Respiratory Syncytial Virus Nucleoprotein ... - Journal of Virology

Download PDF
4 downloads 0 Views 1MB Size Report
Comment
Nov 15, 2002 - field strains representing the diversity of circulating genotypes. ...... Adamus, M. L. Clements-Mann, D. O. Harris, V. B. Randolph, S. A. Udem,.
JOURNAL OF VIROLOGY, July 2003, p. 7319–7329 0022-538X/03/$08.00⫹0 DOI: 10.1128/JVI.77.13.7319–7329.2003 Copyright © 2003, American Society for Microbiology. All Rights Reserved.

Vol. 77, No. 13

Respiratory Syncytial Virus Nucleoprotein-Specific Cytotoxic T-Cell Epitopes in a South African Population of Diverse HLA Types Are Conserved in Circulating Field Strains Marietjie Venter,1 Michael Rock,2 Adrian J. Puren,1 Caroline T. Tiemessen,1 and James E. Crowe, Jr.2* National Institute for Communicable Diseases, Sandringham, South Africa,1 and Vanderbilt University Medical School, Nashville, Tennessee2 Received 15 November 2002/Accepted 2 April 2003

This study identifies memory cytotoxic T lymphocyte (CTL) epitopes to respiratory syncytial virus (RSV) in healthy South African adults and demonstrates the conservation of those epitopes in circulating field strains of RSV in South Africa. Thirty-seven healthy adults from a population with diverse HLA backgrounds were screened by gamma interferon (IFN-␥) enzyme-linked immunospot for memory CTL activity in response to overlapping peptides representing the complete nucleoprotein (N) of RSV. Responses of more than 40 spotforming cells/million cells were detectable in 21 individuals. The significant responses were further characterized, and 14-mer peptides were identified that induced cytolytic activity. Fine mapping of peptides with the highest cytolytic activity identified an HLA-B*08-restricted RSV-specific CTL epitope. The extended 14-mer peptide containing this epitope also induced lysis in the context of A*02-restricted target cells in some individuals. These HLA types are common in the target population; thus, the epitope is useful for studies of CTL responses to RSV in humans. The epitope was detected in healthy adults, reflecting the response generated in the course of previous natural RSV infection. We obtained a large panel of naturally occurring isolates of RSV to determine whether there was evidence of escape from CTL activity in circulating strains. We found that this epitope and a previously identified B*07-restricted N protein epitope were conserved in RSV field strains representing the diversity of circulating genotypes. This work suggests that escape from CTL activity is not common for this acute respiratory infection. in humans (6, 17). Studies in humans have been limited by difficulties imposed by the classical techniques used to enumerate CTL responses. These limitations include the small blood sample volume that can be obtained safely from acutely infected young children and the low frequency of RSV-specific memory CTL response in adults who are not recently infected (24). The availability of more-sensitive technologies has now made it possible to study virus-specific CTL responses in humans. These techniques include the use of peptide epitope stimulation of T cells followed by measurement of gamma interferon (IFN-␥) secretion (34) by enzyme-linked immunospot (ELIspot) or intracellular cytokine staining (ICS) by flow cytometry. The enumeration of peptide-specific responses is also possible by using tetrameric major histocompatibility complex (MHC) class I molecules for binding specific epitopes (4). In order to use these techniques to study infants, RSV-specific CTL epitopes need to be identified that can be used as peptide reagents in such assays. Recent studies have shown that memory responses in adults can be used to identify RSV-specific epitopes with IFN-␥ ELIspot (24). The attraction of this strategy is the availability of larger volumes of blood from adults, which allows further characterization of the identified epitopes by using chromium release assays to confirm HLA-restricted cytotoxicity. To date, three human CTL epitopes have been defined for RSV: one HLAB*07-restricted epitope in the N protein (24) and two epitopes in the F protein restricted to HLAB*57 and HLACw*12, respectively (12). South Africa has a very diverse population consisting of Southern African black ethnic groups,

Respiratory syncytial virus (RSV) is the major viral cause of pneumonia and wheezing in infants and children in both developed and developing countries (20, 49, 50). Previous studies with the murine model suggest that an RSV-specific type 1 T-helper cell response and CD8⫹ cytotoxic T lymphocytes (CTLs) are critical for the control of RSV infection (15). It is thought that activation of Th2 CD4⫹ T cells is a key factor in disease pathogenesis, while a Th1 CD4⫹ T-cell response with concomitant induction of CD8⫹ T cells is needed for an effective immune response that resolves infection (reviewed in reference 25). However, the adoptive transfer of activated CTLinduced immunopathology in mice suggests that a fine balance exists between protective and disease-enhancing effects of RSV-specific T cells (15, 16). In the past, most studies of the CTL response to RSV focused on murine T cells (reviewed in reference 49). CTLs also have been documented in humans following RSV infection (6, 18) and may play an important role in viral clearance. RSV proteins previously shown to induce CTL activity include the N protein, the fusion (F) protein, the matrix 2 (M2) protein, and the short hydrophobic (SH) protein (6, 14, 17). The M2 protein induces the dominant response in mice of the H-2d haplotype (36). The N protein, which is the most highly conserved protein between the two RSV antigenic subgroups, appears to be the major target for the memory CTL response * Corresponding author. Mailing address: D-7235 Medical Center North, 1161 21st Ave. South, Nashville, TN 37232-2581. Phone: (615) 343-8064. Fax: (615) 343-9723. E-mail: [email protected]. 7319

7320

VENTER ET AL.

Caucasians of Dutch, French, German, and British origin, Cape Coloreds, and Indians with a wide diversity of HLA types (22, 26). In order to further characterize the CTL response in children, commonly recognized epitopes that are restricted to HLA types of the study population need to be defined. The use of epitope-specific CTL assays will contribute to vaccine development and provide a better understanding of immuneinduced disease pathogenesis. In this study, peripheral blood samples from healthy South African adults were screened with overlapping peptides covering the complete N protein of RSV by IFN-␥ ELIspot. The significant responses were further characterized, and the epitope that elicited the strongest cytolytic activity was mapped with truncated peptides to identify the optimal epitope. We determined the sequence of a major part of the N protein gene of a panel of isolates representative of all RSV genotypes identified in South Africa to date in order to determine if evidence of immune-induced selection could be found in the identified CTL epitopes. MATERIALS AND METHODS Subjects. Thirty-seven healthy South African adult volunteers between 18 and 50 years of age who represented a wide range of South African ethnic groups were recruited from Johannesburg, South Africa, for this study. These subjects included South-Sotho, North-Sotho, Tswana, Zulu, Swazi, Ndebele, Xhosa, Cape Colored, Caucasian, and Indian individuals. The study was approved by the Committee for Research in Human Subjects (Medical) of the University of the Witwatersrand, Johannesburg, South Africa, and the Committee for the Protection of Human Subjects of Vanderbilt University Medical Center. EBV-transformed BCL. Blood was collected from volunteers, and peripheral blood mononuclear cells (PBMCs) were isolated by Ficoll-Hypaque density gradient centrifugation. PBMCs were used to establish Epstein-Barr virus (EBV)-transformed B cell lines (BCL) by standard techniques. Briefly, 1 ⫻ 106 to 10 ⫻ 106 PBMCs were resuspended in 5 ml of RPMI medium containing 20% heat-inactivated fetal calf serum (FCS), 0.01% 200 mM L-glutamine, 0.01% 100 mM Na pyruvate, and 0.01% (vol/vol) 10,000-U/ml penicillin and 10-mg/ml streptomycin stocks. Five milliliters of filtered EBV supernatant prepared from the EBV-infected marmoset cell line B95-8 and 10 ␮l of cyclosporine A (1 mg/ ml) were added, and the cells were incubated at 37°C with 5% CO2. Once transformation was apparent, EBV-transformed BCL were expanded and maintained in complete medium. Peptide-specific T cell lines. Peptide-stimulated lymphocyte cultures were established from healthy adult subjects as previously described (31). Briefly, 20 ⫻ 106 PBMCs were pulsed with the peptide of interest at a concentration of 100 ␮M for 1 h at 37°C in 5% CO2. Cells were diluted to 2 ⫻ 106 in RPMI containing 10% FCS and 25 ng of recombinant human interleukin 7 (Endogen, Woburn, Mass.)/ml, plated out at 4 ⫻ 106 cells/well in 24-well plates, and incubated at 37°C in 5% CO2. Recombinant interleukin 2 (10 U/ml) (Roche Diagnostics, Mannheim, Germany) was added on days 3 and 7. These concentrations were chosen because they were previously shown to enhance peptide-specific lysis without increasing nonspecific lysis. The expanded T cell lines were tested for cytolytic activity on day 10. Synthetic peptides. A panel of 127 synthetic peptides was generated that spans the complete RSV N protein of the RSV wild-type strain A2. The overlapping peptides were 14-mers offset by 3 amino acids (i.e., overlapping by 11) and were named N1 through N127. The starting position of each peptide corresponds to amino acid (n ⫻ 3 ⫺ 2), where n is the peptide number. Immunograde 14-mer peptides (Cleaved PepSet; Mimotopes, Clayton, Australia) were used for screening purposes. After the initial ELIspot screen, selected 9-mer or 10-mer peptides were synthesized and high-performance liquid chromatography purified to more than 90% purity (Natural and Medical Sciences Institute at the University of Tuebingen, Tuebingen, Germany). Purified peptides were used for chromium release assays and ICS assays. The CTL epitope prediction algorithm software programs BIMAS (http://bimas.dcrt.nih.gov/molbio/hla_bind/) (37) and SYFP EITHI (http://syfpeithi.bmi-heidelberg.com/) (40) were used to assist in the selection of 9- or 10-mer peptides used for mapping of the optimal length of epitopes identified. IFN-␥ ELIspot assay. ELIspot assays were performed as described previously (24). Briefly, 96-well plates (IP Opaque; Millipore, Bedford, Mass.) were coated

J. VIROL. with anti-IFN-␥ monoclonal antibody (clone 1-DIK; Mabtech, Stockholm, Sweden) at 4°C overnight. The following day, plates were washed three times with sterile phosphate-buffered saline before blocking with 10% FCS–RPMI. Fresh PBMCs (2 ⫻ 105) and 10 ␮M individual peptides were added to the wells and incubated overnight at 37°C. The next day, cells were washed and incubated for 3 h with a biotinylated anti-IFN-␥ monoclonal antibody (clone 7-B6-1 biotin; Mabtech). Following washing, streptavidin-conjugated horseradish peroxidase (Pharmingen BD Bioscience) was added for 1 h. The plates were washed, and spot-forming cells (SFC) were detected after a 40-min reaction with NovaRed developing substrate (Vector, Burlingame, Calif.). The number of IFN-␥-specific T cells was calculated by subtracting the negative-control value, and the results were expressed as SFC per million PBMCs (SFC/106 PBMCs). ICS. For ICS, 2 ⫻ 106 freshly isolated PBMCs were incubated with 10 ␮M peptides and 1 ␮g each of the stimulatory monoclonal antibodies anti-CD28 and anti-CD49d (Becton Dickinson)/ml for 2 h at 37°C in 5% CO2. Brefeldin A (10 ␮g/ml; Sigma, St. Louis, Mo.) was added, and the cells were incubated for a further 4 h before being placed at 4°C overnight. The next day, phosphatebuffered saline–2 mM EDTA solution was added, and the cells were vortexed and incubated for 10 min at 37°C before washing. PBMCs were lysed and permeabilized with BD fluorescence-activated cell sorter (FACS) lysing and permeabilization 2 solutions (Becton Dickinson) according to the manufacturer’s directions. Cells were blocked in 4% mouse serum for 30 min at room temperature before being washed and stained with peridinin chlorophyll protein (PerCP), allophycocyanin (APC), fluorescein isothiocyanate (FITC), and phycoerythrin (PE) fluorescent conjugated antibodies, anti-CD3-APC, anti-CD8PerCP, anti-CD69-FITC, and anti-IFN-␥-PE (Becton Dickinson), for 30 min. Cells were washed before being fixed in 1% paraformaldehyde. Control conditions were established by the use of PBMCs that had not been stimulated by peptides. Cells were analyzed on a FACSort flow cytometer (Becton Dickinson Immunocytometry Systems), with CellQuest and Paint-A-Gate software to measure PerCP, APC, FITC, and PE as fluorescent parameters. Cytotoxicity assay. Chromium release assays were performed by using peptidestimulated CTL lines as effector cells and autologous BCL as target cells. Briefly, BCL were washed 12 h prior to the assay in RPMI 1640 containing 20% FCS (R20) and resuspended in fresh R20. On the day of the assay, 2 ⫻ 106 BCL target cells were labeled with 100 ␮Ci of Na251CrO4 for 1 h at 37°C. After washing cells three times with cold RPMI 1640 containing 10% FCS (R10), selected peptides were added at 10 ␮M for a further 1 h. Cells were washed again before dilution to 1 ⫻ 105 cells/ml. Effector cells were added at various effector-to-target cell (E:T) ratios (50:1, 25:1, and 12.5:1) to a fixed number (1 ⫻ 105) of 51Cr-labeled target cells in 96-well culture plates in a final volume of 200 ␮l. In control wells, only R10 was added to target cells (negative control) or 5% Triton X was added with no effector cells (maximum lysis). After 4 h of incubation at 37°C in 5% CO2, the supernatants were harvested and transferred to 96-well Luma plates for solid scintillation counting. Chromium release was measured by using a Top Count Microplate scintillation counter (Packard Instrument Company, Meridien, Conn.), and the percent specific lysis was calculated by the following formula: percent specific lysis ⫽ [(experimental release ⫺ spontaneous release)/ (maximum release ⫺ spontaneous release)] ⫻ 100. The results are reported as the means of triplicate values. To determine the HLA restriction of epitopes, BCL were used as target cells that matched the HLA type of the effector cells in only one allele. To demonstrate that peptide-specific effector cells recognized the epitope after proteolytic processing of the whole N protein in the cell, target cells were prepared by infecting BCLs with a recombinant vaccinia virus expressing the RSV N protein (obtained from J. Beeler, Food and Drug Administration), or a vaccinia-human immunodeficiency virus (HIV) Nef (vT197) construct or wildtype NYCBH strain vaccinia virus (negative controls) at a multiplicity of infection of 5 PFU in 1 ml for 1 h at 37°C in 5% CO2. The vaccinia-HIV Nef recombinant virus (96ZM651 Nef, subtype C) was obtained through the AIDS Research and Reference Reagent Program, Division of AIDS, National Institute of Allergy and Infectious Diseases, National Institutes of Health; vT197 was from Therion Biologics Corporation. Excess vaccinia virus was removed by washing with R10, and cells were placed in 10 ml of R10 overnight in a 25-cm2 area tissue culture flask. The following day, a chromium release assay was performed as described previously. Cold unlabeled competitor target cells were added to the assay plates at a cold/hot target cell ratio of 40:1. Subset depletion studies. Cells were sorted by using magnetic Dynabeads coated with anti-CD4 or anti-CD8 monoclonal antibodies according to the manufacturer’s recommendations (Dynal Biotech, Oslo, Norway). Briefly, PBMCs or peptide-specific cell lines were labeled with prewashed T cell subset-specific Dynabeads at a bead-to-cell ratio of 5:1. Following 15 min of incubation at 4°C,

VOL. 77, 2003

RSV HUMAN CTL EPITOPES IN SOUTH AFRICAN SUBJECTS

cells were washed with RPMI containing 1% FCS before the beads were adhered to a magnet for 2 min. The supernatant was collected; the cells were washed and then concentrated by centrifugation for their subsequent use in chromium release or ELIspot assays. Cells sorted with anti-immunoglobulin G or anti-CD19 beads were used to control for the amount of cells used in further experiments. The depletion efficiency was measured by staining with fluorescent labeled CD4or CD8-specific antibodies followed by flow cytometric analysis. HLA typing. HLA typing was performed by low-resolution sequence-specific primer PCR in the HLA Laboratory, National Institute for Communicable Diseases, Sandringham, South Africa. Viruses and RNA. We selected 14 South African RSV isolates isolated from 1997 to 2001 for N protein sequence analysis in this study from a large panel of isolates that had been recently described (45, 46). We chose isolates that represented each of the subgroup A and B genotypes described to date in South Africa, based on genotyping with sequences from the variable surface protein G (the attachment protein). These included subgroup A genotypes GA2, GA5, GA7, and SAA1 and subgroup B genotypes GB4, GB3, SAB1, SAB2, and SAB3. RT-PCR for amplification of viral genomic sequences. Total RNA was isolated from frozen nasopharyngeal aspirates as described previously (45, 46). In brief, nasopharyngeal aspirates stored at ⫺70°C were thawed on ice and immediately supplemented with 20 U of RNase inhibitor (Roche Molecular Biochemicals). Total RNA was extracted directly from 200 ␮l of clinical specimen by use of the viral RNA kit (Roche Molecular Biochemicals) according to the manufacturer’s recommendations. In the case of very viscous specimens, the lysates were homogenized by using QIAshredder homogenizer columns (Qiagen, Hilden, Germany). Primers were designed with Primer 3 software (http://www-genome.wi .mit.edu/cgi-bin/primer/primer3_www.cgi/) by using the prototype sequences A2 and B1 for subgroups A and B, respectively. The primers were as follows: subgroup A 5⬘ primer, N73AF (TCCAGCAAATACACCATCCA); subgroup A 3⬘ primer, N1017AR (CATTATGCCTAGGCCAGCAG); subgroup B 5⬘ primer, N2BF (GGGAAATACAAAGATGGCT); subgroup B 3⬘ primer, N1031BR (A TGCCTAGACCTGCTGCATT). The reactions were performed separately for subgroups A and B in a one-step reverse transcription (RT)-PCR by using the Titan One Tube RT-PCR system (Roche Molecular Biochemicals) according to the manufacturer’s instructions. In brief, 10 ␮l of RNA was added to 10 ␮l of standard 5⫻ reaction buffer, 10 ␮M concentrations of each deoxynucleoside triphosphate, 20 pmol of each of the 5⬘ and 3⬘ primers, 5 mM dithiothreitol solution, 10 U of RNase inhibitor, and 1 ␮l of the Titan enzyme mix, and the volume was brought up to 50 ␮l with distilled water. The reaction was performed in a GeneAmp PCR system 9600 thermocycler (Applied Biosystems, Foster City, Calif.) according to the following program: 50°C for 30 min; 94°C for 2 min; 35 cycles of 94°C for 30 s, 55°C for 30 s, and 68°C for 1 min; and 68°C for 7 min. PCR products were visualized by agarose gel electrophoresis. Nucleotide sequence determination and analysis. PCR products were purified by using the Qiaquick 8 PCR purification kit (Qiagen). DNA products for sequence analysis were purified with Autoseq GE-50 columns (AEC-Amersham, Sandton, South Africa). Nucleotide sequence analysis was carried out on both DNA strands by using ABI Prism big dye terminator cycle sequencing on an ABI 3100 sequencer model according to the manufacturer’s specifications (Applied Biosystems). Forward and reverse reactions were carried out with the same primers that were used for RT-PCR. An additional primer, N454F (CAGAAT ACAGGCATGACTCT), was designed to assist in sequencing the 3⬘ ends where necessary. Sequences were analyzed with Sequencher software, version 4.0.5. Sequence alignments were performed with Clustal X, version 1.64b (42). Translations and phylogenetic analysis were performed with MEGA, version 2.1 (30). Shading of amino acid alignments was carried out with Genedoc, version 2.6.002 (www.psc .edu/biomed/genedoc). Sequences used for comparison were obtained from GenBank: human RSV prototype subgroup A (A2) (19) and subgroup B (B1) (28), bovine RSV (13, 32, 43), and ovine RSV (2). Little conservation was noted when comparisons were performed with sequences of related pneumoviruses: mouse pneumovirus (8), human metapneumovirus (44), or avian metapneumovirus (41) (data not shown). Statistical analysis. Data were analyzed by using SPSS (Chicago, Ill.) software, version 11.5.0. A general linear model analysis of repeated measures was used to generate P values. Probabilities of ⬍0.01 were considered statistically significant. All data are presented with standard deviations. Nucleotide sequence accession numbers. The 14 N sequences from distinct RSV genotypes that we determined have been deposited in GenBank with the accession numbers AY151194 through AY151207.

7321

FIG. 1. An ELIspot screening assay was used to test PBMCs obtained from 37 healthy adult subjects for IFN-␥ production against a panel of 127 individual peptides spanning the RSV N protein. Peptides inducing a response of ⬎40 SFC/million PBMCs in at least four subjects are indicated.

RESULTS Identification of RSV-specific T-cell responses by IFN-␥ ELIspot and ICS. IFN-␥ ELIspot and ICS focused on the RSV nucleoprotein (N) as a potential target based on earlier work. The N protein is the most highly conserved protein between the two antigenic subgroups and appears to be the major target for the memory CTL response in humans (6, 17). An HLArestricted RSV N-protein-specific CTL epitope recently has been characterized from a healthy adult subject (24). The approach we adopted for identification of novel RSV-specific CTL epitopes included the use of a panel of 127 overlapping 14-mer peptides spanning the N protein. These peptides were used to screen 37 healthy South African adults from various ethnic groups by IFN-␥ ELIspot assays. Of those subjects tested, 21 subjects exhibited a peptide-specific response greater than 40 SFC/106 cells while 17 reacted with at least one peptide with more than 50 SFC/106 cells. The 14-mer peptides that exhibited reactivity with the highest number of subjects by ELIspot were peptides N16, N78, N84, N85, N86, and N100 (Fig. 1 and Table 1), and these were selected for further characterization. The HLA types of donors that reacted to these peptides are listed in Table 1. Low frequencies of IFN-␥-producing CD8⫹ T cells could be detected in some subjects by ICS after stimulation with the individual peptides N16, N78, N84, N85, and N100 (data not shown). The frequency of RSV N-peptide-specific CD8⫹ T cells induced by the selected peptides was low, ranging from 0.02 to 0.09% of CD8⫹ T lymphocytes. The low precursor frequency of responding cells was not surprising given that these responses were elicited in healthy subjects who were not known to be recently reinfected. The ICS data confirmed the ELIspot results and suggested that the 14-mer peptides contain RSV N-specific CTL epitopes. The N84/N85/N86 cluster of nested peptides that was detected in a significant number of subjects by ELIspot (Fig. 1) also induced the strongest and most reproducible IFN-␥ production as detected by ICS (Fig.

7322

VENTER ET AL.

J. VIROL. TABLE 1. Peptides that elicited IFN-␥ ELIspot responses in four or more subjects

Position on N protein

Peptide

Subject

46

N16

D1 D5 D7 D8 D10 D11 D16 D24

KLCGMLLITEDANH

D1 D5 D8 D10 D16 D24 D25

STRGGSRVEGIFAG

D16 D35 D37 D38

AYGAGQVMLRWGVL

D5 D16 D28 D31 D35 D37

AGQVMLRWGVLAKS

232

250

253

N78

N84

N85

Sequence

HLA type(s)a

No. of SFCb

Ethnic group

3, 7 7, 17

176* 182* 105* 75* 90* 65* 65* 40*

South Sotho Tswana Zulu South Sotho Cape Colored Tswana Caucasian North Sotho

44, 57 15, 35 7, 58 15, 40 8, 15 8, 42 15, 49

4, 4 6, 2, 3, 7, 3,

59* 112* 35* 120* 135* 40 268*

South Sotho Tswana South Sotho Cape Colored Caucasian North Sotho North Sotho

8, 15

3, 7

8, 58

6, 7

24*c 250* 45 105

Caucasian Caucasian Zulu Caucasian

15, 35 8, 15 7, 55 18, 44

4 3, 7 7, 9 7, 12

8, 58

6, 7

59 153 90 146 233 75

Tswana Caucasian Caucasian Caucasian Caucasian Zulu Tswana Caucasian Caucasian Caucasian Caucasian

A

B

Cw

30, 34 43, 68 NDd 23, 68 2, 26 ND 2, 24 30, 68

44, 57 15, 35

4, 18 4

7, 58 15, 40

6, 7 2, 4

8, 15 8, 42

30, 34 43, 68 23, 68 2, 26 2, 24 30, 68 2, 68 2, 24 ND 30 ND 43, 68 2, 24 3, 11 2, 25 ND 30

18 7 4 7 17 7

256

N86

D5 D16 D12 D31 D38

VMLRWGVLAKSVKN

43, 68 2, 24 31 2, 25 ND

15, 35 8, 15 8, 60 18, 44

4 3, 7 7, 10 7, 12

82 112 45 190 56

298

N100

D10 D25 D31 D40

AGFYHILNNPKASL

2, 26 2, 68 2, 25 24, 31

15, 15, 18, 35,

2, 4 3, 7 7, 12 12

40 43 56 50

40 49 44 52

Cape Colored North Sotho Caucasian Indian

a

HLA typing was performed on cells from subjects for whom EBV cell lines could be established. IFN-␥ ELIspot results are numbers of SFC/million PBMCs. The fine-spot phenotype is indicated by an asterisk; the rest of the spots had a large-spot phenotype. Subject repeatedly displayed reactivity to this peptide in subsequent assays. d ND, not determined. b c

2) Therefore these peptides were characterized in the most detail. Characterization of an HLA-B*08-restricted RSV N-proteinspecific CTL epitope. ELIspot and ICS assays measure the ability of peptide-stimulated T cells to produce IFN-␥ but not necessarily the ability to kill target cells. Therefore, the selected peptides were characterized further by standard chromium release assays. Peptide-specific CTL lines were generated by in vitro stimulation and used as effector cells after 10 days in culture. The highly characterized HLA-A*02-restricted influenza matrix protein epitope, designated MP.58 (GILG FVFTL) (9), was used as a positive control, and the HLAA*02-restricted HIV reverse transcriptase epitope, designated RT.476 (ILKEPVHGV) (48), was used as a negative control in subject D16, for whom the HIV status was known to be negative, with permission of the subject. N84- and N85-peptide-specific cell lines displayed the most significant lysis of peptide-pulsed autologous BCL targets at

multiple E:T ratios. The cytolytic activity observed from the N84 and N85 cell lines was comparable to that observed from the MP.58-peptide-specific cell line generated from the same subject (Fig. 3a). As expected, cytolytic activity was not detected for RT.476-specific cell lines against autologous RT.476-pulsed BCL targets or against target cells that were not pulsed with peptide. We next sought to determine the HLA restriction(s) of the epitope(s) contained within the N84/N85 peptide sequences by comparing lysis of autologous BCL target cells with lysis of single HLA class I-matched peptide-pulsed target cells (Fig. 3b and data not shown). N84- and N85-peptide-stimulated cells from donor 16 reliably lysed HLA-B*08-matched target cells or cells that were HLA-B*08/Cw*07 matched but not cells that were only HLA-Cw*07 matched (data not shown). Interestingly, N85-peptide-specific cell lines also displayed significant cytolytic activity against HLA-A*02-matched BCL target cells in the majority of assays tested with subject D16. However, the

VOL. 77, 2003

RSV HUMAN CTL EPITOPES IN SOUTH AFRICAN SUBJECTS

FIG. 2. Analysis of peptide-reactive CD8⫹ T cells from subject D16. PBMCs were incubated in the presence or absence of the indicated peptides for 6 h in an IFN-␥ ICS assay. The data are presented as the percentages of CD3⫹ CD8⫹ lymphocytes that were CD69⫹ and scored positive for IFN-␥ production. The results in the upper left panel (labeled control) were the results obtained before peptide stimulation.

9- or 10-mer peptide-specific cell lines generated within the overlap of N84 and N85 did not induce significant HLA-A*02restricted cytolytic activity (Fig. 3d and data not shown), suggesting a potential HLA-A*02-restricted epitope within the C-terminal region of N85 or that the epitope is longer than 10 amino acids. The previous data indicated that an HLA-B*08-restricted T-cell epitope exists within the overlap of peptides N84 and N85. To identify the optimal peptide that induced the MHCrestricted CTL response, peptide-specific cell lines for the 9and 10-mer peptides corresponding to the overlapping region of the two 14-mer peptides N84 (amino acids 250 to 263) and N85 (amino acids 253 to 266) were generated from subject D16. These cell lines were then assayed for cytolytic activity against peptide-pulsed HLA-B*08 single matched BCL targets (Fig. 3c). Significant cytolytic activity was observed for the cell lines generated from the 9- and 10-mer peptides at amino acid position 255 (designated N09.255 and N10.255, respectively). However, the N09.255-peptide-specific cell line exhibited a nearly twofold-higher cytolytic activity against peptide-pulsed HLA-B*08 targets, suggesting that this is the optimal epitope. The HLA-B*08 restriction for peptide N09.255 was further confirmed against partially matched peptide-pulsed BCL targets (Fig. 3d and data not shown). Significantly, cytolytic activity was only observed for the N09.255-peptide-specific cell line against peptide-pulsed autologous or HLA-B*08-matched target cells, further suggesting that N09.255 is an HLA-B*08restricted CTL epitope. The 9- and 10-mer peptides that were shown to contain the HLA class I epitope were again tested by IFN-␥ ELIspot to confirm a reaction is induced to a CD8⫹ epitope by our screening method. Reaction mixtures containing 40 and 45 SFC/106 cells were induced by these peptides in

7323

donor 16, which was a higher number of SFC than that used in the reaction mixture induced by N84. To confirm that the cytolytic activity observed was mediated by CD8⫹ T cells, N09.255-peptide-specific lines were depleted on day 10 of culture by using paramagnetic beads for CD4⫹, CD8⫹, or CD19⫹ cells and then used as effector cells in chromium release assays (Fig. 3e). N09.255-peptide-specific cell lines depleted of CD4⫹ T cells or depleted of CD19⫹ cells were equally capable of cytolysis of peptide-pulsed BCL targets. In contrast, cytolytic activity of N09.255 cell lines depleted of CD8⫹ T cells was reduced to that seen for the negative control. The efficacy of the depletion was confirmed by flow cytometry and found to be more than 99% effective (data not shown). These data confirm that N09.255 is an MHC class I-restricted CTL epitope for CD8⫹ T cells. To determine whether the putative CTL epitope identified with synthetic peptides is presented after processing of the viral protein in cells, target cells were infected with a recombinant vaccinia virus expressing the RSV N protein (Vac-N) and tested by chromium release assay. N09.255-peptide-specific effector cells lysed target cells infected with Vac-N but not target cells infected with wild-type vaccinia virus or an HIV Nef recombinant vaccinia virus (Fig. 3f). Effector cells stimulated with the N13 control peptide did not lyse Vac-N-infected target cells (data not shown). This finding suggested that the N09.255 HLA-B*08-restricted CTL epitope is naturally processed and presented in target cells. Lack of evidence for immune selection at RSV N protein CTL epitopes. Isolates were selected from all subgroup A and B genotypes identified to date in South Africa as well as the two prototype strains A2 and B1 for inclusion in the study. The RT-PCR resulted in a specific band of 944 bp for all subgroup A isolates or 1,029 bp for all subgroup B isolates, confirming the sensitivity and specificity of the selected primers. The amino acid sequences of the N proteins of the selected subgroup A and B isolates were compared in alignments with the amino acid sequences of other members of the Pneumovirinae, including the Pneumovirus genus (human RSV, bovine RSV, and ovine RSV). Similar alignments with sequences from pneumovirus of mice and the metapneumovirus genus (avian and human metapneumoviruses) suggested little conservation of sequence between RSV and these viruses at the site of the RSV epitopes (data not shown). The 14 N sequences from distinct RSV genotypes that we determined (Fig. 4) are deposited in GenBank with the accession numbers AY151194 through AY151207. The average between-group amino acid P distances (calculated with Mega, version 2) between human RSV subgroups A and B was 5% while the distances between human RSV and bovine or ovine RSV were 6 and 7%, respectively. Much less conservation existed with other members of the Pneumovirinae. The distances between RSV and pneumovirus of mice and the metapneumoviruses were 38 and 59%, respectively. When only taking the human RSV genotypes into consideration, much more variation was found between the two subgroups on the nucleotide level (P ⫽ 0.14) than on the amino acid level (0.05), with the average synonymous substitutions per synonymous site (Ks ⫽ 0.23) greatly exceeding the nonsynonymous substitutions per nonsynonymous site (Ka ⫽ 0.01) (Ka/Ks ⫽ 0.04) and transitions exceeding transversions (Ts/Tv ⫽ 4) (calculated with Mega, version 2) (30). This find-

FIG. 3. Recognition of viral epitopes by peptide-specific effector cells generated from subject D16 (HLA type indicated) in standard 51Cr release assays. (a) Effectors were Flu MP.58, HIV RT.476, RSV N.84, or RSV N.85 peptide-specific cell lines. Target cells were autologous BCL pulsed in the presence (closed symbols) or absence (open symbols) of the indicated peptides. (b, d) Effector cells were RSV N.84 or RSV N.85 (b) or RSV N09.254 or RSV N09.255 (d) peptide-specific cell lines. Target cells were autologous or HLA single-matched BCL (corresponding HLA alleles are indicated) pulsed in the presence of the indicated peptides. (c) Peptide-specific effector cells were generated and assayed for cytotoxicity against HLA-B*08-matched BCL target cells pulsed with the indicated peptide. (e) RSV N09.255-peptide-specific effector cells were depleted of CD4⫹, CD8⫹, or CD19⫹ cells by paramagnetic beads and assayed for cytotoxicity against HLA-B*08-restricted BCL pulsed in the presence or absence of peptide RSV N09.255. (f) Effector cells were RSV N09.255-peptide-specific cell lines. Target cells were HLA-B*08restricted BCL that were pulsed in the presence or absence of peptide RSV N09.255 or infected with recombinant vaccinia virus expressing the HIV Nef protein or the RSV N protein. The data are presented as the percentages of specific lysis at an E:T ratio of 25/1 (b to e) or at the indicated E:T ratios (a and f). *, percent specific lysis results are statistically significant from each other (P ⬍ 0.01). 7324

7325 RSV HUMAN CTL EPITOPES IN SOUTH AFRICAN SUBJECTS VOL. 77, 2003

FIG. 4. Amino acid alignment of the N protein sequences of human RSV subgroup A and B genotypes with those of bovine RSV, ovine RSV, pneumovirus of mice, and avian and human metapneumoviruses (MPV). The genotype assignment of the South African isolates is indicated in the first letters of the virus designation in the figure: subgroup A, GA2, GA5, SAA1, and GA7; subgroup B, SAB1, SAB2, SAB3, GB3, and GB4. The RSV A2 and B1 prototype strains from subgroups A and B, respectively, were included for comparative purposes. The positions of the epitopes identified in this study are highlighted. Epitopes identified at the 14-mer level are indicated in gray while the mapped 9-mer epitopes N9.255, and the epitope NL9 from reference 24 are highlighted in black.

7326

VENTER ET AL.

ing suggests that there is little selection pressure in nature for the emergence of virus strains resistant to N-protein-specific CTLs. Figure 4 shows the amino acid alignment of the N proteins. The positions of the epitopes identified in this study are highlighted. Epitopes mapped to the 14-mer level are shown in gray while the mapped 9-mer epitope N09.255 and an N protein epitope identified by Goulder et al. (24), designated NL9, are highlighted in black. The positions of the epitope-containing peptides were mostly conserved, with the 14-mers N78 and N100 and the mapped 9-mers N09.255 and NL9 being identical between all human RSV genotypes in both subgroups as well as identical to the animal RSVs (ovine and bovine RSV). One amino acid change was detected in peptide N16 in genotype SAB3. Peptides N84 and N85 had a single amino acid difference between subgroups A and B that were conserved in the respective subgroups. One amino acid difference was detected in N86 in all RSV isolates relative to the A2 prototype, although this mutation resulted in a minor amino acid change (Ile to Val). The positions of the mapped epitopes N09.255 and NL9 were conserved between all human and animal RSVs. The sequence of NL9 was also conserved in the pneumovirus of mice. DISCUSSION This study examined the T-cell memory response to RSV N protein in healthy South African adults by using IFN-␥ ELIspot and identified putative CTL epitopes. Although the responses were usually present at a low frequency in these healthy adults who were not recently infected, the screening procedure effectively identified potential epitopes that could be further characterized with classical techniques. Thirty-seven adults with a wide diversity of HLA types were screened for reactivity to overlapping peptides that represent the complete RSV N protein. Previous studies have indicated that screening of viral peptide libraries in 30 to 50 previously infected individuals rapidly defines the epitopes for the most common HLA types (3, 21). The RSV N protein was selected because this is the most conserved protein both within and between the two RSV antigenic subgroups and because it was previously reported to be a major target for the CTL memory response in humans (17). Using IFN-␥ ELIspot as a marker for CTL activity, 21 subjects were identified with a response of more than 40 SFC/106 cells to at least one 14-mer N peptide. This frequency of SFC was selected as a cutoff point for a positive response. Seventeen of these subjects exhibited responses of more than 50 SFC/106 cells. In some cases, CTL epitopes contained within 14-mer peptides may induce SFC frequencies of less than 40 SFC/106 cells. For example, the identified B*08restricted CTL epitope that was located within the 14-mer N84 induced only 24 SFC/106 cells in the ELIspot screen with PBMCs from subject D16; however, when tested by chromium release assay after 10 days in culture, the peptide induced very strong lysis by this same subject’s effector cells. This finding suggested that the number of spots observed in ELIspot does not always correlate directly with the strength of cytolytic activity that can be induced by the peptides. Peptides that induced strong IFN-␥ release in a smaller number of cells, observed as a small number of large spots in ELIspot (e.g.,

J. VIROL.

peptides N84 and N85), appeared to induce stronger specific cytolysis than peptides that induced low levels of IFN-␥ in many cells (suggested by a high number of small spots in ELIspot, for example, peptide N16). These results suggest that the T-cell response to RSV N protein in healthy adult memory cells is induced by a relatively small number of peptides. These data are in keeping with previous findings for HIV-specific CTLs where a small number of immunogenic regions on the Gag protein were identified that contain the majority of the Gag-specific epitopes irrespective of virus clade, ethnicity, or age group studied (22). The strongest RSV N epitope identified in the present study was mapped to a 9-mer starting at position 255. This epitope was contained in both experimental peptides N84 and N85 (spanning amino acids 250 to 266). Seven subjects had a reaction to either one or both of these peptides during the IFN-␥ ELIspot screening with a response of more than 40 SFC/million cells. The 14-mer peptides were shown to induce strong B*08-restricted responses and, to a lesser extent, A*02-restricted reactions. When studying the HLA types of subjects that reacted to either of these peptides in IFN-␥ ELIspot, the MHC class I alleles B*08 and A*02 were identified in both Caucasians and Africans (Table 1). The A*68 molecule, a member of the A*02 HLA supertype, was identified in certain African individuals that had neither B*08 or A*02 alleles. Since only low-resolution HLA typing was performed, the exact allelic specificity of the subjects was not known. The identified epitope on N09.255 corresponds to a predicted A*0205-restricted epitope. The BIMAS algorithm predicted B*08- and A*02-restricted epitopes in position 255 on the N protein that scored among the top 10 predicted N protein epitopes for these HLA types. These results suggest the usefulness of epitope prediction algorithms to help select peptides as putative epitopes for testing. Our experience suggests, however, that algorithm prediction may be most effective after an initial experimental screening process in which the individual’s responses are determined with longer peptides. The South African population has a very diverse HLA repertoire. HLA A*02 is detected frequently in both South African black (16.06%) and Caucasian populations (29.7%) (A. Puren, HLA laboratory, National Institute for Communicable Diseases, unpublished data). The A*02 allelic frequency ranges from 7.8 to 25% in the different South African ethnic groups (Xhosa, Sotho, Zulu, Shona, and Matabele), but the A*0201 subtype is rare in Zulus, for whom A*0207 is detected frequently (26). A*68 is detected in 2.48% of South African Caucasians and in 12.04% of the black population. B*08 is also relatively common in both South African Caucasian (11.62%) and South African black populations (7.44%) (Puren, unpublished). This suggests that the epitope recognized by the B*08 and A*02 alleles identified in this study will be useful for studies of RSV CTL responses in South African subjects. Although only the epitopes located on the peptides that induced the strongest lysis (N84 and N85) were mapped down to the 9-mer level, the position of further potential epitopes located on the 14-mers N16, N78, N86, N100, and N108 that may be restricted to other HLA types were also identified. These peptides induced ELIspot responses, suggesting they might contain epitopes restricted to HLA A*24 (found in 1.64% of the South African black population and 8.42% of Caucasians),

VOL. 77, 2003

RSV HUMAN CTL EPITOPES IN SOUTH AFRICAN SUBJECTS

A*30 (22.26% of South African blacks, 8.42% of the Caucasian population), B*35 (1.91% of South African blacks, 8.08% of the Caucasian population), and B*44 (6.30% of South African blacks, 12.63% of the Caucasian population) (Puren, unpublished). The other potential epitopes we identified did not induce lysis as strong as that induced by N84 and N85 by memory CTLs in healthy adults. Nevertheless, these peptides might be considered when studying the CTL response in infected infants and children, the target population for future studies of RSV-specific CTL kinetics. In light of the genetic and antigenic variation detected in RSV, the identification of a broadly cross-strain-protective immune response might be of significant value for vaccine development. This makes identification of CTL epitopes that will provide protection against the different genotypes and to both subgroups an important goal. However, one of the methods used by viruses to evade immune surveillance is the introduction of amino acid mutations within CTL epitopes or in sequences flanking these epitopes (35). The epitope sequences themselves are essential for association with the MHC class I molecule and for recognition by the virus-specific CTLs while the flanking sequences are important in cytosolic processing of the proteins to yield the peptides that constitute CTL epitopes. Mutations in these regions may be associated with loss in CTL-mediated lysis of virus-infected cells (5, 51) or may generate peptides that antagonize CTL function (10, 29). Previous reports have suggested that the N protein is the major target of the memory CTL response to RSV in humans (6, 17). Although it is known that the N protein is the most conserved of all RSV proteins (27), it is not known if the variation that does exist occurs in the regions that contains the CTL epitopes or if immune pressure may induce positive selection in the CTL epitopes. Escape mutations in CTL epitopes have been described for several persistent infections like HIV type 1 (7, 23, 38) and EBV (10). Although CTL escape has been documented in chronic diseases, it was until recently thought to be unlikely to occur in acute viral infections. In influenza virus infections, the CD8⫹ CTL responses are usually cross-reactive between subtypes (1, 33). Consistent with this is the relative conservation of the dominant CTL target antigens, which are usually internal viral proteins that are expressed early in infected cells before viral release and thus not accessible to antibody selection pressure (39). However, several recent reports of CTL escape mutations in influenza virus that abolish MHC class I presentation and recognition from specific CTLs have been made (11, 39, 47). These reports suggest that antigenic drift resulting from immune pressure mediated by specific CTLs occurs that can result in escape from CTL-mediated immunity to influenza (11). Limited data are available about CTL epitopes to RSV in humans, and to date, no study has looked at the sequence variation in CTL epitopes. To determine if the CTL epitopes identified in South Africa are conserved between the genotypes and subgroups, isolates were selected from all genotypes identified in South Africa from 1997 to 2001 and a major part of the N protein gene was sequenced and compared with historic RSV isolates published in GenBank. To determine the extent of conservation, the N proteins were also compared to other members of the Pneumovirinae. We found that RSV

7327

N-protein-specific CTL epitopes were conserved in all South African isolates in the panel that we sequenced, which included representative strains from the major genotypes of virus circulating in the country at the time these CTL studies were performed. These data suggest that immune selection of RSV strains resulting in the induction of CTL escape mutations is not common. Our data were obtained from memory responses in healthy adults. The greatest morbidity due to RSV, however, occurs in infants and the fragile elderly. Future studies will need to determine whether the kinetics and patterns of immunodominant epitopes in RSV CTL reactivity in those groups differs from the patterns observed in healthy adults. In summary, we identified RSV-specific memory CTL epitopes in healthy South African adults by IFN-␥ ELIspot and confirmed them by ICS and cytolytic assays. A population with a diverse range of HLA types was screened with RSV N peptides, and the peptides that elicited a dominant response were selected for further characterization. Fine mapping of the 14mer peptides that induced the highest cytolytic activity identified an MHC class I restricted T-cell epitope, N09.255, that was recognized strongly by B*08- restricted cells. These alleles are relatively common in the South African population and will be useful for future studies in this population. Identification of the B*08 and putative A*02 epitopes, in combination with a previously identified RSV N protein B*07-restricted epitope (24), contributes to our ability to characterize the role of RSV-specific CTLs in pediatric infection or immunization. RSV field strains were examined for evidence of immune selection at the epitopes discovered, and we did not find evidence for immune selection mediated by CTLs directed to RSV N protein. ACKNOWLEDGMENTS This study was supported by grant no. 00/04 from the Poliomyelitis Research Foundation of South Africa and by International Research Collaboration Award R03 TW001497 from the Fogarty International Center, U.S. National Institutes of Health. M.R. was supported by NIH training grant T32 AI007474. We thank Clive M. Gray for technical advice. We also thank Maria Paximadis for HLA typing, Stephina Nyoka for assistance with FACS analysis, and the medical staff at the National Institute for Communicable Diseases for obtaining blood from subjects. We also thank each subject who took part in this study. REFERENCES 1. Ada, G. L., and P. D. Jones. 1987. The cell-mediated and humoral responses to influenza virus infection in the mouse lung. Adv. Exp. Med. Biol. 216B: 1033–1042. 2. Alansari, H., and L. N. Potgieter. 1994. Nucleotide and predicted amino acid sequence analysis of the ovine respiratory syncytial virus non-structural 1C and 1B genes and the small hydrophobic protein gene. J. Gen. Virol. 75(Pt 2):401–404. 3. 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, B. D. Walker, and P. J. Goulder. 2000. Identification of dominant optimal HLAB60- and HLA-B61-restricted cytotoxic T-lymphocyte (CTL) epitopes: rapid characterization of CTL responses by enzyme-linked immunospot assay. J. Virol. 74:8541–8549. 4. Altman, J. D., P. A. Moss, P. J. Goulder, D. H. Barouch, M. G. McHeyzerWilliams, J. I. Bell, A. J. McMichael, and M. M. Davis. 1996. Phenotypic analysis of antigen-specific T lymphocytes. Science 274:94–96. (Erratum 280: 1821, 1998.) 5. Apolloni, A., D. Moss, R. Stumm, S. Burrows, A. Suhrbier, I. Misko, C. Schmidt, and T. Sculley. 1992. Sequence variation of cytotoxic T cell epitopes in different isolates of Epstein-Barr virus. Eur. J. Immunol. 22:183– 189.

7328

VENTER ET AL.

6. Bangham, C. R., P. J. Openshaw, L. A. Ball, A. M. King, G. W. Wertz, and B. A. Askonas. 1986. Human and murine cytotoxic T cells specific to respiratory syncytial virus recognize the viral nucleoprotein (N), but not the major glycoprotein (G), expressed by vaccinia virus recombinants. J. Immunol. 137:3973–3977. 7. Barouch, D. H., J. Kunstman, M. J. Kuroda, J. E. Schmitz, S. Santra, F. W. Peyerl, G. R. Krivulka, K. Beaudry, M. A. Lifton, D. A. Gorgone, D. C. Montefiori, M. G. Lewis, S. M. Wolinsky, and N. L. Letvin. 2002. Eventual AIDS vaccine failure in a rhesus monkey by viral escape from cytotoxic T lymphocytes. Nature 415:335–339. 8. Barr, J., P. Chambers, C. R. Pringle, and A. J. Easton. 1991. Sequence of the major nucleocapsid protein gene of pneumonia virus of mice: sequence comparisons suggest structural homology between nucleocapsid proteins of pneumoviruses, paramyxoviruses, rhabdoviruses and filoviruses. J. Gen. Virol. 72(Pt 3):677–685. 9. Bednarek, M. A., S. Y. Sauma, M. C. Gammon, G. Porter, S. Tamhankar, A. R. Williamson, and H. J. Zweerink. 1991. The minimum peptide epitope from the influenza virus matrix protein. Extra and intracellular loading of HLA-A2. J. Immunol. 147:4047–4053. 10. Bertoletti, A., A. Sette, F. V. Chisari, A. Penna, M. Levrero, M. De Carli, F. Fiaccadori, and C. Ferrari. 1994. Natural variants of cytotoxic epitopes are T cell receptor antagonists for antiviral cytotoxic T cells. Nature 369:407– 410. 11. Boon, A. C., G. de Mutsert, Y. M. Graus, R. A. Fouchier, K. Sintnicolaas, A. D. Osterhaus, and G. F. Rimmelzwaan. 2002. Sequence variation in a newly identified HLA-B35-restricted epitope in the influenza A virus nucleoprotein associated with escape from cytotoxic T lymphocytes. J. Virol. 76:2567–2572. 12. Brandenburg, A. H., L. de Waal, H. H. Timmerman, P. Hoogerhout, R. L. de Swart, and A. D. Osterhaus. 2000. HLA class I-restricted cytotoxic T cell epitopes of the respiratory syncytial virus fusion protein. J. Virol. 74:10240– 10244. 13. Buchholz, U. J., S. Finke, and K. K. Conzelmann. 1999. Generation of bovine respiratory syncytial virus (BRSV) from cDNA: BRSV NS2 is not essential for virus replication in tissue culture, and the human RSV leader region acts as a functional BRSV genome promoter. J. Virol. 73:251–259. 14. Cannon, M. J., and C. R. Bangham. 1989. Recognition of respiratory syncytial virus fusion protein by mouse cytotoxic T cell clones and a human cytotoxic T cell line. J. Gen. Virol. 70:79–87. 15. Cannon, M. J., P. J. Openshaw, and B. A. Askonas. 1988. Cytotoxic T cells clear virus but augment lung pathology in mice infected with respiratory syncytial virus. J. Exp. Med 168:1163–1168. 16. Cannon, M. J., E. J. Stott, G. Taylor, and B. A. Askonas. 1987. Clearance of persistent respiratory syncytial virus infections in immunodeficient mice following transfer of primed T cells. Immunology 62:133–138. 17. Cherrie, A. H., K. Anderson, G. W. Wertz, and P. J. Openshaw. 1992. Human cytotoxic T cells stimulated by antigen on dendritic cells recognize the N, SH, F, M, 22K, and 1b proteins of respiratory syncytial virus. J. Virol. 66:2102– 2110. 18. Chiba, Y., Y. Higashidate, K. Suga, K. Honjo, H. Tsutsumi, and P. L. Ogra. 1989. Development of cell-mediated cytotoxic immunity to respiratory syncytial virus in human infants following naturally acquired infection. J. Med. Virol. 28:133–139. 19. Collins, P. L., K. Anderson, S. J. Langer, and G. W. Wertz. 1985. Correct sequence for the major nucleocapsid protein mRNA of respiratory syncytial virus. Virology 146:69–77. 20. Collins, P. L., K. McIntosh, and R. M. Chanock. 1996. Respiratory syncytial virus, p. 1313–1351. In B. N. Fields, D. M. Knipe, P. M. Howley, and R. M. Chanock (ed.), Fields virology, 3rd ed. Lippincott Raven Press, Philadelphia, Pa. 21. 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, M. Bunce, H. M. Coovadia, and B. D. Walker. 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. 22. Goulder, P. J., C. Brander, K. Annamalai, N. Mngqundaniso, U. Govender, Y. Tang, S. He, K. E. Hartman, C. A. O’Callaghan, G. S. Ogg, M. A. Altfeld, E. S. Rosenberg, H. Cao, S. A. Kalams, M. Hammond, M. Bunce, S. I. Pelton, S. A. Burchett, K. McIntosh, H. M. Coovadia, and B. D. Walker. 2000. Differential narrow focusing of immunodominant human immunodeficiency virus gag-specific cytotoxic T-lymphocyte responses in infected African and caucasoid adults and children. J. Virol. 74:5679–5690. 23. Goulder, P. J., C. Brander, Y. Tang, C. Tremblay, R. A. Colbert, M. M. Addo, E. S. Rosenberg, T. Nguyen, R. Allen, A. Trocha, M. Altfeld, S. He, M. Bunce, R. Funkhouser, S. I. Pelton, S. K. Burchett, K. McIntosh, B. T. Korber, and B. D. Walker. 2001. Evolution and transmission of stable CTL escape mutations in HIV infection. Nature 412:334–338. 24. Goulder, P. J., F. Lechner, P. Klenerman, K. McIntosh, and B. D. Walker. 2000. Characterization of a novel respiratory syncytial virus-specific human cytotoxic T-lymphocyte epitope. J. Virol. 74:7694–7697.

J. VIROL. 25. Graham, B. S. 1996. Immunological determinants of disease caused by respiratory syncytial virus. Trends Microbiol. 4:290–293. 26. Hammond, M. G., E. D. du Toit, A. Sanchez-Mazas, M. Andrien, M. Coluzzi, M. de Pablo, G. de Stefano, C. Kaplan, L. J. Kennedy, L. Louie, and F. Migot. 1997. HLA in sub-Saharan Africa, p. 345–353. In D. Charon (ed.), Proceedings of the 12th International Histocompatibility Workshop and Conference. SSAF report, Paris, France. 27. Johnson, P. R., and P. L. Collins. 1989. The 1B (NS2), 1C (NS1) and N proteins of human respiratory syncytial virus (RSV) of antigenic subgroups A and B: sequence conservation and divergence within RSV genomic RNA. J. Gen. Virol. 70(Pt 6):1539–1547. 28. Karron, R. A., D. A. Buonagurio, A. F. Georgiu, S. S. Whitehead, J. E. Adamus, M. L. Clements-Mann, D. O. Harris, V. B. Randolph, S. A. Udem, B. R. Murphy, and M. S. Sidhu. 1997. Respiratory syncytial virus (RSV) SH and G proteins are not essential for viral replication in vitro: clinical evaluation and molecular characterization of a cold-passaged, attenuated RSV subgroup B mutant. Proc. Natl. Acad. Sci. USA 94:13961–13966. 29. Klenerman, P., S. Rowland-Jones, S. McAdam, J. Edwards, S. Daenke, D. Lalloo, B. Koppe, W. Rosenberg, D. Boyd, A. Edwards, et al. 1994. Cytotoxic T cell activity antagonized by naturally occurring HIV-1 Gag variants. Nature 369:403–407. 30. Kumar, S., K. Tamura, I. B. Jakobsen, and M. Nei. 2001. MEGA2: molecular evolutionary genetics analysis software. Bioinformatics 17:1244–1245. 31. Lalvani, A., T. Dong, G. Ogg, A. A. Patham, H. Newell, A. V. Hill, A. J. McMichael, and S. Rowland-Jones. 1997. Optimization of a peptide-based protocol employing IL-7 for in vitro restimulation of human cytotoxic T lymphocyte precursors. J. Immunol. Methods 210:65–77. 32. Larsen, L. E., A. Uttenthal, P. Arctander, K. Tjornehoj, B. Viuff, C. Rontved, L. Ronsholt, S. Alexandersen, and M. Blixenkrone-Moller. 1998. Serological and genetic characterisation of bovine respiratory syncytial virus (BRSV) indicates that Danish isolates belong to the intermediate subgroup: no evidence of a selective effect on the variability of G protein nucleotide sequence by prior cell culture adaption and passages in cell culture or calves. Vet. Microbiol. 62:265–279. 33. McMichael, A. 1994. Cytotoxic T lymphocytes specific for influenza virus. Curr. Top. Microbiol. Immunol. 189:75–91. 34. Murali-Krishna, K., J. D. Altman, M. Suresh, D. Sourdive, A. Zajac, and R. Ahmed. 1998. In vivo dynamics of anti-viral CD8 T cell responses to different epitopes. An evaluation of bystander activation in primary and secondary responses to viral infection. Adv. Exp. Med. Biol. 452:123–142. 35. Oldstone, M. B. 1997. How viruses escape from cytotoxic T lymphocytes: molecular parameters and players. Virology 234:179–185. 36. Openshaw, P. J., K. Anderson, G. W. Wertz, and B. A. Askonas. 1990. The 22,000-kilodalton protein of respiratory syncytial virus is a major target for Kd-restricted cytotoxic T lymphocytes from mice primed by infection. J. Virol. 64:1683–1689. 37. Parker, K. C., M. A. Bednarek, and J. E. Coligan. 1994. Scheme for ranking potential HLA-A2 binding peptides based on independent binding of individual peptide side-chains. J. Immunol. 152:163–175. 38. Price, D. A., P. J. Goulder, P. Klenerman, A. K. Sewell, P. J. Easterbrook, M. Troop, C. R. Bangham, and R. E. Phillips. 1997. Positive selection of HIV-1 cytotoxic T lymphocyte escape variants during primary infection. Proc. Natl. Acad. Sci. USA 94:1890–1895. 39. Price, G. E., R. Ou, H. Jiang, L. Huang, and D. Moskophidis. 2000. Viral escape by selection of cytotoxic T cell-resistant variants in influenza A virus pneumonia. J. Exp. Med. 191:1853–1867. 40. Rammensee, H., J. Bachmann, N. P. Emmerich, O. A. Bachor, and S. Stevanovic. 1999. SYFPEITHI: database for MHC ligands and peptide motifs. Immunogenetics 50:213–219. 41. Shin, H.-J., K. T. Cameron, J. A. Jacobs, E. A. Turpin, D. A. Halvorson, S. M. Goyal, K. V. Nagaraja, M. C. Kumar, D. C. Lauer, B. S. Seal, and M. K. Njenga. 2002. Molecular epidemiology of subgroup C avian pneumoviruses isolated in the United States and comparison with subgroup A and B viruses. J. Clin. Microbiol. 40:1687–1693. 42. Thompson, J. D., T. J. Gibson, F. Plewniak, F. Jeanmougin, and D. G. Higgins. 1997. The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 25:4876–4882. 43. Valarcher, J. F., H. Bourhy, A. Lavenu, N. Bourges-Abella, M. Roth, O. Andreoletti, P. Ave, and F. Schelcher. 2001. Persistent infection of B lymphocytes by bovine respiratory syncytial virus. Virology 291:55–67. 44. van den Hoogen, B. G., T. M. Bestebroer, A. D. Osterhaus, and R. A. Fouchier. 2002. Analysis of the genomic sequence of a human metapneumovirus. Virology 295:119–132. 45. Venter, M., M. Collinson, and B. D. Schoub. 2002. Molecular epidemiological analysis of community circulating respiratory syncytial virus in rural South Africa: comparison of viruses and genotypes responsible for different disease manifestations. J. Med. Virol. 68:452–461. 46. Venter, M., S. A. Madhi, C. T. Tiemessen, and B. D. Schoub. 2001. Genetic diversity and molecular epidemiology of respiratory syncytial virus over four consecutive seasons in South Africa: identification of new subgroup A and B genotypes. J. Gen. Virol. 82:2117–2124.

VOL. 77, 2003

RSV HUMAN CTL EPITOPES IN SOUTH AFRICAN SUBJECTS

47. Voeten, J. T., T. M. Bestebroer, N. J. Nieuwkoop, R. A. Fouchier, A. D. Osterhaus, and G. F. Rimmelzwaan. 2000. Antigenic drift in the influenza A virus (H3N2) nucleoprotein and escape from recognition by cytotoxic T lymphocytes. J. Virol. 74:6800–6807. 48. Walker, B. D., C. Flexner, K. Birch-Limberger, L. Fisher, T. J. Paradis, A. Aldovini, R. Young, B. Moss, and R. T. Schooley. 1989. Long-term culture and fine specificity of human cytotoxic T-lymphocyte clones reactive with human immunodeficiency virus type 1. Proc. Natl. Acad. Sci. USA 86:9514– 9518.

7329

49. Walsh, E., and B. Graham. 1999. Respiratory syncytial viruses, p. 161–203. In Viral infections of the respiratory tract, vol. 27. Marcel Dekker, Inc., New York, N.Y. 50. Weber, M. W., E. K. Mulholland, and B. M. Greenwood. 1998. Respiratory syncytial virus infection in tropical and developing countries. Trop. Med. Int. Health 3:268–280. 51. Yellen-Shaw, A. J., E. J. Wherry, G. C. Dubois, and L. C. Eisenlohr. 1997. Point mutation flanking a CTL epitope ablates in vitro and in vivo recognition of a full-length viral protein. J. Immunol. 158:3227–3234.