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 Honeyborne, I.
Right arrow Articles by Goulder, P. J. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Honeyborne, I.
Right arrow Articles by Goulder, P. J. R.
Right arrowPubmed/NCBI databases
*Substance via MeSH
The Journal of Immunology, 2006, 176: 4699-4705.
Copyright © 2006 by The American Association of Immunologists

Motif Inference Reveals Optimal CTL Epitopes Presented by HLA Class I Alleles Highly Prevalent in Southern Africa1

Isobella Honeyborne*, Almas Rathod{dagger}, Rico Buchli{ddagger}, Dhanwanthie Ramduth§, Eshia Moodley§, Prinisha Rathnavalu§, Senica Chetty§, Cheryl Day*,§, Christian Brander{dagger}, William Hildebrand, Bruce D. Walker{dagger},||, Photini Kiepiela§ and Philip J. R. Goulder2,*,{dagger}

* Department of Paediatrics, Nuffield Department of Medicine, Peter Medawar Building for Pathogen Research, Oxford, United Kingdom; {dagger} Partners AIDS Research Center and Infectious Disease Division, Massachusetts General Hospital and Harvard Medical School, Boston, MA 02129; {ddagger} Pure Protein LLC, Oklahoma City, OK 73104; § HIV Pathogenesis Programme, Doris Duke Medical Research Institute, University of KwaZuluNatal, Durban, South Africa; Department of Microbiology and Immunology, University of Oklahoma Health Sciences Center, Oklahoma City, OK 73104; and || Howard Hughes Medical Institute, Harvard Medical School, Boston, MA 02129


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
HIV-specific CTL play a central role in immune control of HIV. The basis for understanding the success or failure of this immune response requires identification of the specific epitopes targeted by CTL. However, in populations most severely affected by the global epidemic, this fundamental knowledge is hindered by the lack of characterization of many of the HLA class I alleles highly prevalent in such populations. Overall, the peptide-binding motif has been determined for a small minority (9%) of HLA class I alleles, with a strong bias toward those alleles prevalent in Caucasoid populations. These studies therefore set out to define, in a South African Zulu/Xhosa population at the epicenter of the epidemic, the epitopes presented by alleles highly prevalent, but for which the peptide-binding motif had not been characterized. Using a method of motif inference, epitopes presented by four such alleles prevalent in the Zulu/Xhosa population of Durban, South Africa, namely, B*3910, B*4201, B*8101, and Cw*1801, are described. Importantly, this approach may additionally facilitate optimization of epitopes in certain instances where conflicting reports in the literature exist regarding the peptide-binding motif, such as for HLA-A*2902, also highly prevalent in southern African populations. These data indicate that the previously anomalous position of HLA-A*2902 among HLA-A alleles, outside any recognized HLA-A supertype, is artifactual, and the true position of the A*2902 motif overlaps those of the A1 and A24 supertypes.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
In the past decade, there has been a rapid increase in the number of virus-specific CTL epitopes that have been defined. This is well illustrated for HIV-1, for which the first epitope was defined in 1988 (1). By 1993, an additional 10 epitopes had been fully defined (2) and in the current HIV Immunology database 214 epitopes are listed (3). A recent single study of a population of Zulu/Xhosa in Durban, South Africa (4), identified 180 CTL responses to epitopes within 18-mer peptides, of which only 45 had previously been described. Thus, although a large number of epitopes have been defined, an even greater number have yet to be determined. In addition, the number of HLA class I alleles that have been described has increased, from <200 in 1993 to >1100 in 2003 (<www.anthonynolan.org.uk/HIG/index.html>). The majority of uncharacterized new HLA class I alleles and the CTL epitopes they present are expressed in non-Caucasoids that include the very populations most afflicted by the HIV-1 epidemic and toward which vaccine-directed efforts need to be most urgently directed.

Knowledge of the CTL epitopes targeted forms the basis for understanding the mechanisms of success or failure of immune control of a pathogen, such as HIV-1, where elimination or containment is dependent on CTL activity. Vaccine-directed efforts therefore need to focus first on identifying the epitopes presented by HLA alleles prevalent in populations severely affected by the global HIV epidemic. This study set out to define the dominant epitopes restricted by previously uncharacterized HLA alleles highly prevalent in the Zulu/Xhosa population in KwaZuluNatal, South Africa, and which, from previous studies, play a prominent role in the HIV-specific CTL response (4).

The opportunity to rapidly define novel epitopes results primarily from use of the ELISPOT assay and the use of matrices of pooled peptides that allow identification of an individual peptide that is recognized by CTL (5). The peptides used typically are 15–20 aa in length and are so constructed to span the entire expressed genome of the virus in question. Each peptide overlaps by 10 aa with the next to ensure that, theoretically, CTL responses to all epitopes up to 11 aa in length are detectable (6). Definition of the optimal epitope, 8–11 aa long, from the 15- to 20-mer is greatly facilitated by knowledge of the relevant HLA class I molecule motif. The motif, determined by elution and sequencing of mostly self-peptides (<www.syfpeithi.de/> and Ref.7), has been defined for 97 of the 1,111 (8.7%) HLA-A, -B, and -C alleles (<www.syfpeithi.de/> and Ref.7). The motif has not been determined for many of the alleles known to play an important role in the HIV-specific response in Southern Africa. Notable examples are HLA-B alleles such as HLA-B*4201 (expressed in 20.4% of the population), B*8101 (12.0%), and HLA-C alleles such as HLA-Cw*1801 (14.4%) that dominated the HLA-B and HLA-C-restricted anti-HIV response, respectively, in a recent study of 375 C clade-infected South Africans (4).

To determine the peptide-binding motifs for these uncharacterized HLA class I alleles, it is necessary to know the HLA residues forming the primary anchor pockets. This may be inferred from knowledge of the amino acid sequence of the alleles and of the residues that, from crystallographic studies, characteristically form these binding pockets (7, 8, 9, 10). We hypothesized that the motif may then be successfully inferred by comparison of these pocket-forming HLA residues in uncharacterized alleles, such as HLA-B*4201, with those equivalent HLA residues that form the same pockets in related alleles, such as HLA-B*0702, for which the peptide-binding motif has been determined. We tested this hypothesis for one such HLA-C and three such HLA-B alleles. For a fifth allele, HLA-A*2902, phenotypic frequency was 16.4% in Zulu/Xhosa, where a marked discrepancy exists between the defined motif (11) and defined epitopes, we applied the same approach to determine whether the inferred motif was consistent with the actual epitopes or with the defined motif.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
ELISPOT assays and peptides

As previously described (4, 5), 410 overlapping peptides, 18 aa in length and overlapping by 10 aa with the adjacent peptide, were synthesized. These peptides spanned the entire length of the expressed HIV-1 C clade consensus sequence, that being the clade prevalent in Durban. These peptides were arranged in a matrix of 10–12 peptides per pool, such that the recognition of a single peptide would be observed uniquely by detection of responses in two wells of an IFN-{gamma} ELISPOT assay. Recognition of the individual peptide was confirmed in a subsequent independent IFN-{gamma} ELISPOT assay.

Determination of HLA restriction by intracellular cytokine staining assays and ELISPOT

Confirmation of the statistically determined HLA restriction of the response was undertaken as previously described in intracellular IFN-{gamma} staining assays (12) or by ELISPOT assay (13) using a panel of EBV-transformed B cell lines (BCL)3 matched individually through only one class I allele with the effector cells being tested. Briefly, the peptide of interest was incubated with each matched BCL for 60 min at 37°C, followed by washing to remove unbound peptide. Each peptide-pulsed BCL and corresponding negative control, BCL with no peptide added, was incubated with specific effectors (CTL lines or PBMC) for 90 min. Brefeldin A was added followed by a further incubation of 4.5 h. FACS analysis was used to determine percentage of IFN-{gamma}-FITC-, CD8-PE-, and CD4-APC-positive cells. HLA restriction by IFN-{gamma} ELISPOT assay utilized a similar procedure, as previously described (5, 13).

HLA typing and HLA frequencies

Patients HLA type was determined to the oligo-allelic level using Dynal RELITM Reverse Sequence-Specific Oligonucleotide kits for the HLA-A, -B, and -C loci (Dynal Biotech). To obtain four-digit typing, Dynal Biotech Sequence-Specific priming kits were used, in conjunction with the Sequence-Specific Oligonucleotide type. Where alleles were still not defined to the four-digit level, the sequence-specific priming primer mixes were used. All HLA class I alleles in the International Immunogenetics (IMGT) allele release 2.4.0 were considered in the typing. Phenotypic frequencies quoted were determined from HLA typing of 250 consecutively typed, HIV-infected mothers presenting at antenatal clinic and consenting to participate in the study.

Motif inference

The amino acids lining the six pockets, critical for epitope binding, have previously been determined (7, 8, 9, 10). The residues at the second position and at the carboxyl terminus of the epitope, that bind within the B and F pockets, respectively, form the primary anchor residues in 86 of the 97 motifs determined by analysis of eluted peptides (<www.syfpeithi.de/> and Ref.7). For the great majority of HLA class I molecules, therefore, the B and F pockets are the most important in determining the peptide-binding motif. The remaining 11 alleles have additional or alternative anchor residues. Of these 11, there are 9, which still rely on the carboxyl terminus of the epitope as a primary anchor; but have additions to the B pocket as anchors; e.g., Cw*0101 uses the D pocket, accommodating position 3 of the epitope, as a primary anchor. B*0801 and B*0802 are the only defined alleles where residues 3 and 5 of the motif are the primary anchors and the B and F pockets of these alleles therefore play a lesser role.

To infer the motif for uncharacterized alleles, the HLA residues anticipated to line the B and F pockets were compared with those of alleles for which the motif has already been determined. Where these HLA residues were identical for the two class I molecules being compared, an identical motif was inferred; for example, the residues anticipated to form the F pocket of HLA-B*0702 and B*4201 (Table I). In the absence of identical pockets, the motif for the uncharacterized allele was predicted by reference to a closely related allele for which the motif had been defined, taking into account differences in charge and size of the side chains of the amino acids lining the pockets.


View this table:
[in this window]
[in a new window]
  		Table I. Motif inference and facilitated optimization of seven epitopesa

 
Validation of inferred motifs

To test the validity and utility of this approach for predicting peptide-binding motifs, the HIV-specific CTL responses were characterized in a cohort of 375 study subjects from Durban, South Africa (4), which included a number of HLA class I alleles whose motif had not been determined but which were present at high prevalence. Initial IFN-{gamma} ELISPOT assays identified which of 410 18-mer overlapping peptides were targeted. The HLA restriction was determined using a statistical method comparing the frequency of the HLA class I alleles expressed in study subjects making a response to the 18-mer peptide with the frequency of those alleles in the nonresponders (Ref.4 and Fig. 1A).


Figure 1
View larger version (33K):
[in this window]
[in a new window]
  		FIGURE 1. Definition of novel B*4201, B*3910, and B*8101 epitopes facilitated by motif inference (see Table I). A, Statistical association between recognition of overlapping peptide 244, MASEFNLPPIVAKEIVA, and expression of HLA-B*4201. B, Definition of the optimal B*4201-restricted epitope LPPIVAKEI (integrase residues 28–36). C, Definition of the optimal B*3910-restricted epitope TPQDLNTML (p24 residues 48–56). D, Definition of the optimal B*8101-restricted epitope TPVNIIGRNML (protease residues 80–90).

 
Determination of the optimal epitope within the 18-mer overlapping peptide recognized was undertaken in two steps. First, the optimal 8-to 11-mer epitope was predicted using the established or predicted peptide-binding motif for the relevant allele. Second, truncated peptides, one amino acid longer or shorter than the predicted optimal 8- to 11-mer, were used in further IFN-{gamma} ELISPOT assays. These peptides were serially diluted and incubated with PBMC to ascertain the truncation causing an IFN-{gamma} response at the lowest dilution, thereby determining the identity of the optimal epitope.

HLA-A*2902-binding assay

To assess the ability of synthetically defined peptide epitopes to associate with the specific class I complex HLA-A*2902, a screening assay based on inhibition of binding of a fluorescent standard peptide was used. Binding events were determined using the technique of fluorescence polarization, which is unique among methods used to analyze molecular-binding events because it gives a direct, nearly instantaneous measurement of a ligand’s bound:free ratio. The technique is based on the principle that small molecules rotate faster than large molecules. During binding, the small fluorescent peptide, which has free rotational mobility, is converted to a larger fluorescent peptide-HLA complex with restricted rotational mobility, resulting in an increase in polarization in absence of a competitor. At a selected high threshold concentration of 80 µM, various competitor peptides were incubated with activated soluble (s) HLA-A*2902 (50 µg/ml) in the presence of 1 nM FlTC-labeled reference peptide YTDGVPL-K(FITC)-Y and excess beta2-microglobulin (24.8 µg/ml) and peptide-MHC interactions were monitored over time. Final equilibrium polarization levels were transformed into percent inhibition to indicate the extent of binding to sHLA-A*2902.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
Definition of novel epitopes presented by HLA-B*4201, B*8101, and B*3910

Initial focus was placed on HLA-B*4201, the HLA class I allele that previous studies had shown makes the strongest individual contribution to the HIV-specific CTL response in the study population in Durban, South Africa (4). The B*4201 allele was sequenced in 1988 (14) but its motif has not been determined. Approximately 20% of the Zulu-Xhosa population express B*4201 and it is thus one of the most prevalent alleles. The approach adopted for defining B*4201-restricted epitopes is illustrated in Fig. 1, A and B. A strong association between recognition of the 18-mer peptide MASEFNLPPIVAKEIA and expression of HLA-B*4201 was observed (p = 3.2 x 10–18), and the identical motif to HLA-B*0702 (15, 16, 17) was inferred for B*4201 (Table I). The peptide-binding motif of Pro at position 2 (P2) and Leu at the C-terminal position (PC) predicted LPPIVAKEI (LI9) as the optimal epitope, as was subsequently confirmed (Fig. 1B).

Similarly, recognition of the 18-mer LVGPTPVNIIGRNMLTQL was strongly associated with expression of the HLA-B*8101 allele (p = 4.5 x 10–19). The sequence of B*8101 was defined in 1996 (18), and again no motif has been characterized to date. Comparison of this uncharacterized allele to other defined alleles indicated B*8101 would have the same B and F pocket amino acid preferences as B*0702. Potential optimal epitopes within this 18-mer satisfying the peptide-binding motif of Pro at P2 and Leu (or a medium-sized hydrophobic residue) at the PC, assuming an optimal epitope of 8–11 aa, are the 11-mer, TPVNIIGRNML (TL11), the 10-mer, TPVNIIGRNM (TM10), and the 8-mer GPTPVNII (GI8). The optimal epitope in this case proved to be the 11-mer, TL11 (Fig. 1C and data not shown), and B*8101 was formally shown to be the restriction element for this response (data not shown).

The utility of this approach was tested for one additional HLA-B allele for which the peptide-binding motif is unknown. For HLA-B*3910, the prevalent subtype of HLA-B39 in the Durban cohort (<www.ncbi.nlm.nih.gov/projects/mhc>), the motif was expected to differ significantly from the B*3901 and B*3902 subtypes (19) for which the binding motif has been characterized (Table I), in particular as a result of Tyr67 in B*3910 replacing Cys67 or Ser67 in B*3901 and B*3902, respectively. The B pocket for B*3910 would thus be expected to be considerably smaller and very similar in size and charge to the B pocket of B*0702. The F pocket for B*3910 is identical to that of B*3901 and B*3902 and differs only marginally from B*0702 in the substitution of Phe for Tyr at residue 116, thus a medium-sized hydrophobic residue such as Leu or Ile would be predicted to bind favorably into the F pocket of B*3910. The predicted optimal epitope TPQDLTNML from the 17-mer GATPQDLNTMLNTVGGH was confirmed (Fig. 1D) and HLA-B*3910 restriction having also been confirmed (4).

Definition of novel epitopes presented by Cw*1801

Definition of the peptide-binding motifs for HLA-C alleles has been particularly neglected; only 3.2% of C allele motifs have been formally defined. To establish the validity of the approach of motif inference as applied to HLA-C alleles, HLA-Cw*1801, the HLA-C allele that makes the greatest contribution to the HIV-specific CTL response (4) and for which the motif has not been defined, was studied. The predicted Cw*1801 B pocket motif was Arg/Gln by analogy with the defined motif for the very similar Cw*0602 B pocket (20), the only difference being a change of Phe->Tyr at HLA residue 99. The Cw*1801 F pocket motif would be predicted to be identical to that of Cw*0401 (20), whose motif has been defined (Table I). In each of four cases, the optimal epitope based on this predicted motif for Cw*1801 proved to be as anticipated (Fig. 2). In each case, the optimal epitope was determined from studies of at least two subjects in whom the response was detected.


Figure 2
View larger version (32K):
[in this window]
[in a new window]
  		FIGURE 2. Optimization of four novel HLA-Cw*1801-restricted epitopes. A, The epitope FRDYVDRFF (p24 residues 161–169). B, The epitope VRMYSPVSI (p24 residues 142–150). C, The epitope YRLGVGALI (gp41 residues 1–8). The epitope VRDQAEHL (integrase residues 165–172).

 
Discrepancies between defined motif and defined epitopes

Previous reports have illustrated the fact that the sequence of an epitope peptide may not conform precisely to the binding motif that has been defined for a particular allele, which may point to the difference between the sequences of self-peptides and those of epitope peptides. However, for one particular allele, HLA-A*2902, for which the peptide-binding motif has been determined by sequencing of eluted peptides, none of the A*2902-restricted CTL epitopes that have been defined conform to the motif defined in 1996 (Ref.11 and Table II). From this study, the sequences of self-peptides eluted from A*2902 in 16 of 16 cases had the negatively charged Glu at P2, 6 of 16 cases had Tyr at PC, and 5 of 16 cases had Leu at PC. However, alignment of the five defined HLA-A*2902-restricted epitopes show, that in each case, they have a large hydrophobic residue (Phe or Tyr) anchoring the peptide in the B pocket (Refs.3 , 6 , and 21, 22, 23 and Table II).


View this table:
[in this window]
[in a new window]
  		Table II. B pocket motifs for the closely related A*3001, A*3002, A*3003, A*3101, A*3303 and A*2902 alleles and the 5 A*2902-restricted published epitopes.

 
More recently, the peptide-binding motif for A*2902 has been defined (24) and also conflicts with the originally defined motif. These data indicate a broad specificity at P2, but with positively and negatively charged residues not tolerated; and a strong preference for Tyr at PC. To examine the motif that would be inferred for HLA-A*2902 based solely on the HLA amino acid sequence, a comparison was made between HLA-A*2902 and the closely related A19 alleles, A*3001, A*3002, A*3003, A*3004, A*3101, and A*3303 (25, 26), for which the motifs have also been defined. From these data, one would predict a similarly capacious B pocket characteristically containing large to medium-sized hydrophobic residues at P2 for A*2902, consistent with what is observed for the defined A*2902-restricted epitopes that all have Phe or Tyr at P2 (Table II). Thus, the approach of generating an inferred motif may facilitate rapid optimization of epitopes where a discrepancy may exist between the defined motif and the sequence of peptides restricted by the relevant allele.

To examine the possibility that the previously defined motif for A*2902 might have erroneously led to this allele being categorized as an anomaly among HLA-A alleles, being included in none of the four major HLA-A supertypes (27, 28), we subsequently performed binding assays using peptides conforming to the defined motif (Glu at P2 and Tyr/Leu at PC) and the peptides defined as A*2902-restricted epitopes (Fig. 3). The reference A*2902-binding peptide was outcompeted by all five A*2902-restricted epitope peptides, but by none of 10 peptides conforming to the published motif of Glu at P2 and Tyr/Leu at PC. These data suggest that the motif determined by inference in this case is closer to the true motif than that determined by peptide elution. The recent definition of the A*2902 motif using single amino acid-substituted peptides (24) supports the data presented here and suggests that the A*2902 motif overlaps those of the A1 and A24 supertypes.


Figure 3
View larger version (17K):
[in this window]
[in a new window]
  		FIGURE 3. Binding of A*2902-restricted epitope peptides to A*2902; lack of binding of B*4403- and B*4001-restricted peptides to A*2902. Competition of binding of the FITC-labeled reference peptide YTDGVPL-K(FITC)-Y by various nonlabeled peptides at a threshold concentration of 80 µM. The unlabeled test peptides (80 µM) were incubated along with the FITC-labeled peptide (2 nM), sHLA-A*2902 (50 µg/ml), and excess beta2-microglobulin until reaching equilibrium. Fluorescence polarization was measured and the data were analyzed by nonlinear regression analysis using the GraphPad software program to determine the maximal FP values from independent association curves. Data are expressed in percent inhibition relative to values obtained without addition of a competitor (0%). A*0101-related control peptides: C1 (ATDFKFAMY), C2 (IADMGHLKY); A29-binding peptides: #1 (YFPDWQNY), #2 (LYNTVATLY), #3 (SFDPIPIHY), #4 (SFNCRGEFFY); B*18-binding peptide: NETTPGIRYQY; B*4001-binding peptides: #1 (KDREPLTSL), #2 (KEKGGLEGL), #3 (IEELREHLL), #4 (IEVRDTKEAL); and B*44-binding peptides: #1 (REPYNEWTL), #2 (AEQATQDVKNW), #3 (GEVGFPVRPQV), #4 (IEEKAFSPEVI), #5 (AETFYVDGA).

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
These studies demonstrate the value of HLA class I allele motif inference, determined by the HLA amino acids sequence and previous x-ray crystallographic definition of the HLA residues forming the critical B and F pockets into which the peptide residues at P2 and PC preferentially bind. There is an exponential increase in the number of new alleles that are being discovered. This means that definition of motif by the labor-intensive method of peptide elution and sequencing from particular class I molecules is not feasible to keep pace with the need to understand characteristics of alleles that may be highly prevalent in particular populations of interest. In addition, the HLA nomenclature does not necessarily reflect peptide-binding motif similarities. As demonstrated here, the motif for B*3910 is more closely related to B*0702, B*4201, and B*8101 than to B*3901 or B*3902. These studies show that, for five HLA class I alleles that are highly prevalent in the Zulu/Xhosa population of South Africa, an epicenter of the HIV epidemic, the motif can be successfully inferred in this way.

The second principal finding of these studies is that HLA-A*2902, previously believed to be unique among HLA-A alleles in that it did not fit into one of the four major supertypes (27, 28), in fact has a motif that overlaps that of other HLA-A class I molecules (24). Like the closely related A*3002, Tyr/Phe are the preferred residues at P2 and Tyr is preferred at PC. There is no HLA-A allele with a motif that incorporates a negatively charged residue at P2, and thus A*2902 would be unique if, as previously indicated (11), it were so. However, the failure to demonstrate binding to A*2902 of peptides conforming to this motif (with Glu at P2), in combination with the statistical implausibility of 16 of 16 eluted peptides having Glu at P2 compared with 0 of 5 defined A*2902-restricted epitopes (p = 4.9 x 10–5), would indicate that the true motif for A*2902 is Tyr/Phe at P2 and Tyr at PC. This new motif is fully consistent with that determined for A*2902 recently by single amino acid peptide substitutions (24).

The method of motif inference requires two assumptions that require caution in this approach. First, the success of the approach depends on correctly identifying the HLA residues that form the B and F pocket for the allele in question. Furthermore, greater weight needs to be placed on particular residues lining these pockets than others. The residues used here are 9, 24, 45, 63, 66, 67, and 99 for the B pocket and 77, 80, 81, 95, and 116 for the F pocket. The second note of caution relates to the motif as being determined by the B and F pocket alone. The other pockets, in particular A, D, and C pockets, may in certain instances play a greater role than the B and F pockets. However, as noted above, it is rare for a class I allele not to have a clear preference for particular residues to bind within the B and F pockets. With these notes of caution, motif inferences for all uncharacterized HLA-A and HLA-B alleles present at a prevalence of >1% in the South Africa Zulu/Xhosa study population are shown in Tables III and IV. This approach provides a peptide-binding motif, where one previously did not exist, for 9 of 23 HLA-A alleles and for 9 of 21 HLA-B alleles. Similarly inferred motifs for HLA-C alleles have been determined (29).


View this table:
[in this window]
[in a new window]
  		Table III. Defined and predicted motifs for the HLA-A alleles present in the study of the South African patient cohort at a phenotypic frequency of >1%

 

View this table:
[in this window]
[in a new window]
  		Table IV. Defined and predicted motifs for the HLA-B alleles present in the South African patient cohort at a phenotypic frequency of >1%

 
In summary, motif inference for the increasing number of new HLA class I alleles greatly facilitates the rapid and accurate identification of optimal CTL epitopes. This knowledge of the CTL epitopes being targeted forms the fundamental starting point for subsequent studies needed to understand the pathogenesis of autoimmune diseases and viral infections such as HIV.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
The authors have no financial conflict of interest.


   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 supported by the National Institutes of Health (Contract N01-Al-15422 (HLA Typing and CTL Epitope Mapping to Guide HIV Vaccine Development) and AI46995-01A1), the Wellcome Trust (to P.J.R. Goulder and A. Leslie), the Elizabeth Glaser Pediatric AIDS Foundation (to P.J.R. Goulder), and the Doris Duke Charitable Foundation. Back

2 Address correspondence and reprint requests to Dr. Philip Goulder, Peter Medawar Building for Pathogen Research, South Parks Road, Oxford, U.K. E-mail address: philip.goulder{at}ndm.ox.ac.uk Back

3 Abbreviations used in this paper: BCL, B cell line; P2, position 2; PC, C-terminal position; s, soluble. Back

Received for publication December 5, 2006. Accepted for publication February 14, 2006.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Nixon, D. F., A. R. Townsend, J. G. Elvin, C. R. Rizza, J. Gallwey, A. J. McMichael. 1988. HIV-1 gag-specific cytotoxic T lymphocytes defined with recombinant vaccinia virus and synthetic peptides. Nature 336: 484-487. [Medline]
  2. Venet, A., B. D. Walker. 1993. Cytotoxic T-cell epitopes in HIV/SIV infection. AIDS 7: (Suppl. 1):S117-S126.
  3. Frahm, P. G. Nicole, C. Brander. 2003. Broad HIV-1 specific CTL responses reveal extensive class I binding promiscuity of HIV-1 derived optimally, defines CTL epitopes. B. T. Korber, and C. Brander, and B. F. Haynes, and R. Koup, and J. P. Moore, and B. D. Walker, and D. I. Watkins, eds. HIV Molecular Immunology and HIV/SIV Vaccine Databases 3-24. Los Alamos National Laboratory, Theoretical Biology and Biophysics, Los Alamos, NM. LA-UR 04-8162.
  4. Kiepiela, P., A. J. Leslie, I. Honeyborne, D. Ramduth, C. Thobakgale, S. Chetty, P. Rathnavalu, C. Moore, K. J. Pfafferott, L. Hilton, et al 2004. Dominant influence of HLA-B in mediating the potential co-evolution of HIV and HLA. Nature 432: 769-775. [Medline]
  5. Addo, M. M., X. G. Yu, A. Rathod, D. Cohen, R. L. Eldridge, D. Strick, M. N. Johnston, C. Corcoran, A. G. Wurcel, C. A. Fitzpatrick, et al 2003. Comprehensive epitope analysis of human immunodeficiency virus type 1 (HIV-1)-specific T-cell responses directed against the entire expressed HIV-1 genome demonstrate broadly directed responses, but no correlation to viral load. J. Virol. 77: 2081-2092. [Abstract/Free Full Text]
  6. Draenert, R., M. Altfeld, C. Brander, N. Basgoz, C. Corcoran, A. G. Wurcel, D. R. Stone, S. A. Kalams, A. Trocha, M. M. Addo, et al 2003. Comparison of overlapping peptide sets for detection of antiviral CD8 and CD4 T cell responses. J. Immunol. Methods 275: 19-29. [Medline]
  7. Steven, G. E., P. P. Marsh, L. D. Barber. 2000. The HLA Facts Book Academic Press, San Diego.
  8. Guo, H. C., D. R. Madden, M. L. Silver, T. S. Jardetzky, J. C. Gorga, J. L. Strominger, D. C. Wiley. 1993. Comparison of the P2 specificity pocket in three human histocompatibility antigens: HLA-A*6801, HLA-A*0201, and HLA-B*2705. Proc. Natl. Acad. Sci. USA 90: 8053-8057. [Abstract/Free Full Text]
  9. Madden, D. R., J. C. Gorga, J. L. Strominger, D. C. Wiley. 1992. The three-dimensional structure of HLA-B27 at 2.1 Å resolution suggests a general mechanism for tight peptide binding to MHC. Cell 70: 1035-1048. [Medline]
  10. Madden, D. R., D. N. Garboczi, D. C. Wiley. 1993. The antigenic identity of peptide-MHC complexes: a comparison of the conformations of five viral peptides presented by HLA-A2. Cell 75: 693-708. [Medline]
  11. Boisgerault, F., I. Khalil, V. Tieng, F. Connan, T. Tabary, J. H. Cohen, J. Choppin, D. Charron, A. Toubert. 1996. Definition of the HLA-A29 peptide ligand motif allows prediction of potential T-cell epitopes from the retinal soluble antigen, a candidate autoantigen in birdshot retinopathy. Proc. Natl. Acad. Sci. USA 93: 3466-3470. [Abstract/Free Full Text]
  12. Goulder, P. J., M. M. Addo, M. A. Altfeld, E. S. Rosenberg, Y. Tang, U. Govender, N. Mngqundaniso, K. Annamalai, T. U. Vogel, M. Hammond, et al 2001. Rapid definition of five novel HLA-A*3002-restricted human immunodeficiency virus-specific cytotoxic T-lymphocyte epitopes by ELISPOT and intracellular cytokine staining assays. J. Virol. 75: 1339-1347. [Abstract/Free Full Text]
  13. Altfeld, M. A., A. Trocha, R. L. Eldridge, E. S. Rosenberg, M. N. Phillips, M. M. Addo, R. P. Sekaly, S. A. Kalams, S. A. Burchett, K. McIntosh, et al 2000. Identification of dominant optimal HLA-B60- and HLA-B61-restricted cytotoxic T-lymphocyte (CTL) epitopes: rapid characterization of CTL responses by enzyme-linked immunospot assay. J. Virol. 74: 8541-8549. [Abstract/Free Full Text]
  14. Parham, P., C. E. Lomen, D. A. Lawlor, J. P. Ways, N. Holmes, H. L. Coppin, R. D. Salter, A. M. Wan, P. D. Ennis. 1988. Nature of polymorphism in HLA-A, -B, and -C molecules. Proc. Natl. Acad. Sci. USA 85: 4005-4009. [Abstract/Free Full Text]
  15. Maier, R., K. Falk, O. Rotzschke, B. Maier, V. Gnau, S. Stevanovic, G. Jung, H. G. Rammensee, A. Meyerhans. 1994. Peptide motifs of HLA-A3, -A24, and -B7 molecules as determined by pool sequencing. Immunogenetics 40: 306-308. [Medline]
  16. Barber, L. D., B. Gillece-Castro, L. Percival, X. Li, C. Clayberger, P. Parham. 1995. Overlap in the repertoires of peptides bound in vivo by a group of related class I HLA-B allotypes. Curr. Biol. 5: 179-190. [Medline]
  17. Huczko, E. L., W. M. Bodnar, D. Benjamin, K. Sakaguchi, N. Z. Zhu, J. Shabanowitz, R. A. Henderson, E. Appella, D. F. Hunt, V. H. Engelhard. 1993. Characteristics of endogenous peptides eluted from the class I MHC molecule HLA-B7 determined by mass spectrometry and computer modeling. J. Immunol. 151: 2572-2587. [Abstract]
  18. Vilches, C., L. Sanz, R. de Pablo, M. E. Moreno, S. Puente, M. Kreisler. 1996. Molecular characterization of the new alleles HLA-B*8101 and B*4407. Tissue Antigens 47: 139-142. [Medline]
  19. Falk, K., O. Rotzschke, M. Takiguchi, V. Gnau, S. Stevanovic, G. Jung, H. G. Rammensee. 1995. Peptide motifs of HLA-B38 and B39 molecules. Immunogenetics 41: 162-164. [Medline]
  20. Falk, K., O. Rotzschke, B. Grahovac, D. Schendel, S. Stevanovic, V. Gnau, G. Jung, J. L. Strominger, H. G. Rammensee. 1993. Allele-specific peptide ligand motifs of HLA-C molecules. Proc. Natl. Acad. Sci. USA 90: 12005-12009. [Abstract/Free Full Text]
  21. Rickinson, A. B., D. J. Moss. 1997. Human cytotoxic T lymphocyte responses to Epstein-Barr virus infection. Annu. Rev. Immunol. 15: 405-431. [Medline]
  22. Masemola, A. M., T. N. Mashishi, G. Khoury, H. Bredell, M. Paximadis, T. Mathebula, D. Barkhan, A. Puren, E. Vardas, M. Colvin, et al 2004. Novel and promiscuous CTL epitopes in conserved regions of Gag targeted by individuals with early subtype C HIV type 1 infection from southern Africa. J. Immunol. 173: 4607-4617. [Abstract/Free Full Text]
  23. Wilson, C. C., S. A. Kalams, B. M. Wilkes, D. J. Ruhl, F. Gao, B. H. Hahn, I. C. Hanson, K. Luzuriaga, S. Wolinsky, R. Koup, et al 1997. Overlapping epitopes in human immunodeficiency virus type 1 gp120 presented by HLA A, B, and C molecules: effects of viral variation on cytotoxic T-lymphocyte recognition. J. Virol. 71: 1256-1264. [Abstract]
  24. Sidney, J., S. Southwood, A. Sette. 2005. Classification of A1- and A24-supertype molecules by analysis of their MHC-peptide binding repertoires. Immunogenetics 57: 393-408. [Medline]
  25. Krausa, P., C. Munz, W. Keilholz, S. Stevanovic, E. Y. Jones, M. Browning, M. Bunce, H. G. Rammensee, A. McMichael. 2000. Definition of peptide binding motifs amongst the HLA-A*30 allelic group. Tissue Antigens 56: 10-18. [Medline]
  26. Falk, K., O. Rotzschke, M. Takiguchi, B. Grahovac, V. Gnau, S. Stevanovic, G. Jung, H. G. Rammensee. 1994. Peptide motifs of HLA-A1, -A11, -A31, and -A33 molecules. Immunogenetics 40: 238-241. [Medline]
  27. Lund, O., M. Nielsen, C. Kesmir, A. G. Petersen, C. Lundegaard, P. Worning, C. Sylvester-Hvid, K. Lamberth, G. Roder, S. Justesen, et al 2004. Definition of supertypes for HLA molecules using clustering of specificity matrices. Immunogenetics 55: 797-810. [Medline]
  28. Sette, A., J. Sidney. 1999. Nine major HLA class I supertypes account for the vast preponderance of HLA-A and -B polymorphism. Immunogenetics 50: 201-212. [Medline]
  29. Yusim, K. S. J., I. Honeyborne, C. Calef, P. J. Goulder, B. T. Korber. 2004. Enhanced motif scan: a tool to scan for HLA anchor residues in proteins. B. T. Korber, and C. Brander, and B. F. Haynes, and R. Koup, and J. P. Moore, and B. D. Walker, and D. I. Watkins, eds. Immunology and HIV/SIV Vaccine Databases 24-36. Los Alamos National Laboratory, Theoretical and Biophysics, Los Alamos, NM. LA-UR 04-8162.



This article has been cited by other articles:


Home page
 J. Virol.Home page
I. Honeyborne, A. Prendergast, F. Pereyra, A. Leslie, H. Crawford, R. Payne, S. Reddy, K. Bishop, E. Moodley, K. Nair, et al.
Control of Human Immunodeficiency Virus Type 1 Is Associated with HLA-B*13 and Targeting of Multiple Gag-Specific CD8+ T-Cell Epitopes
J. Virol., April 1, 2007; 81(7): 3667 - 3672.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
A. Leslie, D. A. Price, P. Mkhize, K. Bishop, A. Rathod, C. Day, H. Crawford, I. Honeyborne, T. E. Asher, G. Luzzi, et al.
Differential Selection Pressure Exerted on HIV by CTL Targeting Identical Epitopes but Restricted by Distinct HLA Alleles from the Same HLA Supertype
J. Immunol., October 1, 2006; 177(7): 4699 - 4708.
[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 Honeyborne, I.
Right arrow Articles by Goulder, P. J. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Honeyborne, I.
Right arrow Articles by Goulder, P. J. R.
Right arrowPubmed/NCBI databases
*Substance via MeSH

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS