Human immunodeficiency virus type 2 (HIV-2)/HIV-1 envelope chimeras detect high titers of broadly reactive HIV-1 V3-specific antibodies in human plasma

Abstract

Deciphering antibody specificities that constrain human immunodeficiency virus type 1 (HIV-1) envelope (Env) diversity, limit virus replication, and contribute to neutralization breadth and potency is an important goal of current HIV/AIDS vaccine research. Transplantation of discrete HIV-1 neutralizing epitopes into HIV-2 scaffolds may provide a sensitive, biologically functional context by which to quantify specific antibody reactivities even in complex sera. Here, we describe a novel HIV-2 proviral scaffold (pHIV-2(KR.X7)) into which we substituted the complete variable region 3 (V3) of the env gene of HIV-1(YU2) or HIV-1(Ccon) to yield the chimeric proviruses pHIV-2(KR.X7) YU2 V3 and pHIV-2(KR.X7) Ccon V3. These HIV-2/HIV-1 chimeras were replication competent and sensitive to selective pharmacological inhibitors of virus entry. V3 chimeric viruses were resistant to neutralization by HIV-1 monoclonal antibodies directed against the CD4 binding site, coreceptor binding site, and gp41 membrane proximal external region but exhibited striking sensitivity to HIV-1 V3-specific monoclonal antibodies, 447-52D and F425 B4e8 (50% inhibitory concentration of [IC(50)] <0.005 microg/ml for each). Plasma specimens from 11 HIV-1 clade B- and 10 HIV-1 clade C-infected subjects showed no neutralizing activity against HIV-2 but exhibited high-titer V3-specific neutralization against both HIV-2/HIV-1 V3 chimeras with IC(50) measurements ranging from 1:50 to greater than 1:40,000. Neutralization titers of B clade plasmas were as much as 1,000-fold lower when tested against the primary HIV-1(YU2) virus than with the HIV-2(KR.X7) YU2 V3 chimera, demonstrating highly effective shielding of V3 epitopes in the native Env trimer. This finding was replicated using a second primary HIV-1 strain (HIV-1(BORI)) and the corresponding HIV-2(KR.X7) BORI V3 chimera. We conclude that V3 is highly immunogenic in vivo, eliciting antibodies with substantial breadth of reactivity and neutralizing potential. These antibodies constrain HIV-1 Env to a structure(s) in which V3 epitopes are concealed prior to CD4 engagement but do not otherwise contribute to neutralization breadth and potency against most primary virus strains. Triggering of the viral spike to reveal V3 epitopes may be required if V3 immunogens are to be components of an effective HIV-1 vaccine.

FIG. 1.

Construction of HIV-2_KR.X7 Env scaffold and HIV-2/HIV-1 V3 chimeras. (A) The parental pHIV-2_KR.P1 env backbone was modified to include unique silent restriction sequences for XhoI, SnaBI (located within the leader peptide region [LP]), XmaI (located in the C3 coding region), and XbaI (located 3′ of the transmembrane region [TM]) to create the parental shuttle vector pHIV-2_KR.X4. Nonsynonymous mutations (see Materials and Methods) were introduced at amino acid positions 204 (E204K), 360 (T360A), and 668 (Y668H) to create the final scaffold vector pHIV-2_KR.X7. BS, bridging sheet. (B) V3 sequences for HIV-1_MN, HIV-1_YU2, HIV-1_Ccon, and HIV-1_BORI were incorporated into the pHIV-2_KR.X7 env cassette as described in Materials and Methods to generate the chimeric proviruses pHIV-2_KR.X7 MN V3, pHIV-2_KR.X7 YU2 V3, pHIV-2_KR.X7 Ccon V3, and pHIV-2_KR.X7 BORI V3.

FIG. 2.

Infectivity and Env processing of HIV-2_KR.X7/HIV-1 V3 chimeras. (A) Proviral constructs were transfected into 293T cells to generate infectious virus stocks. Virus infectivity was assessed by luciferase production (RLU) in the TZM-bl single-cycle entry assay (96, 97). Luciferase readout was normalized by HIV-2 p27 antigen quantification. Data are presented as the mean and standard deviation of three independent experiments. Env-deficient HIV-2 was tested as a negative control. (B) Virus stocks were prepared by 293T transfection. At 48 h after transfection, virus was harvested from culture supernatants, pelleted, solubilized, and subjected to sodium dodecyl sulfate-gel electrophoresis and immunoblotting with a guinea pig anti-HIV-2 gp120 polyclonal antibody. Envelope-deficient HIV-2 was included as a negative control. The positions of the gp160 precursor glycoprotein and processed gp120 are identified by arrows.

FIG. 3.

Coreceptor tropism of HIV-2_KR.X7/HIV-1 V3 chimeras and fusion inhibition by T1249. (A) TZM-bl reporter cells were incubated with 10 μM TAK-779 (CCR5 antagonist), 1.2 μM AMD3100 (CXCR4 antagonist), medium only (untreated), or a combination of both coreceptor inhibitors for 30 min prior to the addition of infectious virus stock. TZM-bl cells express high levels of surface CD4, CCR5, and CXCR4, rendering them susceptible to infection by both CCR5- and CXCR4-tropic viruses. Entry was assessed by luciferase production (RLU) measured 48 h after infection. Values are presented as percent infectivity compared to the untreated control. HIV-1_NL4.3 is a CXCR4-tropic control virus, and HIV-1_YU2 is a CCR5-tropic control virus. Data are presented as the mean and standard deviation of 4 to 12 determinations. (B) Serial dilutions of T1249 were combined with an equal volume of infectious virus stock, incubated at 37°C for 1 h, and transferred to TZM-bl reporter cells. Virus entry was measured by luciferase production 48 h later and normalized to luciferase expression in the absence of T1249. Inhibition curves for HIV-2_KR.X4, HIV-2_KR.X7 YU2 V3, and HIV-2_KR.X7 Ccon V3 are shown. IC₅₀ values are presented in Table 1. Data represented are the mean and standard deviation of three independent experiments.

FIG. 4.

Neutralization of HIV-2_KR.X7/HIV-1 V3 chimeras by HIV-1 MAbs. HIV-2_KR.X7/HIV-1 V3 chimeras and control viruses were tested for neutralization susceptibility to HIV-1 MAbs targeting the CD4bs (b12), MPER (2F5, 4E10), and CD4i (17b, 19e, 21c, E51, 4.12D, ED47, ED49) epitopes and V3 (447-52D and F425 B4e8). Fivefold serial MAb dilutions were prepared at a starting concentration of 20 μg/ml, mixed with an equal volume of infectious virus stock to give the final concentrations shown, and incubated at 37°C for 1 h prior to transfer to TZM-bl reporter cells. Virus entry was measured by luciferase production 48 h after infection and normalized to luciferase expression in the absence of MAb. Non-V3-specific HIV-1 MAbs are represented by black lines. HIV-1 V3-specific MAbs (447-52D and F425 B4e8) are shown by red lines. IC₅₀ neutralization values are presented in Table 1. Data represented are the mean and standard deviation of three independent experiments.

FIG. 5.

Epitope specificity of HIV-2_KR.X7 YU2 V3 chimera neutralization by HIV-1 V3 MAbs. HIV-2_KR.X7 YU2 V3 neutralization by 447-52D or F425 B4e8 is competed by V3_JR-FL 24-mer peptide (A), V3_YU2 24-mer peptide (B), or by Fc-V3_B FP or Fc-V3_C FP (C). V3 peptides and fusion protein (FP) competitors are shown in Table 2. For all experiments, fivefold serial dilutions of HIV-1 MAbs were combined with peptide or fusion protein and incubated at 37°C for 30 min. Virus was then added and incubated at 37°C for 1 h, and the mixture was transferred to TZM-bl reporter cells. The final peptide and fusion protein concentrations in all wells were 50 μg/ml and 10 μg/ml, respectively. Luciferase expression was assessed 48 h later and was normalized to that in the absence of antibody or inhibitor. Scrambled (scr) V3 peptide, fusion proteins lacking complete or partial V3 sequences, and medium-only controls were included.

FIG. 6.

Epitope specificity of neutralization of HIV-2_KR.X7/HIV-1 V3 chimeras by HIV-1 subtypes B and C plasmas. (A) Neutralization of HIV-2_KR.X7/HIV-1 YU2 and Ccon V3 chimeras but not parental HIV-2 strains by plasma from subjects infected by HIV-1 subtype B and C. Neutralization of the HIV-2_KR.X7 YU2 V3 chimera by HIV-1 clade B plasma is inhibited modestly by the V3_JR-FL 24-mer peptide (B) and nearly completely by Fc-V3_B FP (C, left). Neutralization of the HIV-2_KR.X7 YU2 V3 chimera by HIV-1 clade C plasma is inhibited equally by Fc-V3_B FP and Fc-V3_C FP (C, right). V3 peptides and fusion protein (FP) competitors are shown in Table 2. For all experiments, fivefold serial dilutions of plasma were combined with peptide or fusion protein and incubated at 37°C for 30 min. Virus was then added and incubated at 37°C for 1 h, and the mixture transferred to TZM-bl reporter cells. The final peptide and fusion protein concentrations in all wells were 50 μg/ml and 10 μg/ml, respectively. Luciferase expression was assessed 48 h later and was normalized to that in the absence of plasma or inhibitor. Scrambled (scr) V3 peptide, fusion proteins lacking complete or partial V3 sequences, and medium-only controls had no effect on virus neutralization.

FIG. 7.

Breadth and potency of V3-specific NAbs in plasma of clade B- and clade C-infected subjects. Eleven clade B plasmas (blue) and 10 clade C plasmas (red) were tested for V3-specific neutralizing activity against HIV-2_KR.X7 YU2 V3, HIV-2_KR.X7 Ccon V3, and HIV-1_YU2 (A) and HIV-2_KR.X7 BORI V3 and HIV-1_BORI (B). Reciprocal IC₅₀s and median values (horizontal lines) are plotted for each plasma-virus combination. IC₅₀s for all plasmas tested against control HIV-2 viruses were all <1:20. Comparisons showing statistical significance are indicated.

FIG. 8.

Exposure of V3 neutralization epitopes in the native HIV-1 Env trimer. The exposure of V3 epitopes from the primary viruses HIV-1_YU2, HIV-1_{BORI d9-4F8_1413}, HIV-1_11006-11, and HIV-1_63068-05 (49) was tested by 447-52D and F425 B4e8 neutralization in three contexts: primary HIV-1 Env (open symbol; solid line), primary HIV-1 Env after sCD4 triggering (open symbol; dotted line), and in the HIV-2_KR.X7/HIV-1 V3 Env scaffold (closed symbol). The HIV-2_KR.X7 BORI V3 chimera (B) was constructed similarly to the HIV-2_KR.X7 YU2 V3 chimera (A) (see Materials and Methods). Virus entry was measured by luciferase production 48 h after infection of TZM-bl reporter cells and normalized to luciferase expression in the absence of MAb.