Spatiotemporal hierarchy in antibody recognition against transmitted HIV-1 envelope glycoprotein during natural infection

Abstract

Background: Majority of HIV-1 infection is established by one transmitted/founder virus and understanding how the neutralizing antibodies develop against this virus is critical for our rational design an HIV-1 vaccine.

Results: We report here antibody profiling of sequential plasma samples against transmitted/founder HIV-1 envelope glycoprotein in an epidemiologically linked transmission pair using our previously reported antigen library approach. We have decomposed the antibody recognition into three major subdomains on the envelope and showed their development in vivo followed a spatiotemporal hierarchy: starting with the ectodomain of gp41 at membrane proximal region, then the V3C3V4 and the V1V2 of gp120 at the membrane distal region. While antibodies to these subdomains appeared to undergo avidity maturation, the early anti-gp41 antibodies failed to translate into detectable autologous neutralization. Instead, it was the much delayed anti-V3C3V4 and anti-V1V2 antibodies constituted the major neutralizing activities.

Conclusions: Our results indicate that the initial antibody response was severely misguided by the transmitted/founder virus and future vaccine design would need to avoid the ectodomain of gp41 and focus on the neutralizing targets in the V3C3V4 and V1V2 subdomains of gp120.

Fig. 1

Unrooted neighbor-joining tree depicting the genetic relatedness of full-length envelope sequences from P08 and P11 with the controls. Sequences from P08 and P11 are colored in red, those previously obtained from China are in green, those provided to us by John Mascola of Vaccine Research Center at NIH are in blue, and those commonly used reference sequences for classifying HIV-1 subtypes and CRFs were in black. The branch length is drawn to scale so that the relatedness between different sequences can be readily assessed. Individual sequences are named at the tip of the branches and their assignment to specific subtype or CRF is also indicated

Fig. 2

Robustness and specificity of combinatorial library approach in the selection of linear and conformational antigenic fragments by bnAbs with known epitope specificity. a, c Overlapping sequences of the positive yeast clones selected by bnAbs (PG09, 12A12, VRCPG04, 3BNC60, 3BNC117, 10E8, 4E10, and 2F5) and aligned to the original full-length P08-gp160 and P11-gp160 sequence used in the construction of combinatorial yeast library. Vertical dashed lines highlight the V1V2, V3V4V5, and MPER regions in which the epitopes of bnAbs are located. b, d The number of amino acid residues among the selected fragments along their corresponding positions in the P08 and P11 envelope glycoprotein. The dominant stretch of amino acid residues is numbered on the top of each graph according to HXB2 and the red vertical bars separate gp120 and gp41. The hypervarible regions V1–V5, interspersed conserved regions C1–C5 in gp120, immune dominant (ID), MPER and transmembrane domain (TM) in gp41 are also indicated

Fig. 3

Analysis of major antigenic subdomains on the transmitted/founder HIV-1 envelope based on the positive yeast clones selected by the sequential plasma samples from P08 and P11. a, c Overlapping sequences of the positive yeast clones selected by the sequential plasma samples and aligned to the original full-length P08-gp160 and P11-gp160 sequence used in the construction of combinatorial yeast library. Vertical dashed lines highlight the V1V2, V3C3V4, and the ectodomain of gp41 in which the selected fragments are predominantly located. The four red horizontal lines from each subdomain represent the selected clones for studying avidity maturation of plasma antibodies during infection (see Fig. 5). b, d Quantitative measurement of antibody recognition against distinct subdomains on transmitted/founder HIV-1 envelope during natural infection. Each subdomain was identified based on the frequency of amino acid residues among the selected fragments and colored coded. The locations of the most frequently recognized residues within V1V2, V3C3V4 and ectodomain of gp41 are indicated in parentheses according to HXB2 and highlighted in red, blue, and green, respectively. The percentages represent the area under the curve (AUC) of each antigenic subdomain verse the total AUC (shaded area). The hypervarible regions V1–V5, interspersed conserved regions C1–C5 in gp120, immune dominant (ID), MPER and transmembrane domain (TM) in gp41 are also indicated

Fig. 4

The most frequently recognized amino acid sequences on the envelope of P08 and P11. The sequences are aligned to the full-length transmitted/founder HIV-1 envelope from P08 and P11 and compared with and numbered according to HXB2. The predominantly recognized amino acid sequences are underlines. Those highlighted in red, blue and green boxes represent the piece of sequences transplanted into the CNE6 backbone to generate various chimeric viruses. The signal peptide, hypervarible regions V1–V5, interspersed conserved regions C1–C5 in gp120, immune dominant (ID), MPER and transmembrane domain (TM) in gp41 are also indicated

Fig. 5

Avidity maturation of subdomain-specific antibodies during natural HIV-1 infection. a For each distinct subdomain, four representative yeast clones were used (see Fig. 3a, c) to measure the mean fluorescence intensity (MFI) of all sequential plasma samples from P08 and P11, and from which the association (Ka) and disassociation (Kd) constants were calculated using nonlinear least squares fit. b Dynamic changes in Ka of subdomain-specific antibodies during the first 2–2.4 years of HIV-1 infection. Ka is the reciprocal of Kd

Fig. 6

Standard ELISA assay to measure the dynamic changes in plasma antibody binding of P08 (a) and P11 (b) against three forms of envelope glycoprotein; trimeric ectodomain of NL4-3 (subtype B) constructed based on BG505 SOSIP.664, monomeric gp120 and monomeric gp140 (CRF01_AE)

Fig. 7

Determination of major antigenic subdomains for autologous neutralization by comparing neutralizing sensitivity of parental and subdomain-specific chimeras to sequential plasma samples from P08 and P11. a Percent neutralization on the y axis indicates the reduction in luciferase activity in the presence of plasma verse those without. b Half-maximal inhibitory dilutions (ID₅₀) were calculated as the dilution at which the plasma demonstrated inhibition of 50 % infection compared with the controls. The actual Ka values for corresponding plasma samples against each subdomain were also included for comparative purpose. Different color intensities correspond to different ID₅₀ and Ka values for clarity, lighter for lower and darker for higher. ND not detectable