphoma cell APC only, as the T cells are irradiated. The assay is at least 10-fold more sensitive than a conventional T cell proliferation assay. The APC had been transfected with genes encoding IgDa (13) and were thus targets for the IgD-specific mutants. Irradiated 7A10B2 Th1 cells were mixed with secreted rAb mutants and APC as described in Materials and Methods (Fig. 5).


Figure 5
View larger version (18K):
[in this window]
[in a new window]

  		FIGURE 5. Induction of T cell activation. Mutants are compared in a growth inhibition assay. Irradiated {lambda}2315-specific 7A10B2 CD4+ T cells were cultured with A20{delta} B cells and IgD-specific rAbs at various concentrations. A pulse with [3H]dThd was given after 16–24 h and incorporation of [3H]dThd in A20{delta} cells was measured, reflecting growth of residual A20{delta} that had not been killed by activated TH1 cells. Mutants are named as in Figs. 2–4. A, CH1 mutants; B, CH2 mutants; and C, CH3 mutants. Error bars indicate SD of triplicates. The graphs are from one representative experiment of three.

 
We found that the APC mixed with all rAb-harboring epitopes in CH2 or CH3 loops activated specific T cells, with CH2 replacements being more effective and with less variation than CH3 replacements. Thus, the epitope was presented to T cells from all positions in CH2 and CH3. However, from CH1, the epitope was detectably presented from loop 6 only. Notably, a high amount of loop 6CH1 was needed to obtain the effect. At 0.01 µg/ml, a concentration that was sufficient to induce the effect of all CH2 and CH3 mutants, no growth inhibition was detected. Thus, the epitopes in loops of CH1 seemed to be somewhat protected from presentation.

T cell proliferation assay

To confirm the results from the growth inhibition assay, eight mutants were selected, namely, loop 6CH1, all CH2 mutants, and loop 6CH3, all of which were secreted at levels above 20% of wt. These were further tested in a T cell proliferation assay. BALB/c spleen cells as APC, T cells, and rAbs were combined. The APC were irradiated and, thus, in this case, incorporation of radioactivity reflects T cell proliferation upon Ag stimulation. All mutants tested induced proliferation of the specific T cells (Fig. 6A). As in the growth inhibition assays, the mutants with epitopes in either of loops 2, 3, or 4CH2 were the most efficient T cell activators, followed by those with epitope in loops 1, 5, or 6 CH2, which showed intermediate activation ability. Both loop 6CH1 and loop 6CH3 mutants required ~100x higher concentration than the best mutants to induce proper activation. A hierarchy appeared where loop 2CH2 = 3CH2 = 4CH2 > 5CH2 > 1CH2 > 6CH2 > 6 CH3 > 6CH1. The results correspond well to those obtained in the APC growth inhibition assay. A summary of results from the secretion and T cell activation assays are presented in Table III.


Figure 6
View larger version (19K):
[in this window]
[in a new window]

  		FIGURE 6. Induction of T cell activation. Mutants are compared in a T cell proliferation assay. 7A10B2 T cells were cultured with irradiated BALB/c spleen cells and rAbs at various concentrations. A pulse with [3H]dThd was given after 48 h and activation was measured as incorporation of [3H]dThd in T cells. A, IgD-specific rAb with aa 89–105 {lambda}2315. B and C, IgD-specific rAb with long (89–105) or short (91–101) {lambda}2315 aa sequence in loop 4CH1 (B) and loop 6CH1 (C). rAb mutants are named as in Figs. 2–4. Error bars indicate SD of triplicates. The graphs are derived from one representative experiment of two.

 

View this table:
[in this window]
[in a new window]

  		Table III. List of hIgG3 mutants

 
A loop 4CH1 mutant has previously proved to induce T cell activation (8). This mutant was identical to the one tested here, except for the fact that a short sequence of 11 aa, namely, 91–101 {lambda}2315, was introduced. To test whether the 89–105 sequence of 17 aa was less efficient than 91–101, two mutants with the long and two mutants with the short sequence in loop 4 or 6 of CH1, respectively, were compared side-by-side in the T cell proliferation assay. In each case, the mutant with short sequence was indeed found to be more efficient than the corresponding mutant with a long sequence (Fig. 6, B and C).

T cell activation after in vivo targeting

One of the most active CH2 mutants, namely, loop 4CH2, was selected for further characterization in vivo and compared side-by-side with the loop 6CH1 mutant. Both were injected in BALB/c mice at various concentrations, and after 90 min the spleens were removed and whole spleen cell preparations mixed with specific T cells in vitro. T cell activation was measured as proliferation, IFN-{gamma} secretion or IL-2 secretion, as a function of dose injected. Thus, in vivo APC targeting and processability of the two rAbs was compared in normal mice that express the relevant MHC class II molecule as previously described (11). Two different T cell preparations were used, 7A10B2 as described above and a T cell hybridoma derived from a mouse transgenic for the TCR of the T cell line 4B2A1. 7A10B2 and 4B2A1 have the same specificity (7). The dose-response curves are shown in Fig. 7. In all cases, whether measuring proliferation (Fig. 7A), IFN-{gamma} production (Fig. 7B), or IL-2 production (Fig. 7C), the loop 4CH2 mutant induced specific T cell activation at concentrations at least 102-fold lower than loop 6CH1, and detectable activation was found after injection of ~1 µg of loop 4CH2 rAb per mouse.


Figure 7
View larger version (13K):
[in this window]
[in a new window]

  		FIGURE 7. Induction of T cell activation after rAb injection in BALB/c mice and in vivo targeting. Titrated amounts of loop 4CH2 and loop 6CH1 rAbs were injected i.v. (two mice per point). After 90 min, the spleens were removed and the splenocytes were irradiated and used as APC in coculture with specific T cells: 7A10B2 for 72 h (A and B) or BW4B2A1 T cell hybridomas for 48 h (C). A, The cells were pulsed with [3H]dThd and T cell proliferation was measured as incorporation of radioactivity. B, IFN-{gamma} produced was measured by ELISA. C, IL-2 produced was measured by ELISA. Error bars indicate SD of triplicates.

 
Binding studies

The eight rAbs were tested for binding to Fc{gamma}RIIA and protein G. In general, IgGs bind ligands in either of two Fc locations, namely, the lower hinge (all Fc{gamma}R as well as C1q) or the elbow region between CH2 and CH3 (protein A, protein G, FcRn) (39). To investigate the integrity of these two regions in the mutants, they were tested in ELISA for binding to recombinant soluble Fc{gamma}RIIA and protein G, respectively. Mouse IgD was coated in the wells followed by the various IgD-specific rAbs. As was to be expected, the loop 6CH1 mutant bound both protein G and Fc{gamma}RIIA (Fig. 8, A and B). Although none of the other seven mutants bound Fc{gamma}RIIA, the loop 4CH2 and loop 6CH2 mutants bound protein G. The relative amount of rAb that was added to each well was visualized by Western blotting, detected with a hIgG3 hinge- specific Ab (Fig. 8C).


Figure 8
View larger version (19K):
[in this window]
[in a new window]

  		FIGURE 8. Ab binding protein G and soluble Fc{gamma}RIIA in ELISA. A and B, Mouse IgD was coat and IgD-specific rAb were added as well as protein G (A) and soluble recombinant Fc{gamma}RIIA (B). Error bars indicate SD of triplicates. The graphs are derived from one representative experiment of two. C, Western blot detection of each rAb preparation with a hIgG3 hinge-specific Ab. Mutants are named as in Figs. 2–4.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
We expected that substitution of constant region loops with a different amino acid sequence would influence the stability and, therefore, secretion, of the fusion molecules. In general, amino acids that are important for thermodynamic stability can be involved in packing of the hydrophobic core, charge interactions, hydrogen bonding, desolvation upon folding, and be more or less compatible with the enforced local structure (40). Regarding CH1, this domain interacts with CL through a large, well-packed hydrophobic interface that enhances the stability of both domains (34). Furthermore, the variability and surface accessibility of loop amino acids in the CH1 domain is notable. Therefore, it was not surprising that the mutants with substitution of loops in this domain were well secreted. An exception is the loop 2CH1 mutant that was completely retained. This is in agreement with previous results (9) and confirms that removal of the ball-and-socket joint (35) between VH and CH1 that involves this loop, is detrimental to secretion of hIgG3. A corresponding mutant in mIgG2b is secreted, however (9), and, therefore, the requirement for the interaction is not absolute. The amino acids involved in the CH1/CL interface are mostly situated in the A, B, D, and E strands, separated by loops 1 and 4, respectively. Substitution of these loops was less well tolerated than substitution of loops 3 and 6.

The CH2 domains do not physically interact with each other, they are heavily glycosylated on the Asn297 residue situated in loop 4, and the carbohydrate moiety normally fills the space between the domains. Mutating loop 4 decreased the secretion level 4-fold. However, all CH2 loop mutants were secreted in amounts between 20 and 60% of wt level, and the reduction may therefore not easily be attributed to removal of the carbohydrate. Of the CH2 loop mutants, that in loop 5 showed the lowest secretion, probably due to removal of the conserved and buried tryptophan in the native loop sequence.

The CH3 domain contains several amino acids that contribute to either the interface between CH2 and CH3 or dimerization through CH3-CH3 domain interaction and consequently have low accessibility scores. Variability analysis demonstrated the presence of many conserved residues in this domain, both in frameworks and in loops. The CH3-CH3 interface involves 16 residues that make interchain contacts (41). Of these, dimer stabilization is largely mediated by two loop 4 residues and four β strand residues. Thus, substitution of the involved loop 4CH3 may have destabilized the domain interface, whereas the other loop mutations may well have altered the configuration of β strands to reduce overall domain stability. It is therefore not surprising that the secretion of all but one mutant, namely, that involving loop 6, was very low. In this study, the new sequence is introduced at the outer edge of the domain and destabilizing effects introduced may have left the domain core mostly unaffected. The introduced 17 aa increased the loop length of all mutants from 3 to 13 aa. Those secreted in high amounts were all elongated from 10 to 13 aa (Table III). In a previous study, loop 6CH1 was exchanged with epitopes as long as 37 aa without interrupting secretion (42). Thus, long sequences may be introduced.

Proline residues at the framework/loop boundaries were defined as part of the loop by the Swiss-Pdb Viewer program and removed upon introduction of the 89–105 {lambda}2315 sequence. In one case, the loop 5CH1 mutant was made both with and without an N-terminal proline and the presence of proline improved secretion 4-fold (M. Flobakk, unpublished data). Thus, secretion may be further improved by keeping or possibly adding prolines to the framework/loop boundaries.

As we addressed ligand binding, we observed that the mutant with the new sequence in Fab bound both Fc{gamma}RIIA and protein G. None of the mutants with insert in the Fc region bound Fc{gamma}RIIA, whereas two bound protein G, namely, the loop 4CH2 and the loop 6CH1 mutants. A lack of Fc{gamma}R binding may be an advantage, as the function of IgG in the context of the studies reported here, is solely to target APC by way of V region specificity. Protein G binds in a region remote from the Fc{gamma}R binding site, namely, at the interface of CH2 and CH3 which is also the site of FcRn binding, and alterations here are likely to affect biodistribution and half-life of the mutant in vivo (36).

Specific growth inhibition and T cell proliferation assays were used to monitor peptide presentation on MHC class II and thus processing and loading of all mutants in vitro. The epitope was excised and presented to T cells from 13 of 18 positions in the rAb, and there were variations in activation efficiency. The mutants with substitution in loops of CH1 did not induce T cell activation, with the exception of loop 6CH1, which did so with a relatively low activity. All of the CH2 mutants induced with a high activity and, in particular, the mutant with epitope in loop 4, which was the better inducer in both the growth inhibition and T cell proliferation experiment. To induce the same in vitro effect, 100x less rAb of the loop 4CH2 than the loop 6CH1 variant was needed. The mutants with epitope in either of the CH3 domain loops activated at various levels. Of these, the loop 6CH3 mutant, which unlike the other CH3 mutants, was secreted well, had low in vitro activity, almost as low as loop 6CH1.

The CH1 domain seemed to be somewhat resistant to intracellular processing. This was not an effect of the epitope chosen, as the same result was obtained for the unrelated sequence of the OVA epitope, aa 323–339. Although an OVA323–339 loop 6CH1 mutant induced activation (10), an OVA323–326 loop 1CH1 mutant did not (M. Flobakk, unpublished data). Furthermore, we have previously observed very poor presentation of the {lambda}2315 sequence from loop 2 and 4CH1 in mIgG2, and the observation is therefore not restricted to hIgG3 (9) (Note that mutant terminology differs in this article.) In all cases were the CH2 domain mutants well secreted from producing cells and induced T cell activation at low concentrations. Again, the finding was not restricted to the amino acid sequence chosen for the systematic substitution experiments, as analogous fusion molecules with aa 110–120 from the hemagglutinin epitope in loop 1CH2 or aa 46–61 from the hen egg lysozyme in loop 6CH2 were well secreted and induced T cell activation in vitro at the same low rAb concentrations (M. Flobakk, I. B. Rasmussen, E. Lunde, T. E., Michaelsen, B. Bogen, and I. Sandlie, submitted for publication). In conclusion, a total of eight loop mutants were found to be useful for T cell activation purposes when both secretion from producing cells and the ability to induce activation in vitro were considered.

Of these, two were chosen for in vivo investigation, namely, loop 6CH1 and loop 4CH 2. One harbors the epitope in Fab and one in Fc and they have a remarkable 100-fold difference in T cell activation efficiency in vitro. Importantly, this difference was also seen in vivo. Others have demonstrated that stability is inversely related to processing and. thus, to presentation in vitro. In vivo, the situation was reversed and the stable molecule induced a stronger T cell response than the unstable variant (43, 44). In the experiment presented here, on the other hand, the mutants were actively targeted to APC in vivo. Because we have previously shown that presentation peaks when the targeted spleen cells are removed and tested for T cell activation ability after 1–2 h (11), it is reasonable to believe that both mutants are stable in serum until internalized.

We were intrigued by the fact that the CH1 domain mutants were poor inducers of T cell activation and speculated that there might not be recognition sites for proteolytic enzymes adjacent to the epitope in the CH1 domain. Although most proteases have broad specificity, making processing predictions difficult, AEP has restricted specificity (45). It cleaves at selected asparagine residues and unlocks the globular protein structure for MHC class II screening (46). There are 14 asparagine residues in the C{gamma}3 amino acid sequence (26), and the AEP cleavage server (http://theory.bio.uu.nl/keshmir/AEP) predicts five cleavage sites in the Fc region and only one in CH1. Thus, additional recognition sites in CH1 might improve the activation potential of the CH1 mutants. Notably, the 89–105 {lambda}2315 sequence also contains an asparagine residue (Asn96), but according to the AEP cleavage server, it is not susceptible to AEP cleavage regardless of its position within any CH domain.

Peptides presented on I-Ed are anchored at positions P1, P4, P6, and P9 (31). The anchor residues in 89–105 {lambda}2315 are Leu92, Phe94, Arg95, and Phe100 within the epitope (91–101). Competing I-Ed binding motifs could disturb 91–101 epitope loading. The PredBALB/c server (http://Ag.i2r.a-star.edu.sg/predBALBc/) predicts nonamer peptide binding to the H-2d haplotype of BALB/c mice (30). Focusing on nonamers with an equal to or higher score than 92–100, 25 peptides in C{gamma}3wt were identified that may compete for I-Ed binding, but there does not seem to be an obvious correlation between the location of these peptides and activation efficiency. Importantly, the 89–105 sequence was found to encompass an additional nonamer, namely, 97–105 (HFVFGGGTK) that fits better in I-Ed than 92–100. These two are mutually exclusive and this may explain the finding that the long peptide of 17 aa induced activation of aa 91–101-specific T cells at a lower level than the short peptide of only 11 aa in two different mutants (loop 4CH1 and loop 6CH1).

In the experiments shown here, APC were targeted by anti-mouse IgDa, i.e., to the Ag receptor of B cells (13). This is an endocytic receptor that directs Troybodies to intracellular vesicles where Ag processing and epitope loading on MHC class II occur. However, the B cell receptors are not optimal targets in vivo and a major disadvantage is the possibility for polyclonal activation of B cells (47). Alternative targets are surface molecules such as CD19 on B cells (48), MHC class II (48, 49), and CD14 (50). Receptors on dendritic cells may be even better targets since dendritic cells are able to prime naive T cells and induce TH, TC, and B cell responses. CD40 (51, 52), TLR (53, 54), CD91 (55), several chemokine receptors (56, 57), and C-type lectins, such as DC-SIGN (58, 59), mannose receptor (60), and DEC-205 (61, 62), have all been shown to be attractive candidates for targeting.

Our results are clear regarding an activation hierarchy in vitro and in vivo. rAbs with epitope loaded in the CH2 domain are secreted from producing cells and processed very efficiently after targeting to APC. Thus, whenever the goal is strong enhancement of T cell responses, epitope grafting in the CH2 domain seems preferable. For vaccination purposes, integration of several different sequences may be desirable. The results presented here suggest that multiple loop replacements within a single rAb may be a useful strategy using a combination of loop substitutions in two or possibly three domains.


   Acknowledgments
 
We are grateful to Hilde Omholt, Peter Hofgaard, and Karoline W. Schjetne for help during the T cell assays and Kristine Ustgård for Fc{gamma}RIIA.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
Inger Sandlie and Bjarne Bogen are coinventors on U.S. patent 6.249.654 granted September 25, 2001. Title: Modified immunoglobulin molecule incorporating antigen in a non-CDR loop region.


   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 work was funded by the Norwegian Cancer Society (to M.F. and I.B.R.) and the Research Council of Norway (to E.L., T.F., and G.B.). Back

2 Current address: BioInn, 2317 Hamar, Norway. Back

3 Address correspondence and reprint requests to Dr. Inger Sandlie, Department of Molecular Biosciences, University of Oslo, PO Box 1041 Blindern, NO-0316 Oslo, Norway. E-mail address: inger.sandlie{at}imbv.uio.no Back

4 Abbreviations used in this paper: CH, constant region of Ig H chain; RT, room temperature; AEP, asparaginyl endopeptidase; wt, wild type. Back

Received for publication July 11, 2007. Accepted for publication September 6, 2008.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Snider, D. P., D. M. Segal. 1987. Targeted antigen presentation using cross-linked antibody heteroaggregates. J. Immunol. 139: 1609-1616. [Abstract]
  2. Bona, C. A., S. Casares, T. D. Brumeanu. 1998. Towards development of T-cell vaccines. Immunol. Today 19: 126-133. [Medline]
  3. Baier, G., G. Baier-Bitterlich, D. J. Looney, A. Altman. 1995. Immunogenic targeting of recombinant peptide vaccines to human antigen- presenting cells by chimeric anti-HLA-DR and anti-surface immunoglobulin D antibody Fab fragments in vitro. J. Virol. 69: 2357-2365. [Abstract]
  4. Hawiger, D., K. Inaba, Y. Dorsett, M. Guo, K. Mahnke, M. Rivera, J. V. Ravetch, R. M. Steinman, M. C. Nussenzweig. 2001. Dendritic cells induce peripheral T cell unresponsiveness under steady state conditions in vivo. J. Exp. Med. 194: 769-779. [Abstract/Free Full Text]
  5. Trumpfheller, C., J. S. Finke, C. B. Lopez, T. M. Moran, B. Moltedo, H. Soares, Y. Huang, S. J. Schlesinger, C. G. Park, M. C. Nussenzweig, et al 2006. Intensified and protective CD4+ T cell immunity in mice with anti-dendritic cell HIV gag fusion antibody vaccine. J. Exp. Med. 203: 607-617. [Abstract/Free Full Text]
  6. Bogen, B., J. D. Lambris. 1989. Minimum length of an idiotypic peptide and a model for its binding to a major histocompatibility complex class II molecule. EMBO J. 8: 1947-1952. [Medline]
  7. Bogen, B., B. Malissen, W. Haas. 1986. Idiotope-specific T cell clones that recognize syngeneic immunoglobulin fragments in the context of class II molecules. Eur. J. Immunol. 16: 1373-1378. [Medline]
  8. Lunde, E., B. Bogen, I. Sandlie. 1997. Immunoglobulin as a vehicle for foreign antigenic peptides immunogenic to T cells. Mol. Immunol. 34: 1167-1176. [Medline]
  9. Eidem, J. K., I. B. Rasmussen, E. Lunde, T. F. Gregers, A. R. Rees, B. Bogen, I. Sandlie. 2000. Recombinant antibodies as carrier proteins for sub-unit vaccines: influence of mode of fusion on protein production and T-cell activation. J. Immunol. Methods 245: 119-131. [Medline]
  10. Rasmussen, I. B., E. Lunde, T. E. Michaelsen, B. Bogen, I. Sandlie. 2001. The principle of delivery of T cell epitopes to antigen-presenting cells applied to peptides from influenza virus, ovalbumin, and hen egg lysozyme: implications for peptide vaccination. Proc. Natl. Acad. Sci. USA 98: 10296-10301. [Abstract/Free Full Text]
  11. Lunde, E., L. A. Munthe, A. Vabo, I. Sandlie, B. Bogen. 1999. Antibodies engineered with IgD specificity efficiently deliver integrated T-cell epitopes for antigen presentation by B cells. Nat. Biotechnol. 17: 670-675. [Medline]
  12. Lunde, E., K. H. Western, I. B. Rasmussen, I. Sandlie, B. Bogen. 2002. Efficient delivery of T cell epitopes to APC by use of MHC class II-specific Troy bodies. J. Immunol. 168: 2154-2162. [Abstract/Free Full Text]
  13. Tisch, R., M. Watanabe, M. Letarte, N. Hozumi. 1987. Assessment of antigen-specific receptor function of surface immunoglobulin M and D with identical hapten specificity. Proc. Natl. Acad. Sci. USA 84: 3831-3835. [Abstract/Free Full Text]
  14. Bogen, B., L. Gleditsch, S. Weiss, Z. Dembic. 1992. Weak positive selection of transgenic T cell receptor-bearing thymocytes: importance of major histocompatibility complex class II, T cell receptor and CD4 surface molecule densities. Eur. J. Immunol. 22: 703-709. [Medline]
  15. White, J., M. Blackman, J. Bill, J. Kappler, P. Marrack, D. P. Gold, W. Born. 1989. Two better cell lines for making hybridomas expressing specific T cell receptors. J. Immunol. 143: 1822-1825. [Abstract]
  16. Prat, M., G. Gribaudo, P. M. Comoglio, G. Cavallo, S. Landolfo. 1984. Monoclonal antibodies against murine {gamma} interferon. Proc. Natl. Acad. Sci. USA 81: 4515-4519. [Abstract/Free Full Text]
  17. Cherwinski, H. M., J. H. Schumacher, K. D. Brown, T. R. Mosmann. 1987. Two types of mouse helper T cell clone. III. Further differences in lymphokine synthesis between Th1 and Th2 clones revealed by RNA hybridization, functionally monospecific bioassays, and monoclonal antibodies. J. Exp. Med. 166: 1229-1244. [Abstract/Free Full Text]
  18. Fjeld, J. G., T. E. Michaelsen, K. Nustad. 1992. The binding parameters of radiolabelled monoclonal F(ab')2 and Fab' fragments relative to immunoglobulin G in reactions with surface-bound antigens. Eur. J. Nucl. Med. 19: 402-408. [Medline]
  19. Jefferis, R., C. B. Reimer, F. Skvaril, G. de Lange, N. R. Ling, J. Lowe, M. R. Walker, D. J. Phillips, C. H. Aloisio, T. W. Wells, et al 1985. Evaluation of monoclonal antibodies having specificity for human IgG sub-classes: results of an IUIS/WHO collaborative study. Immunol. Lett. 10: 223-252. [Medline]
  20. Aaberge, I. S., T. E. Michaelsen, A. K. Rolstad, E. C. Groeng, P. Solberg, M. Lovik. 1992. SCID-Hu mice immunized with a pneumococcal vaccine produce specific human antibodies and show increased resistance to infection. Infect. Immun. 60: 4146-4153. [Abstract/Free Full Text]
  21. Michaelsen, T. E., A. Aase, C. Westby, I. Sandlie. 1990. Enhancement of complement activation and cytolysis of human IgG3 by deletion of hinge exons. Scand. J. Immunol. 32: 517-528. [Medline]
  22. Berman, H. M., J. Westbrook, Z. Feng, G. Gilliland, T. N. Bhat, H. Weissig, I. N. Shindyalov, P. E. Bourne. 2000. The Protein Data Bank. Nucleic Acids Res. 28: 235-242. [Abstract/Free Full Text]
  23. Saphire, E. O., R. L. Stanfield, M. D. Crispin, G. Morris, M. B. Zwick, R. A. Pantophlet, P. W. Parren, P. M. Rudd, R. A. Dwek, D. R. Burton, I. A. Wilson. 2003. Crystal structure of an intact human IgG: antibody asymmetry, flexibility, and a guide for HIV-1 vaccine design. Adv. Exp. Med. Biol. 535: 55-66. [Medline]
  24. Gasteiger, E., C. Hoogland, A. Gattiker, S. Duvaud, M. R. Wilkins, R. D. Appel, A. Bairoch. 2005. Protein identification and analysis tools on the ExPASy. John M. Walker, ed. The Proteomics Protocols Handbook 571-607. Humana Press, Totowa, NJ.
  25. Kyte, J., R. F. Doolittle. 1982. A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157: 105-132. [Medline]
  26. Huck, S., P. Fort, D. H. Crawford, M. P. Lefranc, G. Lefranc. 1986. Sequence of a human immunoglobulin {gamma} 3 heavy chain constant region gene: comparison with the other human C {gamma} genes. Nucleic Acids Res. 14: 1779-1789. [Abstract/Free Full Text]
  27. Norderhaug, L., T. Olafsen, T. E. Michaelsen, I. Sandlie. 1997. Versatile vectors for transient and stable expression of recombinant antibody molecules in mammalian cells. J. Immunol. Methods 204: 77-87. [Medline]
  28. Berntzen, G., E. Lunde, M. Flobakk, J. T. Andersen, V. Lauvrak, I. Sandlie. 2005. Prolonged and increased expression of soluble Fc receptors, IgG and a TCR-Ig fusion protein by transiently transfected adherent 293E cells. J. Immunol. Methods 298: 93-104. [Medline]
  29. Berntzen, G., O. H. Brekke, S. A. Mousavi, J. T. Andersen, T. E. Michaelsen, T. Berg, I. Sandlie, V. Lauvrak. 2006. Characterization of an Fc{gamma}RI-binding peptide selected by phage display. Protein Eng. Des. Sel. 19: 121-128. [Abstract/Free Full Text]
  30. Zhang, G. L., K. N. Srinivasan, A. Veeramani, J. T. August, V. Brusic. 2005. PREDBALB/c: a system for the prediction of peptide binding to H2d molecules, a haplotype of the BALB/c mouse. Nucleic Acids Res. 33: W180-W183. [Abstract/Free Full Text]
  31. Schild, H., U. Gruneberg, G. Pougialis, H. J. Wallny, W. Keilholz, S. Stevanovic, H. G. Rammensee. 1995. Natural ligand motifs of H-2E molecules are allele specific and illustrate homology to HLA-DR molecules. Int. Immunol. 7: 1957-1965. [Abstract/Free Full Text]
  32. Ellgaard, L., A. Helenius. 2003. Quality control in the endoplasmic reticulum. Nat. Rev. Mol. Cell Biol. 4: 181-191. [Medline]
  33. Bothwell, A. L., M. Paskind, M. Reth, T. Imanishi-Kari, K. Rajewsky, D. Baltimore. 1982. Somatic variants of murine immunoglobulin {lambda} light chains. Nature 298: 380-382. [Medline]
  34. Rothlisberger, D., A. Honegger, A. Pluckthun. 2005. Domain interactions in the Fab fragment: a comparative evaluation of the single-chain Fv and Fab format engineered with variable domains of different stability. J. Mol. Biol. 347: 773-789. [Medline]
  35. Lesk, A. M., C. Chothia. 1988. Elbow motion in the immunoglobulins involves a molecular ball-and-socket joint. Nature 335: 188-190. [Medline]
  36. Kim, J. K., M. Firan, C. G. Radu, C. H. Kim, V. Ghetie, E. S. Ward. 1999. Mapping the site on human IgG for binding of the MHC class I-related receptor, FcRn. Eur. J. Immunol. 29: 2819-2825. [Medline]
  37. Sondermann, P., R. Huber, V. Oosthuizen, U. Jacob. 2000. The 3.2-Å crystal structure of the human IgG1 Fc fragment-Fc {gamma}RIII complex. Nature 406: 267-273. [Medline]
  38. Idusogie, E. E., L. G. Presta, H. Gazzano-Santoro, K. Totpal, P. Y. Wong, M. Ultsch, Y. G. Meng, M. G. Mulkerrin. 2000. Mapping of the C1q binding site on rituxan, a chimeric antibody with a human IgG1 Fc. J. Immunol. 164: 4178-4184. [Abstract/Free Full Text]
  39. Nezlin, R., V. Ghetie. 2004. Interactions of immunoglobulins outside the antigen-combining site. Adv. Immunol. 82: 155-215. [Medline]
  40. Ewert, S., A. Honegger, A. Pluckthun. 2003. Structure-based improvement of the biophysical properties of immunoglobulin VH domains with a generalizable approach. Biochemistry 42: 1517-1528. [Medline]
  41. Dall'Acqua, W., A. L. Simon, M. G. Mulkerrin, P. Carter. 1998. Contribution of domain interface residues to the stability of antibody CH3 domain homodimers. Biochemistry 37: 9266-9273. [Medline]
  42. Tunheim, G., K. W. Schjetne, I. B. Rasmussen, L. M. Sollid, I. Sandlie, B. Bogen. 2008. Recombinant antibodies for delivery of antigen: a single loop between β-strands in the constant region can accommodate long, complex and tandem T cell epitopes. Int. Immunol. 20: 295-306. [Abstract/Free Full Text]
  43. Thai, R., G. Moine, M. Desmadril, D. Servent, J. L. Tarride, A. Menez, M. Leonetti. 2004. Antigen stability controls antigen presentation. J. Biol. Chem. 279: 50257-50266. [Abstract/Free Full Text]
  44. So, T., H. O. Ito, T. Koga, S. Watanabe, T. Ueda, T. Imoto. 1997. Depression of T-cell epitope generation by stabilizing hen lysozyme. J. Biol. Chem. 272: 32136-32140. [Abstract/Free Full Text]
  45. Chen, J. M., P. M. Dando, N. D. Rawlings, M. A. Brown, N. E. Young, R. A. Stevens, E. Hewitt, C. Watts, A. J. Barrett. 1997. Cloning, isolation, and characterization of mammalian legumain, an asparaginyl endopeptidase. J. Biol. Chem. 272: 8090-8098. [Abstract/Free Full Text]
  46. Manoury, B., E. W. Hewitt, N. Morrice, P. M. Dando, A. J. Barrett, C. Watts. 1998. An asparaginyl endopeptidase processes a microbial antigen for class II MHC presentation. Nature 396: 695-699. [Medline]
  47. Finkelman, F. D., I. Scher, J. J. Mond, S. Kessler, J. T. Kung, E. S. Metcalf. 1982. Polyclonal activation of the murine immune system by an antibody to IgD: II. Generation of polyclonal antibody production and cells with surface IgG. J. Immunol. 129: 638-646. [Medline]
  48. Yan, J., M. J. Wolff, J. Unternaehrer, I. Mellman, M. J. Mamula. 2005. Targeting antigen to CD19 on B cells efficiently activates T cells. Int. Immunol. 17: 869-877. [Abstract/Free Full Text]
  49. Casten, L. A., S. K. Pierce. 1988. Receptor-mediated B cell antigen processing: increased antigenicity of a globular protein covalently coupled to antibodies specific for B cell surface structures. J. Immunol. 140: 404-410. [Abstract]
  50. Tunheim, G., K. W. Schjetne, A. B. Fredriksen, I. Sandlie, B. Bogen. 2005. Human CD14 is an efficient target for recombinant immunoglobulin vaccine constructs that deliver T cell epitopes. J. Leukocyte Biol. 77: 303-310. [Abstract/Free Full Text]
  51. Tillman, B. W., T. D. de Gruijl, S. A. Luykx-de Bakker, R. J. Scheper, H. M. Pinedo, T. J. Curiel, W. R. Gerritsen, D. T. Curiel. 1999. Maturation of dendritic cells accompanies high-efficiency gene transfer by a CD40-targeted adenoviral vector. J. Immunol. 162: 6378-6383. [Abstract/Free Full Text]
  52. Schjetne, K. W., A. B. Fredriksen, B. Bogen. 2007. Delivery of antigen to CD40 induces protective immune responses against tumors. J. Immunol. 178: 4169-4176. [Abstract/Free Full Text]
  53. Schjetne, K. W., K. M. Thompson, N. Nilsen, T. H. Flo, B. Fleckenstein, J. G. Iversen, T. Espevik, B. Bogen. 2003. Cutting edge: link between innate and adaptive immunity: Toll-like receptor 2 internalizes antigen for presentation to CD4+ T cells and could be an efficient vaccine target. J. Immunol. 171: 32-36. [Abstract/Free Full Text]
  54. Jackson, D. C., Y. F. Lau, T. Le, A. Suhrbier, G. Deliyannis, C. Cheers, C. Smith, W. Zeng, L. E. Brown. 2004. A totally synthetic vaccine of generic structure that targets Toll-like receptor 2 on dendritic cells and promotes antibody or cytotoxic T cell responses. Proc. Natl. Acad. Sci. USA 101: 15440-15445. [Abstract/Free Full Text]
  55. Hart, J. P., M. D. Gunn, S. V. Pizzo. 2004. A CD91-positive subset of CD11c+ blood dendritic cells: characterization of the APC that functions to enhance adaptive immune responses against CD91-targeted antigens. J. Immunol. 172: 70-78. [Abstract/Free Full Text]
  56. Schjetne, K. W., H. T. Gundersen, J. G. Iversen, K. M. Thompson, B. Bogen. 2003. Antibody-mediated delivery of antigen to chemokine receptors on antigen-presenting cells results in enhanced CD4+ T cell responses. Eur. J. Immunol. 33: 3101-3108. [Medline]
  57. Biragyn, A., K. Tani, M. C. Grimm, S. Weeks, L. W. Kwak. 1999. Genetic fusion of chemokines to a self tumor antigen induces protective, T-cell dependent antitumor immunity. Nat. Biotechnol. 17: 253-258. [Medline]
  58. Schjetne, K. W., K. M. Thompson, T. Aarvak, B. Fleckenstein, L. M. Sollid, B. Bogen. 2002. A mouse C {kappa}-specific T cell clone indicates that DC-SIGN is an efficient target for antibody-mediated delivery of T cell epitopes for MHC class II presentation. Int. Immunol. 14: 1423-1430. [Abstract/Free Full Text]
  59. Tacken, P. J., I. J. de Vries, K. Gijzen, B. Joosten, D. Wu, R. P. Rother, S. J. Faas, C. J. Punt, R. Torensma, G. J. Adema, C. G. Figdor. 2005. Effective induction of naive and recall T-cell responses by targeting antigen to human dendritic cells via a humanized anti-DC-SIGN antibody. Blood 106: 1278-1285. [Abstract/Free Full Text]
  60. He, L. Z., V. Ramakrishna, J. E. Connolly, X. T. Wang, P. A. Smith, C. L. Jones, M. Valkova-Valchanova, A. Arunakumari, J. F. Treml, J. Goldstein, et al 2004. A novel human cancer vaccine elicits cellular responses to the tumor-associated antigen, human chorionic gonadotropin β. Clin. Cancer Res. 10: 1920-1927. [Abstract/Free Full Text]
  61. Mahnke, K., Y. Qian, S. Fondel, J. Brueck, C. Becker, A. H. Enk. 2005. Targeting of antigens to activated dendritic cells in vivo cures metastatic melanoma in mice. Cancer Res. 65: 7007-7012. [Abstract/Free Full Text]
  62. Bonifaz, L. C., D. P. Bonnyay, A. Charalambous, D. I. Darguste, S. Fujii, H. Soares, M. K. Brimnes, B. Moltedo, T. M. Moran, R. M. Steinman. 2004. In vivo targeting of antigens to maturing dendritic cells via the DEC-205 receptor improves T cell vaccination. J. Exp. Med. 199: 815-824. [Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Flobakk, M.
Right arrow Articles by Sandlie, I.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Flobakk, M.
Right arrow Articles by Sandlie, I.
Right arrowPubmed/NCBI databases
*Substance via MeSH

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS