Fig 1.
Identification of a signature of shared mutations in animals with higher viral loads.
(A) Env sequences through 20 weeks of infection in three RM with higher viral load setpoints reveal several shared mutations. Amino acid highlighter plot of SGS-derived gp160 Env sequences generated from plasma virus collected at weeks 4, 10, and 20 post-infection in SHIV.C.CH505-infected RM, with SHIV.C.CH505.TF as master sequence. Red tics indicate amino acid substitutions, with six high frequency mutations seen across RM annotated below. Black tics indicate deletions. (B) Plasma viral load determined by quantitative real-time reverse-transcription PCR. The dotted line indicates the assay’s limit of detection (LoD) (62 copies/ml). Euthanasia timepoints are marked with the x symbol. (C) Frequency of the six shared mutations through 20 weeks.
Fig 2.
Signature confirmation in a second SHIV.C.CH505-infected RM cohort.
(A) In vivo viral plasma kinetics from a published vaccination study of 22 RM immunized with a CH505 Env DNA vaccine or sham control and challenged with SHIV.C.CH505 [50]. Average viremia threshold of >104 copies/ml distinguished animals with higher viral loads (n = 5) from those with lower viral loads (n = 17). Geometric mean viral loads are indicated by the thick red and black lines, respectively. The dotted line indicates the assay’s limit of detection (LoD) (62 copies/ml). (B) Percentage of RM with 100% penetrance of specified mutations determined by SGS of gp160 env in high vs. low VL RM.
Fig 3.
In vivo competition of longitudinal plasmas from SHIV.C.CH505-infected RM.
(A) RM 5695 was intravenously inoculated with 1.4 mL plasma from three timepoints between weeks 4 and 20 from RM 6069, 6070, and 6072 (specifically, 200 ul each from week 4, 10, and 20 from RM 6072, weeks 10 and 20 from RM 6069, and weeks 10 and 20 from RM 6070). RM 5181 was inoculated with 0.8 mL of RM5695 plasma (200 uL each from weeks 4, 8, 12, and 24). Plasma viral load kinetics from RMs 5695 and 5181 are depicted. The dotted line indicates the assay’s LoD (62 copies/ml). Euthanasia timepoints are marked with the x symbol. (B) Highlighter plot of SGS-derived gp160 Env sequences from longitudinal plasma collected from RM 5695 and 5181. (C) Frequency of mutations in intravenously inoculated RM 5695 and 5181. Fig 3A was created with BioRender.com.
Fig 4.
Surface projection and conservation of signature mutations.
(A) The seven signature mutations are shown on a surface model of the non-glycosylated CH505 SOSIP Env trimer, PDB ID 6vy2. The CD4 binding site is highlighted in orange, the CCR5 binding site in yellow, and the seven mutations are color-coded, with K302N and V255I indicated with arrows, as they are largely hidden in this view. (B) Ribbon structure of CH505 SOSIP, with conservation shown by color and the seven mutations depicted as spheres. Conservation is defined by Shannon entropy, ranging from 0 to 3.67, as determined from a 208-virus panel [81].
Fig 5.
In vitro replication of infectious HIV-1 molecular clones bearing signature mutations.
(A) The top panel represents the gp160 Env sequence of the SHIV.C.CH505.TF IMC depicted along with the positions of signal peptide (SP), variable loops (V1-V5), CD4 binding loop (CD4BL), fusion peptide (FP), membrane-proximal external region (MPER), and mutations N130D, V255I, N279D, K302N, Y330H, N334S, and H417R. IMCs containing different combinations of signature mutations are depicted underneath. (B) Replication of mutant viruses containing single mutations compared with SHIV.C.CH505.TF are shown in primary, activated RM CD4 T cells over 9 days of culture as determined by supernatant p27 Ag levels (ng/ml). The negative control virus was Env-defective SHIV. (C) Fold-difference at peak replication vs. TF is shown. **** p < 0.0001—One-way ANOVA followed by Dunnett’s multiple comparisons tests. (D) The replication of viral clones containing combinations of mutant alleles (5MA, 5MB, 4MA, 4MB, 3MA, 3MB, 3MC, 3MD) is shown compared with SHIV.C.CH505.TF in RM CD4 T cells over 9 days of culture as determined by supernatant p27 Ag levels (ng/ml). (E) Fold-difference at peak replication vs. TF is shown. * p ≤ 0.05, *** p ≤ 0.001- One-way ANOVA followed by Dunnett’s multiple comparisons tests.
Fig 6.
Cell entry efficiency and Env conformation of single and combination mutant SHIV.C.CH505 IMCs.
(A, B). Entry efficiency by single and combination mutant clones was evaluated in a single cycle entry assay in ZB5 cells that express rhesus CD4 and CCR5 [82,83]. (C-F) Env trimer conformation or “openness” was assayed by measuring virus neutralization sensitivity to anti-HIV mAbs that target linear V3 (3869) or CD4-induced bridging sheet (17b) epitopes. The single mutant K302N and the combination mutants 3MB, 3MC and 4MA are “open” compared with the closed structure of the TF CH505 Env and other mutants. Values are mean (SD); *** p ≤ 0.001, **** p < 0.0001—One-way ANOVA followed by Dunnett’s multiple comparisons test.
Fig 7.
In vivo competition of candidate SHIV.C.CH505 combination clones.
Three outbred RM were infected with 80 ng p27 Ag of an inoculum stock comprised of approximately equal amounts of 8 variants (A, left) and followed longitudinally for plasma virus load (B) and for relative proportions of each variant at weeks 1, 2, and 4 post-infection as determined by Illumina sequencing and shown in the donut plots (A, right). Mutant 5MA replicated most efficiently in all three animals. In this experiment, the wildtype TF SHIV.C.CH505 was not tested to allow for a more direct comparison of the replication efficiency of the different combination mutants. Fig 7A was created with BioRender.com.
Fig 8.
In vivo competition of SHIV.C.CH505.v2 vs. SHIV.C.CH505.TF.
(A) Four RM were IV inoculated with approximately equal amounts of SHIV.C.CH505.v2 (variant 5MA) and SHIV.C.CH505.TF. Two animals (RM T276 and T277) received 15 ng p27 Ag of each variant whereas two other animals (RM T278, and T279) received 150 ng p27 Ag of each variant. The relative proportion of the different viruses in the challenge stock and at weeks 1, 2 and 4 post-infection as determined by illumina sequencing are shown in donut plots. (B) In vivo plasma virus kinetics are shown over time. The dotted line indicates the assay’s LoD (62 copies/ml). Euthanasia timepoints are marked with the x symbol (C) Modeling of the relative frequency of V2 vs. TF in the 4 RM over time is shown with a best-fit approximation in purple. Fig 8A was created with BioRender.com.
Fig 9.
SHIV.C.CH505.v2 persists through suppressive ART.
(A) In vivo viral kinetics of 4 RM IV infected with barcoded SHIV.C.CH505.v2, with daily ART initiated at 9 months post-infection, maintained for 5 months, then monitored for an additional 6 months post-ATI. The grey box indicates the 5-month period of daily antiretroviral therapy (dolutegravir, emtricitabine and tenofovir) administered subcutaneously. (B) Intact and total proviral DNA levels for two RM sampled on suppressive ART at day 330 (PBMC) and day 358 (LN) post-infection. (C) SGS of pre-ART and rebound plasma env reveals multiple distinct rebound virus lineages reactivating from latency.