Broad neutralization of human immunodeficiency virus type 1 (HIV-1) elicited from human rhinoviruses that display the HIV-1 gp41 ELDKWA epitope

Abstract

In efforts to develop AIDS vaccine components, we generated combinatorial libraries of recombinant human rhinoviruses that display the well-conserved ELDKWA epitope of the membrane-proximal external region of human immunodeficiency virus type 1 (HIV-1) gp41. The broadly neutralizing human monoclonal antibody 2F5 was used to select for viruses whose ELDKWA conformations resemble those of HIV. Immunization of guinea pigs with different chimeras, some boosted with ELDKWA-based peptides, elicited antibodies capable of neutralizing HIV-1 pseudoviruses of diverse subtypes and coreceptor usages. These recombinant immunogens are the first reported that elicit broad, albeit modest, neutralization of HIV-1 using an ELDKWA-based epitope and are among the few reported that elicit broad neutralization directed against any recombinant HIV epitope, providing a critical advance in developing effective AIDS vaccine components.

FIG. 1.

Immunoselection schemes used to enrich for chimeric viruses with optimal binding to immobilized MAb 2F5. Pools I, II, and III (consisting, respectively, of libraries 12, 14, 16, 42, 44, 46, and 64, libraries 14 and 44, and libraries 11, 12, 14, 16, 21, 22, 24, 26, 41, 44, 61, 62, and 64) were independently subjected to increasingly stringent binding conditions. (These pools were originally designed to consist of chimeras of differing insert sizes, but a few of the oligonucleotides used to generate the viruses were cross-contaminated at an early stage.) A, B, and C correspond to the first, second, and third rounds of selection, respectively. Information in the boxes indicates the following: 2F5, the concentration (nanomolar) of 2F5 used for the immobilization of viruses to 96-well plates; Pep, the concentration (micromolar) of Ac-LELDKWASL-NH₂ peptide used for competition; elute, the concentration (micromolar) of Ac-LELDKWASL-NH₂ peptide used for elution of bound chimeric viruses; large N, not subjected to competitive immunoselection. The default “eluent” was cells.

FIG. 2.

Design of ELDKWA library III and observed sequences. Subscript numbers correspond to relative percentages encoded (in the case of the HIV residue, reflecting the values among HIV-1 isolates at the times of the library design. The chimeric virus names reflect the number of biased residues on the N- and C-terminal sides of the ELDKWA insert, respectively (1, 2, 4, or 6), the immunoselection round (A, B, or C), the concentration of Ac-LELDKWASL-NH₂ peptide used for competition (0 to 40 μM) or the possibility of no competitive peptide used for the immunoselection (N) but used for elution instead (75 to 300 μM), and the clone number. The N-terminal side of the insert of UN-14-26 has a deletion of at least a cysteine residue. Positions with a ? were unreadable and may comprise any of the 20 amino acids.

FIG. 3.

ELISA curves representing the effects of immunoselection on the binding of pools of chimeric viruses to immobilized MAb 2F5. In this example, the starting set was pool II. Subsets are designated with an A or B to indicate their derivation after one or two rounds of immunoselection, respectively. The presence of an N indicates that no competitive peptide was used. The subset numbers reflect the micromolar concentration of Ac-LELDKWASL-NH₂ peptide used to compete with the chimeric viruses for binding to immobilized 2F5.

FIG. 4.

ELISA (A) and neutralization (B) data representing the effects of immunoselection on the individual immunoselected chimeric viruses compared to those for their unselected parent pool. In this example, the starting set was pool II. The individual chimeras shown have been through two rounds of immunoselection. The presence of an N indicates that no competitive peptide was used. The subset numbers reflect the micromolar concentration of Ac-LELDKWASL-NH₂ peptide used to compete with the chimeric viruses for binding to immobilized 2F5. In the panel A key, the final number is the virus clone number.

FIG. 5.

HRV14:HIV-1 gp41 ELDKWA virus-derived reciprocal serum neutralization titers at which there was 50% inhibition of HIV-1 pseudovirus replication (luciferase activity). Sera were obtained via immunization of three guinea pigs according to the following schedules: (i) for chimeric virus with or without peptide, week 0, initial immunization, ∼50 μg virus, no adjuvant; week 4, first boost, ∼50 μg virus, no adjuvant; week 9, second boost, either ∼50 μg virus, no adjuvant, or ∼25 μg virus plus 40 μg KLH-conjugated 14-mer peptide (EQELLELDKWASLW) with CFA; in some cases, week 13, no virus, 80 μg 14-mer peptide or 9-mer peptide (LELDKWASL) with IFA; (ii) for the peptide control, 80 μg 14-mer peptide at weeks 0 (with CFA), 4 (with IFA), and 8 (with IFA). (The peptide control immunization involved larger and more frequent doses of peptide than those used for any of the peptide-boosted virus immunizations.) Data in the figure correspond to sera collected 2 weeks after the final immunization. Serum codes with the same number are from the same animal; the number of Ps indicates the number of peptide boosts given before sample collection. The core insert sequence (E/A)LDKWA is shown in red. Residues marked in green appeared at the same relative positions in one or more HIV-1 isolates represented in the Los Alamos database in 2003; residues marked in black did not. The titers are averaged from ≥3 independent experiments with the exceptions of the IgG samples, the 147 to 149 series of sera, and the clade C tests. The same 67-PP serum sample was tested in its normal, heat-inactivated form and after protein A purification. A sampling of representative curves is shown in Fig. S2 of the supplemental material.

FIG. 6.

ELISA and neutralization data for guinea pig antisera as a function of the immunization protocol. A. ELISA data show binding of sera to immobilized 14-mer peptide, Ac-EQELLELDKWASLW-NH₂, for prebleeds, intermediate bleeds, and final bleeds from individual guinea pigs. B. A subset of the neutralization data of Fig. 5 is shown for the same samples used in the ELISAs (designated with the same symbols). The samples for which reciprocal neutralization titers were <20 are illustrated with values of slightly less than 20 for simplicity. Guinea pigs were immunized (as illustrated with arrowheads) at week 0 (V, 2.8 × 10⁹ PFU chimeric virus), week 4 (V, 2.8 × 10⁹ PFU chimeric virus), and week 9 (V, 2.8 × 10⁹ PFU chimeric virus [for HRV14 and guinea pigs 63, 147, and 149] or half as much virus [1/2 V] plus half a dose of KLH-conjugated 14-mer peptide [1/2 P; 40 μg of the peptide moiety] for guinea pig 67), and week 13 (P, 80 μg of the 14-mer peptide moiety alone [or HRV14 alone]). Serum samples were collected at weeks −1, 12, and 16.