Molecular and Immunological Significance of Chimpanzee Major Histocompatibility Complex Haplotypes for Hepatitis C Virus Immune Response and Vaccination Studies

Abstract

The chimpanzee is a critical animal model for studying cellular immune responses to infectious pathogens such as hepatitis B and C viruses, human immunodeficiency virus, and malaria. Several candidate vaccines and immunotherapies for these infections aim at the induction or enhancement of cellular immune responses against viral epitopes presented by common human major histocompatibility complex (MHC) alleles. To identify and characterize chimpanzee MHC class I molecules that are functionally related to human alleles, we sequenced 18 different Pan troglodytes (Patr) alleles of 14 chimpanzees, 2 of them previously unknown and 3 with only partially reported sequences. Comparative analysis of Patr binding pockets and binding assays with biotinylated peptides demonstrated a molecular homology between the binding grooves of individual Patr alleles and the common human alleles HLA-A1, -A2, -A3, and -B7. Using cytotoxic T cells isolated from the blood of hepatitis C virus (HCV)-infected chimpanzees, we then mapped the Patr restriction of these HCV peptides and demonstrated functional homology between the Patr-HLA orthologues in cytotoxicity and gamma interferon (IFN-γ) release assays. Based on these results, 21 HCV epitopes were selected to characterize the chimpanzees' cellular immune response to HCV. In each case, IFN-γ-producing T cells were detectable in the blood after but not prior to HCV infection and were specifically targeted against those HCV peptides predicted by Patr-HLA homology. This study demonstrates a close functional homology between individual Patr and HLA alleles and shows that HCV infection generates HCV peptides that are recognized by both chimpanzees and humans with Patr and HLA orthologues. These results are relevant for the design and evaluation of vaccines in chimpanzees that can now be selected according to the most frequent human MHC haplotypes.

The presentation of pathogen-derived peptides on major histocompatibility complex (MHC) class I molecules of infected cells is a crucial step in host defense that initiates the adaptive cellular immune response and controls the expansion and effector function of pathogen-specific T cells. It is especially important in the defense against noncytopathic viruses, such as the hepatitis C virus (HCV), that reside and amplify intracellularly and cannot be completely cleared by the humoral arm of the immune response alone.

Although MHC class I loci are among the most polymorphic and variable genes in the genome (38), certain common features that determine the nature of binding peptides have been identified. First, the peptide-binding site is formed by two parallel α-helices on top of antiparallel oriented β-strands. The MHC class I binding groove allows binding of peptides 8 to 11 amino acids in length. Second, the specificity of binding is determined by polymorphic amino acid residues in the α1 and α2 domains whose side chains protrude into the peptide-binding groove. Third, based on these residues, certain MHC supertypes have been defined that bind peptides with the corresponding binding motif, i.e., specific anchor amino acid residues that interact with the polymorphic complementary pockets of the MHC peptide-binding grooves (54).

The chimpanzee is an important animal model for studying infections with HCV, human immunodeficiency virus (HIV), malaria, and other pathogens and for evaluating candidate vaccines. Several relevant vaccination strategies aim at the induction or enhancement of cellular immune responses against viral epitopes that are presented by common human MHC alleles. Although chimpanzee MHC alleles are closely related to human alleles, distinct differences do exist. For example, the chimpanzee Patr-A locus is less polymorphic than the human HLA-A locus (35). In addition, all known Patr-A alleles appear to be related to only one of the two sublineages of HLA-A alleles, namely to the HLA-A3 lineage, which comprises the HLA-A1, -A3, -A9, -A11, and -A80 family (15, 31, 34, 35, 45); a specific Patr lineage or allele related to the most frequent human allele, HLA-A2, has not been identified at the molecular level. Moreover, it is not known whether those HCV epitopes that are well characterized in HCV-infected patients and therefore represent promising vaccine candidates (4, 6, 9, 40) are also endogenously processed and recognized by T cells of HCV-infected chimpanzees.

The aim of this study was therefore to characterize individual chimpanzee MHC class I alleles both at the molecular level, to identify HLA class I orthologues, and at the functional level, to characterize peptide presentation and recognition by HCV-specific T cells. The results are relevant for the design and evaluation of vaccines in chimpanzees that can now be selected according to the most frequent human MHC haplotypes.

MATERIALS AND METHODS

Chimpanzees.

Fourteen chimpanzees were used in this study (Table 1). All animals tested negative for serum HCV RNA by reverse transcriptase PCR and had normal liver enzyme levels prior to inoculation with HCV. Chimpanzees Ch6390, Ch6394, and Ch6413 were not inoculated with HCV and were studied as additional naive control animals. In Ch1536, Ch1606, and Ch1552, HCV infection was established by intrahepatic inoculation of full-length uncapped HCV RNA transcripts of genotype 1a (25). In Ch4X0132 and Ch4X0142, HCV infection was established by intrahepatic inoculation of full-length uncapped HCV RNA transcripts of genotype 1b (50). Ch1601, Ch1605, Ch1629, and Ch6411 were infected by intravenous inoculation of plasma (100 50% chimpanzee infectious doses) from chimpanzee Ch1536; Ch4X0234 was infected with plasma from chimpanzee Ch4X0142. Ch4X0186 was infected with a passage of H77 plasma (genotype 1a). HCV infection was diagnosed by detection of serum HCV RNA by nested reverse transcriptase PCR, and HCV-specific antibodies were detected by third-generation enzyme-linked immunosorbent assay as previously described (33, 50). Lymphocyte samples for immunological analysis were obtained prior to and after HCV inoculation. After HCV inoculation, lymphocyte samples were obtained during the first 2 to 10 months of viremia from animals Ch1536, Ch1601, Ch1605, Ch1606, Ch1629, Ch6411, Ch6413, Ch4X0132, Ch4X0142, Ch4X0186, and Ch4X0234 and after 14 months and 5.3 years of viral clearance from animals Ch1552 and Ch4X0186, respectively.

TABLE 1.

Chimpanzee MHC class I alleles

Chimpanzee	Allele(s)a
	Patr-A	Patr-B
1536	A*1401	B*0101, B*0901
1552	A*0701, A*0901	B*0101, B*1301
1601	A*0601, A*0701	B*0302
1605	A*0501	B*0501, B*1301
1606	A*0301, A*0501	B*1301, B*2401
1629	A*0301, A*1501	B*2202
6390	A*0301, A*1401	B*0101, B*2901
6394	A*0901	B*0101, B*0302
6411	A*0401	B*0101
6413	A*0301, A*0701	B*2401
4X0142	A*0101, A*0301	B*0101, B*0302
4X0186	A*0401	B*0101, B*1601
4X0132	A*0501	B*0501, B*1301
4X0234	A*0301, A*0701	B*0101

^aSingle underline indicates a new allele; double underline indicates an allele for which only a partial sequence was previously reported.

All chimpanzees were housed at ALAAC-accredited facilities, namely the Southwest Foundation for Biomedical Research, San Antonio, Tex.; the Coulston Foundation, Alamogordo, N.Mex.; and the Food and Drug Administration, Bethesda, Md. Animal protocols were reviewed and approved by the animal care and use committees from each institution involved with the chimpanzee studies. The chimpanzee experiments were also reviewed and approved by the Public Health Service Interagency Animal Model Committee.

Isolation of PBMCs and generation of Epstein-Barr virus-transformed B-cell lines (EBV-BCLs).

Peripheral blood mononuclear cells (PBMCs) were separated on Ficoll-Histopaque (Sigma, St. Louis, Mo.) as described previously (49), washed three times in phosphate-buffered saline (PBS), resuspended at 2 × 10⁷ PBMC/ml in RPMI 1640 medium (Gibco Laboratories, Grand Island, N.Y.) containing 70% fetal calf serum (FCS) and 10% dimethyl sulfoxide (Sigma), and cryopreserved in liquid nitrogen until use.

Chimpanzee EBV-BCLs were established from PBMCs as previously described (31). Human EBV-BCLs used in this study were BCL 888 (HLA-A1, -A24, -B52, and -B55 positive; kindly provided by Paul F. Robbins), JY (HLA-A2 and -B7 positive), and GM3107 (HLA-A3 and -B7 positive).

Cloning, sequencing, and analysis of Patr class I alleles.

Isolation of total RNA from EBV-BCLs or PBMCs, first-strand cDNA synthesis, and amplification of Patr class I sequences were performed as previously described (16). Amplified Patr class I cDNA was cloned into the pGEM11Zf(+) Vector (Promega, Madison, Wis.) and sequenced in its full length by using the T7 forward primer and primers 3S, 3N, 4S, and 4N (16). At least five molecular clones per Patr allele of each chimpanzee were sequenced and analyzed with DNASIS software (Hitachi Software Engineering Co., Ltd., Yokohama, Japan).

Generation of Patr class I transfectants.

Patr class I transfectants were established as follows. Error-free Patr class I cDNA inserts were excised from pGEM11Zf(+) by digestion with EcoRI and HindIII purified from 1.0% agarose-Tris-acetate-EDTA gels and subcloned into the EcoRI- and HindIII-digested pCMV-Script (Stratagene, La Jolla, Calif.). After amplification in DH5α cells, recombinants were purified and Patr class I inserts were confirmed by sequencing. Recombinants containing Patr class I cDNA were transfected into the MHC class I-deficient human B-cell line 721.221 by electroporation as previously described (10). After electroporation, cells were cultured in RPMI 1640 medium containing 30% FCS for 3 days and then selected in RPMI medium containing 10% FCS and 1 mg of Geneticin (Sigma)/ml. Expanding G418-resistant cells were stained with the anti-MHC class I antibody W6/32 and fluorescein isothiocyanate (FITC)-labeled goat anti-mouse immunoglobulin G and immunoglobulin M antibody (Jackson ImmunoResearch, West Grove, Pa.), sorted, and autocloned on a Coulter flow cytometer. Three autocloned cells per well of a 96-well plate were expanded in the presence of 10⁴ irradiated allogeneic PBMCs. Expanded transfectants that showed high expression of Patr class I were selected for immunologic assays.

Synthetic peptides.

Peptides selected for the binding studies were known to bind to individual MHC proteins with high affinity (see Table 2 for peptide sequences and references). Biotinylated peptides were synthesized at the Facility for Biotechnology Resources, Center for Biologics Evaluation and Research at the Food and Drug Administration, Bethesda, Md., on an Applied Biosystems Inc. (Foster City, Calif.) ABI Model 432 peptide synthesizer at the 25-mmol synthesis scale by using 9-fluoroenylmethoxycarbonyl (Fmoc) chemistry (2, 37) mediated by 2-[H-benzotriazole-1-yl]-1.13.3-tetramethyluronium hexafluorophosphate/1-hydroxybenzotriazole (HBTU/HOBT) activation. Amino acid derivatives were from ABI, except for N-a-Fmoc-N-e-(d-biotin-6-amidocaproate)-l-lysine, which was purchased from Anaspec Incorporated (San Jose, Calif.). Double coupling was performed at biotinylated lysine positions. Following deprotection of the N-terminal amino group, peptides were cleaved with a standard ABI mixture of trifluoroacetic acid, ethanedithiol, thioanisole, triisopropylsilane, and phenol-water (1:1) at a ratio of 90:1.5:2.5:1.5:4.5. Following cleavage, peptides were precipitated and washed in ice-cold methyl tert-butyl ether. Peptide pellets were resuspended in 1 to 5% acetic acid, clarified through a 0.45-mm-pore-size filter, and lyophilized.

TABLE 2.

Biotinylated peptides used for MHC binding assays

Peptide	Source protein	Amino acid sequencea	MHC restriction	Reference
1	Human J chain 102-110	YTAV(V→K)PLVY	HLA-A1	26, 36
2	HBVcore 18-27b	FLPS(D→K)YFPSV	HLA-A2	43
3	Consensus peptide sequence of high binders	KVFP(V→K)ALINK	HLA-A3	29, 36
4	HLA-A*0201 signal sequencec	APRT(L→K)VYLL	HLA-B7	22, 36
5	HCV E2 542-551	TRPP(G→K)GNWF	Patr-B13	27

^aAmino acids in parentheses are replaced by lysine (K) attached to a 6C linker and biotin.

^bThe HBV core 18-27 peptides contain an F in position 6, which was replaced by a Y (43).

^cThe HLA-A*0201 signal sequence contains an L in position 7, which was replaced by a Y for the binding assay (48).

Peptides selected for analysis of HCV-specific T-cell responses were derived from the HCV genotype 1a polyprotein and had been described to be recognized by cytotoxic T lymphocytes (CTL) of HCV-infected humans or chimpanzees (see Table 3 for peptide sequences and references). Peptides were synthesized at >80% purity with free amino and carboxy termini (Research Genetics, Huntsville, Ala.).

TABLE 3.

HCV-derived HLA-A1-, -A2-, -A3-, -B7-, and Patr-restricted peptides for analysis of chimpanzee T-cell responses

Peptide	Protein	Amino acid position	Amino acid sequence	HLA or Patr restriction	Source or reference
6	Core	126	TLCGFADLMGY	HLA-A1	New epitope
7	NS3	1128	CTCGSSDLY	HLA-A1	New epitope
8	NS3	1305	LADGGCSGGAY	HLA-A1	New epitope
9	NS5A	2180	LTDPSHITA	HLA-A1	New epitope
10	NS5B	2922	LSAFSLHSY	HLA-A1	New epitope
11	Core	35	YLLPRRGPRL	HLA-A2	4, 6, 8, 39, 49
12	Core	131	ADLMGYIPLV	HLA-A2	6, 8, 39
13	Core	167	GNLPGCSFSI	HLA-A2	44
14	NS3	1073	CVNGVCWTV	HLA-A2	6, 8, 32, 39, 49
15	NS3	1406	KLVALGINAV	HLA-A2	6, 8, 39, 49
16	NS4B	1807	LLFNILGGWV	HLA-A2	4, 6, 8, 39, 49
17	NS5B	2727	GLQDCTMLV	HLA-A2	4, 6, 8, 39, 49
18	NS1	632	RMYVGGVEHR	HLA-A3	7
19	NS3	1396	LIFCHSKKK	HLA-A3	7
20	NS3	1632	VTLTHPITK	HLA-A3	New epitope
21	NS4B	1863	GVAGALVAFK	HLA-A3	7
22	NS4B	1864	VAGALVAFK	HLA-A3	7
23	Core	41	GPRLGVRAT	HLA-B7	28
24	Core	111	DPRRRSRNL	HLA-B7	12
25	E1	234	NASRCWVAM	HLA-B35a	28
26	NS3	1379	IPFYGKAI	HLA-B7	New epitope
27	E2	651	RCDLEDRDRSELSPL	Patr-A0601	10
28	P7	781	RWVPGAVYAFYb	Patr-A0601	10
29	NS5A	2055	MWSGTFPINAY	Patr-A06	10
30	NS3	1444	FTGDFDSVIc	Patr-B01	27
31	E1	542	TRPPLGNWF	Patr-B13	27
32	NS3	1357	VPHPNIEEV	Patr-B13	27

^aHLA-B35 belongs to the HLA-B7 supertype family.

^bPublished as KWVPGAVYTKY. In this study, the sequence was changed to match the sequence of the infecting virus.

^cPublished as YTGDFDSVI. In this study, the sequence was changed to match the sequence of the infecting virus.

The following HLA-A2 restricted epitopes derived from the hepatitis B virus (HBV) were used as control peptides to test for the virus specificity of T-cell responses (Fig. 4E): B-1, HBVpol575-583, FLLSLGIHL; B-2, HBVcore18-27, FLPSDFFPSV; B-3, HBVenv183-191, FLLTRILTI; B-4, HBVenv335-343, WLSLLVPFV; B-5, HBVpol455-463, GLSRYVARL (42).

FIG. 4.

Analysis of in vivo primed T-cell responses of HCV-infected chimpanzees with HLA-A1-, -A2-, -A3-, and -B7-restricted HCV peptides and known Patr-restricted HCV epitopes. Thirty-hour IFN-γ ELISPOT assays were performed with PBMCs obtained from chimpanzees before and after HCV infection. Asterisks indicate additional animals that were studied at a single time point only, i.e., either prior to (∗) or after (∗∗) HCV inoculation. HCV peptides were selected according to the animals' Patr haplotype (Table 1), the proposed Patr-HLA homology (Tables 4 and 5), and the results of human studies indicating that these HCV peptides are recognized in the context of specific HLA alleles (Table 3). The sequence and amino acid position of each peptide within the HCV polyprotein are provided in Table 3. All bars indicate the number of specific IFN-γ spots in response to a peptide, i.e., the number of spots in the presence of the peptide minus the number of spots in the absence of the peptide. Responses to HLA-restricted HCV epitopes are indicated by filled bars; responses to Patr-restricted HCV epitopes are indicated by open bars. Each chimpanzee was studied at least at three separate time points; representative results are shown. The sequences of the HLA-A2-restricted control peptides derived from HBV and shown in panel E are provided in Materials and Methods. n.t., not tested.

Analysis of peptide-MHC binding affinity.

A total of 3 × 10⁵ EBV-transformed B cells or Patr class I transfectants per well of a 96-well flat-bottom plate (Nunc, Naperville, Ill.) were incubated with 100 μg of biotinylated peptide/ml at 37°C for 12 h followed by labeling with 10 μg of FITC-avidin D (Vector Labs, Burlingame, Calif.)/ml at 4°C for 30 min. For Patr class I transfectants, incubation with FITC-avidin D was followed by incubation with 10 μg of biotinylated anti-avidin D (Vector Labs)/ml at 4°C for 30 min and by a second incubation with FITC-avidin D to increase the FITC signal, because MHC class I expression of some Patr transfectants, i.e., Patr-A*0601- and Patr-B*1301-transfected cell lines, was lower than that of EBV-BCLs (Table 4). After each incubation, excess reagents were washed off at 4°C with PBS. Stained cells were analyzed by flow cytometry on a FACSCalibur with CellQuest software (Becton Dickinson, San Jose, Calif.). Dead cells were excluded from the analysis by propidium iodide staining. To measure the relative amount of FITC-avidin D bound, the mean fluorescence of 5,000 stained cells was determined. Binding assays were repeated and confirmed at least once.

TABLE 4.

Binding of biotinylated peptides to Patr proteins

Allelea	Mean fluorescence intensity
	No peptide	Peptide 1 (HLA-A1)	Peptide 2 (HLA-A2)	Peptide 3 (HLA-A3)	Peptide 4 (HLA-B7)	MHC I expression
Patr-A*0301	0	19.3	0	195.0	0	381.1
Patr-A*0601	0.2	82.1	2.1	0.4	0.5	111.6
Patr-A*0501	0	0	15.4	192.0	0	421.0
Patr-B*0101	0	4.4	56.3	245.4	0	816.5
Patr-B*0901	0	43.6	24.3	33.0	0	381.1
Patr-B*1301	0	5.5	7.7	0	17.3	187.2

^a721.221 cells were transfected with the indicated allele.

ELISPOT assay.

Ninety-six-well plates (Millititer; Millipore, Bedford, Mass.) were coated with 0.5 μg of anti-human gamma interferon (IFN-γ) (Endogen, Woburn, Mass.)/ml at 4°C overnight and then washed four times with sterile PBS. The plates were blocked with RPMI medium and 1% bovine serum albumin (Sigma) for 1 h at 25°C. Cryopreserved PBMCs were thawed, and 3 × 10⁵ PBMCs were added in duplicate cultures of RPMI 1640, 5% AB serum, and 2 mM l-glutamine together with MHC class I-restricted HCV peptides (10 μg/ml). In the case of Ch4X0132, enzyme-linked immunospot assays (ELISPOTs) were performed with 140,000 PBMCs. After 30 h, the plates were washed seven times and incubated overnight with 0.25 μg of biotin-conjugated anti-human-IFN-γ antibody (Endogen)/ml. After four washes, streptavidin-AP (1:2,000) (DAKO, Glostrup, Denmark) was added for 2 h. Finally, the plates were washed four times with PBS and developed with freshly prepared NBT/BCIP (nitroblue tetrazolium-5-bromo-4-chloro-3-indolylphosphate) solution (Bio-Rad, Hercules, Calif.). The reaction was stopped by washing with distilled water, and the spots were counted with an ImmunoSpot series 1 analyzer (Cellular Technology, Cleveland, Ohio). The number of specific spots was determined by subtracting the number of spots in the absence of peptides from the number of spots in the presence of peptides. Positive controls for IFN-γ ELISPOTs consisted of 1 μg of phytohemagglutinin (Murex Biotech Limited, Dartford, England)/ml.

Stimulation of PBMCs with synthetic peptides.

HCV peptide-specific T cells were expanded from PBMCs in 96-well round-bottom plates (Nunc). Two hundred thousand cells per well were stimulated with synthetic peptides (10 μg/ml), rIL-7 (10 ng/ml), and rIL-12 (100 pg/ml) (PeproTech Inc., Rocky Hill, N.J.) in RPMI 1640 supplemented with 10% heat-inactivated human AB serum, l-glutamine (2 mM), penicillin (100 U/ml), and streptomycin (100 μg/ml). Cultures were restimulated with 10 μg of peptide/ml, 20 U of rIL-2 (Chiron Corp., Emeryville, Calif.)/ml, and 10⁵ irradiated (3,000 rad) autologous PBMCs as feeder cells on days 7 and 14. On days 3, 10, and 18, 100 μl of RPMI medium with 10% human AB serum and rIL-2 at a final concentration of 10 U/ml was added to each well.

Cytotoxicity assay.

A standard 6-h CTL assay was performed with autologous EBV B-cell lines or Patr class I transfectants as target cells that were incubated overnight with 10 μg of synthetic peptide/ml and labeled with 25 μCi of ⁵¹Cr (Amersham Corp., Arlington Heights, Ill.) for 1 h. After three washes with PBS, target cells were plated at 3,000 cells/well in complete medium in round-bottom 96-well plates. Unlabeled K562 (120,000 cells/well) was added to reduce nonspecific lysis. Peptide-stimulated T-cell lines from chimpanzees were added at various effector-to-target ratios as indicated. The percent cytotoxicity was determined by the following formula: 100 × [(experimental release − spontaneous release)/(maximum release − spontaneous release)]. Maximum release was determined by lysis of ⁵¹Cr-labeled targets with 10% Triton X-100 (Sigma). Spontaneous release was <15% of the maximum release in all experiments. The specific cytotoxic activity was calculated as follows: (cytotoxic activity in the presence of peptide) − (cytotoxic activity in the absence of peptide). A specific cytotoxic activity of >10% was considered positive.

Nucleotide sequence accession numbers. The sequences for Patr-B*2202 and Patr-B*0302 are available under GenBank accession numbers AF500292 and AFF500291, respectively. The complete sequences for Patr-A*0301, -A*0501, and -A*0601 are now reported (GenBank accession numbers AF500288, AF500289, and AF500290).

RESULTS

Identification of Patr class I alleles of chimpanzees.

Nine different Patr A and nine different Patr B molecules were identified from a sequence analysis of the 14 chimpanzees included in this study (Table 1). Patr-B*2202 and Patr-B*0302 constituted new alleles. Three Patr-A sequences, those of the Patr-A*0301, -A*0501, and -A*0601 alleles, had only been partially reported previously. Patr-A*0301 was the most frequent allele in this group of chimpanzees and was identified in 6 of 14 animals. Patr-B*0101 was the most common of the Patr-B alleles and was displayed by 8 of 14 animals.

Pocket analysis of Patr class I molecules.

Recent studies on the classification of HLA class I molecules suggest that the peptide-binding specificity of MHC class I molecules can be predicted by the analysis of MHC pocket sequences and specifically by the presence of certain amino acid motifs in key MHC residues (54). To analyze whether supertypes also exist for Patr molecules, we compared 14 key amino acid residues of the B and F pockets of Patr class I molecules with those of four HLA class I molecules (Table 5). The four HLA alleles belong to the HLA-A1, -A2, -A3, and -B7 supertypes, which cover more than 95% of the human population (45). Thus, identification and characterization of Patr orthologues of those HLA alleles would be invaluable for infection and vaccine studies.

TABLE 5.

Comparison of B and F pockets of HLA and Patr proteins^a

Allele	B-pocket amino acid residue								F-pocket amino acid residue						No. of HLA-Patr shared residues/total no. of residues
	9	24	45	63	66	67	70	99	74	77	95	97	114	116	Identical	Similar (* and **)
HLA-B*0701	Y	S	E	N	I	Y	Q	Y	D	S	L	S	D	Y
Patr-B*1301	Y	S	E	N	I	Y	Q	Y	D	S	L	S	E*b	Y	13/14	14/14
HLA-A*0101	F	A	M	E	N	M	H	Y	D	N	I	I	R	D
Patr-A*0301	Y**b	A	M	E	N	M	Sc	Y	D	D	I	I	R	D	11/14	13/14
Patr-A*0601	S	A	M	E	S	V*	S	Y	D	N	I	I	R	D	10/14	11/14
Patr-B*0901	Y**	T*	M	E	N	M	S	Y	D	N	I	R	Y*	Y	8/14	11/14
HLA-A*0301	F	A	M	E	N	M	Q	Y	D	D	I	I	R	D
Patr-A*0301	Y**	A	M	E	N	M	S	Y	D	D	I	I	R	D	12/14	13/14
Patr-B*0901	Y**	T*	M	E	N	M	S	Y	D	D	I	I	R	D	11/14	13/14
Patr-A*0501	S	A	M	E	S	V**	F	F**	D	D	I	L**	R	D	8/14	11/14
Patr-B*0101	Y**	T*	M	E	N	M	S	Y	D	N	W*	T*	G	Y	6/14	10/14
HLA-A*0201	F	A	M	E	K	V	H	Y	H	D	V	R	H	Y
Patr-A*0301	Y**	A	M	E	N	M**	S	Y	D	D	I**	I	R*	D	5/14	9/14
Patr-B*0501	S	A	M	E	S	V	F	F**	D	D	I**	L	R*	D	5/14	8/14
Patr-B*0101	Y**	T*	M	E	N	M**	S	Y	D	N	W	T	G	Y	4/14	7/14
Patr-B*2401	Y	T	K	E	I	S	N	F	D	N	W	T	R	F
Patr-A*0701	S	A	M	V	E	S	G	F	Y	N	I	L	E	L

^aSpecific B- and F-pocket residues were selected for comparison based on the publication of Zhang et al. (54). In addition, B-pocket residues 70 and 99 were included in the analysis based on publications from Sette and Sidney (45) and Hashimoto et al. (20). Other residues, i.e., B-pocket residues 7 and 34 and F-pocket residues 73, 80, 81, 84, 118, 123, 143, 146, and 147 (20, 45), did not differ between the analyzed HLA and Patr molecules and are therefore not listed in this table.

^b* and **, amino acids of similar size and broadly related physicochemical properties (54).

^cUnderline indicates amino acids that are not similar in size and physicochemical properties (54).

We analyzed eight polymorphic amino acid residues in the B pocket of the MHC class I complex (positions 9, 24, 45, 63, 66, 67, 70, and 99) and six residues in the F pocket (positions 74, 77, 95, 97, 114, and 116). These amino acids represent key residues for contact with the MHC binding peptide (20, 45, 54) and are listed in Table 5. In contrast, B-pocket residues 7 and 34 and F-pocket residues 73, 80, 81, 84, 118, 123, 143, 146, and 147, which have also been used for analysis in previous studies (20, 45, 54), did not differ between the HLA and Patr molecules analyzed.

As shown in Table 5, Patr-B*1301 displayed the same sequence as HLA-B*0701 in 13 of the 14 residues. The only differing amino acids, aspartic acid and glutamic acid, are of similar size and share broadly related physicochemical properties (54). None of the other Patr molecules of this study displayed a similar degree of homology in its amino acid residues at these positions.

HLA-A*0101 shared 6 of 8 key B-pocket residues and 5 of 6 key F-pocket residues with Patr-A*0301. Moreover, the nonidentical residues at position 9 of the B pocket displayed similar size and physicochemical properties. When HLA-A*0101 was compared to Patr-A*0601, all key F-pocket residues were conserved and four B-pocket residues differed. HLA-A*0101 was also related to Patr-B*0901 except for two residues at positions 97 and 116 of the F pocket, one residue at position 70 of the B pocket, and three additional residues that contained similar amino acids in both MHC proteins.

HLA-A*0301 shared identical amino acid residues with Patr-A*0301 and Patr-B*0901 in all key positions of the F pocket and displayed a very close homology in the B pocket. Patr-A*0501 and Patr-B*0101 were also related to HLA-A*0301, except for three dissimilar amino acids at positions 9, 66, and 70 of the B pocket of Patr-A*0501 and four dissimilar amino acids at positions 70, 77, 114, and 116 of Patr-B*0101.

A less stringent match was found for HLA-A*0201. Nine of 14 key pocket residues were broadly related to those of Patr-A*0301, 8 of 14 to those of Patr-B*0501, and 7 of 14 to those of the Patr-B*0101 allele, but only approximately one-third of all analyzed residues were identical.

It is notable that Patr-A*0301 was closely related to at least three human alleles, namely to HLA-A*0101, HLA-A*0301, and HLA-A*0201, and is therefore listed three times in Table 5. Likewise, Patr-B*0101 was related to two HLA alleles, namely HLA-A*0301 and HLA-A*0201, but displayed a lower degree of similarity to each of them. These results demonstrate structural similarities in the peptide-binding pocket of individual Patr and HLA molecules and suggest that some HLA-restricted peptides may also bind to Patr molecules. Other alleles, e.g., Patr-A*0701 and Patr-B*2401, differed more considerably (Table 5).

Binding of biotinylated peptides to HLA- and Patr-expressing cell lines.

Based on this sequence analysis, we examined the binding of HLA-restricted peptides to the appropriate Patr alleles experimentally. Four different HLA-restricted peptides and one Patr-restricted peptide, known to display a high binding affinity to individual MHC class I molecules, were selected (Table 2). For each peptide, the amino acid at position 5 was exchanged with a lysine residue that was attached to biotin via a 6C linker. Because the binding affinity had been determined with radiolabeled peptides and purified HLA molecules in previous studies (47), we first confirmed the binding of these biotinylated peptides to well-characterized EBV-BCLs expressing different HLA molecules. Figure 1 displays representative flow cytometry histograms evidencing the binding of the biotinylated HLA-A2-restricted peptide 2 to the HLA-A2-positive cell line JY-EBV (Fig. 1A) and to the Patr-B*0101 transfectant (Fig. 1B). The mean fluorescence intensities of both JY-EBV and Patr-B*0101 transfectants were more than fivefold higher in the presence of biotinylated peptide than in the absence of peptide, and both graphs show single peaks well separated from each other. In contrast, HLA-A1- and HLA-B7-restricted peptides did not bind (Table 4).

FIG. 1.

Flow cytometric analysis of binding of a biotinylated HLA-A2-restricted peptide to HLA-A2-expressing JY-EBV B-cells (A) and Patr-B*0101 transfectants (B). The bold profiles in the middle of the graphs represent the binding of the biotinylated HLA-A2-restricted peptide followed by avidin amplification for the Patr transfectants. The profiles on the left of the graphs represent fluorescence intensity after staining with avidin in the absence of peptide. The profiles on the far right of the graphs represent staining with an antibody targeted against MHC class I. Fluorescence-activated cell sorter analysis was repeated twice; representative experiments are shown.

The results of all binding assays performed with human EBV B-cell lines expressing different HLA molecules are summarized in Table 6. In each case, the mean fluorescence intensity (MFI) was significantly increased when the biotinylated peptide with known HLA restriction was incubated with an EBV B-cell line that expressed the corresponding HLA type but not if it expressed nonrelevant HLA types. For example, the highest MFI of HLA-A1-expressing BCLs (BCL 888) was observed in the presence of the biotinylated HLA-A1-restricted peptide. Similarly, in the binding assays with HLA-A2- and -B7-positive JY cells and HLA-A3- and -B7-positive GM 3107 cells, high MFIs were observed with the HLA-A2-, HLA-B7-, and HLA-A3-restricted peptides, respectively, but not with other peptides. Thus, the assay was suitable for specifically determining peptide binding and MHC class I restriction.

TABLE 6.

Binding of biotinylated peptides to HLA proteins

EBV cell line	Expressed HLA	Mean fluorescence intensity
		No peptide	Peptide 1 (HLA-A1)	Peptide 2 (HLA-A2)	Peptide 3 (HLA-A3)	Peptide 4 (HLA-B7)	MHC I expression
BCL 888	A1, A24 B52, B55a	4.3	41.7	7.4	4.3	9.8	759.0
JY	A2, B7	5.0	6.6	25.1	9.3	27.0	383.8
GM3101	A3, B7	3.9	4.7	4.5	15.6	9.7	398.0

^aHLA-B52 and HLA-B55 belong to the HLA-B7 supertype.

The results of the binding analysis with Patr transfectants are shown in Table 4. Because the MHC class I expression level of some transfectant cell lines was lower than that of human EBV-BCLs (Table 6), an additional incubation step with anti-avidin D and FITC-avidin D was performed to increase the FITC signal. The indicated MFI values represent the MFI of transfected cell lines minus the MFI of the parental, nontransfected cell line 721.221. The results clearly demonstrate that the Patr-A*0301 transfectant exhibits the highest affinity for the HLA-A3-restricted peptide and significantly weaker affinity for the HLA-A1-restricted peptide. Interestingly, this is consistent with the analysis of the Patr B- and F-pocket residues, which demonstrated a closer similarity of the key B- and F-pocket residues of Patr-A*0301 to those of HLA-A*0301 than to those of HLA-A*0101.

Patr-A*0601 and Patr-A*0501 displayed the highest binding affinities to the HLA-A1- and -A3-restricted peptides, respectively, which is in accordance with the high number of identical and/or similar key amino acid residues in the B and F pockets (Table 5). The Patr-B*0101 transfectant bound both HLA-A2- and -A3-restricted peptides, but the binding of the HLA-A3-restricted peptide resulted in a higher mean fluorescence intensity in the fluorescence-activated cell sorter analysis, reflecting a higher degree of similarity between the key amino acid residues of HLA-A*0301 and Patr-B*0101 than between those of HLA-A*0201 and Patr-B*0101 (Table 5). Finally, the Patr-B*0901 transfectant cell line bound the HLA-A1-, HLA-A2-, and HLA-A3-restricted peptides and the Patr-B*1301 transfected cell line bound the HLA-B*0701-restricted peptide but none of the other peptides tested.

HLA-A1-, -A2-, -A3-, and -B7-restricted HCV epitopes bound to orthologous Patr molecules are recognized by cytotoxic T cells.

After demonstrating the binding of HLA-restricted peptides to orthologous Patr molecules, we asked whether these peptide-Patr complexes would also be recognized by CD8⁺ T cells, resulting in lysis of the target cells that express the specific Patr protein.

For this purpose, individual chimpanzee Patr class I alleles, representing the animal's MHC I haplotype, were transfected into MHC-negative 721.221 target cell lines and loaded with peptides that had been identified as HLA-restricted HCV epitopes in HCV-infected patients (Table 3) and tested in a standard ⁵¹Cr release assay as targets of peptide-specific cytotoxic T-cell lines.

Cytotoxic T cells derived from chimpanzee Ch1601, for example, were tested against peptide-pulsed cell lines transfected with either Patr-A*0601, -A*0701, or -B*0302, consistent with the chimpanzee's MHC haplotype. The nontransfected parental cell line 721.221 was tested as a negative control. As demonstrated in Fig. 2, the HLA-A1-restricted HCV peptide 6 was specifically recognized in the context of the Patr-A*0601 molecule but not in the context of any of the other molecules and therefore confirms a close functional homology of HLA-A1 with Patr-A*0601. Similarly, this analysis clearly demonstrates that CTL from chimpanzee Ch1536 specifically recognized the HLA-A1-restricted HCV peptide 7 in the context of Patr-B*0901 and not in the context of any of the other MHC molecules. Finally, CTL from chimpanzees Ch1606 and Ch1605 specifically recognized an HLA-A3-restricted HCV peptide and an HLA-B7-restricted peptide only in the contexts of Patr-A*0501 and Patr-B*1301, respectively.

FIG. 2.

Patr restriction of HCV peptides. The cytotoxicity of HCV-specific, in vivo primed chimpanzee T-cell lines was determined against peptide-loaded 721.221 target cells that expressed selected individual Patr alleles. Cytotoxicity was determined in a standard 6-h cytotoxicity assay.

Efficient recognition of Patr-peptide complexes was confirmed in an autologous system with cytotoxic T cells and peptide-loaded, autologous EBV-BCLs derived from the same chimpanzee. Figure 3A demonstrates that two T-cell lines from the Patr-A*0601-positive chimpanzee Ch1601 and one T-cell line from the Patr-B*0901-positive chimpanzee Ch1536 exerted cytotoxic effector functions against autologous B-cell lines pulsed with two HLA-A*0101-restricted HCV peptides at effector-to-target ratios as low as 12.5 and 6.3 to 1. Similarly, CTL lines derived from the Patr-B*0101-positive chimpanzee Ch1552, the Patr-A*0301-positive chimpanzee Ch1606, and the Patr-B*1301-positive chimpanzee Ch1605 recognized autologous B-cell lines pulsed with HLA-A2-, HLA-A3-, and HLA-B7-restricted HCV peptides, respectively (Fig. 3B to D). As described in Table 3, all peptides have been described to be recognized by HCV-infected patients during the acute and chronic phase of hepatitis C (4, 6-8, 28, 39, 44, 49), confirming that the panel of recognized HCV epitopes is shared by humans and chimpanzees.

FIG. 3.

Cytotoxic-T-cell responses of HCV-infected chimpanzees are targeted against the same HLA-A1-, -A2-, -A3-, and -B7-restricted peptides that are recognized by HCV-infected patients. The cytotoxicity of HCV-specific T-cell lines or Patr-typed, HCV-infected chimpanzees was determined in a standard 6-h cytotoxicity assay at various effector-to-target (E/T) ratios against autologous B cells pulsed with one of the following types of peptide: (A) HLA-A1-restricted HCV peptides (filled squares, peptide 7 with chimpanzee 1601; filled circles, peptide 6 with chimpanzee 1601; open squares, peptide 7 with chimpanzee 1536); (B) HLA-A2-restricted HCV peptides (squares, peptide 11 with chimpanzee 1552; circles, peptide 13 with chimpanzee 1552); (C) HLA-A3-restricted HCV peptide (peptide 18 with chimpanzee 1606); or (D) HLA-B7-restricted HCV peptides (squares, peptide 26 with chimpanzee 1605; circles, peptide 25 with chimpanzee 1605). The percent specific cytotoxicity (cytotoxicity in the presence of the specific peptide minus cytotoxicity in the absence of the peptide) is indicated.

IFN-γ-producing T cells induced in vivo in HCV-infected chimpanzees recognize a panel of HLA-restricted epitopes in direct ex vivo assays.

Because two important functions of virus-specific T cells, i.e., cytokine production and proliferation of virus-specific T cells, require a more stringent T-cell receptor activation than induction of cytotoxicity (51), we asked whether the presentation of HLA-restricted HCV epitopes by Patr orthologues also induced these T-cell effector functions. These are particularly relevant as part of the immune response to viruses that are sensitive to cytokines. We employed the ELISPOT technique, which allows direct ex vivo quantification of cytokine-producing, antigen-specific cells from the blood without the establishment of T-cell lines. Based on the analysis of the Patr B- and F-pocket amino acids and peptide binding assays, we selected a panel of HLA-restricted HCV peptides (Table 3) according to the Patr haplotype of the individual chimpanzees and tested PBMCs for direct ex vivo IFN-γ ELISPOT analysis with PBMCs prior to and/or after HCV inoculation of chimpanzees (Fig. 4).

Specifically, PBMCs from chimpanzees that displayed the Patr-A*0601 and Patr-B*0901 haplotypes were tested with a panel of HLA-A1-restricted HCV peptides. HLA-A2-, -A3-, and -B7-restricted peptides were tested for recognition by PBMCs from chimpanzees with the Patr-B*0101, Patr-A*0301, and Patr-B*1301 haplotypes, respectively. PBMCs of naïve chimpanzees prior to HCV inoculation were used as negative controls.

Figure 4A demonstrates that the Patr-A*0601-positive chimpanzee Ch1601 mounted IFN-γ responses to 2 of 5 HLA-A1- and 3 of 3 Patr-A*0601-restricted peptides after but not prior to HCV infection. Similarly, chimpanzee 1536, which was Patr-B*0901 positive, showed strong IFN-γ responses to 4 of 5 HLA-A*0101- and 2 of 3 Patr-A*0601-restricted peptides after HCV infection. These results demonstrate that HLA-A1-restricted peptides not only bind to Patr-A*0601 and -B*0901 but are also endogenously processed in vivo and induce specific T cells in HCV-infected chimpanzees. As regards the HLA-B7 supertype (Fig. 4B), we examined four Patr-B*1301-positive, HCV-infected chimpanzees. Significant IFN-γ production by PBMCs in response to HLA-B*0701 (filled bars)- or Patr-B*1301 (open bars)-restricted peptides was not seen in any animal prior to HCV infection but was observed in all chimpanzees after inoculation with HCV. As regards the HLA-A2 supertype (Fig. 4C), Patr-B*0101-positive chimpanzees recognized the same HCV peptides as HLA-A2-positive patients with hepatitis C. While no significant IFN-γ production was observed in any chimpanzee prior to HCV infection, ELISPOT analysis with chimpanzee PBMCs after HCV inoculation showed IFN-γ production to HLA-A2-restricted peptides in all cases, and two animals also recognized the Patr-B*01-restricted peptide 30 (open bars). These results suggest that HLA-A2-restricted HCV peptides were endogenously processed and recognized by Patr-B*0101-positive chimpanzees during HCV infection.

Finally, four Patr-B*0301-positive chimpanzees (Ch1606, Ch4X0142, Ch4X0234, and Ch1629) were examined for recognition of HLA-A3-restricted HCV peptides before and after HCV inoculation (Fig. 4D). Two additional Patr-A*0301-positive chimpanzees (Ch6390 and Ch6413) were analyzed in the naive state to increase the number of negative controls. No significant peptide-specific IFN-γ response was seen in any chimpanzee prior to HCV infection, but after HCV inoculation a significant response was clearly detectable in all animals studied. In addition, three Patr-A*0501-positive chimpanzees (Ch1605, Ch1606, and Ch4X0132) were studied with these HLA-A3-restricted HCV peptides. Consistent with the sequence homology between Patr-A*0501 and HLA-A3 (Table 5), these animals mounted IFN-γ responses after but not prior to HCV inoculation. Interestingly, one other animal, Ch1536, displayed both the Patr-B*0901 and Patr-B*0101 alleles, and both alleles share key amino acids with the HLA-A3 allele (Table 5). Thus, these results demonstrate that HLA-A3-restricted HCV peptides were presented in the context of Patr-A*0301, Patr-A*0501, and potentially Patr-B*0101 and/or Patr-B*0901 in vivo and were recognized by in vivo induced CTL.

Notably, all animals mounted multispecific T-cell responses against HCV epitopes located in all HCV proteins (Fig. 4), and peptides were still recognized at a 100-fold dilution, i.e., at 0.1 μg/ml (data not shown). IFN-γ production was specific for the infecting virus and the Patr haplotype of the chimpanzee, because HLA-A2-restricted HCV epitopes but not HLA-A2-restricted HBV epitopes were recognized by Ch4X0186 (Fig. 4E). Similarly, other peptides restricted by nonrelevant Patr alleles that did not belong to the haplotype of the given chimpanzee were not recognized (Fig. 4E).

DISCUSSION

This study demonstrates a close functional homology between individual chimpanzee Patr class I and human HLA class I alleles that extends beyond genetic and structural similarities of MHC complexes and results in a common panel of endogenously processed peptides that are generated in vivo during HCV infection, bound and presented by Patr-HLA orthologues, and recognized by T cells. Since this cross-reactivity not only exists at the level of peptide binding but also at the level of antigen processing and induction of T-cell responses in vivo, it allows selection of animals according to the most common human HLA supertypes for experimental infection and vaccine studies. As evidenced by molecular and functional analyses performed in this study (Tables 4 and 5), including a formal restriction analysis with single Patr transfectant cell lines (Fig. 2) and in vivo induced CTL (Fig. 4), Patr-B*1301 functions as the chimpanzee orthologue to the human HLA-B*0701 allele, and HLA-B7-restricted HCV peptides tested in this study were also presented by Patr-B*1301. Similarly, Patr-A*0601 and Patr-B*0901 presented peptides known to bind to the human HLA-A*0101 protein, and Patr-A*0501 presented peptides known to bind to HLA-A*0301. In addition, MHC sequence comparison (Table 5), peptide binding assay (Table 4), and ELISPOT analysis of in vivo induced CD8⁺ T cells (Fig. 4) suggest that Patr-A*0301 also presents HLA-A*0101-restricted peptides and that Patr-A*0301 and Patr-B*0901 also present HLA-A*0301-restricted peptides. Finally, Patr-B*0101 and Patr-A*0301 displayed a remarkable amount of functional similarity, but less stringent sequence homology, with HLA-A*0201, the most common HLA haplotype in the Caucasian population.

While it remains possible that additional peptides not included in this study could be recognized by chimpanzees but not by humans and vice versa, the Patr-HLA homologies described and the panel of endogenously processed and presented peptides identified are valuable tools for immunological studies in the chimpanzee model. Notably, the HLA-A1, -A2, -A3, and -B7 supertypes cover more than 95% of the human population—an important consideration for the development of epitope-based vaccines and immunotherapies for infectious diseases.

At the molecular level, the peptide specificity of individual MHC molecules is determined by specific amino acid residues in the peptide-binding groove (17, 41, 46, 53). In fact, the antigen recognition sites of the MHC class I and II complexes are characterized by a higher number of nonsynonymous amino acid substitutions than are other parts of the molecules (23, 24). Moreover, the number of existing peptide-binding pockets is significantly lower than the number that is theoretically possible (13). Interestingly, genetic variation is even lower for HLA-B and -C alleles and other nuclear and mitochondrial DNA genes in humans than for the corresponding chimpanzee genes (11, 14, 19). On the other hand, human HLA-A alleles display a higher degree of polymorphism than do chimpanzee Patr alleles (35). Collectively, these observations may indicate specific adaptations after Homo sapiens and Pan troglodytes diverged from a common ancestor five million years ago (21), and it is possible that the diversity, sequence, and function of MHC alleles have been influenced by pathogen-mediated selection. For example, it has been demonstrated that the cotton-top tamarin, a primate species characterized by very low MHC class I polymorphism, is very susceptible to fatal infections with various pathogens (52). Similarly, it has been suggested that P. troglodytes may have maintained certain MHC alleles that are not present in humans (1) but may confer a selective advantage during infections with pathogens such as HCV and HIV, because both infections are clinically less severe and more frequently cleared in chimpanzees than in humans (13).

The present study does not find any evidence for this hypothesis, at least as regards the alleles studied here. In fact, up to four different Patr-A and -B alleles are shown to display sequence and functional homology to individual human MHC alleles. These results indicate that, like the human HLA supertypes, several independent chimpanzee Patr alleles with similar functional characteristics may also be grouped into supertypes. For example, the human HLA-A3 supertype consists of the HLA-A*0301, -A*1101, -A*3101, -A*3301, and -A*6801 alleles, while the corresponding chimpanzee Patr supertype consists of the Patr-A*0301, -B*0901, -A*0501, and -B*0101 alleles. This functional similarity between certain Patr and HLA alleles, which confirms and extends the results of previous studies (27, 36), is probably not limited to the presentation and recognition of peptides from HCV. It has been described, although not analyzed at the molecular level, that HIV-infected human long-term survivors and chimpanzees recognize the same conserved HIV epitopes (3) and that patients and chimpanzees with acute hepatitis B recognize shared HBV epitopes (5). Moreover, the maintenance of independent MHC class I alleles with identical function has also been demonstrated in other species, such as the red-crested tamarin (Saguinus geoffroyi) and the cotton-top tamarin (Saguinus oedipus)—lineages that diverged approximately 0.7 million years ago. For example, it has been demonstrated that influenza virus epitope-specific CTL from the cotton-top tamarin killed lymphocytes from the red-crested tamarin that expressed the influenza virus nucleoprotein (18).

Apart from the identification of HLA-Patr orthologues that display similar sequence and functional specificity in vitro, we showed that overlapping repertoires of HCV peptides and T-cell specificities are also induced in vivo in response to HCV infection. Previous studies had raised this hypothesis (5, 36) but either did not define and analyze the corresponding HLA and Patr molecules that mediate this cross-reactivity at the molecular level or did not demonstrate effector functions such as IFN-γ production, proliferation, and cytotoxicity for peptide-specific in vivo induced T cells. The present study confirms and extends these reports and suggests that the selection of specific animals for vaccine studies is feasible. The most frequent MHC class I allele in the Caucasian population is HLA-A2, and indeed, most studies on the pathogenesis of HCV infection in humans have been performed with HLA-A2-restricted peptides. However, the Patr class I orthologue of the HLA-A2 family has not been reported to date, because the sequences of all known Patr-A alleles are more closely related to the HLA-A3 lineage, which comprises the HLA-A1, -A3, and -A11 family, than to the HLA-A2 lineage (31, 35). In this study, we demonstrate that the Patr-B*0101 molecule binds all HLA-A2-restricted peptides tested and that all Patr-B*0101-positive, HCV-infected chimpanzees mount T-cell responses that produce IFN-γ and lyse autologous EBV B-cells pulsed with the HLA-A2-restricted HCV peptides tested. These results concur with a previous observation that suggested the existence of a previously unidentified HLA-A2-like Patr molecule in an individual chimpanzee, 1558 (5). Indeed, this animal was also Patr-B*0101 positive (E. Mizukoshi and B. Rehermann, unpublished data). Thus, our data show that both alleles display a considerable degree of functional similarity.

While the natural course and outcome of HCV infection is certainly influenced by additional differences between humans and chimpanzees, the data presented here will be invaluable for selecting animals on the basis of the most frequent human MHC types for vaccine studies.