HIV-1 vaccine design through minimizing envelope metastability

A coherent HIV-1 vaccine strategy addresses envelope stabilization, nanoparticle display, antibody response, and manufacture.


This PDF file includes:
Fig. S1. Effect of expression system on antigenicity and glycosylation of BG505 Env trimers.  Table S1. X-ray data collection and refinement statistics.  S1. Effect of expression system on antigenicity and glycosylation of BG505 Env trimers. (A) Additional antigenic profiles for SOSIP, HR1-redesigned, and UFO trimers derived from clade A BG505. Sensorgrams were obtained from an Octet RED96 instrument using a titration series of six trimer concentrations (200-6.25 nM by 2-fold dilution). (B) Total oligomannose content for a BG505 trimer produced in HEK293 F and ExpiCHO cells. (C) Ion mobility mass spectrometry analysis of a BG505 trimer produced in HEK293 F and ExpiCHO cells. Mobility-extracted singly charged negative-ion electrospray spectra are shown for N-linked glycans on gp140. (D) Quantitative site-specific N-glycosylation of a BG505 trimer produced in HEK293 F and ExpiCHO cells, with relative quantification shown for N-glycosylation sites on both gp120 and gp41 ECTO . The gp140 construct used in (B)-(D) is the HR1 redesign 1 (or termed gp140.664.R1), which is the immediate predecessor of the UFO design.

Fig. S3. Structural characterization of diverse UFO-BG trimers. (A)
Crystal structure of clade B tier 3 H078.14 UFO-BG trimer bound to bNAbs PGT124 and 35O22 at 4.6 Å resolution. A 2Fo-Fc electron density map (contoured at 1.0 ) showing gp120 (in yellow backbone tube model) and gp41 ECTO (in green backbone tube model) of the H078.14 UFO-BG protomer in the crystal asymmetric unit bound to PGT124 (grey and pink) and 35O22 (cyan and magenta). (B) Negative-stain EM of UFO-BG trimers produced in ExpiCHO cells followed by GNL and SEC purification. The full sets of reference-free 2D class averages are shown for clade A BG505 (tier 2) and Q842-d12 (tier 2), clade B 6240.08.TA5.4622 (tier 2) and H078.14 (tier 2), clade C Du172.17 (tier 2) and 16055-2.3 (tier 2), clade B/C CH115.12 (tier 3) and clade A/E 93JP_NH1. The percentage of native-like trimers is indicated below each panel. CD4-Ig binding to SOSIP, UFO, and UFO-BG trimers measured for five selected Envs of clades A, B, and C. BLI was performed using the same protocol as described above for antibodies.

Fig. S5. Evolutionary root of metastability and design of UFO-C trimers containing a database-derived ancestral gp41 ECTO . (A) Connection between HIV-1 evolution and Env metastability. (B) SEC profiles of UFO and UFO-BG trimers derived from three transmitted/founder (T/F) viruses, including clade B B41
, clade C CH505, and clade C 1086. The yield (mg) of SEC-purified trimer protein (fractions corresponding to 53-57ml) obtained from 100-ml ExpiCHO expression is listed for both trimer designs. (C) Antigenic profiles of B41 UFO and CH505 UFO-BG trimers measured against three bNAbs and two non-NAbs. Sensorgrams were obtained on an Octet RED96 using a trimer titration series of six concentrations (200-6.25 nM by two-fold dilution). (D) Design (left) and schematic representation (right) of the UFO-C trimers. As shown on the left, a database-derived consensus gp41 ECTO in conjunction of a generic UFO design is used to stabilize gp120s from diverse HIV-1 Envs in a hybrid form of gp140 trimer designated UFO-C. The generic HR1 redesign, an 8-aa (GS) 4 linker, is highlighted in green. (E) SEC profiles of UFO-C trimers derived from five representative stains of clade A, B, C, B/C, and A/E origins. The yield (mg) of SEC-purified trimer protein (fractions corresponding to 53-57ml) obtained from 100-ml ExpiCHO expression is listed for each of the three trimer designs (SOSIP, UFO, and UFO-C). (F) Thermal stability of 4 ExpiCHO-expressed UFO-C trimers derived from four subtypes by differential scanning calorimetry (DSC). Three thermal parameters (T m , T 1/2 and T onset ) are labeled on the DSC profiles. Data fitting was performed using the standard software provided by the vendor. The raw data and the fitted curve are plotted as black dotted lines and red lines, respectively. (G) BN-PAGE of Env protein after GNL purification but prior to SEC and of purified trimer following SEC and BN-PAGE for two UFO-C constructs derived from clade A BG505 and clade B 6240.08.TA5.4622. (H) Reference-free 2D class averages derived from negative-stain EM of the BG505 and 6240.08.TA5.4622 UFO-C trimers. Percentage of native-like trimers is indicated below each panel. (I) Antigenic profiles of five purified UFO-C trimers measured against three bNAbs and two non-NAbs. Sensorgrams were obtained on an Octet RED96 using a trimer titration series of six concentrations (200-6.25 nM by two-fold dilution). The strains tested here include clade A BG505 (tier 2), clade B 6240.08.TA5.4622 (tier 2), clade C 16055-2.3 (tier 2), clade B/C CH115.12 (tier 3), and clade A/E 93JP_NH1. The peak values at the highest concentration are summarized in a matrix, in which cells are colored in red and green for bNAbs and non-NAbs, respectively. Higher color intensity indicates greater binding signal measured by Octet.

Fig. S6. Characterization of gp41 ECTO -stabilized trimer-presenting nanoparticles.
(A) Micrographs derived from negative-stain EM of UFO-BG-FR nanoparticles of five subtypes. (B) Thermal stability of four UFO-BG-FR nanoparticles derived from three subtypes by differential scanning calorimetry (DSC). Three thermal parameters (T m , T 1/2 and T onset ) are labeled on the DSC profiles. Data fitting was performed using the standard software provided by the vendor. The raw data and the fitted curve are plotted as black dotted lines and red lines, respectively. (C) Antigenic profiles of UFO-BG-FR nanoparticles derived from five subtypes evaluated against a panel of six bNAbs and four non-NAbs. Sensorgrams were obtained from an Octet RED96 instrument using a titration series of 6 concentrations (starting at 35 nM by 2-fold dilution). The strains tested here include clade A Q842-d12 (tier 2), clade B H078.14 (tier 2), clade C Du172.17 (tier 2), clade B/C CH115.12 (tier 3), and clade A/E 93JP_NH1. (D) Sequence alignment of I3-01 (PDBID: 5KP9) and 2-dehydro-3-deoxyphosphogluconate aldolase/4hydroxy-2-oxoglutarate aldolase (TM0066) from Thermotoga maritima (PDBID: 1VLW), with the five amino acids that differ in the two sequences colored in red and the 11 amino acids at the I3-01 particle-forming interface highlighted in yellow. (E) Sequences of ten 1VLW variants (1VLW-V i , i=1 to 10) by altering amino acids at the corresponding I3-01 particle-forming interface. (F) SEC profiles of ten BG505 gp140.664.R1-PADRE-1VLW-V i constructs obtained from a Superose 6 10/300 GL column. The fusion constructs were expressed in ExpiCHO cells and purified using a 2G12 affinity column.

Fig. S7. B cell activation and in vivo evaluation of gp41 ECTO -stabilized trimers and nanoparticles.
(A) Ca 2+ mobilization in B cell transfectants carrying PGT145, PGT121, and VRC01 bNAb receptors. WEHI231 cells expressing a doxycyclin-inducible form of bNAb B cell receptor (BCR) were stimulated with anti-BCR antibodies or the indicated antigens at a concentration of 10 g ml -1 : anti-human Ig -chain F(ab') 2 ; anti-mouse IgM; an UFO-BG trimer derived from a clade A, B, C, B/C, or A/E strain, or a gp140-PADRE construct containing a redesigned HR1 bend within gp41 ECTO . (B) Testing gp41 ECTO -stabilized BG505 trimer and nanoparticle immunogens in mice and ELISA binding of purified mouse IgGs to three HIV-1 antigens, including a BG505 UFO trimer, an N332 supersite probe based on ferritin nanoparticle (1GUT_A_ES-FR) or I3-01 nanoparticle (1KIG_L_ES-2-I3-01), and a V1V2 apex probe based on ferritin nanoparticle (clade C V1V2-FR). EC 50 values are labeled for all ELISA plots except for instances in which the highest OD 450 value is below 0.1 or in cases of ambiguous data fitting. (C) HIV-1 neutralization by group-combined mouse IgGs, with IC 50 s color-coded according to the value ranging from green (weak neutralization) to red (potent neutralization). (D) ELISA binding of rabbit plasma to BG505 UFO trimer and an I3-01 nanoparticle presenting 60 N332 scaffolds (1KIG_L_ES-2-I3-01). The heat-inactivated plasma was diluted by 50-fold and subjected to a 10-fold dilution series in the assay. EC 50 binding antibody titers are labeled for all plots except for the time point of day -10 (-d10). (E) Neutralization of clade A tier 2 BG505.T332N and clade B tier 1 SF162 by rabbit plasma. The heat-inactivated plasma was diluted by 50-fold and subjected to a 3-fold dilution series in the TZM-bl assay. ID 50 NAb titers are labeled for all plots except for the time points of day -10 (-d10) and week 1 (w1), as well as week 6 (w6) for the trimer group tested against BG505.T332N. In (D) and (E), samples obtained from eight rabbits in the trimer group (34, 35, 36, and 37) and the ferritin group (62, 63, 64, and 65) at the time points of day -10 (-d10) and weeks 1, 6, 14, 22, and 30 (w1, w6, w14, w22, and w30) were analyzed in ELISA and neutralization assays. (F) Neutralization of 12 isolates in the global panel by rabbit plasma from the trimer group. (G) Neutralization of 12 isolates in the global panel by rabbit plasma from the ferritin group. In (F) and (G), samples at day -10 (-d10) and week 22 (w22) were analyzed to examine plasma neutralization of heterologous HIV-1 clones. To increase the sensitivity of detection, rabbit plasma was diluted by 20-fold followed by a 3-fold dilution series in TZM-bl assays. To reduce false positives caused by non-specific antiviral proteins present in the rabbit plasma, a simple metric was utilized that directly compares the percent neutralization of the w22 sample to that of the -d10 sample from the same animal at the 20-and 60-fold dilutions. The w22 signals at both dilutions must be 2-fold greater than the corresponding -d10 signals for a sample to be considered neutralizing a virus clone in the global panel. By this metric, the w22 samples in both trimer and ferritin groups neutralized p398F1 (clade A), p25710 (clade C), pX2278 (clade B), and pCNE8 (clade A/E) among others. ID 50 NAb titers are only labeled for p398F1, for which 60-80% neutralization was observed for w22 compared to -d10, approaching ID 50 values observed for the autologous neutralization.

PDB ID
6CE0 a Numbers in parentheses are for highest resolution shell b R sym = Σ hkl Σ i | I hkl,i -<I hkl > | / Σ hkl Σ i I hkl,i, where I hkl,i is the scaled intensity of the i th measurement of reflection h, k, l, and < I hkl > is the average intensity for that reflection c R pim = Σ hkl (1/(n-1))1/2 Σ i | I hkl,i -<I hkl > | / Σ hkl Σ i I hkl,i, where n is the redundancy d CC 1/2 = Pearson Correlation Coefficient between two random half datasets e R cryst = Σ hkl | Fo -Fc | / Σ hkl | Fo | x 100 f R free was calculated as for R cryst , but on a test set comprising 5% of the data excluded from refinement