A single amino acid difference within the alpha-2 domain of two naturally occurring equine MHC class I molecules alters the recognition of Gag and Rev epitopes by equine infectious anemia virus-specific CTL

Abstract

Although CTL are critical for control of lentiviruses, including equine infectious anemia virus, relatively little is known regarding the MHC class I molecules that present important epitopes to equine infectious anemia virus-specific CTL. The equine class I molecule 7-6 is associated with the equine leukocyte Ag (ELA)-A1 haplotype and presents the Env-RW12 and Gag-GW12 CTL epitopes. Some ELA-A1 target cells present both epitopes, whereas others are not recognized by Gag-GW12-specific CTL, suggesting that the ELA-A1 haplotype comprises functionally distinct alleles. The Rev-QW11 CTL epitope is also ELA-A1-restricted, but the molecule that presents Rev-QW11 is unknown. To determine whether functionally distinct class I molecules present ELA-A1-restricted CTL epitopes, we sequenced and expressed MHC class I genes from three ELA-A1 horses. Two horses had the 7-6 allele, which when expressed, presented Env-RW12, Gag-GW12, and Rev-QW11 to CTL. The other horse had a distinct allele, designated 141, encoding a molecule that differed from 7-6 by a single amino acid within the alpha-2 domain. This substitution did not affect recognition of Env-RW12, but resulted in more efficient recognition of Rev-QW11. Significantly, CTL recognition of Gag-GW12 was abrogated, despite Gag-GW12 binding to 141. Molecular modeling suggested that conformational changes in the 141/Gag-GW12 complex led to a loss of TCR recognition. These results confirmed that the ELA-A1 haplotype is comprised of functionally distinct alleles, and demonstrated for the first time that naturally occurring MHC class I molecules that vary by only a single amino acid can result in significantly different patterns of epitope recognition by lentivirus-specific CTL.

FIGURE 1

Differential CTL recognition efficiencies and sequences of MHC I alleles. a, CTL recognized Rev-QW11-pulsed A2150 EK target cells more efficiently than A2152 and A2140 EK target cells. A2150 CTL were stimulated with Rev-QW11 peptide and percent-specific lysis was determined on A2150, A2152, and A2140 EK targets pulsed with increasing concentrations of Rev-QW11 peptide. E:T cell ratio was 20:1. EC₅₀, Peptide concentration resulting in 50% maximal-specific lysis. Actual minimum and maximum percent-specific lysis for A2150, A2152, and A2140 targets was 7.1 and 49.3, 4.9 and 32.5, and 2.7 and 34.7, respectively. b, Equine MHC I molecules 7-6 and 141 differed in only one amino acid in the α-2 domain. Amino acid sequences of 7-6 and 141 are shown with domains (46, 49) indicated. The E→ V substitution at position 152 is shown in bold. c, CTL recognized Env-RW12 more efficiently than Gag-GW12 when presented by equine MHC I molecule 7-6. A2140 CTL were stimulated with Env-RW12 or Gag-GW12 peptides and percent-specific lysis was determined on 7-6-transduced 721.221 cells pulsed with increasing concentrations of Env-RW12 or Gag-GW12 peptides. E:T cell ratios were 20:1. Actual minimum and maximum percent-specific lysis for Env-RW12 and Gag-GW12 targets was 0 and 43.8, and 2.1 and 21.9, respectively.

FIGURE 2

Presentation of CTL epitopes by equine MHC I molecules 7-6 and 141. a and b, MHC I molecule 7-6 presented Env-RW12, Gag-GW12, and Rev-QW11 to CTL. A2140 Env-RW12-, A2140 Gag-GW12-, and A2150 Rev-QW11-stimulated CTL on 7-6-transduced 721.221 cells (a) pulsed with 10⁴ nM of the corresponding peptide and on 7-6-transduced H585 EK cells (b) pulsed with 10⁴ nM of the corresponding peptide. c and d, CTL recognized Env-RW12 and Rev-QW11 when presented by equine MHC I molecule 141, but did not recognize Gag-GW12-pulsed targets expressing 141. A2140 Env-RW12-, A2140 Gag-GW12-, and A2150 Rev-QW11-stimulated CTL on 141-transduced 721.221 cells (c) pulsed with 10⁴ nM of the corresponding peptide and on 141-transduced H585 EK cells (d) pulsed with 10⁴ nM of the corresponding peptide. a-d, vLXSN, empty vector. Error bars are SE for the assay shown, derived as described in Materials and Methods. E:T ratio is 50:1.

FIGURE 3

Env-RW12- and Rev-QW11-specific CTL recognition efficiencies and competitive peptide-binding inhibition. a, CTL recognized Env-RW12 with similar efficiency when presented by equine MHC I molecules 7-6 and 141. A2140 CTL were stimulated with Env-RW12 peptide and percent-specific lysis was determined on 7-6- and 141-transduced 721.221 cells pulsed with increasing concentrations of Env-RW12 peptide. E:T cell ratios were 20:1. EC₅₀, peptide concentration resulted in 50% maximal-specific lysis. Actual minimum and maximum percent-specific lysis for 7-6 and 141 targets was 0 and 63.5, and 0 and 49.1, respectively. b, CTL recognized Rev-QW11 more efficiently when presented by equine MHC I molecule 141 than when presented by 7-6. A2150 CTL were stimulated with Rev-QW11 peptide and percent-specific lysis was determined on 7-6- and 141-transduced H585 EK cells pulsed with increasing concentrations of Rev-QW11 peptide. E:T cell ratios were 20:1. Actual minimum and maximum percent-specific lysis for 7-6 and 141 targets was 4.0 and 38.1, and 7.8 and 53.4, respectively. c and d, Unlabeled Env-RW12, Gag-GW12, and Rev-QW11 inhibited ¹²⁵I-labeled Env-RW12 binding to live 721.221 cells expressing MHC I molecule 7-6 (c), and live 721.221 cells expressing MHC I molecule 141 (d). IC₅₀, peptide concentration resulted in 50% inhibition of radiolabeled Env-RW12 binding.

FIGURE 4

Molecular modeling of the Env-RW12, Rev-QW11, and Gag-GW12 peptides bound to MHC I molecules 7-6 and 141. Top views of Env-RW12 bound to 7-6 (a) and 141 (b), Rev-QW11 bound to 7-6 (c) and 141 (d), and Gag-GW12 bound to 7-6 (e) and 141 (f). The binding clefts of 7-6 and 141 are shown in green, the peptides in red, and the ¹⁵²E(V) residue in blue. The first residue of the bound peptides is oriented down.

FIGURE 5

Molecular modeling suggested that the Env-RW12, Rev-QW11, and Gag-GW12 peptides bound to MHC I molecules 7-6 and 141 in bulged conformations, and that Env-RW12 (a) and Rev-QW11 (b) maintained similar conformations when bound to 7-6 and 141, whereas the conformation of Gag-GW12 (c) changed when bound to 141. For each peptide, the conformation when bound to 7-6 is shown in red, and the conformation when bound to 141 is shown in blue. The first residue of the bound peptides is oriented to the right.

FIGURE 6

Numbers of interactions by category between each residue of Env-RW12 and its contact residues in the 7-6 (a) and 141 (b) complex, each residue of Gag-GW12 and its contact residues in the 7-6 (c) and 141 (d) complex, and each residue of Rev-QW11 and its contact residues in the 7-6 (e) and 141 complex (f).

FIGURE 7

Interactions between contact residues of Gag-GW12 and the 7-6 and 141 MHC I molecules. a, Two salt bridges (bright green dotted lines) between ⁶⁹E (bright green) of 7-6 and the K4 residue (white) of Gag-GW12 (red ribbon). ⁶⁹E protrudes from the α-1 helix (turquoise coiled ribbon) of 7-6. The other green and blue ribbons represent the floor of the peptide-binding cleft. b, A single hydrophobic interaction (purple dotted line) between ⁶⁹E (purple) of 141 and the K4 residue (white) of Gag-GW12 (red ribbon). Other designations are the same as in Fig. 1a. c, Seven hydrophobic interactions (purple dotted lines) and three hydrogen bonds (tan dotted lines) among ⁷Y (tan), ⁹⁹Y (purple), and ¹⁵⁹Y (tan) of 7-6 and the G1 residue (white) of Gag-GW12 (red ribbon). ⁷Y and ⁹⁹Y protrude from the floor of the peptide-binding cleft (blue and green ribbons), and ¹⁵⁹Y protrudes from the α-2 helix (green coiled ribbon) of 7-6. d, Three hydrophobic interactions (purple dotted lines) between ¹⁵⁹Y (purple) of 141 and the G1 residue (white) of Gag-GW12 (red ribbon). Other designations are the same as in Fig. 1c. Images a-d were generated with the STING Millennium Suite (77, 78), based on the docking models.