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


     
 

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 HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Godkin, A. J.
Right arrow Articles by Hill, A. V. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Godkin, A. J.
Right arrow Articles by Hill, A. V. S.
The Journal of Immunology, 2001, 166: 6720-6727.
Copyright © 2001 by The American Association of Immunologists

Naturally Processed HLA Class II Peptides Reveal Highly Conserved Immunogenic Flanking Region Sequence Preferences That Reflect Antigen Processing Rather Than Peptide-MHC Interactions1

Andrew J. Godkin2,*, Kathrine J. Smith{dagger}, Anthony Willis{ddagger}, Maria V. Tejada-Simon§, Jingwu Zhang§, Tim Elliott* and Adrian V. S. Hill*

* Nuffield Department of Medicine, Molecular Immunology Group, Institute of Molecular Medicine, University of Oxford, John Radcliffe Hospital, Oxford, United Kingdom; {dagger} Department of Medicine, Laboratory of Molecular Medicine, The Children’s Hospital, Howard Hughes Medical Institute, Boston, MA 02115; {ddagger} Department of Biochemistry, Medical Research Council Immunochemistry Unit, Oxford, United Kingdom; § Department of Neurology, Baylor College of Medicine, Houston, TX 77030; and Department of Medicine, Imperial College, London, U.K.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
MHC class II heterodimers bind peptides 12–20 aa in length. The peptide flanking residues (PFRs) of these ligands extend from a central binding core consisting of nine amino acids. Increasing evidence suggests that the PFRs can alter the immunogenicity of T cell epitopes. We have previously noted that eluted peptide pool sequence data derived from an MHC class II Ag reflect patterns of enrichment not only in the core binding region but also in the PFRs. We sought to distinguish whether these enrichments reflect cellular processes or direct MHC-peptide interactions. Using the multiple sclerosis-associated allele HLA-DR2, pool sequence data from naturally processed ligands were compared with the patterns of enrichment obtained by binding semicombinatorial peptide libraries to empty HLA-DR2 molecules. Naturally processed ligands revealed patterns of enrichment reflecting both the binding motif of HLA-DR2 (position (P)1, aliphatic; P4, bulky hydrophobic; and P6, polar) as well as the nonbound flanking regions, including acidic residues at the N terminus and basic residues at the C terminus. These PFR enrichments were independent of MHC-peptide interactions. Further studies revealed similar patterns in nine other HLA alleles, with the C-terminal basic residues being as highly conserved as the previously described N-terminal prolines of MHC class II ligands. There is evidence that addition of C-terminal basic PFRs to known peptide epitopes is able to enhance both processing as well as T cell activation. Recognition of these allele-transcending patterns in the PFRs may prove useful in epitope identification and vaccine design.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Human histocompatibility leukocyte antigens glycoproteins present peptides to cognate T cells, leading to specific immune activation. HLA class I molecules bind peptides that are usually 8 or 9 aa in length with the N- and C-termini bound in conserved pockets at either end of a groove. In contrast, HLA class II molecules tend to bind nested sets of longer peptides, 12–20 aa in length. These longer peptides bind with a core region of similar length, with the nonbound flanking residues of the peptide extending from the ends of an open groove (1). The side chains of these core residues bind in a series of pockets along the groove, spaced from the bound N terminus of the peptide at position 1 (P1),3 P4, P6, P7, and P9 (2, 3, 4, 5). In HLA-DR molecules, the pockets in the groove are lined with the polymorphic residues of the HLA-DR {beta}-chain, generating specific binding motifs for individual HLA-types and subtypes. Eluted peptide sequence data reveal patterns of enrichment of amino acids, which reflect the binding requirements of these allele-specific pockets (6, 7, 8, 9).

The peptide flanking residues (PFRs) not bound in the central core region can interact with the HLA class II heterodimer and enhance the binding affinity and stability/conformation of the trimolecular complex (8, 10, 11). However, it is becoming increasingly clear that the PFRs may also have a more specific role in immune recognition (12). TCR activation can be influenced by these PFRs as well as the residues in the core binding region (13, 14). Furthermore, we have shown that the additional information in the PFRs is necessary in certain instances for accurate prediction of T cell epitopes from the parent protein, possibly reflecting patterns of amino acid enrichments in these flanking regions distinct from the core binding region (9).

Initially, to examine further the generation of flanking regions, the binding of peptides to the heterodimer encoded by HLA-DRA1*0101, B1*1501 (HLA-DR2(b)) was studied. HLA-DR2 has been associated with susceptibility to multiple sclerosis (15). In the context of multiple sclerosis, the interaction of the immunodominant myelin basic protein (MBP) peptide 84–102 (DENPVVHFFKNIVTPRTPP) has been well characterized (14, 16, 17). The core binding region is 89–97, with two important anchors at P1 and P4 (Val89 and Phe92). The preferences of other anchor positions in HLA-DR2 are not so clear. These anchors alone are not enough to ensure high affinity binding, as demonstrated by the poor binding of the 9-aa core peptide compared with a peptide with the addition of either N- or C-terminal flanking residues. These flanking residues may also influence the activation of T cells. The majority of CD4+ T cell clones interact with the residues in the core region His90, Phe91, and Lys93. However, many clones are also sensitive to truncations or substitutions of the flanking residues in the MBP peptide (14, 18, 19). The altered T cell activation does not correlate with a change in binding affinities. It appears that the TCR is either directly interacting with the PFRs or the alterations are changing the physical contours of the core peptide.

The binding of peptides to HLA-DR2 was studied by a combination of eluted peptide sequence data, in vitro peptide binding assays, and combinatorial peptide technology. Complete eluted peptide sequence data identified both the nature and position of anchor residues in the core binding region of HLA-DR2, as well as the characteristics of the N- and C-terminal PFRs that appeared to be highly conserved among nine additional HLA alleles studied. Using in vitro binding assays and semirandom combinatorial peptide libraries bound to empty HLA-DR2 molecules purified from an insect cell expression system, the following questions were addressed. First, is there a difference between the patterns seen in the termini of the natural processed peptides bound in vivo and synthetic peptides selected in vitro; second, does the core binding region of the peptide influence these termini; and finally, do naturally occurring residues in the PFRs enhance binding in vitro and offer an explanation for any enrichments found?


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Peptide synthesis

The candidate peptides were synthesized by F-moc chemistry using an automated synthesizer (Apex 396 multiple peptide synthesizer; Advanced ChemTech, Cambridge, U.K.). The purity was assessed by reversed phase HPLC, and the concentration was measured using a bicinchoninic acid assay. The sequence XXXVVHFFKNIVXXXA was synthesized corresponding to the MBP core peptide binding sequence with random termini. X represents any of 19 amino acids (cysteine not incorporated). For the X position, all 19 amino acids were mixed into a single well of the peptide synthesizer block. The other semirandom peptides were synthesized with modified methodology as previously described (20) to encourage equimolar incorporation of amino acids in the random positions (Echaz Microcollections, Tübingen, Germany). In brief, double couplings were performed with equimolar mixtures of F-moc-L-amino acids, which were used in an equimolar ratio with respect to the coupling sites of the resins, and extended coupling times were used. The following sequences were made: XXXXVHAFKNIVXXXX and XXXXIHQLKNVIXXXX. The first MBP sequence was modified (Ala for Phe at position 91). This was to discourage any possible sliding of the peptide binding by the use of F91 as an anchor rather than the usual F92. The second sequence contained a different core binding peptide to ensure that bias was not introduced by a particular sequence.

Purification of naturally produced HLA class II molecules

The method used was based on the original description by Gorga et al. (21). EBV-transformed HLA-homozygous B cell lines (obtained from the European Collection of Cell Culture, Salisbury, U.K.) were grown for each HLA-type. Cells were grown up to a final concentration of 1–2 x 106 cells/ml in 2 L roller bottles with RPMI 1640 supplemented by 10% FCS, L-glutamine (10 mM), penicillin (100 IU/ml), and streptomycin (50 µg/ml). Between 1011 and 1012 cells were spun down, washed in cold PBS, and then lysed on ice with 3% Nonidet P-40 in PBS containing leupeptin (1 µg/ml), pepstatin (1 µg/ml), and 5 mM EDTA. The lysate was spun at 100,000 x g for 90 min. The supernatant was passed over a precolumn of Sepharose CL 4B followed by the affinity column with cyanogen bromide-activated Sepharose beads linked to the monoclonal Ab to HLA-DR L243 (22). After washing, the class II molecules were eluted from each column with 0.05 M diethylamine (pH 11.5). The eluates were immediately neutralized with 1 M Tris (pH 6.8) and concentrated by ultrafiltration (Centriprep; Amicon, Beverly, MA). The purity was analyzed by 12% SDS-PAGE.

Purification of HLA-DR2 molecules from Schneider cells and binding to semirandom libraries

Drosophila (S2) cells transfected with HLA-DR2 (DRA1*0101, B1*1501) (constructs given in Ref. 5) were grown in spinner flasks in ExCell 401 medium (Sigma, St. Louis, MO) supplemented with 1% FBS. At a density of 4 x 106 cells/ml, protein expression was induced with 1 mM CuSO4. The supernatant was harvested after 6 days by centrifugation at 4°C. HLA-DR2 was purified from the supernatant by affinity chromatography; DR2 was eluted from the column with 50 mM glycine (pH 11.5), and fractions were neutralized with 1 M Tris (pH 7.0). For peptide binding, 2 mg of each random peptide was dissolved in 2.2-M urea, and 100 µl was removed for analysis. The remainder was bound to 5 mg of empty HLA-DR2 at 37°C for 60 h. After peptide binding, the HLA-DR2 was concentrated using centricons (Amicon), and DR2-peptide complexes were purified by gel filtration using a Superdex 200 column (Amersham Pharmacia Biotech, Little Chalfont, U.K.) equilibrated in PBS.

Pool Sequencing

Naturally processed peptide ligands (as well as synthetic random peptides) were released from the HLA molecules by acid elution with 0.1% trifluoroacetic acid. The HLA molecules were removed from the peptides by ultrafiltration using a centricon microconcentrator (Amicon). The peptides were separated on an Aquapore RP-300 (C8/100 x 2 mm) reversed phase column (Perkin-Elmer Applied Biosystems, Warrington, U.K.) connected to a Perkin-Elmer 172A narrow-bore HPLC system. The column was equilibrated in 0.1% trifluoroacetic acid, and the peptides were eluted with a linear gradient of acetonitrile increasing at 1% per min. The fractions containing peptides were pooled and sequenced for 15 cycles on an ABI 473A or an ABI Procise 494A protein sequencer (Perkin-Elmer Applied Biosystems). Any disproportionately large peaks were removed so as not to bias the sequencing of the rest of the pool. Sequencing data was integrated using Model 610A (version 2.1a) data analysis software obtained from Perkin-Elmer Applied Biosystems.

Epitope prediction

Total yields of amino acids diminish over each cycle of Edman degradation due to the nature of this process. To allow for this, the pool sequence data were organized using a procedure previously described to generate a matrix of relative enrichments (23). Briefly, the amount of each amino acid at each cycle was described as a percentage of the total quantity of all amino acids in that cycle. The average of each amino acid was then taken over the 15 cycles, and each position in the cycle was expressed as a proportion of the average. The resulting table was used with a computer program that scans given protein sequences iteratively and calculates for each 15 mer a probability of being processed and presented score. The protein sequences that were scanned were obtained from the Swiss-Prot and Protein Information Resource databases.

Peptide binding assays

Binding of synthetic peptides to purified HLA-DR2 molecules was assayed in vitro in a competitive inhibition assay against the promiscuously binding class II-associated invariant chain peptide (CLIP) (96–114), which was biotinylated. All incubations were conducted at 37°C. The 96-well flat-bottom ELISA immunoplate was blocked with PBS containing 3% BSA for 3 h. The plate was emptied, and serial 1/10 dilutions of candidate peptide were then added to the wells with 0.02 µg biotinylated marker CLIP and 0.1 µg HLA-DR2 in pH 5 buffer (1.95 g 2-(morpholino)ethanesulfonic acid + 0.5 g 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate + 4.1 g NaCl + 0.02% azide in 500 ml of water) for a total volume of 60 µl/well. The plate was incubated for 18 h. After this time, 10 µl of neutralizing buffer (1 M Tris (pH 8) + 10% BSA + 1% Tween 20 + 0.02% azide) was added to each well. The 70-µl aliquots were transferred to a new 96-well plate (precoated with anti-DR Ab (L243) and blocked as described above with PBS containing 3% BSA) and incubated for 1 h. The wells were emptied, washed extensively with PBS containing 1% Tween 20, and then washed with PBS alone. A total of 50 µl of 1/1000 avidin-HRP solution (Sigma) was added to each well and incubated for 15 min. The plate was emptied, washed, and dried as described above. A total of 50 µl/well of 1/1000 biotinylated anti-avidin solution (Sigma) was added and incubated for 30 min. The plate was again emptied, washed, and dried. A total of 50 µl/well of avidin-HRP solution was then added and incubated for 15 min.

The plate was emptied and washed with PBS with 1% Tween 20, then washed again with H20. After the plate was dried, 100 µl of freshly made developing solution was added (5 ml of 0.1 M citrate buffer (pH 9) + 5 ml of 0.2 M phosphate buffer (pH 5) + 100 µl o-phenylenediamine at 40 mg/ml + 20 µl hydrogen peroxide). The plate was wrapped in silver foil and incubated (without stacking) for 30–120 min (until color changes were clearly visible). Then the plate was blocked with 100 µl of 12.5% H2SO4, and the color changes were measured at 492 nm in an ELISA reader.

The IC50 of unlabeled peptide required to inhibit binding of the CLIP was calculated as a concentration (µM). All the binding assays were performed at least twice and a mean for the IC50 calculated.

IFN-{gamma} enzyme-linked immunospot (ELISPOT) assays

The biological effect of peptides on T cells was studied in a direct ex vivo assay using single cell IFN-{gamma} release as a measure of Ag-specific effector function in a highly sensitive ELISPOT assay (Abs and streptavidin-alkaline phosphatase were obtained from Mabtech, Nacka, Sweden). PBMCs were extracted from whole blood by centrifugation over Lymphoprep (Nycomed, Oslo, Norway). A total of 5 x 105 cells in RPMI 1640 with 5% heat-treated FCS were added per well of a special polymer-backed 96-well filtration plate MAIP-S-45 (Millipore, Moslheim, France). The wells were precoated with monoclonal anti-IFN-{gamma} (1-D1K) according to the manufacturer’s instructions. The peptides were tested in duplicate wells, having been added to a final concentration of 10 µg/ml, and compared with control wells with no peptide. The plate was incubated at 37°C, 5% CO2 for 18 h. The plate was then washed with PBS and incubated with a second layer biotinylated anti-IFN-{gamma} (7-B6-1; 1/1000 in PBS, 50 ml/well) for 2 h at room temperature. The plate was rinsed and developed with streptavidin-alkaline phosphatase and a chromogenic substrate. When the spots had developed, the reaction was stopped by washing with cold tap water. The plate was thoroughly dried, and activated T cells were enumerated at the single-cell level by counting the number of spots per well. The plates were assessed by an independent observer blinded to the well contents.

Generation of short-term T cell lines

A total of 2 x 105 PBMCs were plated per well in a 96-well round-bottom plate with 10 µg/ml of peptide, using RPMI 1640 plus 10% human AB serum plus an antibiotics/L-glutamine supplement. PBMCs were then incubated in a humidified atmosphere containing 5% CO2 at 37°C. After 7 days, all cells were restimulated with peptide-pulsed irradiated autologous PBMCs as a source of APC. After another 7 days, each well was tested for peptide-specific proliferation in a split-well assay. Duplicate wells containing 105 T cells and 5 x 105 irradiated autologous PBMCs (± pulsing with peptide followed by washing) were incubated for 72 h. Specific proliferation was measured by [3H]thymidine incorporation over the final 17 h. A T cell line was considered specific when the stimulation index was >3 and the cpm were >1500. Occasionally, IL-2 was added to the lines on day 9 (10% Lymphocult-T; Biotest U.K. Ltd., Solihull, U.K.).

Generation of MBP 83–99-specific CD4+ T cell clones

A positive T cell line was generated from a patient with multiple sclerosis, as described above. The line was plated out at 0.3 cells/well in the presence of 105 irradiated autologous PBMCs and 2 µg/ml PHA. Cultures were fed with fresh medium containing 50 IU/ml rIL-2 every 3–4 days. Growth-positive cells were tested in a proliferation assay after 10–12 days to identify Ag-specific clones. For further proliferation assays against MBP peptides and analogs, transfected L cells (HLA-DR2) were used. Irradiated L cells (2.5 x 105/well) were incubated in a microtiter plate for 2–3 h, pulsed with peptides for 1–2 h, and then washed. Cultures were set up in duplicate for each peptide, and [3H]thymidine incorporation measured after 72 h


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Pool sequence data from naturally processed peptides reveals enrichments in both the core binding region and the PFRs

Naturally processed peptides were eluted from multiple batches of affinity purified HLA class II molecules and sequenced en masse. Each pool was obtained from ~1 mg of purified protein, and, for HLA-DR2, an example of one pool is shown in Table IGo. To allow for the diminishing yield seen over each cycle of degradation and to emphasize changes in each amino acid, the data was organized as described in Materials and Methods to generate a table showing the relative enrichments of the amino acids over 15 cycles (Table IIGo). As has been previously described, the pattern of enrichments reflect the binding preferences of the pockets starting with cycles 4/5 corresponding approximately to the first anchor position P1 (6, 24). In keeping with other studied HLA-DR molecules with a Val at position 86 of the {beta}-chain (25), there was clearly a hydrophobic pocket, P1, corresponding to the first position in the binding core, which accommodates aliphatic hydrophobic side chains, reflected by the enrichment of Ile, Leu, and Val. Two other anchor positions emerged, P4, which favors large bulky aromatic hydrophobic residues, and P6, which favors polar residues. The striking enrichment of Tyr and Phe in cycles 7 and 8 suggests that this is a key pocket and may account for increases seen in earlier cycles. The features of this pool sequencing data correspond with the described crystal structure of HLA-DR2 (5). This structure revealed the hydrophobic pockets at P1 and the additional large deep hydrophobic pocket at P4, and, although a small pocket was identified at P9, the amino acid side chain did not occupy this pocket. It appears that the key pockets for HLA-DR2 are the P1 and P4 pockets, confirming previous reports (14, 16), with the pocket at P6 acting as an additional anchor.


View this table:
[in this window]
[in a new window]
  		Table I. An example of pool sequence data for HLA-DR21

 

View this table:
[in this window]
[in a new window]
  		Table II. Matrix derived from pool peptide sequence data from HLA-DR21

 
Table IIGo also reveals marked patterns of enrichment in cycles that do not correspond to residues binding in the HLA groove but to the PFRs. The early cycles, 1–3, favored the acidic residues Asp and Glu, and Pro was increased in cycle 2, which probably reflects aminopeptidase activity (26). Ile and Leu were increased in cycle 1, and this may reflect the cleavage site of the Ag-processing enzymes cathepsin D and E (27). The later cycles, 14–15, showed a marked preference for the basic residues His, Arg, and Lys, along with Gln.

To address the question of whether these PFR enrichments were allele specific or not, we purified a series of different HLA class II molecules and eluted and sequenced the peptides as described above (HLA-DR3, -DR4(B1*0407), -DR7, -DR11, -DR13, -DQ2, and -DQ8) (our unpublished data and Refs. 8, 9, 24, 28). The data from HLA-DR1 and -DR4(B1*0401) has been published by another group (6, 7). The PFR enrichments revealed highly conserved features, which are summarized in Table IIIGo. The Pro at P-2 (already recognized as a feature of pool sequence data (6)) and Lys and Glu at P-1 were seen, along with the striking enrichment of C-terminal basic amino acids. The presence of Arg in the late cycles of the PFR was uniformly present in all HLA types studied. The pool sequence data represent a summation of several thousand ligands; not every individual peptide will have all the features, but clearly a significant proportion must have these patterns of enrichment. Reviewing the sequences of HLA class II ligands from a variety of alleles (25) reveals that ~26% contain C-terminal basic residues, and 16% contain a Pro at P2 from the N terminus of the peptide.


View this table:
[in this window]
[in a new window]
  		Table III. Summary of peptide flanking residue enrichments from the pool sequencing data for nine HLA molecules1

 
Further validation of the HLA-DR2 pool sequencing data

We have previously shown that pool sequencing data can be used to identify HLA-specific T cell epitopes from parent proteins (9). If the HLA-DR2 pool sequence data are biologically valid, they should also allow the prediction of DR2-restricted epitopes using the same approach. We tested the integrity of the eluted sequence peptide data by observing whether they were capable of identifying known HLA-DR2 ligands/epitopes using a previously described computer program (23). Table IVGo shows the accurate and specific identification of epitopes/ligands that are HLA-DR2(b) restricted (14, 29), suggesting that the data correctly reflect the characteristics of DR2-derived peptides. For comparison, the same ligands were scanned with a probability matrix derived from pool sequence data from HLA-DR3 (the matrix is published Ref. 9), confirming the quality and specificity of the pool sequencing data for HLA-DR2.


View this table:
[in this window]
[in a new window]
  		Table IV. HLA-DR2 pool sequence data can accurately identify known HLA-DR2 ligands/epitopes1

 
Analysis of the binding of semirandom peptides to empty HLA-DR2 molecules to examine the requirements of the terminal regions

Empty HLA-DR2 molecules were purified from transfected insect Schneider cells. The peptides XXXVVHFFKNIVXXXA, XXXXIHQLKNVIXXXX, and XXXXVHAFKNIVXXXX (P1 and P4 anchor residues underlined) were dissolved in 2.2 M urea and bound to the empty HLA-DR2 molecules as described in Materials and Methods. Samples of the peptides dissolved in urea were sequenced and the random nature of the termini confirmed. The gel filtration profile and elution time confirmed that the complexes were stable heterodimers bound to peptide. The purified complexes were acidified, and the eluted peptides were collected. The specificity of the eluted peptides was further revealed by comparing the HPLC profiles of the eluted peptides with the initial profile of the sample peptide (Fig. 1Go). The eluted peptides were sequenced, and, as expected, the same core binding region was identified. The relative amounts of amino acids were compared in the noncore N- and C-terminal cycles. The amino acids were expressed as a percentage of the total amount obtained in each cycle. A deviation between peptides of 50% or more for a particular amino acid before and after binding to HLA-DR2 was arbitrarily denoted as significant.



View larger version (17K):
[in this window]
[in a new window]
  		FIGURE 1. HPLC elution profiles of a semirandom combinatorial peptide (XXXVVHFFKNIVXXXA) before and after binding to empty HLA-DR2 molecules. The 16-mer peptide libraries eluted between 30 and 40 min (hatched line, before binding; solid line, after binding).

 
The results are summarized in Table VGo. At the N-terminal flanking region, the consensus from the three peptides is a decrease in the presence of bulky hydrophobic residues after binding to HLA-DR2, which is similar to the findings in the natural pool sequence (Table IIGo). The enrichment of acidic residues seen in the eluted naturally processed peptides was only seen in one of the peptides in two of the four early cycles and probably is not significant.


View this table:
[in this window]
[in a new window]
  		Table V. Summary of the changes in the peptide flanking residues of three different peptides after binding to empty HLA-DR2 molecules1

 
The C-terminal flanking residues indicated by the late cycles of Edman sequencing reveal a decrease in basic residues for all three semicombinatorial peptides. This is in marked contrast to the enrichments found in natural ligands (compare Tables IIGo and VGo). This striking difference confirms that these C-terminal PFRs are genuinely enriched in naturally processed ligands and argues for a role for cellular processes in generating such enrichments, despite less favorable peptide-MHC interactions.

The later cycles did favor aliphatic amino acids such as Ile, Leu, and Val. At P9, small residues such as Gly and Ala were found, along with the Leu. Larger amino acids such as Met, Trp, and Tyr were decreased. It appears that the P9 pocket allows smaller residues and does not show any marked preferences for more distinctive residues, in keeping with the structural data discussed above.

Binding of synthetic peptides in vitro confirms the natural enrichments in the PFRs are not due to specific interactions with the HLA heterodimer

A range of synthetic peptides, as shown in Table VIGo, were tested for the ability to bind to purified HLA-DR2 molecules. Both the C-terminal and N-terminal extensions nonspecifically enhance binding, as seen before with the MBP peptide (14). Substitution of Thr with a His at position P9 did not significantly alter the binding, in keeping with the degenerate nature of this pocket reflected by the pool sequence data. This also suggests that the increase in His found in the natural pool in cycles 12 and 13 is not a result of binding to a pocket but probably reflects the overall increase in basic amino acids in the flanking region.


View this table:
[in this window]
[in a new window]
  		Table VI. In vitro binding of synthetic peptides to purified HLA-DR2 molecules1

 
Substitutions with Lys at different positions in the C-terminal region did not significantly alter binding, although there was a tendency to reduce the binding affinity, which may contribute to the decrease yields seen in Table VGo. This finding offers further evidence that the natural enrichments are the result of mechanisms not involving the HLA-peptide interaction. Indeed, these results suggest that the naturally processed ligands must interact in a pathway that can overcome unfavorable interactions. A second set of peptides containing N-terminal substitutions with Asp bound overall with a reduced affinity, although the amino acids at core residues 8 and 9 were switched to allow a more favorable residue at P9 (i.e., Val). It appears that the C-terminal extensions tend to favor binding to a greater degree.

Alterations in the C-terminal PFRs can lead to changes in T cell activation

Basic amino acids in the C-terminal flanking region are highly conserved across a range of alleles. There is already some evidence that these residues can alter TCR activation (13, 14). To examine this further, we initially attempted to generate HLA-DR2-restricted CD4+ T cell clones to the immunodominant epitope within MBP 83–99 (QDENPVVHFFKNIVTPR), as outlined in Materials and Methods. Three clones were generated, of which two were specific for MPB 83–99. The proliferative responses of the clones were tested against four additional peptides: MBP-1 (VVHFFKNIVHAA), MBP-2 (VVHFFKNIVTAAA), MBP-3 (VVHFFKNIVTKAA), and MBP-4 (VVHFFKNIVTAKA). Both clones showed increased proliferation with peptide MBP-4. A representative result with one clone (IC3 1) is shown in Fig. 2Go. The increase in T cell reactivity to peptide MBP-4 (Lys at P11) does not correspond to an altered affinity for HLA-DR2 (Table VIGo). These results are corroborated by another study on the response of an MBP 86–96-specific clone that was more reactive to peptides with a Pro to Lys substitution in the C-terminal flanking position (18). Interestingly, this clone was again far more reactive to additional non-MBP peptides with a Lys substitution in the C-terminal flanking region.



View larger version (26K):
[in this window]
[in a new window]
  		FIGURE 2. Proliferation measured by [3H]thymidine incorporation of an MBP 83–99-specific T cell clone to a series of peptides with altered C-terminal PFRs. The core-binding region of the peptides is underlined. The maximum response is obtained to MBP-4 with a basic residue at position P11. Error bars shown ± SD. MBP-1, VVHFFKNIVHAAA; MBP-2, VVHFFKNIVTAAA; MBP-3: VVHFFKNIVTKAA; MBP-4, VVHFFKNIVTAKA.

 
To extend this study further, an ex vivo analysis of T cell activation was conducted in 11 individuals using a CD4+ T cell epitope from influenza A virus HA 305–320 (ACPKYVKQNTLKLATG). This epitope has been shown to bind promiscuously to a range of HLA class II molecules using the core region Tyr309 to Leu317 (30). Two C-terminally modified peptides were synthesized with an Arg substituted for either Ala318 or Thr319 (labeled HA-1 and HA-2, respectively).

The T cell responses were measured by IFN-{gamma} release in an 18-h ELISPOT assay. The sensitivity of the ELISPOT assay allows low frequency responses to be enumerated ex vivo from PBMCs. The individual responses to HA-1 were similar, but there was an increase in responses to HA-2 in 10 of the 11 individuals tested (Fig. 3GoA). Here, the substituting Arg is placed at the same position as the Lys in the MBP peptide (i.e., P11). TCLs were generated using peptide HA-2. A representative experiment is shown in Fig. 3GoB, revealing that it was possible to select out T cells that may be favorably activated by the altered peptide HA-2. Thus, a population of T cells with preferences for basic residues at P11 can be selected out from a memory pool that was initially generated by exposure to native HA during influenza A infection.



View larger version (16K):
[in this window]
[in a new window]
  		FIGURE 3. A, Ex vivo T cell responses in 11 individuals measured by IFN-{gamma} ELISPOT assays to the promiscuously binding universal CD4+ T cell epitope hemagglutinin (HA) 305–320 (ACPKYVKQNTLKLATG); the core-binding region is underlined. The variant peptides HA1 (ACPKYVKQNTLKLRTG) and HA-2 (ACPKYVKQNTLKLARG) were also tested. A total of 10 of 11 individuals showed an increased frequency of T cells activated with HA-2. Responses to the HA-1 variant were not significantly altered. HA 305–320 vs HA-1, p = 0.7; HA 305–320 vs HA-2, p < 0.01 (two-tailed Student’s t test). B, PBMCs were cultured for 7 days with HA-2 and then restimulated and expanded for a further 7 days. The T cell lines were then assayed by [3H]thymidine incorporation in a standard proliferation assay. The lines were tested against the native peptide HA 305–320 and HA-2; a representative example is shown. T cells, which clearly revealed a preference for HA-2 containing the basic Arg at position P11, could thus be selected out and expanded from the peripheral pool. Error bars shown ± SD.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The data outlined above strongly suggest that certain enrichments of the PFRs result from cellular mechanisms involved in Ag processing and are not the result of specific interactions of the PFRs with the HLA heterodimer. The presence of allele-transcending changes may be particularly important if there is a proportion of TCRs that favorably recognize these noncore residues. A study of mouse MHC class II molecules has found that the C-terminal PFRs can have a profound effect on TCR V region use and T cell function. Intriguingly, the substitution of basic residues for the original residues in the C-terminal flanking region markedly altered T cell function (13). It appears that the MHC class II-peptide complex can interact with the TCR in such a manner that the PFRs (in these examples, the C-terminal residues) can influence T cell activation. The exact physical nature of the interaction awaits further structural studies (31). MBP-reactive clones have been described that also favor C-terminal basic flanking residues in both the MBP peptide as well as non-MBP-stimulating peptides (18). We also generated MBP-specific clones that demonstrated an increased response following the substitution of a basic for another residue at the C terminus. We extended this by an analysis of ex vivo and cultured responses to a universal epitope derived from HA. This suggested that there was a population of T cells that could be favorably activated by C-terminal flanking basic residues, and these cells could be expanded from a memory pool derived from exposure to the original epitope.

The realization that TCR activation may be altered by peptides whose flanking regions contain particular residues, and that these PFRs are generated by cellular processes in a fixed fashion, may be of fundamental as well as practical importance. Indeed, these cellular processes appear to overcome a less favorable HLA-peptide interaction to generate certain PFRs. If T cells are initially selected in the thymus on a series of ligands that often have these PFRs, and these flanking regions can alter TCR activation, this may account for later-favored activation of a certain proportion of the cognate T cells.

The rules dictating crypticity and dominance of epitopes are not fully understood (32). It is clear that the flanking residues are important for the immunogenicity of epitopes for a variety of reasons (12). Recently, it has been reported that the introduction of basic residues to the C-terminal flanking region of a subdominant epitope from hen egg lysozyme enhanced the processing and presentation of this epitope (33). There is very little known about cryptic CD4+ T cell epitopes in humans. The best examples are from comparisons of T cell responses to whole protein Ags, such as acetylcholine receptor and collagen compared with overlapping peptides covering these proteins (34, 35). Certain epitopes are only recognized when peptides are used to stimulate T cell lines but are not processed from the whole Ag, suggesting that they may be subdominant or cryptic epitopes. Using the epitope-prediction computer program described above and previously (23), the addition of a basic residue to the C-terminal flanking region of these cryptic epitopes can markedly increase the probability of being processed and presented score (data not shown). Theoretically, this may reflect an alteration in status of the epitope from cryptic to dominant, as practically described in the mouse model with hen egg lysozyme (33).

It seems that these favored C-terminal basic residues may possibly contribute to the selection of epitopes by at least two mechanisms: first, by enhancing the processing of epitopes from a candidate protein and, second, by increased activation of certain T cells. Future studies are now planned to assess the immunogenicity of epitopes and proteins incorporating these preferred PFRs.


   Acknowledgments
 
This research was conducted while K.J.S. was in D. Wiley’s laboratory. We thank K. Wucherpfennig for providing the S2/DR2 transfectants, P. Pietras for assistance with tissue culture, and N. Kirilcuk for purification of empty DR2. We also thank A. McMichael for useful discussion and for allowing the use of facilities in his laboratory, and N. Willcox and K. Gould for comment on the manuscript.


   Footnotes
 
1 A.J.G. was supported by the United Kingdom Medical Research Council. K.J.S. was supported by a Wellcome Trust Traveling Fellowship. A.V.S.H. is a Wellcome Trust Principal Research Fellow. Back

2 Address correspondence and reprint requests to Dr. Andrew Godkin at his current address, Imperial College, St. Mary’s Campus, Room 366, Norfolk Place, Paddington, London W2 1PG, U.K. E-mail address: a.godkin{at}ic.ac.uk Back

3 Abbreviations used in this paper: P, position; PFR, peptide flanking residue; MBP, myelin basic protein; CLIP, class II-associated invariant chain peptide; ELISPOT, enzyme-linked immunospot; HA, hemagglutinin. Back

Received for publication November 21, 2000. Accepted for publication March 29, 2001.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Rammensee, H.-G.. 1995. Chemistry of peptides associated with MHC class I and class II molecules. Curr. Opin. Immunol. 7:85.[Medline]
  2. Stern, L., J. Brown, T. Jardetzky, J. Gorga, R. Urban, J. Strominger, D. Wiley. 1994. Crystal structure of the human class II MHC protein HLA-DR1 complexed with an influenza virus peptide. Nature 368:215.[Medline]
  3. Ghosh, P., M. Amaya, E. Mellins, D. Wiley. 1995. The structure of an intermediate in class II MHC maturation: CLIP bound to HLA-DR3. Nature 378:457.[Medline]
  4. Dessen, A., M. Lawrence, S. Cupo, D. Zaller, D. Wiley. 1997. X-ray crystal structure of HLA-DR4(DRA*0101, B1*0401) complexed with a peptide from human collagen II. Immunity 7:473.[Medline]
  5. Smith, K., J. Pyrdol, L. Gauthier, D. Wiley, K. Wucherpfennig. 1998. Crystal structure of HLA-DR2 (DRA*0101, DRB1*1501) complexed with a peptide from human myelin basic protein. J. Exp. Med. 188:1511.[Abstract/Free Full Text]
  6. Falk, K., O. Rotzschke, S. Stevanovic, G. Jung, H.-G. Rammensee. 1994. Pool sequencing of natural HLA-DR, DQ, and DP ligands reveals detailed peptide motifs, constraints of processing, and general rules. Immunogenetics 39:230.[Medline]
  7. Friede, T., V. Gnau, G. Jung, W. Keilholz, S. Stevanovic, H.-G. Rammensee. 1996. Natural ligand motifs of closely related HLA-DR4 molecules predict features of rheumatoid arthritis associated peptides. Biochim. Biophys. Acta 1316:85.[Medline]
  8. Godkin, A., T. Friede, M. Davenport, S. Stevanovic, A. Willis, J. Jewell, A. Hill, H.-G. Rammensee. 1997. Use of eluted peptide sequence data to identify the binding characteristics of peptides to the insulin-dependant diabetes susceptibility allele HLA-DQ8 (DQ 3.2). Int. Immunol. 9:905.[Abstract/Free Full Text]
  9. Godkin, A., M. Davenport, A. Willis, D. Jewell, A. Hill. 1998. Use of complete eluted peptide sequence data from HLA-DR and -DQ molecules to predict T-cell epitopes, and the influence of the nonbinding terminal regions of ligands in epitope selection. J. Immunol. 161:850.[Abstract/Free Full Text]
  10. Malcherek, G., V. Gnau, G. Jung, H.-G. Rammensee, A. Melms. 1995. Supermotifs enable natural invariant chain-derived peptides to interact with many major histocompatibility complex-class II molecules. J. Exp. Med. 181:527.[Abstract/Free Full Text]
  11. Rotzschke, O., K. Falk, J. Lau, G. Jung, J. Strominger. 1999. Conformation variants of class II MHC/peptide complexes induced by N- and C-terminal extensions of minimal peptide epitopes. Proc. Natl. Sci. USA 96:7445.[Abstract/Free Full Text]
  12. Moudgil, K., E. Sercarz, I. Grewal. 1998. Modulation of the immunogenicity of antigenic determinants by their flanking residues. Immunol. Today 19:217.[Medline]
  13. Carson, R., K. Vignali, D. Woodland, D. Vignali. 1997. T cell receptor recognition of MHC class II-bound peptide flanking residues enhances immunogenicity and results in altered TCR V region usage. Immunity 7:387.[Medline]
  14. Wucherpfennig, K., A. Sette, S. Southwood, C. Oseroff, M. Matsui, J. Strominger, D. Hafler. 1994. Structural requirements for binding of an immunodominant myelin basic peptide to DR2 isotypes and for its recognition by human T cell clones. J. Exp. Med. 179:279.[Abstract/Free Full Text]
  15. Olerup, O., S. Hillert, S. Fredrickson, T. Olsson, E. Kam-Hansen, B. Carlsson, J. Wallin. 1989. Primary chronic progressive and relapsing/remitting multiple sclerosis: two immunogenetically distinct disease entities. Proc. Natl. Acad. Sci. USA 86:7113.[Abstract/Free Full Text]
  16. Vogt, A., H. Kropshofer, H. Kalbacher, M. Kalbus, H.-G. Rammensee, J. Coligan, R. Martin. 1994. Ligand motifs of HLA-DRB5*0101 and DRB1*1501 molecules delineated from self-peptides. J. Immunol. 153:1665.[Abstract]
  17. Ota, K., M. Matsui, E. Milford, G. Mackin, H. Weiner, D. Hafler. 1990. T-cell recognition of an immunodominant myelin basic protein epitope in multiple sclerosis. Nature 346:183.[Medline]
  18. Hemmer, B., B. Fleckenstein, M. Vergelli, G. Jung, H. McFarland, R. Martin, K.-H. Wiesmuller. 1997. Identification of high potency microbial and self ligands for a human autoreactive class II-restricted T-cell clone. J. Exp. Med. 185:1651.[Abstract/Free Full Text]
  19. Kozovska, M., Y. Zang, I. Aebischer, S. Lnu, V. Rivera, P. Crowe, S. Boehme, J. Zhang. 1998. T cell recognition motifs of an immunodominant peptide of myelin basic protein in patients with multiple sclerosis: structural requirements and clinical implications. Eur. J. Immunol. 28:1894.[Medline]
  20. Fleckenstein, B., H. Kalbacher, C. Muller, D. Stoll, T. Halder, G. Jung, K.-H. Wiesmuller. 1996. New ligands binding to the human leukocyte antigen class II molecule DRB1*0101 based on the activity pattern of an undecapeptide library. Eur. J. Biochem. 240:71.[Medline]
  21. Gorga, J. C., V. Horejsi, D. R. Johnson, R. Raghupathy, J. L. Strominger. 1987. Purification and characterization of class II histocompatibility antigens from a homozygous human B cell line. J. Biol. Chem. 262:16087.[Abstract/Free Full Text]
  22. Robbins, P., E. Evans, A. Ding, N. Warner, F. Brodsky. 1987. Monoclonal antibodies that distinguish between class II antigens (HLA-DP, DQ, and DR) in 14 haplotypes. Hum. Immunol. 18:301.[Medline]
  23. Davenport, M., I. Ho Shon, A. Hill. 1995. An empirical method for the prediction of T-cell epitopes. Immunogenetics 42:392.[Medline]
  24. Davenport, M., C. Quinn, R. Chicz, B. Green, A. Willis, W. Lane, J. Bell, A. Hill. 1995. Naturally processed peptides from two disease-resistance-associated HLA-DR13 alleles show related sequence motifs and the effects of the dimorphism at position 86 of the HLA-DR {beta} chain. Proc. Natl. Acad. Sci. USA 92:6567.[Abstract/Free Full Text]
  25. Rammensee, H.-G., T. Friede, S. Stevanovic. 1995. MHC ligands and peptide motifs: first listing. Immunogenetics 41:178.[Medline]
  26. Nelson, C., I. Vidavsky, N. Viner, M. Gross, E. Unanue. 1997. Amino-terminal trimming of peptides for presentation on major histocompatibility complex class II molecules. Proc. Natl. Acad. Sci. USA 94:628.[Abstract/Free Full Text]
  27. Arnold, D., W. Keilholz, H. Schild, T. Dumreses, S. Stevanovic, H.-G. Rammensee. 1997. Substrate specificity of cathepsins D and E determined by N-terminal and C-terminal sequencing of peptide pools. Eur. J. Biochem. 1:171.
  28. Davenport, M., A. Godkin, T. Friede, S. Stevanovich, A. Willis, A. Hill, H.-G. Rammensee. 1997. A distinctive binding motif for HLA-DRB1*0407, an HLA-DR4 subtype not associated with rheumatoid arthritis. Immunogenetics 45:229.[Medline]
  29. Wucherpfennig, K., J. Strominger. 1995. Molecular mimicry in T cell-mediated autoimmunity: viral peptides activate human T cell clones specific for myelin basic protein. Cell 80:695.[Medline]
  30. O’Sullivan, D., T. Arrhenius, J. Sidney, G. M. Del, M. Albertson, M. Wall, C. Oseroff, S. Southwood, S. M. Colon, F. C. Gaeta, et al 1991. On the interaction of promiscuous antigenic peptides with different DR alleles: identification of common structural motifs. J. Immunol. 147:2663.[Abstract/Free Full Text]
  31. Reinherz, E., K. Tan, L. Tang, P. Kern, J. Liu, Y. Xiong, R. Hussey, A. Smolyar, B. Hare. 1999. The crystal structure of a T cell receptor in complex with peptide and MHC class II. Science 286:1913.[Abstract/Free Full Text]
  32. Sercarz, E., P. Lehmann, A. Ametani, G. Benichou, A. Miller, K. Moudgil. 1993. Dominance and crypticity of T cell antigenic determinants. Annu. Rev. Immunol. 11:729.[Medline]
  33. Schneider, S., J. Ohmen, L. Fosdick, B. Gladstone, J. Guo, A. Ametani, E. Sercarz, H. Deng. 2000. Cutting edge: introduction of an endopeptidase cleavage motif into a determinant flanking region of hen egg lysozyme results in enhanced T cell determinant display. J. Immunol. 165:20.[Abstract/Free Full Text]
  34. Cope, A., S. Patel, F. Hall, M. Congia, H. Hubers, G. Verheijden, A. Boots, R. Menon, M. Trucco, A. Rijinders, G. Sonderstrup. 1999. T cell responses to a human cartilage autoantigen in the context of rheumatoid arthritis-associated and nonassociated HLA-DR4 alleles. Arthritis Rheum. 42:1497.[Medline]
  35. Matsuo, H., A.-P. Batocchi, S. Hawke, M. Nicolle, L. Jacobson, A. Vincent, J. Newsome-Davis, N. Willcox. 1995. Peptide-selected T cell lines from myasthenia gravis patients and controls recognise epitopes that are not processed from whole acetylcholine receptor. J. Immunol. 155:3683.[Abstract]



This article has been cited by other articles:


Home page
 CVIHome page
S. Gaseitsiwe, D. Valentini, R. Ahmed, S. Mahdavifar, I. Magalhaes, J. Zerweck, M. Schutkowski, E. Gautherot, F. Montero, A. Ehrnst, et al.
Major Histocompatibility Complex Class II Molecule-Human Immunodeficiency Virus Peptide Analysis Using a Microarray Chip
Clin. Vaccine Immunol., April 1, 2009; 16(4): 567 - 573.
[Abstract] [Full Text] [PDF]

Home page
 BioinformaticsHome page
C. Lundegaard, O. Lund, C. Kesmir, S. Brunak, and M. Nielsen
Modeling the adaptive immune system: predictions and simulations
Bioinformatics, December 15, 2007; 23(24): 3265 - 3275.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
A. P. Fontenot, T. S. Keizer, M. McCleskey, D. G. Mack, R. Meza-Romero, J. Huan, D. M. Edwards, Y. K. Chou, A. A. Vandenbark, B. Scott, et al.
Recombinant HLA-DP2 Binds Beryllium and Tolerizes Beryllium-Specific Pathogenic CD4+ T Cells
J. Immunol., September 15, 2006; 177(6): 3874 - 3883.
[Abstract] [Full Text] [PDF]

Home page
 Infect. Immun.Home page
K. M. Williams and E. C. Bigley III
Identification of an I-Ed-Restricted T-Cell Epitope of Escherichia coli Outer Membrane Protein F
Infect. Immun., July 1, 2004; 72(7): 3907 - 3913.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
A. J. Godkin, H. C. Thomas, and P. J. Openshaw
Evolution of Epitope-Specific Memory CD4+ T Cells After Clearance of Hepatitis C Virus
J. Immunol., August 15, 2002; 169(4): 2210 - 2214.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
P. Y. Arnold, N. L. La Gruta, T. Miller, K. M. Vignali, P. S. Adams, D. L. Woodland, and D. A. A. Vignali
The Majority of Immunogenic Epitopes Generate CD4+ T Cells That Are Dependent on MHC Class II-Bound Peptide-Flanking Residues
J. Immunol., July 15, 2002; 169(2): 739 - 749.
[Abstract] [Full Text] [PDF]

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 HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Godkin, A. J.
Right arrow Articles by Hill, A. V. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Godkin, A. J.
Right arrow Articles by Hill, A. V. S.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS